Water Use Efficiency and Water Productivity - University of Malta

O ptions Méditerranéennes

W ater Use Efficiency and Water Productivity WASAMED Project

CIHEAM

SERIES B: Studies and Research Number 57

Water Use Efficiency and Water Productivity WASAMED Project (EU contract ICA3-CT-2002-10013)

Proceedings of 4th WASAMED Workshop Edited by: Nicola Lamaddalena, Muhammad Shatanawi, M laden Todorovic, Claudio Bogliotti, Rossella Albrizio 2007 Number B 57

CIHEAM / IAM B -EU DG Research

CIHEAM

Centre International de Hautes Etudes Agronom iques Méditerranéennes International Centre for Advanc ed Mediterranean Agronomic Studies Président / Chairm an: Mouïn HAMZÉ Secrétariat Général / Secretary General: Bertrand HERVIEU 11, rue Newton 75116 PARIS Tel. + 33 1 5323 9100 - Fax + 33 1 5323 9101 Email : [email protected]

IAM

Instituts Agronom iques Méditerranéens Mediterranean Agronomic Institutes

Ba ri - Chania - Montpellier - Za ra goza IAM - Bari Directeur : Cosimo LACIRIGNOLA Via Ceglie 9 - 70010 Valenzano, Bari, ITALIE Tel. + 39 080 4606 111 - Fax + 39 080 4606 206 Email : [email protected] www.iamb.it IAM - Chania Directeur : Alkinoos NIKOLAIDIS P.O. Box 85 - 73100 Chania, Crete, GREECE Tel. 30 2821 03 50 00 - Fax 30 2821 03 50 01 Email : [email protected] www.maich.gr IAM - Montpellier Directeur : Vincent DOLLÉ 3191, route de Mende 34033 Montpellier Cedex 5, FRANCE Tel. 33 04 6704 60 00 - Fax 33 04 6754 25 27 Email : [email protected] www.iamm.fr IAM - Zaragoza Directeur : Luis ESTERUELAS Apartado, 202 - 50080 Zaragoza, SPAIN Tel. 34 976 71 60 00 - Fax 34 976 71 60 01 Email : [email protected] Www.iamz.ciheam.org

Options Méditerranéennes, Séries B n. 57

Water Use Efficiency and Water Productivity

Proceedings of 4th WASAMED (WAter SAving in MEDiterranean agriculture) Workshop Amman (Jordan), 30 Sept. - 4 Oct. 2005 EU contract ICA- CT - 2002- 10013

Opinions, data and facts exposed in this number are under the responsibility of the authors and do not engage either CIHEAM or the Member-countries.

Les opinions, les données et les faits exposés dans ce numéro sont sous la responsabilité des auteurs et n’engagent ni le CIHEAM, ni les Pays membres. 2

CIHEAM Centre International de Hautes Etudes Agronomiques Méditerranéennes

Options

méditerranéenne

Directeur de la publication: Bertrand Hervieu

SERIES B: Studies and Research Number 57

Water Use Efficiency and Water Productivity Proceedings of 4th WASAMED (WAter SAving in MEDiterranean agriculture) Workshop Amman (Jordan), 30 Sept. – 4 Oct. 2005 EU contract ICA- CT - 2002- 10013 Edited by:

Nicola Lamaddalena, Muhammad Shatanawi, Mladen Todorovic, Claudio Bogliotti, Rossella Albrizio

2007

La maquette et la mise en page de ce volume de Options Méditerranéennes Séries B ont été réalisées à l’Atelier d’Edition de l’IAM Bari This volume of Options Méditerranéennes Series B has been formatted and paged up by the IAM Bari Editing Board

Water Use Efficiency and Water Productivity Proceedings of 4th WASAMED (WAter SAving in MEDiterranean agriculture) Workshop Amman (Jordan), 30 Sept. - 4 Oct. 2005 EU contract ICA- CT - 2002- 10013 Edited by:

Nicola Lamaddalena, Muhammad Shatanawi, Mladen Todorovic, Claudio Bogliotti, Rossella Albrizio

Bari: CIHEAM (Centre International de Hautes Etudes Agronomiques Méditerranéennes) p. 294, 2007 Options Méditerranéennes, Séries B, N. 57

ISSN : 1016-1228 ISBN : 2-85352-355-1

© CIHEAM, 2007 4

Reproduction partielle ou totale interdite sans l’autorisation d’ « Options Méditerranéennes »

Reproduction in whole or in part is not permitted without the consent of « Options Méditerranéennes »

Options méditerranéennes, Series B, n°57


CONTENTS FOREWORD........................................................................................................................... 3 INTRODUCTION .................................................................................................................... 5

KEYNOTE PAPERS WATER USE EFFICIENCY IN IRRIGATED AGRICULTURE: AN ANALYTICAL REVIEW . 9 A. HAMDY

FUTURE OPTIONS AND RESEARCH NEEDS OF WATER USES FOR SUSTAINABLE AGRICULTURE .................................................................................................................... 21 M.R. SHATANAWI

RELATING WATER PRODUCTIVITY AND CROP EVAPOTRANSPIRATION................... 31 L.S. PEREIRA

SYSTEMATIC APPROACH TO THE IMPROVEMENT OF AGRICULTURAL WATER USE EFFICIENCY......................................................................................................................... 51 T.C. HSIAO

ON THE CONSERVATIVE BEHAVIOR OF BIOMASS WATER PRODUCTIVITY ............ 59 P. STEDUTO, T.C. HSIAO, E. FERERES

TECHNICAL INTERVENTIONS TO IMPROVE WATER USE EFFICIENCY IN IRRIGATED AGRICULTURE .................................................................................................................... 63 A. HAMDY

COUNTRY REPORTS TECHNIQUES FOR IMPROVING WATER USE EFFICIENCY IN GREENHOUSE CULTIVATION IN CYPRUS ................................................................................................. 74 P. POLYCARPOU, D. CHIMONIDOU, I PAPADOPOULOS

REVIEW AND ANALYSIS OF WATER USE EFFICIENCY AND WATER PRODUCTIVITY IN EGYPT ............................................................................................................................. 82 M. NASR ALLAM, R. ABDEL-AZIM

WATER USE EFFICIENCY AND WATER PRODUCTIVITY IN GREECE.......................... 92 A. KARAMANOS, S. AGGELIDES, P. LONDRA

IRRIGATED AGRICULTURE AND WATER USE EFFICIENCY IN ITALY....................... 102 M. TODOROVIC, A. CALIANDRO, R. ALBRIZIO

EFFECTS OF DEFICIT IRRIGATION ON YIELD AND WATER USE EFFICIENCY OF SOME CROPS UNDER SEMI-ARID CONDITIONS OF THE BEKAA VALLEY OF LEBANON .......................................................................................................................... 139 F. KARAM, K. KARAA, N. TARABEY

1



WATER USE EFFICIENCY AND WATER PRODUCTIVITY IN MALTA .......................... 156 G. ATTARD, J. MANGION, P. MICALLEF

IRRIGATION RESEARCH RESULTS IN THE SYRIAN ARAB REPUBLIC ..................... 166 A. KAISI, Y. MOHAMMAD, Y. MAHROUSEH

WATER USE EFFICIENCY IN TURKEY............................................................................ 178 R. KANBER, M. UNLU, E.H. CAKMAK, M. TUZUN

OTHER CONTRIBUTIONS WATER USE EFFICIENCY IN C3 CEREALS UNDER MEDITERRANEAN CONDITIONS: A REVIEW OF SOME PHYSIOLOGICAL ASPECTS............................................................ 192 E.A. TAMBUSSI, J. BORT, J.L. ARAUS

PRODUCTIVITY OF THE POTATO CROP UNDER IRRIGATION WITH LOW QUALITY WATERS ............................................................................................................................ 208 N. BEN MECHLIA, K. NAGAZ, J. ABID-KARRAY, M.M. MASMOUDI

USE OF THE HEAT DISSIPATION TECHNIQUE FOR ESTIMATING THE TRANSPIRATION OF OLIVE TREES................................................................................ 215 J. ABID KARRAY, M.M. MASMOUDI, J.P. LUC, N. BEN MECHLIA

WATER RESOURCES MANAGEMENT AT THE RIVER BASIN LEVEL: AN INSTITUTIONAL ANALYSIS.............................................................................................. 221 A. BILLI, A. QUARTO, E. ZINI

THE ECONOMICS OF WATER EFFICIENCY: A REVIEW OF THEORIES, MEASUREMENT ISSUES AND INTEGRATED MODELS ................................................ 231 A. BILLI, G. CANITANO, A. QUARTO

PROPOSAL FOR THE INTEGRATION OF IRRIGATION EFFICIENCY AND AGRICULTURAL WATER PRODUCTIVITY...................................................................... 269 B. BLÜMLING*, H. YANG**, C. PAHL-WOSTL

EFFECTIVENESS OF INDICATORS FOR SUSTAINABLE WATER USE IN AGRICULTURE .................................................................................................................. 287 G. ÖZEROL

2



FOREWORD The escalating water crisis in arid and semi-arid areas of the Mediterranean constitutes a major threat to global progress towards sustainable development of the entire region. The region is one of the largest food importers and the forecasts indicate an increase of food import due to strong population growth and increased demand for limited freshwater resources. Modern strategies for water saving in the Mediterranean region are focusing awareness on the “demand management” of irrigated agriculture, while looking at improvement of water use efficiency, throughout the whole water path from the source to the irrigation fields, and on a better productivity of 3 each m of water used in agriculture with the primary objective of water saving and increasing crop production and income of farmers. This requires considerable efforts in the implementation of the best management practices in the region which are compatible with the technical, financial and socioeconomic capabilities of the irrigation environment. In the Mediterranean, the success in water saving strategies depends on the level of understanding and integration of cultural, economic, institutional and environmental contexts. Nevertheless, water saving is still below expectations due to the lack of effective regional coordination, communication and dialogue among all the relevant stakeholders, and the lack of a common-shared knowledge to support formulation of adequate national and regional water saving programmes and sustainable water policies. In this view, WASAMED (WAter SAving in MEDiterranean agriculture) project has been envisaged and implemented at the beginning of 2003. WASAMED is a 48-month Thematic Network (ICA3-CT2002-10013; http://wasamed.iamb.it), granted by the International Scientific Cooperation of the Directorate General of Research, European Commission, in the frame of the 5th Framework Programme. The Network is coordinated by CIHEAM-Bari Institute and involves 42 partners from 16 countries including scientific institutions, decision and policy makers, researchers, end-users, Water User Associations (WUAs) and NGOs. One of the main targets of the WASAMED is seeking consensus on best options, goals and indicators for water saving in the region mainly through the exchange of knowledge, dissemination of information and realization of five Euro-Mediterranean Workshops and an International Conference by the end of project. The Workshop on ”Water Use Efficiency and Water Productivity”, held in Amman (Jordan), in October 2005, is the fourth of a series of five Thematic Workshops and presents the result of shared efforts of all the WASAMED partners and, particularly, of the University of Jordan, the National Center for Agricultural Research and Technology Transfer (NCARTT), and CIHEAM-IAMB which jointly organized the event. This volume of Option Méditerranéennes includes twenty-one contributions presented in the Workshop by both the partners of WASAMED and by external speakers, invited on the basis of their reputation and experience in the subject. In this occasion, I would like to express my deepest thanks to all the institutions and scientists and water experts contributing in the realization of the Workshop and this Proceedings. I hope that the papers presented in this volume and the workshop recommendations can respond to the need of building a regional knowledge on Water Use Efficiency and Water Productivity and the implementation of best management practices on the ground in the Mediterranean region. Finally, on behalf of CIHEAM-Bari Institute and in the name of all institutions involved in the WASAMED project, I would like to express my sincere appreciations and gratitude to the European Commission INCO-MED programme for its financial support and valuable work in following the progress of the project. Cosimo Lacirignola Director, CIHEAM-IAM Bari

3


4




INTRODUCTION The contemporary world is strongly linked to the economic parameterization and the water-related “efficiency” terms are commonly used indicating “the level of performance” of a system when water is transported, consumed and/or used in the process of production of a specific good. In each specific process, the “efficiency” refers to the output/input ratio and it is usually expressed as a nondimensional value ranging between 0 and 1. Throughout the years, several water-related efficiency terms have been applied to different fields (agriculture, engineering, economy, industry, etc). From an engineering point of view, the concept of efficiency is mainly linked to the hydraulic performance of water conveyance and distribution systems. Both indicates the ratio between the water output at the outlet of the system and water input at the inlet of the hydraulic structure/system under consideration. Consequently, in the engineering sector are commonly used the terms as: water conveyance efficiency (the ratio between the volumes of water at the outlet and inlet of a water conveyance network), water distribution efficiency (the ratio between the volumes of water at the outlet and inlet of a water distribution network), water storage efficiency (the ratio between the volume of water diverted for irrigation and the volume entering a storage reservoir for the same purpose), irrigation system efficiency (the ratio between the volume of water effectively available for crop use and the volume applied on the field), etc. In the agricultural sector, the “Water-Use Efficiency” (WUE) term has been widely in use since the middle of sixties when Viets has introduced it in his article on “Fertilizers and the efficient use of water” (Viets, F.G., 1962, Adv. Agron., 14: 223-264) and on soil management practices (Viets, F.G., 1966, “Increasing water use efficiency by soil management”. In Plant environment and efficient water use, eds. W.H.Pierre et al. Madison, Wisconsin, USA: American Society of Agronomy). Since that time, the WUE term has become a common tool to describe, at different scales, the relationship between the crop growth development and the amount of water used. For example, at the leaf scale, the plant physiologists use the Photosynthetic Water-Use Efficiency referring to the ratio of net assimilation to transpiration; at the plant (canopy) scale, the agronomists employ both Biomass and Yield Water-Use Efficiency indicating the ratio between the biomass and yield, respectively, and crop (evapo)transpiration. Nevertheless, in all these cases, the WUE terms are not non-dimensional values and they do not represent an output/input ratio of only one entity. In fact, they describe the processes in which water is consumed and/or used to produce new entities (e.g. biomass, yield, etc.), indicating the quantity “produced” per surface area from the unit amount of water. For this reason, several alternatives have been proposed in the recent years to convert these WUE terms into some others, more appropriate terms. Among such attempts, the “Water Productivity” (WP) term is going to assert and to spread in the agricultural scientific community as it describes better the ratio between the quantity of a product (biomass or yield) and the amount of water depleted or diverted. Certainly, the WP term can be used from leaf to plant (canopy) and field scale whereas the choice of both, the nominator and the denominator of the WP ratio, may vary with the objectives and domain of interest of the study. The Water Productivity term plays a crucial role in modern agriculture which aims to increase yield production per hectare per unit of water used, both under rainfed and irrigated conditions. This objective can be pursued either increasing the marketable yield of the crops for each unit of water transpired, or reducing the outflows and the atmospheric water depletion, or enhancing the effective use of rainfall, of the water stored in the soil, and of the marginal quality water. The first option concerns to the need for improving crop yield; the second one intends to increase the beneficial depletion (transpiration) of water supply against the non-beneficial portion (evaporation); the third aims to utilize efficiently the water resources, further than the water diverted from reservoirs, streams or groundwater sources. All these principles lead to the improvement of the on-field management aspects of crop growth, emphasizing the importance of the application of the best crop management practices which will permit to use less water for irrigation, decrease evaporation losses, optimize fertilizer supply, pest control, energy consumption, soil conditions, etc. This is of particular importance in arid and semi-arid regions with limited water supply, where the farmers are frequently constrained to apply deficit irrigation strategies and to manage water supply in accordance with the sensitivity of crop’s growing stages to water stress. In those situations, the economic aspects of WUE would get additional importance due to farmer’s interest to improve economic return from the investments in irrigation water supply.

5



The 4th WASAMED Workshop on “WUE and WP”, held in Amman (Jordan) from Sept. 30 to Oct. 4, 2005, was dedicated to both Water-Use Efficiency and Water Productivity. The main objective of this Workshop were to assess the actual situation of research and agricultural practices related to the WUE and WP on the leaf, canopy and field scale in the arid and semi-arid areas of the Mediterranean region and to address the opportunities for water saving in irrigated agriculture. Accordingly, the Workshop highlighted and discussed the following main issues: a) the hydrological aspects and agronomic management strategies of WUE and WP (e.g. evapotranspiration estimates, crop water requirements, irrigation scheduling and soil water balance), b) eco-physiological aspects of WUE and WP (e.g. plant-water relationships, crop growth, capture and/or use efficiency of resources and yield production) and c) economic aspects of WUE and WP (e.g. water allocation – cropping pattern optimization at different scales and costs of management practices); d) sustainability of WUE and WP options, accounting social, economic, environmental and policy dimensions and inter-linkages among them. This volume of “Option Méditerranéennes” represents only a part of presentations carried out during the Workshop and consists of twenty-one scientific contributions. In the first part are given six keynote documents covering different aspects of water use efficiency and water productivity in irrigated agriculture and indicating future options, research needs and possible technical interventions to improve water use in agriculture sector. The second chapter is dedicated to the country reports presented for eight Mediterranean countries (Cyprus, Egypt, Greece, Italy, Lebanon, Malta, Syria and Turkey). The last part of this volume is related to the research results of several experiments carried out in the Mediterranean region and the analysis of socio-economic and institutional and policy aspects of water resources management. We truly hope that this volume of “Options Méditerranéennes” will contribute in searching for a consensus on the framework of interventions (measures) to disseminate the best-performing WUE and WP practices accounting for different socio-economic conditions, environmental scenarios and scaling aspects characterizing the Mediterranean region.

Nicola Lamaddalena Muhammad Shatanawi Mladen Todorovic Claudio Bogliotti Rossella Albrizio

6



KEYNOTE PAPERS

7


8




WATER USE EFFICIENCY IN IRRIGATED AGRICULTURE: AN ANALYTICAL REVIEW

A. Hamdy CIHEAM - Mediterranean Agronomic Institute-Bari, Italy SUMMARY – The growing water scarcity and the misuse of available water resources are nowadays major threats to sustainable development for most developing arid and semiarid countries of the Mediterranean. For most countries of the region, the important role agriculture could play in not only feeding and clothing burgeoning population is well recognized, but also in increasing the limited available water supply by reducing water losses and by increasing the water use efficiency in the irrigation sector. Avoiding water conflicts among the water user sectors, achieving water security and food security is fundamentally a matter of water use efficient rate in the irrigation sector. The importance of efficiency in water use clearly varies across regions and nations as well as through time. The prevailing geographic, economic and social conditions of the nation play an important role in examining the efficient use of water. In agriculture, water use efficiency may be defined quite differently by a farmer, a manager of an irrigation project, or a river basin authority. Efficiencies in the use of water for irrigation consists in various components and takes into account losses during storage, conveyance and application to irrigation plots. Identifying the various components and knowing what improvements can be made is essential in making most effective use of this vital and scarce source in the Mediterranean agriculture areas. This what will be addressed in the presented analytical review paper. Key words: water use efficiency, definitions, irrigation.

INTRODUCTION Water scarcity has been reflected in traditional social and economic systems in arid areas of the Mediterranean region. During the past thirty to forty years, population growth, urbanization and economic development have depleted the region's economically exploitable water resources. Water scarcity, exacerbated by water quality deterioration and the lack of effective water management, has become a major problem in several arid countries, and even in the humid ones. Water shortage is not a new phenomenon in the Mediterranean countries. What is new, however, is that it is occurring in an increasingly changed environment and this makes it more serious and longlasting. The most recent droughts in the summers of 1989 and 1990 marked a turning point. They highlighted the vulnerability of water supplies even in the industrialized northern Mediterranean countries which had always relied on adequate per capita of rainfall. Water crisis is endemic or permanent in some southern Mediterranean areas, but it has now even reached towns and villages in France, Spain, Italy and Greece, obliging them to impose temporary restrictions. The shortfall in quantity has been compounded by a decrease in quality due to contamination of surface or underground water. Why scarcity emerges as a major problem in most of arid and semi-arid countries of the Mediterranean? The answer to this question implies several reasons that could be outlined in the following: water withdrawals reach the physical limits of available natural water resources (Palestine, Israel, Egypt, Libya and Jordan); physical conditions make inter-basin transfers or development of deep aquifers to balance supply and demand very costly, often requiring special provisions for maintenance and operation combined with uncertainties regarding sustainability of the new sources (Morocco, Tunisia);

9



low water production efficiencies due to low cost of water or weaknesses in the deployment of recent advances in technology (Algeria); and loss of affordable potable water because of pollution and environmental degradation.

Regardless of the specific causes, existing institutions are not amenable to cope with the spiralling increases in aggregate demands for water by municipal, industrial, tourist, and agricultural uses while preventing pollution.

WHAT IS EFFICIENT WATER USE? The term water use efficiency originates in the economic concept of productivity. Productivity measures the same amount of any given resource that must be expended to produce one unit of any goods or service. Thus, water productivity might be measured by the volume of water taken into a plant to produce a unit of the output. In general, the lower the resource input requirement per unit, the higher the efficiency. In any environmental resource context, however, the efficiency concept must be extended to include considerations of quality. Any effort to improve water use efficiency should be consistent with maintaining or improving water quality. Taking both quantity and quality into account, the following definition applies. Water use efficiency includes any measure that reduces the amount of water used per unit of any given activity, consistent with the maintenance or enhancement of water quality. The importance of efficiency in water use clearly varies across regions and nations, as well as through time. Geographically, for instance, water availability will condition the manner in which use patterns develop. Other things being equal, arid and semi-arid regions require a greater efficiency of water use than humid ones. Economic conditions will often lead to greater or lesser water use efficiency. Many regions of the world have been assisted in their development through public financing of water development. While the benefits or costs of such projects in efficiency terms are often debatable, the main point here is that economic factors can influence water use efficiency. Social conditions also play an important role in examining the efficient use of water resources. The literature reveals many areas where public education has led to conservation and better use of available water supplies. Here it is of interest to report some of the definitions for efficient water use presented by several authors during the International Seminar on "Efficient Water Use", Mexico City (1991). J. Bau (1991) affirms that the efficient water use consists in optimizing water use. Walker, Richardson and Seveback (1991) pointed that efficient water use means optimizing water usage and ensuring efficiency in its use. Arreguin (1991) stated that efficiency may be obtained by optimizing the use of water and infrastructures through active participation by users with a sense of social responsibility. Gloss (1991) indicated that efficient water use should be considered from different points of view. There is absolute efficiency to use the least amount of water possible, economic efficiency, which seeks to derive maximum economic benefits, social efficiency which strives to fulfill the needs of the user community, ecological efficiency which guarantees natural resource conservation, and institutional efficiency which qualifies the function of an institution regarding its water related tasks. Depending on the conditions of each user system, these non-exclusive definitions can be achieved simultaneously. The abovementioned definitions indicate that an examination of water use efficiency requires a multi-dimensional approach. In addition to the physical elements, social, economic and environmental factors must be carefully considered.

10



WATER USE EFFICIENCY IN AGRICULTURE In agriculture, water use efficiency may be defined quite differently by a farmer, a manager of an irrigation project or a river basin authority. For example, on-farm irrigation efficiencies and project efficiencies may be low, but substantial water losses may infiltrate in the soil, recharge the acquifers and may be pumped up again for re-use, either in the same project area or in another downstream. Other losses, such as overland flow, may feed drainage systems or rivers, and may be pumped or diverted for re-use. By recycling losses, river basin efficiencies could become very high. The water saving gained from introducing new technologies would be restricted to saving in evaporation losses from wetted land surfaces and water puddles, and evaporation losses from non-beneficial vegetation, which may be substantially less than the savings experienced on the farm. Clearly, any water conservation project should be carefully appraised by using adequate geo-hydrological information to study the project's effect on the water balance in the river basin. In water use a distinction should be made between technical efficiency and economic efficiency. On one hand, technical efficiency may be low in a project area, but may be high in the river basin if water is recycled. On the other hand, water losses in project area and recycling particularly when high pump lifts are involved may reduce economic efficiency. Initial water losses may lead to undesirable effects, such as waterlogging and salinity. A third way to express water use efficiency is through production per cubic meter of water.

IRRIGATION EFFICIENCIES IN THE MEDITERRANEAN REGION In the Mediterranean area irrigation represents 72% of the total water withdrawals. Irrigation is extremely water intensive. It takes about 1000 tons of water to grow one ton of grain and 2000 tons to grow one ton of rice. Despite the high priority and massive resources invested in the water resources development, the performance of large public irrigation systems has fallen short to expectation in developing and developed countries of the Mediterranean area. Competitive and inefficient use of limited regional water supplies by irrigated agriculture is a major threat to sustainability of water supplies. Very often the conveyance losses of conduits (unlined canals or leak pipes) are much too large, a 30% loss percentage of the available water is common in irrigation systems. Another cause of inefficient water use is the emphasis on meeting demand by constructing new supply facilities rather than improving the efficiencies of the existing ones. In most countries of the region, significantly more water is delivered per unit area than is required, leading to low irrigation efficiencies. The area irrigated in many irrigation systems is much less than the area commanded and annual cropping intensities are lower than anticipated. Water deliveries rarely correspond in quantity and timing to the true requirements of the farmer's crops, leading to loss in productivity. In many irrigation systems water is distributed inequitably between farmers near the head-end reaches of the system, where water is short, and those less fortunate farmers located downstream. The quantities of water consumed by the crops in an irrigation project are considerable. But the volumes of water handled by the project system have to take account of system efficiency, a product of efficiency during: (i) conveyance, (ii) distribution, (iii) field application. Average losses in irrigation projects suggest that only about 45% of water diverted or extracted for irrigation actually reaches the crops. But losses vary widely, those in the conveyance system taking

11



the water to the irrigation site may vary between 5 and 50 percent. (Table 1) shows the technical irrigation efficiencies that could be expected for conveyance and distribution efficiency in the Mediterranean region. Project efficiencies range from 30 to 65 percent, depending on the sophistication of the irrigation system and the on-farm irrigation technology in use. Table 1 - Typical project irrigation efficiencies (in percent) Category

A. Large-scale irrigation 1. Traditional open canal system (manual control) (e.g. Turkey) 2. Open canal systems with hydraulic control and surface irrigation (e.g. Morocco) 3. Open canal systems with manual control, on-farm storage and sprinkler/drip (e.g.Jordan) 4. Open canal systems with hydraulic control, buffer or on-farm storage and sprinkler/drip 5. Pipe conveyance systems with sprinkler/drip (e.g. Cyprus) B. Groundwater irrigation 6. Lined field channels and on-farm surface (gravity) 7. Pipe systems and on-farm sprinkler/drip

Conveyance and Distribution

Field Application

Project *

60

50

30

70

60**

40

75

70

55

85

70

60

95

70

65

80 95

50 70

40 65

Notes: * Gravity (surface) irrigation on the farm ** Project efficiencies are rounded to nearest 5 percent Efficiency in the use of water for irrigation consists of various components and takes into account losses during storage, conveyance and application to irrigation plots. Identifying the various components and knowing what improvements can be made is essential to making the most effective use of this vital but scarce source in Mediterranean agricultural areas.

IRRIGATION EFFICIENCY DEFINITIONS An efficiency is generally defined as the ratio of output over input, and is expressed as a percentage. In irrigation, efficiency was first defined by Israelsen (1932) as “the ratio of irrigation water transpired by the crops of an irrigation farm or project during their growth period, over the water diverted from a river or other natural source into the farm or project channel or channels during the same period of time”. This definition has not essentially changed in the years that have since passed although numerous definitions of irrigation efficiency were developed (see Appendix). Basically, they can be divided into three main groups: Definitions based on measured volumes of water; Definitions based on measured depths of application; Definitions based on other criteria, mainly related to yield.

A – Definitions based on volumes of water Most of definitions presented in Appendix are based on ratios of water volumes. These definitions have the advantage of being relatively easy to use: the volume of water delivered to the soil, a field, or a distribution system can be measured, and the volume of water “stored in the rootzone during the irrigation” or “evapotranspired by the crop” can be either measured or calculated. Definitions based on total volumes delivered to or in a system have disadvantage that, for some efficiencies, the uniformity of application or supply cannot be taken into account: the volume of water

12



delivered can match the volume required, but the division of water can be such that a part of the system or field is under-irrigated and another part is over-irrigated. B – Definitions based on measured depths of application The ASCE (1978) “on-farm” efficiency definitions use depths of application. If the application depths are only average values found by dividing the total volume delivered to the system by the area there is no difference between these efficiencies and those based on volumes of water. If, however, the depths are actual values measured at specific – representative – locations in the field, these definitions can, after the measured depths have been processed, also take into account the uniformity of application. The definitions based on measured depth of field application have the disadvantage that they cannot be used for a whole system: it is impractical or impossible to measure the depth of application in all the fields of irrigation system.

C – Definitions based on other criteria, mainly related to yield These definitions are seldom used due to the fact that crop yield are influenced by many factors other than water supply alone (fertilizer application, treatment of plant diseases etc.). Crop yields are measure if agricultural production and not of water supply, although managers of irrigation water supply systems are sometimes judged by crop yields. Many ratios are presented as efficiencies and are called efficiencies although they are not efficiencies in the sense of a ratio of output to input, whereby the output (of some quantity) is a conversion of an input (of the same quantity). Examples are: a “deep percolation efficiency” (Hart et al, 1979), the ratio of “the volume of water required to fill the available rootzone water storage minus the deficit” to “the volume of water absorbed by the soil through infiltration” (the ratio, moreover, is high when the deep percolation is low); an “operation efficiency” (Schuurmans, 1989), as the ratio “effective volume of water supplied” to actually supplied volume of water”.

DEFINITIONS RELATED TO THE CONVEYANCE OF WATER The efficiencies related to the conveyance of water have not changed much since Israelsen (1932) defined the “water conveyance and delivery efficiency”. Jensen et al. (1967), Bos and Nugteren (1974), and ICID (1978) use the same definitions, except where Isrealsen mentions “water delivered to farms” (in the numerator), Jensen uses “water delivered by the conveyance system”, and ICID uses “water delivered to the distribution system”. The distribution system consists of tertiary units, each of which can be one farm only, or can incorporate many farms. The distribution system is usually under the control of farmers or groups of farmers. These efficiencies express how much water is delivered from the conveyance system to the distribution system. The not-delivered water includes operational spills, seepage, and evaporation. These definitions do not intend to cover the uniformity of the division of water over the various parts of the distribution system.

DEFINITIONS RELATED TO WATER STORAGE IN THE ROOTZONE These definitions have the problem that rootzone is not exactly defined. Its dept varies with the growing stage of the crop, the type of crop, the prevailing groundwater depth, etc. The term “water stored in the rootzone” is used by Hansen (1960) in the numerator of the “water storage efficiency”: the ratio of water stored in the rootzone during irrigation, over water needed in the rootzone prior to irrigation. This ration was introduced because of the inability of the Isrealsen’s “water application efficiency” to reflect conditions of under-irrigation. However, rootzone water storage efficiencies do not cover conditions of over-irrigation: if more water is applied than needed, the excess is not accounted for.

13



The three definitions of Hart et al. (1979) deal neither with the conveyance of water nor with crop growth. These definitions seem to originate from irrigation at field level because the area to which water is delivered is not explicitly defined. If that area is a field or a farm, the efficiencies do not, for example, cover operation spills in the system or seepage from conveyance channels.

DEFINITIONS RELATED TO WATER USED FOR PLANT GROWTH These definitions take into consideration: a) Volume of water used for plant growth, and b) Volume of water supplied.

Volume of water used for plant growth The difficult point in irrigation efficiency is to determine the amount of water used for plant growth. In the definitions, this amount is expressed in various ways: • “irrigation water transpired by the crops” (Isrealsen, 1932), • “normal consumptive use of water” (Hansen, 1960), • “useful water applied” (Hall, 1960), • “volume of water transpired by plants, plus volume of water evaporated from soil” (Jensen et al., 1967), • “consumptive use – effective rainfall” (Erie, 1968), • “beneficially applied depth of water” (ASCE, 1978), including applications for such purposes as salt leaching, frost projection, crop cooling, • “volume of water needed, and made available, for evapotranspiration by the crop to avoid undesirable water stress in the plants throughout the growing cycle” (Bos, 1980). The actually used volume of water for plant growth will always be an estimate and it can only be measured on an experimental scale. Crop transpiration and evaporation of water the soil are usually combined in evapotranspiration. The reason for the formulation of Bos is that the link between crop water use and “water storage in the rootzone” is not easily made. With this definitions, the problem of how to account for crop water use is shifted from rootzone moisture to evapotranspiration. A water application efficiency defined as water stored in soil, or in the rootzone, over water delivered to that rootzone is not useful is not related to evapotranspiration. And it is exactly this relationship that is made by Isrealsen’ consumptive use efficiency. The product of this consumptive use and the water application efficiency is the ratio between irrigation water transpired by the crop and irrigation water delivered to the farm. The time period considered is important too. The efficiency of one application at field level, measured with either volumes or depths, gives no information on an average application efficiency for a growing season, irrigation season, or for an area other than the measured field. Hall (1960) defined a “season application efficiency” to extend his “application efficiency” to the entire irrigation season. Israelsen and ICID explicitly mention “growth period” or “growing cycle” in their definitions.

Volume of water supplied When the numerator is concerned with water used for plant growth, the scope of the denominator determines which efficiency is evaluated. For example Isrealsen’s “water application efficiency” concerns “irrigation water delivered to the farm”, and Hall’s “application efficiency” concerns “gross volume of water delivered to the field”. Typically, the broadest scope of “volume of water supplied” is the scheme: Jensen’s “irrigation efficiency” deals with “total volume of water diverted, stored, or pumped for irrigation”, and ICID’s “project or overall efficiency” deals with “volume of water diverted or pumped from the river”.

RELATIVE WATER SUPPLY

14



Relative Water Supply (RWS), proposed by Levine (1982), is the ration of “water supply” to “the demand for water”. Defined as input over output, the RWS resembles the reciprocal of an irrigation efficiency, but there is a difference. Generally, the crop irrigation water requirements is arrived at by subtracting the effective precipitation to be expected from the crop water requirement (potential evapotranspiration), whereas in the definition of RWS, the effective precipitation is taken as a water delivery. The inclusion of seepage and percolation on the demand side (denominator) of the definition suggests that the RWS ratio has been especially formulated to evaluate rice cultivation. Levine (1982) uses rainfall in the numerator of the definition, whereas Weller et al. (1988) use effective rainfall.

THE ICID DEFINITIONS AND WATER SHORTAGE Water shortage usually implies that the supply of water is not enough for crop growth. The terminology used includes “under-irrigation” or “under-sipply”. In the ICID’s irrigation efficiencies, the crop irrigation water requirements is Vm – the volume of water needed, and made available for evapotranspiration by the crop to avoid undesirable water stress in the plants throughout the growing cycle. The starting point for Vm is the theoretical crop irrigation water requirement. However, the definition restricts itself to well-watered conditions. This conditions are presumed, and if this presumption were to be strictly adhered to, the ICID (crop related) definitions could be only used for planning and designing irrigation. When Vm is taken as potential evapotranspiration (ETpot), and Vc is the volume diverted or pumped, the value of ETpot/Vc can become greater than 100% under conditions of water shortage. Then, it is not longer an efficiency in the sense of a ratio of output to input. If, however, Vm is taken as actual evapotranspiration (lower than ETpot), values will be high, although never higher than 100%. Beside “conveyance efficiency”, “distribution efficiency”, “field application efficiency” and “overall or project efficiency” (see Appendix), the ICID also defined a “tertiary unit efficiency” and an “irrigation system efficiency”. They are, when the non-irrigation supplies are neglected, combinations of respectively the distribution and field application processes and the conveyance and distribution processes.

EFFICIENCY AND UNIFORMITY In irrigation, uniformity is used to express the variation in depths of application or supplied volumes. Christiansen (1942) defined uniformity coefficient CU for the comparison of sprinkler patterns. ICID (1978) extended the use of this coefficient to cover infiltration water too. Efficiency and uniformity can be described at different levels in an irrigation system: for a field, for a tertiary unit, and for a scheme. Within a river basin, the division of flow over successive irrigated areas can be important, especially if the river flows through several countries or states.

Efficiency and uniformity at field level Most efficiency definitions concerns volumes only, because these are relatively easy to measure, but, for a field, the total volume delivered is not an absolute measure of how much water each part of the field receives. For sprinkler-irrigated fields, uniformity has been frequently investigated with cans on the field catching the water from the sprinkler. The uniformity of the actual field application is then assumed to be equal to the application depths measured in the cans. For other methods of field application (e.g. basin or furrow), uniformity is more difficult to measure and is then usually estimated. The actual depth of water applied to the soil can be found by measuring the soil moisture content at several – representative – locations in the field before and after irrigation. Because of the practical limitations of these measurements, the depth of water is commonly taken as a function of the opportunity time for water to infiltrate. For the latter assumption to give reliable results, the soil within a field must be reasonable homogeneous.

15



Till and Bos (1985) assumed a normal distribution of applied water depths to parts of an irrigated field. If the volume of irrigation water equals the volume of water needed for evapotranspiration (field application efficiency is equal to 100%), 50% of the area is under-irrigated and the other 50% is overirrigated. And about 84% of the area receives less water than the mean plus the standard deviation. If the uniformity of application is not changed and the volume supplied to the field is increased, the efficiency will decrease, but the under-irrigated area of the field will also decrease. The assumption of a normally distributed soil moisture over the field implies that, to give 75% of the area an average supply, the supply should be raised to mean plus 0.67 times the standard deviation. By increasing the volume of water delivered to the field, the field application efficiency decreases, but an increasing part of the field receives more than the mean water supply.

Efficiency and uniformity at tertiary unit level Within a tertiary unit, water is delivered to fields. The deliveries to these fields can be regarded as observations in a population to which the same statistics can be applied as for fields, provided there are enough data.

Efficiency and uniformity at scheme level Within a scheme, the uniformity of flow division over the main and secondary channels and tertiary units follows from measured flows. Efficiencies and uniformities can be calculated from the same data. Wolters et al. (1987) reported on the uniformity of the division of flow over the irrigated lands of the Fayoum Governorate, in Egypt, with gross irrigated area of 151, 865 ha. The division of flow was measured from the main intake down to where the area is subdivided into five parts which are served by a secondary channel (4,155 ha). For each sub-area, the uniformity was expressed by the ratio of actually delivered flow over intended flow. The intended flow for the Fayoum is based on proportional flow division over the irrigable area, for which the system was designed. The results were that: a) Areas with a disproportional supply are easily spotted; b) The operation needs improvement throughout the system; c) The most important improvement in the investigated secondary channel would be to provide it with an adequate quantity of irrigation water. Making any other improvements in the area would be of little use as long as it remains underirrigated. The timeliness specifications of water deliveries might be usefull in schemes where the objective, the design, and the operation rules are very strict as to the timing of the water delivery when water delivered before or after pre-defined period of time is considered as lost.

INCREASES IN THE EFFICIENCY OF IRRIGATION WATER USE When an increase in the efficiency of irrigation water use is being considered, the following questions arise: Is an increased efficiency needed? Is an increased efficiency possible? It is realistic to expect an increased efficiency from the proposed measures?

The “need for increases” in the efficiency of irrigation water use The need for increases in the efficiency of water use depends on the balance between the following positive and negative effects of increased efficiencies. The positive effects of increased efficiency of irrigation water use are: a larger area can be irrigated with the same volume of water;

16



the competition between water users can be reduced; the effect of a water shortage will be less severe; water can be kept in storage for the current (or another) season; groundwater levels will be lower, which can lead to lower investments costs for the control of waterlogging and salinity; there will be less flooding; better use will be made of fertilizers and pesticides and there will be less contamination of groundwater and less leaching of minerals; health hazards can be reduced; energy can be saved; there will be fewer irrecoverable losses; instream flows, after withdrawals, will be larger, thereby benefitting aquatic life, recreation, and water quality.

The negative effects of increased efficiency of irrigation water use are: soil salinity can increase because of reduced leaching; wetlands and other wildlife habitats may cease to exist; groundwater levels will fall and aquifers will receive less recharge; water retention in upstream river basin areas will be reduced; there will be a need for more accurate operation and monitoring; and a need for a more expensive infrastructure. When considering measures that could lead to an increase in irrigation efficiencies, the possible effects of the proposed measures on the factors in this list have to be investigated. Many factors in the list of positive and negative effects of increased irrigation efficiencies have a relationship with the water quality of the components of the water balance, which means that water quality investigations should complement efficiency investigations.

The “possibilities” of increasing the efficiency of irrigation water use The possibilities of increasing efficiencies are influenced by what is technically possible and, moreover, by the general rule of the economic feasibility of improving the system. The benefits from improvements should outweigh the costs. Investigating the possibility of increasing the efficiency of irrigation water use implies establishing the components of the water balance of the system. This will often reveal whether, how, and where the efficiency can be increased. If such a comparison reveals that improvements are possible, targets values for one or more efficiencies have to be set and the measures needed to reach these target values have to be taken. Finally, the results of the actions have to be assessed.

Effect of proposed measures to increase irrigation efficiencies Several issues were recommended as proper tools for the improvement of irrigation efficiencies. Among those issues water charges, lining canals, improvement of irrigation structures and modernization of irrigation systems, … etc. The analysis of the literature and comparing the results obtained on the irrigation efficiencies by the implementation of the previous mentioned measures, the contraries in the data found by the researchers demonstrates that it is not always realistic to expert increased efficiencies from measures aimed at that goal. This could be explained on the ground that irrigation efficiencies are basically ratios of volumes in the water balance of an irrigation scheme. The studies carried concentrated mainly on the relationships between the components of the waterbalance of irrigation systems and the factors that might influence these relationships. However, there are other factors that are related to irrigation efficiencies such as the management of crops, the socio-economic and legal environments of irrigation systems, the capacity building in the irrigation sector and the quality of water.

17



Those factors, together with those related to the components of the water balance of irrigation systems should be fully considered when final decisions are being made on water use in the agricultural sector.

CONCLUSIONS AND RECOMMENDATIONS

Irrigated agriculture is by far the greater user of water. The limits to the availability of water and land for irrigated agriculture necessitate the careful use of these resources. Nevertheless, an increased water can have negative as well as positive effects. Low irrigation efficiencies are mainly attributed to the following common problems: 1) Lack of measurement devices 2) Lack of data on water flows and cropping patterns, and 3) Low irrigation efficiencies because of seepage or excessive water applications

Evaluation of development irrigation “projects” identify additional problems (e.g. inadequate project formulation and discontinuity after donor involvement). Lack of data is a serious constraint to improving irrigation. There is a continuous need for data on water flows, crops, state of system repair, and agricultural practices. Monitoring systems should be an integrated part of irrigation system management. Many irrigation schemes, especially those in the formal “modern” sector, do not live up to expectations because of over-optimistic assumptions on: 1) the time needed for the design, construction, and realization of benefits, leading to higher than expected project costs, and benefits being realized later in time; 2) irrigation efficiencies, leading to, for instance, a conveyance system that delivers less water than planned, and a lower than expected performance in economic terms. There are considerable seasonal variations in efficiencies whereas, usually, the values are only high, and only to be high, for a short period of one or two months in the season. That period is the critical part of the season for design and operation. Further research is needed into the implications of this phenomenon for the design of irrigation systems. When considering an increased irrigation efficiency in a certain system, it is more useful to regard that system as unique, and to use the list of positive and negative effects of increased irrigation efficiencies on which to base a decision, rather than to rely on general relationships between characteristics and efficiencies. Many factors in the list of positive and negative effects of increased irrigation efficiencies bear a relationship to the water quality of the components of the water balance, which means that water quality investigations should complement efficiency investigations.

REFERENCES American Society of Civil Engineers/ASCE. 1973. Consumptive Use of water and Irrigation Water Requirements. The Technical Committee on Irrigation Water Requirements of the Irrigation and Drainage Division of the ASCE. American Society of Civil Engineers/ASCE. 1978. Describing Irrigation Efficiency and Uniformity. The On-farm Irrigation Committee of the Irrigation and Drainage Division of the ASCE. Proceedings of the ASCE 104, IR 1:35-41. Bos, M.G. 1980. Irrigation Efficiencies at Crop Production Level. ICID Bulletin 29.2: 18-26. New Delhi. Bos, M.G. and W. Wolters. 1990. Water Charges and Irrigation Efficiencies. Irrigation and Drainage Systems 4: 267-278. Kluwer Academic Publishers. Dordrecht. The Netherlands. Erie, L.J. 1968. Management : A Key to Irrigation Efficiency. Proceedings of the ASCE 94, IR3:285293. Greenland, D.J.and S.I. Bhuiyan. 1980. Rice Research Strategies in Selected Areas: Environment Management and Utilization. Special Internationl Symposium on Rice Research Strategies for the Future, 21-25 April 1980. IRRI, Manila, Philippines. Cited in: Water Management Study at Kaudalla Irrigation Scheme.

18



Hall, W.A. 1960. Performance Parameters of Irrigation Systems. Transactions of the American Society of Agricultural Engineers/ASAE 3(1): 75-76, 81. Hamdy, A. and C. Lacirignola. 1999. Mediterranean Water Resources: Major Challenges towards 21st Century. Mediterranean Agronomic Institute of Bari, Italy. Hansen, V.E. 1960. New Concepts in Irrigation Efficiency. Transactions of the American Society of Agricultural Engineers/ASAE 1960: 55-64. Hart, W.E., Peri, G. and Skogerboe. 1979. Irrigation Performance: An Evaluation. Journal of the Irrigation and Drainage Division of the ASCE. 105, IR3: 275-288. International Commission on Irrigation and Drainage/ICID. 1978. Standards for the Calculation of Irrigation Efficiencie. ICID Bulletin 27. 1:91-101. New Delhi. Israelsen, O.W. 1932. (1st Edition). Irrigation Principles and Practices. John Wiley, New York. Jensen, M.E., Swarner, L.R. and J.T. Phelan. 1967. Improving Irrigation Efficiencies. In: Irrigation of Agricultural Lands. Agronomy Series: 11, American Society of Agronomy, Wisconsin, USA. Keller, J. 1986. Irrigation System Management. In: Irrigation Management in Developing Countries. K.C. Nobe and Sampath, R.K. (Editors). Studies in Water Policy and Management, 8: 329-352. Westview press. Levine, G. 1982. Relative Water Supply: An Explanatory Variable. Technical Note 1. The Determinants of Developing Country Irrigation Problems Project. Cornell and Rutgers University, Ithaca, N.Y. Schuurmans, W. 1989. Impact of Unsteady Flow on Irrigation Water Distribution. In, Irrigation: Theory and Practice. J.R. Rydzewski and C.F. Ward (Editors), Pentech Press, London. Weller, J.A.; Payawal, E.B. and Salandanan, S.. 1988. Performance Assessment of the Porac River Irrigation System. Asian Regional Symposium on the Modernisation and Rehabilitation of Irrigation and Drainage Schemes (held at the Development Academy of The Philippines, 13-15 Feb. 1989). ODU/Hydraulics Research Ltd. Wallingford, UK. Wolters, W. , Nadi Selim Ghobrial and Bos, M.G.. 1987a. Division of Irrigation Water in The Fayoum. Egypt. Irrigation and Drainage Systems. 1:159-172. Kluwer Academic Publishers. Dordrecht. The Netherlands.

19


20




FUTURE OPTIONS AND RESEARCH NEEDS OF WATER USES FOR SUSTAINABLE AGRICULTURE

Muhammad R. Shatanawi Professor of Water Resources and Irrigation, University of Jordan, Amman, Jordan, Email: [email protected] SUMMARY - The increased demand for other uses coupled with recurrent drought and climatic changes in countries of limited water resources, is producing unprecedented pressure for reducing the share of fresh water used in irrigation. Many countries in the Mediterranean region, give priority of water allocation to the domestic sector followed by tourism and industry, and what is left is allocated to agriculture. At the same time that agriculture is asked to give water to other uses, the increasing population demand requires increase in food production. This creates a conflict that should be resolved and should be alleviated by examining different options of water uses for sustainable agriculture. Increasing the efficient use of water is a key non-structural approach to water resources management. The agriculture water use efficiency and water productivity is very important as they are largely inefficient in so many countries due to poor distribution systems and excess irrigation. The over all global average agricultural water use efficiency in the region is in the order 40%. This paper analyzes the water situation in the region and shows how agriculture will be affected by water shortages and giving priority of water uses to other sectors. Water planners and decision makers as well as researchers are faced with different challenges including; resources, economical, environmental and institutional. There are some options that can be used to sustain agriculture in the region considering the above constraints and challenges. They include: improving water use efficiency, reducing crop consumption of water, irrigation with reclaimed water, practicing deficit irrigation and irrigation with desalinated water. Lessons learned from Jordan and other countries will be illustrated in this presentation. The research vision for the next 25 years could include the following set of actions: have more efficient use and allocation for water use in irrigation; improved water productivity by introduction of new management measures such as deficit irrigation; and the introduction of high yielding low water demanding varieties. Desalination of brackish and sea water can offer limitless fresh water that can be used for agriculture. Therefore, there is a need to find cheap methods of desalination such as the use of solar and renewable energy. As reclaimed water of urban wastewater is becoming a new source, it is necessary to maintain efficient and sustainable agriculture production while using them. Therefore, researchers should consider these needs in setting up their research priorities. Key words: irrigation, water saving, research, education, sustainability.

INTRODUCTION Many countries in the Mediterranean region are located in the arid to semi arid regions of the world that are classified with limited water resources and increasing water scarcity. Managing these resources for sustainability will become increasing complex and difficult in the future as climate changes increases the frequency and intensity of drought and water shortages. The decline in the available water supplies as a result of the above biophysical factors as well as other geopolitical factors, rapid expansion in population, urbanization and economic development will result in depletion of the exploitable water resources. These conditions will produce an unprecedented pressure on the share of fresh water for irrigating the agricultural sector and will force many countries to reform their water allocation policy by giving priorities to the domestic and industrial water demand. As such, irrigated agriculture will be the most effect sector claiming what is left and using the reclaimed wastewater. In many cases such as Jordan, agriculture is asked to give its share to other uses in spite of increasing demand for food production. This situation creates a conflict that should be resolved and could be alleviated by adopting different options of water uses for sustainable agriculture. Increasing the efficient use of water in agriculture would be a prime option as it is considered a key non-structural approach to water resources management.

21



The issues of agricultural water use efficiency and water productivity are very important as they are largely inefficient in many countries due to many factors such as poor distribution system, excess irrigation and lack of proper irrigation water management. Therefore, improving water uses efficiency and productivity on sustainable basis is an enormous challenge for the water stress and scare countries of the Mediterranean. They should set forth strategies and plan to design and carry out programs that increase water use efficiency by farmers while increasing farm incomes. This will require a combination of technology, education and extension services coupled with research programs and effective policy framework in order to reflect the real opportunity cost of water. The focus should be concentrated on crop selection for better water use efficiency and productivity with improvement of marketing performance to add value and increase the return on agricultural investment in the irrigated sector. The problem is not limited to countries of the south Mediterranean but it has reached other humid high rainfall countries like France, Spain, Italy and Greece obliging them to impose temporary restrictions. The agricultural productivity and competitiveness in the whole region is adversely affected by water scarcity and inefficient use of water. The average water use efficiency for the sector in general ranges from 40-45% (Hamdy 2005 and Osman 2006). Efficiencies in the use of water for irrigation consists of various elements taking into account losses during storage, conveyance and application to irrigated area while optimal water productivity can take into consideration, proper water allocation and scheduling, selection of high value crop, timing of irrigation and other field practices. These practices and elements have been discussed in many articles but this paper concentrate on the options of research, education and information to improve irrigation water efficiency and its productivity.

CHALLENGING FACING RESEARCH Water scarcity is the single most important resource management challenge in the region. Inspite of that, most countries do not treat their water as a scarce resource. Before exploring different options in research, education and information, it is necessary to discuss the issues and challenges facing them. Researchers and decision makers should be aware of the resources and management challenges as well as socio-economic and environmental issues and institutional setup.

Resources Challenges The water shortage of the region has been traditionally addressed by increasing supply of water. The most common approach was to extend exploration and make massive investment in water resources development. Over the years, most of water resources have been almost developed, so the rate of investments is currently shrinking. Expanding the supply is unlikely to make dramatic changes in the future because their development will technically be unfeasible and economically expensive. Therefore, an essential part of any resources program must focus on managing water demand. Most of the water saving will come from agriculture by improving water use efficiency.

Economic and Social Challenges The impact of reduced share of water to the agricultural sector will directly affect the issues of food self reliance. This may cause much economical and social reform that can create additional challenges, thus encourage decision makers, researchers and farmers for the efficient use of water and improved productivity.

Environmental Challenges With improved water use efficiency, the extra amount of water applied for leaching can be reduced thus increasing soil salinity. Also, the agricultural sector will relay on a great percentage of its water supply on reclaimed wastewater and low quality water. These conditions will create environmental problems that should be addressed in research and management. Management Challenges

22



Supply and demand management, through the more economically efficient supply and use of water and through changes in production practices as well as reduction of loses, is a vital issue in resource management strategies and planning. Legislation an economic approach to water demand management is properly the most important instrument that water resources managers has to develop and use. They should work together with researcher in order to come up with an optimum approach to water saving.

Institutional Challenges For any water management program to be successful and effective, there should be a close link between researchers and decision makers, between extension workers and researchers and between farmers or farmer’s associations and extension workers. For these reasons, an institutional set up must be created where the four groups are integrated; researchers focus on on-farm research and extension agents focuses on adoption process. In developing countries, irrigation extension services dose not exist thus establishing extension agencies for irrigation and water management imposes a great challenge.

RESEARCH OPTIONS Universities and research institution are asked to participate in the improved management of irrigation water and they should demonstrate that they are capable of developing sustainable and integrated research programs. On the other hand, public sectors (represented by ministries of agriculture and water and irrigation) and the private sector should give the universities the leading role in research to gain experience in improving agricultural water use efficiency and productivity. Research results in many developing countries demonstrates that agricultural demand for water can be reduced without decreasing the total irrigated areas or the value and the quality of the agriculture production. Proper water usages along with crop selection might actually increase agriculture’s contribution to the economy and at the same time decrease its water usage. However all research efforts and results are considered worthless unless they can reach and be adopted by the end users and farmers. Therefore, universities and research institutions should have extension units linked to the public sector extension services that are capable of disseminating the results to the farmers. The later are in need of information which are site specific and time dependant because many of them have realized that optimal agricultural production requires good water management. The research on supply management are limited to improving the quantity of supply by using various management tools through modeling and enhancing water supply such as water harvesting and recharge of groundwater. However, other minor options such as cloud seeding, fog harvesting and water desalination should not be ignored. On the demand side, the research challenges and options are attractive because water saving and increasing the value per cubic meter of water are highly achievable. Any improvement in the irrigation efficiency means expanding irrigated area by the same percentage. Below are some options that might be adopted by researchers in order to orient their research program in water demand management aiming at increasing water use efficiency and improving its productivity.

Crop Water Requirements Previous calculations of crop evapotranspiration using different formulae and procedures have always overestimated the actual consumptive use of different crops. Due to lack of data and research results, these methods were applied to be the basis for the design of many existing projects. In these projects, the system capacity and the delivery schedule allow for over irrigation most of the time. The FAO paper 56 has suggested a new procedure in calculating ET based on the revision of previous methods. Although, FAO procedure is general but the results of field trials have shown that about 20% of ET have been overestimated by previous methods. For example, the average peak ET for citrus in the northern part of the Jordan Valley was calculated as 5.4 mm/day using FAO methods compared to previous calculation of 6.5 mm/day according to Blaney-Criddle method. The results of

23



recent research (Shatanawi et al 2006) and the farmer’s practices in the same area have shown that the average ET has drop to about 4.8 mm/day without any impact on the produce quality and quantity. On the other hand, crop requirement for near maximum yield is determined by plant physiology, there are associated management factors manipulating the microclimate environment that can provide some advantages. For example, the irrigation requirement for open field vegetables may be twice or triple that for crops grown in plastic houses. A complementary approach is to select planting times and growing seasons that minimize the atmospheric demand for water consumption. Another possible action aimed at reducing ET is to change cropping patterns in favor of high value crops indented to the export that have relatively smaller water requirement. With the availability of new technologies like real time automatic weather stations and modern devices like the Eddy correlation, it would be possible to determine the exact amount of daily ET. The use of remote sensing data using satellite images coupled with ground truthing, ET and crop coefficient can be determined at district and regional level. The EU has supported a research project (STRP) entitled “Improved Management Tools for Water-Limited Irrigation: Combining ground and satellite information through models, (IRRIMED)” with the participation of 6 Mediterranean countries under FP6. The aim of this project is the establishment of tools to support efficient management for water used for irrigation as well as to test scenarios for long term sustainable policies. Accurate knowledge of water demand and use by irrigated agriculture is the key to an effective water management strategy. The general scientific objective is the assessment of temporal and spatial variability of water consumption of irrigated agriculture under limited water resources condition. Intensive measurement campaigns with eddy correlation equipment will allow combining ground and satellite measurements into models, to ultimately produce simple methods to assess evapotranspiration (ET) over large areas. The accurate assessment of actual ET over selected crop during the growing season, will allow validating models and to update the crop calendar and crop water requirements. Also, remote sensing of crop extension and evolution during the growing season will help to measure the actual acreages of the different crops. Refining existing methods for simple ET estimation will be used to deriving ET maps from satellite data. This line of research will continuously update information that can be revised annually based on agro-climatic conditions.

Precision Irrigation The issue of irrigation scheduling (in when to irrigate and how much water to apply) is a matter of delivery schedule and farmer’s decision. With the availability of soil moisture sensors and stem water potential devices, it is possible to irrigate at the exact time when water is needed by the plant. These devices can be installed in the soil at two depths or can measure the tension in the leaves or fresh stems may be connected to electronic control panel that can tell the farmers the need to irrigate. Research on who to integrate these modern sensors to the irrigation systems is an option that should be exploited in the future. Precision irrigation is not limited to irrigation scheduling but can be extended to incorporating them into the design of various irrigation systems. In surface irrigation, laser land leveling can insure good uniformity distribution and improved irrigation efficiency. In pressurized irrigation system, the systems can be operated through automatic control panel. Also, leaks and uneven distribution of irrigation water along the lateral and subunits can be detected easily. The introduction of such technology will certainly improve irrigation efficiency and water productivity as well as reducing water losses.

Use of Reclaimed Water Reclaimed wastewater has become a significant source of the water resources in many countries of the Mediterranean like Jordan were it contribution to the irrigation sector has reached about 15% in 2005 and will reach 40% by the year 2020. Research in this area is scattered and is limited to treating this source as low quality water. Research in this regards should be extended to include long-term impact of using the reclaimed water on soil and the environment, changes in on-farm practices

24



especially those related water use efficiency, adopting more high value crops and social and economical impact of the reuse of treated effluent. The research should focus on finding appropriate tools to help farmers to overcome problems that will face them and the sector and to develop new attitude and behavioral patterns.

Desalination of Brackish Water The area of the Mediterranean, especially the south has a significant reserve of saline water that is considered a potential resource in the future. It is possible to irrigate certain crops with this kind of water provided that the soil exhibits good drainage conditions while applying extra water for leaching purposes. Research results have shown that there are few success cases where the production is economically feasible. The reduction in yield of up to 50% may not justify the investment provided for the irrigation and drainage systems as well cost of pumping and delivery. There are few cases where it is possible to irrigate fodder crops in sandy soils with good natural drainage system. An alternative to that would be to desalinize this water in which the cost is justified. Experience from the Jordan Valley has shown that the cost of desalination of saline water (2000 to 5000 ppm) can reach as low as 0.2 $/m3 using medium size reverse osmosis plants with a capacity of 40 to 50 m3/hr. Irrigation with the blended water of 500 ppm has increased the yield of high consumptive crops more than twice. Banana yield has increased from 20 ton/ha to 40 ton/ha with good quality produce while the irrigation water requirements have been reduced from 2500 mm to 1800 mm. The investment can be farther justified if this water is used to irrigate seedling nurseries and cash crops like strawberry. Therefore, new research ideas should be explored on conducting comparative studies, reducing the cost of desalination, and evaluating the environmental and economical impact.

Deficit Irrigation Deficit irrigation means applying less water than cumulative ET, thereby allowing roots to utilize stored soil water in the winter or pre-season irrigation. Therefore, the irrigation water requirements in early irrigation in the spring season can be less than that indicated by ET calculation. Also, deficit irrigation may be regulated for the rest of the season avoiding critical periods. Such management practices results in water saving in irrigation without affecting or reducing yield. There are two types of deficit irrigation; sustained and regulated. In sustained deficit irrigation, the irrigation is reduced during the whole season while regulated deficit irrigation starts with normal irrigation and then gradually irrigation is reduced. Regulated deficit irrigation is an irrigation strategy based on limiting non beneficial water losses by reducing the amount of water for crop during non-critical phonological stages. The deficit irrigation is controlled during times when the adverse effects on productivity are minimized. There are a lot of research activities on DI that are going on field crops and vegetables. Field demonstration conducted by Shatanawi and the French agriculture Mission in Jordan (1996) showed that 40% reduction in water consumptive from the farmer's practices did not affect the yield. Observation and communication with some farmers concluded that reducing water application by 3040% during drought years did not reduce yield economically. However, research on fruit trees is limited and should be evaluated in estimating the actual ET under deficit irrigation in order to maximize the water unit productivity. Such research should include applying different level of irrigation while measuring the soil moisture content and leave water potential. It is worth mentioning at this point that EU has supported a research project on deficit irrigation entitled “Deficit Irrigation for Mediterranean Agricultural Systems (DIMAS)”. The objective of this project is to evaluate the concept of deficit irrigation (DI) as a means of reducing irrigation water use while maintaining or increasing farmers profits. The DI concept will be the subject of multidisciplinary research at different scales, geographic locations, and with different perennial and annual crops. The objective will be to develop a workable, comprehensive set of irrigation (DI) strategies that can be disseminated quickly among the various agricultural systems of the Mediterranean Region. The project addresses directly the first topic of the FP6- INCO-2002-B1.2 specific measure, ‘research on sustainable irrigation, including deficit irrigation’. Eleven partners from seven different countries (Greece, Italy, Jordan, Morocco, Spain, Tunisia and Turkey), including research and water association institutions will work for three years on the project. Their main activities will be: a) the

25



development of a general summary model of crop yield as a function of water supply, b) the validation of the model for the main irrigated annual (wheat, sunflower, cotton,) and perennial crops (olive, pistachio, citrus), using common research protocols, c) a survey on physical, socio-economic and cultural conditions for each crop and irrigated area, and d) scaling up by combining the yield model with economic optimization modules that will generate optimum DI strategies compatible with the specific socio-economic characteristics of each area under study. The results of the project will provide recommendations for reducing irrigation water use while ensuring the sustainability of irrigated agricultural systems in the Mediterranean basin. Feedback with project end-users will take place via participation of farmers associations and irrigation water agencies who will contribute their expertise in managing water scarcity, thus ensuring that all relevant issues are addressed.

Irrigation Techniques The efficiency of the on-farm water use and the water productivity can be increased with improved irrigation techniques. Innovation in this area should be pursued between researchers and the irrigation industry. Although micro irrigation is known to be high efficient irrigation system but experience shows that if the system is not well design and not operated probably, the efficiency can be as low as 50%. In addition to that, research on irrigation accessories such as filters, pressure regulators should be incorporated into the system design and management. In this area, the use of sand filter with proper sand gradation automatic filter systems, emitters, acid and chlorine injection should be tested and experimented on crops highly sensitive to water stress. For other irrigation systems like surface irrigation and sprinkler irrigation, there is a high potential for innovation such as molding of surface irrigation, irrigation cut back and surge flow irrigation. The design of all irrigation system should provide flexibility and simplicity required for successful operation under different soil variables and topographic variation. Research should be oriented toward proper and careful selection of pumps, pipes and on-farm sprinkler equipment in order to sustain high uniformity at a specified application rate. The research in irrigation system should also concentrate on the energy aspect by introducing and testing low pressure micro and sprinkler irrigation in order to reduce the cost of operation and maintenance. Technology so far, has produced sprinkler system of low energy precision application (LEPA) and low pressure compensating emitters that can give high uniform application rate and efficient irrigation. A probably designed and managed system incorporating all of the above technology can have efficiency as high as 98%. Research should be further pursued to explore along with the industry new technology that can save in water and produce uniform irrigation.

EDUCATION AND TRAINING OPTIONS The overarching goal in promoting the efficient and the effective management of water is the investment in human resources development. The venue in this regards has many options ranging from University diploma and higher graduate research to in-job training and tailor training programs. Most universities in the region offer B.Sc degrees and M.Sc Degrees in water and irrigation and few Ph.D programs. However, capacity building is not only limited to new graduate and extension agents but should be extended to the decision makers, legislators and stakeholders. Training on the state of art on water management can help in establishing water resources management agencies and creation of irrigation advisory units. Experience from Jordan has shown that farmers receiving training from the University of Jordan in irrigation scheduling and management have reported improvement in irrigation efficiency by 30% (Shatanawi, 2004). The link between farmers and educators as well as researcher should be the responsibility of the extension services who will convey the result of the research to the farmers. Therefore training of extension personal will facilitate the flow information to the stakeholder groups such as water user associations. They should be also trained to carry on activities such as managing demonstration sites, organizing working session with farmers groups, hosting educational opportunities for professional, disseminating information and encouraging individuals to participate in improving water use efficiency and productivity. The results of the demonstration programs that have been undertaken on pilot basis, aiming at improving on-farm water use efficiency can be disseminated by extension services.

26



The irrigation sector in many countries of the Mediterranean needs high quality extension services with innovative ideas to translate policies into action plans. The extension services should not be limited to the public sector but the various irrigation companies and private agricultural enterprises who are using the state of art technology must all of them take action in training extension services as well as farmers.

INFORMATION AND TECHNOLOGY TRANSFER Information is considered a key element in the day to day management, decision making and undertaken current and future water balance and flow. The flow and availability of information to researchers, operators and farmers is also of prime important to determine irrigation scheduling by farmers and water delivery schedule by farmers. There are different levels of information required. For operators of the irrigation system at the project, it is necessary that information regarding flows of surface water and reclaimed wastewater, and stocks of groundwater be available at the present and future. This kind of information coupled the projected cropping pattern and agro-climatic data can help in better water allocation and optimization, thus improving water use efficiency and productivity. The role of the irrigation authority should not be limited to operation at the sub-unit level, which is responsibility of the farmers group, but they should work at higher level of planning and management at the project level. The Water Management and Information System (example from the Jordan Valley) can provide such a model for irrigation districts. Research and technology transfer institution can play an important role in utilizing the field and climatic data, information of crop pattern and meteorological data into daily out put that can be used by farmers. An example on that is the Irrigation Information System established by the National Center for Research and Technology Transfer (NCARTT) where the whole country is covered by a network of automatic weather stations. Daily ET is calculated for different location from the data transmitted from these stations using the FAO method of paper 56. On the other hand, expansion of pilot projects and demonstration farms will help in obtaining accurate information on the impact of improved irrigation practices on water use and economic return of unit volume of water. They will serve as a venue for comprehensive extension and technology transfer programs. The demonstration sites of concentrated water practices could be divided to cover the different climatic zone of any region. Demonstration activities can include the following: 1. Demonstration on micro irrigation operation and maintenance. 2. Demonstration of deficit irrigation strategies. 3. Demonstration of winter soil moisture conservation technique. 4. Demonstration on optimum irrigation practices. Each country should have an entity responsible to coordinate national agriculture research on water management and technology transfer activities. The entity can act as a focal point for capitalizing and facilitating information and technology transfer including social and economic marking. The focal points should work among a network of partners such as universities, irrigation authorities, extension services and other environmental societies. The experience of the Scientific Irrigation Scheduling Services (SISS) from California State and Washington State can be technically adopted by some countries. SISS are enterprises that market full service or self service SIS products such as a low cost tensiometers, water marks and portable soil water sensors (Neutron Probe). Also, the technology of Aqua-Card have been widely used in Italy could be adopted by other Mediterranean countries.

CONCLUSIONS Many countries in the region are characterized by limited water supplies while providing additional resources will be an expensive option. Therefore, efforts in the optimal management of water resources should concentrate on the demand side management. As the agricultural sector is consuming the bulk of water supply, good management of irrigation water can be translated into significant amount of saving in the water resources. In addition, the agriculture sector will be most

27



affected by water shortage and would be asked to give water to other uses such as the domestic and the industrial sector. Therefore, increasing the efficient use of water in the agricultural sector would be an overarching goal in changing certain policies or adopting new ones with the objectives of improving on-farm water management and maximize agricultural return per unit of water. Optimization of water use at the farm level involves getting the maximum value output for minimum amount of water. There are several activities that can be carried out to achieve this goal ranging from field management practices by the farmers to water management approach by decision makers. However, these activities can attain high degree of success if it is supported by good research programs and extension services where both need the availability and flow of information. Research program and activities should be oriented towards the adoption of improved technologies, management practices and policies which contribute to the national development as well as addressing the problems and needs of the irrigated agriculture. Research projects must be planned and formulated in such a way that these activities are directed towards effectively conserving and managing the national resources utilized in agricultural development so it would not contribute to the loss of and/or deterioration of these resources. In order to cope with future challenges facing irrigated agriculture, research and technology transfer, system must be capable of responding to the needs of an increasingly more sophisticated and more competitive agricultural sector. Research projects will be demand driven and should be applied and/or adaptive in nature. In order to ensure the most efficient use of the limited resources available for research and technology transfer activities, project should be multidisciplinary in their approach. The role of institutions and agencies engaged in agricultural research must be clearly defined, along with the areas of research and technology transfer they are capable of implementing and carrying out. Agricultural research should take into consideration population increase, changing consumer habits, limited water resources and the need to increase water use efficiency and improve production practices to reduce production costs and increase the competitiveness of the production. The goal of research and technology transfer in the area of irrigated agricultural is to achieve an efficient and economic production, diversification and marketing crops for domestic and export market. In order to achieve the above goals, research strategy and technology transfer activities should be directed towards the following objectives: 2. Improve water use efficiency 3. Utilize non-traditional water resources 4. Intensify cropping systems 5. Improve and test new management practices

REFERENCES Al-Jayyousi, O. and M. Shatanawi. (1995).”Analysis of Future Water Policies in Jordan. “Water Resources Development, Vol. 11, No. 3, pp. 315-330. Allen, R., A. Clemmens, C. Burt, K. Solomon, and T. O'Halloran. (2005). “Prediction Accuracy for Project-Wide Evapotranspiration Using Crop Coefficients and Reference ET.” J. Irrig. Drain. Eng. 1(24), 24- 36. ESCWA, (2004). The Optimization of Water Resources Management in the West Asia Countries, Economic and Social Commission for Western Asia, UN, New York. Fereres, E., Goldhamer, D. A., and Parsons, L. R. (2003). “Irrigation Water Management of Horticultural Crops.” HortScience. 38(5), 1036-1042. Fereres, E. (2005). “Managing Deficit Irrigation: from the Crop to the District”. a paper presented at the 4th WASAMED Workshop on Water Use Efficiency and Productivity, University of Jordan, Amman, Jordan, 1-4 October, 2005. Hamdy, A. (2001). “Sectorial Water Use Conflict and Water Saving Challenges” in the proceeding of the workshop entitled “Water Saving and Increasing Water Productivity; Challenges and Options. March 10-23, 2001. Amman, Jordan.

28



Hamdy, A. (2005). “Water use Efficiency in Irrigated Agriculture: An Analytical Review” a paper presented at the 4th WASAMED Workshop on Water Use Efficiency and Productivity, University of Jordan, Amman, Jordan, 1-4 October, 2005. Osman, M.E. (2005). “Reflection on the Status of on-Farm Water Use Efficiency in Selected Countries of West Asia, a paper presented at the 4th WASAMED Workshop on Water Use Efficiency and Productivity, University of Jordan, Amman, Jordan, 1-4 October, 2005. Shatanawi, M. R. (1986). “Efficiency of the Jordan Valley Irrigation System.” DIRASAT (Agricultural Sciences). 13 (5), 121-142. Shatanawi, M. R. et al., (1994). "Irrigation Management and Water Quality in the Central Jordan Valley", A Baseline Report Prepared for the USAID Mission to Jordan, by the Irrigation Support Project for Asia and the Near East (ISPAN) and the Water and Environment Research and Study Center, University of Jordan, Amman, Jordan. Shatanawi, M. (2002). Policy Analysis of Water, Food Security and Agriculture Policies in Jordan, Review paper submitted to the World Bank. Shatanawi, M. (2004). Improved Management Tools in Irrigation for the Jordan Valley, an internal report submitted to the Deanship of Scientific Research, University of Jordan, Amman, Jordan. Shatanawi, M., M. Duqqah and S. Naber. (2006). “Agriculture and Irrigation Water Policies for Water Conservation in Jordan”. A paper presented in the 5th WASAMED Workshop on Harmonization and Integration of Water Saving Options: Convention and Promotion of Water Saving Policies and Guidelines. Malta, 3-7 May, 2006. World Bank. (2002). Water Sector Review Update, Main Report, The Hashemite Kingdom of Jordan, World Bank, Washington D.C.

29


30




RELATING WATER PRODUCTIVITY AND CROP EVAPOTRANSPIRATION

Luis S. Pereira* *Center for Agricultural Engineering Research, Institute of Agronomy, Technical University of Lisbon, Tapada da Ajuda, 1349-017 Lisbon, Portugal, Fax: +351 21 362 1575; email: [email protected]

SUMMARY - Water productivity became an important issue in improving the performance of irrigation, including when focusing water saving issues. However, various concepts may be considered, which requires appropriate definitions and related analysis. Because it represents a ratio between harvesting yield and water use, a main component of the latter is crop evapotranspiration. This calls for a discussion relative to its concepts and computation, thus to identify both how crop evapotranspiration may be managed to improve water productivity and existing gaps in its knowledge. In addition, a discussion on the economic aspects relative to improved water productivity and saving is presented, including a simple analysis of economic issues and respective gaps in knowledge. Keywords: water productivity, land productivity, evapotranspiration, aerodynamic resistance, surface resistance, crop coefficients, stress coefficients.

WATER PRODUCTIVITY VS. IRRIGATION EFFICIENCY The term efficiency is commonly applied an irrigation systems or sub-system: water storage, conveyance, distribution off- and on-farm, and application at the farm. It can be defined by the ratio between the water depth delivered by the sub-system under consideration and the water depth supplied to that sub-system, usually expressed as a percentage. Adopting an output/input nondimensional ratio, the term efficiency could be applied to evaluate the performance of any irrigation and non-irrigation water system but the term is almost exclusive of irrigation (Pereira et al., 2002a). Misleading interpretations are therefore common by water managers, which should be avoided. For farm irrigation systems, the application efficiency Ea may be defined by the ratio between the average water depth added to the root zone storage in the quarter of the field receiving less water to the average water applied. However, this indicator should be used together with others, mainly those relative to distribution uniformity (Burt et al., 1997; Pereira, 1999; Pereira and Trout, 1999). Moreover, this indicator Ea should not be used for characterizing seasonal irrigation but only each irrigation event because soil water availability and water depths applied vary from an irrigation to the next and they highly influence this performance indicator as analysed by those authors. In addition, weather conditions, mainly wind speed and temperature in case of sprinkler irrigation, may also vary from an irrigation event to another and largely influence Ea. Farmers do not see the improvement of farm application efficiencies as a must. Application efficiencies become higher when farmers apply water timely and the distribution uniformity is higher. Improved uniformities decrease differences in amounts of water made available for the crop in the under-and over-irrigated parts of the field. As discussed by many authors, e.g. Keller and Bliesner (1990) and Mantovani et al. (1995), this leads to more even crop development and higher yields. When the farmer adopts an appropriate irrigation scheduling, then yields are positively impacted; in addition, the application efficiency becomes higher as well as the economic results of irrigation (Ortega et al., 2005). Thus, improving irrigation efficiency is not a farmer’s objective but to achieve higher yields and economic profit. Improving transport and distribution efficiencies may be an objective of farmers management of irrigation systems when seepage, leaking or overflow would decrease the availability of water to tailend distributor canals and tail-end farmers, or when improvements aim at an easier control of deliveries to branch canals, distributors and farms. In other words, the interest of farmers is to have improved service performances. Thus, the former efficiency terms are being replaced by indicators of

31



canal and pipe systems that refer to service performance, such as reliability, dependability and equity (Molden and Gates, 1990; Lamaddalena and Pereira, 1998; Lamaddalena and Sagardoy, 2000; Pereira et al., 2003a; Bos et al., 2005). Improving irrigation efficiencies is often said to be an objective associated with water savings. However this is only true when the farmers and canal managers have the appropriate tools and farm and off farm irrigation systems are designed and managed in such a way that delivery and irrigation scheduling can be applied effectively, as discussed by Pereira et al. (2002a, b). Otherwise water savings may not be achieved in that way but through under-irrigating the crop. Nowadays, there is a trend to call for increasing water productivity as a must (FAO, 2002; Molden et al., 2003). The attention formerly paid to irrigation efficiency issues is therefore being transferred to water productivity. However, this term is used with different meanings (Fig. 1). Water productivity may be generically defined as the ratio between the actual yield achieved (Ya) and the water use, expressed in kg/m3, but the denominator may refer to the total water use (TWU), including rainfall, which is referred herein with the symbol WP:

WP =

Ya TWU

(1)

Effective rain WATER DIVERSION Agriculture and landscape

Total WP

Seepage + runoff

Conveyance + distribution reuse

Non-crop ET

Application to cropped field

Percolation + runoff

Irrig WP

Farm WP

Transpiration Soil evapor WUE YIELD Fig. 1. Different definitions of water productivity and water use efficiency. However, it may refer only to the irrigation water used (IWU) mobilized for at system level (WPI),

Ya WP = I IWU

(2)

or to the total water use at farm or field level (TWUFarm), thus including rainfall and irrigation (WPFarm),

Ya WP = Farm TWU Farm or relate to irrigation water (IWUFarm) only, thus (WPI-Farm):

32

(3)



Ya WP = I − Farm IWU Farm

(4)

The meaning of these indicators is necessarily different and may cause contradictions when the wording water productivity is used without specifying which target is being considered. The term water use efficiency (WUE) is also commonly used in irrigation but often with different meanings. Some authors refer to it as a synonymous of application efficiency, thus as a nondimensional output/input ratio; others adopt it to express the water productivity of the irrigation water, as a yield to water ratio. To avoid misunderstandings, the term water use efficiency should be limited to physiological and eco-physiological purposes (Steduto, 1996) or, as some do, may be replaced by the term transpiration ratio or similar. The idea that improving water productivity or the water use efficiency leads to water savings is also not entirely true because it is also required to distinguish between consumptive and nonconsumptive uses. The same amount of grain yield depends not only on the amount of irrigation water used but also on the amount of rainfall water that the crop could use, which relates to rainfall distribution during the crop season. Moreover, the pathways to improve yields are often not related with water management but with agronomic practices and the adaptation of the crop variety to the cropping environment. However, a crop variety that has a higher WUE than another has the potential for using less water than the second when achieving the same yield. But this is a characteristic intrinsic to the crop and is not depending upon irrigation management.

WATER PRODUCTIVITY CONCEPTS Considering Fig. 1, one may approach the different concepts relative to water productivity and assume some definitions aimed at irrigation management. Then the following base definition is adopted:

WP =

Ya TWU

(1a)

where Ya is the actual harvestable yield in kg, and TWU is the total seasonal water use by the crop in m3 or, referring to the unit surface, in mm. Eq. 1a may take a different form

WP =

Ya ETa + LF + NBWU

(5)

where Ya is the actual harvestable yield and the denominator refers to the water use components; ETa is the actual season evapotranspiration in mm, LF is the water used for leaching in mm and NBWU is the non-beneficial water use in mm. This concerns the percolation through the bottom of the root zone, runoff out of the irrigated fields, and losses by evaporation and wind drift, The beneficial water use (BWU) is then constituted of ETa and LF. If the seasonal water use is considered through the respective and diverse water sources, then Eq. 5 is replaced by the following equation:

WP =

Ya P + CR + ΔSW + I

(6)

where P is the season precipitation, CR is the amount of capillary rise, ΔSW is the difference in soil water content between planting and harvest and I is the seasonal irrigation depth, all expressed in mm.

33



One can observe that maximizing water productivity is to find out its limit when the maximal yield Ymax is attained, which means that ETa = ETc, where ETc is defined as the ET of a healthy crop, well managed and not short of water, thus cultivated under pristine conditions (Allen et al., 1998), and that NBWU is at its minimum value:

max(WP ) =

Y max ETc + LF + min ( NBWU )

(7)

An high WP may also be obtained when a crop is water stressed (up to acceptable limits); then the yield is reduced as well as the denominator terms in Eq. 5. But such an high productivity is obtained with ETa < ETc, and with LF below its target value. If this option is non-controlled, and control is difficult to be achieved including when farm systems have appropriate distribution uniformity, yields may decreased below an acceptable level and therefore induce appreciable income loss to the farmer. Observing Fig. 2 it may be seen that if the objective would be to maximize WP and not the land productivity often farmers would not have advantage in practicing supplemental irrigation. The figure also shows that the highest WP are obtained in rainfed wheat production. This may be understood if the irrigation systems they use are poorly performing. However, the figure also show that yields under rainfed production may often be below the economic viability, which clearly justifies the farmers option to adopt supplemental irrigation.

20

WP (kg ha-1 mm-1)

Irrigated

Rainfed

15

10

5 2

y = -0.4278x + 4.7328x - 0.543 2 R = 0.7611

0 0

2

4 6 8 Grain yield (t ha-1)

10

Fig. 2. Relationship between water and land productivity for durum wheat in northern Syria (source: Zhang and Oweis, 1999). It is therefore important to consider the economic issues relative to water productivity since the objective of a farmer is to achieve high income and profit. Replacing the numerator of equations above by the monetary value of the achieved yield Ya, the economic water productivity (EWP) is expressed as €/m3 and defined by:

EWP =

Value(Ya ) TWU

(8)

However, the economics of production is less visible in this form than when both the numerator and the denominator are expressed in monetary (€) terms, respectively the yield value and the TWU cost, thus yielding the following definition:

34



EWP =

Value(Ya ) Cost (TWU )

(9)

Alternatively, this definition may be expressed assuming that all water costs are due to the costs of irrigation, thus:

EWP =

Value(Ya ) P + CR + ΔSW + Cost (I )

(10)

This may not be true if water conservation measure are used and therefore there are costs associated with water harvesting or soil management that create additional mobilization of rainfall, mainly increasing the infiltrated fraction and/or reducing soil water evaporation losses. Alternatively, considering Eq. 5, the following definition may be used:

EWP =

Value(Ya ) Costs (ETa + LF + NBWU )

(11)

Determining the costs associated with the water use components as in Eq. 11 may be difficult but ideally this equation may support the economic evaluation of measures to control the NBWU. Maximizing EWP, when all costs not referring to water use are kept constant, means to find the limit to the ratio between the yield value and the yield costs associated with water use, which corresponds to maximize the crop revenue in which concerns water use:

max(EWP ) = max

Value(Y max ) = max(Income) Costs (Y max)

(12)

This maximal EWP or maximal revenue is often different from the maximal yield and depends upon the structure of the production costs (Fig. 3).

Fig. 3. Schematic representation comparing how maximizing farm incomes for a commercial and a family farm lead to different approaches to economic water productivity (costs relative to water volumes used are not considered for simplification). For a farm where labour is by workers or using somewhat sophisticated equipment associated with energy costs, then the irrigation costs grow almost linearly with the amount of irrigation water use. Contrarily, for a farm using surface irrigation without energy costs associated nor large capital investment, and where labour is provided by the family, thus is remunerated by the final yield, the

35



effective costs relative to irrigation are not depending upon the amount of irrigation water use. Therefore, the maximum net income for the first may be close to the maximal EWP, while the maximum income for the family farm are close to the maximal land productivity since land, not water, is the main limiting factor determining farm income. The related economic impacts are however less well known, insufficient data are available and tools for the respective analysis are insufficiently developed (Victoria et al., 2005). It is therefore important to understand how WP could be improved. Knowing that yields depend upon the seasonal evapotranspiration, this analysis focuses this component of the water use.

CROP EVAPOTRANSPIRATION AND RESISTANCES TO VAPOUR FLUXES The Penman-Monteith equation (Monteith, 1965) is generally considered to be able to represent the evapotranspiration from any vegetated surface (Jensen et al., 1990; Allen et al., 1994, 1998; Pereira et al., 1999). It can be expressed by the following combination equation:

λ ET =

(es −ea ) ra ⎛ r ⎞ Δ + γ ⎜⎜ 1 + s ⎟⎟ ra ⎠ ⎝

Δ (R n − G) + ρ a c p

(13)

ai r

flo w

where Rn is the net radiation, G is the soil heat flux, (es  ea) represents the vapour pressure deficit of the air, ρa is the mean air density at constant pressure, cp is the specific heat of the air, Δ represents the slope of the saturation vapour pressure -temperature relationship, γ is the psychrometric constant, and rs and ra are the (bulk) surface and aerodynamic resistances (Fig. 4)

stomatal cuticular

soil

ra

reference level

aerodynamic resistance

rs

evaporating surface

(bulk) surface resistance

Fig. 4. Schematic representation of the resistances to vapour fluxes (Allen et al., 1998) The Penman-Monteith approach as formulated above includes all parameters that govern energy exchange and corresponding latent heat flux (evapotranspiration) from uniform vegetation canopies. Most of the parameters are measured or can be readily calculated from weather data. The equation can be utilized for the direct calculation of any crop evapotranspiration as the surface and aerodynamic resistances are crop specific. Aerodynamic resistance (ra) determines the transfer of heat and water vapour from the evaporating surface into the air above the canopy. It can be expressed as:

36


ra =


ln[( zh − d ) / zoH ]ln[( zm − d ) / zom ] k 2uz

(14)

height of wind measurements [m], zh is the where ra is the aerodynamic resistance [s m-1], zm is height of air humidity measurements [m], d is the zero plane displacement height [m], zom is the roughness length governing momentum transfer [m], zoh is the roughness length governing transfer of heat and vapour [m], k is the von Karman's constant, 0.41 [-], and uz is the wind speed at height z [m s-1]. As discussed by Alves et al. (1998), the assumption that heat and vapour escape from the canopy from the level d+zoH, as it is implied in Eq. 14, can be questioned. In alternative, ra can be calculated from the top of the canopy to the reference height, using (Perrier, 1975; Stockle and Kjelgaard, 1996):

ra =

ln[( zh − d ) /(hc −d )]ln[( zm −d ) / zom ] k 2 uz

(15)

where hc is the crop height [m]. These equations are restricted for neutral stability conditions, i.e., where temperature, atmospheric pressure, and wind velocity distributions follow nearly adiabatic conditions (no heat exchange). The application of the ra equations for short time periods (hourly or less) may require the inclusion of corrections for stability. However, when predicting ETo in the well-watered reference surface, heat exchanged is small, and therefore stability correction is normally not required. For its practical application, the parameters d and zo, if not measured, can be estimated from the crop height hc [m] and LAI [-] (e.g. Brutsaert, 1982; Perrier, 1982):

2 ⎛ (1 − exp(− LAI / 2))⎞⎟ d =hc ⎜1 − ⎝ LAI ⎠ zom =hc exp(− LAI / 2 )[1 − exp( − LAI / 2 )]

(16)

(17)

Genetic improvements and crop management influence these parameters through acting on hc [m] and LAI but impacts are relatively small. Aiming at increasing WP (Eqs. 5 and 7) changes should focus on decreasing ETa and ETc, thus on increasing the aerodynamic resistance, thus decreasing both d and zom (vd. Eqs. 14 and 15); however the main impact on ra depends on weather conditions through wind speed. Eqs. 16 and 17 show that d and zom increase for high and fully cover crops and are smaller for low crop heights and partial cover crops. Thus, plant breeding improvements may favour higher ra when crops become lower in height and LAI. Surface resistance is more complex. The ‘bulk’ surface resistance describes the resistance of vapour flow through the transpiring crop and evaporating soil surface. Where the vegetation does not completely cover the soil, the resistance factor should indeed include the effects of the evaporation from the soil surface. If the crop is not transpiring at a potential rate, the resistance depends also on the water status of the vegetation. Plant physiologists consider rs to be a purely physiological parameter that accounts for the stomatal control of transpiration. Stomata have been carefully studied and the factors that determine their functioning are well known. Some of them, like radiation, either solar radiation Rn or PAR, temperature (T) and vapour pressure deficit (VPD) are those that govern the physical process of evaporation. Others, like soil (or plant) water potential ( ψ ) represent the true physiological control by stomata which takes place mainly in water stress conditions. Other factors, like the age of the leaf, the previous history of water stress of the plant and the position of the leaf in the plant, are also important but less quantifiable (Alves and Pereira, 2000). Scaling resistances from leaf to canopy, which constitutes the "bottom up" approach to rs, is full of controversy. The standard procedure is to average stomatal resistance rst at different levels in the

37



canopy, weighted by leaf area index (Monteith, 1973). However, the values of rs determined this way even with measured stomatal resistances seem to give good results only in very rough surfaces, like forests, and partial cover crops with a dry soil. On complete cover crops, especially when the soil is wet, average stomatal resistance can greatly depart, being normally lower, from the values of rs obtained as a residual term of the Penman-Monteith equation using the "top down" approach (e.g. Baldocchi et al., 1991; Rochette et al., 1991). The following equation establishes the essential relations between rs and weather variables (Alves and Pereira, 2000):

ρ c VPD ⎞ ⎛Δ rs = ra ⎜⎜ β −1⎟⎟ + (1 + β ) a p γ (Rn −G ) ⎠ ⎝γ

(18)

where Δ is the slope of the vapour pressure curve (Pa/ºC), γ is the psychrometric constant (Pa/ºC), ρa is the atmospheric density (kg/m3), cp is the specific heat of moist air (J kg-1 ºC-1), Rn-G is the energy available at the crop surface, and β is the Bowen ratio. This discrepancy has been regarded as to indicate that not all leaves actually contribute to the total evaporation fluxes from the canopy. The concept of "effective" leaf area was therefore introduced and linked to radiation interception, the upper, well illuminated leaves being those that most contribute to transpiration. The surface resistance rs is crop specific and relates to the stomatal resistance rst and to the "effective" leaf area; increasing resistance to water stress implies increased stomatal control, thus higher rst and rs. plant breeding for increased resistance to water stress may lead to increased rst and rs; however, acting on these crop characteristics is difficult and should avoid that increasing rst and rs would lead to lower photosynthetic efficiency, which would decrease WUE. Referring to Eq. 18, non considering the role of climate that may be the essential factor determining rs, it becomes evident that the main factor to act on is the Bowen ratio β: since it represents the ratio between sensible and latent heat fluxes, a low β indicates high water availability to the crop and an high β is representative of water stress conditions. Therefore, for unchanged climate conditions, high rs values are obtained when the aerodynamic resistance is high and the crop is water stressed. However, limiting the availability of water to the crop may lead to lowering the photosynthesis and to reduced yields. As for ra, acting on rs is difficult, could have contradictory results in terms of crop yields, and may be less efficient in increasing WP.

CROP EVAPOTRANSPIRATION AND CROP COEFFICIENTS Crop evapotranspiration can be derived from meteorological and crop data by means of the Penman-Monteith equation (Eq. 13). By adjusting the albedo and the aerodynamic and canopy surface resistances to the growing characteristics of the specific crop, the evapotranspiration rate can be directly estimated. The albedo and resistances are, however, difficult to estimate accurately as they may vary continually during the growing season as climatic conditions change, as the crop develops, and with soil surface wetness and soil water availability. The canopy resistance will further be influenced by the soil water availability, and it increases strongly if the crop is subjected to water stress. As there is still a considerable lack of consolidated information on the aerodynamic and canopy resistances for the various cropped surfaces, the crop coefficient approach is generally used to estimate the crop evapotranspiration, ETc, which is calculated by multiplying the reference crop evapotranspiration, ETo, by a crop coefficient, Kc:

ETc = K c ETo

(19)

where ETc is the crop evapotranspiration [mm d-1], Kc is the crop coefficient [dimensionless], and ETo is reference crop evapotranspiration [mm d-1]. Considering the original Penman-Monteith equation (Eq. 13) and the equations of the aerodynamic and surface resistance described above, the FAO-PM reference ETo equation for daily time-step computations has the following form (Allen et al., 1994):

38



900 u 2 (es − ea ) T + 273 Δ + γ ( 1 + 0.34 u 2 )

0.408 Δ (Rn − G) + γ ET = o

(20)

where ETo is the reference evapotranspiration [mm day-1], Rn is net radiation at the crop surface [MJ m-2 day-1], G is soil heat flux density [MJ m-2 day-1], T is air temperature at 2 m height [°C], u2 is wind speed at 2 m height [m s-1], es is saturation vapour pressure [kPa], ea is actual vapour pressure [kPa], es-ea is saturation vapour pressure deficit [kPa], Δ is the slope of the vapour pressure curve [kPa °C-1], and γ is the psychrometric constant [kPa °C-1]. The FAO-PM equation (Eq. 20) determines the evapotranspiration from the hypothetical grass reference surface and provides a standard to which evapotranspiration in different periods of the year and in other regions can be compared and to which the evapotranspiration from other crops can be related. Most of the effects of the various weather conditions are incorporated into the ETo estimate. Therefore, as ETo represents an index of climatic demand, Kc varies predominately with the specific crop characteristics and only to a limited extent with climate. This enables the transfer of standard values for Kc between locations and between climates. This has been a primary reason for the global acceptance and usefulness of the crop coefficient approach and the Kc developed in past studies. The crop coefficient, Kc, is basically the ratio of the crop ETc to the reference ETo, and it represents an integration of the effects of four primary characteristics that distinguish the crop from reference grass. These characteristics are: -

Crop height. The crop height influences the aerodynamic resistance term, ra, of the FAO Penman-Monteith equation and the turbulent transfer of vapour from the crop into the atmosphere. The ra term appears twice in the full form of the FAO Penman-Monteith equation.

-

Albedo (reflectance) of the crop-soil surface. The albedo is affected by the fraction of ground covered by vegetation and by the soil surface wetness. The albedo of the crop-soil surface influences the net radiation of the surface, Rn, which is the primary source of the energy exchange for the evaporation process.

-

Canopy resistance. The resistance of the crop to vapour transfer is affected by leaf area (number of stomata), leaf age and condition, and the degree of stomatal control. The canopy resistance influences the surface resistance, rs.

-

Evaporation from soil, especially exposed soil.

The soil surface wetness and the fraction of ground covered by vegetation influence the surface resistance, rs. Following soil wetting, the vapour transfer rate from the soil is high, especially for crops having incomplete ground cover. The combined surface resistance of the canopy and of the soil determines the (bulk) surface resistance, rs. The Kc in Equation 19 predicts ETc under standard, pristine conditions. This represents the upper envelope of crop evapotranspiration and represents conditions where no limitations are placed on crop growth or evapotranspiration due to water shortage, crop density, or disease, weed, insect or salinity pressures. The ETc predicted by Kc is adjusted if necessary to non-standard conditions, ETc act or ETa, where any environmental condition or characteristic is known to have an impact on or to limit ETc. For this reason, ETa is used when defining WP (Eq. 5) and ETc is used in Eq. 7 referring to the maximal WP. These aspects are represented in Fig. 5. Actual ETc can be less than the potential ETc under non-potential growing conditions including water stress or high soil salinity. The non-potential ETc is termed “actual ETc” and is represented as ETc act. It is defined as:

39



ETc act = K c act ETo

(21)

where Kc act is the “actual” crop coefficient that includes effects of environmental stresses.

Radiation Temperature Wind speed Humidity

ETo

grass reference crop

climate

+

= well watered grass

K c factor

ETo x

ETc

Yield = Ym

ETc act adj

Yield = Ya

= well watered crop

optimal agronomic

conditions

K x K adjusted c s

ETo x

= water & environmental stress

Fig. 5. Schematic representation on the relationships between reference, potential and actual crop evapotranspiration and crop yields (adapted from Allen et al., 1998) In its dual form, Kc = Kcb + Ke (Allen et al., 1998). The basal crop coefficient Kcb represents the ratio of ETc to ETo under conditions when the soil surface layer is dry, but where the average soil water content of the root zone is adequate to sustain full plant transpiration. Additional evaporation due to wetting of the soil surface by precipitation or irrigation is represented in the evaporation coefficient Ke. The total, actual Kc act is the sum of Kcb and Ke reduced by any occurrence of soil water stress:

K c act = K s K cb + K e

(22)

where Kcb is the basal crop coefficient [0 - 1.4], Ke is a soil water evaporation coefficient [0 - 1.4] and Ks is the stress reduction coefficient [0 - 1], which reduces the value of Kcb when the average soil water content of the root zone is not adequate to sustain full plant transpiration. Ks is equal to 1 when no stress occur, thus then the first term of Eq. 22 becomes equal to Kc. Ke represents the evaporation component from wet soil that occurs in addition to the ET represented in Kcb. The sum of Kcb and Ke can not exceed some maximum value for a crop, based on energy limitations, generally referred as the Kc. the value. An update and extension on the use of Eq. 22 is given by Allen et al. (2005a). The linearized form used for mean Kc and basal Kcb curves in FAO-56 was introduced in FAO-24 (Doorenbos and Pruitt, 1977). The FAO Kc curve is comprised of four straight line segments representing the initial period, the development period, the midseason period and the late season period (Fig. 6). These segments are defined by three primary Kc values: Kc during the initial period (Kc ini), Kc during the midseason (full cover) period (Kc mid) and Kc at harvest (or at the end of the late

40



season) (Kc end). The Kc ini defines the horizontal portion of the Kc curve during the initial period until approximately 10% of the ground is covered by vegetation. The Kc mid defines the value for Kc during the peak period for the crop, which is normally when the crop is at "effective full cover". This period is described by a horizontal line extending through Kc mid. The development period is defined by a sloping line that connects the initial and midseason periods. The late season has a sloping line that connects the end of the midseason period with the harvest (end) date.

Kc mid

1.2 1.0 0.8

Kc 0.6 0.4 0.2

Kc ini Initial Period

Crop Dev. Period

Mid Season Period

Late Season Period

K c end

0.0

Time of Season, days Fig. 6. Schematic of the generalized Kc curve with four crop growth stages and three Kc or Kcb values (Allen et al., 1998) The Kc values vary with a large number of factors as represented in Fig. 7. First they depend on the crop through its characteristics determining the aerodynamic and surface resistances as defined above. Thus the Kc values and the respective crop stage durations vary from crop to crop reflecting the respective heights and LAI determining ra, stomatal control, degree of soil cover by vegetation, and plant density determining the bulk surface resistance and, the latter, influencing the albedo and the soil evaporation component. Secondly, the Kc varies with the crop growth stage (Fig. 6). For annual crops, it varies from planting to about 10% soil cover during the initial phase, then until full crop cover and from then to the start of senescence of leaves, and finally until harvest. All factors mentioned above – height, LAI, number and functioning of stomata, soil cover – gradually change along the crop season, thus also the ra and rs, as well as the albedo and soil evaporation. For deciduous trees and shrubs changes go from leaf initiation to full cover and, later, from starting leaf senesce to the fall of leaves. In addition to these changes there are those in plant density while the crop is aging until attaining the target development. Additional changes have to be considered relative to the occurrence of ground cover by vegetation, which also uses water including during the dormancy period of the crop. For the evergreen trees and shrubs changes in crop characteristics determining ra, rs and albedo only occur during the first years of the crop but the influences of ground cover are also non-negligible. Finally, there are also the permanent pastures whose characteristics vary with cuts and between cuts. Management influences refer to soil management, which may be such that evaporation is reduced, planting densities determining soil cover and vapour fluxes inside the canopy, and harvesting date, since a crop may be harvested fully mature or, as for table crops, before that leaf senescence affect the quality of the product. This influences particularly the Kc end.

41


1.2

sugar cane cotton maize cabbage, onions apples

1.0 frequent 0.8 0.6

0.2

wetting events d ie dr

0.4

ted es h rv res ha f

Kc


infrequent .. 25 . 40 . 60 .. % ground cover

crop initial development

(short)

mid-season

late season (long)

main factors affecting K c in the 4 growth stages soil ground cover plant evaporation development

crop type (humidity) (wind speed)

crop type harvesting date

Fig. 7. Schematic of the variation of the Kc curve with crop, environmental and management factors (Allen et al., 1998) Crop management is also influencing Kc: the crop may be managed for achieving the potential yield Ymax, or some aspects be poorly practiced and affect crop height, LAI and the stomatal control. These factors include seedbed preparation, seeding dates, plant density, fertilising, pest and diseases control, weed control, and irrigation. In addition, the Kc are influenced by the climate although the main climatic influences are incorporated in ETo. The climate largely determines the duration of the crop growth stages. The Kc ini for most crops excepting the evergreen ones is essentially representing soil water evaporation since the crop is not covering but a small fraction of the soil. It is therefore determined by the frequency and amount of wettings during this stage and by the potential evaporation rate from the soil. It is also influenced by the amount of water available in the upper soil layer, of 10 to 15 cm and by the soil water holding capacity and its capillary rise potential to bring water stored below into this evaporative soil layer. This process and formulation is well described by Allen et al. (2005b). Factors referred above may largely vary from one location to another and, for the same location, from one crop season to the next. This explains why a large variation of values is shown in Fig. 7. In addition, differences in management add to those factors as referred in the following. The Kc ini may be largely modified by management practices such as direct seeding (no tillage), soil mulching by straw or plastic, tunneling with plastic as for horticultural crops, and other similar practices. Two effects occur: on the one hand, there is a decrease of energy at the soil surface, thus a decrease of the evaporation rates; on the other hand, there is an increased resistance to vapour transfer from the soil surface into the atmosphere, mainly relative to an increase in the surface resistance. Therefore, Kc ini may be much lower than under conditions when the soil is fully exposed to radiation. However, impacts referred in literature are variable reflecting differences in soil cover; in case of vegetal mulching this highly depend on the amount and distribution of soil coverage; in case of plastic mulch they depend on its transparency to radiation of short and long wave, and on the amount of openings in the plastic through where vapour may escape to the atmosphere. Anyway, in

42



general, soil covering with mulch reduces Kc ini, thus soil evaporation, not affecting transpiration of the canopy. Effects are kept but progressively reduced during the crop development phase and mostly disappear at the mid-season for full cover crops. Adopting the dual crop coefficient approach (Eq. 22) these effects are better studied since it becomes possible to separate the soil evaporation and the transpiration components of ETc In case of tree and shrub crops where the soil is covered by active vegetation, which is often required as a measure to combat erosion, the impacts of soil cover are totally different since this vegetation also uses water and the crop water requirements of such crops are then increased relative to bare soil conditions. The Kc mid essentially varies with the crop but is also affected by the climate, namely when advective conditions occur. This relation with climate is analysed by Pereira et al. (1999). Therefore, an adjustment to climate is proposed by Allen et al. (1998) aimed at exporting to other climates the tabled values for Kc mid. In fact these tabled values refer to a standard climate where the wind speed u2 is 2 m s-1 and the minimum relative humidity RHmin is 45% during the mid season of the considered crop. Then the adjustment to any other climate is performed through the following equation:

⎛h⎞ K c mid = K c mid ( standard climate ) + [0.04 ( u 2 − 2 ) − 0.004 ( RH min − 45 )] ⎜ ⎟ ⎝3⎠

0.3

(23)

where Kc mid and Kc mid (standard climate) refer to the location where the application is performed and to the standard conditions, and h is the crop height [m] during mid-season. This equation shows well the aspects referred above relative to ra since u2 is a main variable controlling ra and determining this equation. he dominant. Kc increases with wind speed and decreases inversely. Therefore, reducing ET at the mid-season may be obtained by avoiding high winds as it is commonly done in arid lands, either through cropping in areas less exposed to wind or using wind breaks. RHmin, for a certain extend, represents well the conditions for diffusion of vapour into the atmosphere, very strong in arid climates where RHmin is low. Changing RH to control evaporation is generally non practical, but RH tends to increase when wind is reduced and thus the transport of vapour from the air layer close to the crop. Eq. 23 shows that these impacts are larger for tall crops and smaller for short ones, which agrees with the above referred increase of ra when the crop is of low height. The Kc end is largely affected by management which determines harvesting, thus the end of the crop season (Fig. 7). Moreover, harvesting earlier increases Kc end relative to a late harvesting of the same crop but shortens the duration of the end-season period. Harvesting earlier is practiced for food crops that should be eaten fresh, and harvesting later is adopted for crops when conservation or preservation is easier when they are stored dry, as for cereals. Climate also impacts the Kc end when the crop is harvested fresh. Then Eq. 23 applies but variables refer to the the late season. Controlling ET during this period refers to the same aspects as for the mid-season. Kc values are well known for the temperature climate crops, mainly the annual crops (Allen et al., 1998); some deficiencies in knowledge occur for tree crops due to differences in plant density, ground cover by vegetation or mulch, and architecture of the plantations. However, literature is producing consistent knowledge that provides for adopting coherent values for Kc in other regions. Main gaps refer to tropical and sub-tropical crops, which research is less abundant and largely published in native languages. A great effort in improving the corresponding base-knowledge is necessary. Frequent gaps in practical knowledge refer to the lack of adoption of appropriate estimation of Kc ini when using the rough indicative values in Tables, and the lack in adjusting the Kc mid and Kc end to climate (Eq. 23). Moreover, it is the use of Kc without referring to the fact that the crop is managed for potential yield or not, thus that crop factor is not differentiate between Kc and Kc act (Eq. 19 and 21). Then is often said that FAO 56 tabled values are not appropriate when the main problem is to use only part of the information produced in the guidelines and omitting the use of all other adjusting tools.

43



It is necessary to underline herein that looking to develop water productivity assessment without fully considering (and understanding) the concepts and calculation tools for crop evapotranspiration is inadequate since the main component in computing irrigation depths is Etc or ETa and, without knowing these, one can only roughly know how much is the non-beneficial water use term NBWU (Eq. 7 and 11).

WATER STRESS AND IMPACTS ON YIELD The basic relationship between crop ET and yields may be represented by the Stewart model (Stewart et al., 1977, Doorenbos and Kassam, 1979) relating the relative yield decrease with the relative evapotranspiration deficit (see Fig. 5)

ETc act ⎞ ⎛ ⎛ Ya ⎞ ⎟ ⎟⎟ = K y ⎜1 − ⎜⎜1 − ⎟ ⎜ Y ET m c ⎠ ⎝ ⎠ ⎝

(24)

where Ky is the yield response factor [-], ETcadj is the adjusted (actual) crop evapotranspiration [mm d1 ], ETc is the crop evapotranspiration for standard conditions (no water stress) [mm d-1], Ya is the crop yield when ET = ETcadj , and Ym is the maximal crop yield corresponding to ETc. A better description of ET impacts on yields is obtained when the yield response factors refer to specific crop phases and the history of the crop stresses is taken into consideration as it is largely reported in the literature. Recombining the terms of Eq. 24, the stress coefficient Ks (Eq. ) may be expressed as

K s = 1−

1 Ky

⎛ Ya ⎞ ⎜⎜1− ⎟⎟ ⎝ Ym ⎠

(25)

which expresses how the relative yield decrease impact the stress coefficient as a function of the yield response factor. However, for operational purposes, it is better to express Ks as a function of the soil water depletion:

⎛ TAW − Dr K s = ⎜⎜ ⎝ TAW − RAW

⎞ ⎟⎟ ⎠

(26)

where TAW and RAW are respectively the total and readily available soil water, and Dr is the cumulated soil water depletion between two wetting events by rain or irrigation. Eq. 26 indicates that Ks < 1 when soil water is depleted below the RAW threshold. Yields may be affected by other environmental and management factors such as salinity (Hamdy and Karajeh, 1999; Minhas, 1996; Rhoades et al., 1992). A simplified approach for salinity impacts for conditions where ECe > ECe threshold is

(

)

Ya b = 1 − ECe − ECe threshold Ym 100

(27)

is the maximum expected crop yield when ECe < ECe threshold, where Ya is the actual crop yield, Ym ECe is the mean electrical conductivity of the saturation extract for the root zone [dS m-1], ECe threshold is the electrical conductivity of the saturation extract at the threshold of ECe when crop yield first reduces below Ym [dS m-1], and b is the reduction in yield per increase in ECe [%/(dS m-1)]. Values for ECe threshold and b are tabled by Rhoades et al. (1992) and Allen et al. (1998). The combined impacts of water and salinity stress are expressed through

⎛ b K s = ⎜1 − ECe − ECe threshold ⎜ K y 100 ⎝

(

44

⎞

TAW − Dr ⎞ ⎟ )⎟⎟ ⎛⎜⎜ TAW − RAW ⎟ ⎠⎝

⎠

(28)



indicating that Ks < 1 when either the ECe threshold or the soil water threshold θt (corresponding to RAW) are attained (Fig. 8). The concepts described by Eq. 24 to 28 are the base for deficit irrigation together with the knowledge of the crop development phases when water stress impacts are smaller. Using those equations it is then possible to easily compute

ETc act = K c act ETo

(21 bis)

from appropriate estimation of Ks given as above

K c act = K s K cb + K e

(22 bis)

or simply

K c act = K s K c

(29)

: soil water content

Ks

1.00

t

FC

low

0.80

with soil salinity 0.60 0.40

WP

no

so il s

al in it y

high

0.20 0.00

0

RAW Dr : depletion from root zone (mm)

TAW

Fig. 8. Schematic of soil water and salinity determining the stress coefficient Ks (Allen et al., 1998) The great difficulty in adopting an irrigation management that allows for crop stress during selected phases of the crop season is the insufficient knowledge about the respective economic impacts. The literature is abundant on which crop phases are less or more sensitive to water and salt stress; numerous field observation tools allow to assess the soil water status and, less often, the salinity conditions; a variety of models may be used to simulate the soil water balance and therefore provide appropriate information for irrigation scheduling. Although a few tentative exist (Victoria et al., 2005), the great gap refers to combining physical assessment and simulation with economic assessment of impacts. In an example referring to Tunisia and Portugal (Rodrigues et al., 2003; Zairi et al., 2003) it is shown that more than following the ET-yield relations, it is necessary to analyse the relationships between ET or irrigation and the economic impacts of deficit irrigation. Reducing ET leads to reduced yields (land productivity) but increased water productivity. This option may be feasible when decreasing Gross Margins per unit land (GM/ha), i.e. the economic land productivity, it results in increased GM per unit water, thus increased economic water productivity as exemplified in Fig. 9

45



(b)

GM (USD / m3)

GM (USD / ha)

(a)

1400 1200 1000 800 600 400 200

2.0 1.5 1.0 0.5 0.0

0 0

40

(1)

80

120

160

200

240

(2)

(3)

(4)

(5)

(6)

0

280

40

(1)

(7)

80

(2)

120

160

200

240

280

(3)

(4)

(5)

(6)

(7)

Season irrigation (mm)

Season irrigation (mm)

Number of irrigations


Fig. 9. Gross margins per unit surface (a) and per unit volume of water applied (b) for alternative ), and very high ( ) deficit irrigation strategies for the wheat crop under average ( climatic demand conditions (Zairi et al., 2003) The farmer incomes then reduces but, when water is lacking, that income is higher than reducing the cropped area. However, for crops growing out of the rainy season, these relations are different as shown for a tomato crop in the same region (Fig. 10)

GM (USD / m3)

GM (USD / ha)

0.8 (b)

(a)

5000 4000 3000 2000 1000

0.6 0.4 0.2 0.0

0 240 320 400 480 560 640 720 800 880

240 320 400 480 560 640 720 800 880

(6)

(6)

(8) (10) (12) (14) (16) (18) (20) (22) Season irrigation (mm)


(8)

(10) (12) (14) (16) (18) (20) (22) Season irrigation (mm)


Fig. 10. Gross margins per unit surface (a) and per unit volume of water applied (b) for alternative ), high ( ) and very deficit irrigation strategies of tomato crop in Siliana for average ( ) demand conditions (Zairi et al., 2003). high ( Thus, for crops having a high demand for water such as summer crops, including if they have a favourable ratio between yield price and water cost, as it is the case for tomato in Tunisia, when GM/ha decreases due to less water application the economic water productivity do not increase but may decrease. It is then questionable to adopt deficit irrigation. This response is different with different structure of production costs as exemplified for center-pivot irrigated maize in Portugal (Fig. 11).

46



Fig.11. Variation of the gross margins per hectare cropped (---) and per m3 of water applied (___) relative to the maize crop when irrigation depths decrease from full irrigation to heavy deficits ), high ( ) and very high ( ) climatic demand conditions (Rodrigues et for average ( al., 2003)

CONCLUSION REMARKS The analysis above shows that the recently adopted concept of water productivity may be advantageous relative to the old concept of irrigation efficiency. However, it is required to well define both the numerator and the denominator of the water productivity term in order to further perform appropriate analysis of irrigation performances in relation to yield. Crop evapotranspiration is an essential term when assessing water productivity. Thus, it is required to well understand which factors lead to potential or non-potential (below the potential) crop evapotranspiration and how these factors may be manipulated or managed to achieve a controlled crop demand with minimal negative impacts on yields. Current knowledge provides the appropriate tools for such assessment despite gaps in knowledge relative to tropical and sub-tropical crops in particular. However the main gaps identified in the practice refer to a less good use of existing tools and concepts. Improved modelling tools may help solving this but making better use of present know how is critical. Moreover, it shows to be essential the appropriate combination of physical assessment tools with economic assessment. Economic water productivity seems to be relevant since farmers decisions are based in income considerations. Therefore, an improved knowledge about the economic relations is required and economic considerations must be integrated with engineering approaches and not left as external to be just analysed by economic specialists.

REFERENCES Allen, R.G., Willardson, L.S., Frederiksen, H.D., 1997. Water use definitions and their use for assessing the impacts of water conservation. In: J. M. de Jager, L.P. Vermes, and R. Ragab (Eds.) Sustainable Irrigation in Areas of Water Scarcity and Drought (Proc. ICID Workshop, Oxford), British Nat. Com. ICID, Oxford, pp. 72-81. Allen, R.G., Smith, M., Perrier, A., Pereira, L.S., 1994. An update for the definition of reference evapotranspiration. ICID Bulletin, 43(2): 1-34.

47



Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration. Guidelines for Computing Crop Water Requirements. FAO Irrig. Drain. Paper 56, FAO, Rome, 300 p. Allen, R.G., Pereira, L.S., Smith, M., Raes, D., Wright, J.L., 2005a. FAO-56 Dual crop coefficient method for estimating evaporation from soil and application extensions. J. Irrig. Drain. Engng. 131(1): 2-13. Allen, R.G., Pruitt, W.O., Raes, D., Smith, M., Pereira, L.S., 2005b. Estimating evaporation from bare soil and the crop coefficient for the initial period using common soils information. J. Irrig. Drain. Engng. 131(1): 14-23. Alves, I, Pereira, L.S., 2000. Modelling surface resistance from climatic variables? Agric. Water Manag. 42: 371-385. Alves, I., Perrier, A., Pereira, L.S., 1998. Aerodynamic and surface resistances of complete cover crops: How good is the “big leaf” approach? Trans. ASAE 41(2): 345-351. Baldocchi, D.D., Luxmoore, R.J., Hatfield, J.L., 1991. Discerning the forest from the trees: an essay on scaling canopy stomatal conductance. Agric. For. Meteorol. 54: 197-226. Bos, M.G., Burton, M.A., Molden, D.J., 2005. Irrigation and drainage Performance Assessment. Practical Guidelines. CABI Publ. Wallingford. Brutsaert, W., 1982. Evaporation into the Atmosphere. R. Deidel Publ. Co, Dordrecht, Burt, C.M., Clemmens, A.J., Strelkoff, T.S., Solomon, K.H., Bliesner, R.D., Hardy, L.A., Howell, T.A., Eisenhauer, D.E., 1997. Irrigation performance measures: efficiency and uniformity. J. Irrig. Drain. Engng. 123: 423-442. FAO, 2002. Crops and drops: making the best use of water for agriculture. FAO, Land and Water Development Division, Rome, Italy Hamdy, A., Karajeh, F., (Eds.) 1999. Marginal Water Management for Sustainable Agriculture in Dry Areas. (Proc. Advanced Short Course, Aleppo, Syria), ICARDA, Aleppo and CIHEAM/IAM-B, Istituto Agronomico Mediterraneo, Bari. Jensen, M.E., 1996. Irrigated agriculture at the crossroads. In: Pereira, L. S., Feddes, R. A., Gilley, J. R., Lesaffre, B. (Eds.) Sustainability of Irrigated Agriculture. Kluwer Acad. Publ., Dordrecht, pp. 1933. Jensen, M.E., Burman, R.D., Allen, R.G. (Eds.), 1990. Evapotranspiration and Irrigation Water Requirements. Am. Soc. Civ. Eng. Manual No. 70, 332 pp. Keller, J., Bliesner, R.D., 1990. Sprinkler and Trickle Irrigation. Van Nostrand Reinhold, New York, 652 pp. Lamaddalena, N., Pereira, L.S., 1998. Performance analysis of on-demand pressurized irrigation systems. In: Pereira, L.S., Gowing, J.W. (Eds.) Water and the Environment: Innovation Issues in Irrigation and Drainage, E &FN Spon, London, pp. 247-255. Lamaddalena, N., Sagardoy, J.A., 2000. Performance Analysis of On-Demand Pressurized Irrigation Systems. FAO Irrigation and Drainage Paper 59, FAO, Rome, 132 pp. Mantovani, E.C., Villalobos, F.J., Orgaz, F., Fereres, E., 1995. Modelling the effects of sprinkler irrigation uniformity on crop yield. Agric. Water Manage. 27: 243-257. Minhas, P.S., 1996. Saline water management for irrigation in India. Agric. Water Manage. 38: 1-24. Molden, D.J. and Gates, T.K., 1990. Performance measures for evaluation of irrigation-water-delivery systems. J. Irrigation and Drainage Engineering 116(6): 804-823. Molden, D., Murray-Rust, H., Sakthivadivel, R., Makin, I., 2003. A water-productivity framework for understanding and action. In: Kijne JW, Barker R, Molden D (eds.), Water Productivity in Agriculture: Limits and Opportunities for Improvement, IWMI and CABI Publ., Wallingford, pp 1-18. Monteith, J.L., 1965. Evaporation and the environment. XIXth Symposia of the Society for Experimental Biology. In the State and Movement of Water in Living Organisms. University Press, Swansea, Cambridge, 205–234. Monteith, J.L. 1973. Principles of Environmental Physics. Edward Arnold, London. Ortega, J.F., de Juan, J.A., Tarjuelo, J.M., 2005. Improving Water Management: The Irrigation Advisory Service of Castilla la Mancha (Spain). Agric. Water Manage. 77: 37-58. Pereira, L.S., 1999. Higher performances through combined improvements in irrigation methods and scheduling: a discussion. Agric. Water Manage. 40 (2): 153-169.

48



Pereira, L.S., 2003. Performance issues and challenges for improving water use and productivity (Keynote). In: T. Hata and A H Abdelhadi (Session Organizers) Participatory Management of Irrigation Systems, Water Utilization Techniques and Hydrology (Proc. Int. Workshop, The 3rd World Water Forum, Kyoto), Water Environment Lab., Kobe University, pp. 1-17. Pereira, L.S., Trout, T.J., 1999. Irrigation methods. In: van Lier, H.N., Pereira, L.S., Steiner, F.R. (Eds.) CIGR Handbook of Agricultural Engineering, vol. I: Land and Water Engineering, ASAE, St. Joseph, MI, pp. 297-379. Pereira, L.S., Perrier, A., Allen, R.G., Alves, I., 1999. Evapotranspiration: Review of concepts and future trends. J. Irrig. Drain. Engng. 125(2): 45-51. Pereira, L.S., Cordery, I., Iacovides, I., 2002a. Coping with Water Scarcity. UNESCO IHP VI, Technical Documents in Hydrology No. 58, UNESCO, Paris, 267 p. (accessible through http://unesdoc.unesco.org/images/0012/001278/127846e.pdf) Pereira, L.S., Oweis, T., Zairi, A., 2002b. Irrigation management under water scarcity. Agric. Water Manag. 57: 175-206. Pereira, L.S., Calejo, M.J., Lamaddalena, N., Douieb, A., Bounoua, R., 2003. Design and performance analysis of low pressure irrigation distribution systems. Irrigation and Drainage Systems 17(4): 305-324. Perrier, A., 1975. Etude physique de l'évapotranspiration dans les conditions naturelles. III évapotranspiration réelle et potentielle des couverts végétaux. Ann. Agronomiques 26: 229-243. Perrier, A., 1982. Land surface processes: vegetation. In: Eagleson, P.S. (ed) - Land Surface Processes in Atmospheric General Circulation Models, Cambridge Univ. Press, Cambridge, Mass., pp 395 - 448. Rhoades, J.D., Kandiah, A., Mashali, A.M., 1992. The Use of Saline Waters for Crop Production. FAO Irrigation and Drainage Paper 48, FAO, Rome. Rochette, P.; Pattey, E.; Desjardins, R.L.; Dwyer, L.M.; Stewart, D.W.; Dube, P.A., 1991. Estimation of maize (Zea mays L.) canopy conductance by scaling up leaf stomatal conductance. Agric. For. Meteorol. 54: 241-261. Rodrigues, P.N., T. Machado, L.S. Pereira, J.L. Teixeira, H. El Amami, A. Zairi, 2003. Feasibility of deficit irrigation with center-pivot to cope with limited water supplies in Alentejo, Portugal. In: G. Rossi, A. Cancelliere, L. S. Pereira, T. Oweis, M. Shatanawi, A. Zairi (Eds.) Tools for Drought Mitigation in Mediterranean Regions. Kluwer, Dordrecht, pp. 203-222. Rossi, G., Cancellieri, A., Pereira, L.S., Oweis, T., Shatanawi, M., Zairi, A. (eds.), 2003. Tools for Drought Mitigation in Mediterranean Regions. Kluwer, Dordrecht, 357 p. Steduto, P., 1996. Water use efficiency. In: Pereira, L. S., Feddes, R. A., Gilley, J. R., Lesaffre, B. (Eds.) Sustainability of Irrigated Agriculture. Kluwer, Dordrecht, pp. 193-209. Stockle, C.O.; Kjelgaard, J., 1996. Parameterizing Penman-Monteith surface resistance for estimating daily crop ET. In: Camp, C.R.; Sadler, E.J.; Yoder, R.E. (eds) Evapotranspiration and Irrigation Scheduling (Proc. Int. Conf., San Antonio, Texas, 3-6 Nov), ASAE, St. Joseph, MI, pp. 697-703. Victoria F.B., Viegas Filho J.S., Pereira L.S., Teixeira J.L., Lanna A.E., 2005. Multi-scale modeling for water resources planning and management in rural basins. Agric. Water Manage. 77: 4-20. Zairi, A., H. El Amami, A. Slatni, L.S. Pereira, P.N. Rodrigues, T. Machado, 2003. Coping with drought: deficit irrigation strategies for cereals and field horticultural crops in Central Tunisia. In: G. Rossi, A. Cancelliere, L. S. Pereira, T. Oweis, M. Shatanawi, A. Zairi (Eds.) Tools for Drought Mitigation in Mediterranean Regions. Kluwer, Dordrecht, pp. 181-201. Zhang, H., Oweis, T., 1999. Water-yield relations and optimal irrigation scheduling of wheat in the Mediterranean region. Agric. Water Manag. 38: 195-211.

49


50




SYSTEMATIC APPROACH TO THE IMPROVEMENT OF AGRICULTURAL WATER USE EFFICIENCY

T.C. Hsiao Professor Emeritus, Dept. of Land, Air, and Water Resources, University of California, Davis. 1 Shields Ave., Davis, California, 95616, U. S. A. Email: [email protected] SUMMARY - Effective management of scarce water resources requires a systems approach. Starting at the source of water, a cascade of events leads to the final production of crops or animal products at the expense of water. These events are mostly sequential, with each process step in the sequence having its own efficiency of output per unit of input. Using a simple sequence of three hypothetical steps, it is shown that the overall efficiency of a process is the product of the efficiencies of each sequential step. That is, efficiencies of individual process steps are multiplicative in determining the overall efficiency. Thus, improvement in any one of the efficiency steps has equal effect in improving the overall efficiency, and the overall improvement is more than the sum of the individual improvements. This principle provides a simple and quantitative means to optimize the allocation of limited resources in improving water use efficiency. Crop production in relation to water use are considered in terms of the pertinent sequence of efficiency steps for irrigated conditions. Rainfed conditions will be considered in another presentation in the Rainfed and Drought session. Efficiency steps and the sequences are outlined and discussed and the likely improvements assessed quantitatively for some scenarios. The universal applicability of this approach to different cropping as well as animal production systems when water is limiting is emphasized. Key words: Irrigation, crop productivity, water saving, management, resource allocation, optimization.

INTRODUCTION The relentless growth of human population, coupled with the intensifying desire for higher living standard, including the continuous shifting to diets based more and more on meat and dairy products, are straining the water resources all over the world, especially in the more arid regimes. Adding to the problem is the increased awareness of the need for water in the preservation of the environment and ecosystems. Since the fresh water resources are essentially finite on earth, making more efficient use of the water must be a major focal point in coping with water shortage. Numerous ways have been devised or advocated and major efforts have been made to improve the efficiency of water use in agriculture. The production of crops and animals with water as a key input involves complicated processes with myriad of facets that are subjected to the impact of management decisions and environmental influence. A systematic and quantitative approach is needed to analyze where the inefficiency lies, to assess the potential improvements, and most importantly, to determine how to allocate limited available resource to maximize the improvement in water productivity. This paper describes briefly a relatively simple and yet quantitative and comprehensive framework for these purposes. A more complete treatment is given in a paper in the forthcoming special issue of Irrigation Science (Hsiao et al., 2005).

THE CONCEPT OF CHAIN OF EFFICIENCY STEPS AND ITS SIGNIFICANCE Generally and as commonly used in economics, efficiency of any production process may be defined as the ratio of input to output for that process, both measured in quantitative units. The units to use vary depending on the situation; if the same units define both, then the efficiency ratio is unitless. For example, if the resource input as well as the production output is measured in monetary units such as dollars or euros, the efficiency ratio (or simply efficiency) would be in fractions or percentage. If the measure of input and output are in different units, then the units for the efficiency

51



must be given for the efficiency to be meaningful. For example, fuel efficiency of a car may be expressed in km per liter, the ratio of the distance traveled to the volume of gasoline consumed. When the production of a product is complicated and the starting resource input goes through many processing steps sequentially ending in the product, a simple approach is available to quantify the overall efficiency of the whole in terms of the efficiency of each of the component steps. Because the processing steps are in sequence and comes one after another, the output of the first step is the input of the second step, and the output of the second step is the input of the third step, etc. In equation form: Output i = Input 1+1 ,

and

E1 =

Output 1 Input1

(1a)

E2 =

Output 2 Output 2 = Input 2 Output 1

(1b)

E3 =

Output 3 Output 3 = Input 3 Output 2

(1c)

where E designates efficiency of a step in the efficiency chain, and the subscripts i, a running number, designates the steps; 1, 2, and 3 refer to the specific steps, 1 being the first step and 3 being the last (third). If there are only three steps in the whole efficiency chain, the overall efficiency (Eall) would be the ratio of the final output (output3) to the initial input (input1), i. e., E all =

Output 3 Input1

Because the steps are sequential, the output of the preceding step is the input of the following step, as can be seen by a close examination of Eq. 1a, 1b, and 1c. This gives rise, inevitably, the following relationship between the efficiency of individual steps and the overall efficiency: E all =

Output 1 Output 2 Output 3 × × Input1 Output 1 Output 2

(2)

It is easily seen from the right side of the equation that the numerator of the first fraction cancels out the denominator of the second fraction, and the numerator of the second fraction cancels out the denominator of the third fraction, leaving only the ratio of the last output (output3) to the first input, (input1), which is Eall. So the overall efficiency is the product of the individual efficiency steps as long as the steps for the whole process are sequential. This simple mathematical outcome holds true regardless of the number of individual steps in the whole process, although Eq 2 is written for an efficiency chain consisting only of three steps. When analyzing a production process, it is important not only to know the efficiencies of the different component steps, but also to know how improvements in the efficiency of the steps affect the overall efficiency. It turned out that by expressing the improvement as a fraction of the original efficiency, a simple equation to calculate the new overall efficiency can be obtained. Denoting the fractional improvement by , an expression for the improved efficiency of a step (Enew) is: E new = (1 + Δ) E original

(3)

Applying Eq. 3 to all the steps in an efficiency chain and designating each step by the running number j (j = 1, 2, 3, etc. depending on the position of the step in the chain), a general expression of the new overall efficiency (Eall,new) in terms of and the original overall efficiency (Eall, original) is as follows:

52



(

E all, new = E all, original × Π 1 + Δ j j

)

(4)

where Π is the multiplication operator over items j. Expressed in words, one plus the fractional improvement for each step, when multiplied together, and multiplied again by the original overall efficiency, is the new overall efficiency. Eq. 4 is general, and can be applied to any efficiency chain. It also applies to cases where there is a reduction in efficiency of some or all the steps, simply by denoting the fractional change in efficiency (Δ) as negative. There are some important features to note regarding Eq. 2 and 4: (1) The treatment is quantitative, and by simple mathematics, demonstrates the fact that the overall efficiency is the products of the efficiencies of individual steps (and not the average of the efficiencies). (2) Even though the efficiency of each step may be high, the overall efficiency is considerably or much lower because of the multiplicative effect of individual efficiencies. (3) By the same token, the same multiplicative effect makes it possible to improve the overall efficiency substantially by making minor improvement in several of the individual efficiencies. (4) The impact of a change in the efficiency of one step on the overall efficiency is strictly according to the proportional change in the efficiency of that step, regardless of where the step is located in the efficiency chain or how efficient the step is originally. Some of these features may not be intuitively obvious until some examples are given. Befitting the objectives of this conference and as an example, the chain of efficiency steps concept is applied in the next section to irrigated crop production to quantify water productivity or water use efficiency.

EFFICIENCY OF IRRIGATED CROPPING AND POTENTIAL FOR MPROVEMENT The chain of efficiency steps approach, though not so called, is sometimes used in the literature to evaluate the delivery of water from a reservoir or other sources to the soil of the root zone of the crop. This covers the water and irrigation engineering aspects but not the agronomic and crop aspects. In this paper the concept is extended all the way to crop yield, starting from water diversion from the reservoir. Beginning with the engineering aspects, one may divide up the processes into some obvious sequential steps. Water, the input, is first conveyed from the reservoir outlet to the farm gate, and this constitutes the first efficiency step in the whole process. The efficiency of this step may be termed conveyance efficiency (Econv) and is calculated as the ratio of the quantity of water (W) diverted out of the reservoir (Wvo) for that farm, to the quantity of water received at the farm gate (Wfg). The water loss along the way is by leakage and also commonly by evaporation. The efficiency of this step depends of course on the circumstances and engineering and management practices, and can vary from very low to very high. In Table 1 the range of efficiency for this first and each following step are given, one for poor situations when the efficiencies are low, and one for good situations when the efficiencies are high. These ranges are based on literature and our general understanding and do not include the more extreme values, especially those in the poor situation category. Also given in Table 1 are the overall efficiency (Eall) for the poor and good situations, calculated according to Eq. 2 from the mid-value (average of the two limits of the range) of each step efficiency. In addition, the numerator and denominator of the efficiency ratio for each step are also given, as well as the efficiency units. After the water arrives at the farm, it is stored or not stored depending on the farmer, and distributed to the fields for irrigation. For simplicity, we will combine the storage and on farm conveyance to the field into one step and call its efficiency farm efficiency (Efarm). The output is water at the field edge (Wfd) and the input is water at the farm gate (Wfg). The ranges of efficiency for the poor and good situations are also given in Table 1. Once the water is at the field edge, it is applied as irrigation to the crop in the field. The crop can only use the water retained in its root zone (Wrz), water that runs off the surface of the field or drains below the root zone represents losses. This step is well known in irrigation engineering and its efficiency is designated as application efficiency (Eappl). The output is Wrz, and the input, Wfd. Applying Eq. 2 to link the three efficiency steps described, as well as the subsequent five steps leading to crop yield to be described later, the whole efficiency chain and the overall efficiency are: Wfg Wvo

×

Wfd Wrz Wet Wtr m as m bm m yld m yld × × × × × × = = E all Wfg Wfd Wrz Wet Wtr m as m bm Wvo

(5)

53



Again, because the output of the preceding step is the input of the following step, all the terms on the left side of Eq. 5 cancel out except for the denominator of the first and numerator of the last efficiency. Note that the efficiency steps do not have to be all in the same units and can involve quantities of different nature. In this case the first five steps are all concerned with quantity of water (W), and the last two steps are concerned with mass of materials of different nature. Efficiency of the sixth step is the mass (m) of carbon dioxide assimilated per unit of water transpired by the crop. Units of each efficiency as used in this paper are given in Table 1. Table 1. Range of efficiencies for the steps in the efficiency chain from water diverted out of the reservoir to yield of annual grain (or fruit) crops. Two ranges are given, one for poor circumstances and practices, and the other for good circumstances and practices. Also given are the overall efficiency for the two situations, calculated from mid-values of the efficiency steps. The denominator of the efficiency ratio is the input, and the numerator, the output, for each efficiency step. Efficiency Poor circumstances and practices

Good circumstances and practices

Efficienc y step

Efficiency ratio

Econv

Wfg/Wvo

unitless

0.5 – 0.7

0.8 – 0.96

Efarm

Wfd/Wfg

unitless

0.4 – 0.6

0.75 – 0.95

Eappl

Wrz/Wfd

unitless

0.3 -0.5

0.7 – 0.95

Eet

Wet/Wrz

unitless

0.85 – 0.92

0.97 – 0.99

Etr

Wtr/Wet

unitless

0.25 – 0.5

0.7 – 0.92

6.0 – 8.0

9 – 14

0.22 – 0.36

0.4 – 0.5

0.24 – 0.36

0.44 – 0.52

0.0242

1.22

Eas

mas/Wtr

Units

-3

kgCO2 mwater

Ebm

mbm/mas

kgbiomass kgCO2

Eyld

myld/mbm

unitless

Eall

myld/Wrz

-3

kg m

-1

With the chain of efficiency steps fully written out in Eq. 5, we now return to describe the remaining steps (from the fourth step onward), which concern the plant and agronomic aspects. The fourth step is consumptive efficiency ( E et = Wet Wrz ), a measure of the proportion of water in the root zone removed by evapotranspiration (Wet). The loss of efficiency in this step is due to water left in the soil at harvest time. The next step is transpiration efficiency ( E tr = Wtr Wet ), a measure of the proportion of water taken up by the crop and transpired (Wtr), as distinguished from water evaporated from the soil. The next step is assimilation efficiency ( E as = m as Wtr ), a measure of the mass of carbon dioxide assimilated by photosynthesis (mas) relative to the volume of water transpired. The measurements here now include the mass of assimilated carbon dioxide as well as the volume of water. The next step is biomass conversion efficiency (Ebm), a measure of the plant biomass produced (mbm) relative to the mass of carbon dioxide assimilated. This efficiency is primarily determined by the chemical composition of the crop and is not easily changed. The last step is yield efficiency (Eyld), a measure of the proportion of plant biomass that ends up in the harvested yield (myld), and is equivalent to harvest index (HI), a well known parameter in the crop and agronomic literature. The most striking results (Table 1) of applying Eq. 2 or 5 to irrigated cropping is that the difference in overall water use efficiency (last line, Table 1) between the poor situation and the good situation is huge, in spite of the fact that for each efficiency step the difference between the two situations is not that large or even minor. Nonetheless, Eall for the poor situations is only 2% of Eall for the good situations. The reason for this huge difference lies in the multiplicative nature of the efficiency chain, 54



as already noted. This 50 fold difference in water use efficiency to produce yield (grain or fruits of annual crops) indicate that there is much room for improvement in many situations. It should also be noted that the comparison is not between the extremely poor and the extremely good situations, but between the mid-values of the efficiency steps for the two situations.

DETERMINANTS OF EFFICIENCY OF THE STEPS AND IMPROVEMENTS OF EFFICIENCY Some of the more important factors that impact the various efficiency steps are now discussed briefly, along with potential improvements that can be made at relatively low costs, and a sample improvement in the overall efficiency is calculated to illustrate the potentials. Starting with the first step of the chain of efficiencies, a poor Econv implies leaky conduits or open conveyance over long distance with much loss by evaporation. Improvement could be very costly (e.g., converting open channel to closed conduits) or at least more than nominal (e.g., repairing cracks widely spread along the conduit length). The next step efficiency, Efarm, is more amenable to improvement. A common cause for low Efarm is water leakage from unlined or poorly lined storage pond and conveyance ditches. Lining with plastic sheeting could be relatively inexpensive and could raise Efarm from poor to the good level in Table 1. The next step, Eappl, may also be improved at nominal cost. One common cause of low Eappl for surface irrigation is applying the water too fast or too slow relative to the infiltration rate of the soil and slope of the land, resulting, respectively, either in too much deep drainage at the head, or too much drainage at the tail end, of the field. Better control of the application rate to match the infiltration rate and slope should entail only minimum cost. For sprinkler irrigation, Eappl may be improved by avoiding irrigating under strong wind, and by pressurizing the sprinkle line adequately to ensure even water distribution. The next step, Eet, is already relatively high for the poor situation; improvement is more readily made in Etr. Low Etr is the result of too much soil evaporation relative to crop transpiration. Since soil evaporation is high when coverage of the ground by foliage canopy of the crop is low and when the soil surface is frequently wetted (Ritchie and Burnett, 1971), Etr is raised if the crop is planted more densely and more uniformly distributed over the soil to provide better canopy cover, and the soil is not irrigated frequently to minimizing wetting of the soil surface. The water transpired by the crop is in exchange for the cabon dioxide assimilated photosynthetically by the crop. Eas is generally higher for C4 species than C3 species, and higher if mineral nutrients, especially nitrogen (Steduto et al. 2005), are not deficient. Eas is also affected by evaporative demand of the atmosphere, being higher under cooler temperature and higher humidly (Hsiao, 1993b; Xu and Hsiao, 2004). If switching from a C3 to a C4 crop is not an option, It may be possible to change the planting time so growth of the crop takes place under the lower evaporative demand of the cooler part of the season. Better fertilization would improve Eas as well and the extra cost of the fertilizer may pay for itself by increasing yield in addition to enhancing Eas. Next step is biomass efficiency, Ebm. Because it is largely a function of the chemical composition of the crop, it is not easily changed except for the possibility of reducing respiratory loss of assimilates by growing the crop under a cooler temperature regime. The last step is Eyld, the ratio of harvested yield to the total crop biomass. Eyld has been improved considerably during the last century as the result of breeding for crops with higher yield. The higher yields turned out to be largely the result of partitioning more biomass to fruit or grain and less to vegetative parts (Evans, 1993). For a number of crops, the partitioning can be modulated by water status of the plant, and hence by irrigation scheduling. Unusually high water status induces more vegetative growth in many species and can reduce Eyld. Mild water deficit after the crop canopy is fully grown may improve Eyld, but very severe water deficit at pollination time would reduce it markedly. Moderate water stress also reduces Eyld during grain filling because of accelerated leaf senescence, especially if the crop is relatively low in nitrogen. These effects are more thoroughly discussed elsewhere (Hsiao, 1993a; Hsiao et al., 2005). It suffices to say that strategically better timed irrigation provides a means to improve yield efficiency (equivalent to harvest index) at a minimum or no additional cost. Just how much increase in Eall of the poor situation in Table 1 can be expected if some of the nominal or low cost improvements in the individual efficiency steps discussed above are carried out? Eq. 4 shows that if the improvement is only in one step, say a 55% increase in the efficiency of that step (Δ = 0.55), then the improvement in Eall is also 55%. This holds regardless which step is being improved. On the other hand, if improvements are made in a number of the steps and none of them are major, there would be marked improvement in Eall. To illustrate, the improved Eall is evaluated by applying Eq. 4 assuming the following: the original Eall is that for the poor situation in Table 1; Efarm is

55



increased 40% (Δ = 0.40) by lining the ditches but not the storage pond with plastic sheeting; Eappl is increased 37% by taking more care to regulate water application for the furrow irrigated field; Etr is increased 25% by reducing irrigation frequency somewhat while increasing the water applied per irrigation to ensure good water supply to the crop; and the other efficiency steps in the chain remain unchanged. The (1 + Δ) values for the improvements in the order given are: 1.4, 1.37, and 1.25, and their product is 2.4. That is, the new overall efficiency is now 2.4 times the original overall efficiency, and calculates out to be 0.058 compared to the original 0.0242 kg of yield per m3 of water. If some additional but still not costly improvements are made in the steps, Eall could be raised still much higher. For example, if the storage pond is spread with clay to reduce the porosity of the soil bottom and Efarm is increased by 78% as the result instead of only 40%, Eas is increased 19% by improved nitrogen fertilization, and Eyld is increased 24% by better control of irrigation to restrict leaf growth after canopy closure. The overall efficiency would be increased 4.5 fold in this case, to 0.109 kg of yield per m3 of water. Note that the overall improvement is marked although still much lower than that for the good situation in Table 1. The point is that a systematic and integrated approach must be taken to produce more crop per unit of water, by examining all the individual steps for potential improvements at nominal cost, and not just focus the attention on one or two of the step. That way limited resources can go a long way in improving water use efficiency.

APPLICATION TO OTHER PRODUCTION SYSTEMS AND ON LARGER SCALES Because the principle and equations are general, the chain of efficiency steps is application to any production systems as long as the steps in the production process are largely sequential. Water is of paramount concern in rainfed cropping systems in less humid areas. To apply this approach, the engineering aspects, from conveyance from the reservoir to placing water in the root zone, are replaced by a couple of efficiency steps involving infiltration of rain water into the soil and retention of the water in the root zone. From that point onward the steps are the same as those starting on line 4 (Eet) of Table 1. The concept is also valid for animal production. By adding animal production steps following the biomass step (for forage fed animal) or yield step (for grain fed animal), the final outcome is animal product instead of crops. These interesting applications are discussed elsewhere (in a presentation in the Rainfed and Drought session of this conference, and in Hsiao et al, 2005). The treatment here is confined implicitly to the local scale. In fact, the unit considered is a single field. For practical use, it is necessary to account for more complex situations such as a farm with a number of fields of different crops, or an irrigation district comprised of many farms and several distribution canals. These situations certainly make the calculations more complicated, but the principle and basic equations still apply. A way to integrate the basic equations for application at large scale has been worked out and is discussed in Hsiao et al. (2005). Another complication is the need to account for the use of recycled runoff and drainage water, also discussed in Hsiao et al. USE IN ECONOMICAL ANALYSIS The ability to quantify the contribution of improvement in any efficiency step to the improvement in overall efficiency makes this approach extremely useful. Different steps have difference efficiencies and the cost of their improvement also differ. Often the cost of raising a step efficiency to a top level is very high, but raising it to a modest level is low or moderate. Eq. 4 indicates that generally it is better to allocate resources to improve the steps with the lowest efficiencies, because the overall improvement is proportional to the fractional improvement of a step. So a given percentage improvement (e.g., 20%) in a low efficiency step (e.g., from 0.4 to 0.48) has exactly the same effect on the overall efficiency as the same percentage improvement in a relatively high efficiency step (e.g., from 0.8 to 0.98). When many step efficiencies are less than the good situation, how to allocate the limited resources for improvement among the steps is not simple and requires optimization. The approach here provides the quantitative fundaments for that process.

Acknowledgement

56



The intellectual development of the approach described here owns much to discussions with and suggestions from Dr. Pasquale Steduto. I also thank Prof. Elias Fereres for valuable input, criticism and encouragement.

REFERENCES Evans, L. T. (1993). Crop evolution, adaptation, and yield. Cambridge University Press, Cambridge. Hsiao, T. C. (1993a). Growth and productivity of crops in relation to water status. Acta Hortuculturae 335:137-148. Hsiao, T. C. (1993b). Effects of drought and elevated CO2 on plant water use efficiency and productivity. In: M. B. Jackson and C. R. Black (eds.) Interacting Stresses on Plants in a Changing Climate, NATO ASI series. Vol. I 16. Springer-Verlag, Berlin. pp. 435-465. Hsiao, T. C., Steduto, P, and Fereres, E. (2005) A Systematic approach to the improvement of water use efficiency. Irrigation Sci., in preparation. Ritchie, J. T. and Burnett, E. (1971). Dryland evaporative flux in a subhumid climate: II. Plant Influences. Agronomy J. 63:56-62. Steduto, P, Hsiao, T. C., and Fereres, E. (2005) On the conservative behavior of biomass water productivity. Irrigation Sci., in preparation. Xu, L.-K. and Hsiao, T. C. (2004). Predicted vs. measured photosynthetic water use efficiency of crops stands under dynamically changing field environments. J. Expt. Bot. 55:2395-2411.

57


58




ON THE CONSERVATIVE BEHAVIOR OF BIOMASS WATER PRODUCTIVITY 1

P. Steduto*, T.C. Hsiao**, E. Fereres*** * Division of Land and Water, FAO, United Nations, via delle terme di Caracalla, Rome, Italy ([email protected]) ** Department of Land, Air and Water Resources, University of California, Davis, CA, USA ([email protected]) *** Instituto de Agricultura Sostenible, University of Cordoba, Spain ([email protected])

INTRODUCTION Food production and water use are two closely linked processes. As the competition for water intensifies worldwide, water in food production must be used more efficiently. Of the different steps in water use in the crop production process, the most fundamental is the exchange of water lost by transpiration for the assimilation of carbon dioxide. The net gain of carbon and energy by the plant in this process then leads to the production of biomass. It turned out that for biomass production, the efficiency of water use is relatively constant after the variation in two key environmental factors, evaporative demand of the atmosphere and air carbon dioxide concentration, are accounted for by normalization. This conservative behaviour has been analyzed and discussed in detail several times in the past half century (e.g., de Wit, 1958; Tanner and Sinclair, 1983). In light of the urgent need to answer the question of how much the efficiency of water use in agriculture can be improved, and to further analyse the implications for agricultural systems sustainability (e.g., Fereres et al., 1993), we are revisiting the issue here to see how conservative biomass water productivity is and the extend of possible improvements. The conceptual basis for the conservative behaviour is reviewed and the ways to normalize for evaporative demand and carbon dioxide concentration illustrated. It is hoped that this discourse will help to focus better the potential means to improve the efficiency of water use, and also lead to a simple means of modelling crop productivity based on water use.

THEORETICAL FRAMEWORK AND EXPERIMENTAL EVIDENCE The focus of this note is biomass water productivity (WPb), also referred to as biomass water use efficiency (WUEb) in the literature. From an agronomic standpoint, it is the amount of crop biomass output per unit of water consumed in transpiration by the crop and evaporation from the soil (together, evapotranspiration). From a physiological standpoint, only the water transpired is considered because evaporation from the soil is not in exchange for carbon assimilated. Here, WPb is defined as the aboveground dry matter (kg m-2) produced per unit of water transpired (m3 m-2, or mm). Therefore, the units of WPb are g m-2 mm-1 or g biomass per m3 of water transpired (g m-3). Only above-ground biomass is considered in our discussion as for most crop species, except root crops, only a small portion of the total biomass is in roots and because there is a general homeostatic growth response towards a near constant root:shoot ratio. In developing the theoretical background and the appropriate framework for analyzing the constancy of WPb, we follow a stepwise scaling-up approach, from leaf to whole crop field, in the analysis of the two basic processes involved, water transpiration (T) and net carbon assimilation (A), and its conversion to biomass. At the leaf level, we define photosynthetic water productivity (WPp) as the ratio of leaf net carbon dioxide assimilation (Al) to leaf transpiration (Tl), both expressed as flux rates on a leaf area basis 1

The present note is abstracted from a paper under revision that will be published in Irrigation Science early in 2006. 59



(μmol m-2 s-1 for Al, and mol m-2 s-1 for Tl) and directly proportional to the gas gradient (CO2 for Al and vapour for Tl) and inversely proportional to the resistance encountered along the path (e.g., boundary layer, stomatal, metabolic). Plants have apparently evolved physiological mechanisms to keep the importation (from ambient air to leaf interior via stomata) and depletion (from leaf interior to the cellular carboxilating sites) of CO2 in balance most of the time so that the leaf-internal CO2 concentration (ci) is conservative. This implies that photosynthetic capacity and stomatal opening are coordinated and operate in concert in the leaf. This suggests that when one of the two opposing processes, either the importation or depletion, is perturbed, the other adjusts with some lag to keep the system in balance and ci nearly constant. There has been substantial experimental evidence showing that for many species, ci tends to remain constant under a range of conditions including temperature, radiation, water and salinity stresses, especially when the stress develops gradually, as it generally occurs in the field. The ample evidence of the tendency of ci to remain constant at a constant ambient CO2 concentration (ca), i.e. a constant ci/ca ratio, is an indication that stomata perform at the leaf scale in a manner that leads to a constant WPp.

WP c

p ) as the ratio of At the canopy level, we define, canopy photosynthetic water productivity ( canopy net carbon dioxide assimilation (Ac) to canopy transpiration (Tc). As we scale up from leaf to canopy, there are additional features that must be taken into account because the consideration is now on a land area basis instead of leaf area basis. The extent of radiation capture by a crop depends on the amount of leaf area, on the geometric arrangement of the leaves within the canopy, as well as on the angle and intensity of incident radiation. As is the case at leaf level, the process of Tc shares the same source of captured energy as Ac. Of the total captured solar radiation, though, only the fraction that is photosynthetically active (PAR) is effective in CO2 assimilation, while the whole spectrum is used for transpiration. PAR, however, is a fairly constant fraction of the incident solar radiation as is the ratio of absorptance of PAR to non-PAR radiation for the leaves of many species. Consequently, any change in the amount of radiation captured by the canopy would affect in

a similar way Ac and Tc so that also

WPpc

tends to remain constant.

At crop field level, the variable we want to focus on is the biomass water productivity (WPb). Changing from Ac to biomass requires an analysis of the respiratory costs in relation to Ac and of the chemical composition and carbon cost of the biomass. A constant WPb, then, would be expected only if the relationship between assimilation and respiration is also linear. This seems the case, provided that the composition of biomass does not change significantly. More and more evidence is appearing indicating an approximate fixed ratio of assimilation to respiration for crops (e.g., Albrizio and Steduto, 2003) where the reproductive organ has no high protein and/or oil content, such as soybean and sunflower. Constant WPb seems to be the case even under varying environmental conditions. Although WPb addresses situations where only aboveground biomass is considered, constant WPb has also been described for root and tuber crops.

NORMALIZATION OF BIOMASS WATER PRODUCTIVITY FOR CLIMATE To extrapolate water productivity values between climatic zones and between atmospheric CO2 statuses, there is a need to normalize them for the climate, specifically, for the evaporative demand of the atmosphere and for the atmospheric CO2 concentration, respectively. Ways of normalizing WPb for the evaporative demand of the atmosphere (Steduto and Albrizio, 2005) and for the atmospheric

WP *

p is CO2 concentrations (Hsiao, 1993) are expressed by Eq. (1) and Eq. (2), respectively, where the normalized value of WPb. In Eq. (1), the summation is over a total number of time intervals (n), with i being the running number designating each interval and ti the length of the interval (in days); Biomass denotes the gain in biomass from the beginning to the end of the summation period. In Eq. (2), the subscript “o” indicates the reference situation; the summation is over a number of days (n); Δw is the water vapour concentration difference between the leaf intercellular air space and the atmosphere.

60



n

Biomass WP = n ⎛T ⎞ t i ⎜⎜ c ⎟⎟ ∑ i =1 ⎝ E 0 ⎠i * b

(1)

WP = WPb ,o * b

∑ (c a ) i i =1 n

∑ (c i =1

)

a ,o i

n

∑ ( Δw i =1 n

)

o i

∑ ( Δw ) i =1

(2) i

CONCLUSION The implications that the near constancy of WPb has in the improvement of water productivity in agriculture cannot be overemphasized. The presented stepwise approach from leaf to the whole crop has provided a conceptual and theoretical framework to explain the basis for the constancy of biomass water productivity. An important implication of normalizing biomass water productivity is that it allows the comparison of water productivity data across the globe on equal footing, after accounting for differences due to variations in evaporative demand of the climate, and in atmospheric carbon dioxide concentration when applicable. Such comparisons will reveal more definitively the intrinsic properties of the crop or the management practices that alter such productivity. Most importantly, normalized WPb will provide a head start in knowing the WPb values at a new location or new time period when CO2 concentration is different, whether in the future or in the past. This offers an invaluable tool for modelling purposes, providing an effective way of extrapolating WPb values between different locations and seasons. Crop modelling based on radiation use efficiency (RUE), in fact, has a limited normalizing capability (Steduto and Albrizio, 2005).

REFERENCES Albrizio R. and Steduto P., 2003. Photosynthesis, respiration and conservative carbon use efficiency of four field grown crops. Agric. For. Meteorol., 116: 19-36. de Wit C.T., 1958. Transpiration and crop yields. Versl. Landbouwk. Onderz. 64.6 Inst. Of Biol. and Chem. Res. on Field Crops and Herbage, Wageningen, The Netherlands. Fereres E., Orgaz F. and Villalobos F.J., 1993. Water use efficiency in sustainable agricultural systems. International Crop Science, CSSA, I: 83-89. Hsiao T.C. (1993) Effects of drought and elevated CO2 on plant water use efficiency and productivity. In Jackson, M.D. and Black, C.R. (eds) Global Environmental Change. Interacting Stresses on Plants in a Changing Climate. NATO ASI Series. Springer-Verlag, New York, pp. 435-465. Steduto P. and Albrizio R., 2005. Resource-use efficiency of field-grown sunflower, sorghum, wheat and chickpea. II Water Use Efficiency and comparison with Radiation Use Efficiency. Agric. For. Meteorol., 130: 269-281. Tanner C.B. and Sinclair T.R., 1983. Efficient water use in crop production: research or re-search? In: Taylor, H.M., Jordan, W.A. and Sinclair, T.R. (eds). Limitations to Efficient Water Use in Crop Production. ASA, Madison, Wisconsin, pp. 1-27.

61


62




TECHNICAL INTERVENTIONS TO IMPROVE WATER USE EFFICIENCY IN IRRIGATED AGRICULTURE

A. Hamdy Professor Emeritus, CIHEAM - Mediterranean Agronomic Institute of Bari, Italy SUMMARY - The increasing scarcity of water in the dry areas of the Mediterranean is now a well recognized problem. High rate of population growth and economical development require continuous diversion of agricultural water to higher priority sectors. The need to produce more food with less water poses enormous challenges to transfer existing supplies, encourage more efficient water use and promote natural resources conservation. On-farm water use efficient-techniques if coupled with improved irrigation management options, better crop selection, appropriate culture practices and timely socio-economic interventions would help achieving this objective. In arid and semiarid countries of the region, water is a more limiting factor to production than land; hence, maximizing water productivity should have higher priority over maximizing yield in the strategies of water management. This implies that planning water and land use should be based on the comparative advantages of the dry areas, but, within the framework of maximizing the return from the limited available water resources. Achieving the greatest water productivity needed to resolve water shortage problems will not happen automatically. There is great need to find appropriate ways and proper tools for water saving and to achieve greater efficiency and equity in the irrigation system. Equally, procedures and practices for the assessment of the performance of irrigation must be improved with better management systems for water conveyance, allocation and distribution. Such issues, beside others, will be addressed in this paper, with major emphasis on the technical interventions that could be adopted in order to increase water productivity through water saving and the improvement of the rates in water use efficiency in irrigated agriculture. Key words: water use efficiency, irrigation.

INTRODUCTION The availability of freshwater is, today, one of the great issues facing the human kind, and in some ways the greatest, because problems associated with its availability affect the lives of millions of peoples, particularly those in the developing countries. In the Southern and Eastern countries of the Mediterranean, the agriculture sector is, by far, the largest water user. On a consumptive-use basis, 80 to 90% of all the available water resources are consumed by the agriculture sector. However, average losses in the irrigation projects suggest that only about 45% of the delivered or extracted water for irrigation actually reaches the crops. Very often, the conveyance losses of conduits (unlined canals or leak pipes) are much too large, a 30% losses of the available water is common in the irrigation systems. Another cause of inefficient water use is the emphasis on meeting the demand by constructing new supply facilities rather than improving the inefficiencies of the existing ones. Furthermore, for most countries of the Mediterranean, on-farm irrigation practices deliver significantly more water per unit area that what is required, leading to such low irrigation efficiencies. Up to now, water conservation and efficient use of water have not been given the attention they deserve. As agriculture is, by far, the largest water user, efficient irrigation management will undoubtedly be a major conservation option in the future. An increase in water use efficiency in the agriculture sector will serve a double purpose: it will free sufficient amounts of water for alternative industrial uses, while at the same time, it is quite possible to take away a certain amount without scarifying food production. In deed, it is possible to produce “more out of less”, but, this requires finding out the appropriate ways to achieve greater efficiency and equity in the irrigation systems.

63



Several issues were recommended as proper tools for the improvement of irrigation efficiencies; among those issues: lining canals, improvement of irrigation structure, modernization of irrigation systems, … etc. However, as irrigation efficiencies are basically ratios of volumes in the water balance of an irrigation system, concentrating mainly on the relationships between the components of the water balance of the irrigation system and the factors that might influence these relationships, it is not always realistic to expect increased efficiencies. There are other factors related to irrigation efficiencies such as management of the crops, socioeconomic and legal environment of the irrigation systems, capacity building in the irrigation sector and quality of the water. These factors, together with the ones related to the components of water balance, should be fully considered in all the programs and projects dealing with water use and efficiency improvement in the irrigation sector. Indeed, efficient use of irrigation water “more crop per drop” and improving the productivity of water in agriculture through an appropriate water management is not an easy process. Significant challenges still remain in the areas of technological, managerial and policy innovation and adaptation, human resources development, information transfer and social environment considerations. Our success and/or failure is a matter of our capability in finding sustainable solutions to the challenges we are facing.

WATER USE EFFICIENCY IN IRRIGATION Efficiency in the use of water for irrigation consists of various components and takes into account losses during storage, conveyance and application to irrigation plots. Identifying the various components and knowing what improvements can be made is essential to making the most effective use of this vital but scarce source in Mediterranean agricultural areas. Generally speaking, efficient water use is defined as the ratio between the actual volume of water used for a specific purpose and the volume extracted or derived from a supply source for that same purpose.

Ef = Functionally expressed, we have:

Vu Ve

where: Ef - Efficiency, dimensionless Vu - Volume utilized, m3 Ve - Volume extracted from the supply source, m3. In this case, water use efficiency refers exclusively to irrigation. In accordance with the definition proposed by the International Irrigation and Drainage Committee (quoted by Burmau et al., 1981) efficiency in the use of water for irrigation has several different components (Table 1). It is not redundant to insist on the fact that one can properly judge the hydraulic performance of irrigation only when both its efficiency and its effectiveness are assessed. The product of the three types of efficiency (Table 1) is the total efficiency of water use of irrigation Ei. Functionally expressed, we have:

Ei = Es Ec Eu

The term of irrigation efficiency, which is different from efficiency of water use in irrigation, is also used. According to Israelsen (1963), such efficiency is expressed as:

Ea =

Vt Up

where Vr is the volume of usable water stored in the exploration zone of the plants' root, Vp is the volume received by the plot and Ea is the irrigation efficiency. Also if we consider the volume of runoff loss and E and the seepage loss as D then this efficiency can also be expressed as follows:

64



Ea =

Up − E − D Up

Irrigation efficiency has several different components (Table 2). Losses in the distribution system are due to leakage and evaporation; losses during application to the field are due to wind, evaporation and runoff, and losses from the soil are due to excess water, applied beyond what the crop uses. Table 1 - Components of efficiency of water use in irrigation 1. Storage efficiency Es.

Es =

Vd – Ratio between Ve

The volume diverted for irrigation (Vd) and the volume entering a storage reservoir (Ve) for the same purpose. 2. Conveyance efficiency Ec

Ec =

Vp – Ratio between Vd

The volume of water delivered to irrigation plots (Vp) and the volume diverted from the supply source (Vd). 3. Irrigation efficiency Eu.

Eu =

Vu (* ) – Ratio between Vp

The volume used by plants throughout evapotranspiration process, Vu, and the volume that reaches the irrigation plot, Vp. * (Vu) is equal to the volume of evapotranspiration by plants minus the volume of effective rainfall. Table 2 - Irrigation efficiency factors 1 - Conveyance efficiency - ratio of water delivered to water diverted from source. 2 - Application efficiency - ratio of water reaching the soil to water diverted. 3 - Water use efficiency - ratio water available for the crop to water applied to soil.

IRRIGATION EFFECTIVENESS As defined above, irrigation efficiency is an indicator of how much water is being lost in the process of irrigation and nothing more. It is a usual and misleading habit to take irrigation efficiency as an indicator of the overall hydraulic performance of a system. There is a tendency to think that as long as efficiency is high, the system is performing well overall. However, one can easily get very high efficiencies if he under-irrigates over the entire field, in which case there is in fact a bad performance. Effectiveness is another criterion of performance in that it indicates the degree of achievement of the desired irrigation objective, consisting of replenishing the root zone after depletion by root extraction. In general terms, effectiveness is usually defined as the ratio of output achieved to the target or desired output. It is thus a measure of the degree of achievement of a fixed desired objective. In the case of irrigation, effectiveness can thus be defined as:

Ef (% ) =

Amount of water stored in the root zone × 100 Soil moisture deficit before irrigation

Ef is a parameter which may vary from 0 to 100% and is actually an indicator of the degree of under-irrigation since it falls below 100% in such a case. Walker and Skogerboe (1987) referred to this same criterion as water requirement efficiency and in other instances it is termed storage efficiency.

65



Another criterion which has also been used for judging the extent of under-irrigation and which is area-based rather than volume-based is the percentage of area adequately irrigated (Pa) which can be defined as:

Pa (% ) =

Area with root zone replenished × 100 Total irrigated area

The drawback of this criterion is that while it indicates the spatial extent of under-irrigation, it does not give any indication on the severeness of under-irrigation. One can thus theoretically have a root zone that has been replenished to a level of 95% over the entire field, in which case Ef will be 95% while Pa will be 0% since no point on the field would have had the root zone entirely replenished. Effectiveness is thus a better indicator of the fulfillment of the irrigation objective than is the % of area adequately irrigated.

GENERAL DISCUSSION AND CONCLUSIONS The way to water saving and whenever possible to its re-use, is still open. From the purely technical point of view, important water savings are possible, if one thinks that under realistic conditions water efficiency can vary depending on the cases, the modes as well as equipment, for instance from 75% to 25% about, one understands that moving from the former to the latter value means to triple the irrigable surface at equal water use efficiency, with technologies and methods available today, agriculture could cut its water demand by 10 to 15 percent. If one considers not only crop requirements and the pedological environment but also the fact that quite often the water saving techniques are labor-capital-energy-multiple-factors consuming techniques, on one hand, and the new concerns about environment and some social problems particularly related to the frequently low school levels of farmers, on the other hand, one understands indicate that the solution through water saving is not simple but a complex one. To achieve a sound use of water for irrigation with higher efficiency and better water saving requires, on one hand: • deeper scientific and technical knowledge which is still far from being perfect although using models which are hoped to be clearer and more and more widespread (Ait Kadi,1992); • a more systematic and permanent monitoring of supplying unbased data at reasonable cost, on the other hand; • a closer participation and collaboration of the whole technical environment and the involvement of the farmers in the implementation of the program for some tariffing criteria, or in the improvement of the modes of use of water (Abu Zeid, 1990-1992). A great equilibrium is necessary in evaluating the needs of the different user's groups and more flexibility and reconversion have to be ensured. As for the technical aspects again, it should be noticed that, however, irrigation is a tool used to maximize the general objectives of the whole agricultural management. These are complex objectives, with a strong although not unique economic component, quite different depending on the type of enterprise and scenarios, changing over time and space. It is then very superficial and misleading to think of optimizing the use of water resource by maximizing only irrigation water efficiency (in terms of money) as such, this could be a utopian ideal of a given hypothetical owner of irrigation water only. In practice some remarks are made: the first one is that the scenario where food demand is greater (countries with a high population rate) and the natural rainfall regime is more limiting (arid and semi-arid areas), the objectives of irrigation approach more the one of maximum water use efficiency; it is less so and more related to income, in the case of more developed countries. The second is that the irrigation efficiency is the result of all interactions with other production factors and this may need adequate investments also in other factors (for instance, fertilizers, new varieties, modern machinery, etc...).

66



The third remark concerns the need to have a team of well specialized technicians available for irrigation, and then also the need to ensure training and update knowledge of these cadres. It is a great effort and not only a technical scientific one. Cultural and social progress of the agricultural world should be promoted as much as possible. Operation and maintenance (O&M) is one of the most underestimated aspects of irrigation projects in developing countries. As a result, the efficiency of the project continues to decline, and during a crisis situation, generally the problem faced is more complex to resolve technically and more funds have to be extended than had the maintenance works been carried out on regular basis. Another issue worth noting is the fact that poor though O&M is for irrigation, it is generally even worse in drainage. The impending crisis of land and water scarcity by the turn of the century particularly in the arid and the semi-arid region of the Mediterranean countries implies that research should be oriented to cope with such alarming situation and to increase the productivity of available land and water resources in a sustainable manner. In this regard, the following teams of research are suggested: 1) - Modern Irrigation methods made appropriate to the least developing technological context: irrigation techniques have to be adapted and made appropriate to self-sustained technological improvements, taking into account economic, technical, maintenance and cultural aspects. Adaptive research is needed to modify the hardware of modern methods to permit local construction and engineering use of local and low cost materials and skills. The process will also help in enabling users to master the technologies and facilitate easier maintenance. Similarly the software modern methods would also need adaptation to accommodate cases where data are inadequate or unreliable. Another topic which should be carefully considered is the regulation of land networks; its sequence regulation of water deliveries not only has the direct advantage of rising the crop yield but also contributes to control waterlogging and salinity, beside other advantages such as the reduction of seepage losses and sedimentation, control of disease vectors and the reduction of weed growth. The introduction of pressurized irrigation systems as a tool for a better water saving and a better economy in fertilizer use and less pollution of ground water, is among the points that should receive the attention of researchers. Research is required to adapt such methods to specific soil, topographic ecological and social conditions in order to ensure cost-effectiveness and sustainability. A part of the research should be devoted to the rehabilitation and modernization of large systems, with the objectives of greater decentralized operations, improved water conservation, and reliable and equitable water delivery to self-managed groups. Systems now in operation will have to meet the challenges of water scarcity and hence of conservation; moreover, rapid changes in all aspects of the economy and in agriculture, call for systems that offer flexibility in crop production. 2 - Decision support at various levels: the potential decision support system (DSS) as planning, policy-information and operational tools at a national, regional and farmer level is becoming evident. At national and regional level, the combined use of models, remote sensing and geographical information system can help predict the effects of different policy scenarios and allow better management of existing land and water resources. At the farmer's level, DSS can help to improve crop production and irrigation efficiency. Existing models should be further developed, improved and tested in different situations; also new models are required. At the local level, research should focus on the operational use of models. Research is needed with respect to parameter estimation for different crops and soils and model calibration and verification in combination with field tests. Models should be adapted to the users by making them user-friendly (generation of input data, understandable output, etc.). Furthermore, attention should be given to the introduction of water quality aspects especially salinization into the soil-vegetationatmosphere (SVAT) model. 3 - Sustainable use of marginal quality water: there is increasing awareness of the need to augment water supply and improve water quality through the use of the treated sewage and drainage water.

67



For a sustainable use of drainage and brackish water for irrigation, research work is needed on the physical parameters governing the flow of water and salt in the irrigated soil profile with an estimation of the risk of salinization through modeling and testing improved methodologies to reduce salinity through drainage. Research should be also directed towards the establishment of new strategies including different crop rotations based on proper soil, water and crop management that favor the reuse of drainage and brackish water in irrigation without deterioration either in crop production or soil productivity. Adaptive research programs should be planned on the recycling of used water which could lead to a great deal of progress, but which is still limited by the lack of research in this field. The approach appears to be promising and will enable the impact of scarcity to be minimized in times to come. A part of the program should include the waste water technologies and methodologies to achieve the most appropriate ones to be used in irrigation to produce an effluent which meets the recommended microbiological and chemical quality guidelines, both at low cost and with minimum operational and maintenance requirements. An important aspect of the use of marginal quality water for plant production is the possibility of ground water contamination with undesiderable pollutants. Further basic and applied research is needed to preserve and/or to improve the quality of ground water from serious contamination. Lack of knowledge about long-term effects of treated waste water has to be searched; this will greatly encourage and facilitate its use in irrigation on a wider scale. 4 - Performance assessment, monitoring and evaluation of irrigation and drainage systems: one main reason for poorer economic performance of irrigation projects is deficiency in project planning and management. The result is a restricted budget for operation and maintenance, crucial the integrity and sustainability of an irrigation system. Research is needed to formulate a methodology for the collection, storage and processing of administrative, technical and environmental data required for an effective and efficient management of soil and water resources. The conveyance of information from evaluators to planners and designers creates a positive feedback mechanism leading to a better design which is another important area of research. 5 - Water optimization: crop water requirements and irrigation scheduling; efficient irrigation: in order to optimize water application by irrigation to different crops under different soil types, irrigation systems and climatic conditions, research should be continued to provide better knowledge of soilwater-plant relationships, reviewing the concept of an optimal water supply. The management of irrigated crops to cope with droughts should receive priorities in research. In addition the link between crop water requirements and irrigation schemes reliability should also be considered. 6 - Environmentally sound practices for sustainable land and water use: new approaches and methods are needed to solve some of the old problems in irrigation. These approaches need to be cost-effective and bearable by farmers in order to be ecologically and economically sustainable. Many fields of knowledge including engineering, biological science, and social science need to be drawn on the development of these approaches and methods. Research is needed in areas like: control of soil erosion and soil fertility conservation, water conservation practices, economic criteria for irrigation and drainage of heavy saline and/or other problem-affected soils; also alternative crops and cropping systems better suited to problem-affected soils and other ecological constraints; integrated pest management and pest control; and development of irrigation management practices to cope with drought and aridity, including water conservation practices and the reuse of drainage and waste water for irrigation. 7) Socio-economic studies: management and design interaction in relation to community dynamic and local managerial capability: The design of schemes has traditionally been the domain of irrigation engineers whose concern has been to optimize technical efficiency of water use. The projected returns to such schemes are too often unrealized because they rely on an unrealistic precision in water allocation and water use across the design of irrigation area. Efforts need to be made to design schemes whose layout is simpler to be operated and managed by farmers and their communities. The challenge is for engineers to work with local farmers at whatever level of technology is appropriate, and design more manageable schemes to economize, where possible, to compare real social and economic impact in relation to the money spent as between simpler and more complex designs under similar conditions.

68



REFERENCES Abu Zeid, M. (1993). Irrigation Cost Recovery In Developing Countries. Cahiers Options Méditerranéennes. Vol. I, n° 1 Water Resources: Development and Management in Mediterranean Countries. pp. 13.1-13.8. Abu-Zeid, M. (1990). Some Technical and Economic Considerations on Irrigation Water Pricing. Water Science Magazine, Issue n° 7, Cairo, Egypt. Ait Kadi, M. (1993). The Application of Optimization Techniques to Water Resources. Cahiers Options Méditerranéennes. Vol. I, n° 1 Water Resources: Development and Management in Mediterranean Countries. pp. 9.1 - 9.15. Arreguin, F. (1991). Efficient Water Use in Cities and Industry. Proceedings of the International Seminar on Efficient Water Use, Mexico, Oct. 1991, pp. 749-756. ASCE (1974). Water Management through Irrigation and Drainage: Progress, Problems and Opportunities. Committee on Research of the Irrigation and Drainage Division, ASCE, IR2, 1974. Bau, J. (1991). Research on Water Conservation in Portugal. Proceedings of the International Seminar on Efficient Water Use. Mexico, Oct. 1991. pp. 736-743. Burman, R.D.; Nixon, P.R.; Wright, J.L. and Pruitt, W.O. (1981). Water Requirements. In: Design and Operation Farm Irrigation Systems. Eds. M.E. Jensen, ASAE, St. Joseph, Michigan, 1981. Caswell, M.F. (1989). The Adoption of Low-Volume Irrigation Technologies as a Water Conservation Tool. Water International, Vol. 14, n° 1, International Water Resources Association, 1989. CEMAGREF, CEP, RNED-HA. (1990). Irrigation-Guide Pratique. Montpellier Falkenmark, M. and Widstrand, C. (1992). Population and Water Resources: a Detailed Balance. Population Bulletin, Population Reference Bureau. FAO (1992). Wastewater Treatment and Use in Agriculture. (ed) Pescod, M.B. Irrigation and Drainage Paper. 47 Rome. Gleick, P.H. (1993). Water in a Crisis: Guide to the Worlds Fresh Water Resources. Oxford University Press. Gloss, S. (1991). The Legal and Institutional Conundrum of Efficient Water Use in the Western United States. Proceedings of the International Seminar on Efficient Water Use, Mexico, Oct. 1991, pp. 523-530. Horst, L. (1983). Irrigation System - Alternative Design Concepts, Irrigation Management Network Paper 7c April. Overseas Development Institute. International Commission on Irrigation and Drainage (ICID). (1989). Planning the Management, Operation and Maintenance of Irrigation and Drainage Systems, World Bank Technical Paper Number. Lyle, W.M. and Dixon, D.R. (1977). Basin Tillage for Rainfall Retention. Trans.Am.Soc. Agri.Engr. 20, pp. 1013-1017 & 1021. Lyle, W.M. and Bordovsky, J.P. (1983). LEPA Irrigation System Evaluation. Transactions of American Society of Agricultural Engineers 26, pp. 776-781. Merriam, J.L. (1977). Efficient Irrigation. California Polytechnic State University. San Luis Obispo, California. Nayan, S.; Pandey, A.D. and Debajyoti, C. (1991). Feasibility for Canal Automation to an Existing Canal -a Case Study. Proceedings of the International Seminar of Efficiency Water Use, Mexico, Oct. 21-25, 1991. Norum, E.M. (1991). Irrigation System Selection: Secondary Considerations. In: Irrigation and Agricultural Development, FAO/Pergammon Press, 1980. Phene, C.J.; Howell, T.A. and Sikorski (1985). A Travelling Trickle Irrigation System. In: Advances in Irrigation. Vol. 2 (Ed) Daniel Hillel Academic Press, London, pp. 2-49. Postel, S. (1986). Effective Water Use for Food Production. Water International. n° 11, 1986.

69



Saffaf, A.Y. (1980). Selection of Appropriate Irrigation Methods for Semi-Arid Regions. In: Irrigation Agricultural Development, FAO/Pergammon Press, 1980. Von Bernuth, R.D. and Gilley, J.R. (1985). Evaluation of Centre Pivot Application Packages Considering Droplet Induced Infiltration Reduction. Trans. Am. Soc. Agr. Engr. 28(6), pp. 19401946. Walker, W.; Ruchardson, S. and Sevebeck, K. (1991). A Comprehensive Approach to Water Conservation. Proceedings of the International Seminar on Efficient Water Use, Mexico, Oct. 1991. pp. 763-769. World Bank (1992). World Bank Development Report. Oxford Press. New York.

70



71



COUNTRY REPORTS

72



73



TECHNIQUES FOR IMPROVING WATER USE EFFICIENCY IN GREENHOUSE CULTIVATION IN CYPRUS

P. Polycarpou, D. Chimonidou, I. Papadopoulos Agricultural Research Institute 1516 Nicosia, Cyprus

SUMMARY - In Mediterranean region, there is an increasing concern about the effective and efficient utilization of water for agriculture and water conservation in general. The promotion of effective water use and on farm water management, were identified as an important contribution to the management strategy needed to address problems of water scarcity and practicing intensive agriculture on environmentally sound grounds. Recently particular emphasis was laid on protected cultivation and more specific on cultivation of vegetables and flowers on substrates and soilless cultures (closed systems and open with minimum drainage). New unites with soilless cultivation (mainly perlite, coconut and rockwool) have been established applying modern greenhouse technology and fully computerized irrigation-fertigation methods. The only way to increase productivity, improve quality and control growth and production, is through the application of modern greenhouse technology, new techniques of cultivation and integrated protection and production management. At the Agricultural Research Institute the use of local materials i.e. perlite, mixtures of perlite with pomace, almond shells, pine bark, gravel, etc. have been tried successfully. In this paper, results of the application of modern techniques, hydroponic cultures, re-circulation of irrigation water and nutrient solution in closed systems and control of the climatic conditions in the greenhouse (temperature, air humidity, CO2, etc) will be discussed. The introduction of modern technology and soilless culture in greenhouse cultivations (vegetables and flowers) resulted in higher production, better quality, efficient and effective use of water and fertilizers and minimize the use of chemicals for pest and disease control. The use of closed recirculation systems has reduced the water needs of the cultivations close to the evapotranspiration levels of the crop. Ongoing research on using a green lagoon to denitrifigate the reject water from the closed system, when the undesired elements in it reach toxic levels, seems to be very promising. The grown Sudan-grass in the lagoon can be used as animal feed or as an energy plant. Key words: soilless culture, hydroponic systems.

INTRODUCTION In Mediterranean region, there is an increasing concern about the effective and efficient utilization of water for agriculture and water conservation in general. The promotion of effective water use and on farm water management, were identified as an important contribution to the management strategy needed to address problems of water scarcity and practicing intensive agriculture on environmentally sound grounds (Papadopoulos and Chimonidou, 1997, 2004). In Cyprus, the irrigated land amounts to 35000 ha (16.2% of the total agricultural land) with provision to be expanded. The irrigated agriculture in semi arid countries like Cyprus demands large amounts of water and faces the serious challenge to increase or at least sustain agricultural production while coping with less and/or lower quality water (Papadopoulos and Chimonidou, 1997, 2004). Recently particular emphasis was laid on protected cultivation and more specific on cultivation of vegetables and flowers on substrates and soilless cultures (closed systems and open with minimum drainage). New unites with soilless cultivation (mainly perlite, coconut and rockwool) have been established applying modern greenhouse technology and fully computerized irrigation-fertigation methods (Chimonidou 1999, 2003a, 2003b).

74



The demand for increased production and better yield quality, the lack of good quality water for irrigation and the need for protection and conservation of the environment require the implementation of new technologies in greenhouse cultivation. In this respect, the first measures to be taken are improvement of greenhouse structures and automation of the systems. Moreover, modern techniques have to be applied such as cultivation on artificial substrates, hydroponic cultures, re-circulation of irrigation water and nutrient solution in closed systems and control of the climatic conditions in the greenhouse (temperature, air humidity, CO2, etc). Apart from higher production of good quality (10 %25 % yield increase in vegetables and up to 30 % in flowers), such technology would result in efficient and effective use of water and fertilizers and in minimizing the use of chemicals for pest and disease control, especially after the prohibition in using Methyl Bromide for soil sterilization. Moreover some cultural practices, like soil cultivation and weed control are avoided, and land not suitable for soil cultivation can be used (Polycarpou and Hadjiantonis 2004).

SOILLESS CULTURE Recently particular attention was given in soilless cultivation and the area under soilless culture is rapidly expanding. The first soilless culture in Cyprus started with rose cultivation on rockwool in 1996 at the area of Monagrouli (Limassol) with 0.2 ha and expanded in 0.4 ha in 1997 with a fully automated computerized open system. Since then, the cultivation of roses on substrates has rapidly expanded. The application of soilless culture in vegetables (tomatoes, cucumber, lettuce, strawberries) came later due to the low and unpredictable market prices of the products. The total area cultivated today in Cyprus using soilless techniques is about 11.6 ha. A break down of this area is given in Table 1. This area is expected to increase rapidly in the near future due to the grants schemes announced by the Ministry of Agriculture as a measure to improve the quality of the agricultural products as a result of Cyprus joining the European Community. Soilless culture has some fundamental differences compared to growth on soil. Unlike soil culture, soil1754less cultivation starts with media free of soil born diseases and gives the grower flexibility to control growth by manipulating water and nutrient supply. The substrate can be chosen to have the desirable characteristics so that with the appropriate water and nutrient management yield can be maximized and quality be improved. Table 1. Soilless Cultivation in Cyprus (Source: Department of Agriculture)

Area of Soilless Culture in Cyprus (ha) Flowers (Roses & Cerbera)

Vegetables

Perlite

0.3

0.2

Coco Peat

3.8

1.6

5.4

Rock wool

0.7

5.0

5.7

Substrate

TOTAL

Total

0.5

11.6

However, there are some difficulties in soilless cultivation like: ¾ High initial installation cost ¾ High skill requirements on growers ¾ Sensitive system with no buffering capacity of nutrients – No error tolerance. ¾ High water quality requirements. Risk for environmental pollution if not properly managed.That is the reason for the slow application of the method until now.

75



EXPERIMENTAL WORK ON THE CULTIVATION ON SUBSTRATES Roses on soilless cultivation A lot of attention was given on the research on roses and experiments started since 1990 on the physiology of roses, the critical levels of development in relation to water stress and the effect of irrigation, shading and salinity on yield and quality (Chimonidou, 1996,1997,1998). Experiments on soilless cultivation of roses started in 1997 using different substrates i.e. perlite, perlite (70%) + pomace (30%), Coconut, 100% rockwool and pine bark (70%) + straw (30%) as inert media. The aim was to compare different inert media (imported and local) and evaluate the cultivation in bags and containers using the technique of shoot bending. Irrigation based on the moisture content of the media, is kept constant at -5 to -8 kPa at the area of the root zone. The total amount of irrigation water was the same but the frequency was different according to the holding capacity of each substrate. The productivity and quality characteristics (stem length, fresh weight, flower bud diameter and height) have been recorded. Results were very encouraging for the local substrates tested. Experimental work on the cultivation of Roses on soilless culture expanded in fully automatic greenhouses at different locations aiming at the use of new substrates to face the problem of low quality waters in Mediterranean climates using the minimum drainage and accumulate the salts at the periphery of the container. A joint programme between the Agricultural University of Athens - Greece and the Agricultural Research Institute of Nicosia - Cyprus (2001-2004), aimed at studying the development and photosynthetic activity of roses cultivated in four different substrates and two irrigation regimes. Roses cv "Eurored" were cultivated on four different substrates in a heated greenhouse at the ARI using local materials i.e. perlite 100%, mixtures of perlite 50% with pomace 50%, perlite 50% with almond shells 50% and almond shells 50% with pine bark 50%. The two irrigation regimes applied, were 800ml (6 times/day X 2 min) and 530ml (4 times/day X 2 min). The photosynthetic rate, stomatal conductance, CO2 concentration and transpiration rate of the rose plants grown in the four substrates and under the two irrigation regimes were measured as well as the total productivity and quality characteristics (stem length, fresh weight) of the roses produced. Results showed that concerning the interaction between substrate and irrigation level, higher production was recorded with the roses growing in the substrate pine bark 50% and almond shells 50% irrigated with the reduced irrigation level and the substrate of perlite 50% and pomace 50% irrespective of irrigation level. The quality characteristics of the roses produced under all treatments were marketable with mean stem length between 75-85cm. No significant differences were observed concerning the photosynthetic rate, the stomatal conductance, the CO2 concentration and the transpiration rate under the four substrates and the two level of irrigation. It seems that the lower irrigation level did not create conditions of water stress and did not affect negatively the physiological activities of the rose plants. Concluding remarks showed that the local substrates could be used successfully as substrates for the rose cultivation in the region.

Cultivation of Lysianthus (Eustoma) on substrates Two different experiments on Lysianthus were conducted during the years 2002-2003 (at the Agricultural Research Institute and at Zygi experimental station), aiming at higher productivity and year round production. The first experiment in cooperation with the Aristotle University of Thessaloniki was aimed at evaluating the productivity and quality characteristics of Eustoma grandiflorum on two substrates and two irrigation regimes. The substrates used were perlite 70% with coco 30% and perlite 50% with pomace 50%. The irrigation was performed using drippers of 4l/h and the irrigation intervals were: 6

76



times x 2 min (800 ml/ day) and 4 times x 2 min (530ml/ day = reduction 33%). The pH and the EC were kept constant at the levels of 6,5 and 1,7-1,8 DS/m respectively. Drainage for both cases was only 5%. Results showed that no significant differences were existed between the different substrates or the stressed and not stressed plants with respect to the total productivity (number of stems) or the quality characteristics (number of flower buds, stem length and fresh weight) of Lysianthus. Vase life of the plants was not affected by the cultivation in different substrates. On the contrary, the plants under the low level of irrigation lasted more days in vase with or without preservative (Chimonidou et al. 2003).

Greenery Cultivated in Different Substrates Experiments on the cultivation of flowers and greenery on substrates started in 1995 with the Greenery Rumohra adiantiformis cultivated on four different substrates in comparison to the cultivation on soil in an unheated greenhouse at Zyghi experimental station. The aim was to increase productivity improve quality (particularly stem length with increasing shading levels) and avoid soil born diseases by improving soil structure and aeration of the root zone using local material (i.e. pine bark, mushroom compost, pomace etc.). Results show that pine bark 30% with peatmoss 70% appeared to be the most promising substrate among those tested during 1995-97, concerning their quality characteristics (Chimonidou, 1999). The experiment was extended in 1998 with the cultivation of greenery and other cut flowers on soilless cultivation. The species Limonium ottolepis, Rumohra adiantiformis, Gerbera jamesonii, and Pteris vitata, were tested in a mixture of perlite 70% and pomace 30% at the Agricultural Research Institute in Nicosia. The total productivity and the quality characteristics (stem length and fresh weight) as well as their keeping quality after harvest were studied. Results appeared to be very encouraging for all species tested and specially for Limonium ottolepis, concerning the total productivity and quality characteristics in comparison to previous research work carried out on soil (Chimonidou 1999).

HYDROPONIC SYSTEMS The open system for soilless culture, Fig.1, is at present most favored commercially in Cyprus due to its simplicity, mainly in managing the nutrient solution. Pollution of the environment (underground water), waste of fertilizers and water are though only some of the problems faced in open hydroponic systems. The leachate is usually collected in a reservoir and is used for the fertigation of open cultures or greenhouse cultivations in the soil. This results in approximately 30 % loss of fertilizers and water from the system. For this reason ARI started a research program in order to develop a locally adopted closed hydroponic system (Fig. 2), using locally available inert substrates, like crashed gravel produced in a copper mine in Cyprus. The leach ate from the substrates is collected in a tank and is recirculated after being sterilized passing through a UV lamp. The EC and pH of the water are regulated using an automatic fertilizer-mixing unit as by the open system. The water consumption of a good managed closed system is reduced to the evapotranspiration level of the plants. The system requires water of very good quality that is difficult to find in Cyprus. At the coastal areas where greenhouse cultivation has developed due to the favorable climatic conditions, the ground water salinity ranges from 1.5 to 4 dS/m, whilst the salinity of water coming from dams is around 1 dS/m. The fresh water supplied to the closed system can be therefore rainwater collected from the greenhouses or water treated by a small reverse osmosis unit. Thus the need for replacing the nutrient solution due to the increasing concentration of chlorides and sodium is minimized. When the nutrient solution comes to the point that it has to be replaced, the possibility is studied to pass it through a green lagoon planted with sudan-grass is studied. By this method the water is denitrificated and can be safely disposed to the environment. The sudan- grass harvested from the lagoon could be used for animal feed or for the production of bio-mass for energy purposes. The experiments are carried out at the ARI research station at Zygi on tomato cultivation (Polycarpou and Hadjiantonis 2004).

77



Reject water

Leachate Tank

Fertilizer Mixer TANK A

AC

B

Fresh Water

Fig. 1. Schematic diagram of an open hydroponics system applied commercially in Cyprus. In addition, an open system using a mixture of locally available organic materials with perlite or peat moss as substrate is being studied in floriculture. In this “zero loss” system the nutrient solution is supplied to the plants, planted in big boxes (substrate volume 15 liters/plant), in such a quantity that leaching just starts. In this way the water and fertilizer loss from the system is minimal. The salts are pushed by the irrigation water away from the root zone and are accumulated in the outer volume of the substrate not affecting the growth of the plants.

UV Leachate Tank

Fertilizer Mixer TANK A

B

ACID

Fresh Water

Fig. 2. Schematic diagram of a closed recirculating hydroponics system applied at the ARI in Cyprus. In designing and operating such a closed hydroponic system the following main parameters are to be considered:

78


¾


Crop related matters such as the life span of the crop, the water and nutrient requirements (recipe), and the cultural practices needed.Method for fertilizer mixing and supply of irrigation water (Using simple volumetric fertilizer injectors or automatic fertilizer mixing units).

Use of locally available inert substrates like perlite, coarse sand, crashed gravel vs. imported inert materials like rock wool. ¾ Climate Control in Greenhouses, like monitoring the aerial climate requirements (temperature, relative humidity, light, CO2, etc), the root zone requirements (root temperature and O2 supply in the root zone) and improving the PAR transmission of covering materials and lowering their NIR transmission. Due to the advantages of the closed hydroponic system compared to the open one, ARI is investing a lot of effort in optimizing its parameters, simplifying its operation and training the growers in its effective management and utilization (Polycarpou and Hadjiantonis 2004). ¾

CONCLUSIONS •

In Mediterranean countries where water is limited and of high cost, diversion to intensive irrigated agriculture, protected cultivation and soilless culture are promising alternative and innovative approaches.

•

Soiless culture using locally available substrate materials (perlite, mixture of perlite + peat, pomace, pine bark etc) could be the solution. Experimental results so far in terms of yield, quality and water use efficiency are very encouraging.

•

It is important that a suitable closed system is developed that is based on low cost local materials, which are both effective and easily disposable after use in order to avoid environmental pollution. The system should be easily adaptable to the growers according to their potential skills. The technology used should be locally supported in order to avoid longterm maintenance problems.

REFERENCES Chimonidou-Pavlidou Dora. 1996. Effects of water stress at different stages of rose development. Acta Horticulturae 424:45-51. Chimonidou-Pavlidou Dora. 1997. Use of saline waters for irrigation in Cyprus: New developments and Management practices. In Proceedings of the International Conference on “Water management, salinity and pollution control towards sustainable irrigation in the Mediterranean Region”. Italy, 22-26 Sept.,1997.p. 21-33. Chimonidou-Pavlidou Dora. 1998. Yield and quality of rose Madelon as affected by four irrigation and three shading regimes. Acta Horticulturae 458:95-103. Chimonidou-Pavlidou Dora. 1999. Protected Cultivation and Soilless Culture in Cyprus. First meeting of FAO thematic Working group for Soilless Culture, August 31- Sept. 6, 1999, Halkidiki, Macedonia, Greece. Chimonidou Dora 2002a. Flowers under protected cultivation with special emphasis on roses and new cut flowers. FAO/AUB First National Conference on Integrated Production and Prodection Management of Greenhouse Crops, p.105-114. Chimonidou Dora, 2002b. Country report of Cyprus. In Proceedings of the Regional Experts Meeting on Flowers for the Future. Izmir-Turkey, 8-10 Oct,2002, p.15-30. Chimonidou Dora, M. Stavrinides, I. Papadopoulos and P. Polycarpou, 2003. Intensive cultivation of floricultural greenhouse crops. Proceedings of the Conference on Greenhouse Crop Production in the Mediterranean Region. Nicosia, Cyprus 10th November 2003. Papadopoulos,I. and Chimonidou Dora, 1997. Nutrient and Agro- chemical management for pollution control under intensive irrigated agriculture. In Proceedings of the International Conference on "Water management, salinity and pollution control towards sustainable irrigation in the Mediterranean Region". Italy, 22-26 Sept, 1997. p. 45-65.

79



Papadopoulos I. and Chimonidou Dora, 2004. Participatory Water Saving Management and Water Cultural Heritage: Cyprus Country Report. Option Mediterraneennes Series B, No: 48, p. 97-111. Polycarpou P. and Hadjiantonis Ch.,2004. Cyprus Country Report. FAO Regional Training Workshop on Soilless Culture Technologies. Izmir-Turkey, 3-5 March 2004, p.12-15.

80



81



REVIEW AND ANALYSIS OF WATER USE EFFICIENCY AND WATER PRODUCTIVITY IN EGYPT

M. Nasr Allam* and R. Abdel-Azim** * Prof., Head of the Irrigation and Hydraulics Dept., Faculty of Eng., Cairo University, Egypt ** PhD, Ministry of Water Resources and Irrigation, Egypt SUMMARY - This paper presents analysis of water resources availability and water uses in Egypt. More focus will be given to agricultural water use and productivity since agriculture sector represents the major water-consuming sector in Egypt. Nile River is the major source of water where rainfall is rare and groundwater is limited. The nature of the irrigation network and system of the Nile river is rather unique. Water lost from one point is usually used in the downstream, and hence the global water use efficiency is relatively high. Moreover, there is a multiple system for water use in Egypt. Water is used for agriculture, municipal and industry, navigation, hydropower generation, and fisheries. Emphasis in this paper is given to agriculture water use and productivity. Agriculture sector consumes about 85% of the total available water. There are several applied policies for improving water use efficiency. Irrigation Improvement project is currently implemented to enhance water distribution and minimize water losses through physical and institutional measures. Participatory approach is also being implemented in water management followed by modification of laws and institutional reform. Future water policies include introducing integrated water management approach to increase water use efficiency and maximize water productivity. The paper presents the productivity of the main crops in Egypt and water crop productivity compared to some international figures. The results show that crops have a relatively high productivity particularly rice crop which has the highest productivity over the world. Key words: water resources, cropping pattern, crop production, Egypt.

INTRODUCTION Water, always, plays an essential role for providing the basis of population stability and civilization. The Nile River in Egypt has supported the longest civilization over the world which lasted more than seven thousands years. Egyptians, throughout the history, were skillful enough to utilize the Nile water. During this century, they installed an invaluable water structure; High Aswan Dam (HAD). This dam helped in providing more controllable water releases pattern over the year, serving the Egyptian population who are living on the small batch along the river. Egypt’s annual share of Nile water that controlled by the dam is 55.5 billion cubic meters as stipulated and agreed upon between Egypt and Sudan in 1959. After the construction of the HAD in 1986, the cultivated area has been expanded to reach 8.0 million feddans, which is cultivated about twice a year. The government of Egypt continues to invest heavily in expanding the cultivated area, and is planning to add another 3.4 million feddans of the cultivated land by year 2017. Water Supply augmentation has been practiced in Egypt since several decades through recycling the drainage water and use of shallow groundwater. Drainage reuse started in 1970s and reached now a level of 4.0 billion cubic meters annually and it is planned by the government to increase the reuse up to 8.0 billion cubic meters annually. Simultaneously, the groundwater withdrawals are planed to increase from 4.0 to 7.0 billion cubic meters per year in Nile valley and Delta. The groundwater aquifers in the Nile Delta and Nile valley are replenished from the leakage water from the river, irrigation canals and drains. Therefore water supply augmentation through recycling (drainage water reuse) and capturing of the water losses (groundwater utilization) increases the overall efficiency of the water resources system in Egypt. Although the water requirements of municipal and industrial sectors have the first priority to be met, the agricultural policies and innovations had its impact and pressure on the operation of the Nile water system. During the last decade the government of Egypt has liberalized the cropping pattern

82



and farmers became free to select the crops they like to grow, except for rice which is restricted to the permission of the government. The Nile system is viewed as two sub-global systems; Nile valley and Delta. In Nile valley, drainage water returns back to the river and drainage reuse practices are limited and occur only on a very small scale. Delta region is the potential area for drainage reuse promotion. The linkage between the different hydrological areas, i.e. between global or sub-global and the local levels, is the water flows and change in salts concentration. The water and salt balance analyses are, therefore, viewed essential for evaluating these levels. At present, operation of the Nile system is successful in meeting the current water demands. However, Egypt must do more with less water (Abu Zeid, 1997) to cope with future development plans for the country and with projected future increase in population. Egypt has introduced different innovations to the existing system in order to save water from old lands to be diverted to the new lands. These innovations included Irrigation Improvement Project which started in 1980’s, and drainage water reuse program which started in 1970’s. The irrigation improvement project activities include improvement of secondary and tertiary irrigation delivery network and leveling of agricultural fields. It is expected that this project will result in saving irrigation water and improving the agricultural productivity. Drainage water is one of the valuable water resources in Egypt created by the intensive and large irrigation/drainage system. It accumulates the excess of irrigation water with appropriate quality that can be reused within the system.

WATER RESOURCES Water resources in Egypt are represented with the annual quota from the Nile water; the limited amount of rainfall; the shallow and renewable groundwater reservoirs in the Nile Valley, the Nile Delta and the coastal strip; and the deep nonrenewable groundwater in the eastern desert, the western desert and Sinai. The non-traditional water resources include reuse of drainage and municipal waste water, and desalination of seawater and brackish groundwater. The rainfall is very limited, and mainly at the north coast on the order of 200 mm/year. The more we go southward to Cairo the less this amount is, then the decrease is rapid until the southern areas where there is almost no rainfall. Rainstorms in Egypt take place in autumn, winter and spring and their frequency and intensity differ from year to year. Table 1 shows the available water resources in Egypt and the water use of the different sectors at year 2000. Table 1. Water Uses and Available Resources in Year 2000 Water Uses (bcm/year)

Water Resources (bcm/year)

Sector Municipalities Industry

Amount 5.25 3.5

River Transport

0.25

Fisheries

-

Hydropower

-

Agriculture

Total

63.00

72.00

Resource Nile river Groundwater (Delta and Valley) Deep Groundwater Drainage Water Reuse - Canals in the Delta Region - Nile river and Bahr Youssef - Illegal Uses Waste Water Reuse Rainfall and Flash Floods Evaporation Losses Total

Amount 55.5 5.5 0.8

4.5 5.0 3.0 0.2 0.5 (3) 72.00

83



GOVERNMENT POLICIES AND FARMER PRACTICES ON CROPPING PATTERN Crop Liberalization Policy Selection of crops, mainly cotton and basic food grains, to be grown every year was controlled by the government till year 1985. Farmers were responsible to deliver a preset quota of grains or cotton to the government. The government was controlling all marketing channels and prices. This policy resulted in an average of 30% net tax on agricultural outputs, which enabled the government to continue subsidizing consumers and financing industrialization. The government controlled also the crop rotation among growers. Figure (1) shows a typical crop rotation types (3-turn crop rotation). During the period 1985-1990, partial free cropping pattern policy was applied. It included liberalizing most of crops except for cotton and rice. Jan

Feb

Mar

Apr

May

Jun

Wheat/Long Berseem

Jul

Aug

Sep

Oct

Cotton

Beans/Berseem

Year 2

Beans/Berseem Year 3

Maize

Dec

Short Berseem

Rice

Year 1

Short Berseem

Nov

Nili

Long Berseem /Wheat

Fig. 1 Three-Turn Crop Rotation By year 1995, all crops were liberalized except for rice and sugar cane, which was restricted to the availability of water resources and future expansion plans. In fact this policy resulted in a wide variation in crop selection among growers. Cotton and rice areas were clearly affected, as most of growers shifted from cultivating cotton to rice. Although the MPWWR restricted the rice area not to exceed 1.0 million feddans, but the actual cultivated rice area reached 1.5 million feddans in year 1995, i.e. 50% more than the targeted area. On contrary, the cotton area declined to reach about 0.7 million feddans. Then, the government started to motivate cotton growers through creating new mechanisms for cotton marketing. Among different regions in Egypt, crop diversification varies according to the climatic and soil conditions. In Upper Egypt sugar cane is the largest single crop in summer, particularly in Aswan and Qena governorates. The second summer crop which competes with sugar cane in this region is sorghum and maize. In winter, wheat and clover are the major crops. In middle Egypt region, maize is major summer crop where it covers about 40% of the cultivated area. In winter, wheat and clover cover, situation looks different as the temperature is rather lower and soil conditions are different. In southern part of the Delta, rice is not allowed to be grown. Then, maize is the major summer crop and wheat is the major crop in winter. In the northern part of Delta rice is allowed to be grown where the water table is rather high. Figure (2) shows crop diversification for the different regions of Egypt. Introducing short-duration crops The government of Egypt has launched, in 1999, a program for introducing the short-term varieties of rice, to replace the log-duration one. The short-duration rice can be harvested after 120 days

84



instead of 160 days of the long-term rice. It is assumed that the short-term varieties will consume less water compared to the traditional varieties. Transplanting and Salt-Tolerant Crops Cultivating the drought tolerant varieties will result in maximizing the crop water use efficiency. Moreover, using the transplanting method in cultivating some crops instead of direct seed sowing will reduce the water requirements of those crops. In the case of rice, the reduction in the water requirement is estimated by 10-15%. Replacement sugarcane by sugar beet for sugar production, is another governmental plan. The government already introduced with great success the sugar beet in the western delta. The problem in upper Egypt that sugar cane is necessary for the sugar factories in the region. Sugar cane agriculture in upper Egypt therefore is expected to continue along the life time horizon of these factories. Introduction of modern irrigation in the sugar cane fields is being carried out to reduce the water application and to increase water use efficiency of this high water requirement crop. Cultivating the salt tolerant crop varieties will save the fresh water as they can be irrigated by marginal low quality water. Agricultural drainage water can be used either directly or after mixing with fresh water, based on the salinity of drainage water. This is practiced in Egypt on a large scale and mainly in the north of the Nile Delta as well as in EI-Salam canal project in the eastern Delta and Sinai.

Intensifying Cropping Pattern The crop intensity indicates how land is cultivated. In many case the land is cultivated twice in one year; winter and summer crops. The crop intensity ranges from 1.3 to 2.36. The higher intensity indicates that the land is cultivated three times a year. Crop intensity is higher in Lower Egypt compared to other regions. The overall crop intensity in Egypt is about 1.81. Winter Crops Long-duration clover occupies about 85% of the total cultivated area in Port Said. Damietta and Menoufia Governorates showed also about 50% and 42% of its area, respectively, grown with longduration clover. Wheat and Long-duration clover occupies about 38% for each, of the cultivated in Fayoum. While in Beni Suef, wheat occupies about 40% of the cultivated area and long duration clover occupies less than 20%. Long-duration clover is the dominant crop in Giza that constitutes more than 35% of the cultivated area, but wheat represents less than 10%. n upper Egypt governorates, the dominant crop is wheat which occupies 40%, 57%, 37%, 22% of the cultivated area in Assuit, Sohag, Qena, and Aswan. Long-duration Clover is rather less than 20 % in these governorates. Summer Crops Maize, Rice, Cotton and Sorghum are the major summer crops in Egypt as illustrated in Figure 2. Rice occupies the largest area in Lower Egypt (35% of the total area). Maize Occupies 37% of the cultivated area in Middle Egypt, 24% in Lower Egypt and 23% in Upper Egypt. Cotton occupies 18% of the cultivated area in Lower Egypt, 13% in Middle Egypt and 6% in Upper Egypt. Sorghum is a dominant crop in Upper Egypt where it occupies 25% of the total cultivated area and 4% in Middle Egypt. Nili Crops Nili Crops are planted during the period July – September. These crops are minors and occupy 7%, 21% and 8% in Lower, Middle and Upper Egypt, respectively. Nili crops include mainly maize.

85



Perennial Crops Perennial crops include Sugar Cane and Orchards. Sugar Cane is the most dominant crop in Upper Egypt where it occupies 24% of the total cultivated area. In Middle Egypt Sugar Cane occupies only 8%. Orchard occupies 9%, 8% and 4% in Lower Egypt, Middle Egypt and Upper Egypt, respectively. Orchards showed a high percentage in Desert Governorates; 47% of the

Diversity of Winter Crops among Regions in Egypt 50% % of Command Area

Long-term Clover

Wheat

Short-Term Clover

40% 30% 20% 10% 0% Lower Egypt

Middle Egypt

Upper Egypt

Desert Governorates

Diversity of Summer Crops among Regions in Egypt % of Command Area

50% Maize

40%

Rice

Cotton

Sorghum

30% 20% 10% 0% Lower Egypt

Middle Egypt

Upper Egypt

Desert Governorates

Diversity of Perennial Crops among Regions % of Command Area

50% 40%

Sugar Cane

Orchards

30% 20% 10% 0% Lower Egypt

Middle Egypt

Upper Egypt

Desert Governorates

Fig. 2 Crop diversity in Different Regions CROP PRODUCTION Crop yield and production is affected by different factors such as availability of water, quality of irrigation water, soil type and climatic conditions. As mentioned before, irrigation water deteriorates as moving northwards downstream the system but still within the safe limits. However, the new

86



reclaimed areas at the end of the system such as in the north of Delta are depending, on a great extent, on drainage water which has salinity of higher than 1500 ppm. Winter Crop Yield Table 2 presents the crop yield on governorate basis (MALR, 1996). It includes Wheat, Barely, Beans, Clover and Tomato. It could be concluded from these data that there is no big difference of wheat yield in Lower, Middle and Upper Egypt where the average yield are 17.03, 17.71 and 16.5 Ardeb/fed, respectively. But, there is significant difference in yield for tomato crop. The average yield of tomato was 14.3, 12.24 and 20.29 in Lower, Middle and Upper Egypt regions, respectively. This may indicate the impact of low water quality on tomato production. Table 2. Winter Crop Yield at Governorate Level (Source: MALR, 1996) Governorate

Nobaria Alxandria Beheira Gharbia Kafr Elsheikh Dakahlia Damietta Sharkia Ismailia Port Said Suez Menoufia Qaliobia Cairo

Wheat (Ardeb/fed) Old New Land Land 11.03 15.54

13.00

17.04

Barely (Ardeb/fed) Old New Land Land 8.87 8.49

Clover (Tons/fed) Old New Land Land 19.40

7.91

28.27

8.18

13.83

16.75

Beans (Ardeb/fed) Old New Land Land 9.44

17.05

Tomato (Tons/fed) Old New Land Land 13.77 8.90 7.23

10.72

23.84

17.58

13.63

11.97

8.23

10.06

25.77

20.00

9.17

17.05

14.42

17.50

12.87

9.75

21.38

14.50

7.24

8.59

21.96

14.41

9.22

17.1

13.36

16.52

13.87

13.01

13.52

13.19

13.13

13.99

17.99

14.18

16.00

8.92

17.74

8.91

8

4.00

16.79 12.06

8.24

6.76

10.30

9.64

7.00

25.35

6.36

4.00

14.11

7.00

15.08

6.88

19.98 6.00

28.18

9.73

6.55 14.00 17.00

6.00 8.70

26.60

33.34

13.50

13.76 29.18

8.98 14.16

19.51

7.91

25.40

14.28

8.50

32.73

14.24

13.00

5.91

32.80

13.00

14.26

12.00

7.64

18.60

20.00

9.90

6.88

22.00

12.00

11.02

6.92

24.41

Total for Lower Egypt

17.03

14.16

9.34

14.00

18.55

Giza Beni Suef Fayoum Menya

20.67

13.77

16.81

14.00

11.70

Total for Middle Egypt

19.39 17.71

13.00

12.79

10.64

Assuit Sohag Qena Aswan Luxor

16.49

12.44

14.10

16.96

14.00

12.63

8.84

6.96

6.98

34.26

39.50

25.07

15.79

11.45

10.95

11.22

6.16

5.64

23.58

23.34

25.64

15.99

5.00

8.94

6.02

18.13

10.56

0.00

7.50

6.50

26.11

13.00

Total for Upper Egypt

16.49

11.00

7.59

6.01

34.25

21.77

9.28

25.84

15.86

Total for Nile Valley & Delta 17.06

New Valley Matrouh North Sinai South Sinai

12.83

Total for Desert Govern.

12.08

12.11

9.37

8.05

12.24

35.52

15.35

21.65

13.50

9.38

8.55

10.24

12.31

7.49

23.65

8.50

22.00

6.06

20.29

16.78 13.15

1.17

1.73

1.56

1.81

4.00

4.14

0.00

10.53

3.95

2.23

12.31

7.84

23.32

13.77

10.07

Note: Ardeb = 150 kg; feddan = 0.42 hectares

87



Summer Crop Yield Table 3 shows the crop yield for Sorghum, maize, rice, sugar cane and cotton. There is no significant difference in maize yield among different regions in Egypt. But, cotton yield showed significant variation among regions. Cotton yield was found 6.97 Qentar/ fed1. In Lower Egypt, 8.2 Qentar/fed. In Middle Egypt and 11.42 Qentar/fed. In upper Egypt. This indicates the impact of water quality on cotton yield among different regions. Table 3. Summer Crop Yield at Governorate (Source: MALR, 1996)

Governorate

Nobaria Alxandria Beheira Gharbia Kafr Elsheikh Dakahlia Damietta Sharkia Ismailia Port Said Suez Menoufia Qaliobia Cairo

Sorghum (Ardeb/fed) Old New Land Land 8.25

Maize (Ardeb/fed) Old New Land Land 21.22

Rice (Tons/fed) Old New Land Land

Sugar Cane (Tons/fed) Old New Land Land 21.76

18.10

3.21

40.00

23.40

3.65

24.73

22.70

8.25

Giza Beni Suef Fayoum Menya

15.82

Total for Middle Egypt

Assuit Sohag Qena Aswan Luxor

14.09

3.41

41.75

5.81

14.47

3.39

31.94

5.39

21.85

17.00

3.65

31.20

5.85

19.98

3.05

31.25

6.21

20.28

3.48

39.37

7.95

18.72

3.04

8.48

2.29

15.73

2.00

29.46

16.44

2.71

36.01

8.19

20.00

3.41

40.30

9.06

Total for Upper Egypt

13.4

21.25

5.95

14.78

21.92

19.50

11.74

21.07 20.12 11.65

11.65

20.60

20.00

18.72

13.61

15.63 11.07

20.98 8.81

Total for Desert Govern.

3.95

6.97 4.14

25.63

8.75

31.33

6.17

46.54

9.04

2.96

44.33

8.20

3.49

40.77

11.80

47.66

10.64

46.88

42.49

4.18

47.22

42.49

11.42

46.73

40.47

47.42 8.00

48.95

19.77

11.49

21.76

2.96

15.95

10.37

36.07 33.51

18.75

New Valley Matrouh North Sinai South Sinai

88

3.51

19.13

Total for Nile Valley & Delta 13.11

Qnetar = 50 kg for cotton

28.81

19

12.74

4.74

20.33

11.1

12.45

3.78

19.79

22.74

12.89

6.52

20.63

12.27

Total for Lower Egypt

Cotton (hair) (Qenta/fed) Old New Land Land

19.57 2.76

20.00

7.84

23.32

14.00

2.23

12.31

10.07



Crop Yield in New Lands New lands have more deteriorated water quality in delta region. The current cultivated new lands is still have safe limits for water quality, but the future reclamation areas may have more deteriorated water quality due to using drainage water on a great extent particularly on Northern Delta. The current production of crops in new lands showed nearly equal levels of production particularly for vegetables and maize. It could be concluded that the crop production is higher in the upstream reaches of the Nile (Upper Egypt) and then decline as moving northwards to Delta region. But the difference is not very big. The current production of new lands is not far less than the old lands.

Crop yield compared to international Figures Table 4 presents a comparison for crop yields for different countries. Egypt showed a high level of crop yield compared to other countries. Egypt is ranked on the top for sugar cane, rice, sesame and sorghum yield. Table 4 Egypt's Rank among World Countries According to Productivity of Main Crops Egypt's Rank Crop Yield (ton/fed) 1 Sugar cane 50.4 1 Rice 4.133 1 sesame .531 1 Sorghum 2.397 2 Peanut 1.329 2 Broad bean 1.372 3 Lentil .74 4 Wheat 2.755 7 Dry onion 14.064 9 Cotton 1.1 10 Maize 3.466 10 Barely 1.153 12 Potatoes 10.27 13 Sugar beet 20.29 Note: Number of countries used for comparison are 20 Source: Compiled by the General Department of Agricultural Studies (From FAO data)

Water Use Efficiency The overall efficiency of the water system in year 2000, a shown in Figure 3, equals the consumption as a percentage of the total inflow, is about 71%. This efficiency is relatively high taking into consideration that the prevailing irrigation method is surface irrigation, which has a low efficiency. This high system efficiency is probably attributed to the intensive efforts of government in O&M, and to the current recycling practices, in addition to the considerable experience of Egyptian farmers.

89


Evapotranspiration


Evaporation 3.0

38

Rainfall 0.5

1.75

Agricultre

`

Consumption

Municipalities

Old

and Industry

13.255

55.5

Out Flow HAD

Sea Fig. 3 Water Status in Year 2000

Note: all figures are in BCM/y

CONCLUSIONS Although water resources are limited and scarce in Egypt, great efforts have been and are being conducted to increase water use efficiency and water productivities. The current water use efficiency exceeds 70% on the national level. However, plans are being prepared to increase this level of efficiency in order to increase the cultivated area by about 40% by year 2017. These plans include improvement of irrigation delivery systems, introducing low-water-consuming crops, introducing salttolerant crops, and reuse of drainage water. Intensifying cropping pattern was one of the factors contributed to increasing the water productivity.

REFERENCES Ministry of Water Resources and Irrigation, (1998). “National Policy for Drainage Water Reuse,” APRP-Water Policy Reform Project, Report No. 8, Cairo, Egypt. Abdel Azim, R. A., (1999). “Agricultural Drainage Water Reuse in Egypt: Current Practices and a Vision for Future Development,” Ph.D. thesis, Faculty of Engineering, Cairo University, Cairo, Egypt. Ministry of Agriculture and Land Reclamation (1999), The Economic Sector, Agro-economic Bulletins, several issues, Cairo, Egypt. Allam, M.N. (2000), “Water Resources: Utilization and Management”, Cairo, Egypt.

90



91



WATER USE EFFICIENCY AND WATER PRODUCTIVITY IN GREECE

A. Karamanos*, S. Aggelides**, P. Londra** * Laboratory of Crop Production. Agricultural University of Athens, 75 Iera Odos, 11855 Botanicos, Greece. E-mail: [email protected] ** Laboratory of Agricultural Hydraulics. Agricultural University of Athens, 75 Iera Odos, 11855 Botanicos, Greece. E-mails: [email protected] and [email protected] SUMMARY – This work presents a summary on water resource availability in Greece with an emphasis on irrigation water use since about 85% of water withdrawal is for agricultural purposes. Irrigation methods, crop water requirements, crop production and crop water productivity data are elaborated are discussed. The state of art of research and agricultural activity as related to Water Use Efficiency (WUE) and Water Productivity (WP) in Greece is analyzed by means of: the hydrological aspects and agronomic management strategies, the eco-physiological aspects and sustainability including social, economic, environmental and political measures. Furthermore, the initiatives and activities to improve Water Use Efficiency and to increase food production by using less water are reported. Key words: irrigation, water use efficiency, crop production.

INTRODUCTION Greece has a population of about 11,000,000 and occupies an area of 131,962 Km2. The types of land use are: a. Arable 30% b. Forest 19% c. Pastures 43,4% d. Others 7,6% The cultivated area is 34,638 Km2 out of the total arable area of 38,986 Km2. The irrigated land occupies 14,305 Km2 with the following percentage of cultivations (National Statistical Service of Greece, 2001):

Total irrigated land 2,81% 24,19%

7,94%

crops

92

65,06%

vegetables

trees

vines



Fig. 1. Total irrigated land: percentage of cultivation The climate is typically Mediterranean with all the subtypes of Mediterranean climate due to topographical relief, from the warm and dry type (e.g. Cyclades) to the alpine type (mountainous areas above 1500 m). The annual precipitation ranges from 400 mm in Athens and Cyclades islands to more than 1500 mm in the high mountainous areas with values of 700 mm in Eastern Greece and 1000-1200 mm in Western Greece and the Ionian islands. The total annual precipitation is estimated to be 116,689 hm3 with 50.9% (59,371 hm3) and 31.9% (37,190 hm3) to be lost by evaporation and runoff respectively and only 17.2% (20,133 hm3) infiltrated into soils. The total water consumption was estimated at 5,500 hm3/year increasing by more than 3% annually (Ministry of Development, 2002). The major water use in Greece is for agricultural purposes (85%) whereas the domestic and industrial uses are 13% and 2% respectively. The increased demand for water cannot always be met despite adequate precipitation. Water imbalance is often experienced, especially in the coastal and south-eastern regions, due to the temporal and spatial variations of precipitation. This results from the increased water demand during the summer months and the difficulty of transporting water due to the mountainous terrain. The irrigation methods used in Greece are surface, sprinkler and drip irrigation. The general trend is to abandon gradually surface irrigation and move on to more efficient methods like the sprinkler and drip irrigation (Fig. 2).

300,000

250,000

Irrigated area (ha)

200,000

150,000

100,000

50,000

1998

1996

1994

1992

1990

1988

1986

1984

0

Year Surface irrigation

Sprinkler irrigation

Drip irrigation

Fig.2. The trends of irrigation methods used in common reclamation works It is accepted that the water application efficiencies for surface, sprinkler and drip irrigation are 75%, 85% and 90% respectively. The water conveyance and distribution efficiency is 70%, 85% and 95% for transportation of water with earthen, concrete channels and pipelines respectively (Ministry of Agriculture, Land Reclamation Service, 1991).

93



The amounts of annual reference evapotranspiration (ETo) have been calculated in national level and ranged from 1058 mm in Northern Greece up to 2015 mm in the arid and semiarid zone of Southeastern Greece. The irrigation requirements depend on the crop and the local soil-climatic conditions and they reach 170 mm for vines in the temperate regions of Northern Greece (Thessaloniki) and 440 mm for maize in the warm and wet regions of Southern Greece (Pyrgos) (Ministry of Agriculture, 1990). To meet these requirements, irrigation has been extended (irrigation covers 40% of the total cultivated area) under common reclamation works and private irrigation systems. The water sources used differ between public and private networks. The public networks mainly use surface water, while the private ones use underground water. 3 The maximum calculated crop water requirements reach a value of 4,089 Km of water and the 3 actual water use is 6,833 Km . This means that the Total Water Use Efficiency is 60%. This low value is due to a poor land levelling, the aged water distribution systems, the high percentage of surface irrigation systems and transportation and distribution of water (Tsanis et al., 1996).

In Greece, there are values of crop water productivity in terms of crop production (kg) in relation to the volume (m3) of water used. These values come from irrigation experiments conducted in several parts of the country on behalf of the Ministry of Agriculture. The values given below are a representative selection and refer to “good” years, i.e. in seasons when crops gave the best production for the region (Ministry of Agriculture, 1990). Table 1. Crop production, water use and crop water productivity in Greece (Source: Ministry of Agriculture, 1990) Crop Production Cubic meters of water used for (kg/ha) actual evapotranspiration-ETc (m³/ha)

Crop water productivity (kg/ m³)

Maize Cotton (not ginned) Industrial tomato

12000

4980

2.41

3550

4500

0.79

97000

3900

24.87

Watermelons Vines Potatoes Lettuce

43000 14800 30000 36000

3570 1850 4800 4600

12.04 8.00 6.25 7.82

WATER USE EFFICIENCY AND WATER PRODUCTIVITY The state of research and agricultural activity as related to Water Use Efficiency (WUE) and Water Productivity (WP) in Greece can be focused in the following three aspects: The hydrological aspects and agronomic management strategies of WUE and WP. The eco-physiological aspects of WUE and WP. Sustainability and economic aspects of WUE and WP. The hydrological and agronomic management aspects of WUE and WP •

94

Agricultural and irrigation engineers facing the problem of rational irrigation planning, rarely have at their disposal extensive information on soil-plant-atmosphere system. Therefore, there is demand for simpler approaches to estimate actual evapotranspiration. The attempts that have been made in Greece related to the above issue are:


1.


The crop coefficient (Kc) for sweet shorgum was estimated under the conditions of Biotia area (Dercas et al., 1996). Poulovassilis et al. (1998) compared the estimated heat flux densities obtained through the various methods used with actual heat flux densities, as these are measured into the soil profiles. The comparison was encouraging and therefore heat flux densities calculation methods may be used for following heat flow into the soil profile. Another attempt to help rational irrigation planning was made by Anadranistakis et al. (1999a), who estimated actual and maximum evapotranspiration for neutral instead of actual atmospheric stability, employing a model based on the equation of Shuttleworth and Wallace (1985). Anadranistakis et al. (2000) investigated two parametarizations of the aerodynamic resistance control vapor transfer and they have applied to an evapotranspiration estimating model in order to find out their effect on the estimated (separate) values of transpiration and evaporation throughout the biological cycle of a crop. A semi-empirical approach was proposed for estimating actual water losses from crops (cotton, wheat and maize) assuming that the ratio of actual to maximum evapotranspiration (ET/ETm) is an exponential function of the water content in the root zone (Poulovassilis et al., 2001). Spanomitsios and Paraskevopoulou-Parousi (2001) showed that the comparison of strawberry plant transpiration rates with a linear equation of transpiration rate and the simplified form of Penman-Monteith equation agreed well with the last equation.

2.

9

9

9

9

•

Furthermore several attempts have been made by studying the reference evapotranspiration equations under the arid and semi-arid conditions in Greece. The results of the comparative evaluation of reference evapotranspiration estimated according to seven different versions of the Penman method were presented. The Penman 1963 VPD#3 was considered as the most reliable for the climatic conditions of Copais (Greece) (Poulovassilis et al., 1996). An attempt was made to estimate the wind function parameters, entering the aerodynamic term of Penman equation for hourly evapotranspiration estimates. Useful results were obtained concerning the possibility of using the Penman equation in estimating hourly or daily reference evapotranspiration (Kerkides et al., 1997). The establishment of a relationship between G/RN and L both for day and night was attempted during the development of a wheat crop, under varying soil moisture regimes (Anadranistakis et al., 1997). An attempt was made to estimate the wind function parameters entering the aerodynamic term of Penman equation for hourly evapotranspiration estimates. Useful results were obtained concerning the possibility of using the Penman equation in estimating hourly or daily reference evapotranspiration (Kerkides et al., 1999). The Penman-Monteith method for estimating reference evapotranspiration (ETo) is considered as the most reliable. Using measurements from the Copais (Greece) area, the expression of Rn as a function of Rs and of Rs as a function of the daily sunshine duration (n) is attempted, considering that G is a portion of Rn. Then ETo is calculated both with measured and estimated Rn. The comparison of the corresponding ETo values was proved satisfactory (Alexandris, 2000). A new empirical equation for estimating hourly reference evapotranspiration ETo is proposed. For the calibration of the proposed model, data collected from Copais (Greece) and from CIMIS equation (Davis Sacramento, CA) were used (Alexandris and Kerkides, 2003).

9

9

9

9

9

9

•

For an appropriate irrigation scheduling, the atmospheric demand for water vapor, the soil characteristics and the plant specific features, i.e. the soil-plant-atmosphere continuum as a dynamic system, should be considered. In semi arid regions, such as Greece, rainfall is unevenly distributed over the year. More than 80% of the mean annual precipitation falls during the months October-February. Meanwhile during the dry seasons of spring and summer tourist mobility is at its peak, exercising a substantial pressure on good quality water reserves. The fact that irrigation is also required mostly in spring and summer means that water allocation and

95



water use efficiency are very important. It is therefore necessary to develop models to conveniently estimate actual Crop Water Requirements. For the Mediterranean countries the future challenge will be the increase in food production using less water. In Greece many attempts has been made to this direction. Some of them are: Louizakis (1994) estimated the water requirements and water productivity for tobacco in Greece. Crop water requirements were determined using the Penman-Monteith method with climatic data from meteorological stations in Greece and the USDA method. It was found that the total irrigation requirements for Greece for the year 1991 were ranged from 3073 to 4069 Mm3/yr (Tsanis et al., 1996). A relationship between maize yield and water consumption has been developed in Thessaloniki plain during a four year experiment. Maximum yield was attained when the depth of applied water ranged from 700 to 900 mm (Panoras et al., 1997a). The same relationship in the same place but for winter wheat (cv. Yecora) was established after a five year experiment. It has been shown that maximum yield was attained when the amount of applied water ranged from 400 to 500 mm and the best application time of water was the second and the third critical stage of the winter wheat (Panoras et al., 1997b). Water productivity and water requirements were studied for cotton (cvs Zeta-Z and Korina) in Central Greece for the year 1997 (Polychronidis, 1998). Anadranistakis et al. (1999b) represented a model for estimating crop water requirements throughout crop development. The model has been validated with meteorological and crop data collected from experimental fields of the Agricultural University of Athens. The results taken from three crops (cotton, wheat and maize) against soil moisture profile changes were very satisfactory. The agreement between observed and estimated evapotranspiration was within 8%. Crop water requirements were studied for sorghum during a two-year experiment in Central Greece (Dercas and Liakatas, 1999). An estimation of the total water requirements in the Prefecture of Larissa irrigated both by private and public boreholes and by surface waters during the year 1999 by using FAO Penman-Monteith method was carried out (Sakellariou-Markantonaki and Vagenas, 2003).

9 9

9

9

9 9

9 9

•

Saving irrigation water in arid and semiarid zones where there is scarcity of water is a demand of major priority. The method of irrigation ultimately chosen must be the best as far as the water use efficiency is concerned. The water related efficiencies from water conveyance, storage, distribution and application must be high. Some works related to this subject are: Water use efficiency was studied under the soil-plant-atmosphere conditions of Greece by Poulovassilis et al. (1991) and they proposed measures to increase it. Papamichail and Papadimos (1996) studied the water use efficiency for graded border irrigation. The same authors (Papamichail and Papadimos, 1996) studied the water use efficiency for furrow irrigation used for irrigation. Ampas (1998) studied the irrigation efficiency in furrows used for row crops. He found that the calculation of inflow rate at each time step maximizes the application efficiency. Mayropoulos (2001) studied water use efficiency and he recommended several measures at technical, economic and legislation level.

9 9 9 9 9

The eco-physiological aspect of WUE and WP •

9

96

Here there are some aspects of the actual situation of research and agricultural practices concerning crop water productivity in Greece. The water productivity of potatoes was studied experimentally in the region of Attika and it was found to be 6.25 kg/m3 (Aggelides et al., 1984).



The water productivity of vine (Cardinal) was studied experimentally in the region of Attika and it was found to be 8.0 kg/m3 (Panagiotou and Aggelides, 1987). Poulovassilis et al. (1993a) studied the crop water productivity reduction due to irrigation with brackish groundwater. The crop water productivity of sweet sorghum was experimentally studied in Copais area with a high water table. Results showed a high crop water productivity due to the small amount of irrigation water Dercas et al., 1994). The crop water productivity of sweet sorghum was studied in relation to plant density (Dalianis et al., 1996). It was found that decreasing plant density increased water productivity. The water productivity of winter wheat (cv. Yecora E) was studied in the plain of Thessaloniki and it was found that the maximum yield was attained when the amount of applied water ranged from 400 to 500 mm (Panoras et al., 1997). Crop water productivity and water use efficiency were studied for sweet sorghum (cv. Keller) by Dercas et al. (1998). They found a good relationship between dry matter yield and the quantity of applied water. The water productivity of cotton was studied and compared with water consumption (evapotranspiration, ET), simulated by GOSSYM, a simulation model of cotton growth and yield (Gertsis et al., 1998). The water productivity of lettuce was studied in the area of Achaia (Peloponnese) and it was found to be 7.82 kg/m3 (Aggelides et al., 1999). The water productivity of two varieties of sweet sorghum (cvs Keller and MN 1500) was studied after two-year experiments in Boeotia. Drip irrigation was applied. The water productivity was very high (Dercas et al., 2000). Water productivity of Medicago sativa studied by Lazaridou and Noitsakis (2002) in pure and mixed crop grown with grass. It was found that water productivity exhibited higher values during early spring in relation to the measurements of the remaining period and in the mixed crops despite their smaller productivity compared with pure crop. Fisarakis (2002) studied the reduction of growth observed at relatively low salinities, often before the appearance of visible symptoms. The unfavorable effects of salinity on grapevines were increasing as the level of salinity increased and were demonstrated in terms of a reduced plant growth, bunch number, berry size and yield.

9 9 9

9 9

9

9

9 9

9

9

Sustainability and economic aspects of WUE and WP •

9 9

9

9

9

9 9

This issue contains social, economic, environmental and political measures that add additional importance to the sustainability of agricultural production. Some of these works are: Poulovassilis et al. (1993b), studying sea water intrusion in the area of Irria (East Peloponnese), found the same results as in the case of Argolis. Poulovassilis et al. (1994), studying sea water intrusion in the area of Argolis, showed that the salinity status of the soil is correlated with that of the ground water used for irrigation and that both of them affect adversely crop yields according to their sensitivity. A comprehensive study has been undertaken for in Argolis (Peloponnese) in an attempt to identify the factors which caused soil degradation due to salinity and to promote remedial measures (Poulovassilis et al., 1996). This study presents the climate change, the water availability, the crop water requirements, the improvement of water use efficiency and the necessity of high priority in water resources policy (Chartzoulakis et al., 1997). Mimides et al. (1997) studied ground water degradation due to the sea intrusion in the region of Irria (East Peloponnese). Various softwares such as SURFER, MINEQL and DUROV diagram were utilized. The plain of Argos in Southern Greece was investigated for sea water intrusion, ground water pollution and the suitability of ground water for irrigation purposes (Alexandris et al., 1998). An attempt was made to assess the effect of irrigating tomatoes with saline water in a greenhouse (Kerkides et al., 1998).

97


9 9

9 9


The influence of low quality irrigation water to agricultural production was studied by Giannoulopoulos et al. (2002). Kountouras and van Leeuwen (2002) studied the environmental effects on grapevine grown in the Nemea region and especially a) the vine water status and physiological parameters and b) vegetative growth, fruit ripening and yield. An initiative for Mediterranean water policy is presented to overcome water competition between agriculture and tourism in Cyclades and especially in Naxos island (Epp et al., 2003). Chartzoulakis and Psarras (2005) studied the current trends and predictions for the future climate changes in the area of Crete and discuss the factors that will most probably limit or enhance photosynthesis and productivity of the most important tree crops in a semi-arid Mediterranean environment.

REFERENCES Aggelides, S., G.K. Chardas, P.D. Tsakaleris and G.I. Stamos, 1984. Irrigation of potatoes based on soil water negative pressure measurements. Agricultural Research, 8:45-55. Aggelides, S., I. Assimakopoulos, P. Kerkides and A. Skondras, 1999. Effects of soil water potential on nitrates content and the yield of lettuce. Communication in Soil Science and Plant Analysis, Vol. 30 (1&2). Alexandris, S., 2000. Reference evapotranspiration estimation with the Penman-Monteith method using standard meteorological parameters. 5o Hellenic conference of Meteorological Climatology and Atmospheric Physics. Thessaloniki 2000, pp.439-446. Alexandris, S. and P. kerkides, 2003. New empirical formula for hourly estimations of reference evapotranspiration. Agricultural Water Management, 60:157-180. Alexandris, S., P.M. Allen, I. Black, N. Kalamaras, P. Giannoulopoulos, M. Lemon, T. Mimides, A. Poulovassilis, M. Psychoyou and R.A.F. Seaton, 1998. Agricultural production and water quality in the Argolic valley, Greece. The Archaeomedes project, Understanding the natural and anthropogenic causes of degradation and desertification in the Mediterranean basin. Research results. EUR 18181. ISSN 1018-5593, pp 281-326. Ampas, V., 1998. Optimum application efficiency of furrow irrigation. Geotechnical Scientific Issues, Volume 9, Issue III, pp.4-12. Anadranistakis, M., A. Liakatas and P. Fragouli, 2000. Parametraization of aerodynamic resistances in evapotranspiration estimating models. 5o Hellenic conference of Meteorological Climatology and Atmospheric Physics. Thessaloniki 2000. Anadranistakis, M., P. kerkides, A. Liakatas, S. Alexandris and A. Poulovassilis, 1999a. How significant is the usual assumption of neutral stability in evapotranspiration estimating models? Meteorol. Appl. 6:155-158. Anadranistakis, M., A. Liakatas, P. kerkides, S. Rizos, J. Gavanosis and A. Poulovassilis, 1999b. Crop water requirements model tested for crops grown in Greece. Agricultural Water Management, 45:297-316. Chartzoulakis, K. and G. Psarras, 2005. Global change effects on crop photosynthesis and production in Mediterranean: the case of Crete, Greece. Agriculture, Ecosystems and Environment, 106:147157. Chartzoulakis, K., A.N. Angelakis and P. Skylourakis, 1997. Irrigation of Horticultural crops in the island of Crete, Greece. ISHS Acta Horticulturae 449: II International Symposium on Irrigation of Horticultural Crops. Chania, Crete, Greece, 1 August 1997. Dalianis, C., E. Alexopoulou, N. Dercas and Ch. Sooter, 1996. Effect of plant density on growth, productivity and sugar yields of sweet sorghum in Greece. Biomass for Energy and Environment. Proc. 9th European Biomass Conference, Ed. Chartier et al., Pergamon Press, Oxford, UK, pp. 582-587. Dercas, N., G. Kavadakis and A. Nikolaou, 2000. Evaluation of productivity, water and radiation use efficiency of two sweet sorghum varieties under Greek conditions. 1st World Conference on Biomass for Energy and Industry, Sevilla, Spain, Ed. James & James (Science Publishers) Ltd., pp. 1654-1657.

98



Dercas, N. and A. Liakatas, 1999. Sorghum water loss in relation to irrigation treatment. Water Resources Management, 13:39-57. Dercas, N., C. Panoutsou and A. Liakatas, 1998. Sorghum yield simulation according to irrigation treatment. IFAC-CAEA ‘98, Control Applications and Ergonomics in Agriculture, Athens, pp. 179184. Dercas, N., C. Panoutsou and C. Dalianis, 1996. Water and nitrogen effects on sweet sorghum growth and productivity. Biomass for Energy and Environment. Proc. 9th European Biomass Conference, Ed. Chartier et al., Pergamon Press, Oxford, UK, pp. 54-60. Dercas, N., C. Panoutsou, C. Dalianis and Ch. Sooter, 1994. Sweet sorghum (Sorghum Bicolor (L.) Moench) response to four irrigation and two nitrogen fertilization rates. Biomass for Energy, Environment, Agriculture and Industry. Proc. 8th European Biomass Conference, Ed. Chartier et al., Pergamon Press, Oxford, UK, pp. 629-639. Fisarakis, I.K., 2002. Salinity effects on grapevine. Geotechnical Scientific Issues, No 4/2002, Volume 13, Issue I, pp.73-83. Gertsis, A., A. Symeonakis, S. Varsami, S. Galanopoulou-Sendouca and G. Papathanasiou, 1998. Evaluation of water use efficiency with GOSSYM, a simulation model of cotton growth and yield. 7ο Hellenic Soil Science Symposium, Agrinio 27-30 May, 1998, pp. 99-108. Giannoulopoulos P., S. Alexandris, M. Psychoyou and A. Poulovassilis, 2002. Salinization and quality characteristics of Argolic plain groundwater. Proccedings of 6o Hellenic Hydrogeology symposium, pp1-12. Kerkides, P., C. Olympios and M. Psychoyou, 1998. The effect of irrigation with saline water on greenhouse grown vegetables. 7ο Hellenic Soil Science Symposium, Agrinio 27-30 May, 1998, pp. 120-131. Koundouras, S. and C. van Leeuwen, 2002. Environmental effects on winegrape cv Agiorgitiko (Vitis vinifera L.) grown in the Nemea region (Greece). 2. Vegetative growth, fruit ripening and wine chemical and sensory characteristics. Geotechnical Scientific Issues, No 4/2002, Volume 13, Issue I, pp.28-38. Lazaridou, M.G. and V. Noitsakis, 2002. Effects of drought on seasonal changes of water-use efficiency in a mixture crop of Medicago sativa. Geotechnical Scientific Issues, No 3/2002, Volume 13, Issue II, pp.59-66. Louisakis, A.D., 1994. The water requirements and leaf quantity of tobacco in Greece. Aspects of Applied Biology. Conference on Efficiency of water use in crop systems. Ed. M. C. Heath. Vol 38:209-216. Mayropoulos, T.I., 2001. Saving irrigation water in scarcity of water conditions. Geotechnical Scientific Issues No 2/2001, Volume 12, Issue II, pp.113-126. Mimides T., S. Aggelides, S. Koutsomitros, M. Psychoyou, G. Kargas and A. Sgoumbopoulou, 1997. Study of Iria valley aquifer of Argolis with emphasis to hydrochemical processes after the sea intrusion. Proccedings of 4o Hellenic Hydrogeology symposium, Thessaloniki 11-14 November 1997.. Ministry of Agriculture, 1990. Experimental application of new irrigation methods. Athens, Greece. Ministry of Agriculture, Land Reclamation Service, 1991. Technical guidelines for conducting agrotechnical studies for land reclamation works. Athens, Greece. Ministry of Development, 2002. Master Plan for the Greek Water Resources Management. Athens, Greece. National Statistical Service of Greece, 2001. Annual agricultural statistics survey for the year 2001. Panayiotou, C and S. Aggelides, 1987. Influence of irrigation on the production characteristics of the Cardinal vine variety. Agricultural Research, 11:393-401. Panoras, A., A. Giannakaris, M. Dellios, S. Dimov and S. Eneva, 1997a. Water winter wheat yield relations in Thessaloniki plain. Geotechnical Scientific Issues, Volume 8, Issue IV, pp.45-51. Panoras, A., A. Giannakaris, M. Dellios, S. Dimov and S. Eneva, 1997b. Corn yield to response to water in Thessaloniki plain. Hydrotechnica. Volume 7, pp.39-51. Papamichail, D.M. and D. K. Papadimos, 1996. Efficiency evaluation of the SCS design equations for border irrigation. Geotechnical Scientific Issues, Volume 7, Issue II, pp.45-57.

99



Polychronides, M., S. Galanopoulou, S.Aggelides and N. Danalatos, 1998. The effect of irrigation and fertilisation practice on cotton growth and development under greek conditions. Second Conference of World Cotton Research, September 6-12, 1998, Athens, Greece. Poulovassilis, A., M. Anadranistakis, A. Liakatas, S. Alexandris and P. kerkides, 2001. Semi-empirical approach for estimating actual evapotranspiration in Greece. Agricultural Water Management, 51:143-152. Poulovassilis, A., P. kerkides, S. Alexandris and S. Rizos, 1997. A contribution to the study of the water and energy balances of an irrigated soil profile. A. Heat flux estimates. Soil & Tillage Research, 45:189-198. Poulovassilis, A., T. Mimides, A. Nikolopoulos, M. Psychoyou, N. Sgoupopoulou, P. Kerkides, S. Alexandris, S. Aggelides, G. Kargas and P. Giannoulopoulos, 1994. Validity, limits and possible trends of coastal south mediterrannean traditional groundwater irrigated agriculture. International Conference on Land and Water Management in the Mediterranean Region. Bari, 4 - 8 Sept. Poulovassilis, S. Aggelides, P. Kerkides, T. Mimides, M. Psychoyou, S. Alexandris, G. Cargas and A. Sgoumbopoulou, 1993a. Soil salt accumulation in the valey of Iria- Peloponnese due to irrigation with brackish groundwaters. First International Congress on the Environment. Abstracts Geotechnical Chamber of Greece, Athens, March, 21 - 24. Poulovassilis, A., S. Aggelides, P. Kerkides, T. Mimides, M. Psychoyou, S. Alexandris, G. Cargas and A. Sgoumbopoulou, 1993b. Sea-water intrusion in the coastal aquifers of Iria – Peloponnese due to overpumping. First International Congress on the Environment. Abstracts Geotechnical Chamber of Greece, Athens, March, 21 - 24. Poulovassilis, A., P. Kerkides and S. Aggelides, 1991. Insufficiency of soil water - Crops. Conference of Hellenic Soil Society: Drought - Productivity of Soil. Centre Helexpo, Thessalonica, 1991. Sakellariou-Makrantonaki, M.S. and I.N. Vagenas, 2003. Crop water requirements in the prefecture of Larissa. Hydrotechnika. Journal of the Hellenic Hydrotechnical Association. Vol.13:13-28. Spanomitsios, G.K. and G. Paraskevopoulou-Parousi, 2001. Strawberry transpiration: comparison of measurements with the output of two mathematical estimation methods. Geotechnical Scientific Issues No 2/2001, Volume 12, Issue I, pp.83-93. Tsanis, K., P.A. Londra and A.N. Angelakis, 1996. Assesment of Water Needs for Irrigation in the Island of Crete. 2nd International Symposium on Irrigation of Horticulture Crops, 8-13 September, 1996, pp.41-48.

100



101



IRRIGATED AGRICULTURE AND WATER USE EFFICIENCY IN ITALY

M. Todorovic*, A. Caliandro** and R. Albrizio* *Mediterranean Agronomic Institute of Bari, CIHEAM-IAMB, Italy **University of Bari, Faculty of Agriculture, Italy

SUMMARY – This document aims to provide a comprehensive review of irrigated agriculture in Italy with a particular emphasis on the Water Use Efficiency (WUE) and water productivity (WP). The data presented include the country climatic characterization, water availability and withdrawal, sources of irrigation water, irrigable and irrigated lands, irrigation methods, irrigated crops and their growing parameters, experimental data on both biomass and yield WUE. The agronomic data of main crops, grown principally in Southern Italy environment, are taken into consideration including tree crops (grapevine, olives, citrus, peach, etc.), field crops (wheat, maize, sugarbeet, sunflower, etc.) and horticultural crops (tomato, potato, watermelon, beans, spinach, etc.). An analysis of: crop production functions, application of irrigation methods, and crop water requirements (in terms of estimation of reference evapotranspiration and crop coefficients), in Italian agriculture is given. The presented crop coefficient data vary for a crop in respect to local climatic conditions, latitude, altitude, time of sowing and applied agronomic practices. Moreover, these data differ notable from those presented in scientific literature: it indicates a necessity for a local calibration and eventual revision of well-known existing FAO documents on crop water requirements and crop response to water. Finally, some common agronomic practices for enhancing WUE & WP have been described, focusing mainly on the situation of Southern Italy. This analysis, based on the evaluation of the national scientific literature and technical reports, has shown how these strategies should aim at increase of beneficial water consumption (transpiration) against the non-beneficial losses by: (i) increasing of marketable yield per unit of water transpired; (ii) maximizing transpiration consumption relative to evaporation losses; (iii) enhancing effective use of rainfall and water stored in the soil. Key words: water resources, irrigation, water use efficiency, water saving, Italy.

INTRODUCTION 2 Italy, with a surface area of 301,277 km , occupies a central location in the Mediterranean basin. Stretching over 1,200 km between North and South, Italy has shores on four Mediterranean sees (the Ligurian, the Tyrrhenian, the Jonian and the Adriatic) and it has an exceptionally long coastline of almost 7,500 km. About 27% of Italian territory (8,136,207 ha) is along the coast line and 73% (21,997,893 ha) is considered the inland.

The Italian territory can be subdivided naturally into four main physiographic regions: 6. the Alps mountains chain in the North, extending from west to east and reaching up to 4,800 m a.s.l. (with Monte Bianco, the highest peak of Europe);

102



7. the lowland of the Po river basin, located on the South of the Alps; 8. the peninsula, including the central Apennine massive with the peaks rising up to 2,900 m a.s.l. and the coastline, and 9. two large islands, Sicily on the South and Sardinia on the West of the peninsula The lowlands, flat and valley areas, cover 6,976,373 ha (23.2% of the territory); the mountain areas occupy 10,611,957 ha (35.25% of the country), while the hill areas cover about 12,542,779 ha (41.55% of the territory). The precipitations in Italy are relatively abundant (on average about 1,000 mm/year), but as often, they are not evenly distributed between seasons and regions, and high evapo-transpiration in coastal areas causes significant losses. Due to the range of rainfall, hydrological and climatic regimes (from Mediterranean to continental and Alpine), Italy presents a wide diversity of ecosystems, landscapes and agricultural practices. In fact, Italy's agriculture is a typical example of the division between the agricultures of the northern and southern European countries: the northern part produces primarily grains, sugar-beet, soybeans, meat, and dairy products, while the south is specialized in producing fruits, vegetables, olive oil, wine, and durum wheat. Inasmuch as Italian agriculture is very intensive and market oriented it preserves many local peculiarities especially in the Southern regions. In fact, most farms are small, with an average size of only 7 ha whereas a large working force (more than 1.5 million) is employed. Irrigation represents a common practices in all parts of the country due to market oriented agricultural production and strong variability and uncertainty of climatic factors. However, the cropping pattern, irrigation methods, agronomic practices and water use efficiency vary significantly from region to region and, also, from farm to farm. This paper reports the data describing the irrigated agriculture, crop water requirements and water use efficiency in Italy emphasizing the practices that improve the efficiency of water use and save water for other purposes.

CLIMATIC CHARACTERIZATION The Italian climate is highly varied due to variety of hydrographic and orographic factors, its NorthSouth elongation and exposition to four Mediterranean seas. These factors influence substantial variation of the main climatic variables as illustrated in Figures 1 and 2. The average temperature in January (the coldest month) varies from several degrees below zero in the Alpine area to more than 6°C in the coastal Mediterranean regions while the average temperature in July (the warmest month) spans between less than 15°C in the Northern Alps and about 30°C in the Southern Mediterranean zones (Fig. 1).

103



Fig. 1. Spatial distribution of temperature of the coldest month (on the left) and of the warmest month (on the right) based 30 years averages (Source: SAIN-UCEA, Rome, 1995) The sunshine hours cumulated on annual basis ranges between less than 1800 in the Alps to more than 2200 in the South (Fig 2). The average annual precipitation is relatively abundant and it is estimated to about 1000 mm per year although it is unevenly distributed among regions and seasons. In fact, average annual precipitation goes from less than 400 mm in the coastal Southern zones, receiving almost all precipitation input during the winter season (between October and March), and to almost 3000 mm in the Northern Alpine areas (Fig. 2). The Southern Adriatic regions receive much less precipitation than the Tyrrhenian side due to the characteristic movements of the humid air masses and orographic characteristics of the peninsula. According to the above mentioned parameters and the Köppen climatic classification, the overall Italian climate can be described as moist, mid-latitude subtropical although eight climatic zones can be observed moving from the Northern Alps regions to the South and from the coastal areas to the inner Apennine massive as illustrated in Figure 3.

Fig. 2. Spatial distribution of sunshine hours per year (on the left) and of the total annual precipitations (on the right) based 30 years averages (Source: SAIN-UCEA, Rome, 1995)

104



Fig. 3. The main climatic zones of Italy according to the Köppen climatic classification (Source: www.italocorotondo.it/tequila/partner_section/italy_english/) The coastal zones of Italy are characterized by dry semi-arid Mediterranean climate which passes to sub-littoral and sub-continental as moving into the inner areas of Apennines. The central Northern regions, including the Po river valley (the most important Italian river basin where live about 15.5 million inhabitants), are characterized by sub-continental climatic conditions while the climate of the Alpine mountains goes from cool temperate to cold polar. Some high peaks of Apennines are also characterized by cool temperate climatic conditions.

WATER RESOURCES AVAILABILITY AND USE IN ITALY Water resources availability The analysis of water resources availability in Italy is based on the data coming from several sources (ANPA, 2001; IRSA CNR, 1999; EUOSTAT, 1998; AQUASTAT, 1998, Blue Plan, 2001) and a synthesis of results is presented in Table 1. 3 The precipitation over the Italian territory generates every year a total flow of about 296 km . However, due to the presence of large areas characterized by semi-arid Mediterranean climate, the evapotranspiration losses are estimated to 129 km3/year while the subsurface flow to the sea is in average of about 12 km3/year. This means that the internal renewable water resources account to approximately 155 km3/year which represents about 52.3% of total flow generated by precipitation. External runoff is calculated to 7.6 km3/year (from Switzerland 51%, from Slovenia 43% and from France 6%) while spring outflow contribution from local aquifers is estimated to about 3.5 km3/year. This means that the total renewable water resources of Italy are about 166.1 km3/year. It is estimated that only two/thirds of that volume (or about 110 km3/year) are technically and economically available for exploitation.

105



The total groundwater availability is about 40 km3/year but the greatest part of it (about 30 km /year or 75%) contributes to the recharge of regional aquifers and only 25% (10 km3/year) represents the recharge of local aquifers. Only one/third of it (about 3.5 km3/year) is related to the spring outflow as mentioned previously. 3

Table 1. A synthesis of water resources availability in Italy (data elaborated from the following sources: ANPA, 2001; IRSA CNR, 1999; EUOSTAT, 1998; AQUASTAT, 1998, Blue Plan, 2001) Average precipitation [mm/year] 982 3 Flow generated by average precipitation [km /year] 296 Average evaporation [mm/year]

428 (438*)

Evaporation losses [km3/year]

129 (132*)

Subsurface flow to the sea [km3/year]

12 (9*)

Internal renewable water resources [km3/year]

155 (=296-129-12) 7.6

External runoff – inflow from other countries [km3/year] Total groundwater availability [km3/year]

40

3

Groundwater recharge of local aquifers [km /year]

10 to12

3

Spring outflow from local aquifers [km /year] 3

Total renewable water resources [km /year]

3.5 166.1 (=155+7.6+3.5)

3

Potentially usable water resources [km /year] *

110

there is some difference between data coming from different sources

Therefore, the total renewable water resources availability per person can be estimated as about 2914 m3/year/capita, or 1930 m3/year/capita by means of potentially usable resources. These values are much more greater than those of the Southern Mediterranean countries (e.g. total renewable water resources availability in Middle East and North Africa Region is about 1250 m3/year/capita, or about 43% of Italian availability). However, they are significantly lower than the average renewable water resources of Western Europe countries, which is estimated to about 5183 m3/year/capita (World Resources Institute, 2000). Furthermore, it is important to highlight that water resources are not regularly distributed over the Italian territory (Fig. 4): in the Northern part is located about 59.1% of potentially usable water resources whereas the rest of the country accounts on the 40.9% of resources. This disparity becomes even more evident when expressed by the availability of potentially usable resources per capita (Fig. 4b) which indicates that water resources availability per capita in the North is almost 3.5 times greater than in the Islands and it represents about 175% of water availability in the continental Southern regions. These data emphasize the seriousness of water problems in the Southern regions especially during the summer months when in those areas water demand is strongly increased due to important vocation to tourism and consequent high population inflow.

106



Islands 4.5% South 18.2%

North 59.1%

Center 18.2%

a) 3000

2542 2500

1930

1834

2000

1451

1500 1000

743

500 0 North

Center

South

Islands

Italy

b)

Fig. 4. Regional distribution of potentially usable water resources in Italy as a percentage of total resources (a) and as water availability in m3/year/capita (b) (Source: IRSA CNR, 1999)

Water withdrawal and sectorial water use The average water withdrawal in Italy is estimated to about 51.820 km3/year which represents about 31% of the gross annual available water resources and 47% of the water resources technically and economically available for exploitation. (IRSA, CNR, 1999). This amount, translated to a mean annual per capita withdrawal of 910 m3, is significantly greater than EU average of 662 m3/capita/year and it is, together with Egyptian water withdrawal per capita (however, Egypt uses 100% of exploitable resources), the greatest in the Mediterranean region. Nonetheless, it is significantly lower than in some other highly developed countries (e.g. USA - 1873 m3/capita/year and Canada - 1736 m3/capita/year). The greatest part of water withdrawal belongs to surface water resources (39.673 km3/year or 76.6%) which includes the storage capacity of artificial accumulation reservoirs of about 8.426 km3. The contribution of groundwater is estimated to about 12.147 km3/year, which corresponds to 23.4% of total water withdrawal. Nevertheless, it is important to underline that the knowledge about groundwater resources is far from accurate due to frequent non-authorized water abstraction for irrigation especially in the Southern regions. The water withdrawal varies from year to year between 40 and 56 km3/year according to the availability and demand, and also, it is very variable from region to region. In general, about 65% of withdrawal belongs to the Northern part of the country, 15% to the Central regions and 25% to the South and the Inlands. Water withdrawal is the highest in the North-East region of 1975 m3/capita/year (even greater than in the USA and Canada), and it the lowest in the Apulia region (220 m3/capita/year). In some regions, water shortage is attenuated with the water transfer from other regions, as it is the case of the Puglia region, which receives more than half of its water demand from Basilicata region and partially from Campania region. This was possible thanking to the “CASSA PER IL MEZZOGIORNO” (Southern Italy Development Fund), promoted and implemented by Italian

107



authorities during the 50-ties, 60-ties and 70-ties of the last Century. The realization of new accumulations and water delivery systems is still in progress and, together with an inter-regional action program for management of common water resources, represents the keystone of strategies for facing water shortage problems in the South. The partitioning of water withdrawal between different sectors changes from year to year (Table 2) depending on the overall availability and water demand. Nonetheless, on the basis of average historical data, can be stated that, in general, about 60 per cent of water withdrawal is used for irrigation, 25 per cent for industry, and 15 per cent for domestic use (Fig. 5). Certainly, when water availability is scarce, the reduction is applied primarily to irrigation sector as illustrated in Table 2. Table 2. Sectorial water use in Italy for a hydrological normal (1991) and a dry (1999) year (Source: ISTAT, 1991; IRSA-CNR, 1999; MPAF, 2004) Sectors of water use 1991 (a normal year) 1999 (a dry year) Water use Water use [%] Water use Water use [%] [km3] [km3] Domestic Industry Energy*

8 12 -

16 24 -

7.9 8 4.5

19.6 19.7 11.1

Agriculture 30 60 20.1 Total 50 100 40.6 * includes only the use of freshwater for thermoelectric plant cooling

49.6 100

Domestic 15%

Industry 25%

Agriculture 60%

Fig. 5. Water withdrawal by sectors in Italy The use of water for irrigation is not regularly distributed all over the country: 67% of it belongs to the Northern Italy, 28% to Southern Italy with islands and only 5% to the Central part of the country. Main sources of irrigation water are rivers (67%), followed by groundwater from wells (27%) and by reservoirs (6%). The water withdrawal for domestic purposes reaches almost 370 liters/person/year and it is obtained mainly from groundwater aquifers (50%) and springs (40%) and only marginally from surface water (10%). Groundwater withdrawals in the Po Basin are considered to have reached their maximum, with over-exploitation already occurring in some sub-basins (e.g. Lambro-SvesoOlana, Parma, Panaro rivers).

IRRIGATED LAND AND IRRIGATION PRACTICES IN ITALY Agricultural, irrigable and irrigated land 2 The total utilized agricultural area (UAA) in Italy is estimated to about 131,941 km which corresponds to 43.8% of total surface area. The agricultural land area is continuously decreasing: since 1970 the utilized agricultural area diminished by 2.8 million hectares (-16%) according to data from the most recent survey of farm structures. This is a phenomenon which affects all developed and

108



industrialized countries. Between 1991 and 2001, the utilized agricultural area has decreased progressively by 11.1% per inhabitant, from 0.3 to 0.26 hectares per capita (INEA, 2003), which is in the range of other EU countries (-10.9%). Land is, thus, becoming an ever-more precious resource, especially in countries which, like Italy, have a high population density and where national territory is subject to considerable variation in altitude. A synthesis of land use in Italy and other European countries is given in Table 3 as a percentage of total surface area. In general, the data indicate an intensive use of land in Italy and a substantial difference in respect to other EU Mediterranean countries and to the EU territory. This is probably due to the fact that Italian territory is exposed to many very different climatic zones (eight), from cold polar to subtropical, which caused a strong variation in land use. Approximately 37% of the Italian territory is used for arable agriculture, which is much more greater than EU average of 27%. Nevertheless, due to many arid and semi-arid zones, the percentage of bare ground is two times greater than the EU average. Moreover, the urban, unproductive areas, cover about 2.1 million hectares which is 7% of the country, while the EU average is 5% and average of other EU Mediterranean countries is 4%. Table 3. Land use in Italy and EU countries (%) (Source: INEA, 2003, on the basis of EUROSTAT survey) Italy (1)

Arable land Permanent crops (2) Moorland (areas over 20% covered by shrubs) Permanent meadows and pastures Bare ground Inland waterways and wetlands (3) Unproductive areas and other land (4) TOTAL AREA (000 ha)

37 29 8 10 6 3 7 30,133

Other EU Central EU Mediterranean countries (**) (*) countries 33 32 26 32 20 4 11 20 5 3 1 3 4 6 72,988

110,172

Total EU (***)

27 37 8 12 3 8 5 292,105

(*)

Greece, Spain, Portugal. (**) France, Germany, Belgium, Luxemburg, Denmark, The Netherlands, (***) Excluding UK and Ireland, (1) Including temporary forage crops and set aside. (2) Tree and other permanent crops (woods and forests). (3) Including glaciers and eternal snow. (4) Man-made and industrial settlements, infrastructure, rocks and barren land; ornamental parks and gardens, roads, railways, etc.

According to the General Agriculture Census carried out in 2000, the irrigable land amounts to 3,887,387 hectares which is equivalent to 29% of total national utilized agricultural area (UAA). A comparison with the 1990 Census, indicates that irrigable land has remained almost the same although it varies considerable from region to region (Table 4). The Northern regions, endowed with significantly greater water resources than Central and Southern regions, could potentially irrigate about 50% of their UAA. The average irrigated area is estimated to approximately 2.65 million hectares which corresponds to 68% of the total irrigable land and to about 20% of UAA. According to the Census, the irrigated area in 2000 was slightly smaller (2.47 million hectares), with substantial differences between the regions (Table 4). Slightly less than two thirds of the irrigated area is in the Northern Regions, involving 34.9% of farms with UAA and with an average area per farm of 6.5 hectares. In the Centre, only about 17.9% of farms are irrigated, whereas in the South the practice is carried out on 25% of farms with a total area of 758 thousand hectares, equivalent on average to 2.2 hectares per farm. According to the official data of National Institute for Statistics (ISTAT), about 63.2% of irrigated land in located in the North, 7.2% is in the Central part of the country while 29.6% is situated in the South. There are two main factors limiting irrigation in Italy: the availability of water resources and the presence of infrastructures for water accumulation and delivery to the fields. Accordingly, the largest irrigated areas are located in Lombardia Region, covering about 554,382 ha and corresponding to

109



almost 80% of UAA. Then, irrigation is fully developed in Piemonte Region (on 335,800 ha), Veneto (265,253 ha), Emilia Romagna (252,377 ha), and Puglia Region (248,814 ha). Nonetheless, it is necessary to recognize a drawbacks of official statistics which have difficulties to consider the farms, located mainly in the South, subjected to non-authorized irrigation from private wells. Table 4. Irrigable land in Italy and area irrigated in 2000 (Source: ISTAT, 2002) Region Irrigable land Irrigated land Irrigated/Irrigable [ha] [ha] [%] Piemonte 448,947 335,800 79.25 Valle d’Aosta 26,212 23,623 90.12 Lombardia 700,140 554,382 79.18 Liguria 11,244 7,191 63.96 Trentino Alto Adige Veneto Friuli Venezia Giulia Emilia Romagna Toscana Umbria Marche Lazio Abruzzo Molise Campania Puglia Basilicata Calabria Sicilia Sardegna ITALY

61,774 435,845 91,876

57,768 265,253 63,202

93.51 60.86 68.79

565,573 111,603 66,927 49,470 150,088 59,358 20,881 125,305 389,617 80,640 117,143 209,036 165,709 3,887,387

252,377 47,286 32,117 25,070 74,052 29,995 11,812 86,414 248,814 42,325 66,922 161,044 62,315 2,467,763

44.62 42.37 47.99 50.68 49.34 50.53 56.57 68.96 63.86 52.49 57.13 77.04 37.6 63.48

Irrigated crops The Census on agriculture, referring to the year 2000, provides the data about irrigated crops in Italy and a synthesis of elaborations is given in Figure 6. The data indicate that almost 86% of cultivated citrus crops were irrigated (corresponding to 113,600 ha in respect to total cultivated area of 132,500 hectares). Then, the irrigation was very intensive in the areas cultivated with vegetables (70%), potato (67.4%) and maize (58%), followed by fruit-tree crops (38%), sugarbeet (36.2%), soya (34.5%), vineyards (25.5%), etc. The maize is the crop which is irrigated on the greatest surface areas in Italy, i.e. on 622,000 ha, mainly located in the North-West regions. Then, large irrigated areas are covered by forage crops (267,000 ha), vegetables and potato (217,000 ha), fruit-tree crops (189,000 ha), vineyards (183,000 ha), sugarbeet (81,000 ha), etc. Inasmuch as the cereal cultivation covers the greatest part of UAA (2,233,00 ha), the cereal crops are irrigated on 99,500 ha which represents only 4.5% of their total cultivation.

110


90


85.8

80

70.0

70

67.4

58.2

60 50 40 30 20 10

38.0

36.2

34.5 25.5

17.5 4.5

6.7

10.0

so i v in a ot e he ya rc rd ul tiv s at io ns

ci tru s

cr op su ga s rb fo e ra ge et cr op s ce re a fru ls it tre es su nf lo we r m ve aize ge ta bl es po ta to

0

Fig. 6. Irrigated crops in Italy (as percentage of total cultivated area of each crop) according to the Census in 2000 (Source: ISTAT, 2002) The citrus crops are almost fully irrigated (up to 95%) in the Southern regions, especially in Sicily and Basilicata. The fruit-tree crops are irrigated almost completely in Trentino Alto Adige (93%), while the percentage is lower in other regions: 72% in Veneto, Friuli Venezia Giulia and Basilicata and 61% in Emilia Romagna. Sugabeet is irrigated principally in Trentino Alto Adige (96%), Sardegna (83%), Campania (83%) and Umbria (81%). Vineyards are irrigated particularly in Trentino Alto Adige (67%), Puglia (62%) and Valle d’Aosta (54%). The irrigation practices are strongly related to the availability of water resources, especially in the South, where the irrigation strategies and irrigated crops are selected on the basis of economic parameters and increase of profit. In fact, the irrigated area for the most crops, except maize and vineyards, has decreased substantially in respect to the census in 1990. The most significant decrease of irrigated land was observed for soya and forage crops, of about 123,000 ha (60%) and 172,000 ha (40%) respectively. On the other side, an increase of irrigated land was observed for maize, of about 115,000 ha (23%) and for vineyards, of about 20,000 ha (13%). The irrigated land in 2000 was for about 100,000 ha lower than in 1990.

Irrigation methods The irrigation methods vary in respect to the irrigated crops, quantity and quality of available water, size and type of management of irrigated farms, and soil and climatic characteristics. In general, the sprinkler irrigation method is the most utilized (on 1,047,000 ha), followed by surface and furrow irrigation (850,480 ha), localized irrigation (366,018 ha) - mainly drippers (290,700 ha), flooding irrigation (202,000 ha) and other methods (2,300 ha) as illustrated in Fig. 7. During the last twenty years, there is a general trend of almost all irrigation methods, except localized irrigation, to shrink the surface area of application.

111



Fig. 7. Irrigation methods in Italy The sprinkler irrigation method is frequently used in Emilia Romagna (162,500 ha), in Veneto (157,500 ha) and in Lombardia (138,500 ha) where field crops as maize, forage crops, sugarbeet, etc are cultivated.(Fig. 8). The surface and furrow irrigation, characterized by low application efficiency, high volumes of water supply, well-managed and dense water distribution networks and well-leveled irrigation fields, are extended mainly for irrigation of herbaceous crops in Lombardia (350,000 ha), Piemonte (211,500 ha), Veneto (86,000 ha) and Emilia Romagna (45,000 ha). The furrow irrigation method is utilized also in Campania, on the surface area of 40,000 ha, for irrigation of vegetables. In this case, short furrows (about 10 m length) with the water flow between 5 and 10 l/s are utilized, realizing in such a way a sort of flooding by furrows.

Fig. 8. The most utilized irrigation methods in the Italian regions with the highest irrigation surfaces

112



Moreover, in Lombardia, where the winter temperatures are very low and frequently below zero, the surface irrigation is utilized with anti-frost purposes on permanent forage crops in order to have green forage during the winter season. For this purpose it is necessary to provide an appropriate field land leveling which permits fast flow of water in the normal direction to the longitudinal axis of irrigation units. In these cases, the irrigation is performed by using a single or double lateral land grading (Fig. 9). The slope of land along the axis perpendicular to the longitudinal irrigation unit is 4 to 10% and the length of water course is between 5 and 20 m. In such a way, the time of flow-off is lower than the time necessary for the conversion of water from liquid to solid state, allowing the superficial soil layers to have temperature greater than zero and to permit the growing of vegetation having green forage also during the winter time.

a)

b)

Fig. 9. Surface irrigation method with double (a) and single (b) lateral land grading (Source: Giardini, 2002) The flooding irrigation method is utilized almost exclusively to irrigate rise, in Piemonte on the surface area of more than 110,000 ha, in Lombardia on the surface area of about 89,500 ha, and in Veneto, Emilia-Romagna, Sardegna and Calabria on a total surface area of about 15,000 ha (Fig. 8). The localized low-pressure irrigation methods (drip, sprayers and “capillary” sub-irrigation) are extended mainly in the Southern regions of Italy, and particularly in Puglia (143,000 ha) and in Sicilia (62,000 ha) while in the North they are utilized prevalently in Emilia-Romagna (38,000 ha). These methods guarantee a high water application efficiency and they are used mainly for the irrigation of orchards and vegetables in the areas where water supply is limited. The sub-irrigation method by regulation of water table depth is used in Veneto, in the areas where shallow water table is controlled by sub-surface drainage systems, and it is applied as a supplementary intervention to rise water table when necessary. The capillary subsurface irrigation is practiced on orchards in Emilia-Romagna, Puglia, Sicilia and Basilicata, burying the dripping laterals with drippers that release slowly herbicides (Trifluralin) to avoid intrusion of roots into drippers. Sprinkler irrigation is realized mainly with self-propelled devices which use side-roll laterals with long jets (sprinklers) which can be substituted sometimes with sprinkling laterals in order to improve water use efficiency and to reduce the working pressure of the system. These equipments have been widely used by farmers for irrigation of field crops due to their capacity to adapt at different field conditions, to move easily, to limit labor requirements and application cost. Recently, there is an attempt to improve distribution efficiency of the high pressure sprinklers with large wetted diameter in windy areas through the application of new generation turbine sprinklers with slow return fluctuating arms and with adjustable angle of the jet until reaching the horizontal position (Fig. 10)

113



Fig. 10. New generation turbine sprinklers with slow return fluctuating arms and adjustable angle of jet The devices with mobile and fixed wings (lateral sprinklers) are presented rarely for irrigation of vegetables while permanents irrigation devices are used prevalently for irrigation of orchards. The irrigation devices like “rangers” and “center pivots” are not frequently used due to small size of farms and presence of obstacles in the field (trenches, windbreaks, electrical cables, etc.). Surface irrigation is applied provided that land leveling was done with adequate furrow distances and sometimes by open ditches 20-30 m far away. This type of lateral infiltration irrigation is used in soils which crack superficially and water can run laterally over long distances.

Crop yield response to irrigation water A more significant development of irrigation techniques in Italy coincides with the general reconstruction of country after the World War 2nd. It was particularly relevant in the Southern parts of the country, where the water shortage problems imposed the construction of dams and water accumulation lakes. At the same time, an intensive research in the field of irrigation had been promoted by the National Research Council (Consiglio Nazionale delle Ricerche - CNR). In 1962-63, these activities resulted in the constitution of a Group for Irrigation Studies (Gruppo di Studio sull’Irrigazione – GRU.S.I.) which has been operated up-to-date in an informal way. At the beginning, GRU.S.I. conducted research prevalently on the yield response to irrigation of herbaceous and tree crops with the aims to evaluate crop water requirements from the agronomic point of view and to optimize both the quantitative and qualitative aspects of crop production under different Italian environments. In fact, it is well known that optimal agronomic crop irrigation requirements do not coincide with the maximum evapotranspiration. The research activities on irrigation have been conducted mainly in Southern Italy where the crop productivity is strongly influenced with limited precipitation, and irrigation represents a fundamental practice in order to increase and stabilize agricultural production over the years. These researches have been conducted prevalently on vegetables and field crops (tomato, pepper, bean, sugar-beet, maize, sorghum, etc.) and, also on the olive trees and vineyards since they are well-adaptable to water stress conditions. The results of numerous experimental works highlighted that the seasonal irrigation volume represents the most important irrigation parameter in the determination of the production of crops under specific environmental conditions. Accordingly, the crop responses to water are presented in this document as the variation of yield, expressed as a percentage of the maximum obtainable yield, in relation to the specific seasonal volume of irrigation. In order to make possible a comparison between the crop productivity of different cultivars in different years and under different environmental

114



conditions, the specific seasonal irrigation volume is expressed as a percentage of the maximum crop evapotranspiration (ETc). In most of the experimental works on the crop response to water, the irrigation events have been programmed using the soil water balance approach with the reference to the maximum crop evapotranspiration (ETc), estimated with different methodologies and with the crop coefficient values (Kc) adopted from the literature or defined for the study areas. The methods based on the monitoring of the soil water content and/or the plant water status have been rarely adopted in the past. In general, different irrigation strategies have been compared maintaining fixed the irrigation intervals and changing the volumes of water applied as a percentage of the optimum water supply corresponding to the 100% of crop evapotranspiration. An example is given for some herbaceous crops in Figure 11 showing the relations between the crop yield, expressed as a percentage of the maximum yield obtained during the experimental period, and the specific seasonal irrigation volume, expressed as a percentage of ETc, obtained in Metapontino (Policoro, Southern Italy). The relationships reported in Figure 11 are obtained adapting to the experimental points the Mitcherlich model modified by Giardini and Borin (1985) as:

[

][

y = y m 1 − 10 − c (b −d ) ⋅ 10 − k (b + d )

2

] [1 + 10

1− c (b − d )

]

where: y is crop yield; ym is the maximum obtainable crop yield under non-limited supply of the factor (parameter) under study; c is a coefficient of action (or of increase), indicating the rapidity of the achievement of the maximum yield; k is a coefficient of depression, indicating the tendency of y to decrease after the achievement of the maximum value; b is the quantity of the factor under study available for the crop in natural conditions, and d is the quantity of the factor under study applied under specific experimental conditions. Yield (% of the maximum)

ETc (%)

Fig. 11. Trend of some herbaceous crops yield expressed as percentage of the maximum obtainable yield in relation to the seasonal irrigation volumes expressed as percentage of ETc. The curves have been obtained adapting to the experimental points Mitscherlich model modified by Giardini and Borin. Negative values indicate the quantities of natural water, from precipitation, groundwater table and soil water content, utilized by the crops (adapted after Venezian Scarascia et al., 1987). The research work was carried out in a deep, silty clay loam soil with moisture levels at field capacity and wilting point equal to 31.5 and 15% of dry soil weight, respectively; the water table was between 150 and 200 cm below the ground surface during the rainy season and in the dry months, respectively. The climate is typically Mediterranean with 600 mm average annual rainfall and 16°C as

115



mean annual temperature. As an average, 79% of rainfall occurs in autumn-winter season (from October to March) while the highest averages temperatures (between 22° and 25°C) are recorded in June, July and August; consequently, the dry period extends from the beginning of May to the end of August. The crops considered in the study were autumnal and spring-sown sugar-beet, shell bean and dry bean, tomato, pepper, eggplant, spring and summer-sown grain maize. Utilization of natural water resources (rainfall and ground water) by crops increased as the cycle extended into the rainy season. In this regard, figure 11 shows that the amount of natural water actually used by summer-cycle crops (shell bean, sown in June) is only about 2-3% of ETc and rises to as much as 50% with crops sown in autumn and harvested in summer (autumnal sown sugar beet): the corresponding water volumes are 3 -1 150-180 and 3000 m ha , respectively. Moreover, the yield irrigation water efficiency is much greater for crops whose cycle extends - at least in part - into the rainy period (maize grown as the main crop, eggplant, sugar beet whether sown in spring or in autumn) than far spring-summer, or summer crops (pepper, tomato, maize grown as cash crop); for the first group of crops indeed the curves are steeper, as compared to the second group, because of the higher values of the action coefficient (c) which means better water use efficiency (Fig. 11 and Table 5). Table 5. The parameters of the Mitscherlich equation, flex point coordinates and seasonal irrigation volumes at 100% of yield Equation-parameters

Flex point coordinates

Ym

b

c

(% of the max yield)

(% of the ETc)

(ha / %ETc⋅10-3)

Water volume % of ETc

Yield

3 m /ha

(% of the max)

Seasonal irrigation volume at 100% of (m3/ha)

Tomato

78.5

13.9

23.0

29.6

1435

35.3

4734

Pepper

84.9

15.1

20.1

34.6

2004

38.2

5800

Spring maize

91.0

15.4

27.6

20.8

994

41.0

4181

Summer maize

94.9

10.0

16.4

51.0

1603

42.7

3085

Eggplant

82.9

23.1

30.4

9.8

441

37.3

4795

Shell-bean

96.9

2.2

15.8

61.1

1898

43.6

3109

102.0

4.9

14.9

62.1

2121

45.9

3413

Spring sugar beet

75.9

39.1

32.7

-8.6

-

34.2

7193

Autumnal sugar beet

82.0

49.4

23.1

-6.2

-

36.9

4961

Dry-bean

Consequently, the greatest increments in yield were recorded with seasonal irrigation volumes around 61-62% of ETc (1898-2191 m3 ha-1) for bean (a typically summer crop) between 34 and 20.8% of ETc (2004-994 m3 ha-1) for pepper and maize grown as the main crop (spring-summer cycle crops), and without irrigation for sugar beet (grown either as spring or autumnal crop): the yields corresponding to such maximal increments were respectively 43.6, 45.9, 38.2, 41.0, 34.2 and 36.9% of peak yields recorded during the trial period (Fig. 12 and Table 5 to compare the flex point coordinates of the curves: the amounts of water and the corresponding yields). These results stress the fact that yields are less affected by irrigation when dealing with springsummer and autumn-summer crops, than with summer crops. Fig. 13 shows indeed that to obtain as much as 70% of the yield recorded during the trial period the seasonal amount of irrigation water had to be as high as 75% of the calculated ETc for summer and spring-summer crops and about 25% of the calculated ETc for autumn-summer or winter-summer crops. Seasonal irrigation volumes 3 -1 corresponding to 100% of estimated ETc ranged from minimum of 3100 m ha to a maximum of 3 -1 7200 m ha according to the length of the crop cycle and the season of the year during which the crop cycle develops. The lowest seasonal irrigation volumes were recorded for very short cycle crops (72 days) – including summer crops like shell bean - and also for those crops which crop cycles develop during seasons with a low evaporative demand of the atmosphere, as it happens in the case of maize grown as a forage crop. Conversely, the heaviest seasonal volumes were recorded for longer-cycle crops (more than 150 days) growing during the months when the evaporative demand increases, such as spring sown sugar beet (Fig. 14).

116



Yield (% of the maximum)

ETc (%)

Fig. 12. Yield of some herbaceous crops as a function of seasonal water volumes expressed as percentage of estimated ETc with the indication of the flex points of different curves (adapted after Venezian Scarascia et al., 1987).

Yield (% of the maximum)

ETc (%)

Fig. 13. Yield of some herbaceous crops in relation to the seasonal irrigation volumes, with the indication of the seasonal irrigation volumes corresponding to the 70% of the maximum yield (adapted after Venezian Scarascia et al., 1987).

117



Fig. 14. Seasonal irrigation volumes of several horticultural crops in relation to the adopted irrigation regime (adapted after Venezian Scarascia et al., 1987). In conclusion, a very short cycle of crops makes the best use of irrigation water, irrespective of the season of their growth cycle. Similar inference can be drown for crops sown in autumn or early in spring as they make a good use of natural available water resources. When irrigation water is limited and crops that respond rapidly to irrigation (such as sugar beet and maize grown as main crop) are grown simultaneously to crops that respond gradually (such as pepper, tomato, maize grown as forage crop and shell bean), then, the latter group of crops should be irrigated more than the former.

Deficit irrigation strategies A particular attention has been given to the studies on regulated water stress based on different crop sensitivity to water supply during various phenological stages and on the crop physiological mechanisms of response to water stress. Deficit irrigation techniques have demonstrated a high validity for water saving in the case of various tree crops without particular negative effects on crop production and farmer’s income in both Southern and Northern Italy. However, the technique of controlled deficit irrigation can be applied on the already grown trees since the deficit irrigation can provoke negative impacts (later start of production and overall decrease of productivity) if applied during the first three-four years since plantation. A synthesis of results of the numerous deficit irrigation experiments carried out in Emilia-Romagna (Northern Italy) on peach tree is given in Fig. 15 subdividing the vegetative cycle of peach tree in 4 principal phases: phase 1 – from the start of flowering to the formation of small fruits (of 3-4 cm of diameter); phase 2 – from the end of the previous phase until the hardening of the pit; phase 3 – from the hardening of the pit until the harvesting; phase 4 – from the harvesting until the fall of the leaves. Figure 15 illustrates that the water stress was induced during the phases 2 and 4. A controlled water stress during the phase 2 does not favour development of shoots which reduces the competition for assimilates between the shoots and fruits; similarly, during the phase 4 it reduces vegetative growth and favours the induction of buds to flowers and fruit leader. The overall reductions

118



of irrigation volumes in respect to full irrigation in a normal year under Emilia-Romagna climatic conditions were estimated between 56 and 68% for the medium early and early cultivars and clay soil and between 20 and 23% for the late cultivars without significant differences related to the soil type (Table 6). The results (Fig. 16) indicate that the regulated deficit irrigation technique has increased crop production in respect to traditional irrigation, has maintained the average weight of fruits, has improved the flowering in the successive years and has reduced the necessity for pruning. Similar results have been obtained also in the experiments on peach and nectarine trees carried out under Southern Italy climatic conditions.

Fig. 15. Graphical presentation of the stress thresholds to apply on the peach tree under regulated deficit irrigation treatments (adapted after Mannini, 2004) Table 6. Percentage of seasonal irrigation volumes saved by controlled water stress on peach in respect to normal irrigation regime (Source: Mannini, 2004) Interspace between rows cultivated

Interspace between rows grassy

SOIL

Early cultivars

Medium early cultivars

Late cultivars

Early cultivars

Medium early cultivars

Late cultivars

Sandy

44

38

20

38

34

20

Loam

58

59

20

52

46

23

Clay

68

56

22

60

51

23

The studies of regulated deficit irrigation has been done also on the herbaceous crops in Southern Italy giving different results in respect to those obtained with orchards. In fact, serious drops of production can be observed even in the cases of limited water reduction during the non-critical phenological stages. Four years of investigation on the regulated deficit irrigation of maize have been done in Southern Italy (Policoro, Basilicata). The experiment was based on suspending one or two irrigations or doubling irrigation volumes in correspondence to different phenological phases (a – when crop has achieved 1 m height, during the crop growing stage; b – at the tassel emission; c – at beginning of the milky stage; d – at the beginning of the waxy stage). The results have shown that all phenological phases demonstrated certain sensitivity to water stress. Anyway, the most sensitive phase almost always corresponded to the tassel emission and, in particularly dry years, to the phase of intensive crop growth. These results indicated that maize is not well adaptable to the Southern Italy climatic conditions where the spring-summer periods are characterized with scarce precipitations and very high evapotranspiration demand. Consequently, maize should be fully irrigated under these climatic conditions. In fact, maize is rarely cultivated under Southern Italy climatic conditions because this crop is very sensitive to water stress and it should not be grown under deficit irrigation practices.

119


Yield

Average fruit size

135

37 130

36 g

Fruit (t ha-1)

38

35

125

34 33 32

R.D.I.

120

Test

R.D.I.

Fruit set

76 74 72 70 68

Test Prunings

Wood (N. m-1)

78 Fruit (N. m-1)


R.D.I.

3 2 1 0

Test

R.D.I.

Test

Fig. 16. Productive and vegetative effects of water stress on peach tree grown as an espalier (by Chalmers) (adapted after Mannini, 2004). B

qg

8.0

C

AB

g

Grain yield(t ha-1)

7.0

q

B

qq

6.0

B

B B

x (s+b)

B

x(s)

qA

B

g

B

gA 5.0

B

B

x

B

x x

x C

u

BC

u

4.0 AB

3.0 A

u

2.0 x

A

u

A

x(b)

q q g

x

1.0

u

Valenzano 1986 g Valenzano 1987 x Policoro u Gaudiano

0.0 0

1000

2000

3000

Seasonal volume of irrigation water (m ha-1 ) Fig. 17. Variations of wheat production under different irrigation treatments. The values assigned with the same letter are not significantly different at 0.01 P according to the Newman-Keuls method. (s) – irrigation only at the sowing; (b) – irrigation only at the booting phase; (s+b) – irrigation at sowing and booting phase. Finally, in the Southern Italy environments, characterized with high precipitation variability which contributes to the instability of agricultural production even of non-irrigated autumn-spring crops (e.g. wheat), there is a frequent application of supplemental irrigation strategies. This helps in stabilizing agricultural production and improving the water use efficiency of precipitation. Several experiments were carried out in Southern Italy on different wheat cultivars grown in deep soils with water

120



availability of 8.4% of dry soil weight (Gaudiano di Lavello – Basilicata) and of 16.6% of dry soil weight (Policoro – Basilicata) and on shallow soil with water availability of 13.0% of dry soil weight (Valenzano – Puglia). The irrigation strategies included the application of water only during the critical phenological stages (at sowing, at the booting phase and at both sowing and booting phase) and during the whole growing cycle with different levels of limitations. The results of these investigations, shown in Fig. 17, indicated that in particularly dry years one irrigation immediately after sowing (example of Policoro in 1986) with water volume of 770 m3 ha-1 can be sufficient to increase production from 2.0 t ha-1 to 4.7 t ha-1, while any additional irrigation did not contribute to further augment of grain yield.

STUDIES ON CROP WATER REQUIREMENTS The researches on crop yield response to irrigation water required the intensification of the studies on the adaptability of empirical methods for the estimation of reference evapotranspiration to different Italian agro-climatic conditions. These studies were necessary in order to estimate and/or foreseen better crop water requirements for both the irrigation management purposes and the realization of irrigation projects. A particular attention has been given to the methods indicated in both the FAO Irrigation and Drainage paper n°24 (Dorenboos and Pruit, 1977, 1987) and in n°56 (Allen et al., 1998). For the implementation of studies on crop water requirements, in many Italian regions have been constructed the lysimeters of different characteristics by means of both functionality and size. Type, dimensions and number of lysimeters used in various Italian locations are reported in Table 7, while the spatial distribution of the lysimetric stations is indicated in Fig. 18. Table 7. Type, dimensions and number of lysimeters used in various Italian locations yp , y Type

Surface area 2 (m )

Depth (m)

Presence of guard

Underground(U) or Aboveground (A)

Location and number

2x2 = 4

1,30

yes

U

Policoro (6) Metaponto (2) Foggia (4) S. Prospero (4) Guiglia (4) Gela (2) Roma

2x2 = 4.

2,20

yes

U

Cadriano (2)

2x2 = 4

1,00

yes

U

Polignano (2) Cadriano (2)

1,25x1,25 = 1,56

1,40

no

A

Pisa (6)

1x1 = 1

1,50

yes

U

Legnaro (20)

2x2 = 4

0,50

no

U

Vitulazio (16)

2,75 m*; 5,94

1,50

yes

U

Sassari (4) **

2x2 = 4

1,30

yes

U

Policoro (2) Rutigliano (1) Gaudiano (1) Villa d'Agri (1)

3 m**; 7,07

2,15

yes

U

Campo Volturno (4)

l) DRAINAGE a) groundwater (70-110)

b) free percolation

2) WEIGHING a) mechanical b) with loading cells * circular ** for tree crops

Water consumption have been valued with drainage lysimeters by using the water balance equation weekly or 10-days period, whereas it was measured with weighing lysimeters as a difference in weight at the beginning and the end of the period under consideration, generally on a daily basis, taking into account natural hydrological inputs, irrigation, and the quantities of drained water (Tarantino and Onofrii, 1991).

121



Fig. 18. Location of some lysimeter stations in Italy Lysimeters have not been used only for the research on the adaptability of different methods for reference evapotranspiration estimates under various Italian climatic conditions, but also for the studies on crop water requirements, or maximum crop evapotranspiration (ETc), during the growing cycle of numerous herbaceous crops and some tree-crop species. Daily values of ETc measured for various species have been rationed with the equivalent values of class “A” pan evaporation (E) and/or reference evapotranspiration (ETo), calculated with different methods, in order to obtain corresponding crop coefficients:

Kc' =

ETc E

Kc =

ETc ETo

The research locations, corresponding cultivars, years of experiments and some growing and productive information are given in Table 8, the reference parameters used for the calculation of crop coefficients (Kc’ and Kc) are reported in Table 9, while in Tables 10 and 11 are presented the crop coefficient values (Kc’) related to the class “A” pan evaporation (E). In Figures 19, 20 21 is given the variation of Kc (derived from the ratio between the measured ETc and ETo calculated by the Penman-Monteith equation) for some vegetables (muskmelon and eggplant) cultivated under plastic mulches and without them. Data reported in Table 10 confirmed that the lowest Kc’ values, in the range between 0.1 and 0.6, were observed during the initial growing stage, about 30 days after sowing or planting, when the water losses are prevalently due to soil evaporation. The highest values, between 0.85 and 1.50, were observed when the full crop development has achieved and LAI reached the maximum values, i.e. when the water is almost exclusively consumed in the process of transpiration. The Kc’ values were decreasing gradually with the approximation of the end of crop growing cycle, in relation to the vegetative state of the crops at harvesting.

122



Variability of Kc’ values during the initial crop growing stage is related to the humidity of the superficial soil layers. In fact, the highest value (0.62) was registered for wheat, an autumnal sowing crop, when the frequency of precipitation was relatively high and ETo was limited, and, therefore, the soil water content in the superficial soil layers was pretty elevated. In fact, is well-noted that direct water losses by soil evaporation increases with the increase of humidity of superficial soil layers. The Kc’ values resulted substantially different in the Northern Italy environments, where the crop growing cycles tend to make longer, in respect to the Southern Italy, where they are shorter: evident examples are spring sowing tomato and sugar beet (Table 10). Notable differences are also observed on the values of Kc’ of maize and sorghum grown under different climatic conditions: higher values were in the Northern Italy (locations of S. Prospero and Guiglia) and lower in the Southern Italy (locations of Policoro and Foggia), lower values for the early cultivars (FAO class 200-400), higher values for the hybrids with longer growing cycle (FAO class 600-700). Table 8. Crop growing parameters of some experiments on Kc carried out in Italy Crop

Location

Years of experiment

Cultivar

Sowing date (2)

harvesting

Plant density

Yield (t/ha)

Average ETc (mm)

September August Decem. April Decem. April

10 10 1 1

96.5 117.5 11.4 29.3

669 652 540 557

6 4 16 NR NR 65

13.6 16.4 17.0 20.8 (1) 16.9 (1) 14.8

120 374 233 939 692 276

On-field growing Herbaceous crops Sugarbeet

Artichoke Cabbage Broccolo Summer cultivar Winter cultivar Cetrioli Alfalfa (l) String bean Bean (type borlotto) Fresh Dry Wheat st Sunflower 1 harvesting

Policoro Cadriano Policoro Polignano

Monohill Monogen Locale di mola Locale di mola

Policoro Policoro Policoro S.Prospero Guiglia Policoro

Green duke Clipper Pioner e Bounty Bresaola Bresaola LIT 551

Policoro Policoro Policoro Foggia S. Prospero Guiglia

Lingua di fuoco Lingua di fuoco Salapia Luciole Luciole Luciole

1984 1984 1985-86 1981-82 1981-82 1981-82

June June November April April April

August September June August September September

44 44 49 ears/m2 5.0 5.1 5.0

7.5 3.0 6.7 3.9 3.6 3.4

432 479 475 710 571 605

Mirage Mirage

1986 1986

June July

October October

6 6

3.3 2.9

537 452

1974-75-76

April

September

8

10.1

511

1976-77 1976-77 1976-77

April April April

September October October

6 6 6

12.0 13.5 12.3

686 589 587

1973-7

May

September

6

10.0

450

1986 1986 1978-79 1976-77-78 1977-78 1983-84-85 1975-76 1984 1984

June July March April April May May May June

October October August September September October October October October

8 8 4 6 40 35 30 40 35

13.2 8.6 50.0 87.0 80.0 3.7 4.0 5.2 2.9

582 457 600 546 451 861 500 618 420

Sunflower 2nd harvesting Pisa after barley Pisa after wheat st maize from granella 1 harvesting

Foggia Guiglia Guiglia Guiglia S. Prospero Legnaro Policoro

Dekalb XL 304 FAO 200 Dedalo 95 FAO 400 Titano FAO 700 Titano FAO 700 Dekalb XL 342 FAO 606 Leveret 400 Leveret 400 Bintje Ventura Roma VF Kig SOY TXR 505 Hodson 78 Arrok Dekalb XL FAO 200 NK 121 FAO 200 54BR FAO 200 NK 180 FAO 400 Savanna 5 FAO 600 Savanna 5 FAO 600 NK 180 FAO 400 Seven R

Gaudiano

Policoro Foggia S. Prospero Guiglia Legnaro Maize 2nd harvesting Potato Tomato Soya 1st harvesting Soya 2nd harvesting Sorghum

Spinach Muskmelon mulched Muskmelon non-mulched Muskmelon mulched Muskmelon non-mulched Eggplant mulched Eggplant non-mulched

Pisa Pisa Legnaro Policoro Legnaro S. Prospero Legnaro Cadriano S. Prospero Policoro

1975-76 1981 1974-75-76 1974-75-76-77

March March August August

1977 September December 1986-87 July Sept. November 1977-78-79 July October 1970-79-80-81-82 April October 1970-79-00-81-82 April October 1977-78-79 April July

1977-78

May

October

25

12.7

690

1978-79-80 1983-84-85 1983-84-85 1983-84-85 1983-84-05 1974- 75 1978

April May May May April May February

September September September September September October April

30 50 35 35 35 16 64

Nabucco

2001-2003

June

August

0.5

Policoro

Campero

1999

May

August

1.0

Policoro

Tasca

2003

May

July-August

2.0

10.6 8.2 6.9 6.2 8.9 9.0 31.8 39.4 26.7 30.2 27.8 96.5 55.6

648 512 589 630 624 465 153 310 257 229 320 720 703

Tree crops Orange tree Sassari Washington navel 1987 4° year December Apricot tree Ponticelli Cafona 1981-82-83-84-85 6° year July-January 400 plants/ha from 4 to 28 380 Olive tree Sassari Tondo di Cagliari 1987 5° year January Peach tree Livorno 1600 plants/ha 1 Yield of alfalfa refers to the total dry matter of 5 years of experiments obtained from 4-5 cutting per year; average annual consumptions refers to the period May-June. 2 For tree crops it is intended as the years after planting.

123



Table 9. Crops and methods used for evaporation measurement and reference evapotranspiration estimates at different locations in Italy Evaporation (E) Crops

Sugar beet

Artichoke

Locations

Class “A”

Policoro

X

Cadriano

X

Policoro

X

Putignano

X

Policoro

X

Wild

Reference evapotranspiration (ETo) BlaneyRadiat. Penman Grass Criddle FAO FAO festuca FAO X

X

Policoro

X

Bean

Policoro

X

X

Wheat

Policoro

X

X

X

X

X

X

X

Foggia

X

X

X

X

S. Prospero

X

X

X

X

X

Guiglia

X

X

X

X

X

Pisa

X

X

Policoro

X

X

X

X

Foggia

X

X

X

X

X

S. Prospero

X

X

X

X

X

X

Guiglia

X

X

X

X

X

X

Legnaro

X

X

X

X1

X

X1

X X

X

X

Pisa

X

Legnaro

X

X

Tomato

Poticoro

X

X

X

Legnaro

X

X

X

Soya

Cadriano

X

S. Prospero

X

Legnaro

X

Pisa

X

Policoro

X

Spinach

Penman- Other Monteith methods

X

Potato

Sorghum da granella

X

X

Policoro

Maize da granella

X

Turc Thornthwaite

X

Cabbage Broccolo winter Cabbage Broccolo summer String bean

Sunflower

X

Epan FAO

X

X

X2

X

X2

X X X

X

X X

X1

X X

X

X

X X

X

X1

X

X

X X

X

Foggia

X

Guiglia

X

X

X

X X

X1

S. Prospero

X

X

X

X

X1

Legnaro

X

X

X

X1

Policoro

X

Eggplant Pepper Muskmelon mulched and non Gaudiano

X

Policoro

X

Eggplant

Policoro

Apricot tree

Ponticelli

X X

Orange tree

Sassari

X

Olive tree

Sassari

X

Peach tree

Livorno

X

1

Formula of Tombesi-Lauciani. 2 Formula of Blaney-Morin, Hannon, Hargreaves, Hedke, Ivanov, Helse, Loury-Jensen.

124



Table 10. Measured Kc’ values (ETc/E ratio) of some crops grown under different conditions in Italy Days after sowing or planting Crop

Type of crop

Location

Sugarbeet

Spring

Policoro Cadriano Policoro Polignano

0.28 0.40 0.24 0.44 0.52 0.61 0.68 . 0.73

Winter

Policoro

0.31 0.34 0.40 0.50 0.70 0.86 0.95

0.95 0.94

0.37 0.28 0.36 0.60 0.62 0.30 0.25 0.15 0.48 0.56 0.38 0.41 .

0.87

0.86 0.82 0.80

0.84 0.83 0.80 1.00 1.20 1.45 1.46 1.38 1.03 1.14 1.20 1.02 0.91 1.60 1.30 0.79 1.07 0.62 0.80 0.94 1.15 1.47 1.04 1.00 0.85 1.10 0.95 0.97 0.95 0.82

0.70 0.85 1.16 1.50 1.50 1.38 1.20 1.00 1.14 1.27 1.20 1.00 1.68 .1.34 0.87 1.10 0.71 1.00 1.10 1.13 1.44 1.07 1.00 0.90 1.22 1.10 1.18 0.97 0.88

10

Artichoke (1) Cabbage Broccolo (2)

Summer

Policoro Policoro Policoro Policoro Policoro Foggia 1st harvesting Guiglia S. Prospero Pisa 2nd harvest. after barley Pisa 2nd harvest. after wheat Policoro Maize 1st harvesting Foggia S. Prospero Guiglia Legnaro Pisa 2nd harvest. after barley Pisa 2nd harvest. after wheat Potato Legnaro Tomato Policoro Legnaro Cadriano Soya 1st harvesting Legnaro S. Prospero 2nd harvesting Pisa 2nd harvest. after wheat Policoro Sorghum 1st harvesting early Foggia 1st harvesting medium Guiglia 1st harvesting early Guiglia 1st harvesting medium Guiglia 1st harvesting late S. Prospero 1st harvesting late Legnaro 1st harvesting medium Spinach Policoro 1 for artichoke, n° of days of the vegetative recover Cetriolo String bean Bean Wheat Sunflower

2

0.40 0.54 0.52 . 0.35 . 0.34 0.45 0.50 0.46 0.43

20

0.42 0.45 0.54 0.51 0.62 0.40 0.28 0.20 0.64 0.69 0.42 0.47 .

30

0.45 0.62 0.70 0.53 0.64 0.53 0.32 0.38 0.86 0.84 0.51 0.60

0.15 0.43 0.48 0.52 0.64 0.58 0.70 0.40 0.45 0.41 0.42

0.40 0.63 0.56 0.49 0.47 0.30 0.30 0.30 0.35 0.30 0.43 0.38 0.44

0.50 0.92 0.68 0.59 0.56 0.34 0.34 0.34 0.45 0.55 0.52

40

0.67 0.98 0.82 0.70 0.66 0.64 0.50 0.60 1.08 1.04 0.70 0.80 0.40 0.36 0.57 0.85 0.86 0.70 0.55 0.45 0.25 0.60 1.05 0.94 0.71 0.65 0.45 0.50 0.55 0.60 0.67 0.61

50

60

0.49 0.28 0.75 0.81

0.60 0.32 0.82 0.88

0.71 0.90 0.94 0.82 0.68 0.75 0.77 0.90 1.36 1.24 0.87 0.95 0.70 0.60 0.67 1.15 1.04 0.70 0.80 0.50 0.47 0.74 1.12 1.22 0.85 0.77 0.60 0.70 0.73 0.72 0.80 0.68

0.84 0.70 0.88 0.86 0.74 0.87 1.00 1.20 1.46 1.36 0.98 1.06 1.00 0.85 0.80 1.40 1.19 0.72 0.99 0.55 0.70 0.85 1.15 1.42 0.97 0.95 0.75 0.90 0.90 0.85 0.90 0.75

70

80

90

140

150

160

170

1.00 0.91 1.07 1.20

0.94 0.96 1.06 1.20

0.84 1.00 1.04 1.15

0.68 1.05 1.08 1.00 0.96 0.90 0.85 0.78 0.70 0.65 1.09 1.02 0.92 0.88 0.80 0.75 0.70 0.65

0.95 0.95 1.00 0.50 0.68

0.95 0.90 0.77 0.60 0.57 0.38 0.28 0.50 0.50 0.40

0.56 0.78 1.18 1.26 0.90 1.40

0.45 0.55 1.05 1.00 1.10 0.80 0.48 0.75

0.90 0.78 1.08 1.40 1.30

0.81 0.71 0.69 1.13 1.14 1.14 1.10 1.30 1.18 0.98 0.75

0.95 1.10 1.20 0.60 0.72 0.60 0.71 0.85 1.20 1.34 1.01 1.36 1.40 0.95 0.90 1.00 1.45 1.35 0.64 0.88 0.92 0.75 0.90 1.10 0.76 0.91 1.02

. 0.72 0.86 0.50 0.75 0.90 0.63 0.86

190

200

210

240

130 1.02 0.86 1.08 1.15

0.25 0.92 1.25 1.40 0.90 0.92 0.76 0.82 0.96 1.25 1.35 1.07 1.46 1.30 0.95 1.01 0.91 1.50 1.30 0.70 0.90 1.00 0.96 0.85 1.12 1.20 1.17 0.95 0.98

180

230

120 1.03 0.78 1.07 1.12

0.43 0.88 1.25 1.55 1.20 1.30 1.00 0.93 1.07 1.28 1.30 1.05 1.64 1.31 0.93 1.09 0.82 1.30 1.20 1.00 1.20 1.00 0.99 0.90 1.15 1.23 1.24 0.97 0.93

100

220

110

0.72 0.81 0.90 0.97 1.01 0.38 0.46 0.54 0.63 0.70 0.88 0.93 0.97 1.02 1.05 0.93 0.99 1.03 1.07 1.10

0.55 0.78 0.60 0.75 0.80

for these crops, n° of days after planting

125



Table 11. Crop coefficient values of some tree crops related to the class “A” pan evaporation Month Crop

Location

Apr May June July Aug Sept Oct

Authors

Apricot tree (cv. Cafona) Drip irrigation Ponticelli (NA)

0.70

0.33

0.55

0.64

0.68

0.73

0.81

Ruggiero, 1986

Sprinkler irrigation Ponticelli (NA)

0.64

0.52

1.13

0.80

0.80

0.91

0.68

Ruggiero, 1986

Sassari

-

-

0.17

0.28

0.35

0.38

0.40

Dettori (unpublished data)

Sassari

-

-

0.47

0.46

0.51

0.52

0.40

Dettori (unpublished data)

0.55

0.81

1.01

1.00

Orange tree (cv. Washington navel; 4th year after planting, G.C.I. 20%) Olive tree (tondo di Cagliari; 5th year after planting, G.C.I. 30%) Peach tree (1)

Livorno

Natali et al., 1984

1

On-field data. G.C.I. – Ground Cover Index

In Table 10 is shown that the peak Kc’ values of sunflower were anticipated a) in the case of sunflower intercropping after barley and wheat in respect to the main crop and b) in the case of growing in a valley in respect to hilly area (S. Prospero in respect to Guiglia). Moreover, the Kc’ values of sunflower are higher in the case of cultivation under Northern Italy conditions (Guiglia) in respect to Southern Italy (Foggia). For soya, the peak Kc’ values resulted more anticipated and lower at the second harvesting which is related to the time of sowing and to the local environmental conditions. The peak Kc’ values of some herbaceous crops (such as spinach, potato, bean, cucumber, cabbage, broccoli, wheat and artichoke) were almost always lower than 1.0, except for the artichoke with the values around 1.1. The Kc’ values of tree crops change slightly during the vegetative cycle, although they can vary notable between the species in relation to the density and the age of plants and applied irrigation method: the Kc’ values are greater in the case of irrigation with sprinkler method than with drip irrigation. The Kc’ values obtained under Italian climatic conditions result higher than those recommended in the FAO Irrigation and Drainage papers, especially for the herbaceous crops during the full development phase. In fact, the Kc values reported in the FAO document represent the average data from different environmental conditions and cultivars, while the data given in this document refer to the specific environmental conditions, agricultural practices, cultivars and irrigation methods which can notable influence the Kc values. The Kc values of crops cultivated under plastic mulches (muskmelon and eggplant) have been obtained for the Southern Italy in Lavello (Potenza) and Policoro (Matera). The data obtained for muskmelon in the location of Lavello (Fig. 19) indicate that the growing cycle of mulched crops (Fig. 19a) is shorter than of non-mulched crops (Fig. 19b) and that the Kc values at the beginning of the full development phase (10 days after planting) and immediately after the start of harvesting are greater, while during almost the whole period of harvesting are lower. On the other side, the Kc values of nonmulched crops were higher only during the first 10 days after planting. The higher Kc values of muskmelon grown under plastic mulches during almost the whole growing cycle are related to the greater vegetative development of mulched crops; it is also confirmed by the greater LAI values. However, the mulched crops as had a rapid and anticipated development manifested the symptoms of an earlier senescence of leaves which resulted in a fast reduction of Kc values. Furthermore, these data indicate how the duration of phenological phases of muskmelon is notable shorter than that reported in the FAO Irrigation and Drainage Paper n° 56 (Allen et al., 1998), independently of mulching. Moreover, as it is clearly demonstrated in Figure19, the Kc values obtained at location of Lavello are notably higher than those indicated in the FAO documents.

126



Fig. 19. Relation between estimated and measured by lysimeter crop coefficient Kc during muskmelon cycle cultivated with (a) and without mulch (b) in 2001 and 2003 in Lovello – Southern Italy (from Lovelli et al., 2004).

Fig. 20. Crop coefficient data (ETc/ETo ratio) of muskmelon cultivated under mulches and without mulches in Policoro – Southern Italy (from Cantore et al., 2005). The Kc values obtained by Cantore et al. (2005) on mulched and non-mulched muskmelon (Fig. 20) grown in Policoro (Southern Italy) are very similar to those obtained in Lavello. In fact, the Kc values of muskmelon cultivated under mulches are lower during the initial development phase, in the

127



first 10-15 days after planting, in respect to the non-mulched crops. However, the mulched muskmelon reached more rapidly the full development phase and the Kc values for mulched crops are higher than those for the non-mulched crops. Moreover, the mulched crops have demonstrated faster and more intensive development as compared to the non-mulched crops, followed by a rapid and anticipated senescence of leaves. In Policoro, the Kc values of mulched and non-mulched eggplant (Yared Tesfagaber, 2004) were very similar to those of muskmelon, although with less remarkable differences. In fact, the Kc values of mulched crops were slightly lower during the first 20 days after the planting and they were slightly higher during the successive growing phase, with the very similar phenological phases (Fig. 21). It is interesting to emphasize that in Policoro, the yield production of both mulched eggplant and muskmelon crops resulted greater than the yield of the non-mulched crops, although the water consumption was slightly higher. In fact, in the case of the cultivars grown under mulches, the yield water use efficiency was higher. Furthermore, the Kc values of these crops grown in South Italy are higher than those reported in the FAO documents which indicates that they are influenced non only by the environment in which they are cultivated but also by the cultivars and adopted agronomic practices.

Fig. 21. Crop coefficient data (ETc/ETo ratio) of eggplant cultivated under mulches and without mulches in Policoro – Southern Italy (from Yared Tesfagaber, 2004).

WATER USE EFFICIENCY AND AGRONOMIC PRACTICES FOR IMPROVEMENT In the agriculture field, the term “Water-Use Efficiency” (WUE) was introduced by Viets in the middle of sixties (Viets, 1962). Since that time, it has been generally used to describe the relationship between the crop growth development and the amount of water consumed, thus Stanhill (1986) called it “physiological water use efficiency”. The physiological water use efficiency is more difficult to be conceived as a proper efficiency, as it is not a non-dimensional value and it does not represent an output/input ratio of only one entity. In fact, it describes a process in which water is consumed to produce new entities (e.g. biomass, yield, etc.), and a maximum value attainable by theory does not exist for reference (Monteith, 1984). The physiological efficiency is largely utilized by a wide community of scientists (plant and crop-eco-physiologists, agronomists) and it can be applied at different space- and time-scales as illustrated in Table 12.

128



Table 12. Major definitions of water use efficiency terms, as reported by Steduto (1996). Term

Definition

A T

Photosynthetic WUE

Time scale

Space scale

Minutes, hours

Leaf

Hour, day, season

Canopy

Week, season

Plant, canopy

Season

Plant, canopy

tf

∫

Carbon Water Flux Ratio (CWFR)

NCF

t0 tf

∫

ET

t0 tf

Biomass WUE (BWUE)

∫

biomass

t0

tf

∫

ET

t0

Yield WUE (YWUE)

biomass WUE × HI

In this paragraph is given the state of art of WUE and agronomic practices to improve WUE in Italian agriculture under field conditions. It is based on the evaluation of the national scientific literature and technical reports especially focusing on the Southern Italian region. Water use efficiency values of many field crops, grown under optimal conditions (Table 13) and submitted to some agronomic techniques, such as irrigation (Table 14), fertilization (Table 15), rotations (Table 16), mulching & early sowing (Table 17) are reported. In all the tables, water use efficiency is calculated as the ratio of the above ground biomass and/or the yield over the amount of water used, determined by different methods, and expressed as kg m-3. Table 13 shows as, although all the studies refer to no-limiting environmental conditions and to environments with similar weather conditions in Southern Italy, there exist a great variability in the above-ground biomass WUE values among crops. In fact, although it is quite widely acquainted from the literature the superiority of C4 species to use water more efficiently than C3 species, due to the higher efficiency to fix CO2, their values may overlap or overcome those normally found for the C3, as it occurs in the study of Rubino et al. (1999). In this case the very high values of biomass WUE of sugarbeet (8.0 kg m-3) and rapa (14.0 kg m-3) are explained on the basis of the high net assimilation rate linked to the high translocation efficiency of yielded sucrose to the roots in the former crop and of the very low transpiration rate during the winter season in the latter crop. Nevertheless, it is important to highlight that the biomass WUE value of sugarbeet refers to the total biomass, including the heavy roots, and consequently it is difficultly comparable with the others. In the same study, very high yield WUE values are found for celery, lettuce, rapa, pepper and ascribed to the short crop cycles associated with the very elevated water content (about 85-95%) in the marketable parts of all these crops. The results obtained in a recent work carried out by Steduto and Albrizio (2005) to compare biomass WUE among different crops (sunflower wheat, chickpea and sorghum) indicate large variability in WUE values, also within the same C3 group. From this study it is emerged the need to normalize the amount of water evapotranspired by the climate (vapour pressure deficit and/or reference evapotranspiration), in order to compare the WUE values of crops grown in different season and/or year and climatic conditions. Similar conclusions have been reached also by Rubino et al. (1999). The effect of irrigation practice on both BWUE and YWUE is not obvious, as it is shown in Table 13 for several crops submitted to different water regimes, including deficit irrigation (ID). Irrigation is considered among those strategies allowing to increase the water available for the crops: it may

129



increase growth and, consequently, WUE, provided that the water supplied by irrigation is transpired and not lost as evaporation from the soil, drainage and runoff. Tarantino et al. (1997) compared BWUE and YWUE among six crops and investigated the effect of four irrigation regimes (rainfed, restitution of 50 and 100% of the crop evapotranspiration, and deficit irrigation) on both BWUE and YWUE, showing the great variability among BWUE values of C3 species and a different effect of irrigation regimes on the species. Concerning BWUE, it emerged that: (i) among all the treatments, the highest values have been obtained on average by sweet sorghum (a C4) and durum wheat (a C3); (ii) among the rainfed treatments of all the crops, the highest value was reached by durum wheat; (iii) among the most watered treatments of all the crops, the highest value was reached by sweet sorghum. Comparing the effect of water supply on YWUE, the best results have been obtained by the restoration of minimum 50% of the crop evapotranspiration in sweet sorghum, kenaf and tomato, while no significant variations have been noticed with increasing irrigation regimes in sunflower and cotton. Nevertheless, for both crops excellent results have been reached in the treatment irrigated by deficit irrigation method. Also durum wheat reached high YWUE values by applying deficit irrigation method, further than without any irrigation. The results achieved in this study are very important to highlight the importance of deficit irrigation practice for some crops grown in environments with water restrictions. In deficit irrigation strategy, in fact, “water is applied to create a certain water deficit, which results in a small yield reduction that is less than the consequent reduction in transpiration, and therefore a gain in WUE per unit water transpired, and possible lower production costs if one or more irrigations can be eliminated” (Kijne et al., 2001).

Table 13. Above-ground Biomass water use efficiency, yield water use efficiency, total water used of field-grown crops under optimal conditions. Method to determine the water used, experimental location and reference are also reported. Yield Total Above-ground Determination WUE water used Location Reference Crop Biomass WUE of water used -3 -3 (kg m ) (mm) (kg m ) 1.7 450 Durum wheat 4.0 1.0 477 Soybean * 11.0 862 Spring sugarbeet 8.0 2.9 ** Policoro, 1.4 ** 920 Artichoke weighing Rubino et 14.0 7.8 Matera, 180 Rapa lysimeter al., 1999 4.8 4.2 Basilicata 360 Broccoli 2.0 7.4 536 Pepper 19.5 161 Lettuce 316 2.3 ** 27.4 ** Celery Lavello, weighing Rivelli et 765 Potenza, Kenaf 1.8 ** lysimeter al., 1998 Basilicata 891 Sunflower 2.6 canopy Valenzano, Steduto & 485 Grain sorghum 5.7 chambers Bari, Albrizio, Durum wheat 4.5 230 Puglia 2005 *** *** Chickpea 3.0 320 ** Sweet sorghum 4.8 532 pan Metaponto, 2.3 ** Losavio et Kenaf 631 evaporation & Matera, ** 2.6 al., 1999 Jerusalem Kc Basilicata 556 artichoke *

Roots are included.

**

Avg of more years.

***

130

Incomplete crop cycle.



Table 14. Effect of application of irrigation on above-ground biomass water use efficiency, yield water use efficiency, total water used of field-grown crops. Method to determine the water used, experimental location and reference are also reported. I0 indicates the control; I33, I50, I66, I100, indicate irrigation treatments with 33, 50, 100 percentage of ETc restoration; ID indicates treatment supplied by deficit irrigation method. Irrigation treatment

Crop

Tomato

Sweet sorghum

Sunflower

Cotton

Durum Wheat*

Kenaf

Tomato

Muskmelon

I100 I0 ID I100 I0

Pepper

Sunflower

I0 I33 I66 I100 I0 ID I50 I100 I0 ID I50 I100 I0 ID I50 I100 I0 ID I50 I100 I0 ID I50 I100 I0 ID I50 I100 I0 ID

**

No-flood Rice

Above-ground Biomass WUE -3 (kg m ) 1.0 0.9 0.8 0.8 1.3 2.5 3.0 3.4 1.2 1.8 1.4 1.6 1.8 1.8 1.6 1.7 3.8 3.2 2.6 2.2 1.2 1.3 1.5 1.5 1.6 1.5 1.3 1.3 2.1 1.7 1.1 1.0 0.9 0.6

Yield WUE -3 (kg m ) 0.9 0.8 0.8 0.8 2.9 7.3 8.2 7.2 0.5 0.6 0.6 0.7 0.9 0.7 0.7 0.5 1.2 1.2 1.0 0.8 2.2 4.0 4.9 5.0 8.1 8.2 14.4 13.4

Total water used (mm) 213 246 305 361 195 547 564 826 217 464 534 859 176 438 421 546 281 339 454 641 281 697 570 859 115 359 369 635

435

I33

2.4

522

I66

2.1

611

I100 I70

1.8 3.5

0.5

700 501

I100

3.4

0.5 5.9 8.9 9.1 8.1 7.4 8.4 7.8 9.2 10.6 11.6 9.2

Sunflower***

****

Cotton

Location

Reference

Gaudiano, water balance Potenza, Basilicata

Candido et al., 2000

pan evaporation & Kc weighing lysimeter

weight lysimeter pan evaporation & Kc

594 446 202 212 276 284 306

Gaudiano, Potenza, Basilicata

Tarantino et al., 1997

Matera, Basilicata

Rivelli et al., 2004

Villa d’Agri, Potenza, Basilicata

Rivelli & Perniola, 1997

weighing lysimeter pan evaporation & Kc

water balance

4.9

A B C D E F A B C D E

Determination of water used

Seasonal irrigation volume + rainfall

Metaponto, water balance Matera, Basilicata

Losavio et al., 2001

Pozzallo, water balance Ragusa, Sicilia

Cosentino et al., 1992


Pozzallo, Ragusa, Sicilia

Foti et al., 1992

avg of 2 years; avg of 3 years.

*

**

131



***

Letters indicate 6 irrigation treatments, the same amount of water was given at different stages of crop cycle, as follow: A: full-irrigated; B: one irrigation at the head visible stage; C: one irrigation at the beginning of flowering; D: two irrigations at stage of tenth leaf and at the beginning of flowering; E: three irrigations at stage of tenth leaf, at the head visible stage and at the beginning of flowering; F: four irrigations at stages of tenth leaf, at the head visible stage, beginning and end of flowering. **** Letters indicate five irrigation treatments, different numbers of watering were given at different stages of crop cycle.

Contrasting results with respect to Tarantino et al. (1997) have been shown by Candido et al. (2000) on YWUE of tomato crop: in fact, YWUE was highest in the control and lowest in the treatment with 100% evapotranspiration restoration. It is interesting to notice that in the rainfed treatment of experiment of Candido et al. (2000) the amount of water used is about 2-fold higher than in Tarantino et al. (1997), while opposite behaviour occurred in the well-irrigated treatment, although both the studies have been carried out in the same location. It may be due to the very different cultivars used, but the method utilized to determine the amount of water used plays a crucial role too. In a recent study of Rivelli et al. (2004) the water use efficiency response of two important vegetables (muskmelon and pepper), widely cultivated in the Southern Italy, have been compared, under three different water regimes. The findings have indicated that BWUE was much higher in muskmelon than in pepper in all the compared treatments, demonstrating a greater efficiency of the former crop in using water and its better adaptability to tolerate water deficit conditions. Table 15. Effect of application of fertilizers on above-ground biomass water use efficiency, yield water use efficiency, total water used of field-grown crops. Method to determine the water used, experimental location and reference are also reported. Above-ground Biomass WUE -3 (kg m )

Crop Grain sorghum Sugarbeet

Soybean

Wheat

Sunflower

Soybean*

Sunflower

Sunflower Grain sorghum *

Total Determination water used of water used (mm)

-

1.6

489

+

1.9

466

-

1.3

639

+

1.5

655

-

0.4

309

+

0.6

322

-

1.2

298

+

0.8

283

-

3.3

+

4.2

water balance

Location

Reference

Foggia, Puglia

Rizzo et al., 1990

Campiglia & water balance Viterbo, Lazio Caporali, 1992

-

1.3

0.4

399

+

1.5

0.5

408

-

2.0

0.6

538

+

2.1

0.6

549

-

2.2

837

+

2.6

891

-

4.4

485

+

5.7

510

Avg. of 2 years

132

Yield WUE -3 (kg m )

water balance

Foggia, Puglia

Rinaldi et al., 1996

water balance

Foggia, Puglia

Rinaldi & Rizzo, 1999

canopy chambers

Valenzano, Bari, Puglia

Steduto & Albrizio, 2005



Both the works carried out by Rivelli and Perniola (1997) and Cosentino et al. (1992) on sunflower, under different water supplies, show how much it is difficult to compare the results among experiments. The YWUE values found in these two experiments on sunflower greatly differentiated from those reported by Tarantino et al. (1997), demonstrating that, for the same crop, YWUE varies over a wide range. Similar consideration is valid for the findings of Foti et al. (1992) on cotton, as compared to those of Tarantino et al. (1997). The causes of such great variability may be ascribed to the application of different methods to determine the total “water used” and to the use of different denominators in the WUE ratio. Many times, indeed, as “water used” by the crop, which represents the denominator of WUE and/or WP ratios, is not considered the amount of water effectively lost by transpiration, but the total amount of water supplied by irrigation plus the rainfall. Of course, this amount is not all necessarily used by the crops for transpiration. Table 16. Effect of crop rotations and intercropping on above-ground biomass water use efficiency, yield water use efficiency, total water used of field-grown crops. Method to determine the water used, experimental location and reference are also reported. Crop

Above-ground Biomass WUE -3 (kg m )



Sorghum-Wheat

1.5

463

SorghumWheat+Soybean

1.9

466

Sugarbeet-Wheat

1.3

691

SugarbeetWheat+Soybean

1.5

655

Sunflower-Wheat

0.5

466

SunflowerWheat+Soybean SugarbeetWheat+Soybean

0.6

487

0.5

327

0.5

385

0.5

344

0.6

323

Wheat

0.5

266

Wheat+Soybean

1.2

286

Wheat+Sorghum

1.2

284

Sugarbeet-Wheat

0.7

242

SugarbeetWheat+Soybean

0.9

260

Sunflower-Wheat

0.8

258

SunflowerWheat+Soybean

0.9

267

Sorghum-Wheat

0.7

275

Sorghum-Wheat+Soybean

0.7

258

Wheat+Soybean SunflowerWheat+Soybean SorghumWheat+Soybean

Soybean* as main crop

1.0

0.4

861

Soybean* as catch crop after barley

1.3

0.7

420

Sunflower-Wheat**

2.0

0.5

246

SunflowerWheat+Soybean** * Avg. of 3 years

2.1

0.6

239

*

water balance

drainage lysimeters

water balance

Location

Reference

Foggia, Puglia

Rizzo et al., 1990

Modena, Emilia Romagna

Costantini & Melotti, 1991

Foggia, Puglia

Rinaldi & Rizzo, 1999

Avg. of 12 years

133



Application of fertilizers may not only result in increased growth but also in increased WUE, as it is shown in Table 15. Fertilizers use may increase slightly the total amount of water used (e.g. Rinaldi et al., 1996; Rinaldi and Rizzo, 1999), but the main effect is to increase early canopy growth so that it shades the surface and therefore reduces the evaporation as a proportion of the total water that is lost (Steduto and Albrizio, 2005). However, the positive effect of fertilizer in increasing both growth and water used, and reducing the evaporation is not universal. In fact, in the study of Rizzo et al. (1990) on wheat and grain sorghum an opposite behaviour in water use was observed between treatments with either low or high application of fertilizers, despite large positive effect of fertilizers on biomass production. A proper choice of the crop rotation is of fundamental importance for an appropriate use of water, and it affects the length of crop cycle (to be chosen), the efficiency for water uptake, the amount and the quality of crop residuals, the number and type of soil tillage practices. All these factors influence some important physical properties of the soil, such as the porosity, the water retention, the infiltration rate and the evaporation from the bare soil. Consequently, also the WUE and WP result to be strongly affected by the crop rotations, as it is shown in Table 16. Rizzo et al. (1990) compared YWUE among rotations of wheat cultivated in monoculture, with or without catch crop of soybean or sorghum, and three two-years rotations (sugarbeet-wheat; sunflower-wheat; sorghum-wheat, with or without catch crop of soybean). For wheat the best YWUE was reached in the monoculture with the catch crop of soybean or sorghum. For both sorghum and sugarbeet as main crops, the best results were obtained with soybean as catch crop, while the YWUE of both sunflower and soybean did not significantly differentiate among rotations. Also in the experiment of Rinaldi and Rizzo (1999) all the investigated parameters (BWUE, YWUE and the water used) for sunflower did not significantly varied in the rotation sunflower-wheat as compared to the same rotation, but with soybean as catch crop. Costantini and Melotti (1991) compared both BWUE and YWUE and the water requirement of soybean cultivated for three years as main crop (spring sowing) and as catch crop (summer sowing). From this study, it is emerged that the amount of water used by soybean as main crop was nearly double in comparison with that used by soybean as catch crop, as consequence of a longer crop cycle and the highest temperatures during the summer months. Soybean as main crop produced more biomass and yield dry matter, as compared to the catch crop, but it showed a lower BWUE and YWUE. The effect of mulching and early sowing on BWUE, YWUE and the amount of water used is shown in Table 17. Mulching practice is a common way to reduce evaporation from the soil surface, further than decrease the soil temperature. In terms of water conservation, the main effect of mulches is to reduce the rate of evaporation when the soil surface is damp and then to extend the duration of this stage (Gregory, 2004). In a recent study of Cantore et al. (2005) the use of plastic mulches positively affected both biomass and yield WUE of muskmelon; this effect was mainly due to the reduction of about 40% of the evapotranspiration, as both the evaporation from the soil and the length of the crop cycle were strongly reduced in the mulching treatment. Early sowing of crops is a very important mean of maximizing crop yield and WUE. In fact, increasing the early growth of the canopy when the soil surface is usually damp and the vapour pressure deficit is low has proved effective in increasing WUE. Bonari et al. (1989) found that an early sowing of ten days increased the yield of 54, 35 and 17% for maize, soybean and sunflower, respectively. Hence also biomass and yield water use efficiencies increased significantly in all the crops except of sunflower, although the water use in early sowing was higher than in the normal sowing. Differently, Rivelli and Perniola (1997) dealing with sunflower found that the increase in yield water use efficiency was strictly linked to the decrease in the amount of water used, as effect of a reduced evaporation from the soil.

134



Table 17. Effect of both mulching and early sowing on above-ground biomass water use efficiency, yield water use efficiency, total water used of field-grown crops. Method to determine the water used, experimental location and reference are also reported. Above-ground Biomass WUE -3 (kg m )


Muskmelon

1.7

8.7

320

Mulching Muskmelon

2.8

13.2

229

0.7

487

1.0

385

4.0

1.9

457

4.5

2.3

582

2.0

1.0

457

2.3

1.2

547

1.8

0.6

452

1.7

0.6

537

Crop

Sunflower

Maize

Soybean

Sunflower

Normal sowing Early sowing Normal sowing Early sowing Normal sowing Early sowing Normal sowing Early sowing


Loca tion

Reference

weighing lysimeter

Policoro, Matera, Basilicata

Cantore et al., 2005


Matera, Basilicata

Rivelli & Perniola, 1997

Drainage lysimeter with Pisa, variable water Toscana table

Bonari et al., 1989

CONCLUSIONS During the last 20-30 years, irrigated agriculture has been expanded over the whole Italian territory assuring a more stable agricultural production. In the same period, an important development of various irrigation techniques and agronomic practices have been occurred and followed by numerous research activities especially in two relevant agricultural regions: Puglia region in the South, and Emilia Romagna in the North – in the delta of river Po. The research activities on water saving practices in irrigation have been conducted mainly in Southern Italy where the crop productivity is strongly influenced with limited precipitation, and irrigation represents a fundamental practice in order to increase and stabilize agricultural production over the years. Accordingly, a particular attention was given to the research related to crop water requirements (estimation of reference evapotranspiration and crop coefficients), crop production functions and application of irrigation methods and practices that improve water use efficiency. A review of published data on crop water requirements revealed that the lowest irrigation volumes were recorded for short crop-cycle crops (e.g. shell bean) while the highest volumes were observed for the long-term crops especially if their growing cycle coincides with the summer season (e.g. spring sugar beet). The presented data on the crop coefficients pointed out a large divergence between the data measured under Italian environmental conditions and those published in the FAO Technical documents. This is specially true in the cases of application of specific agronomic practices (e.g. mulching) when the length of growing season and corresponding Kc values substantially differ from those published in the literature. Therefore, further research is needed to revise the existing data and to match better the modern agricultural practices, new varieties and recently adopted standard method for reference evapotranspiration estimate (FAO Penman-Monteith approach). This document reports the most important data related to irrigated agriculture in Italy and biomass and yield water use efficiency values found in many experiments carried out mainly in Southern Italy. Inasmuch as a large amount of data on WUE is available, there is a difficulty to compare them. In fact, it has clearly emerged how for each particular crop both BWUE and YWUE vary over a wide range. Possible reasons of it are: (i) the application of different methods to estimate the “water used”; (ii) the use of different nominators/denominators in WUE and WP ratios. In such perspective, more efforts should be done by the scientific community to make data comparable, using standardized procedures

135



and units of measurements. Certainly, a more clear conceptualisation of WUE and WP terms is necessary at national and regional scale. Agronomic practices to improve WUE rely on the improvement of water use efficiency (WUE) defined in terms of the yield or the biomass per unit area divided by the amount of water used (or transpired) to produce that yield or biomass. Hence, WUE can be enhanced from either crop improvement that increases yield per unit of water transpired (increased transpiration efficiency), or from crop management practices that minimize transpiration relative to other losses, or both. While transpiration efficiency is strictly linked to crop species and it varies among cultivars, there is a wide range of management practices that can reduce the loss of water by evaporation from the soil surface (such us mulching, application of fertilizers, early sowing and the choice of cultivars with rapid early growth) and/or increase the amount of water available to a crop (such as irrigation, fallowing and suitable rotations, weeds control, and the choice of cultivars having deep roots). The success of these practices at specific locations depends on soil properties, crop characteristics and climatic factors.

REFERENCES Allen R.G., Pereira L.S., Raes D. and Smith M., 1998. Crop evapotranspiration – Guidelines for computing crop water requirements. Irrigation and Drainage Paper 24, Food and Agricultural Organization of United Nations, Rome, 300 p. Angus J.F. and van Herwaarden A.F., 2001. Increasing water use efficiency in dry land wheat. Agron. J., 93: 290-298. Campiglia E. and Caporali F., 1992. Effetto della disposizione spaziale degli individui, delle modalità di semina e della concimazione azotata sulla consociazione girasole (Helianthus annuus L.) – cece (Cicer arietinum L.). Nota II. Complementarietà per l’uso della risorsa acqua. Riv. di Agron., 26, 4: 508-516. Candido V., Miccolis V., Perniola M., Rivelli A.R., 2000. Water use, water use efficiency and yield response of “long time storage” tomato (Lycopersicon esculentum MILL.). Proceedings 3rd ISHS on: “Irrigation Hort. Crops”, Ferreira and Jones (Eds), Acta Horticolturae, 537: 789-797. Cantore V., Boari F., Albrizio R., De Palma E., 2005. Influenza della pacciamatura sui consumi idrici e sull’efficienza d’uso dell’acqua del melone. Convegno SIA, su: “Ricerca ed innovazione per le produzioni vegetali e la gestione delle risorse agro-ambientali”, Foggia, 20-22 Settembre. Cosentino S., Sortino O., Litrico P.G., 1992. Risposta produttiva, temperatura radiativi della copertura vegetale e stato idrico della pianta nel girasole (Helianthus annuus L.) in secondo raccolto con differenti regimi irrigui. Riv. di Agron., 26, 4: 633-640. Costantini E.A.C. and Melotti M., 1991. Consumi idrici e risposte quanti-qualitative all’irrigazione della soia in coltura principale e intercalare nella bassa pianura emiliana. Riv. Irr. e Dren., 38, 1: 23-32. Doorenbos, J. and Pruitt. W.O., 1977. Guidelines for predicting crop water requirements, Irrigation and Drainage Paper 24, Food and Agricultural Organization of United Nations, Rome, 179 p. Foti S., Copani V., Guarnaccia P., 1992. Interventi irrigui in momenti significativi del ciclo biologico del cotone (Gossypium hirsutum L.) per una più efficace valorizzazione dell’acqua. Riv. di Agron., 26, 4, 663-670. Giardini, L., 2002. Agronomia generale ambientale e aziendale. Patron Editore, Bologna, 742 p. Giardini, L. and M. Borin, 1985. “Proposta metodologica per l’esame delle curve di risposta produttiva all’irrigazione”. Riv. di Agron., XIX, 4, 239-250. Gregory P.J., 2004. Agronomic approaches to increasing water use efficiency. In: Bacon M.A. (Eds.), Water use efficiency in plant biology. Blackwell Publishing Ltd, CRC Press, 327 p. Hatfield J.L., Sauer T.J., Prueger J.H., 2001. Managing soils to achieve greater water use efficiency: A review. Agron. J., 93: 271-280. INEA – Istituto Nazionale di Economia Agraria, 2003. Italian Agriculture in Figures. Italian Ministry of Agricultural Policies and Forestry, National Institute for Agricultural Economy, Rome, 172 pp.

136



ISTAT, 2002. Census on Agriculture, ISTAT, Rome IRSA-CNR, 1999. Un futuro per l'acqua in Italia, Eds. M. Benedini, A. Di Pinto, A. Massarutto, R. Pagnotta, R. Passino, Quaderni IRSA-CNR, Roma. Kijne J.W., Barker R., Molden D., 2003. Water productivity in Agriculture: Limits and Opportunities for Improvement. CABI Publishing, UK, 332 pp. Kijne J.W.,Tuong T.P., Bennett J., Bouman B., Oweis T., 2001. Ensuring food security via improvement in crop water productivity. Available on: http://www.iwmi.cgiar.org/challengeprogram/pdf/paper1.pdf Losavio N., Ventrella D., Vonella A.V., 1999. Consumi idrici, efficienza dell’uso dell’acqua e della conversione dell’energia in biomassa: parametri per valutare l’introduzione di nuove colture nell’ambiente mediterraneo. Riv. Irr. e Dren., 46, 2: 34-38. Lovelli S., Pizza S., Caponio T., Rivelli A.R., Perniola M., 2005. Lysimetric determination of muskmelon crop coefficients cultivated under plastic mulches. Agric.Water Manag., 72:147-159. Mannini P., 2004. Le buone pratiche agricole per risparmiare acqua. I supplementi di Agricoltura 18. Regione Emilia-Romagna, Assessorato Agricoltura, Ambiente e Sviluppo Sostenibile, p.178. Monteith J.L., 1984. Consistency and convenience in the choice of units for agricultural science. Exp. Agric., 20: 105-117. MPAF (Ministero delle Politiche Agricole e Forestali), Italian Ministry of Agricultural and Forestry Policies, 2004. Irrigazione sostenibile la buona pratica irrigua (P. Scandella and G. Mecella, eds.). Editorial project Panda, L’Informatore agrario, 300pp. Passioura J., 2004. Increasing crop productivity when water is scarce. From breeding to field management. Proceedings 4th International Crop Science Congress on: “New directions for a diverse planet”. Brisbane, Australia, 26 September – 1 October. Published on CDROM. Web site: www.regional.org.au/au/cs Perniola M., 1994. Ecophysilogical parameters and water relations of sweet sorghum, cotton and sunflower during drought cycles. Proceedings of the International Conference on: “Land and water resources management in the Mediterranean region”. Mediterranean Agronomic Institute, Valenzano, Bari, Italy. 4-8 September. Rinaldi M. and Rizzo V., 1999. La coltura del girasole inserita in due avvicendamenti e sottoposta a due livelli di input agrotecnico: produzione e uso dell’acqua. Riv. di Agron., 33: 265-273. Rinaldi M., Ventrella D., Fornaro F., 1996. Analisi di crescita, bilancio idrico e produzione di soia (Glicine max (L.) Merr.) in secondo raccolto dopo frumento duro (Triticum durum Desf.) sottoposta a due livelli di input agrotecnico. Riv. di Agron., 30, 2: 160-167. Rivelli A.R. and Perniola M., 1997. Effetti del regime irriguo e dell’epoca di semina su alcune cultivar di girasole (Helianthus annuus L.) in tre ambienti della Basilicata. Riv. di Irr. e Dren., 44, 1: 17-25. Rivelli A.R., Albrizio R., Lovelli S., Perniola M., 2004. Water use efficiency response of field-grown muskmelon and pepper to environmental water status. Proceedings 4th International Crop Science Congress on: “New directions for a diverse planet”. Brisbane, Australia, 26 Settembre – 1 Ottobre 2004. Published on CDROM. Web site: www.regional.org.au/au/cs Rivelli A.R., Perniola M., Tarantino E., Disciglio G., 1998. Consumi idrici e irrigui del kenaf (Hibiscus cannabinus L.) in un ambiente meridionale. Riv. Irr. e Dren., 45, 3: 29-36. Rizzo V., Castrignanò A., Stelluti M., Ventrella D., Carlone G., 1990. Prime valutazioni sui bilanci idrici di colture in rotazione. Ann. Ist. Sper. Agron., Bari, XXI, suppl. 2, 95-106. Rosegrant M.W., Cai X., Cline S.A., 2002. Global water outlook to 2025: averting an impending crisis. Food Policy Report 2020 Vision. Washington, D.C., IFPRI, 26 pp. Rubino P., Cantore V., Mastro M.A., 1999. Studio dell’efficienza dell’uso dell’acqua di alcune specie erbacee in un ambiente dell’Italia meridionale. Riv. Irr. e Dren., 46, 2: 39-46. Stanhill G., 1986. Water Use Efficiency. Adv. in Agron., 39: 53-85.

137



Steduto P. and Albrizio R., 2005. Resource use efficiency of field-grown sunflower, sorghum, wheat and chickpea. II. Water Use Efficiency and comparison with Radiation Use Efficiency. Agric. and For. Meteor., 130: 269-281. Steduto P., 1996. Water Use Efficiency. In: Pereira, L.S., Feddes, R.A., Gilley, J.R., Lesaffre, B. (Eds.), Sustainability of irrigated agriculture, NATO ASI Series E: Applied Sciences. Kluwer Academic Publ., Dordrecht, pp. 193-209. Tarantino E., Rivelli A.R., Perniola M., Nardiello I., 1997. Efficienza nell’uso dell’acqua di alcune colture erbacee sottoposte a differenti regimi irrigui: valutazione a livello di pieno campo. Riv. di Irr. e Dren., 44, 1: 8-16. Venezian Scarascia, M.E., Caliandro A., Giardini L., Quaglietta Chiaranda F., Rubino P., Giovanardi R., Losavio N.and d’Andria R., 1987. Yield response to different amounts of irrigation water for its best utilization. Thirteenth International Congress on Irrigation and Drainage, Rabat, Morocco, September 1987. Transaction actes, Volume I-D: 189-224. Viets F.G., Jr., 1962. Fertilizers and the efficient use of water. Adv. Agron., 14: 223-264. World Resources Institute, 2000. World Resources 2000-2001. People and ecosystems: The fraying web of life. United Nations Development Programme, United Nations Environment Programme, World Bank, Washington DC, 400pp. Yared Tesfagaber L., 2005. Studio dell’efficienza d’uso dell’acqua di colture erbacee in Italia meridionale. Tesi di Laurea in Agronomia Generale, Facoltà di Agraria, Università degli studi di Bari.

138



EFFECTS OF DEFICIT IRRIGATION ON YIELD AND WATER USE EFFICIENCY OF SOME CROPS UNDER SEMI-ARID CONDITIONS OF THE BEKAA VALLEY OF LEBANON

F. Karam* , K. Karaa**, N. Tarabey*** * Lebanese Agricultural Research Institute, Dept. of Irrigation and Agro-Meteorology, P.O. Box 287, Zahlé, Lebanon ** Litani River Authority, Dept. of Rural Development, P.O. Box 3732, Bechara El Khoury, Beirut, Lebanon *** Lake Share Communities Union, Association of Irrigation Water Users in South Bekaa Valley, Lala, Lebanon SUMMARY - A six-year experiment (1998-2003) was conducted at Tal Amara Research Station in the Bekaa Valley of Lebanon to determine water use, yield and water use efficiency in four annual crops with contrasting response to deficit irrigation (DI); maize (1998-1999) a determinate species with a limited capacity to adjust grain yield in response to water availability; soybean (2000-2001), an indeterminate species with a high capacity to compensate the effects of early water stresses; cotton (2001-2002), an indeterminate species with a larger capacity to adjust the number of dehiscent bolls under stressful conditions, and sunflower (2002-2003), a determinate species with a single inflorescence and an aptitude to tolerate moderate water stresses. Crop evapotranspiration (ET) was measured using drainage and weighing lysimeters. In the plots, ET was measured using a simple soil water balance model. Yield and its components were determined in sampling areas reserved for harvest. Water use efficiency at grain (WUEg) or seed (WUEs) basis was calculated as the ratio of dry yield to crop evapotranspiration (Y/ET), while water use efficiency at biomass-basis (WUEb) was calculated as the ratio of dry biomass to ET (B/ET). For cotton, water use efficiency (WUEl) was calculated as the ratio of dry lint yield to ET. Furthermore, the relationships between yield (Y) and biomass (B) in one hand, and crop evapotranspiration (ET) in the other hand were examined using linear models. Results of the experiments showed that corn seasonal ET reached on the lysimeter 952 mm in 1998 and 920 mm in 1999. Furthermore, grain-related water use efficiency (WUEg) varied with corn treatments from 1.34 kg m-3 to 1.88 kg m-3, while at biomass-basis (WUEb) the values varied from 2.34 kg m-3 to 3.23 kg m-3. For soybean, seasonal ET totaled 800 mm in 2000 and 725 mm in 2001. Seed-related water use efficiency of soybean (WUEs) varied from 0.47 kg m-3 to 0.54 kg m-3, while WUEb varied from 1.06 to 1.16 kg m-3. For Cotton, seasonal ET was 641.5 mm in 2001 and 669.0 mm in 2002. Average WUEl values varied among treatments from 0.43 kg m-3 to 0.64 kg m-3, while WUEb varied from 1.82 to 2.16 kg m-3. For sunflower, average across years of evapotranspiration attained 827 mm. WUEs of sunflower varied among treatments from 0.64 kg m-3 to 0.86 kg m-3, while at biomass-basis WUEb varied from 3.23 kg m-3 to 4.8 kg m-3. The results also showed that yield and biomass have positive, though weak, relationships with ET in corn and soybean, while in cotton and sunflower the relationships are negatives. Key words: Drainage lysimeter, weighing lysimeter, crop evapotranspiration, yield, biomass, water use efficiency.

INTRODUCTION Water is fast becoming an economically scare resource in many areas of the world. The need for more efficient agricultural use of irrigation water arises out of increased competition for water resources and increasing environmental concern (Doorenbos and Kassam 1988).

139



The best use of water must be made for efficient crop production and higher yields. Therefore, agriculture under unfavorable climatic conditions and limited water resources can not be profitably practiced unless on-farm water management techniques are designed to meet the present growing demands of water for increased food production (Oad et al., 2001). It is necessary therefore to develop new irrigation scheduling approaches, not necessarily based on full crop water requirement, but ones designed to ensure the optimal use of allocated water. Deficit irrigation (or regulated deficit irrigation, RDI) is one way of maximizing water use efficiency (WUE) for higher yields per unit of irrigation water applied (English et al., 1990; English and Raja, 1996; Kirda et al., 1999). The crop is exposed to a certain level of water stress either during a particular growth period or throughout the whole growing season, without significant reductions in yields. The main objective of deficit irrigation is to increase the WUE of a crop by eliminating irrigations that have little impact on yield. The resulting yield reduction may be small compared to the benefits gained through diverting the saved water to irrigate other crops for which water would normally be insufficient under traditional irrigation practices. In some irrigation schemes, the system is designed to deliver full irrigation in order to meet full crop water demands (Walker and Skogerbe, 1987). In others, the system is designed upon minimum allowable soil water depletion (James, 1988; Keller and Bleisner, 1990). Cuenca (1989) introduced for the first time the concept of partial irrigation within an irrigation scheme, and suggested that under some circumstances the designer might allow for greater soil water depletion, which could result in reduced yields as an economic tradeoff against the higher costs of intensive irrigation water. Deficit irrigation is a common practice in many areas of the world (English and Raja, 1996). A number of researchers have analyzed the economics of deficit irrigation in specific circumstances and have concluded that this technique can increase net farm income (Martin et al., 1989; English, 1990). The potential benefits of deficit irrigation derive from three factors; increased irrigation efficiency, reduced costs of irrigation and the opportunity costs of water. Four levels of applied water could be defined as optimal in one sense or another (English et al., 1990): •

The level of applied water at which crop yields per unit of land are maximized;

•

The level at which yields per unit of water are maximized;

•

The level at which net income per unit of land is maximized;

•

The level at which net income per unit of water is maximized.

The optimum level of applied water for a particular situation will be that which produces maximum profit or crop yield, per unit of land or per unit of water, depending on whether the goal is to maximize profits or food production and whether the most limiting resources is water or land. The two other levels of applied water are the deficit levels at which net returns will be equal to those which would be realized by full irrigation (English and Raja, 1996). Irrigation management of crops involves a balance between vegetative and reproductive growth. Excessive vegetative growth can delay maturity and reduce final yield. One of the irrigation strategies that could be implemented to reduce excessive vegetative growth maintain yield and reduce water use, leading to an improvement in water use efficiency, is regulated deficit irrigation (RDI), which may be implemented during part of the growing season by regulating moisture within a desired deficit range. RDI aims to optimize water use efficiency and therefore maximize the yield returned per unit of water applied. Any minor yield loss which may result from the implementation of a mild moisture deficit/stress under RDI is offset by the benefits of reduced water use leading to a reduction in excessive vegetative growth (Kirnak et al., 2002). A variety of crops have been found to benefit from a RDI strategy including maize, wheat, sunflower, potatoes, tomatoes and cotton. Irrigation using drip is typically able to apply smaller quantities of water more frequently, and is better able to maintain soil moisture at the mild deficit required to implement RDI. The most desirable benefits associated with implementing RDI strategy in crops are: The reduction in excessive vegetative growth Maintenance of soil moisture in the most agronomical desirable range An increase in water use efficiency, and The ability to better capture and use in-season rainfall events after an irrigation event due the maintained deficit.

140



Where water scarcity exists at regional level, irrigation managers should adopt the same approach to sustain regional crop production, and thereby maximize income (Stegman et al., 1990). Burt et al., (1997) defined irrigation efficiency (IE) as the proportion of irrigation water input to the farm for crop production that was used by the crop as evapotranspiration (ET) over the growing season. Tennakoon and Milroy (2003) defined crop water use efficiency (CWUE) as crop production per unit of ET. CWUE quantifies the efficiency with which economic yield is produced as a function of water used by the crop in the field. Furthermore, Tennakoon and Milroy (2003) defined farm water use efficiency (FWUE) as the amount of yield produced per millimeter of total seasonal water input at the farm level. Oad et al., (2001) calculated irrigation water use efficiency (IWUE) from the marketable fruit yields and amount of water applied to the plants. Moreover, they calculated total water use efficiency (TWUE) as the ratio of marketable fruit yields to water use. The objectives of this study were to determine water use and yield in four annual crops with contrasting response to deficit irrigation (DI); corn, soybean, cotton, and sunflower, and to examine the existing relationships between yield and biomass, in one hand, and evapotranspiration in the other hand.

MATERIAL AND METHODS Field studies aiming at examining the response of maize (Zea mays L.), Soybean (Glycine max L. Merril), cotton (Gossypium hirsutum L.) and sunflower (Helianthus annus L.) to deficit irrigation stress were conducted during the period 1998-2003 at Tal Amara Research Station in the Central Bekaa Valley of Lebanon (33° 51' 44'' N lat., 35° 59' 32'' E long., altitude 905 m a.s.l). Tal Amara has a welldefined hot, dry season from May to September and very cold for the remainder of the year. Long-run data indicate an average seasonal rain of 592 mm, with 95% of the rain occurring between November and March. Crops were grown on deep and fairly-drained soil, characterized by high clay content (44%). Measured field capacity (-0.33 bar) and permanent wilting point (-15 bars) averaged 29.5% and 16.0% by weight. Extractable plant water is estimated at 190 mm for 1 m rooting depth and a bulk -3 density of 1.41 g cm . Hybrid corn (cv. Manuel) was sown on 19 May in 1998 and 25 May in 1999 at 10 plants m-2. Soybean hybrid (cv. Asgrow 3803) was sown on 10 May 2000 and 25 April 2001 at a density of 12 plants m-2. Cotton (cv. AgriPro AP 7114) was sown on 5 May in 2001 and on 13 May in 2002 at a density of 10 plants m-2. Sunflower (cv. Melody) was sown on 2 June 2003 and 10 June 2004 at a density of 8 plants m-2. For corn, crop evapotranspiration (ET) in both years was measured using a set of two drainage lysimeters of 4 m2 surface area (2 m × 2 m) by subtracting the volume of drainage from the irrigation amount. The lysimeters, 1.2 m deep, 24 m apart, aligned N-S, are situated in the middle of 1-ha field (200 m N-S by 50 m W-E) (Karam et al., 2003). For soybean, ET was measured by a weighing lysimeter of 16 m² surface area (4 m × 4 m) and 1.2 m deep, containing the same clay soil as in the drainage lysimeters. Watering of the lysimeter was made upon a 30% soil depletion of the available water in the 0-100 cm soil layer. The weight loss of the lysimeter due to soil evaporation and plant transpiration was measured with load cells and recorded at a 15-minute interval on a computer located near the lysimeter. Water was supplied to the lysimeter when the weight loss reached a threshold value. Data were transferred via telephone modem to the irrigation laboratory, 500 m from the lysimeter. ET was determined as the difference between lysimeter weight gains (irrigation and/or rain or dew) and the lysimeter weight loss (from soil 2 evaporation and plant transplantation) divided by the lysimeter surface area (16 m ), so that day/night ET from midnight to midnight was computed as the average of 96 readings (one reading for each 15minute time scale). The lysimeter has an ET accuracy and resolution of 1 kg, which corresponds to 0.062 mm of water for a surface area of the soil in the lysimeter. Water percolating through the soil mass of the lysimeter was collected and measured in a drainage reservoir located at the bottom of the lysimeter, so that drainage was accounted for in the water balance calculation of the lysimeter (Karam et al., 2005).

141



For cotton and sunflower, crop evapotranspiration (ET) was estimated using the FAO method (Doorendos and Pruit, 1977) by multiplying reference evapotranspiration (ETrye-grass) as measured in rye-grass lysimeters by the corresponding crop coefficients (Kc), which were derived for the different growth stages (Doorenbos and Kassam, 1988): ET = ETrye-grass × Kc Reference evapotranspiration (ETrye grass) was measured in a set of two rye-grass drainage lysimeters of 4 m² surface area (2 m × 2 m) and 1m depth. The lysimeters are 24 m distant, aligned W-E, and located inside the weather station (40 m × 40 m), 50 m apart of the experimental plots. Water was distributed in the field uniformly and simultaneously at 100% of field capacity using line source drippers, 16 mm in diameter, 40 m long, aligned W-E and spaced 70 cm apart. The dripper spacing was 40 cm, each delivering 4 l hr-1 of irrigation capacity at 100 kPa pressure. For corn, deficit irrigation was applied continuously during the growing cycle upon the measured crop evapotranspiration (ET). Water was then applied at 100% (I-100 treatment) and 60% (I-60 treatment) of ET. For soybean, cotton and sunflower, deficit irrigations were made by cutting-out irrigation or for a two-week period during one or more of the different growth stages. For soybean, deficit irrigations were applied at full bloom (R2 stage), at seed enlargement (R5 stage) and at mature seeds (R7 stage). For cotton, deficit irrigations were applied at first open boll, at early boll loading, and at mid-boll loading. For sunflower, irrigation was withheld for a two-week period prior to flowering (E2 stage), at mid flowering (F1 stage), and at the beginning of seed formation (M0 stage) and at seed ripening (M2 stage). For all crops, a control was fully-irrigated throughout the growing period. Table 1 illustrates the deficit irrigation treatments for the crops under study. At sowing, crops were irrigated to keep water content at 100% of soil available water. Weeds and insects were adequately controlled. Each species was grown in a different section in a 2-ha contiguous experimental field. Irrigation treatments were laid out within each crop in a block design with three or four replications. In the plots, evapotranspiration was calculated using a simple soil water balance model (Doorenbos and Kassam 1988): ET = I + P – Dr – Rf ± Δs Where ET is evapotranspiration, I is irrigation application, P is effective rainfall, Dr is drainage water, Rf is amount of runoff, and Δs is change in the soil moisture content determined by gravimetric sampling. All terms in this equation are expressed in mm. Moisture content in the 0-90 cm soil profile was measured gravimetrically before irrigations. Since there was no observed runoff during the experiment and the water table was at 4 m depth, capillary flow to the root-zone and runoff flow were assumed to be negligible in the calculation of ET. Drainage below 90 cm, after a number of soil-water content measurements, was considered as negligible. So the above equation was reduced to: ET = I + P ± Δs Soil moisture in the plots was also measured using a Sentry 200-AP TDR (Time Domain Reflectometry). The TDR was calibrated to the soil at Tal Amara over a wide range of soil moistures (Sentry, 200-AP, 1994). Four access PVC tubes, 50 mm in diameter and 1.0 m in length, were inserted in the middle rows of each plot. Readings were taken one day before irrigation and 2-to-3 days after irrigation during the growing seasons at 0-15, 15-30, 30-45, 45-60, 60-75 and 75-90 cm of the soil profile. Readings were then converted to soil moisture content values using a locallycalibrated equation. Gravimetric measurements and TDR readings were used to estimate seasonal ET in the plots using a water balance model as indicated above. At physiological maturity, all individual plants in the sampling areas were harvested to determine above ground biomass production (B) and yield (Y). For corn, grain number per m2 and the 1000grain weight were determined. For soybean and sunflower, seed number per m2 and the 1000-seed weight were also determined. For cotton, yield was determined by weighting lint at dry basis in the sampling areas.

142



Our study is largely based on the linear relationships between Y and B in one hand, and ET in the other hand. For this reason, we fitted linear models to the data: Y = a1 (ET) + b1 B = a2 (ET) + b2 These models are the simplest and more often used models to describe the relationships between yield, biomass and evapotranspiration. These relationships are appropriate frameworks to investigate the pattern of water use efficiency, i.e. WUE = YET-1, or WUE = BET-1. This concept is widely used in agronomic research. Departure from linearity can be tested through regression of log Y or log B on log ET (Thompson et al., 1991). However, as this test can produce misleading results when the yintercept differs from zero, polynomial regressions were preferred, as in Thompson et al. (1991). In corn, soybean and sunflower, water use efficiency at grain or seed-basis (WUEg,s) was calculated as the ratio of yield at dry basis to the amount of crop evapotranspiration (Y/ET), while water use efficiency at biomass-basis (WUEb) was calculated as the ratio of biomass at dry basis to ET (B/ET) (Foroud et al., 1993; Howell et al., 1998). In cotton, water use efficiency at lint-basis (WUEl) was calculated as dry lint yield to the amount of water evapotranspired from the crop -3 -2 -1 (Tennakoon and Milroy, 2003). WUE was expressed in kg m-3 (1 kg m = 1 g m mm ).

143



Table 1: Deficit-irrigation treatments of the various crops under study. Crop Period Treatment Period of irrigation cutout Corn 1998 I-100 No irrigation restriction during the growing period I-60 Deficit irrigation at 6-leaf stage (from d.o.y 164 to d.o.y 255) 1999 I-100 No irrigation restriction during the growing period I-60 Deficit irrigation at 6-leaf stage (from d.o.y 170 to d.o.y 250) Soybean 2000 C No irrigation restriction during the growing period S-1 Irrigation cutout at full bloom (R2 stage) (from d.o.y 215 to d.o.y 227) S-2 Irrigation cutout at seed enlargement (R5 stage) (from d.o.y 236 to d.o.y 250) S-3 Irrigation cutout at mature seeds (R7 stage) (from d.o.y 250 to d.o.y 264) 2001 C No irrigation restriction during the growing period S-1 Irrigation cutout at full bloom (R2 stage) (from d.o.y 197 to d.o.y 211) S-2 Irrigation cutout at seed enlargement (R5 stage) (from d.o.y 218 to d.o.y 232) S-3 Irrigation cutout at mature seeds (R7 stage) (from d.o.y 232 to d.o.y 246) Cotton 2001 C No irrigation restriction during the growing period S-1 Irrigation cutout at first open boll (from d.o.y 217 to d.o.y 259) S-2 Irrigation cutout at early boll loading (from d.o.y 231 to d.o.y 259) S-3 Irrigation cutout at mid boll loading (from d.o.y 245 to d.o.y 259) 2002 C No irrigation restriction during the growing period S-1 Irrigation cutout at first open boll (from d.o.y 232 to d.o.y 274) S-2 Irrigation cutout at early boll loading (from d.o.y 246 to d.o.y 274) S-3 Irrigation cutout at mid boll loading (from d.o.y 260 to d.o.y 274) Sunflower 2003 C No irrigation restriction during the growing period S-1 Irrigation cutout prior to flowering stage (from d.o.y 232 to d.o.y 274) S-2 Irrigation cutout at mid flowering stage (from d.o.y 246 to d.o.y 274) S-3 Irrigation cutout at the beginning of seed formation (from d.o.y 260 to d.o.y 274) S-4 Irrigation cutout at mid seed ripening (from d.o.y 260 to d.o.y 274) C No irrigation restriction during the growing period 2004 S-1 Irrigation cutout prior to flowering stage (from d.o.y 232 to d.o.y 274) S-2 Irrigation cutout at mid flowering stage (from d.o.y 246 to d.o.y 274) S-3 Irrigation cutout at the beginning of seed formation (from d.o.y 260 to d.o.y 274) S-4 Irrigation cutout at mid seed ripening (from d.o.y 260 to d.o.y 274)

144



145



RESULTS AND DISCUSSION Evapotranspiration, yield and water use efficiency Table 2 shows the values of evapotranspiration (ET), yield (Y), biomass (B) and water use efficiency of the various crops under well and deficit irrigation conditions. Corn Seasonal ET reached on the drainage lysimeter amounts of 952 mm and 920 mm in 1998 and 1999, for total growing cycles of 128 days and 120 days, respectively. In the plots, crop evapotranspiration totaled in the well-irrigated treatment (I-100) 863 mm and 833 mm in 1998 and 1999, respectively, while in deficit-irrigated treatment (I-60) ET totaled 575 mm and 556 mm in 1998 and 1999, respectively (Karam et al., 2003). -2 Grain yield on a dry basis declined in 1998 from 1520 gm-2 on the lysimeter to 1450 gm on the -2 full irrigated treatment (I-100) to 1080 gm on the deficit-irrigation treatment (I-60). In 1999 these reductions ranged from 1340 gm-2 on the lysimeter to 1280 gm-2 and 1040 gm-2 on I-100 and I-60, respectively. Total aboveground biomass at harvest was also reduced by deficit irrigation. In 1998, a reduction of 130 gm-2 was observed in I-100 in comparison with the lysimeter, while the reduction on I-60 exceeded 800 gm-2 when compared to I-100. In 1999, these reductions were 100 gm-2 and 400 gm-2, respectively.

Grain-related water use efficiency (WUEg) of lysimeter grown corn was 1.52 kg m-3 in 1998 and 1.34 kg m-3 in 1999. However, fully-irrigated corn had a WUEg of 1.68 kg m-3 in 1998 and 1.54 kg m-3 in 1999. Higher WUEg values of 1.88 kg m-3 and 1.87 kg m-3 were obtained in 1998 and 1999, respectively, from the I-60 treatment. On a biomass basis, I-100 treatment had values of water use efficiency (WUEb) of 3.16 kg m-3 and 2.46 kg m-3 in 1998 and 1999, respectively, while the I-60 treatment had values of 3.23 kg m-3 and 2.97 kg m-3, respectively. On the lysimeter, these values were 3.0 kg m-3 and 2.34 kg m-3, respectively. Soybean Total evapotranspiration (ET) as measured by the drainage lysimeters in 2000 totaled 800 mm for a total growing period of 140 days. However, when ET was measured by the weighing lysimeter in 2001, it was 725 mm during a growing period of 138 days (Karam et al., 2005). Average seed yield was 3.2 t ha-1 in the control treatment, compared to 3.5 t ha-1 in the lysimeter, whereas total aboveground biomass productions were 7.3 t ha-1 and 8.1 t ha-1, respectively. Deficit irrigation during R2 stage (S1) reduced biomass production by 16% (P50% are reported; e.g. Corbeel et al. 1998), in particular in non-fertilised crops (Gregory et al. 2000). A better seedling emergence and a more early vegetative growth (i.e. early or seedling vigour) has been reported as an important trait in terms of WUE in cereals in Mediterranean conditions (Richards et al. 2001; Richards et al. 2002). Evaporation from the soil is negatively correlated with fractional area of shaded soil (e.g. Passioura 2004). For instance, compared with wheat, barley achieve higher leaf area at early stages of the crops (e.g. López-Castañeda et al. 1995; Rebetzke et al. 2004), decreasing the loss of water by soil evaporation. The presence of a canopy can decrease soil evaporation by three main mechanisms (Gregory et al. 2000) (a) reduction of net radiation absorbed by the soil (b) humidification of the air, increasing the aerodynamic resistance to the transfer of water vapour from the soil and (c) the uptake of water from the roots near the soil surface reduces the hydraulic conductance (Gregory et al. 2000). In wheat canopies, the evaporation from the soil is negatively correlated with the fractional shaded area (Passioura 2004) and it is well documented that is lower in barley than wheat (Siddique et al. 1990). In this study, cultivars with higher SLA and better early vigour reduce soil evaporation. Moreover, compared with old cultivars, modern cultivars of wheat had lower soil evaporation rates

197



early in the growing season despite that transpiration efficiency for dry matter production were similar for all cultivars. Moreover, the growth took place under low VPD, decreasing the total transpired water by the crop, and consequently, increasing the WUE (López-Castañeda et al. 1995). However, it has been reported that a closed canopy increases the water losses by interception and evaporation from the canopy (Leuning et al. 1994). In summary, the fast growth of leaf area has little benefit in regions where soil evaporation is a small component of total crop water use (Condon et al. 2004). Similar consideration may be pointed out in areas where evaporation is high, but it is limited by the movement of water in the soil (and non by canopy density; Gregory et al. 2000; Yunusa et al. 1993). According to estimations derived from simulation models, the reduction of soil evaporation by a higher SLA and early vigour seem to occur only in Mediterranean-type environments and when high nitrogen doses are applied (Asseng et al. 2003). Finally, it must be mentioned that an early growth may be achieved by applying some agronomic practices. Early sowing date (Oweis et al. 2000; Richards et al. 2002), non-tillage management (Klein et al. 2002) or small irrigation at the first stages of the crops (Tavakkoli and Oweis 2004), for instance, ensure early germination, seedling establishment and a fast growth. In addition, sedding pattern may be relevant. The use of narrow row spacing and adequate plant population would help conserve water and hence increase the WUE. Works carried out in semi-arid region of Morocco, for instance, revealed that WUE were increased when row spacing was reduced (Karrou 1998). Although the study of the early vigour has received considerable attention in Mediterranean climate, it could be relevant in other dry regions. In the Loess Plateau in China (a semi-arid region), the use of straw mulch (to reduce soil evaporation) has been proposed to improve WUE in bread wheat crops (Huang et al. 2005). In this context, thus, a fast growth at early stage may replace or complement the mulching practice.

Early vigour in ‘non-competitive’ modern cultivars In wheat, the widely use of semi-dwarf cultivars (GA -giberellic acid- insensitive) which it has increased the harvest index of modern cultivars with lower plant height (Austin 1999), is associated with short coleoptiles (Richards et al. 2002). Tall cultivars have longer and wider leaves and produce higher biomass than semi-dwarf genotypes (v.g. Rht) at early stage (Rebetzke et al. 2004). These traits of semi-dwarf genotypes lead a poor seedling establishment and, consequently, higher soil evaporation at the beginning of the crop. At first, it could suggest that a high potential yield and high early vigour are mutually exclusive, i.e. that modern cultivars are ‘non-competitive ideotype’ (Blum 1996). However, a recent report showed the existence of dwarfing genes that promote short shoot but no small coleoptiles, opening the possibility to explore in this way (see references in Passioura 2004). Researches carried out by Richards’ group found that GA-sensitive lines have good partitioning characteristic (i.e. high harvest index) and, at the same time, long coleoptiles (Richards et al. 2002). Consequently, obtain wheat varieties with high yield potential and high WUE at early stage of the crops is feasible.

WATER USE EFFICIENCY, GRAIN YIELD AND ISOTOPIC DISCRIMINATION: BREEDING FOR HIGH O LOW WUE? The Passioura equation as conceptual model In water limited environments, grain yield could be modelled by Passioura equation, which it has been largely discussed in several works (e.g. Blum 2000; Araus et al. 2002): GY = W x WUEbiomass x HI where W is the water used by the crops (evapotranspiration) and HI is the harvest index. Accordingly with this model, grain yield could be increased by (a) the capacity of capture more water, (b) the efficiency for producing dry matter per unit of used water and (c) the ability to devote more assimilates to the grains (Araus et al. 2002). Although at first view this model is attractive, - and sensu Passioura

198



(2004) the three components of the equation are sufficiently independent to make it worthwhile considering them one by one- the terms are not independent and their interrelationship may be complex. A higher WUE may be related with a lower water use (and virtually, lower growth and grain yield) under drought conditions. In addition, although harvest index may be drought-independent in some cases, drought-dependent HI is often a function of post-anthesis W (Richards et al. 2002 and references cited therein; see references in Araus et al. 2002). In short, is the WUE a suitable parameter to use in breeding programs? The answer is complex and not universal, because it seems to depend on the target environment. As it was pointed out by Blum, WUE is therefore a misleading parameter when applied to plant breeding for water-limited environments where soil water extraction capacity is important (Blum 2001). Those traits that could confer a higher water extraction capacity, and then, a higher water use, such as osmotic adjustment (Blum et al. 1999), could have the opposite effect in WUE if higher stomatal conductances were involved. We discuss this point in the following section.

Carbon isotope discrimination and their relationship with WUE and grain yield 13 13 Discrimination of the stable isotope C (Δ C) has been widely accepted as an indicator of WUE (see review of this issue in Araus et al. 2001; Pate 2001). In short, Δ13C in C3 plants is determined by the following equation:

Δ13C = a + (b-a).(ci/ca) where a = 4.4 ‰ represents the isotope fractionation associated with differential diffusivities of 13C versus 12C, b = 27 ‰ is the fractionation by Rubisco carboxilation and ci and ca are the intercellular and ambient CO2 concentration respectively (Pate 2001). The ci/ca ratio is determined by the balance between stomatal conductance and photosynthetic rate, thus, related to the A/g ratio (the demand and supply of CO2 respectively). Since Δ13C and ci/ca are partially determined by the A/g ratio, measurements of Δ13 C provide a relative index of WUEintrinsic or WUEinstantaneous for given VPD conditions (e.g. Pate 2001). In fact, a negative correlation between Δ13C and WUE has been contrasted in several works (e.g. Hubick and Farquhar 1989; Morgan et al. 1993; Johnson 1993). It must be noted that from a mechanistic point of view, Δ13C is related with instantaneous A/E ratio (or rather, with A/g ratio), but it can provide a time and spatially integrated estimate of water use efficiency. However, the correlation of Δ13C with WUE at other scales could be lower (for instance, Shaheen et al. 2005). Although the negative relationship between Δ13C and WUE is widely consistent throughout several studies (see references above), the sign and magnitude of the correlation with grain yield in C3 cereals is complex, and may be strongly influenced by several factors. In Mediterranean conditions, numerous studies reported a positive correlation between kernel Δ13C and grain yield of bread wheat (Morgan et al. 1993) durum wheat (Merah et al. 1999; Merah et al. 2001, Araus et al. 1997; 2003; Fischer et al. 1998; Clay et al. 2001; Royo et al. 2002) and barley (Voltas et al. 1998). In a recent study Monneveux et al. (2005) confirms these previous finding, although a consistent positive correlation between grain Δ13C and yield was observed only under post-anthesis water stress conditions. Under pre-anthesis and limited residual moisture stress, the correlation was weaker. At full irrigation, by contrast, no correlation was found (Monneveux et al. 2005). Summarising the former considerations, Royo et al. (2002) suggests that selecting for higher Δ13C (lower WUE) in breeding programs in Mediterranean basin could take advantage only under wet or irrigation conditions. A high Δ13C in the grains may involve different phenomenon, such as (a) a greater access to soil water (for instance, related to a deeper root systems) or higher water extraction capacity (e.g. occurrence of osmotic adjustment) (b) higher remobilization of stems reserves of pre-anthesis assimilates which may have a lower isotope signature and (c) an earlier flowering (see Condon et al. 2002 and references cited therein). For these reasons, it has been pointed out that kernel Δ13C may be a more cryptic parameter than leaf Δ13C (Condon et al. 2004). However, the correlation between leaf Δ13C and grain yield are poor in some cases (Merah et al. 2001). Several works carried out by CSIRO’ group (Condon et al.) showed that breeding by low leaf Δ13C increased the grain yield under rainfed conditions in Australia (Rebetzke et al. 2002). In fact, the first

199



genotypes with low Δ13C and high grain yield were released during the 2002 and 2003 (v.g. Drydale and Rees cultivars; see web page of CSIRO: www.csiro.au; Condon et al. 2004). In according with this studies, the advantage of low carbon isotope discrimination is higher at low environment mean yields (i.e., when water stress is more severe). The improvement of low Δ13C selection, however, decline at higher mean yield, where higher season rainfall are present (Rebetzke et al. 2002). The apparent discrepancies between studies carried out in Mediterranean climate (e.g. in Southern Europe) versus some regions of Australia seem to arise from the source of water used during crop cycle. In Mediterranean zone, rainfall (although scarce) is present during the crop growth and stored water may be a minor component of the water used by the crop. In this context, genotypes with higher water extraction capacity will have higher grain yield and higher Δ13C. On other hand, in some regions of Australia with summer rainfall, stored water is a main part of the total water used by the crop. In those environments, rainfall are scarce after seeding, and the saving water could be a critical factor to avoid terminal severe water stress (Passioura 2004). In this context, cultivars with conservative strategy may have advantage respect ‘water spender’ ones. The negative correlation between the advantage in grain yield of low Δ13C varieties and rainfall (mentioned above) support this idea (Rebetzke et al. 2002). Simulation of the effect on yield incorporating higher instantaneous WUE (low Δ13C) showed that advantage was significant in environments where dominate stored water (with summer rainfall). In environments with Mediterranean climate (i.e. winter rainfall), by contrast, improved WUE did not confer advantage in yield (Condon et al. 2004). In this case, early vigour seems to be more important (see above). Araus et al. (2003) -analysing the environmental factors determining Δ13C in durum wheat under Mediterranean conditions in several trials with 25 genotypes- found that the correlation between kernel Δ13C and grain yield was steady a positive when mean yield of the trial above 2500 kg ha-1. By contrast, trials with mean yield below 2000 kg ha-1 showed low correlation between Δ13C and grain yield. Thus, in more stressful environments, Δ13C may be a poor indicator of grain yield, as it was suggested by previous reports (e.g. Royo et al. 2002). Where additional water is not available to the crop, to increase WUE (selecting by low Δ13C) appears to be an alternative strategy (Araus et al. 2002 and references therein).

Cereals yield progress and WUE In a retrospective study of wheat Siddique et al. (1990) found that WUEyield increased substantially from old to modern cultivars, with little difference among modern cultivars. WUEbiomass , by contrast, was similar between cultivars. Improved WUEyield in modern cultivars was associated with faster development, earlier flowering, improved canopy structure and higher harvest index (Siddique et al. 1990). More recently, in a study of several cultivars of bread wheat released in Mexico, Fischer et al. (1998) found an increase in photosynthetic rate and stomatal conductance in modern varieties: the rise in the last parameter (g) was higher than the former (A), leading to a decrease in WUEintrinsic of modern cultivars (calculated from date of Fischer et al. 1998). One question to answer is the range of genotypic variability in WUE in cereals. The range in carbon isotope discrimination among cereals cultivars is around 4‰. However, in Aegilops geniculata (closely related to Triticum) the range seem to be higher (ca. 7‰), suggesting that WUE of wheat could be improved by introgression in hybridisation programs (Zaharieva et al. 2001).

WUE AND AGRONOMIC PRACTICES: IRRIGATION AND FERTILIZATION Agronomic options for improving rainfall-use in dry land regions were extensively reviewed by Turner (2004): at least, half of the increase of rainfall-use efficiency may be attributed to improved agronomic management. The adoptions of practices such as minimum tillage, appropriate fertilization, timely planting, in conjunction with new cultivars, has the potential to increase rainfall use efficiency of dry land (such as Mediterranean) crops. Here, we will discuss irrigation and fertilization practices in relation to physiological aspects of WUE.

Irrigation

200



The amount, timing and frequency of irrigation have strong impacts in WUEbiomass and WUEyield. In fact, WUE may drop (e.g. Huang et al. 2005) or increase (Brandypadhyay et al. 2003) at higher irrigation levels. Irrigation may increase the WUEyield without any effect in WUEbiomass, due to the increase of postanthesis water use, which results in a higher harvest index and better grain yield (Zhang et al. 1998). This author pointed out that the lack of an effect in WUEbiomass could result from two counteracted forces: the decrease of Es/T ratio (related with early growth) may be counterbalanced by the decrease of ‘transpiration efficiency’ (in this context, the ratio between dry matter and plant transpiration), which could result from a greater leaf area and higher stomatal conductance in irrigated treatments (Zhang et al. 1998). Oweis et al. 2000 reported that WUEyield of bread wheat under Mediterranean conditions was higher at 2/3 of irrigation requirements (compared with at full irrigation, where WUEyield was lower). Similar results are reported by Tavakkoli and Oweiss (2004). As these authors pointed out, the common practice of supplemental irrigation is not the most efficient in terms of WUEyield for Mediterranean environments. Considering that water is the main limiting resource in dry areas, the loss of grain yield due to deficit irrigation may be negligible compared with the saving in water (Oweis et al. 2000). ‘Deficit irrigation’ is a strategy under which crops are deliberately allowed to sustain some degree of water deficit and yield reduction (Pereira et al. 2002). In Northern Syria, for instance, the maximum in water productivity of wheat are achieved with some grain yield reduction (see references in Pereira et al. 2002). In fact, increases of WUEyield under water limitation are reported in several studies and climatic conditions (e.g. Abbate et al. 2004 and references cited therein). In the last years, considerable attention have received the irrigation systems where a part of root is subjected to dry conditions (known as ‘CAPRI’ or ‘Controlled Alternate Partial Root Irrigation’). This technique is based in two assumptions (a) fully irrigated plants usually have widely opened stomata and (b) roots in the drying soil can respond by sending a root signal to the shoots, where the stomata may be partially closed, increasing WUEinstantaneous (Kang and Zhang 2004). Although these systems could be implemented in several crops, their use in cereals could be more doubtful. As it was mentioned earlier, in dense canopies (such as cereals crops) boundary layer resistance may be high, and exert a main control over the transpiration. Additionally, the increase of leaf temperature coupled with lower transpiration might eliminate any advantage of stomatal close. However, the advantages of ‘deficit irrigation’ systems (see above) suggest that stomatal control is a useful feature to improve WUE at a crop level. In summary, WUEyield in Mediterranean regions may be incremented with some irrigation, in particular at the beginning of the crops, which improves early growth and decreases soil evaporation. Additionally, some degree of water deficit may improve WUEyield, which it could be explained to some extent by stomatal control on the transpiration (Abbate et al. 2004).

Fertilization Fertilizer use has a remarkable effect on crop yield and WUE. There are several studies reporting the increase of cereal WUE with N application (e.g. Zhang et al. 1998). At first, nitrogen fertilization may increase the CO2 assimilation rate capacity, i.e. WUEyield may be ameliorated due to the improvement in WUEnstantaneous. In addition, and probably more importantly, fertilization increases the early growth and the crop cover, protecting the soil from evaporation and, consequently, increasing the proportion of transpired water by the plant (see references above). Additionally, WUE is also ameliorated because more crop growth takes place with lower VPD in early spring (Zhang et al. 1998). Nitrogen and phosphorous nutrition have been shown to increase the early growth in Mediterranean environments (see above; references in Turner 2004). In wheat, nitrogen fertiliser input reduced soil evaporation, increasing the WUEyield (Asseng et al. 2001; Sadras 2002). For barley in Mediterranean conditions, higher WUEbiomass and WUEyield were achieved by nitrogen application (Cantero-Martínez et al. 2003). According to this study, the higher WUE was explained by an increase of preanthesis WUE. WUE was increased by nitrogen fertilization only when water was not limiting: an excess of nitrogen may increase the water use, without improving WUE (CanteroMartínez et al. 2003). Gregory et al. 2000 also reported an increase in WUEbiomass in fertilised barley. Fertilization with N and P increased the shoot dry weight, the T/(Es + T) ratio and the WUEbiomass.

201



Total water use (Es + T) was not modified, but the fertilization treatment increased the WUE changing the partitioning between Es and T (Gregory et al. 2000). Furthermore, although there are several studies that showed the correlation between vigorous growths for fertilization (at early stages) and grain yield (see above), this seem not to be the case in some environments. In dry regions of Australia, a higher vegetative growth can lead to a reduction of yield (phenomenon named as ‘haying off’) in particular when vigorous growth is followed by a terminal severe water stress (Angus and van Herwaarden 2001). Although at first this yield reduction could be explained by the lack of soil water during grain filling -because more water was used in producing additional vegetative material- another causes have been proposed. For instance, the decrease of soluble carbohydrate available for retranslocation associated with nitrogen availability (see Angus and van Herwaarden 2001 and references cited therein). In spite of the subjacent causes, the association between more vigorous growth by nitrogen fertilization and higher WUE seem to be not universal, and peculiarities of the environment target must be considered. Finally, it must be noted that there is a ‘trade-off’’ between a higher WUEinstantaneous and photosynthetic nitrogen-use efficiency, because negative correlations between both parameters have been reported (e.g. Van Den Boogaard 1997 and references cited therein). As it was pointed out by this author, improving WUEinstantaneous may be suitable in environments where water is the limiting factor, but the cost of a less efficient use of nitrogen should be considered.

REFERENCES Abbad H, El Jaafari SA, Bort J, Araus JL. 2004. Comparative relationship of the flag leaf and the ear photosynthesis with the biomass and grain yield of durum wheat under a range of water conditions and different genotypes. Agronomie 24: 19-28. Abbate PE, Dardanelli JL, Cantarero MG, Maturano M, Melchiori RJM, Suero EE (2004) Climatic and water availability effects on water-use efficiency in wheat. Crop Science 44: 474-483. Acevedo EH, Silva PC, Silva HR, Solar BR (1999) Wheat production in Mediterranean environments. In ‘Wheat: ecology and physiology of yield determination’ (Eds Satorre EH and Slafer GA). Food Products Press, New York, pp. 295-323. Angus JF, van Herwaarden AF (2001) Increasing water use and water use efficiency in dryland wheat. Agronomy Journal 93: 290-298. Araus JL, Amaro T, Casadesús J, Asbati A, Nachit MM (1998) Relationship between ash content, carbon isotope discrimination and yield in durum wheat. Australian Journal of Plant Physiology 25: 835-842. Araus JL, Febrero A, Vendrell P (1991) Epidermal conductance in different parts of durum wheat grown under Mediterranean conditions: the role of epicuticular waxes and stomata. Plant, Cell and Environment 14: 545-558. Araus JL, Bort J, Ceccarelli S, Grando S (1997) Relationship between leaf structure and carbon isotope discrimination in field grown barley. Plant Physiology and Biochemistry 35: 533-541. Araus JL, Brown HR, Febrero A, Bort J, Serret MD. 1993. Ear photosynthesis, carbon isotope discrimination and the contribution of respiratory CO2 to differences in grain mass in durum wheat. Plant Cell and Environment 16: 383-392. Araus JL, Santiveri P, Bosch-Serra D, Royo C, Romagosa I. 1992. Carbon isotope ratios in ear parts of Triticale. Plant Physiology 100: 1033-1035. Araus JL, Casadesus J, Bort J (2001) Recent tools for the screening of physiological traits determining yield. In ‘Application of physiology in wheat breeding’ (Reynolds MP, Ortiz-Monasterio JI, McNab A eds.), CIMMYT, México DF. pp. 59-77. Araus JL , Slafer GA, Reynolds MP, Royo C (2002) Plant breeding and drought in C3 cereals: what should we breed for? Annals of Botany 89: 925-940.

202



Araus JL, Villegas D, Aparicio N, García Del Moral LF, El Hani S, Rharrabti Y, Ferrio JP, Royo C (2003) Environmental Factors Determining Carbon Isotope Discrimination and Yield in Durum Wheat Under Mediterranean Conditions. Crop Science 43: 170-180. Ashraf M, Bashir A (2003) Relationship of photosynthetic capacity at the vegetative stage and during grain yield of two hexaploid wheat (Triticum aestivum L. ) cultivars differing in yield. European Journal of Agronomy 19: 277-287. Asseng S, Turner NC, Keating BA (2001) Analysing of water- and nitrogen-use efficiency of wheat in a Mediterranean climate. Plant and Soil 233: 127-143 Asseng S, Turner NC, Botwright T, Condon AG (2003) Evaluating the impact of a trait for increased specific leaf area on wheat yields using a crop simulation model. Agronomy Journal 95: 10-19. Austin RB (1999) Yield of wheat in the United Kingdom: recent advances and prospects. Crop Science 39: 1604-1610. Bolger TP, Turner NC (1998) Transpiration efficiency of three Mediterranean annual pasture species and wheat. Oecologia 115: 32-38. Bort J, Febrero A, Amaro T, Araus JL. 1994. Role of awns in ear water-use efficiency and grain weight in barley. Agronomie 2: 133-139. Blum A (1996) Yield potential and drought resistance: are they mutually exclusive? In ‘Yield potential in wheat: breaking the barriers (Reynolds MPS, Rajaram, Mc Nab A eds.). Mexico DF CIMMYT, pp. 90-100. Blum A, Zhang J, Nguyen HT (1999) Consistent differences among wheat cultivars in osmotic adjustment and their relationship to plant production. Field Crops Research 64: 287-291. Blum A (2000) A molecular perspective of crop adaptation and production under drought stress. V Simposium Hispano-Portugués de relaciones hídricas - Sevilla Bandyopadhyay PK, Mallick S (2002) Actual evapotranspiration and crop coefficients of wheat (Triticum aestivum) under varyiing moisture levels of humid tropical canal command area. Agricultural Water Management 59: 33-47. Bort J, Brown HR, Araus JL. 1996. Refixation of respiratory CO2 in the ears of C3 cereals. Journal of Experimental Botany 47: 1567-1575. Cantero-Martínez C, Angas P, Lampurlanés J (2003) Growth, yield and water productivity of barley (Hordeum vulgare L.) affected by tillage and N fertilization in Mediterranean semiarid, rainfed conditions in Spain. Field Crop Research 84: 341-357. Chaerle L, Saibo N, van Der Straeten (2005) Tunning the pores: towards engineering plants for improved water use efficiency. Trends in Biotechnology 23 (6). Chen C, Payne WA, Smiley RW, Stoltz MA (2003) Yield and water-use efficiency of eight wheat cultivars planted on seven dates in Northeastern Oregon. Agronomy Journal 95: 836-843. Clay DE R. Engel RE, Long DS, Liu Z (2001) Nitrogen and Water Stress Interact to Influence Carbon13 Discrimination in Wheat. Soil Sci. Soc. Am. J. 65: 1823-1828. Condon AG, Richards RA, Rebetzke GJ, Farquhar GD (2002) Improving intrinsic water-use efficiency and crop yield. Crop Science 42: 122-131. Condon AG, Richards RA, Rebetzke GJ, Farquhar GD (2004) Breeding for high water-use efficiency. Journal of Experimental Botany 55 (407): 2447-2460. Corbeels M, Hofman G, Van Cleemput V (1998) Analysis of water use by wheat grown on a cracking clay soil in a semi-arid Mediterranean environment: weather and nitrogen effects. Agricultural Water Management 38 (2): 147-167 del Blanco IA, Rajaram S, Kronstad WE, Reynolds MP (2000) Physiological performance of synthetic hexaploid wheat-derived population. Crop Science 40: 1257-1263. El Hafid R, Smith DH, Karrou M, Samir K (1998) Physiological responses of spring durum wheat cultivars to early-season drought in a Mediterranean environment. Annals of Botany 81: 363-370. Farquhar GD, Richards RA. 1984. Isotopic composition of plant carbon correlates with water-useefficiency of wheat genotypes. Australian Journal of Plant Physiology 11: 539-552. Febrero A, Vendrell P, Alegre L, Araus JL (1991) Epidermal conductance in flag leaves and ears of several landraces and varieties: morphological and anatomical characteristics involved. In Physiology/Breeding of Winter Cereals for Stressed Mediterranean Environments. Les Coloques 55: 143-157.

203



Fischer RA, Turner NC (1978) Plant productivity in arid and semiarid zones. Annual Review of Plant Physiology 29:277-317. Fischer RA, Rees D, Sayre KD, Lu ZM, Condon AG, Larque Saavedra A (1998) Wheat yield progress associated with higher stomatal conductance and photosynthetic rate, and cooler canopies. Crop Science 38: 1467-1475. Gebbing T, Schnyder H (2001) 13C labeling kinetics of sucrose in glumes indicates significant refixation of respiratory CO2 in the wheat ear. Australian Journal of Plant Physiology 28: 10471053. Gregory PJ, Simmonds LP, Pilbeam CJ (2000) Soil type, climatic regime and the response of water use efficiency to crop management. Agronomy Journal 92: 814-820. Hatfield JL, Sauer TJ, Prueger JH (2001) Managing soil to achieve greater water use efficiency: A review. Agronomy Journal 93: 271-280. Havaux M, Tardy F (1999) Loss of chlorophyll with limited reduction of photosynthesis as an adaptive response of Syrian barley landraces to high-light and heat stress. Australian Journal of Plant Physiology 26: 569-578. Huang Y, Chen L, Fu B, Huang Z, Gong J (2005) The wheat yields and water-use efficiency in the Loess Plateau: straw mulch and irrigation effects. Agricultural Water Management 72: 209-222. Hubick KT, Farquhar GD (1989) Carbon isotope discrimination and the ratio of carbon gains to water lost in barley cultivars. Plant, Cell and Environment 12: 795-804. Johnson RC (1993) Carbon isotope discrimination, water relations, and photosynthesis in tall fescue. Crop Science 33: 169-174. Kang S, Zhang J (2004) Controlled alternate partial root-zone irrigation: its physiological consequences and impact on water use efficiency. Journal of Experimental Botany 55 (407): 2437-2446. Karrou M (1998) Observations on effect of seeding pattern on water-use efficiency of durum wheat in semi-arid areas of Morocco. Field Crops Research 59 (3): 175-179. Kato T, Kimura R, Kamichika M (2004) Estimation of evapotranspiration, transpiration ratio and wateruse efficiency from a sparse canopy using a compartment model. Agricultural Water Management 65: 173-191. Klein JD, Mufradi I, Cohen S, Hebbe Y, Asido S, Dolgin B, Bonfil DJ (2002) Establishment of wheat seedlings after early sowing and germination in an arid Mediterranean environment. Agronomy Journal 94: 585-593. Lawlor DW (2002) Limitation to photosynthesis in water-stressed leaves: stomata vs. metabolism and the role of ATP. Annals of Botany 89: 871-885. Leuning R, Condon AG, Dunin FX, Zegelin S Denmead OT (1994) Rainfall interception and evaporation from soil below a wheat canopy. Agricultural and Forest Meteorology 67(3-4): 221238 Li ZZ, Li WD, Li WL (2004) Dry-period irrigation and fertilizer application affect water use and yield of spring wheat in semi-arid regions. Agricultural Water Management 65: 133-143. López-Castañeda C, Richards RA, Farquhar GD (1995) Variation in early vigor between wheat and barley. Crop Science 35: 472-479. Merah O, Deléens E, Moneveaux P (1999) Grain yield, carbon isotope discrimination, mineral and silicon content in durum wheat under different precipitation regimes. Physiologia Plantarum 107: 387-394. Merah O, Deléens E, Souyris I, Nachit M, Monneveux P (2001) Stability of carbon isotope discrimination and grain yield in durum wheat. Crop Science 41: 677-681. Monneveux P, Reynolds MP, Trethowan R, González-Santoyo H, Peña RJ, Zapata F (2005). Relationship between grain yield and carbon isotope discrimination in bread wheat under four water regimes. European Journal of Agronomy 22: 231-242. Morgan JA, LeCain DR (1991) Leaf gas exchange and related leaf trait among 15 winter genotypes. Crop Science 31: 443-448. Morgan JA, LeCain DR, McCaig TN, Quick JS (1993) Gas exchange, carbon isotope discrimination, and productivity in winter wheat. Crop Science 33: 178-186.

204



Oweis T, Zhang H, Pala M. (2000) Water Use Efficiency of Rainfed and Irrigated Bread Wheat in a Mediterranean Environment. Agronomy Journal 92: 231-238. Passioura J (2004) Increasing crop productivity when water is scarce- from breeding to field management. In ‘New directions for a diverse planet’ Proceedings of 4th International Crop Sciences Congress, Brisbane, Australia. Pate JS (2001) Carbon isotope discrimination and plant water-use efficiency. In Stable Isotope Techniques in the Study of Biological Process and Functioning of Ecosystems (Unkovich et al. eds), Kluwer Academic Publishers, The Netherlands, pp 19-36. Pereira LS, Oweis T, Zairi A (2002) Irrigation management under water scarcity. Agricultural Water Management 57: 175-206. Polley WH (2002) Implications of Atmospheric and Climatic Change for Crop Yield and Water Use Efficiency. Crop Science 42: 131-140. Poorter H (1989) Interspecific variation in relative growth rate: on ecological causes and physiological consequences. In ‘Causes and consequences of variation in growth rate and productivity of higher plants?’ (Lambers et al. eds.), Academic Publishing, The Hague, The Netherlands, pp.45-68. Rebetzke GJ, Condon AG, Richards RA, Farquhar GD (2002) Selection for reduced carbon isotope discrimination increases aerial biomass and grain yield of rainfed bread wheat. Crop Science 42: 739-745. Rebetzke GJ, Botwright TL, Moore CS, Richards RA, Condon AG (2004) Genotypic variation in specific leaf area for genetic improvement of early vigour in wheat. Field Crop Research 88: 179189. Reynolds MP, van Ginkel M, Ribaut JM (2000) Avenues for genetic modification of radiation use efficiency in wheat. Journal of Experimental Botany 51: 459-473. Reynolds MP, Nagarajan S, Razzaque MA, Ageeb OAA (2001). Heat tolerance. In ‘Application of physiology in wheat breeding’ (Reynolds MP, Ortiz-Monasterio JI, McNab A eds.), CIMMYT, México DF. pp. 88-100. Richards RA, Rebetzke GJ, Condon AG, van Herwaarden (2002) Breeding opportunities for increasing the efficiency of water use and crop yield in temperature cereals. Crop Science 42: 111-121. Richards RA, Condon AG, Rebetzke (2001) Traits to improve yield in dry environments. In ‘Application of physiology in wheat breeding’ (Reynolds MP, Ortiz-Monasterio JI, McNab A eds.), CIMMYT, México DF. pp. 88-100. Poorter H (1989) Interspecific variation in relative growth rate: on ecological causes and physiological consequences. In Causes and consequences of variation in growth rate and productivity oh higher plants (H Lambers ed.), SPB Academic Publishing, The Hague, The Netherlands, pp. 45-68. Rekika D, Nachit MM, Araus JL, Monneveaux P (1998) Effects of water deficit on photosynthesis rate and osmotic adjustment wheats. Photosynthetica 32 (1): 129-138. Richards RA, Condon AG, Rebetzke (2001) Traits to improve yield in dry environments. In ‘Application of physiology in wheat breeding’ (Reynolds MP, Ortiz-Monasterio JI, McNab A eds.), CIMMYT, México DF. pp. 88-100. Royo C, Villegas D, García del Moral LF, Elhani S, Aparicio N, Rharrabti Y, Araus JL (2002) Comparative performance of carbon isotope discrimination and canopy temperature depression as predictors of genotype differences in durum wheat yield in Spain. Australian Journal of Agricultural Research 53: 561-569. Sadras V (2002) Interaction between rainfall and nitrogen fertilization of wheat in environments prone to terminal drought: economic and environmental risk analysis. Field Crop Research 77: 201-215. Shaheen R, Hood-Nowotny RC (2005) Effect of drought and salinity on carbon isotope discrimination in wheat cultivars. Plant Science 168: 901-909. Shangguan ZP, Shao MA, Dyckmans J (2000) Nitrogen nutrition and water stress effects on leaf photosynthetic gas exchange and water use efficiency in winter wheat. Environmental and Experimental Botany 44: 141-149. Siddique-KHM, Tennant D, PerryMW, Belford RK (1990) Water use and water use efficiency of old and modern wheat cultivars in a Mediterranean-type environment. Australian-Journal-ofAgricultural-Research 41(3): 431-447.

205



Skovmand B, Reynolds MP, Delacy IH (2001) Searching genetic resources for physiological traits with potential for increasing yield. In ‘Application of physiology in wheat breeding’ (Reynolds MP, Ortiz-Monasterio JI, McNab A eds.), CIMMYT, México DF. pp. 17-28. Snyder KA, Richards JH, Donovan LA (2003) Night-time conductance in C3 and C4 species: do plants lose water at night? Journal of Experimental Botany 54 (383): 861-865. Tardy F, Créach A, Havaux M (1998) Photosynthetic pigment concentration, organization and interconversions in a pale green Syrian landrace of barley (Hordeum vulgare L. Tadmor) adapted to harsh climatic conditions. Plant, Cell and Environment 21: 479-489. Tavakkoli AR, Oweis TY (2004) The role of supplemental irrigation and nitrogen in producing bread wheat in the highlands of Iran. Agricultural Water Management 65: 225-236. Teare ID, Sij LW, Waldren RP, Goutz SM. 1972. Comparative data on the rate of photosynthesis, respiration and transpiration of different organs in awned and awnless isogenic lines of wheat. Canadian Journal of Plant Science 52: 965-972. Turner NC (2004) Agronomic options for improving rainfall-use efficiency of crops in dryland farming systems. Journal of Experimental Botany 55 (407): 2413-2425. Van Den Boogaard R, Alewinjnse D, Veneklaas EJ, Lambers H (1997) Growth and water-use efficiency of 10 Triticum aestivum cultivars at different water availability in relation to allocation of biomass. Plant Cell and Environment 20: 200-210. Van Den Boogaard R, Veneklass EJ, Lambers H (1996) The association of biomass allocation with growth and water use efficiency of two Triticum aestivum cultivars. Australian Journal of Plant Physiology (now Functional Plant Biology) 23 (6): 751-761. Voltas J, Romagosa I, Muñoz P, Araus JL (1998) Mineral accumulation, carbon isotope discrimination and indirect selection for grain yield in two-rowed barley grown under semiarid conditions. Yunusa IAM, Sedgley RH, Belford RK Tennant D (1993) Dynamics of water use in a dry mediterranean environment I. Soil evaporation little affected by presence of plant canopy. Agricultural Water Management 24 (3) 205-224. Zhang H, Oweis TY, Garabet S, Pala M (1998) Water-use efficiency and transpiration efficiency of wheat under rain-fed conditions and supplemental irrigation in a Mediterranean-type environment. Plant and Soil 201: 295-305. Zaharieva M, Gaulin E, Havaux M, Acevedo E, Monneveux P (2001) Drought and heat responses in the wild wheat relative Aegilops geniculata Roth: potential interest for wheat improvement. Crop Science 41: 1321-1329.

206



207



PRODUCTIVITY OF THE POTATO CROP UNDER IRRIGATION WITH LOW QUALITY WATERS

N. Ben Mechlia*, K. Nagaz**, J. Abid-Karray*, M.M. Masmoudi* *Institut National Agronomique de Tunisie. **Institut des Régions Arides, 4119 Médenine, Tunisia. SUMMARY - Opportunities to enhance agricultural productivity in arid Tunisia have been the domain of active investigation during the last years. Many field experiments were carried out on the potential of using limited amounts of saline water from wells to grow profitably crops. This paper concerns the performance of the potato crop gown in commercial farms under relatively stressful conditions. The FAO-Water Balance calculation method was applied for irrigation scheduling over three contrasting seasons. Results obtained from full and deficit irrigation treatments show that under optimum irrigation, marketable yields varied from about 40 t/ha to 20 t/ha , and water productivity from 11.7 to 3 9.1 Kg/m , respectively for the spring and autumn periods of the year. We propose here a simple linear relationships between yield (Y, t/ha) and total water application (Q, mm) to be used for quick appraisals of attainable yields of the potato crop when waters of about 3-3.5 dS/m are used. Within a range of 150 to 350 mm total water supply after planting the models are [Y= 0.1Q+5] for the springsummer season and [Y=0.1Q-5] for the autumn-winter season. These models could be valuable tools for estimating yield gaps and potential for water savings concerning the potato crop grown under arid environments. Key words: irrigation scheduling, potato yield, water productivity, salinity.

INTRODUCTION

208



Potato is considered relatively susceptible to salinity (Maas and Hoffman, 1977) and normally is not suited for stressful conditions. However its cultivation has been gaining popularity during the last decade in arid areas as a cash crop because temperature and radiation conditions allow for cropping over the spring, fall, and winter seasons. In the Southern part of Tunisia where mean annual rainfall is less than 200 mm, potato is cultivated primarily on shallow wells in private farms where water is limiting not land. Irrigation is typically applied on a routine basis without scheduling. Surveys carried out recently in the governorate of Medenine show that potato yield vary usually between 10 and 20 t/ha. Inadequate management of irrigation has been identified as an important limiting factor (Nagaz and Ben Mechlia, 2003, Nagaz et al. 2004). In farms where cultivation is practised under drip irrigation water savings by drip seem to be forfeited with inappropriate scheduling. Yield is greatly influenced by timing, amount and frequency of irrigation applied, therefore, precise knowledge of the amount of water required by the crop and the proper timing for supply is essential. Scheduling based on crop water requirements and soil characteristics allows for applying irrigation water when needed during the growing season. However, its application is only possible when water supply and irrigation amounts can be managed independently by farmers (Smith, 1985). In areas where potato is irrigated with well waters, accurate scheduling is manageable. This is precisely the case of our area, therefore chances to achieve tangible productivity improvements are high. Field trials were implemented with the objective to evaluate the applicability of representative irrigation scheduling methods for drip systems. Basically, the investigation had to quantify yield, water use efficiency and soil salinity when the soil water balance method is used instead of the prevailing common approach. With the expectation to enable growers to incorporate more appropriate irrigation scheduling methods in their usual production practices, all field work was conducted with farmers participation.

MATERIALS AND METHODS Irrigation-scheduling experiments were carried out during the years 2000 to 2004 in commercial farms situated in the south-eastern part of Tunisia near the “Institut des Régions Arides, IRAMdenine”. The potato cultivar "Spunta" was used following standard cultural practices as reported by Nagaz et al. (2004) over three distinct growing periods: spring, autumn and winter seasons. For the irrigation-scheduling experiment, the soil water balance (SWB) methodology was adopted. A spreadsheet program estimates the day when the soil readily available water (RAW) would be depleted and estimates the amount of irrigation water needed to replenish the soil profile to field capacity. The target depletion threshold was set to 35 % of total available water in the root zone (TAW). The program calculates the soil water depletion with potato root depths estimated on daily basis. The soil depth of the effective root zone is increased with the program from a minimum depth of 0.15 m at planting to a maximum of 0.60 m in direct proportion to the increase in the potato crop coefficient. Once the maximum root depth is reached, it is held constant. The soil is of a sandy type with low organic matter content. The total soil available water, calculated between field capacity and wilting point for an assumed potato root extracting depth of 0.60 m, is 75 mm. The crop evapotranspiration (ETc) was estimated for daily time step by using reference evapotranspiration (ET0) combined with a potato crop coefficient (Kc), following the FAO-56 method given in Allen et al. (1998). Weather data used in this experiment (Tmax, Tmin and u2) were collected daily from a nearby meteorological station located at IRA-Medenine. The development of quantitative relationships between yield and water supply was an integrated part of this study. To produce experimental data, the layout included four distinct water treatments with decreasing amounts in relation to maximum crop evapotranspiration (ETc). Full irrigation is the

209



treatment that received 100 % of accumulated ETc. Deficit treatments were irrigated at the same frequency as the control, but with quantities equal to 80, 60 and 40 % of accumulated ETc . The producer method consisted in the supply of fixed amounts of water of about 17 mm to the crop every 5 days from planting till harvest. This method corresponds to irrigation practices traditionally implemented by local farmers using drip irrigation. A randomised block design with four replications was used as trial layout. Water, obtained from a well with a conductivity of 3.25 dS/m, was applied by means of polyethylene dripper lines with an emission rate of 4 l h-1. For each block, it passed through a water meter, gate valve, before passing through laterals placed in every potato row. A control mini-valve in the lateral permits use or non-use of the dripper line. Before planting, the soil was set to field capacity over a depth of 0.6 m i.e. all treatments received a start amount of 75 mm. Our investigation was concerned with the effect of different water management tactics corresponding to decreasing irrigation water supplies on yield, water saving and soil salinization. Potato yields were estimated from samples of ten plants per row within each plot. Soil samples were collected after harvest and analyzed for ECe.

RESULTS The SWB scheduling techniques based on limited weather information and standard data provided in FAO guidelines was effective for managing irrigation under ordinary farming conditions. Maximum yields obtained under this scheduling technique (Yo) were respectively 39.7, 30.4 and 22.7 t/ha for spring, autumn and winter crops.

Yield water supply relationships Reduction of irrigation amounts resulted in proportional reduction of yield (Figure 1). Because all treatments were uniformly at field capacity at the beginning of the cropping seasons, the impact of irrigation reduction varied slightly from one season to the other, depending on the contribution of soil reserves to the overall water supply to plants. Intuitively one can assume that the least irrigated treatment would rely more on soil water initial store, provided that sufficient roots were developed over the wetted profile. The little amounts of rainfall waters although relatively limited can also affect the plant water status and therefore mitigate to different extents the effect of irrigation restrictive regimes.

210



Fig. 1. Relative yield decrease of the potato crop in relation to reduction of irrigation, for different production seasons in Southern Tunisia. Figure 2 represents yields obtained under situations of full and deficit irrigation regimes with saline waters. Two simple linear relationships are identified for the spring and for the autumn-winter seasons. The proposed quantitative models are: Y = 0.1 Q + 5, for spring productions

(1)

Y = 0.1 Q - 5, for the autumn-winter season

(2)

with Y the yield of fresh potato tuber in t/ha, and Q the total water supply during the growing season, including the little amounts of rainfall. As empirical relationships, these equations correspond to situations where applied water is comprised between 150-350 mm limits, water salinity about 3 dS/m and rainfall input representing less than 20 % of the total supply. It is worthy to notice that these results were obtained on sandy soils not significantly affected by salinity (ECe less than 4 dS/m). The obtained models have different intercepts because of differences between growing conditions in spring and Autumn seasons. Temperature and radiation regimes affect the partitioning of dry matter between tuber and vegetative pert. High temperatures and low radiation are known to be favorable to vegetative development (Autumn), whereas cool conditions associated with important radiation loads (spring) are suitable for more tuber production. Our results show that the ratio between dry matter of tubers and total dry matter biomass is 10-13% higher for spring productions than for the other two seasons.

211



Fig. 2. Potato yields obtained with irrigation waters having an ECi of 3.25 dS/m over three contrasting growing seasons in Medenine, southern Tunisia. The soil is sandy with ECe lower than 4 dS/m and rainfall represented about 6 to 20% of the total supply. Prior to planting, all treatments received 75 mm to put the top 60 cm soil layer at field capacity. Marginal gains in fresh tuber yield associated with increased water supply seem to be similar for the various cropping periods. The slope of the idealized linear relationships indicates an overall water productivity of 100 kg ha-1mm-1. This value fall within ranges commonly reported in the literature (Bowen 2003).

Water use efficiency and potential for water saving Amounts of irrigation water and total water supply for the two growing seasons are presented in Table 1. With the producer method more water was used than the SWB. Surplus was about 63 mm. The water use efficiency (WUE) expressed as the ratio of potato yield to water supply from 3 planting to harvest varied typically from 11 to 9 Kg/m (Table 1). These values are comparable to those obtained in other field studies (Bowen, 2003; Fabeiro et al., 2001; Ferreira et al., 1999). Maximum values were observed during the spring season (11.7 Kg/m3) under full irrigation with the SWB method. Since water used to prepare the field for planting (75 mm) is not included in the calculation of total water supply, caution should be used in analyzing productivity under deficit irrigation. It would be normal that with differential water reduction levels variable amounts of the soil readily available waters are used. Actually water use efficiency (WUE) was significantly affected by deficit irrigation treatments. The highest values occurred in the 100 % and 80 % treatments. It was also observed that significant reduction in marketable tuber size (tuber weight) were associated with the 60 % and 40 % regimes. The lower yields and WUE values obtained by the producer are attributable to the fact that water was applied regardless of plant changing ETc. Irrigation occurrences relates to days after planting rather than to crop growth stages progress and to ETo values. The relatively high yields and water use efficiency values noted under full irrigation in both seasons indicate the high potential of the potato crop to valorize irrigation waters of limited quality, provided that good management is applied.

212



Table 1. Water supply from planting to harvest and water use efficiency (WUE) measured under SWB scheduling and grower's practices.

Soil Salinity The electrical conductivity (ECe) values measured before planting were, respectively, 1.35 and 3.45 dS/m (0-60 cm depth of soil) for spring and autumn seasons. Under the different irrigation scheduling methods, ECe in both seasons remained lower than ECi of the irrigation water. Low ECe levels were observed after rainfall events. The leaching by rain occurred mainly during September, October and December in autumn season (72 mm), and in April and May in spring season (29.7 mm). Under the prevailing conditions leaching seemed to be sufficient to control the build-up of soluble salts in the profile of this well-drained soil. Singh and Bhumbla (1968) observed that the extent of salt accumulation depended on soil texture and reported that in soils containing less than 10 % clay the ECe values remained lower than ECi. Differences between treatments within a given season were also observed. With SWB full irrigation, the average ECe value was equal to 1.0 dS/m, beneath the emitter in autumn and to 0.75 dS/m in spring season. The zone of highest ECe was moved out to 20 cm from the emitter. With the producer method values of 1.9 and 1.7 dS/m were recorded below the emitter, respectively, in the spring and autumn seasons.

CONCLUSION The potential of irrigation scheduling to improve yield and to save water has been demonstrated in this work. Results, obtained under actual farming conditions, support the practicality and usefulness of using the Soil Water Balance (SWB) scheduling method as simplified by FAO to optimise irrigation in arid regions. Using well waters having an ECi of 3.25 dS/m, it was possible to produce potato at about 20 to 40 3 t/ha, depending on the period of the year. Scheduling can ensure a water productivity of 10 Kg/m of water, in spite of the stressful environmental conditions. The proposed quantitative linear relationship between yield and water applied from planting to harvest could be used as a tool for selecting the appropriate irrigation tactic to respond to irrigation water shortages after planting. Y = 0.1 Q + 5, for spring productions

213



Y = 0.1 Q - 5, for the autumn-winter season These model can be also used to investigate the causal factors for potato productivity variations and yield gaps in areas similar to Southern Tunisia.

REFERENCES Allen R.G., Perreira L.S., Raes D., Smith M., 1998. Crop evapotranspiration: Guidelines for computing crop water requirements. Irrigation and Drainage Paper N° 56, FAO, Rome, Italy, 300p. Bowen W. T., 2003. Water productivty and potato cultivation. In Water Productivity in Agriculture : Limits and Opportunities for Improvement. Ed. J. W. Kinje, R. Baker and D. Molden. CAB international, pp.229-238. Fabeiro C., Martin De Santa Ollala F., De Juan J.A., 2001. Yield and size of deficit irrigated potatoes. Agriculture Water Management, 48, pp.255-266. Ferreira T.C., Malheiro A.N.C, Freixo F.A.M.F.P., Bernardo A.A.S, Carr M.K.V., 1999. Variation in the response to water and nitrogen of potatoes (Solanum tuberosum L.) grown in the highlands of Northeast Portugal. In: Proceedings of 14th Triennial Conference of the European Association for Potato Research, Sorrento, Italy, May, 2-7, 1999, 410-411. Maas E.V., Hoffman G.J., 1977. Crop salt tolerance: Current assessment. J. Irrig. Drain. Div. Am. Soc. Civ. Eng., 103,pp.115-134. Nagaz K., Ben Mechlia N., 2003. Caractérisation de la conduite de pomme de terre en irrigué dans les périmètres privés sur puits de surface. Unpublished data. Nagaz K., Masmoudi M, Ben Mechlia N., 2004. Etude de la Réponse de la pomme de terre (Solanum tuberosum L.) de saison, de primeur et d’arrière-saison aux régimes d’irrigation à l’eau salée dans la région de Médenine. 4th AgroEnviron 2004 int. Symp., Udine-Italy October 20-24, 2004. Singh B. and Bhumbla D.R. 1968. Efeect of quality of irrigation water on soil properties. J. Res. Punjab Agri. Univ., 5, 166. Smith M., 1985. Irrigation scheduling and water distribution. In: Les besoins en eau des cultures. Actes de Conférence Internationale, INRA, Paris, France, 11-14 Septembre 1984, pp.497514.

214



USE OF THE HEAT DISSIPATION TECHNIQUE FOR ESTIMATING THE TRANSPIRATION OF OLIVE TREES

J. Abid Karray*,**, M.M. Masmoudi*, J.P. Luc**, N. Ben Mechlia* * Institut National Agronomique de Tunisie ** Institut de Recherche pour le Développement, UR-DIVHA. SUMMARY - The heat pulse method is used on olive trees for determination of transpiration. It consists of generating heat pulses at a point in the trunk and detecting their effect in another point. In spite of its usefulness its precision has not been demonstrated since the integration from one point measurement to the trunk cross section area could result in large errors. The more recently developed heat dissipation method, which is based on measuring temperature differences between heated and non heated probes, has been used on many forest species. Because it uses long probes (20 mm to 80 mm) it allows for better integration of the radial flow. The objective of this work is to investigate the practicality of using this technique for sap flow measurements of olive trees. To this end an experiment has been carried out on twelve year olive trees grown in central Tunisia. Four trees were selected for transpiration monitoring under changing water regime. On each tree three probes placed at different exposures were used, North (N), South East (SE) and South West (SW). Measurements of all twelve probes were collected every thirty minutes, over a complete growing season. Results obtained over periods where all probes were operating properly show a good coherence between sap flow estimates and environmental conditions. For the 95 days, during which -1 all sensors were operational, daily values of sap flow density vary from 10 to 120 l dm-2 d , about one to five ratio is observed between sensors. The mean value of sensors installed on each tree is highly correlated with the 12 sensors average values. When considering each direction separately, it seems that flux density is not related to the direction of the sensor, correlation between sap flow density of a sensor placed in a given direction and average value within that direction is variable. Variability of sap flow density between probes seems to be related rather to sapwood area heterogeneity. Regression equations relating sap flow density of each probe to the 12 probes average values were established and correlation coefficient exceed 0.85. With such calibrated equations, estimation of olive tree transpiration from a single probe and reconstitution of missing data becomes possible. Key words: olive tree, transpiration, sap flow, heat dissipation technique, model.

INTRODUCTION Estimation of transpiration is required for appropriate irrigation management, particularly for crops with variable planting densities. In orchards, quantifying water used by trees cannot be performed easily by methods based on water balance and micrometeorological measurements. However sap flow methods seem to have the potential of estimating the course of transpiration flux in a continuous manner. Presently a variety of methods are used successfully on many forest and fruit tree species. These methods can provide direct measurements and are easily automated, so continuous records of plant water use with high time resolution can be obtained. Sap flow measurements are also versatile because complex terrain and spatial heterogeneity does not limit their applicability. Sap flow can be estimated by heat-pulse, heat-balance or thermal dissipation methods. All of these techniques use the heat as a tracer for sap movement, but they are fundamentally different in their operating principals and each one have its merits and drawbacks. On olives, a modified heat pulse technique (Compensation Heat Pulse Velocity, CHPV) was first used by Moreno et al (1996) and Fernandez et al. (1997) on either roots to study hydraulic behavior or trunks to estimate whole-tree water consumption respectively. A good agreement was found between the transpiration determined by CHPV system and that predicted by the Penman-Monteith equation. However, Fernandez et al (1998) reviewed the performance of the CHPV technique and outlined its advantages and limitations. Indeed, this system is reliable and of low maintenance, it provides information on the dynamics of both water uptake by roots and tree transpiration and

215



determines the effect of meteorological conditions and soil water status on both processes. Nevertheless, the capability of the technique to estimate the tree transpiration is limited by the considerable heterogeneity of the conductive area in mature olive trees. Because this system uses point sampling, radial variability of sap flow results in large errors when scaling up to the whole tree. Consequently, many probes have to be installed at different depths and azimuths to account for radial variations. Girorio and Giorio, (2003) used the CHPV with six gauges and found a strong correlation between the sap flow measured by each gauge and the average of the remaining ones. Estimation of average sap flow is therefore possible using measurement of a single probe. However, the validity of such models requires appropriate calibration over long period to provide a high resolution response to the physiological and environmental factors. The heat dissipation technique developed by Granier, (1987) is based on the measurement of temperature differences between a heated and unheated probes. Use of probes with high temperature conductance allows integration of radial variability of flow along the probe length, but does not solve problems related to natural temperature gradients and sap flow density across the trunk as reported in the literature. The original method using continuous heating was therefore modified by Do et Rocheteau (2002) to avoid natural gradient effect on measurements. The heat dissipation technique using alternative heating is presently used in two sites in Tunisia having contrasting climates. This paper concerns the work carried out in the arid environment of central Tunisia on four olive trees. The objective is to investigate the variability of measurements among probes on a given tree and the inter-variability of sap flow density among trees. Relationships between single probe measurements and averaged values will be analyzed. It assumed that with quantitative inter-relationships between probes and appropriate calibration good quality data could be derived with fewer sensors and therefore an easy to manage monitoring system could be obtained.

EXPERIMENTAL LAYOUT The experimental work was carried out in a commercial orchard at Chebika near the city of Kairouan in central Tunisia (latitude 35°37’N, longitude 9°55’W, altitude 110 m). The climate is arid with an average annual rainfall of 280 mm. Four olive trees (Olea europea L., cv Chemlali) planted in 1993 and spaced 11×11 m, having similar shape (canopy and trunk diameter) have been chosen to be representative of the plot. Three probes are placed in each tree at three exposures: North (N), South East (SE) and South West (SW). The sensors, associated electronics and worksheets software have been developed locally (Masmoudi et al, 2004). Two data loggers (Campbell CR10X) have been used for continuous monitoring and recording of sensors signals. Measurements of the twelve sensors were collected every fifteen minutes, over a complete growing season from 01/01/2003 to 31/12/2003. The sensors are two cylindrical probes with 2 mm diameter and 20 mm length equipped with heating resistance and thermocouple and associated electronic modules for current regulation and control. Probes are inserted radially in the xylem and spaced vertically by 8 to 10 cm (Granier, 1987). The upper probe is heated with a constant power with an alternative 15 mn heating and cooling cycles while the lower probe is unheated. Sap flow density is calculated according to the Do and Rocheteau equation recalibrated for olive trees.

RESULTS A sample of 95 days of data has been selected and used for investigation. As it was mentioned previously measurements on all 12 sensors were taken every 15 minutes, and half an hour sap flow density were calculated and daily cumulative values were derived. A typical course of sap flow density during a summer day is given in Figure 1.

216



Fig. 1.Sap flow density (Fd) measured on four trees (T1 to T4) on three directions: North (N), South East (SE) and South West (SW) on 31/08/05 Sap flow given by all sensors increased rapidly after 7h15 to peak at about 09h15 and decline from 17h45 when solar radiation decreases. While there is a general consistency between the pattern of hourly measurements, absolute values of sap flow density obtained by individual sensors varied in high proportions. For instance, maximum value given by South Est (SE-T2) oriented sensor of tree 2 was 1.4 l dm-2 day-1 while the North oriented one on the same tree (N-T2) indicated 4.1 l dm-2 day-1. In order to characterize variability between sensors, daily values for individual sensors (Fdi) were compared to average values of all sensors (Fdm). Figure 2 shows that the sensors have the same behavior for the wide range of transpiration levels as observed over the 95 selected days.

217



Fig. 2. Sap flow density (Fdi) of individual probes (12) in relation to mean sap flow density (Fdm), data relates to 95 days covering a large range of environmental conditions. Overall sap flow patterns seem to be related to the probe orientation or position and to the local conditions of sap flow within the trunk. Consistency in the sensors relative change exclude any of random variation. Indeed, Table 1 gives the regression coefficients and shows that apart from sensors on tree 2 (T2) the regression coefficient is higher than 0.6. Table 1. Values of slope (a), intercept (b) and R2 of linear regression between sap flow density of each probe (Fdi) and 12 probes average (Fdm)

As transpiration is related to solar radiation, the existence of preferential orientation of the flow related to sun position is tested. To this end, regression equations have been established for each tree between values corresponding to the different directions (N, SE and SW) and the average sap flow of the considered tree. Figure 3 give results concerning the sap flow density on the three directions (Fdi-T4) versus mean sap flow density for tree 4 (Fdm-T4).

218



Fig. 3. Sap flow density for the three directions : North (N), South East (SE) and South West (SW) for tree (Fdi-T4) versus mean sap flow density of tree 4 (Fdm-T4) Table 2 gives the slope coefficients and R2 of all regressions. Correlation between individual sensors values and 12 sensors mean exceeds 0.7 except for tree 2 (T2). The slope of regression equation indicates that the relative weight of directions is quite random, there is no influence of the direction of the probe on average value of transpiration. The variability observed between signals from different probes is not due to the effect of exposure but to the variability of sap flow conductivity within the trunk section area. Table 2. Slope coefficients (a) and R2 of regressions between sap flow density of a given direction (North, South Est and South West) and mean sap flow density of the correspondent tree (T1, T2, T3, T4)

With increased number of sensors, average values will better represent the effective sap flow and take into account the heterogeneity of flow within the cross section. As shown in figure 4 a good correlation is observed between mean sap flow of each tree (Fdm-Ti) and the value obtained with all sensors (Fdm). The number of three sensors seems to be a good compromise between the validity and the complexity of the measurements.

Fig. 4. Mean sap flow density of each tree (Fdm-Ti) as related to mean sap flow density of all sensors (Fdm) CONCLUSION

219



The use of heat dissipation technique for measurement of sap flow is known as a technique which reduces the effect of radial sap flow heterogeneity as it integrates radial variability of flow along the probe length. The present work concerns the representatiness vity of a single or a set of sensors when installed on one tree or on several trees. Continuous measurements of 12 sensors installed on 4 similar trees grown under the same conditions showed that sap flow densities obtained are coherent with climatic conditions but large differences are observed between absolute values produced by different sensors. It was found that behavior and relative importance of a single sensor was consistent with reference to overall average for a wide range of transpiration. Mean values are well correlated to individual sensor outputs. There is no apparent relationship between orientation of sensors and the flow distribution within the trunk cross section. A number of three sensors seems to be adequate to produce good estimates of effective sap flow and cover heterogeneity of flow in the trunk cross section. Methodology using regression equations seems to be appropriate and could be used for reconstitution of missing values in case of failure of sensors or to reduce the number of sensors for long term monitoring. However the use of such models needs calibration for local ranges of transpiration.

REFERENCES Do F. and Rocheteau A., 2002. Influence of naturel temperature gradients on measurements of xylem sap flow with thermal dissipation probes. 1. Field observations and possible remedies. Tree Physiology, 22, p641-648. Do F. and Rocheteau A., 2002. Influence of naturel temperature gradients on measurements of xylem sap flow with thermal dissipation probes. 2. Advantages and calibration of a noncontinuous heating system. Tree Physiology, 22, p649-654. Giorio P. and Giorio G., 2003. Sap flow of several olive trees estimated with the heat-pulse technique by continuous monitoring of a single gauge. Environmental and Experimental Botany, 49, p9-20. Granier A.,1987. Mesure du flux de sève brute dans le tronc du Douglas par une nouvelle méthode thermique. Annales des Sciences Forestières, 44 (1), p1-14. Masmoudi M. M., Mahjoub I., Charfi-Masmoudi C., Abid-Karray J and Ben Mechlia N., 2004. Mise au point d'un dispositif de mesure du flux de sêve xylémique chez l'olivier. Revue des Régions Arides, ns 2004, p242-251. Moreno F., Fernández JE., Clothier B.E., Green S.R., 1996. Transpiration and root water uptake by olive trees. Plant and soil, 184, p85-96. Fernández JE., Palomo M.J., Díaz-Espejo A. and Girón I.F., 1997. Calibrating the compensation heat-pulse technique for measuring sap flow in olive. Acta Horticulturae, 474, 2, p455-458. Fernández JE., Palomo M.J., Díaz-Espejo A., Girón I.F. and Moreno F., 1998. Measuring sap flow in olive trees: potentialities and limitations of the compensation heat-pulse technique. Proceedings of the 4th Workshop on Measuring Sap Flow in Intact Plants. Zidlochovice, Czech Republic, 3-4 November, p.16-22.

220



WATER RESOURCES MANAGEMENT AT THE RIVER BASIN LEVEL: AN INSTITUTIONAL ANALYSIS

A. Billi*, A. Quarto*, E. Zini** *University of Rome “La Sapienza” ** Graduated at the University of Rome “La Sapienza”

SUMMARY - Water resources management at the river basin scale is, under the perspective of institutional analysis, a decentralization process aiming at maximizing socio-economic benefits related to water resources and minimizing transaction costs, in both the implementation of new institutional arrangements and the adaptation of existent arrangements to changing situations. After describing the two key-principles of decentralization and participation, the current work analyzes a number of important characteristics, factors and variables influencing the successful implementation of river basin management decentralization processes. The intent of this paper is to provide a practical tool for researchers and policy-makers, useful to describe – on a case-by-case basis – the performance of water resources management institutional arrangements at the river basin level. Key words: river basin management, water resources institutional arrangements, decentralization, participation.

INTRODUCTION Integrated water resources management at the river basin level is - since the beginning of 90s one of the key-themes of the world debate on water issues. The paradigm merges the two concepts of integrated management and of river basin management. The first, defined as integrated water resources management (IWRM), broadly refers to the integration of the natural system and the human system into the management of water resources: the integration is aimed at reconciling the aggregate supply and demand for water through structural and non-structural measures, and to achieve sustainable water management under the economic, the 2 social and the environmental profile. The second concept refers to the management of water resources at the river basin level, based upon the consideration that the river basin area constitutes a single inter-connected natural system (even if a complex system) hence it needs a corresponding coordination of collective decisions on that scale. For this reason, integrated river basin management is considered a particularly suitable form of implementation of the IWRM paradigm. The institutional perspective for the analysis of river basin management aims to identify the institutional arrangements associated with concrete sustainable outcomes. Moreover, the institutional asset and the dynamics of the processes of institutional change reveal the complex relations between managing water and the management of agriculture, socio-economic issues, and the natural environment. Thus, it is essential to consider the sustainable integrated management of water resources on a basin scale as a dynamic phenomenon, consisting of the process of balancing decisions through different needs and objectives, in a specific, uniquely-tailored way in every single scenario.

2

For the definition of the IWRM concept, see GWP (2000).

221



The implementation of water management at the basin scale means activating a process of coordination/cooperation between institutions and stakeholders. The process aims to introduce or reinforce institutional arrangements for water management at the basin level – not necessarily to create or strengthen an ad hoc institutional subject (such as a river basin organization). The general target of the process is the maximization of socio-economic welfare related to water resources, through coordination of multiple activities and resolution of conflicts arising between different users about the limited resource base. It is then fundamental to identify – case-by-case – the main problems to be properly faced at the basin level, and especially those connected with negative externalities at the basin scale (e.g. water pollution, allocative problems). Indeed, basin management is the result of interactions between different institutional subjects and different stakeholders in facing a specific set of problems, hence it has to be identified as a highly context-specific process. For this reason, since it is difficult to formulate a standard setting of institutional arrangements 3 (that is, an institutional model) expected to fit the situation of each basin, the recent literature – 4 mainly based upon a transaction costs approach – aims at establishing a set of factors and variables related to successful outcomes, as result of wide case-study experiences of institutional performance at the river basin level in different countries. 5

KEY-PRINCIPLES: DECENTRALIZATION AND PARTICIPATION Under the perspective of institutional analysis, river basin management is a process of decentralization in decision-making, to be implemented with an adequate involvement of the stakeholders in the decision-making process itself. Thus, the two fundamental elements are: i)

The achievement of decentralization at the maximum “appropriate” extent 6

ii)

The activation and strengthening of participation among all the different stakeholders, as main component of the same decentralization process

As regards the first element, decentralization means organizing or re-organizing the institutional arrangements towards managing water resources on a basin scale, a process that generally implies a transition of powers and functions, from the central to the local level. But it does not univocally correspond to devolution. It has to be noticed that, in implementing a decentralization process in a specific setting, several functions can be opportunely provided by the central authorities and not devolved to the basin level. 7 On this purpose, the literature refers to the aspects of water management characterized by a “public good” nature (e.g. weather monitoring and forecasting, but even – to some extent – hydrological / environmental research, or flood control). As regards the second element, participation can take various and different forms: transparency and accountability (participation as openness of decision-making to the public), consultation of stakeholders, negotiation (active involvement of stakeholders in decision-making), full transition in powers and functions from central administrative authorities to stakeholders. 8 An important role is played by the relationships and the different situations among the stakeholders, given that strong

3

This topic is analyzed by Alaerts, with regard to river basin organizations. See Alaerts, Le Moigne (forthcoming). About the transaction costs approach and his relevance for the institutional analysis of water resources management, see Saleth, Dinar (2004). 5 See the findings of the World Bank-supported study “Integrated River Basin Management and the Principle of Managing Water Resources at the Lowest Appropriate Level – When and Why Does It (Not) Work in Practice?”as reported in Blomquist, Dinar, Kemper (2005), Blomquist, Ballestero, Bhat, Kemper (2005), Bhat, Ramu, Kemper (2005). 6 The maximum level of appropriate decentralization is defined as the need for decisions to be taken “at the lowest appropriate level” (ICWE,1992). 7 As observed in Mody (2004) “..the case for decentralization as against central control is not unambiguous..” given that the devolution of functions can accompany benefits with counter-effects. 8 See on this topic Massarutto (2005). 4

222



asymmetries in resources endowment (e.g. financial or political asymmetry) may be associated with an heterogeneous distribution of incentives to cooperative action. As can be clearly noticed, the two elements of decentralization and participation are deeply interrelated. The following paragraphs will detail single aspects connected with these two fundamental concepts, in order to provide an analytical framework to describe the institutional performance of water resources management processes at the local/basin scale.

ANALYTICAL FRAMEWORK FOR INSTITUTIONAL ARRANGEMENTS In order to provide an analytical framework for assessing water management institutional 9 arrangements in a local setting, four main categories of factors and variables will be considered. The framework will be completed with a schematic table and a set of specific questions, intended as a tool for documental research and in-field survey of a case-study.

Contextual factors and initial conditions The success of a decentralization process is influenced by various pre-existent factors, configuring a smaller or greater aptitude, in each specific context, to improve the management of water resources at the river basin scale. As regards the early stages of the decentralization process, the first contextual factor to be 10 observed is the level of economic development at both the central and at the local/basin level. The level of economic development is connected with the potential activation of financial commitment to the decentralization process from the central government and the local stakeholders: sustainable outcomes are linked to the financial viability of the institutional arrangements for water management, and this target will be easier to achieve where economic well-being, at both the central and at the local level, allows the bearing of transition and ongoing costs of the process. The central financial commitment to the decentralization process is an important starting factor, while it is possible to observe that it is not a necessary factor (in theory, a decentralization process can be initiated even with the sole financial commitment of local stakeholders). The central authority can activate funding for initial implementation, in the forms of financing the devolution and/or financing the maintenance of a set of water management-related functions considered to be better managed centrally rather than locally. The local financial commitment to the decentralization process is a significant factor, as connected with the financial autonomy at the local level, one of the main components of successful implementation of a decentralization process. Another initial factor influencing the development and implementation of a decentralization process is the distribution of resources between basin stakeholders. When resources are asymmetrically distributed (in terms of financial power, rights over the water resources, or also political influence over water allocation) it is possible that a cooperative arrangement will be less attractive for the better11 situated subjects than a non-cooperative option. This element acts in a complex way, because the most endowed stakeholders may assume, if attracted by the future benefits deriving from the basin management option, a leadership role and strengthen the process itself rather than making it more difficult. This leadership role can take the form of a strong financial commitment for decentralization 9

According to the theoretical framework presented in Blomquist, Dinar, Kemper (2005). The central level in a decentralization process can correspond, depending on the context, to a nation, a state (in federal countries), a region or a province. 11 When a cooperative arrangement generates significant gains for the group as a whole but results (or is perceived as) less favourable of a non-cooperative scheme for some actors, a redistribution or compensation in benefits can occur. The problem, with regard to the management of international river basins, has been analyzed in Sadoff, Grey (2002). 10

223



by the main stakeholders. Thus, if extreme inequality may be detrimental to the decentralization effort, the assumption of a leadership role by the better-situated stakeholders can foster the process itself. The described factor, to be taken into account, behaves in a multi-directional way. Also the social and cultural distinctions at the local/basin level are a significant contextual factor, because they can affect communication between stakeholders, as well as trust and aptitude to cooperation. All other factors being equal, it is expected that the greater and more contentious are these distinctions, the more difficult it will be to develop and maintain institutional arrangements for the management of water resources on a basin scale. Furthermore, a contextual factor to be considered is the existence of previous experiences of governance at the local level. It is expected that water management decentralization initiatives will be more likely to achieve successful results in settings where there is a local experience in governing and managing other resources and services in a cooperative way. While the main challenge is, in this regard, to strike a balance between the central role and the local role in organizing water management at the “lowest appropriate level”, this ability will also depend on the skills previously developed in other areas of social life.

Characteristics of the decentralization process A decentralization process implies the devolution of authority and responsibility from the centre, and the acceptance of authority and responsibility at the local level. The occurrence of both depends upon the way in which the decentralization takes form, and this form may affect the achievement of successful results. A first characteristic to be considered is the nature of the decentralization initiative. If, in theory, a decentralization process may start from an exclusive top-down initiative (by the central government) or, at the other extreme, bottom-up initiative (by local stakeholders), it is expected that most of the actual settings lie in-between these two extreme examples. A decentralization initiative will be more likely to achieve successful results where devolution is a mutually desired process, shared by basin stakeholders and central government officials. Furthermore, one of the main targets of a decentralization process is to obtain a deep involvement of the stakeholders into the making of decisions. For this reason, incorporating existing local institutions and practices is another important characteristic of the decentralization initiative. Where the traditional institutions are involved, they play a participating and legitimating role for basin management towards the stakeholders. Moreover, it is to be expected that the transaction costs (in terms of time and effort) to basin stakeholders will be smaller in existing organizational forms than in an additional set of organizational arrangements. All other things being equal, decentralization initiatives are more likely to succeed where they involve existing governance institutions and practices. Another significant characteristic of the decentralization initiative is the continuity in central level commitment for the decentralization policy. Usually, a decentralization initiative includes a transition of authority from the central government to the local level.12 In these situations, an important element is how the decentralization policy can survive any changes of power that may occur at the central level during the process. Thus, all other things being equal, when there is a lack of continuity in the central level commitment to the decentralization policy, it will be harder to achieve a successful implementation of the process.

Central government and basin-level relationships and capacities Coordination of central and local actions is an essential element of a successful decentralization process. The respective capacities of central government and basin stakeholders, and the relationships between them, are key to achieve this target. 12

Other cases, expected to be rare, are when the decentralization doesn’t include any kind of transition in powers by the central authority to the local level. In these cases, the mentioned factor is not important.

224



In this direction, a significant factor to be considered is the extent of actual devolution from the central to the local level. Indeed, a central government, while formally pursuing the implementation of a decentralization policy, may act substantially only in a symbolic or even in an abandoning way for a real devolution of authority and responsibility at the local/basin level. 13 This behaviour may also undermine stakeholders’ commitment to the decentralization process. It is reasonable to observe that the degree of actual devolution in resource management responsibilities to the local level is associated with more or less successful results for the decentralization process. The achievement of financial autonomy at the local/basin level is another factor to be considered for a successful decentralization process. Financial autonomy will be better achieved when there is a balance between the central and the local authorities in financing the process, and in managing the financial resources. On one side, a form of financial autonomy is needed at the basin level but, on the other, a complete transfer of financial responsibilities from the central to the local level may be dangerous for the process itself. All other things being equal, favourable prospects of success will occur where there is a balance in funding and control between the central government and the basin level. A third significant characteristic is the basin-level authority to create and modify institutional arrangements. A decentralization process is highly context-specific, and the functions of governing, financing, and monitoring water resources, as well as coordinating the infrastructure construction and maintenance, have to be tailored to the specific settings of the basin area. Moreover, sustainability in efficiently managing these functions in the long run necessitates the power to modify the institutional arrangements in response to changed conditions. These activities will be better performed by the local authorities, for two principal reasons: one is the high requirement of information needed; another is the potential of this form of local autonomy to attract stakeholders and foster their involvement into the process. For these reasons, it is expected that successful and sustainable implementation of a decentralization process could occur where stakeholders are empowered to create and modify the institutional arrangements. Another important element related to what mentioned is the power of local authority to set and modify any form of cross-jurisdictional arrangement useful to efficiently implement the process: the relevance of this factor is high in the case of water management, given that in many cases the administrative boundaries don’t match the basin or sub-basin boundaries. With regard to central/local relationships, the distribution of central-level political influence among local stakeholders is another significant factor of a decentralization process. In a specific context, it is possible that the better-situated stakeholders have a stronger access than others to central government influence: the exercise of this influence, consisting in a block or overturning of disagreed local level decisions, can erode the stakeholders’ collective commitment to the decentralization process. All other things being equal, a more successful implementation of decentralized management will occur in settings where there is a relative symmetrical political influence of the stakeholders upon the central government. Another important factor to be observed is the characteristics of the water rights systems (formal or informal rights, recognized as binding among stakeholders). The water rights can be defined at the local level, but it is more likely that at least in some aspects these rights are defined at the central level (as national, state or provincial rules). The nature of these systems of rights, by which the central and the local level relate, can change the commitment of the local stakeholders to the agreements needed by the collective action. Furthermore, under a transaction costs profile, the adequate time for implementation and adaptation to new institutional arrangements is a significant aspect of a decentralization process related to central and local institutional capacities. Longevity, as well as adaptability to change – in a trial-and-error learning process – are both important factors for the success of a basin management process, although is difficult to generally establish after how much time a given arrangement can be considered apt to achieve its expected results or to be substituted by another. If this aspect can be

13

This is distinct from the above-mentioned factor of continuity in the central action; in this case the problem lies in the gap between formal pursuance and actual implementation of a decentralization policy by the central authority.

225



opportunely analyzed only inside a specific case, it can be observed that time is an influencing factor for the successful implementation of a decentralization process.

Internal configuration of basin-level institutional arrangements The possibility to achieve a successful implementation of decentralized water resources management will depend on the characteristics of the institutional arrangements configured at the basin level. Among these characteristics, a necessary component is the presence of institutional arrangements for the basin-level governance, by which stakeholders articulate interests, share information, communicate and take collective decisions. Nevertheless, it should remain clear that basin-level governance, or the presence of institutional arrangements to enable stakeholders’ actions at multiple levels, doesn’t imply the creation or strengthening of an ad hoc river basin organization. Another significant characteristic of institutional arrangements at the local level is the clarity of institutional boundaries and their matching with the basin boundaries,14 given that decentralized water management at the basin scale is a process of collective decision-making. Unclearly defined or mismatched boundaries create a lack of efficiency and effectiveness of collective decisions (e.g. inadequate information, mismatch between decisions and users involved or excluded). All other things being equal, it is reasonable to expect that successful implementation of decentralized water management will take place where basin-level institutions have clearly defined boundaries and where these boundaries are well-matched to the basin boundaries. Furthermore, an important characteristic of basin-level arrangements is the recognition of subbasin communities of interest. With regard to this characteristic, it can be observed that the basin system naturally configures an inter-relation of interests among users or groups, but water users and groups have different interests: interests are likely to be different among users/groups in the various sectors of activity (agriculture, industry, hydropower generation) or among users/groups differently situated in the basin area (downstream/upstream users). Recognition of communities of interest can include only representation (guaranteed participation to decisions) or even assurance that decisions are the results of agreements reached between the different communities of interest. The recognition of sub-basin communities of interest is not a costless practice: transaction costs are expected to increase for the recognition of each sub-basin community, to the extent that – beyond a given threshold – additional recognitions may become counter-productive. Nevertheless these recognitions, supporting trust and reciprocity between stakeholders, are an important factor to the emergence and sustainability of basin-level arrangements. Among the characteristics of basin-level institutional arrangements, a significant role is represented by the availability of fora where stakeholders can communicate and resolve conflicts.15 These instruments will function as a means to strengthen both cooperation and participation between the different actors. Information sharing and communication between stakeholders are important elements of water resources management, because they reduce information asymmetries and differences of interpretation, thus fostering cooperation between stakeholders. For this reason, the presence of regular fora for information sharing and communication is expected to be, all other things being equal, 14

Matching the administrative (political) boundaries and the basin (natural) boundaries, is one of the most challenging issues in water resources management. In many cases, a solution has been found in creating river basin organizations. Nevertheless, the necessary sustainability of such institutions in the long period suggests the opportunity to consider into single cases (especially in less developed settings), the relation between the benefits to be obtained from a new ad hoc institutional organization, and the costs – included transaction costs – of creating and managing it. 15 If Alaerts indicates how the forum is one of the most specific features of river basin organizations and, in some cases, the essence of such organizations – in Alaerts, Le Moigne (forthcoming) – occasions for ensuring participation of stakeholders at the basin level can be organized even in absence of an ad hoc river basin organization.

226



a factor contributing to the successful implementation of decentralized water management at the basin level. Moreover, the sustainability of decentralized water management depends on the presence of fora for conflict resolution, since disagreements between stakeholders might arise in any conceivable water resources management setting. Thus, all other things being equal, decentralized water management is more likely to achieve sustainable results where there are fora for conflict resolution. TABLE OF RECAPITULATION: CATEGORIES, FACTORS AND RELATED QUESTIONS The following table represents schematically the described categories and factors as a set of elements useful to evaluate the institutional performance of the decentralization process of water resources management at the basin level. Related to the single factors, a number of questions are purposed as a case-study tool for documental research and in-field survey. Table 1. Main categories, single factors and related questions useful to evaluate the institutional performance of the decentralization process of water resources management at the basin level Single factors Related questions Main categories Contextual factors Economic Does the national/state/regional/provincial level of and initial conditions development at the economic development allow a financial commitment to central level the basin management process from the central authorities? Economic Does the level of economic development at the basin level development at the allow a financial commitment to the basin management basin level process from the basin stakeholders? Distribution of Are there consistent asymmetries in financial (or other resources between kind of) resources endowment among local stakeholders? local stakeholders If yes, are these asymmetries expected to weaken (or eventually to foster) the decentralization commitment? Socio-cultural Is the basin area shared by different background cultural/ethnical/religious groups? If yes, are the relations between these groups expected to allow cooperation or are they expected to raise conflictuality in a local water governance process? Previous Are there previous successful experiences of institutional experiences of arrangements for governance at the local/basin level? local governance If yes, have these experiences developed local capacities expected to be useful for water resources management? Characteristics of Top-down / Bottom Does the decentralization process start from a top-down the decentralization –up / Mutually central government officials initiative or from a bottom-up process desired devolution local stakeholders initiative (or in-between the two cases)? If from a top-down initiative, is this initiative likely to activate an adequate stakeholders involvement at the local level? Incorporation or involvement of existing local governance arrangements Consistent central government policy commitment

Does the decentralization initiative adequately incorporate/involve pre-existing local institutions in the process? If yes, is this involvement expected to enhance participation of local stakeholders into the process? Does the decentralization initiative include a transition of authority from the central to the local/basin level? If yes, is (or is expected to be) the decentralization policy commitment at the central level continuous or discontinuous with regard to changes of the political situation ?

227


Central government and basin-level relationships and capacities

Extent of actual devolution

Financial resources and autonomy at the basin level Basin-level authority to create and modify institutional arrangements Distribution of central level political influence among stakeholders Characteristics of the water rights system

Internal configuration of basin-level institutional arrangements

Adequate time for implementation and adaptation Presence of basinlevel governance institutions Clarity of institutional boundaries and match to basin boundaries Recognition of subbasin communities of interest Availability of fora for information sharing and communication Availability of for a for conflict resolution


Is there an adequate degree of actual devolution of responsibilities from the central level to the local/basin level, or is the central government decentralization role only formal? Is there a balance in funding/managing funds between the central and the local/basin level? Is the local level empowered to create institutional arrangements / to modify them in response to changed conditions (especially in regard to cross-jurisdictional arrangements at the basin or sub-basin level)? Is there a relative asymmetrical access to central level political influence among basin stakeholders? If yes, are these asymmetries expected to configure blocks or overturns of local level decisions? Are there (formal or informal) local systems of water rights and rules defined at the central level? If yes, are these rights and rules perceived as certain and clear by local stakeholders (otherwise there can be a lack of stakeholders commitment to the decentralization process)? Is there an adequate longevity and/or adaptability to change of the institutional arrangements? Are there adequate institutional arrangements to enable stakeholders collective decisions for water management at the basin level? Do the water management institutional arrangements act into clearly defined boundaries? Do these boundaries match the basin boundaries?

Is there an adequate recognition of different communities of interest in taking decisions at the local level? Are there regular fora for information sharing and communication between stakeholders at the basin level?

Are there regular fora for conflict resolution at the basin level?

REFERENCES Alaerts, G., and G. Le Moigne (Editors), Forthcoming, “Integrated Water Management at River Basin Level: An Institutional Development Focus on River Basin Organizations”. G.Alaerts, “Chapter 18 Institutions for River Basin Management: A Synthesis of Lessons in Developing Cooperative Arrangements”

228



Billi, A., Y. Meroz, and A. Quarto, 2004, “Water Sector and Institutional Reform”, 3rd WASAMED Workshop on Non- conventional Water Use, Cairo, Egypt, 7 – 11 December 2004 Bhat, A., K. Ramu, and K. Kemper, 2005, “Institutional and Policy Analysis of River Basin Management. The Brantas River Basin, East Java, Indonesia”, World Bank Policy Research Working Paper 3611, May 2005 Blomquist, W., A. Dinar., and K. Kemper, 2005, “Comparison of Institutional Arrangements for River Basin Management in Eight Basins”, World Bank Policy Research Working Paper 3636, June 2005 Blomquist, W., M. Ballestero, A. Bhat, and K. Kemper, 2005, “Institutional and Policy Analysis of River Basin Management. The Tárcoles River Basin, Costa Rica”, World Bank Policy Research Working Paper 3612, May 2005 GWP, 2000, “Integrated Water Resources Management”, Global Water Partnership Technical Committee, TEC Background Paper 4, Global Water Partnership International Conference on Water and the Environment (ICWE),1992, “The Dublin Statement on Water and Sustainable Development”, International Conference on Water and the Environment, 26-31 January 1992 Massarutto, A., 2005, “Partecipazione del Pubblico e Pianificazione nel Settore Idrico”, presented at the Conference “La partecipazione pubblica nell’attuazione della Direttiva Quadro europea sulle acque”, Università Bocconi, 30 June 2005, Milano Mody, J., 2004, “Achieving Accountability through Decentralization: Lessons for Integrated River Basin Management”, World Bank Policy Research Working Paper 3346, June 2004 Sadoff, C.W., and D. Grey, 2002, “Beyond the River: The Benefits of Cooperation on International Rivers”, Water Policy n.4-2002, pp. 389-403, Elsevier Saleth, R.M., and A. Dinar, 2004, “The Institutional Economics of Water. A cross-country Analysis of Institutions and Performance”, Edward Elgar and The World Bank

229


230




THE ECONOMICS OF WATER EFFICIENCY: A REVIEW OF THEORIES, MEASUREMENT ISSUES AND INTEGRATED MODELS

A. Billi*, G. Canitano**, A. Quarto* * University of Rome “La Sapienza”, Italy ** University of Florence, Italy

SUMMARY - The present paper focuses on the economics of water efficiency, in its theoretical foundations, statistical developments, and interactions with hydrology. After defining the various concepts of water use efficiency and the main techniques for its valuation, the paper presents the main indicators that can be used for measuring economic efficiency in water use and the basic characteristics of an integrated hydrologic-economic water accounting system. Finally, the main issues at stake in integrating different perspectives into a comprehensive model are presented, in order to give suggestions on how to design appropriate policies for water resources planning and management. Key words: water economics, river basin modelling, water efficiency.

INTRODUCTION (Section 1) The pressures of societies and the economy, combined with traditional approaches to water supply and management, have led to the unsustainable use of world’s freshwater resources. Indeed, in some areas of the Mediterranean basin, it is essential to implement appropriate water-savings policies, in order to avoid shortages, ecological degradation, and even permanent economic and social consequences. With the growth in population and the demand for economic development, increasing the efficiency of water use is certain to become more and more important. Improving efficiency and increasing conservation are the cheapest, easiest, and least destructive ways to meet future water needs. They are also the most politically- and environmentally-responsible measures to be implemented in the sector. However, there are dissimilar perceptions on what the determinants of water efficiency are in the end, and what policies should be implemented to pursue the goal of efficient and wise water management. Two main approaches, the hydrological/engineering approach and the economic/institutional approach, have usually confronted on what methodologies and performance ratings to use to measure water efficiency. The former approach focuses on the abstraction, storage, distribution, treatment and disposal activities related to the hydrological cycle and its variability. These models give rise to the recommendations of implementing supply-side measures, such as infrastructure expansion and investment in reduction of leakages. On the other hand, the economics of water deals with the ways to improve social gains form the use of a scarce resource. It uses optimization techniques under alternative institutional policies, in order to maximize the benefits of an allocation of an exogenous amounts of water in the economy. This approach follows from the joint analysis of production and environmental costs, and of demand

231



conditions, and is at the basis of the introduction of demand-management policies, such as costrecovery, environmental taxes, water use permits tradable on special markets. In the last decade or so, a trend has emerged toward integrating economic and institutional considerations into complex hydrological modeling, and to represent hydrological relationships to determine the available amount of water into economic models. Such analyses are based on the recognition that there complex interactions exist within a territory of reference, between the user’s system, represented by the economy, and the water resource system, represented by the hydrological cycle. The ongoing integration of different perspectives into a comprehensive model is a fundamental task in designing appropriate policies for water resources planning and management. The present paper focuses on the economics of water use efficiency, in its theoretical foundations, statistical developments, and contacts and interactions with the hydrologic science. Water use efficiency includes any measure that reduces the amount of water used per unit of any given activity, consistent with the maintenance or enhancement of water quality. Anyhow, it is when water prices reflect the full social costs of developing supplies, that the incentives are created to use the resource efficiently and rationally. Hence, when resources are correctly valued, reflecting its contribution to production, the incentive exists, through the forces of supply and demand, to use those resources efficiently though the introduction of technological change. The achievement of economic efficiency in resource use is a major economic policy aim, for it means that the economy is approaching its maximum in the context of available resources. The economics of water resource addresses these issue, both on a sector basis through stand-alone analyses, and in a comprehensive manner, through multi-objective approaches. The purpose here is illustrative, not comprehensive or definitive. Nonetheless, these aspects are considered of special interest to the WASAMED project, since they give suggestions on how to integrate the different approaches in water resources modelling, planning and management in the Mediterranean. From table 1, it is evident that total renewable waster resources pre capita vary widely between participating countries. Moreover, water use efficiency is especially important in this region, since many of the Mediterranean countries suffer high levels of water stress, as shown by the last two indicators in table 1, displaying the pressure on water resources from agriculture. In some instances, the dependency ratio, is equal to the part of the renewable water resources which originates outside the country, is worryingly high. The rest of the paper is organized as follows. Section 2 introduces the different meanings of efficiency in water resources use, focusing on economics but stressing the ways the various approaches can complement each other. It also introduces the basic concepts of water economics, especially in the agriculture sector. Section 3 briefly analyses the basic principles of water supply and demand, carrying out examples of stand-alone economic analyses of water resources. Section 4 examines the techniques used to assess the value of water use, focusing especially on the agricultural production functions that are used in complex modeling. Section 5 reviews water accounting principles and the techniques for modeling the interactions between the hydrologic and the economic conditions and the institutional framework. It also examines the main indicators developed to measure water efficiency and productivity in related sectors, and how they can be used for policy-making. Section 5 draws some policy conclusions and proposes directions for future empirical research.

A MULTI-FACED APPROACH TO WATER EFFICIENCY (Section 2) It is commonly intended that achieving water efficiency consists of optimizing water use. Indeed, different points of view should be considered when investigating water use efficiency. Absolute or physical efficiency means using the least possible amount of water for any activities. Economic efficiency seeks to derive the maximum economic benefit for the society. Institutional efficiency qualifies the functions of an institution regarding its water-related tasks. Social efficiency strives to fulfill the needs of the user community. Environmental efficiency looks at natural resource conservation. Finally, technological efficiency refers to the process of finding ways for extracting more valuable products from the same resources. Depending on the conditions of each users system, these non-exclusive definitions of water use efficiency can be achieved simultaneously. In any case, it is clear that efficient water use should be approached in a multi-objective, cross-sectional and

232



comprehensive manner. In particular, it should include the management of both supply and demand, assigning an economic value to water resources (Garduño and Cortés, 1994).

233



Table 1. Basic water data WASAMED countries (Source: FAO AQUASTAT, 2003, unless specified) Groundwater: Surface water: produced produced internally (10^9 internally (10^9 m3/yr) m3/yr)

Overlap: surface and groundwater (10^9 m3/yr)

Water resources: total internal renewable (10^9 m3/yr)

Water resources: total internal per capita (m3/inhab/yr)

Water Water resources: resources: total renewable total renewable per capita (actual) (10^9 (actual) 3 m /yr) (m3/inhab/yr)

Dependency ratio (%)

Total water Ag water withdrawal withdrawal (1998, as % of (1998, as % of total renewable total renewable water water resources) resources)

Algeria

1.70

13.20

1.00

13.90

429.80

14.32

442.8

2.93

27.51

42.39

Cyprus

0.41

0.56

0.19

0.78

965.30

0.78

965.3

0.00

21.79

30.77

Egypt

1.30

0.50

0.00

1.80

24.53

58.30

794.4

96.91

101.20

117.20

Germany

45.70

106.30

45.00

107.00

1,297.00

154.00

1,866.0

30.52

6.05

30.55

Greece

10.30

55.50

7.80

58.00

5,284.00

74.25

6,764.0

21.89

8.42

10.46

Italy

43.00

170.50

31.00

182.50

3,182.00

191.30

3,336.0

4.60

10.46

23.19

Jordan

0.50

0.40

0.22

0.68

121.10

0.88

156.8

22.73

86.36

114.80

Lebanon

3.20

4.10

2.50

4.80

1,294.00

4.40

1,189.0

0.76

20.88

31.31

Malta

0.05

0.00

0.00

0.05

127.50

0.05

127.5

0.00

19.80

100.00

Morocco

10.00

22.00

3.00

29.00

933.60

29.00

933.6

0.00

37.97

43.45

Palestine

-

-

-

-

-

-

-

-

-

-

Portugal

4.00

38.00

4.00

38.00

3,773.00

68.70

6,821.0

44.69

12.82

16.39

Spain

29.90

109.50

28.20

111.20

2,704.00

111.50

2,711.0

0.27

21.74

31.96

Syria

4.20

4.80

2.00

7.00

384.10

26.26

1,441.0

80.26

72.09

75.97

Tunisia

1.49

3.10

0.40

4.19

422.20

4.595.00

462.4

8.70

47.12

57.45

Turkey

69.00

186.00

28.00

227.00

3,139.00

213.60

2,953.0

1.52

13.05

17.57

Mean

14.98

47.63

10.22

52.39

1,605.00

63.46

2,064.0

21.05

33.82

49.56

Std. Dev

21.52

64.45

14.87

72.41

1,614.00

72.38

2,162.0

30.73

29.87

35.78

234



Such approach is required because both population and water are unevenly distributed, such that different areas experience differing degrees of water stress. Moreover, keeping it constant in terms of quantity, absolute water supply is diminishing in terms of quality. Hence, water allocation is a highly controversial subject, particularly in arid and semiarid zones. Most water planners assign priority to water use in the following hierarchical order: human consumption, food production and industrial production. However, this criterion has often caused conflicts because in many countries the priority of development strategies is not necessarily to improve the quality of life through sanitation and good health, but rather through the development of industry and exports (food and finished products). Above all, people’s fundamental right to clean water and sanitation at and affordable price should be recognized. But in the past, the prevailing lack of awareness about the economic value of water has led to its being wasted and to its use with a negative impact on the environment. Moreover, indirect benefits such as environmental and psychological ones have not been included, as they are difficult to quantify. In practice, the value water is considered inferior to its real worth and this gives rise to inefficient use. For much of the time, water management has focused on manipulating water supplies from natural source to where it was needed. In this supply management, water is thought of as a requirement to be met, not as a commodity the demand for which can be altered. Thus, water use efficiency has often meant satisfying all possible demands for the resource. It is only comparatively recently that the focus has shifted on how demand can be satisfied without massive supply developments. In particular, the concept of allocation of water based on its values in alternative uses has become increasingly important, and, with it, consideration of water use efficiency. Water use in most socioeconomic activities can vary widely depending upon the interplay of many factors. Many of these factors, such as pricing policies, comprise products of public decision-making. Others, such as the selection of production processes, are private decisions, but again are the product of many forces that change through time. Thus, policies and practices which lead to improved water use efficiency bring about an array of possible choices for adapting to local circumstances. The establishment of a dynamic balance between interventions in water supply and demand, taking into account the variability of supply in time and space, the changes in demand, and the limits and opportunities of technology, is the final goal of the society. This will enable to dispose of water supply in adequate amounts for human groups with positive growth rates (increasing population) and provide responses that are better suited to availability and demands.

Economic and institutional issues in water use efficiency Any method of increasing water use efficiency should be subjected to a technical evaluation in order to obtain an estimate of the actual reduction in water demand or discharge resulting from using the method. However, economic and environmental factors are of foremost importance. Where water development costs and competition for available capital are rising, the concept of physical or engineering efficiency is limited by its inability to address the value of any specific use of water in 16 relation to alternative uses for the same water. An exclusive emphasis on improving the engineering efficiency of a given water use, therefore, may lead to unproductive expenditures if the value of that use is less than the value of some other use of the same water. Hence, the economic efficiency concept should be taken into account. While the hydrological cycle is at the basis of physical calculations of water efficiency, the concept of the factors of production is the starting point for analyzing economic efficiency. Three generalized factors underlie all productive activities: natural resources, labour and capital. In particular, natural resources are added to varying amounts of labour and capital, to bring about production of goods and services.

16

Physical efficiency means using the least possible amount of water for any activities. Such technical evaluations basically measure the ratio between water pumped into a system and water delivered to consumers or end uses. More on physical water efficiency is presented in section 5 on water accounting and related indicators.

235



This general paradigm applies not only to the goods and services traded in markets (or sometimes by other means), but also to some which may be quite far removed from market processes, like recreation. In any event, these three factors of production combine in varied combinations to yield products for consumption. Economic efficiency in a production setting involves technical and allocative components. Production is technically efficient when the maximum possible output is generated with a given set of inputs, or when a selected output is produced at minimum cost. Allocative efficiency involves the optimal combination of inputs and occurs when the marginal prices of each of the factor inputs are equal. In turn, the way in which the factors combine depends upon their relative prices. 17 The economic approach to deciding the most desirable allocation of water is to use the principles of economic efficiency in order to ensure that water is supplied to its most valuable uses. In an economic sense, two principles or rules are the criteria which guarantee the greatest efficiency in the allocation of a resource. These are (Agudelo, 2001): The principle of ‘equimarginal value’, which means that the marginal benefit, or incremental value, per unit of resource used should be equal across all uses. When equality of marginal values is achieved, further redistribution of water can make no sector better off without making another sector worse off. The principle, then, is that the resource should be allocated in such a way that all users or consumers derive equal value in use from the marginal (the last) unit used or consumed. It should be noted that this principle presupposes an homogeneous good, which is not really the case with water; surface water, for instance, cannot be interchanged with ground water. The principle of ‘marginal cost pricing’, which means that the marginal benefit of use of the resource should be equal to the marginal cost of its supply. Whether an enterprise is private and unregulated, private and regulated, or public, the condition that the price set should be equal to marginal cost is the desired situation from the point of view of economic efficiency considerations (provided the principle of equimarginal value is also met). Where free competition in the economic sense exists, market processes tend to automatically bring about this optimum. If government policies dominate price-quantity determination because of public ownership or regulation, political processes replace market processes. When water prices are low relative to the costs of other inputs and in relation to the costs of developing supplies, the 18 resource will be overused and efficiency of use will be correspondingly low. Three general considerations emerge. First, the level of attention paid to water use efficiency is directly proportional to the prices charged for water servicing. Second, rising prices lead to increasing attention to water use characteristics, and, over the long run, to more efficient water use, improved productivity and reallocation among users. Finally, when water prices reflect the full social costs of developing supplies, incentives are created to use the resource efficiently and rationally, reflecting its value in production or in its various other uses. In other words, rising prices generate powerful incentives for increasing water use efficiency. This last point leads directly to the issue of water institutions. The legal systems of societies are 19 endlessly complex, and beyond the scope of this paper. Nonetheless, a few characteristics can be pointed out which clearly affect water efficiency decisions. First of all, most nations employ systems of building codes, which specify minimum standards that must be met in new or renovation construction. Until recently, the matter of water efficiency has rarely formed part of these codes. However, until 17

Firms or consumers normally will tend to use relatively more of the cheaper inputs, and relatively fewer of the more costly ones. If any of the required inputs has a very low, or zero price, to the user, then as much as possible of that input will be used. Here lies one of the fundamental problems of water resources management. As outlined earlier, water supplies have been cheap historically in most areas of the world, even in semi-arid areas. This basic factor plays a major role in explaining why water usage per unit of production is high, why recycling rarely reaches its full potential and why water usage per capita is higher in some countries than in others. 18 Basic pricing considerations also play a major role in explaining why pollution occurs. Waste removal, in the majority of cases requires the use of environmental resources, such as water. When this input is available free of charge, it is invariably cheaper than any other option for waste disposal. The resulting overuse leads directly to water pollution problems. 19 For an analysis of the institutional implications of water management, see Billi, Meroz, and Quarto (2004).

236



codes and standards are modified, improved water efficiency will be very difficult to achieve. Similarly, the ability to charge self-supplied water users royalties for the use of water constitutes formal legal arrangements, which can be manipulated to induce adequate incentives. Above all, the fundamental institutional issue underlying water use efficiency is that of the regime of property rights. The rights to natural resources of any kind display varying degrees of ownership on the spectrum from public to private. At the public end of the scale, access is completely open to all citizens. The resource is essentially free for the taking. With open access, no incentive exists to manage the resource in a conserving, efficient manner, except through moral suasion. At the other end of the spectrum, where private ownership pertains, access to the resource belongs exclusively to its owner, is enforceable under law and is both divisible and transferable. Under such conditions, 20 positive incentives do exist for effective management and efficient use. The point is that water typifies common property resources, with non-exclusivity, non-enforceability and low prices. Under these conditions, little incentive exists for conserving, efficient resource use. Indeed, in many cases, the potential for overuse and abuse is strong, and management becomes a very complex and difficult undertaking. But the theory goes further by suggesting that externalities, under such conditions, will rise to socially unacceptable levels, and that, over time, the development of private or quasi-private arrangements of rights will develop. Currently, in some parts of the world, the development of water markets for re-allocating water supplies, and the fledgling use of effluent discharge fees and tradable permits for pollution control reflect the growing reformation of property rights to water. Under such conditions, the development of increasingly efficient water use practices is an accompanying trend. The principle emerging here is that water use efficiency is partially a response to the property rights prevailing in a society. The greater the degree of private ownership, the greater the use of water efficient practices. The aforementioned three key concept of water use efficiency (physical, economic and institutional) can be seen as related to each other in a sequential way. Figure 1 exemplifies this relationship and gives evidence of the fact that the story of changing social uses of water forms a spiral movement, oscillating between a perceived scarcity of the natural resource, and the implementation of the means required to overcome such scarcity. While at the beginning, engineering solutions were put in place to overcome water scarcity, and technical productive efficiency was sought to ameliorate the use of water, nowadays we have moved to demand management, in term of allocative and institutional efficiency. This process has been called by Ohlsson and Turton (2000) “the turning of a screw” and is displayed in Figure 1. The movement to economic and institutional considerations has been key to facing properly water problems, since besides physical scarcity, it is necessary to overcome the conflicts generated by competing uses, that is the so-called “social water scarcity”.

Other aspects of water use efficiency Social and political realities, technological innovations and environmental constraints in different regions or nations of the world also play important roles in the use of water, and therefore, in efficiency considerations. In a sense, the economic factors singled out above form a subset of these realities. Socio-political factors are embedded into the fabric of societies. Many of them are subtle and indirect in their effects on efficient water use, and the best that can be done is to deal with those factors that seem to have particular importance to the issue at hand.

20

Demsetz (1967), and later Pearse (1988), have illustrated that the progression from common property to private ownership of resources reflects a response to social cost externalities. When resources are plentiful relative to demands, no incentive exists to develop property rights systems, and common property characteristics apply. However, as population growth and economic growth occur, conflicts over access to the resource rise in number and seriousness. There comes a time when the social costs, or externalities, of such conflicts rise to such a degree that it becomes worthwhile to reform the basic property rights (a costly undertaking in itself) to bring about increasing degrees of private ownership.

237


Management phases


Management contents

Management objectives

Demand management 2

User-pay, polluter pay

Demand management 1

More value per unit of water

Economic efficiency

Supply management 2

More produce per unit of water

Productive efficiency

Supply management 1

Less water

Institutional efficiency

Physical efficiency

Fig. 1 The different phases of water management strategies The other caveat is that the social factors are quite complex, and would, if dealt with comprehensively, merit a detailed treatment of their own. It is thought, however, that even a brief treatment will yield a few principles and observations. The discussion that follows deals briefly with: a) effects of social tastes and preferences, b) effects of technological innovation, and c) introduction of environmental considerations. The issue of social tastes and preferences is deeply embedded in societies, and may be a major influence on the ways in which individuals and groups view the need for water use efficiency. For example, water supply abundance creates general attitudes that water is very plentiful, and thus the need to conserve is not felt. This makes efforts at water efficiency more difficult than in less waterabundant areas. A further example relates to a characteristic commonly termed “green lawn syndrome”. This term refers to the attitudes that residential landscaping should be green, with healthy lawns, trees and shrubbery. These attitudes have led in the past to excessive water demands, particularly in drier areas, with subsequent overcapitalization of water infrastructure. In drier areas, the use of water efficient landscaping is gradually being accepted as an alternative to the green lawn syndrome. The point here is that deeply ingrained attitudes, tastes and preferences are important considerations in moving towards increased water use efficiency. Technological change also has a great impact on water use efficiency. On the supply side, the progression of technology has vastly increased the resources available, through the discovery of new reserves and stocks. Supplies have been expanded even more by advances enabling the use of less accessible resources, of lower quality, and lesser concentrations. Even land, though limited in the spatial sense, has been augmented enormously in its capacity to produce crops. On the demand side, technology has progressively reduced and eliminated our dependence on particular resources for particular purposes. Technological innovations have more than offset the depletion of resources through consumption: notwithstanding the economic growth of the past century, the demand for almost all natural resources and for food has risen slower than the supply. For present purposes, it is important to understand the forces driving all the creative technological effort that has overcome the limits of nature’s endowment. Owners of land and natural resources are constantly striving to generate the greatest possible value from them. And those who need these resource commodities are constantly searching for cheaper sources of supply, alternative materials that are less costly, and ways of using them more efficiently. Both suppliers and demanders, driven by the financial incentives created by resource commodity markets, direct their creativity towards overcoming scarcity. The lesson for water use and its efficiency is that, when resources are correctly valued, commensurate with their contribution to productivity, the incentive exists to use those

238



resources efficiently though the introduction of technological change. This principle relates back to earlier points made about the economic forces underlying water use efficiency. Finally, the environmental approach to water use efficiency takes a broader view of the issue, emphasizing the need for integrated approaches to its management. This is where water quality considerations become important, as opposed to concentrating on the quantitative aspects. An environmental view highlights the often-ignored fact that water quantity and quality are tightly interlinked, such that actions that affect one dimension have inevitable effects on the other. 21 This principle, as suggested at the outset of the section, means that water use efficiency should only be considered when they maintain or enhance water quality. The extent to which efficient water use can forestall or even prevent development would be considered under environmental appraisal.

Why an economic perspective for analyzing water use efficiency? While the classical concept of water efficiency are helpful in describing changes in the volume and quality of water available, they are not sufficient for describing all of the economic implications of current or alternative allocation practices. This task requires understanding in economic terms the direct and indirect impacts of water use decisions, including opportunity costs and externalities. The primary objective of the economic analysis of water resources is, thus, demand management and efficient allocation among its various uses. Under growing scarcity, valuing water appropriately and allocating it to the uses in which it has the most value, promotes rational use of scarce resources and greater overall societal net benefit. Allocation mechanisms should resolve trade-offs and balance competing demands, both within and between sectors, as well as between countries and regions. A second domain in which water economics plays a fundamental role is the study of the interdependencies of the water sector with the wider economic and social domain. This interdependence generates externalities, which are uncompensated effects of one agent (or group) to another. The fact that someone could be harmed by another, or get a benefit from him, without a counterpart transaction, create inefficiencies. This calls for a careful analysis of externalities. A third objective of the economic analysis of water is cost recovery. This means pricing water at its full long-run marginal cost, which includes O&M costs, capital costs, opportunity costs, and costs of economic and environmental externalities. Finally, economics is the basis for calculating financing needs. In this sense, economic estimation of financing requirements is a precondition for rehabilitation and improvement of sector performance, and for identifying and mobilizing additional financial resources. 22 The first and second of the aforementioned objectives assumes particular importance. The economics of water has devoted much time to the study of allocative trade-offs and externalities. While traditional measures of water efficiency can even indicate that after a certain threshold there is little scope for saving water, there may still be significant opportunities to increase the net value generated by limited resources. This is the role of the economic analysis of water allocation, which aims at both improving the distribution of water among its uses, and reducing the negative external effects of one use on the others. This helps identifying opportunities and designing policies to improve water management practices.

In what follows, a straightforward microeconomic framework of irrigation activities, composed of a production possibilities frontier, is presented to demonstrate how externalities and opportunity costs

21

The common wisdom is that a reduction in water use without an accompanying decline in waste generation will cause wastes to increase in concentration. The consequences of the latter can vary in their effect on quality. Such increases in waste concentration may overwhelm the ability of existing treatment plants to operate effectively. On the other hand, increased waste stream concentrations might actually enhance the operation of waste treatment systems. The point here is not so much the actual answer to this issue, which, in any event, probably depends on local conditions, but rather the illustration of an additional principle of water use efficiency. 22 The third objective is the specific focus of section 3.1, while the fourth objective is overlooked in the present work.

239



can prevent a region or nation from achieving economic efficiency, even when irrigation is described by high measures of technical water efficiency (Wichelns, 2002a). The key concept of water productivity is to be introduced now. It is the quantity of produce (crops or other goods) that can be obtained per each unit of water used (Molden, 1997). Water productivity can be increased by improvements in agronomic practices, varying crop varieties, and supply and demand management, both regionally and at the farm level. In economic words, increasing water productivity means increasing the technical efficiency of production. Technical and allocative efficiency can be jointly represented in a graphical setting such as that in Figure 2. The continuous convex curve depicted in Figure 2 is the production possibilities frontier, i.e. the locus of technically-efficient combinations of outputs that can be obtained with a given set of resources. Points above the frontier are not feasible, while points below it could be achieved with fewer inputs and are thus inefficient. For example, point E is technically efficient, while point F is not feasible and point I is inefficient. The frontier also describes the technical trade-off that must be considered when choosing crop combinations. The shape of the curve reflects diminishing incremental returns in the production of each crop, that is the opportunity cost of augmenting the production of one crop at the expense of the other. Allocative efficiency describes the maximization of the net benefit, for example the revenue from crop production, by the allocation of a resource. This is achieved in the single point where the ratio of output prices is equal to the rate at which the output of one product (for example, cotton) must be reduced in order to increase the output of another product (for example, rice). In Figure 2, this occurs when the line PP, describing the ratio of output prices of cotton (PC) and rice (PR) is tangent to the production possibilities frontier.23 Farmers interested in maximizing revenues will choose to produce C1 units of cotton and R1 units of rice, and they will respond to any changes in output prices by adjusting crop choices and moving along the frontier. 24

Cotton production, C

P

F

C1

E I

C2

E2

P

Rice production, R

R1

Fig. 2. Technical and allocative efficiency in a production possibilities frontier

23

The slope of PP is – PR / PC. Similarly, a production possibilities frontier can also be used to describe production opportunities throughout a region or nation with multiple uses of water resources. 24

240



The economic value of production at point E is both technically and allocatively efficient. Conversely, an inefficient point such as I might be the result of water rationing, degraded water quality, inability to obtain water when needed, unreliable supply, overexploitation by head-end users. The production possibilities frontier is also useful in analyzing the effects of externalities caused by upstream users to a downstream activity. When a negative externality occurs along the water flow, it affects both crops and the result is the dotted production possibilities frontier depicted in figure 1. This new frontier exemplifies a situation where waterlogging and salinization, caused by upstream irrigation activities, reduce the feasible set of downstream production alternatives. This is reduced by the area between the original and the shifted frontiers. The maximum value of crop that can be produced with a given set of resources is less. For example, it is no more possible to produce the pair (C1, R1); if the quantity of rice is to be the same, water should be diverted and cotton production must be reduced to C2. When it is the cultivation of one crop that generates a negative externality on another, the production possibilities frontier assumes the shape represented in Figure 3. The rotation of the frontier indicates that the maximum achievable production of cotton is reduced at all levels of rice production. The revised feasible set includes the inner curve and the line segment BA. The maximum possible output of cotton, point A, can be achieved only if rice is not produced. Cotton production, C

B E1

C1 A C2 E2

Rice production, R

R1

Fig. 3. Efficiency constraints in a production possibilities frontier The analysis of negative externalities is a particularly important, yet very difficult, aspect of the economic analysis of water. As noted by Perry et al. (1997), while surface runoff and subsurface drain water can be used beneficially by downstream farmers, agricultural return flows usually are lower in quality than the water diverted originally, due to higher concentration of salt, pesticides or nutrients. In addition, return flows may not be available to downstream farmers at the time when the water can be used most productively. The intensity of external effects in water use is perhaps greater than in any other sector of the economy. This is why a careful assessment of water values requires accounting for externalities, as explored in more detail in section 4.

241



MAJOR COMPONENTS OF THE ECONOMIC ANALYSIS OF WATER RESOURCES (Section 3) The supply and demand for water can be traced back to the so-called ‘hydro-social water cycle’ (Merrett, 1997). This relates the natural hydrological cycle to the production and consumption activities related to water. The cycle starts from the natural flow in a catchment, which is variable both during the course of a year, and from a year to another. 25 Then, there are the seven main phases of supply: abstraction, storage, treatment, distribution, wastewater collection, treatment, and disposal. These seven phases are the sources of supply costs and require major construction works. Return flows are then released into the freshwater network, such as surface sources, ground reservoirs, and the sea. Between freshwater distribution and wastewater treatment, consumption takes place in several sectors of the economy. Other sources of supply costs, that require engineering interventions, are internal and external reuse, and recycling. Internal re-use refers to the process of treating internally the water used in the production process, in order to employ it in the same or related processes. External re-use means making wastewater suitable for being utilized in other uses. Recycling includes the activities related to releasing wastewater to the freshwater network, supplementing the natural downstream flows. After this brief overview of the hydro-social water cycle, the remaining of the section is intended to give some other elements of the economics of water resources. The starting points are the basic economic concepts of supply and demand. Supply is mainly analyzed in this section, while section 4 gives special focus to valuation techniques of demand components.

The economic perspective of water supply The supply of water, from an economic perspective, is driven by the costs of constructing and operating the infrastructure, the opportunity cost of these resources in alternative uses, and the correction for external side effects on economic agents beyond the transaction in analysis. Figure 4 is the well-known representation of Rogers et al. (2002) of the total cost to societies of supplying water. Environmental externalities are given evidence as fundamental non-economic supply costs, then they are not given special treatment in this paper.

25 The variability of the natural flow calls for the economic analysis of supply reliability, alternative sources and storage facilities. On these issues the paper will return later.

242



Fig. 4. Components of water supply cost Supply costs are one of the focuses of the economic analysis of water; on these costs will be based the discussion in this sub-section. In the literature they are usually referred to as ‘use costs’ and are distinguished in headworks costs, incurred in abstraction, storage and treatment, and network costs, for distribution, wastewater collection and disposal. Water supply costs can be fixed or variable (Merrett, 1997). When costs are converted into a monetary measure, one can single out those capital expenditures that are incurred for the purchase of resources whose expected life is greater than 12 months. They are called fixed costs and include the main construction works. Conversely, current expenditures, paid for the purchase of resources used up routinely in the production process. These variable costs include those for freshwater itself, materials, chemicals, labor, and so on, required to operate and maintain the system. The total cost function is the sum of the two and is expressed as quantitative relationship describing the cost of supplying output in any time-period at each scale of output, from zero to the system’s theoretical capacity. It is function of the quantity of water supplied to the economic system. In the analysis of the water industry, supply can refer, as appropriate, to any single stage of the hydro-social cycle, or to the system as a whole. Unless otherwise specified, in the following the total cost of water supply will be referred to. Total water costs functions are technical relationships that for economic analyses can be approximated by a quadratic function, usually expressed in the following form:

TC ( Q ) = aQ

2

+ bQ + c

(1)

where TC are total costs, Q is the quantity of water, and a, b and c are the parameters of the relationship, estimated through regression analysis. Average costs are equal to total costs divided by the unit of water produced, that is AC = TC / Q. However, costs that are looked upon by economists are usually expressed in marginal terms, i.e. the resources that have to be employed if capacity needs to be expanded to produce another unit of water. The focus on marginal costs is explained by the fact that it is the appreciation of incremental costs of getting one more unit of water, instead of the absolute cost paid for each unit, that contributes to determining the right incentives to proper use.

MC ( Q ) ≡ ∂TC / ∂Q

and, in the short-term, are strictly positive, Marginal costs are expressed as due to scarcity or capacity constraints. Hence, marginal costs tend to be increasing in the short term, at least after a threshold. Combining the three cost concepts gives a measure of economies of scale in water supply. This is given by the output elasticity of total costs, which is defined as the percentage change in total costs per unit percent change in quantity. In symbols:

243



∂ TC

ε TC , Q =

∂ TC TC

∂Q Q

=

TC

∂Q Q

=

MC ATC

(2)

ε The elasticity TC , Q can be lower or higher than unity, depending on whether there are economies or diseconomies of scale respectively; or it may be equal to unity, if costs are constant all along the relevant values. When average costs are falling, as happens in those phases of the value chain where there are economies of scale (treatment and network operation), marginal costs are less than average costs. For raw water abstraction, the opposite is true, since usually the closest, cheapest sources are those which are used first. The slope of the costs curve of abstraction is, therefore, strictly positive, since marginal costs are greater than average costs. The total cost of supplying water can exhibit several slopes, each depending on the relative strength of the two opposite effects. Determining whether the first case or instead the second prevails is an empirical mater and is essential in characterizing each alternative sources of supply, whenever different degrees of scale economies are displayed. Another key aspect is the distinction between short- and long-run costs. In the former, increased daily output is possible through operational changes or by organizational innovations demanding new procedures. In the long term, additional capital equipment is required, either in new projects or for the expansion of existing infrastructure and plants. Whenever economic analysis is used for determining the societal cost of providing water to the whole system, the long-run perspective is required and, therefore, the long-run marginal cost should be calculated and compared to water demand. This is because, in order to produce even modest levels of output, major works are necessary, as long as the infrastructure is operating close to its maximum capacity. This is called the ‘indivisibility’ character of water provision, and allow for higher elasticity of supply in the long run. This perspective is accordingly used throughout the rest of the paper. Increasing long-run marginal costs give rise to an upward-sloping supply schedule, since higher costs are incurred by producers to expand the quantity of water.

Opportunity costs of water supply alternatives The opportunity cost is defined as the value of a resource in its highest-value alternative use (Briscoe, 1996). In this sense, a water supply project imposes two opportunity costs: that of water resources, which could be used in alternative supply projects, and that of other resources used up in the proposed project, that can be used elsewhere. While the latter is beyond the scope of this paper and of many water projects, few remarks can be accounted for in this discussion about the opportunity cost of water itself. First, the opportunity cost of water depends on the specific attributes of the supply, in terms of location and hydraulic connections. If water can be used only by the proposed project, it has a very low opportunity cost, since other uses elsewhere are prevented in any case by the physical characteristics of the source. Conversely, the opportunity cost is maximum when transfers of water from one use to another are relatively easy to implement, and grow up as a source becomes more densely used. Second, the relative importance of opportunity costs is determined by the regime of property rights enforced in each context. Where transfers of water are prohibited by law or customs, then the opportunity costs are close to zero. On the contrary, where private markets are free to operate, opportunity costs assume relevance since welfare maximization requires that the best supply project alternative is chosen. Third, opportunity costs are different in each single water use: high-valued uses impose a lower cost on low-valued ones, than what is the case in the opposite situation. Where opportunity costs are high, conflicts among users may arise, hence proper inter-sector and intra-sector allocation choices have to be made, and even rationing may be an option. This is

244



essential to assure that water flows could be exploited to the benefits of the best alternative use. The estimation of opportunity costs entails calculating water benefits in other alternative uses, hence it will be explored in section 4.

The demand for water The consumption side of the economics of water resources is represented by three major types of economic agents: households, farmers and other firms. The residential sector is composed of households that use water in final consumption, whereas in the agriculture and industrial sectors, water is a raw input required for the production process. In this lays the key difference between the residential demand, which is direct, and agricultural and industrial demand, which is indirect and derived as requirement for the production of other final or intermediate goods. Relevant demand-side economic sectors other then agriculture are hydropower, navigation and waste dilution. In the present context, the focus is on water demand in the agriculture sector, which is the focus of the WASAMED project, and is the major consumer in Mediterranean countries, as evident from the table 2, which shows the latest available data on sectoral water withdrawals.

Table 2. Water withdrawals, total pre-capita and by sectors (Source: FAO AQUASTAT) 1998-2002

Households (%)

Agriculture (%)

Industry (%)

Total water withdrawal per 3 capita (m /inhab/yr)

Algeria

21.91

64.91

13.18

194.1

Cyprus

29.17

70.83

0.00

301.5

7.76

86.38

5.86

968.7

Germany

12.35

19.79

67.86

570.9

Greece

16.34

80.44

3.22

708.3

Italy

18.19

45.10

36.71

771.9

Jordan

20.79

75.25

3.96

189.5

Lebanon

32.61

66.67

0.72

383.8

Malta

79.21

19.80

0.99

128.5

9.76

87.38

2.86

419.0

9.59

78.24

12.17

1,121.0

Spain

13.44

68.03

18.52

869.5

Syria

3.31

94.89

1.81

1,148.0

Tunisia

13.83

82.01

4.17

271.4

Turkey

14.81

74.23

10.95

533.7

Mean

20.20

67.60

12.20

571.9

Std. Dev

18.08

22.63

18.12

342.9

Egypt

Morocco Palestine Portugal

Each demand type has particular estimation techniques, which will be analyzed in section 4. Here it suffices to specify that the demand for water is based on the monetary evaluation of the benefits that an additional unit of water provided to each agent. The inverse demand curves, in which the quantity is a function of the price, are downwards sloping, since the benefits of an additional unit of

245



water are decreasing. Each sector is characterized by a specific relation between water quantity and the benefits derived, hence the demand curves have different slopes. The total demand schedule is the horizontal sum of the marginal benefit of water use in all the relevant sectors. Taking agriculture as the focus sector, a generic equation for the inverse water demand schedule of all farmers is be the following (Tsur and Dinar, 1995):

D( p) =

n

∑

Li qi ( p )

(3)

i =1

where the aggregate water demand D is the sum of each farmer’s optimal demand of water per unit of land, qi ( p ), derived form water being an input in the production process, multiplied by the land endowment, Li.

Welfare analysis of alternative sources of supply In institutional settings characterized by non-competitive markets, the demand and supply schedules can be partially independent of costs and value considerations, since other political and social factors may play a major role in determining the price at which water is sold. Conversely, in competitive markets, strict economic efficiency is guaranteed by the price mechanism. This can be demonstrated by using partial equilibrium analysis of market clearing, carried out by relating water quantity to its price in an agriculture market. The difference between increased costs and benefits gives the net benefits to society. Efficiency equilibrium is attained at a price where supply and demand meets. If the price is lower then the costs required of meeting the current demand, then there 26 is a deadweight loss, that is a decline in net benefits for the society. The purpose here is to use partial equilibrium analysis to explain briefly the advantage of estimating the supply of water form alternative sources, following Zekri and Dinar (2003). In figure 5, the demand for water in agriculture, D, and different curves of water supply are shown, which reflect both use and opportunity costs of supplying water. Public supply, Sp, at service level Qp is provided at price Pp. There, the quantity of water demanded is Qp’, but this quantity cannot be provided by public supply, which is therefore exogenously fixed. The part of consumers that are not satisfied would be willing to pay up to P1 to improve the service. Hence, there are incentives to introduce alternative supply projects, represented by the supply schedules SA1 and SA2. Given the current demand, the two alternative projects can provide the same quantity Q2 of water at the same price P2, but are characterized by different elasticity of supply to price:

ε S = ( ∂ Q / Q ) ( ∂ P / P ) = ( ∂ Q / ∂ P )( P / Q )

(4)

The same formulae applies to the elasticity of demand, ε D . Under public supply, social surplus is given by the area abcQp. Social surplus with SA1 is equal to the area hbf. Hence an alternative supply is socially justifies whenever hbf – abcQp > 0, or alternatively cdf > ahdQp. Social surplus with SA2 is equal to the area cfd’. With two alternatives, the best project is given by the comparison of the net benefits with SA1, | cdf – ahdQp | with those with SA2, | cfd’ – a’d’Qp |.

26

We point to the works of Young (1996) and Briscoe (1996) for the analytics of these results.

246



P Sp b

S A2 SA1

c

P1

f

g

P2

d Pp d h a’ a

Qp

Q

Qp’

Q1

Fig. 5. Partial equilibrium analysis of alternative sources of water supply (Source: Zekri and Dinar (2003) Estimation of supply and demand elasticities makes it possible to empirically calculate the areas under supply and demand functions. When quantities are known, assuming an invariable demand function, one can derive the incremental benefits. The observed quantities of public and alternative

∂Q = Q1 − Qp. Similarly, ∂P = P2 − Pp . Therefore, supply give the incremental quantities, the estimated elasticity can be used to specify supply and demand functions and perform welfare comparison. In particular, the area V under the demand curve between two points of consumption, say Qp and Q1, can be calculated in discrete terms as follows:

[

V = P1 × Q p ( 1

εD )

] [

1 − 1 εD × Qp

Q p (1

ε

D

)

− Q1 Q1 (1

ε

D

)

]

(5)

Box 1. Case study of alternative supply sources in rural Tunisia The estimation of elasticity of supply and demand for water has been carried out by Zekri and Dinar (2003) for rural Tunisia, where the public supplier, SONEDE has expanded production in rural areas, traditionally supplied by associations of joint use called ACI. Their main results are summarized in table 3. According to authors’ estimations, the water sector in Tunisia presents conflicting results. In 1996, ACIs were more efficient than SONEDE form a cost perspective. The price-to-cost ratio, which indicates the cost recovery rate of the supply agency or instead the level of subsidization, clearly shows that ACIs recovered a larger proportion of their O&M costs. Moreover, while ACI members paid 21% of the total costs, SONEDE customers paid only 18%. However, both systems were highly subsidized, in that ACIs receive substantial public subsidies and SONEDE used cross-subsidies among regions. Furthermore, both systems operated at a sub-optimal scale (declining marginal costs), what is evident from the negative values of the producer surpluses. The high consumer surpluses indicate instead that consumers would be willing to pay more for improvements. Hence, there is scope for additional steps, such as price increases, in order to expand supply to a more efficient scale.

247



Table 3 Welfare analysis of supply alternatives in rural Tunisia, 1996 (Source: Zekri and Dinar,2003) Total cost per unit of water (TD/m3) Average O & M cost of water (TD/m3) Average charge per consumer (TD/m3) Total bill per consumer (TD) Price-to-cost ratio (%) Cost elasticity of supply Price elasticity of demand Total consumer surplus Total producer surplus Total social welfare

SONEDE 0.422 0.241 0.190 44.6 45 -1.42 -0.24 2.45 -1.91 0.54

ACI 0.929 0.148 0.195 13.0 132 -0.78 / -0.42 - 1.30 5.04 -10.64 -5.60

METHODS FOR DETERMINING THE VALUE OF WATER (Section 4) In adopting new strategies for water management, a central issue concerns effective evaluative procedures. In the definition of water conservation, two criteria have suggested that the methods adopted must reduce water use or consumption and must also be socially beneficial. This section examines the demand side of the economics of water, exploring the evaluation criteria that can be used in assessing various water demand management measures. Economic or allocative efficiency addresses the value of scarce resources available to society. Thus, concern with the economic efficiency of water use creates a concern about net values of water in alternative uses and whether existing institutions are flexible enough to permit the allocation of existing supplies in such a way that society as a whole derives maximum value from those supplies. In an economically efficient resource allocation, the marginal benefit of the employment of the resource is equal across uses, and thus social welfare is maximized. Hence, there is a case to understand the underlying economics of water demand and value in various economic sectors. The starting point is figure 6, which shows the various components of water value identified by the wellknown study of Rogers, Bhatia and Huber (1998). As evident, in order to arrive at determining the full 27 value of water to societies, several components have to be included in the calculation. A common aspect to valuation techniques is therefore the need to consider many possible benefits that water produces. The economic value of water, in each location, each use and each time, is given by the sum of: o value to users, calculated on the basis of marginal value product, which is an estimate of per unit output for a unit of water used; o net benefits form return flows, derived from aquifer recharge during irrigation, or downstream benefits of water diversion during hydropower generation; o net benefits form indirect uses, derived when the water diverted for one purpose is used for another purpose; o adjustment for societal objectives, in order to take into consideration the wider considerations, such as poverty alleviation, gender empowerment and food security.

27

The issue of water valuation has been also discussed during the Expert Group Meeting on Strategic Approaches to Freshwater Management, held in Harare in 1998. The Meeting recommended considering the value of water within the broader context of Integrated Water Resources Management (IWRM). Hence, in what follows, a basin-wide approach to economic valuation will be used, although priority is given to valuing water for agriculture.

248



Fig. 6. Components of water demand value (Source: Rogers, Bhatia, Huber, 1998) The authors also refer to the intrinsic value, reflecting environmental, social and cultural benefits, in order to arrive to the full value of water. Nonetheless, the present paper focuses on the economic value and presents an overview of the main techniques employed for estimating the marginal benefits to users. In fact, when competition among uses arises, a careful examination of marginal benefits in each use could help identifying large disparities and aid pressure for legal change in allocation rules. In addition, marginal benefits of water use should be compared to marginal costs of water supply proposals, in the interest of promoting economic efficiency and fiscal responsibility (Gibbons, 1986). In general, the economic equilibrium can be achieved through the operation of price signals in a competitive market place. Nevertheless, in the absence of market-clearing prices, there are a number of alternative means of estimating the value of a resource.

Classification of major water use values A long-standing debate on how to value water has led to recognizing the need to have a clear analysis of what this means. In fact, it is widely recognized that water has traditionally been regarded as a free resource of unlimited supply with zero cost at supply point and, at best, water users have been charged only a proportion of the cost of extraction, transfer, treatment, and disposal. All associated externality costs of water have been ignored and users have been offered very little incentive to use water efficiently and not waste it (WWAP, 2003, pp. 327-8). However, water has a value to society in all its uses. To be exact, it has several values, each specific to each location, each use and each time. Hence, valuing water is an exercise that should be undertaken systematically and consistently in water resource planning and management. The key issue is the cost imposed on others by a particular use of the resource, which is called opportunity cost. The economic valuation of water resource starts form the premise that water can be considered a natural asset, the value of which resides in its ability to create flows of goods and services over time (Agudelo, 2001). Values derived from water can be mainly divided into use values, also know as 28 extrinsic or direct values, and nonuse values, called intrinsic or passive or existence values. 28 Option values, that is the desire of individuals who do not actually use water to preserve nonetheless its integrity for future eventual uses, is also a nonuse value.

249



In the present paper, the interest is centered around use values, that is the economic benefits derived from direct use of water by consuming it or its services. Comparing marginal values between sectors, in fact, allows assessing the economic efficiency of allocations among them. For a fair comparison, adjustments are required to express values in commensurate terms of place, time, source and quality. This can be done by further specifying the different categories under which water use values can be classified, as in Figure 7.

Fig. 7. Specifications of water use by three criteria (Source: Agudelo, 2001) According to their subtractability, water use values can be divided into consumptive and nonconsumptive values. If the former case, as in the main sectors of the economy, water is no longer available for further uses. 29 Considerations about competition are paramount, since consumption by one economic agent generate a negative externality on others, in terms of opportunity costs in alternative projects. Complementarities also assume relevance, as water can be used repeatedly or even simultaneously for different uses. Finally, changes in water quality are important too, since they may differently affects beneficial uses elsewhere. 30 Conversely, non-consumptive use values include the benefits received in the activities that leave water and its properties essentially intact. Hydropower and navigational uses are the main examples. Another breakdown of water uses is by location. Those uses occurring in a watercourse and dependent on its flow characteristics are called instream uses. Those uses for which water has to be removed from the watercourse are called offstream uses. The distinction assumes relevance since water is a bulk commodity, that is its transportation is extremely costly per unit of water. Hence, the location becomes a crucial element in assigning values to water, because adjustments have to be made to reflect the site-specific nature of offstream uses. Important offstream uses are those related to the three main economic sectors, while the main instream uses are navigation, hydropower and waste dilution. Finally, by their economic role, water values can be distinguished according to the level of the value chain in which they appear, so as to distinguish between intermediate or final water uses. Water is a final good for households and uses in navigation and waste dilution, while it is an intermediate good in all other economic sectors, which use water as an input in the production process. This is reflected in the different methods that can be used to estimate water use values in the different sectors. The focus of the paper is on agricultural uses. These can be classified, according to the aforementioned scheme, as offstream consumptive uses of an intermediate good. Coherently, the 29

In this sense, it is essential to keep in mind the distinction between withdrawal and consumption, due to the possibility of reuse and recycling of water withdrawn. 30 Accordingly, the waste dilution properties of water can properly be classified as consumptive uses.

250



estimation technique used mainly in agricultural research are explored in the next sub-sections. Beforehand, in Box 2 an overview of the trends in water uses in the two other main sectors of the economy is provided for OECD countries. Table 4 instead presents some indicators related to a broad estimation of potentials for expanding water uses in agriculture in WASAMED countries. It can be firstly noted that the weight of the agriculture sector, in terms of area cultivated and population employed, is clearly dependent on total land and morphological characteristics. More importantly, irrigation potentials are substantially different, implying different scopes for irrigation extensions and diverse opportunity costs of doing so. On the other hand, those countries whose cultivated areas are close to be fully covered by irrigations services, would face higher incremental costs of expanding irrigation. Finally, the percentage of irrigated area on total cultivated land highlight the infrastructure-poor countries and show where potential improvements are likely to cost less at the margin. Hence, a close economic investigation of water resources availability, supply costs and willingness to pay, that take all those elements into account, would be especially helpful to guiding water policies. Box 2. Global trends in water use by sector Registered trends in water use vary among countries and, within countries, among sectors. Globally, agriculture is responsible for about 69% of total freshwater abstraction. The corresponding figure for OECD countries is 45%. Agricultural demand for water is projected to increase substantially over the next few decades, as much of the additional food that will be needed to feed the world’s growing population is expected to come from irrigated land. Over the past 20 years, there has been a continuous upward trend in water use for irrigation in many OECD countries, associated with an increase in irrigated land area that has been mainly encouraged by government investment in irrigation infrastructure and by irrigation water subsidies. For most countries, irrigation water represents over 80% of total agricultural water use, with much of the remainder being accounted for by livestock farming. While agriculture is likely to remain the primary abstractor of freshwater in the near future, industry will be the fastest growing water user overall, largely due to rapid industrialization in many non-OECD countries. Industry is the fastest growing user of freshwater resources worldwide, and demand from this sector is expected to more than double over the next two decades. Industry accounts for 23% of global water abstraction, weighted towards the OECD countries but with industrial use in developing countries growing. The most water-intensive industries include pulp and paper, chemicals, and food and beverages. Another important emerging trend in many OECD countries is the growing use of freshwater for cooling in electricity production. The remaining 8% of global water abstraction is used by households. Source: OECD (2003)

251



Table4. Main agriculture water indicators in WASAMED countries in 1998 (Source: FAO AQUASTAT) Total Cultivated area Drained area as economically (arable land and a percentage of active population permanent cultivated land in agriculture crops) (1000 ha) (%) (1000 inhab)

Area equipped for irrigation as percentage of cultivated land (%)

8,265

2,660

Cyprus

113

31

37

3,400

8,475

4,420

11,997

923

485

4.043

3,846

753

1,422

36.97

11,064

1,220

2,698

24.39

Jordan

400

192

85

Lebanon

313

43

1775

10

2

2

Morocco

9,283

4,274

Palestine

19

Germany Greece Italy

Malta

Portugal

6.97

510

Area equipped Part of area Area equipped for irrigation: full equipped for for irrigation as control - total irrigation actually perc of irrigation (1000 ha) irrigated (%) potential (%)

Algeria Egypt

0.74

Irrigation potential (1000 ha)

1,664

513

79.61

111.60

6.89

3,422

100.00

77.42

100.00

77

1,417

97.50

12

0

86.70

15.54

2,358

609

632

26.80

Spain

18,715

1,220

3,640

19.45

Syria

5,421

1,563

Tunisia

4,908

958

Turkey

28,523

14,697

Mean

6,790

2,508

3.91

Std. Dev

7,885

4,027

3.12

252

4.01

1,250

1,267

560

367

99.75

70.36

8.03

1,721

1,329

75.37

86.52

26.97

2,731

1,269

42.98

18.00

29.60

8,500



Values inferred from the markets: are they reliable? In section 3.4, a technique has been presented that has been used to select the best source of water supply when one or more alternatives are possible. The analysis therein calculated the water demand making use of regression techniques that use data from actual transactions in irrigation water markets. The market approach accomplishes the estimation of the value of water by means of the observation and analysis of market transactions (rentals and sales) of either or both water rights and land properties with irrigation facilities. In the latter case, the value is implicitly packaged in the value of the property and is given by the difference between the price of irrigated land and the price of comparable non-irrigated land. It turns out that, for estimations to be reliable, well-functioning water markets and irrigation land property markets should be in place. This possibility is ruled out if water markets are poorly functioning or are absent at all, as seems to be the case in many countries (Briscoe, 1996; OECD, 2003). The study of the performance of water markets is outside the scope of this paper, being already analyzed in a previous paper of the WASAMED project (Billi, Meroz, and Quarto, 2004). However, it is worth mentioning the limitations, some technical and one theoretical, of the technique used to estimate water values from market observations, when these market are in place. From a technical point of view, it should be firstly noted that observed rental prices are usually referred to the short term, whereas observed prices for perpetual water rights give a more correct picture of long-term incremental marginal value of water. However, for the purposes of planning horizon, the price for perpetual water rights has to be converted in annual values, using capitalization formulas. This in turn requires the selection of appropriate planning period and interest rates, which is a non-trivial task. Moreover, the assumption implicit in capitalization formulas that the annual value is constant over time is also unrealistic from a long-term perspective, since inflation and variation of real values of water can influence expectations and profitable opportunities. From a theoretical point of view, water values estimated from market transactions may be unreliable since they tell little about the actual shape and elasticity of demand to an expansion of the sources of supply. In figure 8, P0 is the price observed from market transactions. However, the historical estimated demand is known only up to this market price. Before undertaking a project, the willingness to pay for more quantities of water is unknown. Hence, form that point onwards, the water demand D may take either of the different shapes D1, D2 or D3 in figure 8, so the actual prices would vary greatly. This effect is not taken into consideration in market estimates, which usually assume an invariant demand. Price

D

S1

P0 P3 P2 P1

S2

D3 D1

D2 Quantity

Fig. 8. Limitation of current market prices as a measure of water value Source: Agudelo (2000)

253



What precedes imply that even when prices are easily discoverable, they are an imperfect and frequently bad-performing estimations of the values of water. This suggests that market estimations should be complemented by valuation techniques that rely on other kind of data. 31

Non-market estimation of the value of water as an intermediate good The framework for valuing water should link changes in the physical characteristics (quantity and quality) of the resources to changes in the level of services or uses of water, and to how each society value these changes. The methods usually assess on-site water values with point estimates. That is, water valuation is a highly-focused activity that should be carried out at least at the level of the single service or basin. Moreover, for off-stream values to be compared to in-stream values, the former values have to be reduced by the cost of transporting or pumping and distributing water to end 32 users. Valuation techniques are based on the concept of shadow pricing. Shadow prices are dollar values a resource display in a given situation, when all internal and external economic (and frequently social and environmental) factors are taken into account. Non market valuation techniques are evolving, in response to social and environmental pressures and the desire of policy-makers to adopt more informed decisions. Shadow pricing of water is base on four alternative approaches: the residual imputation; the change in net income; the value added; and the alternative cost. The second approach is a variant of residual imputation. Both are base on the estimation of the farmers’ production function and use the same analytical apparatus developed in the present paper. They are referred to as the ‘farm budget approach’. In what follows, this approach is emphasized. 33 The basic procedure estimates the farmers’ production functions and then infers the demand function from the analysis of optimal water use patterns, utilizing mathematical programming methods. The technique starts form the theory of cost-minimizing producers that use an input up to the point in which its dollar-value marginal contribution to the production (value of marginal product – VMP) equals its price or marginal cost of supply. This is the basic point of benefit-cost analyses of development projects. In a model agricultural production function, a vector of m crops, Yj, is produced out of employing irrigation water, whose quantity is QW, and n other factors of production, whose quantities are expressed by the vector Xi. Given that, in a competitive equilibrium, the price of each input is equated to its returns at the margin, then the price of the single non-traded input may be derived by making it the residual claimant of the total value product.34 Therefore, solving for the price of water, its shadow P* price, W , is equal to its value of marginal product, VMPW, which is given by:

VMP W

⎡ = PW* = ⎢ ⎢⎣

∑ (Y j m

j =1

× Pj

) − ∑ Qi n

i =1

⎤ × Pi ⎥ ⎥⎦

QW

(6)

The shadow price given by equation (5) is interpreted as the on-site maximum average willingness to pay for water for that combination of crops. To derive the equivalent marginal value, a refinement has been introduced by the ‘change in net income’ approach. The value of water is the change in the net benefits attributable to the project, which in turn is given by the difference between the with- and 31

Other aspects that prevent observation from market transactions to be a good indicator of water values are the following: externalities are not accounted for in the individuals’ transactions; prices do not reflect in-stream values or other environmental considerations; there may be imperfect competition between suppliers; imperfect and asymmetric information may prevent efficient decision-making; and concerns about equity and conflict resolution are left out of the analysis. 32 From a higher perspective, the value-flow concept integrates these point estimates in space and time, yielding values of water in the different stages throughout its flow. See Seyam, Hoekstra and Savenije (2003). 33 The discussion draws heavily from Young (1996). 34 In this case, the Euler’s theorem, that allows the total value of product to be divided into shares, is assumed to hold.

254



without-project net incomes derived from producing that combination of crops. In symbols, calling

Z =

(∑ Y

j

Pj −

∑X

i

Pi

) the net income, with inputs is also including water, then the change

ΔZ = Z − Z

1 0 , where the subscripts 1 and 0 refer respectively to the in net income is equal to with- and the without-project water situations.

The main problem of using the aforementioned approaches involves specifying the production functions, especially in the long term. This leaves room for both omitting relevant variables, and forecasting inaccurately the levels of output associated with a given increase of an input. However, this technique remains a useful instrument for public planning, which is usually oriented towards the short and, at best, the medium terms. This is especially true today since computer technologies have made available sufficient power to run mathematical programming models that make use of several production functions. These models are important instruments where a large set of activities compete for scarce resources. The models find the profit-maximizing set of activities given the resources constraints. They can trace out a set of net total benefit points, from which a set of marginal values 35 can be derived. For the sake of completeness, it is worth describing briefly the basic characteristics of the two other approaches mentioned at the outset of this sub-section. The value added approach starts from the net payments to primary economic resources, 36 in order to build up an input-output matrix, organized on a sector-basis, providing a static picture of production processes in the economy. The estimated value of water is referred to a broad sector, such as agriculture, and is equal to the residual value added per unit of the resource, imputed after subtracting for the value added of all other economic resources. The approach gives broad estimations of average values, hence it is less useful in broad basin-wide efficiency analysis, than what it is in planning allocations within a sector. Finally, the ‘alternative cost approach’ is a variant of cash-flow analysis, aimed at determining the cost of producing an output in the next-best project, and then attributing that cost as the water of water for the proposed project. This technique is better suited for determining the least-cost option of a supply project, than for carrying out complex analyses of efficient allocation in a river-basin context.

Economic analysis of water allocation policies Having defined the main tools for estimating the incremental benefits of water projects, one can turn out to analyzing a proposal to implement a particular agriculture policy, such as the expansion of an irrigated area. The aim is to single out the true cost of the proposed project by including the opportunity cost of water that can be used alternatively by other agents in the same sector. Many other techniques can be applied, as highlighted before. Here, the aim is informative, hence a simple case is analyzed, and then the results of an empirical test conducted in Egypt is acknowledged in Box 3. The starting point is 9, which represents the optimal allocation choice between to competing projects. The evaluation of water in each single project gives the estimations of the value of marginal product of water inputs, which are decreasing in the quantity. The optimal choice is taken when Q1 and Q2 water is allocated to project 1 and 2 respectively. With such allocation, the marginal benefits of water are equated across the economy and take the value of λW. Societal benefits are therefore maximized form an economic efficiency perspective with such an optimal allocation.

35 This aspect, integrated with geographic information technologies, may provide visual comparisons of the net benefits of alternative policy alternatives. This aspect is further explored in the conclusions. 36 That are wages, capital, depreciation, rents to primary natural resources, and payments for government services.

255



Water value

Water value VMPW in project 1

VMP W in project 2

?W

?W

Water volume Q1

Q2

Fig. 9. Optimal allocation of a limited supply of water

Box 3. Allocation scenarios in Egypt with a farm budget approach Wichelns (2002b) estimated the true costs of a water project in Egypt that proposes to divert water away from the delta region. Through a small-scale simulation model of production regions, the author calculates empirically the optimal allocation of irrigation water under different policy objectives between three irrigation expansion projects, using a farm budget approach to maximize total net revenues from agriculture. The main results are summarized in table 5. The net revenue impacts of alternative water allocation policies vary with the total volume of water available for irrigation in the three regions. That volume is expected to decline in the future, as municipal and industrial demands increase. Two sets of scenarios are examined, pertaining to two policy objectives: a) the supply is allocated so as to maximize the sum of net revenues generated in the three regions; b) as supply is reduced, the volume of water is held constant in Toshka and Sinai and reduce in the Delta. The optimal irrigation depths in table 5 are used to show the kind of economic analysis that can be done with farm budget estimations. Under the first policy objective, when water supply is reduced, net revenue is maximized by reducing irrigation depths in all three regions and reducing irrigated area in Sinai, where the marginal productivity of water is the smallest. When water supply is reduced by 10%, it is no longer optimal to irrigate land in Sinai (the case is not reported in the table). The second policy scenario, where all reductions in irrigated area and irrigation depths occur in the Delta, is far less attractive. In this case, the sum of net revenues for each reduction is smaller than under the first policy scenario. The impact of the second policy objective is reflected in the marginal value of water, which increases in the Delta, where the net revenue from crop production declines more sharply. Reallocating water to the Delta is, therefore, welfare improving, since it is the most productive region. For a similar estimation of water values in Namibia, see MacGregor et al. (2000).

256



Table 5 Optimal allocation of water in three district in Egypt (Source: Wichelns 2002b) Policy objectives

Regions Delta Toshka Sinai Delta Toshka Sinai Delta Toshka Sinai

a) Maximizing the sum of b) Maintain full production in net revenues Toshka and Sinai regions Area in Applied Net unit Area in Applied Net unit prod. water revenue prod. water revenue (ha) (cm) (US$/ha) (ha) (cm) (US$/ha) Water supply not limiting 8.3 94.9 711 1.2 9435 435 0.5 93.9 160 Water supply reduced by 5% 8.3 90.7 707 8.3 89.2 704 1.2 89.6 431 1.2 94.5 435 0.4 88.2 155 0.5 93.9 160 Water supply reduced by 20% 7.7 77.4 642 1.2 94.5 435 0.5 93.9 160

ACCOUNTING FOR AND MODELING EFFICIENT WATER USE (Section 5) The starting point of basin-wide water accounting and modeling is that the water sector interacts with all other sectors of the economy, and could potentially become a binding constraint on economic expansion and growth. This is especially true because while the amount of renewable water resource is practically fixed, water demands will continue to grow and diversify. Thus, the economic challenge is to maximize social and economic benefits under varying circumstances, by efficiently using the available resources. Water accounting is based on the water balance approach and focused on the hydro-social cycle described at the outset of section 3, that is on the interactions between two systems within a territory of reference: the user’s system represented by the economy and the water resource system. The territory of reference can be a country, a region or a river basin. Figure 10 describes these interaction, both between the water system and the economy for a given territory, and their interaction with the economies and the environment of other territories. Water accounting is not intended to give an explicit valuation of water. Instead, it registers the flows of water over time in and out of the physical system and the economy. Of special interest here is the fact that water is considered a physical asset as all other economic assets, whose stock should be calculated at the beginning and at the end of each given period. The objective is to provide a large amount of data by disaggregating water inflows and outflows per supply source and demand sector, so that specific water accounting indicators can be calculated, as well as more sophisticated estimations of water use efficiency and productivity can be performed. Modeling differs form accounting in that it explicitly deals with the issue of estimating water values in use and productivity in alternative uses, for the purpose of optimizing its allocation at the basin level. Models developed to this aim use mathematical programming for solving complex equation systems, that usually involve both a simulation model for the hydrologic components, and an optimization model for the economic components. The main advantages and disadvantages of each of the two techniques are presented in turn in what follows. The underlying thesis is that both tools are necessary for appropriate water planning and management. Economic indicators that can be derived from those methods are acknowledged as well.

257



Atmosphere

Territory of Reference

Returns

Returns

Abstraction

Returns

Abstraction

In situ use of precipitation

Abstraction

Physical Water Resource System

Evapotranspiration Sea

Sea

ns tur Re

ti o ac str

Households

Re

Ab

tu rn s

Sewage and refuse disposal...

n

Other Industries (incl. Agriculture) RoW Economy Imports

Collection, purification and distribution of water; Transport via pipeline

Exports

RoW Economy

Economy

Fig. 10. Main flows of water within the economy (Source: UNDESA/UNSD)

Basic features of water accounting techniques As evident from figure 11, the first element to consider carefully in water accounting is the water available for abstraction. This is not the net inflow of water, which is the gross inflow (effective rainfall plus 37 surface and sub-surface flows) plus change in storage. In order to obtain the quantity of water available for supply in a given year, the net inflow must be reduced by the amount of water flowed out of the system because already committed to other uses, such as downstream rights or minimum stream flows for the environment. An interesting extension, following Merrett (1997), is to allow for the possibility of water recycling. The supply of water augmented with recycled water would then be:

SC = E +

n

∑δ

P

IP

(7)

p =1

where SC is the total supply corrected for recycling, E is the net inflow, n is the number of discharge points P, IP is a specific flow of recycled water at a defined discharge point, δP is a parameter that reflect the distance of that particular discharge location from the sea, divided by the distance from the sea of the point where that flow was abstracted. This formulae allows taking into consideration the location of the recycling point, such that those plants that collect wastewater downstream and discharge it treated upstream, can be attributed a higher role in total supply. This is an important aspect when considering the valuation of the net economic benefits of a project intended to provide treatment facilities. Of the available water resources, part flows out of the system in both usable and non-usable form, while part is depleted, either beneficially in a productive activity or non-beneficially, through evaporation, deep percolation into saline aquifers, or waterlogging. In turn, beneficial depletion is composed of both the amount depleted in the process of producing goods and services of value, or the amount that is naturally depleted through evaporation and other causes (Molden and Sakthivadivel, 1999). 37

Effective rainfall is equal to total rainfall in an area minus transpiration and evaporation.

258



Fig. 11. Main water accounting relationships Source: Molden and Sakthivadivel (1999) Water accounts can be compiled at the basin, service or use levels. Maintaining a consistency between these various levels is an ambitious, but unavoidable task of future applied research. It is a promising approach in that it gives evidence of the main water uses in the economy, By disaggregating water inflows and outflows per supply source and demand sector. However, the technique is limited in that, in order to provide meaningful economic information, it needs estimating water values through other techniques. It can therefore be considered as a complement, not a substitute, for more sophisticated analysis of integrated models, as those that follow.

Integrated water basin modeling: rationale and classifications Water resources management modeling represents the most advanced tool for optimal water efficiency, reliability and cost-effective use. Such characteristics as the inherent intricacy of aquifers flows, the host of hydrologic uncertainties, the stochastic nature of water flows, the possibility of using various supply sources simultaneously, the conflicts and complementarities that arise in water use, all make complex mathematical modeling the necessary means to effective planning and efficient management. Results from these models can be interpreted to reveal opportunities for improvement 38 over the status quo. The hydrologic approach to water resource modeling is mainly concerned with simulating the functioning of a hydrological system and/or optimizing water flows under a technically-inferred objective function. On the other hand, the aim of integrated modeling is to characterize, not only the natural and physical processes, but also the proposed projects and institutional strategies, and to optimize for the maximum net benefit to society under a politically-inferred multiple-objective function, whose weights reflect the underlying social values (McKinney et al., 1999).

38 Lee and Dinar (1995) discuss at long the limitations and weaknesses of this sort of models. The results depend on the model assumptions embedded in the objective function. Hence, the degree of accuracy and specification is usually limited. Data limitations, qualitative factors and subjective inference also play a role. Some of these critiques are further explored in the conclusions.

259



The task of these models is to allow generating both physical savings of water and economic gains, by increasing the output per unit of water used. This involves both using water where it has the most value (an allocation decision), and reducing water losses due to evapotranspiration, salinization, pollution and percolation (an evaluation of a proposed project’s benefits). Water resources management models can be referred to the short or the long term. Short-term models estimate the optimal combination of water quantity and quality for one year or single irrigation season. Long-term models also account for the effects of salt accumulation in the soil profile and, in the extended versions, in the groundwater. Simulation model is the preferred technique to assess water resources systems’ response to extreme, non-equilibrium conditions. Optimization models are based on an objective function and a set of constraints that can include social values and institutional settings. However, integrated hydrologic-economic optimization models always need a simulation component, in order to characterize the hydrologic regime, although usually at a considerably simplified level. Integrated models can also follow a compartment approach or a holistic approach. Under the former, the main relationships that characterize the hydrologic and the economic systems are represented as stand-alone systems of computable equations, whose output data are transferred between the two components. In the holistic approach, there is one single unit that contains both components, which are tightly connected to a consistent analytical framework. Finally, it is at the basin level that hydrologic, agronomic, and economic relationships can be best integrated into a comprehensive modeling framework. At this level, in fact, allocation decisions have the widest economic implications. As a result, it is recommended that the design of policy instruments, to make more economically-rational water use, is carried out at this level. The basin as appropriate unit of analysis has long been acknowledged in international fora.

The analytical framework of hydrological-economic optimization A river basin system is made up of three types of components: (1) sources, such as rivers, canals, reservoirs and aquifers; (2) off-stream and in-stream demands; and (3) intermediate facilities such as treatment and recycling plants. The conceptual framework is developed as a node-link network, with nodes representing physical entities and links the connection between these entities (Rosegrant et al., 2000). The nodes are the sites of the three aforementioned components. At each agricultural demand site, water is allocated to a series of crops, according to their water requirements and economic profitability. Figure 12 shows as example the node-link representation of the Maipo river basin in Chile. The graphical representation also makes it easier to detect the spatial relationships among the various elements. Water demand is determined endogenously based on empirical agronomic production functions. Water supply is determined through the hydrologic water balance in the river basin with corrections for distribution to the irrigated crop fields at each irrigation demand site. Water demand and supply are then balanced based on social objective functions, such as maximizing net benefits to water use. The calculation of the salt concentration allows the endogenous consideration of this important externality with respect to upstream and downstream irrigation districts. The major component of integrated hydrologic-economic models is the representation of the production functions for agriculture that include water as an input, and demand functions for domestic and industrial uses, in order to estimate the uses and values of water by sector. Other types of water demand, such as hydropower, recreational and environmental demands, can also be included, though at the cost of complicating the analysis. The specification of the theoretic agronomic production relationships can be done in several ways, some of which have been described in section 4. Here the concept is operationalized in terms of specific econometric estimation models drawn from the literature. The basic framework is a farm model of constraint optimization, where constraints are given by physical characteristics of water and soil (given by the hydrologic model), institutional policies (prices, property regime), and financial/investment restrictions. The objective function is maximizing the

260



present value of farmers’ profits, PA, at each demand site, over a chosen time horizon (Varela-Ortega et al., 1998). The farmers’ net profits are given by crop-water functions, which relate crops yields to water applications. One of the most used crop-water function is that of Dinar and Letey (1996), which for each demand site defines: m

PA =

T

∑

t =1

∑

j =1

A jY j Pj −

(

m

∑

A j Fc

j

(1

t

j =1

+ r)

+ Tc

j

)−

Q Wt PWt (8)

where Yj, Pj, QW and PW indicate as before the jth crop quantity and price and applied water quantity and price, whereas Fcj and Tcj are respectively the fixed and technological water application costs of crop jth, 39 and the subscript t = (1, .., T) refers to the each time period. Fixed and technological costs can be specified in technical terms and be referred to biophysical determinants, other inputs’ costs, and externalities such as salinity in return flows.

Fig. 12. Schematic representation of a river basin (Maipo, Chile) (Source: Rosegrant et al., 2000 It is important to stress that PA is to be referred to each single demand site, in order to reflect differences in hydrologic conditions. Ideally, it should also account for several cropping patterns. This can be done by further specifying the underlying economic functions, by at the expenses of greater model complexity, more rigidity of the hydrologic model, or its complete absence, such as the analysis that has been applied to the case of a water-scarce basin in southern Spain, as described in Box 4.

39

One technological cost parameter is the Christiensen Uniformity Coefficient (CUC), that is used as a proxi for both irrigation technology and management activities. The choice of water application technology can be determined endogenously. See Rosegrant et al. (2000).

261



In a complex system that accounts for all water uses, the objective function solves simultaneously for all the sectoral benefit function. 40 Moreover, the objective function may account for climate uncertainty and varying water storage capacity, using stochastic techniques, 41 or employing more sophisticated tools for incorporating spatial and temporal distribution of the aquifer response to external stresses. 42 The institutional rules are the very key policy variables. They can dictate, on a command-andcontrol basis, specific allocations of water resources among irrigation districts, among sectors, and among in-stream and off-stream uses. Institutional can be set so as to resemble more or less the market mechanisms, in order to introduce efficiency elements into the system. An even the introduction of private property rights, as well as markets for permanent allocations and/or rental markets, can be tested out of these models. Box 4. Accounting for different cropping patterns in water-scarce regions Reca et al. (2001a) consider three resolution levels in an optimization model or an irrigation system, where water scarcity is taken as an exogenous external constraint. The objective function is to maximize the benefits of consuming a volume of water today, or instead store it and consume more water the uncertain future. The model is limited to a detailed economic component, hence the hydrologic component is taken as given in the process of economic optimization. The first sub-problem studies the optimum irrigation timing that maximizes a single crop yields. The second component derives optimal aggregated economic functions associated to each irrigation area, by analyzing optimal land and water allocation for all cropping patterns. Finally, the third level addresses optimal water allocation for a complex distribution system, such as the irrigation system of an entire basin, by taking into account the economic functions for each irrigation area. The authors in a companion paper (2001b) apply the model to the Bémbezar system in the Guadalquivir river basin in southern Spain, using data produced ah hoc from field irrigation evaluations. A deterministic analysis has been carried out in order to compare optimum water and cropping patterns management with actual ones, in a stochastic environment that determines water availability. The authors conclude, among others, for the superiority of water markets in reducing consumption levels.

INDICATORS OF WATER USE EFFICIENCY AND WATER PRODUCTIVITY The overview of water economics principles and techniques developed in the previous sections would not be given full significance unless remarks would be made about the appropriate combination of water data and modeling outputs. The purpose of the present section is providing the reader with insights about the most useful indicators of water efficiency used in irrigation analysis and policy making. Indicators are usually expressed as fractions, that is to say ratios between physical measures of water inflows and outflows. Several comments and remarks arisen in the literature are provided.

Technical indicators of efficient water use The first class of ratios considered herein as a benchmark for reference are the technical irrigation efficiency, which is adimentional (Palacio-Vélez, 1994). Efficient water use, Ef, is defined as the ratio between the actual volume of water used for a specific purpose, Vu, and the volume extracted or diverted from a supply source for that same purpose, Ve. Functionally expressed:

Ef = Vu 40

Ve

See for example the water policy analysis of the Mekong river basin in Ringler et al. (2004). See the work of Mejías et al. (2004) applied to an irrigation district in southern Spain. 42 See for example Faisal et al. (1997). 41

262

(9)



Efficiency in the use of water in irrigation may be separated into three components: storage, conveyance, and irrigation efficiency. Storage efficiency, Es, is the ratio between the volume of water diverted for irrigation, Vd, and the volume entering a storage reservoir for that same purpose, Ve. In symbols:

Es = Vd

Ve

(10)

Conveyance efficiency, Ec, is the ratio between the volume of water diverted to irrigation plots, Vp, and the volume diverted from the supply source, Vd, that is:

Ec = Vp

Vd

(11)

The classical irrigation efficiency, Eu, is defined as the volume of water beneficially used net of evapotranspiration, Vu, divided by the volume of water diverted, Vd, or analogously:

Eu = Vu

Vd

(12)

When large volumes of surface runoff or deep percolation are generated during irrigation events, the value of irrigation efficiency Eu tends to be low, even if a proportion of the drainage water is used by other farmers. This generates potentials for misinterpreting those low values. 43 Keller et al., 1996 introduced the concept of ‘effective efficiency’, EE, as:

E E = Et

Vp

(13)

where Et is the net crop evapotranspiration and Vp is the net volume of water diverted to a field that reaches the irrigation plots. Water that becomes usable surface runoff and deep percolation is subtracted from the total volume delivered. Effective efficiency can be also adjusted to reflect water quality. An aggregate version of the effective efficiency is the concept of ‘basin’ or ‘global efficiency’, whose value increases when farmers reuse drainage water (Seckler, 1996). The total efficiency of water use for irrigation, Ei, is given by the product of (10), (11) and (13), that is:

Ei = ES Ec EE

(14)

A common deficiency of physical indicators of water efficiency is that they may generate misperceptions among policy analysts, who may interpret higher values as preferable to lower values, without examining the economic implications of alternative allocation scenarios. Yet, without considerations of economic variables it is not possible to determine if higher irrigation efficiency generates greater net economic value to society.

Economic indicators of water efficiency and productivity The economic indicators of water use efficiency and water productivity take their rationale from the discussion of the previous sections. They are usually expressed as ratios of physical and economic variables referred to a specific time period, though many economic indicators are based on more complex statistical elaborations and actualization of multi-period relationships. Several indicators have already been given in the previous sections, in particular the measure of economies of scale in water supply (equation (2)) and the elasticity of supply and demand (equation (4)), which have an explanatory meaning in themselves, apart from being used in more sophisticated analyses.

43

When a unit of water in a water basin is diverted from a source to a particular use, three basic things happen to it. First, a part is lost to the atmosphere because of evaporation or evapotranspiration. Second, part of the diverted water may drain to the sea, a deep canyon, or a similar sink where it cannot be captured and reused, in which case it is truly lost to the system. Otherwise, the drainage water flows back into a stream or to other surface and subsurface areas where it can be captured and reused as an additional source of supply. This water is not lost or wasted in physical terms (Seckler, 1996).

263



The first other key indicator that can be taken as an element in the objective function is water productivity. In principle, water productivity can be calculated for any sector or sub-sector of the economy and is equal to a sector’s dollar-value net benefits per unit of time, divided by water used measured in cubic meters per unit period of time. Water productivity can also be calculated at the basin or service level, in order to measure the overall impact of waster policies. Maximizing water productivity across sectors and services within a basin is a major policy objective. Thus, recalling that Zj represents the net benefits for user/sector j and that QWj is the quantity of water used up in the jth production process, water productivity PRW can be written as:

=

PR W

m

∑

j =1

Zj Q Wj

(15)

Other economic ratios come from water accounting techniques, which relate annual output to land and water, and provide the basis for comparison of irrigated agriculture, are: output per cropped area, Yca, output per unit irrigation supply, Yds, and output per unit water consumed, Ywc. Respectively, the following ratios express these relationships (Sakthivadivel et al., 1999):

Yca = Y Yds = Y Ywc = Y

A Vd Et

(16) (17) (18)

where Y is the output of the irrigated area in terms of gross dollar value of production at local prices, A is the sum of the irrigated areas (in ha) under crops during the time period of analysis, Vd and Et have the usual meaning of net volume of water diverted to a field, and evapotranspiration. For international comparison, it is useful to convert all values of production in a standardized term, in order to reflect differences in local prices and of tastes throughout the world. To obtain this value, equivalent yield is calculated based on local prices of the crops grown, compared with the local price of the predominant, locally grown, internationally traded base crop, such as wheat. The Standardized Gross Value of Production, SGVP, is equal to:

⎛ Pi ⎞⎟ SGVP = ⎜ A i Pi P world ⎜ crop ⎟ P b ⎠ ⎝

∑

(19)

where Yi is the yield of crop ith, Pi is its local price, Ai is the area cropped with crop ith, Pb is the local price of the base crop, and Pworld is the value of the base crop traded at average world market price. More detailed analyses can be performed when a comprehensive modeling framework is put in place. In this case, on the basis of the irrigation water demand functions developed in the previous sections, the profit per unit of water consumed at each demand site, PUWdm, and the same at the basin level, PUWb, can be calculated as follows (Rosegrant et al., 2000):

PUW dm =

PA dm

∑

WD dm , t − RF dm , t

∑ dm

PA dm

(20)

time

PUW

b

=

WDP

(21)

where PAdm is the net profit from irrigation given by equation (8) of section 5.3, whereas WDdm and RFdm are respectively water withdrawal and return flows from demand sites, and WDP is total irrigation water depleted, all taken from a water accounting matrix. Finally, some indicators can be useful in assessing the degree of cost recovery and financial sustainability of basin water systems, such as irrigation services. Two main indicators assume relevance: the Cost Recovery from Water Billing, RCR, and the Cost Recovery Rate, CRR (RUB, 2002). The first indicator relates the total revenues from billing, TR, to the estimated theoretical cost

264



of reaching sustainability of technical systems, Csts. The second indicator relate the total revenues, TR, minus the subsidy provided, S, to the total actual cost of providing the service, TC. In symbols:

⎞ × 100 RCR = ⎛⎜ TR ⎟ C sts ⎠ ⎝

(22)

CRR = TR − S

(23)

(

TC

) × 100

Policy-related water indicators and indices Indicators and indices for water policy-making are instruments of simplification in that they summarize large amounts of measurements to a simple and understandable form, in order to highlight the main characteristics of a system. Information is reduced to its elements, maintaining the crucial meaning fro the questions under consideration. Though the aggregation causes a loss of information, if the indicator is planned properly, the loss will not gravely deform the results. A fundamental different exists between indicators and indices (WWAP, 2003). An indicator, comprising a single data (a variable) or an output value from a set of data (aggregation of variables), describes a system or process such that it has significance beyond the face value of its components. It aims to communicate information on the system or process. The dominant criterion behind an indicator’s specification is scientific knowledge and judgment. An index is a mathematical aggregation of variables or indicators, often across different measurement units so that the result is dimensionless. An index aims to provide compact and targeted information for management and policy development. The problem of combining the individual components is overcome by scaling and weighting processes, which will reflect societal preferences. 44 A plenty of indicators exist for evaluating the effectiveness and impact of water policies. The focus here is put on a tool that uses the water accounting technique and poverty analysis to examine water use in relation to specific social goals. The Water-Poverty Accounting Framework (WPAF) is one such tools that gathers the different aspects of water management and use, in specific relation to poverty. The WPAF expands on the water accounting technique outlined in section 5.1, in order to account for how water is used to meet social and economic goals, in particular poverty alleviation. The approach has a bias towards water used for agriculture, sanitation and nutrition, since these are the most significant source of employment for the poor (Biltonen and Dalton, 2003).

Using the WPAF, water allocations required to meet different poverty dimensions can be analyzed for each specific use. The desired and actual situations are then compared to determine options for reallocating water to meet social goals. A set of indicators, based on current and target allocations, has been developed to show the efficiency of water use to meet different demands. There are two classes of indicators: adequacy ratios, which indicate how well either current or future needs are being met; and bias indicators, which show the bias of water allocations either toward or away from meeting certain social goals. The indicators show where surplus water is available for reallocation and where additional water is required to meet other goals. It is possible to compile a set of indicators for any area, country or 45 region. CONCLUSIONS AND PROPOSALS

44 45

For a comprehensive review, see WWAP (2003) and WaterStrategyMan Project (2004). For the details about the indicators, we point to the original work of Biltonen and Dalton (2003).

265



This concise review has tried to show how fundamental the economic analysis of water is, and could further be, in order to improve water use efficiency and water productivity. The discussion started from the fundamental reasons at the basis of water economics, going ahead through the major techniques developed to assess water supply costs and water demand values, with a focus to the agriculture sector, which is the main topic of the WASAMED project. The paper concludes for the superiority of the choice to push the accounting and modeling analyses of water resources, since they can generate a large amount of useful information and can be integrated into user-friendly policy tools. There are barriers to the effective use of integrated river basin management modeling, including informational, physical, and application barriers (Lee and Dinar, 1995). Insufficient data, limitations, and poor information about the cultural, social and political norms often limit the development of an effective planning model. Moreover, because basins are irregular and receive flows from multiple sources, difficulties arise when attempting to dived a basin into discrete, manageable subunits. Temporal and special variability also complicate matters. Application barriers are due to the fact that such models are usually formulated and applied to a particular area, they are finely calibrated to address specific problems, and any modifications can be undertaken with additional costs. However, models are by definition intended to abstract from reality, so the best model is the most adapted to available data and the most transparent in terms of social values embedded in the objective function. Trade-offs exist in model specification. But while an overly simple specification may yield insufficient information to address the problem or unreliable results, a more complex model, that would require extensive data collection, is not always a more preferred option then a simple, cost-effective modelling approach. However, given the demonstrated usefulness of integrated models, two complementary strategies might be suggested. On the one hand, regular data collection should be improved, to allow for the progressive introduction of more complex models. Agricultural extension services can be a useful tool for finding information on farm possibilities and constraints, in so helping specifying production functions to be used in estimations of water use values. Setting up a comprehensive Water Information System in each country, that uses environmental-economic accounting practices, is a key tool. This in developing countries may require foreign aid, in the form of training, technical assistance and financing of collection stations. On the other hand, it is necessary to refine and adapt existing tools, and create new models tailored on specific circumstances. Future directions in integrated modelling include the combination of Decision Support Systems (DSS) and Geographic Information Systems (GIS) into comprehensive Spatial Decision Support Systems (SDSS). DSS are interactive programs with a graphical interface, which embed simulation and optimization models to support users in problem solving. GIS offer a spatial representation of water resources systems using existing datasets. SDSS integrate spatial representations and modelling capacity into a single operational framework. Even the Water-Poverty Accounting Framework, referred to in section , 6.3, can be integrated into a geographical information system, with indicators separated into a range of categories and assigned graduated color codes. In this manner, maps can be constructed that demonstrate the current conditions as related to any of the preferred indicators. Reinforcing these trends would be a decisive step towards helping decision makers with problems that have a spatial dimension, such as water allocation. The combination of reliable datasets, robust models, maps, and statistical analysis components, in fact provides water planners with effective and comprehensive support for taking informed decisions.

REFERENCES Agudelo, J. I., 2001, “The economic valuation of water. Principles and methods”, Value of Water Research Report Series No. 5, IHE Delft, The Netherlands, August Allan, T., 1999, “Productive efficiency and allocative efficiency: why better water management may not solve the problem”, Agricultural Water Management, Volume 40, Issue 1, 1 March, Pages 7175

266



Billi, A., Meroz, and A. Quarto, 2004, “Water sector and institutional reform”, 3rd Workshop on Nonconventional Water Use, Cairo, Egypt, 7 – 11 December 2004 Biltonen, E., J. A. Dalton, 2003, A Water-Poverty Accounting Framework: Analyzing the WaterPoverty Link”, Water International, Vol. 28, No. 4, pp. 467-477, December Boscha, D. J., V. R. Eidmanb, and L. K. Oosthuizen, 1987, “A review of methods for evaluating the economic efficiency of irrigation”, Agricultural Water Management, Volume 12, Issue 3, April, Pages 231-245 Dinar, A., and J. Letey, 1996, Modeling economic management and policy issues of water in irrigated agriculture, Westport, Connecticut: Praeger Publishers Faisal, I. M., R. A. Young, and J. W. Warner, 1997, “Integrated Economic-Hydrologic Modelling for Groundwater Basin Management”, Water Resources Development, Vol. 13, No. 1, pp. 21-34 Garduño, H., and F. Arreguín-Cortés, eds., 1994, Efficient Water Use, Proceedings of the International Seminar on Efficient Water Use, Mexico, October, UNESCO Regional Office for Science and Technology for Latin America and the Caribbean, available at: http://www.unesco.org.uy/phi/libros/efficient_water/tapaefus.html Gibbons, D. C., 1986, The Economic value of Water, Resources from the Future, Washington D.C. Gleick, P., et al., 2003, Waste Not, Want Not: The Potential for Urban Water Conservation in California, Pacific Institute, November Grimble, R. J., 1999, “Economic instruments for improving water use efficiency: theory and practice”, Agricultural Water Management, Volume 40, Issue 1, 1 March, Pages 77-82 Keller, A., J. Keller, and D. Seckler, 1996, “Integrated Water Resource Systems: Theory and Policy Implications”, Research Report 3, International Water Management Institute, Colombo, Sri Lanka Lee, D. J., and A. Dinar, 1995, “Review of Integrated Approaches to River Basin Planning, Development, and Management”, World Bank Policy Research Working Paper, No. 1446, April Mac Gregor, J., S. Masirembu, R. Williams, and C. Munikasu, 2000, “Estimating the Economic Value of Water in Namibia”, paper presented at the 1st WARFSA/Waternet Symposium Sustainable Use of Water Resources, Maputo, 1-2 November 2000 McKinney, D. C., X. Cai, M. W. Rosegrant, C. Ringler, and C. A. Scott, 1999, “Modeling Water Resources Management at the Basin Level: Review and Future Directions”, SWIM Paper No. 6, International Water Management Institute, Colombo, Sri Lanka Mejías, P., C. Varela-Ortega, and G. Flichman, 2004, “Integrating agricultural policies and water policies under water supply and climate uncertainty”, Water Resources Research, Vol. 40, W07S03 Merrett, S., 1997, Introduction to the economics of water resources. An international perspective, Rowman & Littlefield Publishers Molden, D., and R. Sakthivadivel, 1999, “Water Accounting to Assess Use and Productivity of Water”, Water Resources Development, Vol. 15, No. 1-2, pp. 55-71 Morris, B L, Lawrence, A R L, Chilton, P J C, Adams, B, Calow R C and Klinck, B A., 2003, Groundwater and its Susceptibility to Degradation: A Global Assessment of the Problem and Options for Management, Early Warning and Assessment Report Series, RS. 03-3. United Nations Environment Programme, Nairobi, Kenya OECD, 2003, Improving Water Management. Recent OECD Experience, OECD Publications, Paris Ohlsson, L., and A. R. Turton, 2000, “The Turning of a Screw. Social Resource Scarcity as a BottleNeck in Adaptation to Water Scarcity”, ECD News, Swedish International Development Cooperation Agency (Sida), available at: http://www.edcnews.se/Reviews/Turningofascrew-1.html Palacio-Vélez, E., 1994, “Water Use Efficiency in Irrigation Districts”, in H. Garduño and F. ArreguínCortés, 1994, op. cit. Perry, C. J., M. Rock, and D. Seckler, “1997, “Water as an Economic Good: a Solution or a Problem?”, Research Report 14, International Water Management Institute, Colombo, Sri Lanka Reca, J., J. Roldán, M. Alcalde, R. López, and E. Camacho, 2001a, “Optimization model for water allocation in déficit irrigation systems. I. Description of the model”, Agricultural Water Management, Vol. 48, pp. 103-116

267



Reca, J., J. Roldán, M. Alcalde, R. López, and E. Camacho, 2001b, “Optimization model for water allocation in déficit irrigation systems. I. Application to the Bémbezar irrigation system”, Agricultural Water Management, Vol. 48, pp. 117-132 Ringler, C., J. von Braun, and M. K. Rosegrant, 2004, “Water Policy Analysis for the Mekong River Basin”, Water International, Vol. 29, No. 1, pp. 30-42 Rogers, P., R. Bhatia, A. Huber, 1998, “Water as a Social and Economic Good”, Technical Advisory Committee (TAC), Technical paper n 2, Global Water Partnership (GWP) Rosegrant, M. W., C. Ringler, D. C. McKinney, X. Cai, A. Keller and G. Donoso, 2000, “Integrated economic–hydrologic water modeling at the basin scale: the Maipo river basin”, Agricultural Economics, Volume 24, Issue 1, December, Pages 33-46 RUB, 2002, “Methodology Report on the Quantitative Analysis of Water Systems”, WaterStrategyMan Project, Deliverable 7, prepared by Ruhr-University Bochum, November Sakthivadivel, R., C. De Fraiture, D. J. Molden, C. Perry, and W. Kloezen, 1999, “Indicators of Land and Water Productivity in Irrigated Agriculture”, Water Resources Development, Vol. 15, No. ½, pp. 161-179 Seckler, D., 1996, “The New Era of Water Resources Management: From Dry to Wet Water Savings”, Research Report 1, International Water Management Institute, Colombo, Sri Lanka Spulber, N., and A. Sabbaghi, 1994, Economics of Water Resources: From Regulation to Privatization, Kluwer Academic Publishers Sullivan, C., 2002, “Calculating a Water Poverty Index”, World Development, Vol. 30, No. 7, pp. 1195–1210 Tate, D. M., 1994, “Principles of Water Use Efficiency”, in H. Garduño and F. Arreguín-Cortés, Efficient Water Use, Proceedings of the International Seminar on Efficient Water Use, Mexico, October Tsur, Y., and A. Dinar, 1995, “Efficiency and Equity Considerations in Pricing and Allocating Irrigation Water”, World Bank Working Paper, No. 1460, May Varela-Ortega, C., J. M. Sumpsi, A. Garrido, M. Blanco, and E. Iglesias, 1998, “Water pricing policies, public decision making and farmers' response: implications for water policy”, Agricultural Economics, Volume 19, Issues 1-2, September, Pages 193-202 Wang, H., C. Wang, J. Wang, and D. Qin, 2004, “Investigations into the Effects of Human Activities on the Hydrological Cycle in the Yellow River Basin”, Water International, Volume 29, Number 4, December, Pages 499–509 WaterStrategyMan Project, 2004, “Indicators and Indices for decision making in water resources management”, Newsletter, Issue 4, Jan-Mar Wichelns, D., 2002a, “An economic perspective on the potential gains from improvements in irrigation water management”, Agricultural Water Management, Vol. 52, pp. 233-248 Wichelns, D., 2002b, “Economic analysis of water allocation policies regarding Nile River water in Egypt“, Agricultural Water Management, Vol. 52, pp. 155-175 WWAP, 2003, World Water Development Report, World Water Assessment Programme, UNESCO. Young, R. A., 1996, “Measuring Economic Benefits for Water Investments and Policies”, World Bank Technical Paper No. 338, The World Bank, Washington, D.C. Zekri, S., and A. Dinar, 2003, “Welfare consequences of water supply alternatives in rural Tunisia”, Agricultural Economics, Vol. 28, pp. 1-12

268



PROPOSAL FOR THE INTEGRATION OF IRRIGATION EFFICIENCY AND AGRICULTURAL WATER PRODUCTIVITY

B. Blümling*, H. Yang**, C. Pahl-Wostl*** Institute of Environmental Systems Research, Barbarastraße 12, 49076 Osnabrück, Germany; [email protected] ** EAWAG, Dübendorf, Switzerland; [email protected] *** Institute of Environmental Systems Research, Barbarastraße 12, 49076 Osnabrück, Germany; [email protected] *

SUMMARY – In this paper, we provide a concept for the integration of the engineering and agronomic definitions of irrigation efficiency into the concept of Water Productivity. After “Water Productivity” has entered the water policy and research arena, there has been some confusion in its use and delineation from “Efficiency”. We will therefore first make a clear differentiation between the terms, and then actually integrate the different kinds of efficiency into what we call “Agricultural Water Productivity”. “Agricultural Water Productivity” then sets the boundaries within which efficiency indicates the smoothness of the water use process which itself is directed towards high Agricultural Water Productivity. The latter denotes at which points a process has to be efficient in order to get the highest overall value out of water. Applying this system perspective of Water Productivity to agriculture allows going beyond “yield” as the only output from irrigation water use, but considers different outputs with differing values. The conceptualisation of Agricultural Water Productivity provides a sound basis for a harmonized application of irrigation and water use efficiency and water productivity to decision making. Key words: Water Productivity; Agricultural Water Productivity; Classical Efficiency; Irrigation Efficiency.

INTRODUCTION The importance of increasing and securing food production for a growing world population, while at the same time limiting agricultural water use, has been extensively discussed among practitioners and researchers (see for example Rosegrant 1997; IFPRI 2001; FAO 2003; Qadir et al 2003; SIWIIWMI 2004). The debate for a long time focussed on “(agricultural) water use / irrigation efficiency” as the core concept to indicate the successfulness of water policy that aims at increasing the “crop per drop” ratio. In the course of the discourse, many researchers have made an appeal to change the perspective on and thereby modify the conceptualisation of dealing with water resources in agriculture (e.g. Carruthers et al 1997; Perry 1999; Gleick 2000; Molden et al 2001b; Postel 2003). Molden’s concept of “water productivity” was one response to this plea, and was added to the discussion in 1997. With this concept he framed the idea of a group of researchers who thought that “efficiency” underlies a “conceptual blindness” since, what is “waste” from the “efficiency” point of view may be used beneficially elsewhere in the hydrological system. When water is used, not all of it is lost but parts return to the system and may provide input to other uses.

269



There were also other attempts to label the idea which stands behind the “water productivity” definition by Molden. Carruthers et al in 1997 proposed a differentiation into “real” water savings and “paper” water savings for the same phenomenon. Seckler (1996) previously advocated for the terminology of “dry” and “wet” water savings. Keller et al (1996; 1998) put forward the term of “effective irrigation efficiency”. Taking a look at literature, “water productivity” is the concept which some researchers have tried to improve or to apply to their research and therewith seems to be accepted by the community (see for example Sakthivadivel et al 1999; Renault et al 2000; Droogers & Kite 2001; Hamdy et al 2003; Peranginangin et al 2004; Dong et al 2004; Bessembinder et al 2004; Ahmad et al 2004; Kijne et al (without date); Kijne et al. 2004). At the same time, there seems to be a differing use of the term Water Productivity. For some, it is just a new name for what was “originally referred to in literature as ‘water use efficiency’” (Zwart & Bastiaanssen 2004: 116 in their thorough literature review). But the two terms do have different 46 underlying etymologies and concepts, and the one may not just be re-named into the other . We will come to this point under the paragraph about classical efficiency in relation to water productivity. Not only that Water Productivity is used in two different ways, but one may also ask what the one or other meaning may add? Why introducing a term like “Water Productivity” when there are already “irrigation efficiency” and “water use efficiency” which are widely applied? In parallel to Water Productivity entering the water policy and research arena, “efficiency” remains the indicative term to other researchers for the evaluation of water use in agriculture (see for example Skaggs and Samani 2005, Rosenzweig et al 2004, Mo et al 2004; Hatfield et al. 2001) 47. There hence seems to be a need for clarification. We will in the following make a clear differentiation between irrigation and water use efficiency on the one hand and Water Productivity on the other. We then will integrate efficiency into Water Productivity. The system perspective of Water Productivity as defined by Molden is proven to be very useful for meeting the new challenges in agricultural water policy, and applying the concept to irrigation water use allows going beyond “yield” as the only output from irrigation water use. This “Agricultural Water Productivity”, as we call it, sets the boundaries within which efficiency indicates the smoothness of the water use process which itself is directed towards high Water Productivity. By integrating efficiency into the concept of Water Productivity as defined by Molden, we add to the latter and at the same time clarify the difference between Water Productivity and irrigation and water use efficiency. Whereas Molden focuses on definitions of Water Productivity depending on the scales of investigation and their interlinkages (Molden et al 2003), we remain on the scale of a field and make Water Productivity, through the integration of efficiency, a more operational term. Our incorporation of the system perspective refers to a single water user, whereas Molden focuses on various system users and their interrelations on different scales. We will start with outlining three kinds of irrigation efficiency, the so-called “Classical Efficiency”, Water Use Efficiency and Irrigation Water Use Efficiency. We then contrast one of it, “Classical Efficiency, with the concept of Water Productivity. Since the upcoming of Water Productivity to some extent can be regarded as a reaction to “Classical Efficiency”, we will focus on the comparison of these two concepts. We nevertheless will also discuss the relation of Water Use Efficiency to Water Productivity. After a detailed description of the concept of Water Productivity, we come to its modification for irrigation water use. We will show that the different concepts of efficiency and that of Agricultural Water Productivity do not compete against each other, but can be used synergistically. This conception, as we will see, provides a very helpful link to policy making. 46 It is moreover interesting that a definition which previously had been named out of an engineering terminology, now should be replaced by a term out of an economic context. This may also allow for a discussion about who takes the lead of discourse in the domain. 47 According to the advocates of the water productivity approach, when speculating on the reasons for the persistence of what they call “classical efficiency” (CE), they regard it as a matter of training of current irrigation practitioners, the orientation of their professional interests and positions around CE, and also the institutional establishment around CE, as well as the fact that “CE serves the interest of other professions and groups as well. Economists can use low CE as justification for pricing water and water markets; and environmentalists can use it in their battles against large dams, transbasin diversions and other water-development projects.” (Seckler et al 2002: 47f)

270



DEFINITION OF TERMS - “IRRIGATION EFFICIENCY” AND “WATER PRODUCTIVITY” Engineers as well as agronomists use the term “irrigation efficiency”, but denoting two different meanings. The concept of Water Productivity can be seen as a response and critique to the definition by irrigation engineers and practitioners, which is often referred to as “classical efficiency” or “CE” (see Wichelns 2002; Seckler et al 2002). Some “agronomic definitions” seem to be very close to Water Productivity, but, as we will see, only in their parameters, not in their conceptualisation.

Two Terms of Efficiency The “classical” irrigation efficiency, as for example defined by ICID, at each stage of an irrigation 48 scheme relates the volume of incoming water to the volume coming out of the scheme . For the whole irrigation scheme, the amount of water stored in the root zone is related to the amount of water delivered for irrigation. Across different scales, “irrigation efficiency is defined for: irrigation conveyance (farm supply/main system supply), farm irrigation efficiency (field application/farm supply), field irrigation efficiency 49 (rootzone storage/field application), and overall irrigation efficiency (rootzone storage/main system supply)” (Kassam, Smith 2001: 15). To its users, the term has an operational function; it is a management ratio which can be taken for management decision support 50. For agronomists, there are various definitions of irrigation efficiency. Basically, efficiency relates the agricultural yield to water consumption. Therefore, whatever may be integrated into the definition of efficiency as used by agronomists, at the core of it lays “(Crop) Water Use Efficiency”. It is the ratio of crop yield to the water consumed to produce the yield, that is, evapotranspiration or, better, transpiration.

This definition is still widely used (Viets 1962; Hatfield et al 2001; Kang et al 2002 51; Yuan et al 2003 52; Zhang et al 2004 53). The only difference in its use lays in the framing of the nominator, whether yield may be crop dry matter (either total biomass or aboveground dry matter), the economic yield (including the crop price), etc. For the denominator, evapotranspiration is often taken, since the calculation of transpiration is considered difficult. Water Use Efficiency varies with crop species, available energy from sunlight, atmospheric pressure, etc. This definition hence expresses the property of a plant at a certain location, that is, the characteristic of a crop, and therewith is much related to plant breeding. Water Use Efficiency in its strictest sense does not take into account the role of irrigation. It hence is a genuinely agronomic term. 48

The International Commission on Irrigation and Drainage uses about the same definition: “The water used in irrigation passes through successive stages of storages (possible), conveyed up to the head of the area, distributed among the fields and finally applied to each field. During each stage, there is loss of water and the volume coming out is less than the volume entering. The efficiency at each stage is equal to the ratio: volume coming out/volume entering ” (ICID 2000) 49 which often is also referred to as “application efficiency” 50 We thank M.G. Bos for this remark, who also emphasized that efficiency was never intended to carry out, and therewith did not include parameters of a water balance. It would be a term which currently would be disused, “ratio” may taking the lead. 51 “WUE, defined as the ratio between grain yield and total growing season evapotranspiration” (Kang et al 2002: 204) 52 “Water use efficiency (WUE) is the relation between yield or dry matter produced and the quantity of water consumed” (Yuan et al 2003: 164) 53 “Water use efficiency is generally defined in agronomy as the ratio of crop yield (usually economic yield) to water used to produce the yield” (Zhang et al 2004: 113) They set the yield in relation to evapotranspiration.

271



Some agronomists include opportunity costs into the definition of what they then call “economic efficiency”, focussing on financial aspects of irrigation. “Economic efficiency of irrigation water use refers to the economic benefits and costs of agricultural water use in agricultural production. As such, it includes the cost of water delivery, the opportunity cost of irrigation and drainage activities, and potential third-party effects or negative (and positive) externalities [...]. Economic efficiency can be expressed in various forms, for example, as total net benefit, as net benefit per unit of water, or per unit of crop area and its broader approach compared to physical efficiency [which is here referred to Classical Efficiency, the authors] allows an analysis of private and social costs and benefits“ (Cai et al 2001: 6). Since irrigation plays a role only as a cost factor, “economic efficiency” may go too much into the direction of economics. The definition of “Irrigation Water Use Efficiency” by Howell seems to be more suitable from an agronomic perspective. It specifies the above Water Use Efficiency in order to take the benefits of irrigation into account. “Irrigation Water Use Efficiency” (Howell 2003: 471; Howell 2001: 285) is calculated by first subtracting the yield which would be achieved without irrigation from the yield which is produced with the help of irrigation. The same applies for the water fraction in the denominator where evapotranspiration of precipitation input during the growing season is subtracted from evapotranspiration of irrigation water input.

This definition of irrigation efficiency incorporates agronomic aspects of plant characteristics as well as the management of irrigation (e.g. irrigation scheduling or irrigation system). When further referring to the agronomic definition of irrigation efficiency, we will refer to this Irrigation Water Use Efficiency. We will understand the “genuinely” agronomic Water Use Efficiency (WUE) as integrated in Irrigation Water Use Efficiency (IWUE). Whereas WUE considers the variation in the yield of different species of a crop, or even among different crops, under the same input of water, Irrigation Water Use Efficiency looks at the variance of the yield of the same specie / crop under different applications of water. This integration will be of importance when we further below will set the two terms in the frame of Agricultural Water Productivity. Irrigation Water Use Efficiency and Classical Efficiency are relating to “Water Productivity” in different ways, which we will see in the following section.

Efficiency in Relation to Water Productivity As mentioned before, there exists mix-up in the naming of Water Use Efficiency as Water Productivity, and we will first shortly address this question under the chapter about Water Use Efficiency in relation to Water Productivity. But since Classical Efficiency is the definition which is criticized by the advocates of Water Productivity, we will in the following focus on this discussion of distinguishing Water Productivity and Classical Efficiency. Irrigation Water Use Efficiency actually to some extent has the same “conceptual blindness” as CE in that it focuses on the in- and outputs of agricultural production only. The critique under the chapter about Classical Efficiency in relation to Water Productivity hence does not only address Classical Efficiency but implicitly is also related to WUE / IWUE. (Irrigation) Water Use Efficiency in Relation to Water Productivity Some of the confusion in the definition of Water Productivity comes from the fact that researchers use it interchangeably with Water Use Efficiency (see before Zwart & Bastiaanssen 2004). Belder et al (2004) define Water Productivity as “the amount of harvested product per unit water use” (Belder et al 2004: 170), and also Cantero-Martinez (2003) talk about the “water productivity of barley”, when actually referring to Water Use Efficiency in their article. Cabangon et al (2004) differentiate between “irrigation water productivity” (WPI, kg/m3), and “calculated from grain yield divided by the volume of

272



irrigation water input during the crop season” (Cabangon et al 2004: 197) and “water productivity with respect to the total water input (WPI+R, kg/m3), the denominator was the total water input (I + R)” (ibid). This shows that what previously was, and actually still is, defined as Water Use Efficiency, has been renamed in “Water Productivity”. Though the concept of yield per defined unit of water is very useful, the double naming of Water Productivity in two different, yet related scientific communities suggests to return to Water Use Efficiency when it comes to the relation of yield to water consumption, since this term for a long time has been proved useful, as well as the renaming in Water Productivity actually does not add much to it. Applying “Water Productivity” as defined here in contrast will enhance the concept of irrigation water use and agricultural produce. This is why we encourage returning to Water Use Efficiency and leaving Water Productivity for taking a new perspective on irrigation water use. Classical Efficiency In Relation to Water Productivity Taking a step back and looking at the semantic meaning of the term, “productivity” focuses on the 54 result of an action. Being productive implies “yielding or furnishing results” , while the term at the same time has a positive connotation of “resulting in or providing a large amount or supply of something” 55. “Productivity” is a term which is used in an economic context where it means “The rate at which goods or services are produced especially output per unit of labor” 56. In being defined as the "rate of output per unit" 57, that is by referring to a unit, productivity incorporates system boundaries in its definition, as well as it has the notion of getting the most out of a defined limited base, that is the notion of (profit-) maximization. In summary, productivity is result-orientated and focuses on the maximization of output based on a certain unit of input. “Efficiency” as the quality of being efficient may also be expressed as being “productive without waste” 58 or “acting or producing effectively with a minimum of waste, expense, or unnecessary effort” 59. In this sense, the term focuses on the quality of a process, like using water well without wasting any. Efficiency is expressed in a ratio, that is, as “the ratio of the effective or useful output to the total input in any system” 60. In irrigation, Classical Efficiency stems from an ideology of technological process optimization. For irrigation engineers, since “‘efficiency’ is per definition related to comparing input with output during a given process, the same units for input and output should be applied” (van Dam and Malik 2003: 13). Focussing on making the process within the system smoother (or: less wasteful), Classical Efficiency therefore does not necessarily have a fixed reference unit like “water productivity” which relates the yield to, for instance, a cubic meter of water (output of a system per unit of input 61). At any scale of Classical Efficiency, one may increase the ratio by means of technological innovation and better irrigation practices, but, and this is the critique by those using “water productivity”, without referring to system boundaries (see Molden 1997: 2; Perry 54

Webster’s Revised Unabridged Dictionary, © 1996, 1998 MICRA, Inc.; http://dictionary.reference.com/search?q=productive, viewed December 2004 55 Cambridge Advanced Learner's Dictionary, Cambridge University Press 2004, http://dictionary.cambridge.org/define.asp?key=63151&dict=CALD, viewed December 2004 see also for “productive”: “producing or capable of producing (especially abundantly); ‘productive farmland’; ‘his productive years’; ‘a productive collaboration’ [...] marked by great fruitfulness; ‘fertile farmland’” WordNet ® 2.0, © 2003 Princeton University, http://dictionary.reference.com/search?q=productive, viewed December 2004 56 The American Heritage® Dictionary of the English Language: Fourth Edition, 2000. http://www.bartleby.com/61/12/P0581200.html, viewed December 2004; see also: “The ratio of the quantity and quality of units produced to the labor per unit of time.” Ultralingua.Net. http://www.ultralingua.net/index.html?service=ee&text=productivity, viewed December 2004 57 Online Etymology Dictionary, © November 2001 Douglas Harper, http://www.etymonline.com/index.php?term=productive, viewed December 2004 58 Merriam-Webster Online Dictionary, viewed December 2004 http://www.m-w.com/cgibin/ dictionary?book=Dictionary&va=efficiency, viewed December 2004 59 The American Heritage Dictionary of the English Language: Fourth Edition. 2000 http://www.bartleby.com/61/98/E0049800.html, viewed December 2004 60 The American Heritage Dictionary of the English Language: Fourth Edition. 2000. http://www.bartleby.com/61/95/E0049500.html, viewed December 2004 61 see also: “productivity, in economics, the output of any aspect of production per unit of input. […] Output can be measured in output per acre for land, per hour for labor, or as a yearly percent...” Encyclopedia.com © 2004, http://www.encyclopedia.com/searchpool.asp?target="productivity", viewed December 2004

273



1999: 46f). An increase of efficiency simply implies more total water savings within the system. The expansion of the system boundaries, the irrigation scheme, that is, the expansion of efficient irrigation, can only be welcomed. Increasing efficiency in some cases then may even lead to the overexploitation of the resource. A common example for this is that farmers, when increasing their application efficiency by utilising irrigation efficient technologies, most likely extend their irrigated land surface and therewith the overall water use 62. The main differences between Water Productivity and Classical Efficiency are provided in Table 1. In general, Water Productivity takes the hydrological system as a reference unit to set the system boundaries, whereas Classical Efficiency refers to the irrigation scheme, the infrastructure, as the system boundaries, within which efficiency shall be increased. Table 1: Comparison of Water Productivity and Classical Efficiency

Thus, efficiency as “being able to function without wasting resources” may not be a concept integrative enough for dealing with the potential conflicts around scarce water resources, since actually, what is “left over” may be productively used by other stakeholders. “Water Productivity” then again seems to be a suitable terminology in the discussion of how much food or “value” may be secured based on limited water resources. It underlies a more integrative view on water resource use.

THE IDEA BEHIND WATER PRODUCTIVITY Coming from the semantics of “efficiency” and “productivity”, we now want to present Molden’s (1997) concept of Water Productivity and dwell on two points which we consider important. On the one hand, this is the system perspective (see chapter Differentiation of a River Basin’s In and Outflows) which allows differentiating diverse “products” out of irrigation water use. The latter then

62

(Perry 1999: 48) gives an example of a “water scarce country in the Middle East [in which] on-farm investments were made to increase measured ‘efficiency’ from 40 – 50 % to 60 – 70 %, releasing water for further expansion of the irrigated area. Measurements to date show that the improved technology results in increased crop yields and increased water consumption – a direct confirmation of the many existing studies showing the positive relationship between yield and evapotranspiration, but not the hoped-for saving in water!”

274



have to be valuated out of the context of a river basin’s water scarcity and water use situation in the chapter Getting the Highest Value Out of Water. Differentiation of a River Basin’s In- and Outflows The Water Productivity terminology was developed in parallel with the emergence of the “IWMI water resources paradigm” (see Perry 1999) 63. The IWMI-paradigm states, among others, that water is not lost to a larger system though it may be lost to a smaller system. As Seckler et al put it: “One of the cardinal features of water use is that, when water is used, not all of it is ‘used up’. Most of the water remains in the hydrological system, where it is available for reuse or recycling. As water is recycled through the hydrological system, the efficiency of use increases. Thus, while every part of the system may be at low levels of water-use efficiency, the system as a whole can be at high levels of efficiency” (Seckler et al 2002: 37f). Molden and de Fraiture give the following example: “In some cases, when 18” irrigation efficiency is improved downstream users (often the environment) can actually get less water because water gained from farm-level efficiency increases is used upstream. In southern Sri Lanka, cement lining of canals led to reduced groundwater recharge and consequently several shallow drinking-water wells dried out [...]. These shallow wells provide better quality drinking water than fluoride-laden deep wells in the area“ (Molden and de Fraiture 2004: 9). This more integrative view on water use as being situated in an overall context of a river basin certainly is a main improvement by the concept of Water Productivity. The idea actually had been 64 existing for some time (see for example Palacios-Vélez 1994 ), but had not been formulated into a concept. It results in new ways of assessing water use: water accounts are proposed across sectors within a defined hydrological unit, the “receipts” and the “outgoings” of the balance being the inflows to and the outflows from the water body. On the sides of the outflows of a water use, according to Molden (Molden 1997: 7), Classical Efficiency as well as the agronomic definition of irrigation efficiency would only take into consideration water which is lost to the system through transpiration or evapotranspiration during the growth stages of plants. “Water Productivity”, on the other hand, would in addition integrate the water fractions to the outflows which, though even allocated to irrigation, are not consumed by the crop. The conventional definitions of irrigation efficiency would consider this part as something which “is left” when irrigation efficiency would not be high, that is, as a loss for the system. But according to the Water Productivity concept, water is only then lost from the system when it e.g. is deteriorated in its quality, flows to saline sinks or evaporates into the air. The output of a water use, the “left over” according to the efficiency definition, thus is much more differentiated: All fractions of water which deplete from the irrigation site should be integrated; no matter whether they further may render benefits or not. Looking at the inflows to a system, the critique addresses the same “conceptual blindness”. Seckler et al (Seckler et al 2002: 43) criticize that “in all the definitions of efficiency up to this point, precipitation only enters the analysis as effective precipitation (Pe). The difference between total precipitation (P) and Pe(P-Pe) – the amount of ‘ineffective precipitation’, as it were – is lost; it simply vanishes from the system, much like the ‘water losses’ in CE. This is unacceptable in terms of the water balance of the hydrological system as a whole”. The Water Productivity concept hence integrates different kinds of in- and outflows into the water use balance for a defined hydrological unit. Molden structures the flows into a water flow diagram which is reproduced in the illustration below (Fig. 1). We will shortly explain each fraction. Available water is the amount of water available to a service or use, which is equal to the inflow to the system less the share of outflow which is committed to other uses. The inflow is split into gross and net inflow. Gross inflow is defined as the total amount of water flowing into the system (precipitation, surface and subsurface inflows), whereas net inflow includes changes in the storage of 63

See also (van Dam and Malik 2003: 13) “The International Water Management Institute (IWMI) has started a strong lobby to change the nomenclature from water use efficiency into water productivity, which is now also followed by other Consultative Group on International Agricultural Research (CGIAR) institutes and the Food and Agricultural Organization of the United Nations (FAO). 64 Palacios-Vélez (1994) who states in a talk at a seminar held in 1991: “In many cases, however, part of that water can be reused, either in the same system or downstream in another system (…) when considering actions to improve water use efficiency, care must be taken that such actions do not have harmful effects in other parts of the system”

275



the system, that is: net inflow adds water to the gross inflow if water is removed from the storage, or it subtracts water from the gross inflow if water is added to the storage. Depleted water comprises that share of water of the net inflow which becomes unavailable for further use by the system. Molden distinguishes between process depletion and non-process depletion. The former refers to water which gets unavailable for further use in the system during its processing for the production of a certain good (e.g. when it comes to irrigation water use: transpiration during plant growth and water incorporated into plant tissues). Non-process depletion comprises depletion of water from the system without fulfilling a specified use (e.g. evaporation of water from the soil and free water surfaces; water flowing into the sea or into saline groundwater which makes it further unavailable to the system, or water as much polluted that it is not usable anymore). Non-process depletion is further subdivided into beneficial or non-beneficial. A beneficial non-process depletion may be that of fruit trees consuming irrigation water. A non-beneficial non-process depletion may comprise water which is lost to sinks as well as water rendered unusable because of pollution. But it non-beneficial non-process depletion can also be caused by weed which is using up water for evapotranspiration. Hereby, the consideration of how beneficial the depletion may be is defined by the stakeholders in the system. If stakeholders may find out that a plant which was previously considered a weed has beneficial uses in their agriculture or as an herb, the water depletion may then be considered as beneficial. The water still remains defined as “non-process depleted” since depletion by these plants was not the main reason why water was diverted from the system.

Fig. 1: Water Flow Balance (after Molden 1997: 5) The outflow is additionally split into committed and uncommitted outflow. This distinction is important for the integration of the context into the evaluation of the water productivity of a water use. Committed outflow comprises the fraction of water which is allocated to further uses in the system. Uncommitted outflow is further divided into utilizable and non-utilizable. Outflow is utilizable if existing facilities or the improved management could make further use of it but actually doesn’t. Non-utilizable

276



uncommitted outflow then is the fraction which leaves the system since the facilities in any case could not capture it for further use. Water input to, e.g. an irrigation scheme, thus is considered to have many outflows, and not only the one to which it is allocated. Several products out of an allocated water input result, and these products have to be given a value. Getting the Highest Value Out of Water Since the term of productivity concentrates on what may result out of a water use, the labelling of this output is of additional importance. It opens up the discussion about what the value of a produce actually is to different stakeholders, like for different sectoral products that originate from the same source of water such as industrial products, bird habitats or tourism. In Classical Efficiency, since in- and output are of the same entities, no further value would be given to the output. The result of Classical Efficiency is always that the processing is more or less efficient. In Irrigation Water Use Efficiency, the value of agricultural produce of course differs for, e.g. farmers and the government. But the comparison of the value remains within the sector. For Water Productivity, in its broadest sense, an increase “means obtaining more value from each drop of water—whether it is used for agriculture, industry or the environment. Improving agricultural Water Productivity generally refers to increasing crop yield or economic value per unit of water delivered or depleted“(Molden, de Fraiture 2004: 9). But even within one sector, there are different understandings of what the value may be. Should agricultural water use be set into relation with economic values, should Water Productivity be a nutritional concept indicating how much nutritional value is produced out of a certain amount of water (see Renault, Wallender 2000), or should one simply refer to the yield, without assigning a nutritional or economic value? Seckler et al (2002) decide this question by making three distinctions of the Water Productivity terminology. “Pure physical productivity” would be defined as “the quantity of the product divided by the quantity of AWS [available water supply, the -3 authors], diverted water or depleted water, expressed as kg m ” (Seckler et al 2002: 46). “Economic productivity” would be the net present value of the product divided by the net present value of the amount of available water supply, or the water which is diverted or depleted, which can be defined in terms of its value, or opportunity cost, in the highest alternative use 20(Seckler et al 2002: 47). And as a “hybrid” definition, „combined physical and economic productivity is defined in terms of the net present value (NPV) of the product divided by the amount of water diverted or depleted. Thus, the quantity of the product is productivity times the amount of AWS or water depleted“ (Seckler et al 2002: 47). According to Kijne et al, the question of the output of Water Productivity can be dealt with flexibly. “Within one context of water productivity (physical or economic), the choice of the denominator (depleted or diverted water) may vary with the objectives and domain of interest of the study” (Kijne et al 2003: 5). As we understand it here, the choice of the denominator will be subject to the interests and values of the respective groups or organisations having a stake in water use, not so much of the study itself. If the value which different stakeholders denote to certain agricultural products differs, new issues of conflicts of interest enter the water policy arena.

THE APPLICATION OF WATER PRODUCTIVITY TO IRRIGATION WATER USE The concept of Water Productivity up until now has been applied to whole river basins (see, for example, Molden et al 2001a, Peranginangin et al 2004) but not to the detailed analysis of a single user’s Water Productivity. But the systematisation of a river basin’s Water Productivity is also valuable for the single water use irrigation. As we will see, Agricultural Water Productivity proves to be very useful to encompass different kinds of benefits out of irrigation water use under the respective natural resources conditions. We will in the following make Modifications in the theoretical concept of Water Productivity for its applicability to irrigation water use. A general adjustment addresses the differentiation of system

20

“water used in one place has an opportunity cost in terms of the value of its use in another place within the system. The concepts of efficiency and productivity need to reflect the values of all the uses and alternative uses within the system.” (Seckler et al 2002: 45)

277



flows, while the later integration of the two kinds of efficiency into the concept of Water Productivity improves it considerably for its application to irrigation water use.

Adding “Non-Beneficial Process Depletion” To The Water Balance For The Case of Irrigation Water Use Whereas Molden splits the fraction “non-process depletion” further into “non-beneficial” and “beneficial” in his definition of water productivity, the fraction “process depletion” is not. “Process depletion” – in the context of irrigation water use – for us describes the water rendered unusable to the system in the process of crop growth. As soon as water is used by other processes or leaves the field and then is used by other processes, it is rendered to the fraction of non-process uses. From this perspective, process depletion is the minuend for the calculation of non-process depletion. After the subtraction of non-process-depletion, water is returned to the fractions of committed or uncommitted outflow. Under “beneficial process depletion”, we then understand the portion of water that is lost to the system because of a special function it fulfils, that is, in the case of irrigation water use, the water transpired by the crop. “Non-beneficial process depletion” would comprise the fraction of water which e.g. evaporates from the soil surface. Figure 2 illustrates the application of this water productivity systematization to the farm level. On this scale, “available water” defines the Gross Inflow. Inflows at the field level are irrigation application, precipitation, subsurface contributions, and surface seepage flows. The “storage change” in the hydrological system is expressed in soil moisture change in the active root zone. Beneficial process depletion at the field level is set equal with crop transpiration. Non-beneficial process depletion comprises the fraction that evaporates from the soil surface, or water rendered unusable due to the degradation of quality. Non-beneficial non-process depletion for example comprises weed evapotranspiration, beneficial non-process depletion comprises the evapotranspiration from useful plants.

278



Fig. 2: Water flow balance for irrigation water use 21 The subdivision of water fractions is not necessarily something new for irrigation water use. The American Society of Civil Engineer’s On-Farm Irrigation Committee in 1978 defined irrigation efficiency as the ratio of the volume of water which is beneficially used to the volume of irrigation water applied. Beneficial uses would e.g. include crop evapotranspiration, deep percolation for salt control, crop cooling, frost control, or would take place in combination with pesticide or fertilizer applications. The denominator also in this case represented the total volume (which means beneficial as well as non-beneficial uses) of irrigation water. By extending the range of beneficial uses, efficiency still remains high though the water is not used for transpiration only. Still, the definition does not see agriculture as embedded in a context of other water users and therewith does not allow for the valuation of the water fractions.

Problems With The Differentiation of Beneficial / Non-Beneficial Process Depletion Whereas evaporation and transpiration in this conceptualisation are indicated as the fractions “non-beneficial process depletion” and “beneficial process depletion”, respectively, they generally are integrated into “evapotranspiration”, that is, a crop is considered together with certain management practices under which a certain amount of water evapotranspirates. But we need to make a differentiation here since it is important to know how much water the crop itself takes to grow (transpiration), to then examine in how far soil and water conservation practices may change the evaporation of water from the soil. Different human decision making processes and activities are linked to the respective depletion fraction, and, as we will see, efficiency and water productivity are linked to them in different ways. Crop transpiration basically is a result of plant breeding, and on the farm level a consequence of crop and species selection. A change in the amount of transpiration more likely requires making decisions about which crop to grow and which species to select (since transpiration can change across the species of a crop with their respective production-biomass ratios, the length of the growing season, etc.). A farmer here basically has to decide whether he wants to grow a crop which he can use for his livelihood or not, whether he wants to take the risk of producing it (which is also linked to the selection of certain species of a crop), whether he wants to take the time to manage it, whether it is easy to sell etc. Since transpiration stands for how much water a crop needs to produce yield, it can 22 be indicated by Water Use Efficiency as defined before . “Evaporation is always a component related to crop specific growth, tillage and water management practices” Zwart, Bastiaanssen 2004: 116), thus, a reduction in evaporation requires an alteration in management practices, and here, a farmer most likely has to spend more time on agriculture, or invest in infrastructure to reduce this actual water loss. The surplus value out of the management then likely is to play a role if a farmer is a main-income farmer, but maybe not so much if he is a sideincome farmer. Evaporation then is a side-effect / -loss during the course of a water use, that is, nonbeneficial process depletion occurs in connection with an allocation of water to a water use 23. The potential amount of water loss from soil evaporation may best be indicated with a ratio of evaporation to (potential) evapotranspiration (E/ET).

21

terms taken from Molden 1997 and Kijne et al., 2003 Actually, also here exists confusion of terms. What we call “Water Use Efficiency” is also labelled “Transpiration Efficiency” by researchers of a biological science / plant breeding community (Byrd 1997; Turner 2004; Condon et al 2004). To them, Transpiration Efficiency is the the weight of dry matter or biomass produced per unit of water transpired. Since this paper addresses researchers dealing with irrigation, we will stick to Water Use Efficiency as yield per water consumed by transpiration. 23 Precipitation of course does not happen for the sake of watering the crop. Still, non-beneficial process depletion also in the case of precipitation input describes the part of the rain which evaporates from the soil in the proximity of the crop after a precipitation event. It is lost to further use, though it could have been captured by soil and water conservation methods. 22

279



To support our argument for a separation of evaporation and transpiration despite the common integration into evapotranspiration, we will give examples of Zwart and Bastiaanssen’s thorough literature search by which the variety of evapotranspiration values depending on crop management practices should get clear 24. They found out that the variability of the yield for actual crop evapotranspiration of wheat ranges between 0.6 and 1.7 kg m-3 25. The values with the most efficient water use were found by Jin et al (1999) where the application of manure allowed for higher production and straw mulching again improved soil water and soil temperature conditions, reducing evaporation 26. The variability of cotton lint yield again ranged between 0.14 to 0.33 kg m-3. The best values were found in China and Israel, and for China, they were the result of experiments in which cotton was planted in furrows and the soil was covered with plastic leaving holes for the infiltration near the plants (see Jin et al. 1999). By this method, soil evaporation was reduced and the soil water status of the root zone was improved. From these results we conclude that a differentiation into evaporation and transpiration makes sense in two regards: the differentiation shows in how far there is scope for a reduction of the nonbeneficial process-depletion fraction, as well as it provides the opportunity to denote which kind of action the respective farmers should take in order to achieve this reduction, may it be by changing to a different crop (in case of a high transpiration), or by investment in irrigation technologies or effort for crop-, water- or soil-management practices in case of a naturally given high ratio of evaporation.

Integration of Terms In a Concept of “Agricultural Water Productivity” As stated above, irrigation efficiency as a single indicator for water use may not respond to contemporary requirements of harmonized water use planning in water scarce river basins. But the fact that the two terms of irrigation efficiency may not match current needs does not mean that they are not important. In fact, Water Productivity can set the boundaries within which efficiency indicates the smoothness of the process which itself is directed towards high Water Productivity. Water Productivity then denotes at which points a process has to be efficient in order to get the highest overall value out of water. Setting this benchmark does not so much orientate at agriculture, but looks at the context of water use in order to evaluate the respective Water Productivity of agriculture (see Figure 3). Setting the benchmark for Classical Efficiency defines the optimal relation of the water outflow fractions to each other. If we take the above mentioned example of Molden and de Fraiture (2004: 9), high Water Productivity under these conditions may imply that agriculture facilitates the percolation of irrigation water to groundwater, so that the fraction “process depletion” which leaves the system through drainage would be beneficial and its share in overall irrigation input should accordingly be high. The same is true for the case of excessive accumulation of salts in the soil. In arid areas, excess irrigation water is used to leach salt from the root zone. In this case flushing salts with additional water guarantees future fertility of the soil. From a CE perspective, efficiency would consequently be called low, but from a Water Productivity point of view, the “inefficiency” may be rather valuable. In cases in which water from agriculture would otherwise flow to sinks (like when agriculture is located close to the coast), high Water Productivity would imply increasing overall irrigation efficiency. The same holds true if excess irrigation water would leach soluble chemicals below the root zone, as well as if nitrate is carried below the root zone. Often, in arid and semi-arid areas, a gradual salinization occurs due to rising water tables where proper drainage has not been provided and too much water leached underground. This distribution of outflow water fractions from incoming irrigation water is determined by how the process of irrigation is managed. To indicate the smoothness of the process of irrigation water use at the field level, we set “evapotranspiration” and “drainage below the root zone” in relation to “irrigation 24

As mentioned above, Zwart and Bastiaanssen use “water productivity” interchangeably with water use efficiency, which in their research is measured as amount of yield per amount of evapotranspiration. We therefore describe the results listed in their article with “yield per unit evapotranspirated”. We take “actual evapotranspiration” since their listed examples deal with limited water supply. 25 The value of the FAO study by Doorenbos and Kassam ranged between 0.8 – 1.0 kg m-3 26 The “CWP for the experiment with straw mulching was 2.67 and 2.41 kg m-3 for a combination of straw mulching and manure” (Zwart, Bastiaanssen 2004: 118).

280



water input”. As can be seen from Figure 3, these ratios make up the “Evapotranspiration Fraction” and the “Drainage Fraction” respectively. A water balance would actually additionally incorporate soil moisture change and runoff. But we simply neglect soil moisture change which finds only expression in evapotranspiration, and, - if water content is beyond field capacity – drainage from the soil. Runoff would be left as a non-beneficial water fraction. In Figure 3, we additionally make the simplifying assumption that the irrigation water input does not leave the field so that the entire outflow is made up of process-depletion fractions and does not contain non-process depletion fractions. Additional to the Drainage and Evapotranspiration fraction, we think that the E/ET-ratio is an important indicator (see Figure 3). The ratio requires special attention since evaporation in every case is lost to the system without benefit. Especially if a lot of the irrigation water input is allocated to the Evapotranspiration fraction, the E/ET-ratio becomes important because it shows whether there is scope for reducing its size through preventing evaporation. The more the E/ET-ratio approaches 1, the smaller the scope for action will be to reduce evapotranspiration by minimizing evaporation.

Fig. 3. Integration of Water Productivity and Irrigation Efficiency These above efficiency indicators all relate to Classical Efficiency since they indicate the processing of different water fractions. Setting Water Use Efficiency then into the frame of Water Productivity allocates an output to one water fraction (transpiration) already, which would have to be given a value. Efficiency in this case denotes whether this value is achieved with a high or low input of water depending on the cultivar. Irrigation Water Use Efficiency then shows the importance of irrigation to the crop. Table 2 shows the main function of the above mentioned efficiency indicators. As Figure 3 as well as our above explanations show, the system perspective on irrigation water use allows for more products than only agricultural yield. Therewith, also the value of irrigation water

281



use does not only relate to agricultural produce. Setting the benchmark for the efficiency of water processing out of the context in which irrigation water use is embedded thus shall provide value to the different outflow water fractions, to then allot the size of the fractions within the given water input. We will come to the point of valuation in our discussion.

Table 2. Efficiency indicators Efficiency Indicator Evapotranspiration Fraction

Drainage Fraction

E/ET-ratio

Water Use Efficiency

Irrigation Water Use Efficiency

Function shows how much of the irrigation water input meets its primary purpose indicates how much of the irrigation water input drains below the root zone and can potentially be used for other purposes than agricultural production. Whether this water fraction can be named beneficial or non-beneficial depends on the context. shows how much of the evapotranspiration is actual loss since it evaporates without returning a benefit. The ratio also indicates if there is scope for water saving through soil and water management: if the ratio is low, evaporation in overall evapotranspiration is high, so that soil and water conservation may reduce the size of the evapotranspiration fraction, considerably. indicates how much yield a crop returns out of transpiration. It mainly depends on crop breeding (but also agronomic practices like fertilizer input, but this is beyond our subject). If WUE is low, as well as the E/ET-ratio is high, the scope for increasing the beneficial use of water would be rather limited. In this case, a change to other crop species or another crop may be advisable. shows how much of the total water consumption by a crop can be attributed to irrigation. It shows the dependency on irrigation by the cultivar.

DISCUSSION AND OUTLOOK In this conceptualisation of agricultural water productivity as well as irrigation and water use efficiency, water productivity directs the process of optimization, that is efficiency, towards the highest overall value of agricultural water use. Process optimization applies at the system flows of the water source (Classical Efficiency), as well as at the increase of output out of the water transpired ((Irrigation) Water Use Efficiency). Efficiency thus is integrated into the framework of water productivity and therewith relates to the boundary conditions of water use. A farmer can impact the above described efficiencies in different ways. In Figure 4, we have linked the respective efficiency indicators with two main activities: “Crop and crop species selection” will have an impact on how much a crop depends on irrigation (Irrigation Water Use Efficiency), as well as how much yield a farmer may gain out of transpiration (Water Use Efficiency). Soil and water management again will influence the size of evaporation in evapotranspiration (E/ET-ratio), and how

282



much of the irrigation input may be used for evapotranspiration and drainage respectively (Evapotranspiration Fraction and Drainage Fraction). Depending on the value of the agricultural produce in relation to the valuation of the drainage outflow from the field, farmers may take one or the other way in order to increase or decrease the size of the respective water fractions, and therewith raise their Agricultural Water Productivity. The valuation of the outflow will depend on the natural as well as the socio-economic conditions in which agricultural water use is embedded. The valuation of agricultural produce depends on whether it is sold or not. If it is not sold, the valuation becomes difficult since non-monetary aspects come into play. If it is sold on the market, it has a value to the consumers of a local (or global) market. If for example the outflow is valued high since otherwise the soil may turn saline and will make future agriculture less possible, the Drainage Fraction would have to increase. Soil and water management would have to be adjusted accordingly. If a farmer in the respective context values his produce high and does not want to change it, a comparison of the Water Use Efficiency of his crop as well as the E/ET-ratio can indicate whether agronomic practices may better be changed or the respective soil and water management, in order to allow for a big Drainage Fraction and achieve a high Agricultural Water Productivity. If a farmer did not have an interest to stay with a certain crop, he may also change to another with a lower Irrigation Water Use Efficiency so that more water is set free for drainage. The dimension of Agricultural Water Productivity then would depend on the price of the crop and the monetary value of the drainage.

Fig. 4. Field Management Practices and their impact on Agricultural Water Productivity This example also shows that an interesting point is who may set the value for the products of irrigation water use in a river basin. Different stakeholders will provide different meanings and hence values to outcomes, and often, these values may not be easily compared, or monetarised. Even

283



within a sector like agriculture, stakeholders will give different meanings and values to agricultural produce. Our conceptualisation of Agricultural Water Productivity makes it possible to operationalize this “integrative view” on irrigation water use as providing several outputs into an indication of how the different water use flows may achieve a high overall Agricultural Water Productivity. For this purpose, we made use of different kinds of efficiency indicators. We think that this conceptualisation provides a good basis for the integration of irrigation efficiency and water productivity to respond to current needs of dealing with limited water resources and increasing water demands from different sectors, as well as it provides links for policy makers to inspire the optimization of the process of water use in direction of high water productivity. The integration nevertheless still is in its conceptualisation phase. The most prevailing question is how to really integrate values provided by the river basin to the outflow water fraction in order to guide water use flows, and which implications this may have to agriculture.

REFERENCES Ahmad, M.D.; I. Masih, H. Turral (2004). Diagnostic analysis of spatial and temporal variations in crop water productivity: A field scale analysis of the rice-wheat cropping system of Punjab, Pakistan. Journal of Applied Irrigation Science 39.1: 43 – 63 Belder, P; J.H.J. Spiertz, B.A.M. Bouman, G. Lu, T.P. Tuong (2004). Nitrogen economy and water productivity of lowland rice under water-saving irrigation. Field Crops Research 93.2-3: 169-185 Bessembinder J.J.E.; P.A. Leffelaar; A.S. Dhindwal; T.C. Ponsioen (2004). Which crop and which drop, and the scope for improvement of water productivity. In: Agricultural Water Management (in press) Byrd, G.T. (1997): Transpiration efficiency, specific leaf weight, and mineral concentration in peanut and pearl millet. Crop Science 3.1 Cai, Ximing; Claudia Ringler; Mark W. Rosegrant (2001): Does efficient water management matter? Physical and economic efficiency of water use in the river basin. EPTD Discussion Paper No. 72. Washington, USA: IFPRI Cabangon, Romeo J.; To Phuc Tuong, Ernesto G. Castillo, Lang Xing Bao, Guoan Lu, Guangho Wang, Yuanlai Cui, Bas A. M. Bouman, Yuanhua Li, Chongde Chen, Jianzhang Wang (2004): Effect of irrigation method and N-fertilizer management on rice yield, water productivity and nutrient-use efficiencies in typical lowland rice conditions in China. Paddy and Water Environment 2: 195 – 206 Cantero-Martinez, C.; P. Angas, J. Lampurlanés (2003). Growth, yield and water productivity of barley.(Hordeum vulgare L.) affected by tillage and N fertilization in Mediterranean semiarid, rainfed.conditions of Spain. Field Crops Research 84: 341 – 357 Carruthers, Ian; Mark W. Rosegrant; David Seckler (1997). Irrigation and food security in the 21st.century. In: Irrigation and Drainage Systems 11: 83 – 101 Condon, A.G.; R.A. Richards; G.J. Rebetzke; G.D. Farquhar (2004): Breeding for high water-use efficiency. Journal of Experimental Botany 55: 2447 – 2460 Dong, B.; D. Molden; R. Loeve; Y. H. Li; C.D. Chen; J.Z. Wang (2004): Farm level practices and water productivity in Zhanghe Irrigation System [online]. In: Paddy and Water Environment. Published on-line 10 November 2004. Heidelberg: Springer Verlag Droogers, P; G. Kite (2001): Simulation Modeling at Different Scales to Evaluate the Productivity of Water. In: Physics and Chemistry of the Earth, 26.11-12: 877-880 FAO (2003): Agriculture, Food and Water. Rome: FAO FAO (2003): Unlocking the Water Potential of Agriculture. Rome: FAO Gleick, Peter H. (2000): The Changing Water Paradigm – A Look at Twenty-first Century Water Resources Development. In: Water International 25.1: 127 – 138

284



Hamdy, A.; R. Ragab, Elisa Scarascia-Mugnozza (2003): Coping with water scarcity: Water saving and increasing water productivity. Irrigation and Drainage 52: 3 – 20 Hatfield, Jerry L.; Thomas J. Sauer; John H. Prueger (2001): Managing Soils to Achieve Greater Water Use Efficiency: A Review. Agronomy Journal 93: 271 – 280 Howell (2001): Enhancing Water Use Efficiency in Irrigated Agriculture. Agronomic Journal 93: 281 – 289 Howell (2003): Irrigation Efficiency. Encyclopedia of Water Science: 467 – 472 IFPRI (2001): Sustainable Food Security for All by 2020. Proceedings of an International Conference September 4–6, Bonn, Germany Jalota, S.K.; S.S. Prihar (1998): Reducing Soil Water Evaporation with Tillage and Straw Mulching. Ames: Iowa State University Press Kang, Shaozhong; Lu Zhang; Yinli Liang; Xiaotao Hu; Huanjie Cai; Binjie Gu (2002): Effects of limited irrigation on yield and water use efficiency of winter wheat in the Loess Plateau of China. Agricultural Water Management 55: 203 – 216 Kassam, Amir; Martin Smith (2001): FAO Methodologies on Crop Water Use and Crop Water Productivity. Paper No. CWP-M07, submitted to the Expert Meeting on Crop Water Productivity, Rome, 3 to 5 December 2001 Keller, A. ; J. Keller, D. Seckler (1996): Integrated water resource systems: Theory and policy implications. Research Report 3. Colombo, Sri Lanka: International Irrigation Management Institute (IIMI) Keller, A. ; J. Keller, G. Davids (1998): River basin development phases and implications of closure. Journal of Applied Irrigation Science 33.2: 145 – 163 Kijne, Jacob W.; Randolph Barker, David Molden (2003). Water productivity in agriculture: limits and opportunities for improvement. Wallingford: CAB International Mo, X.; S. Liu, Z. Lin, Y. Xu, Y. Xiang, T.R. McVicar (2004). Prediction of crop yield, water consumption and water use efficiency with SVAT-crop growth model using remotely sensed data on the North China Plain. Ecological Modelling Molden, David (1997). Accounting for Water Use and Productivity. SWIM Paper 1. Colombo, Sri Lanka: International Irrigation Management Institute Molden, David; R. Sakthivadel; Christopher J. Perry; Charlotte de Fraiture; Wim H. Kloezen (1998). Indicators for Comparing Performance of Irrigated Agricultural Systems. Molden, David; R. Sakthivadel; Zaigham Habib (2001a). Basin-Level Use and Productivity of Water: Examples from South Asia. Research Report 49. Colombo, Sri Lanka: International Water Management Institute Molden, David; F. Rijsberman; Y. Matsuno; U.A. Amarasinghe (2001b). Increasing Productivity of Water: A Requirement for Food and Environmental Security. Dialogue Working Paper 1. Colombo, Sri Lanka: Dialogue Secretariat Molden, David; Hammond Murray-Rust, R. Sakthivadivel; Ian Makin (2003). A Water Productivity Framework for Understanding and Action. In: Jacob W. Kijne, Randolph Barker, David Molden: Water productivity in agriculture: limits and opportunities for improvement. Wallingford: CAB International Molden, David; Charlotte de Fraiture (2004). Investing in Water for Food, Ecosystems and Livelihoods. Comprehensive Assessment of Water Management in Agriculture, Blue Paper Discussion Draft. Stockholm: IWMI On-Farm Irrigation Committee of the Irrigation and Drainage Division. (1978). Describing irrigation efficiency and uniformity. Journal of Irrigation and Drainage Div., ASCE: 104(1):35-42. Palacios-Vélez, Enrique (1994). Water Use Efficiency in Irrigation Districts. In: Héctor Garduno, Felipe Arreguin-Cortés: Efficient Water Use. International Seminar on Efficient Water Use, Mexico City, October 21-25, 1991. Peranginangin, Natalia; Ramaswamy Sakthivadel; Norman R. Scott; Elise Kendy; Tammo S. Steenhuis (2004). Water accounting for conjunctive groundwater/surface water management: case of Singkarak-Ombilin River basin, Indonesia. In: Journal of Hydrology 292: 1 – 22 Perry, C.J. (1999). The IWMI water resources paradigm – definitions and implications. In: Agricultural Water Management 40: 45 – 50

285



Postel, Sandra L (2003). Securing water for people, crops, and ecosystems: New mindset and new priorities. In: Natural Resources Forum 27.2 (May 2003): 89 – 98 Qadir, M.; Th.M. Boers; S. Schubert; A. Ghafoor; G. Murtaza (2003). Agricultural water management in water-starved countries: challenges and opportunities. In: Agricultural Water Management 62: 165 – 185 Renault, D; W.W. Wallender (2000). Nutritional water productivity and diets. In: Agricultural Water Management 45: 275 – 296 Rosegrant, Mark W. (1997). Water Resources in the Twenty-First Century: Challenges and Implications for Action. Food, Agriculture, and the Environment Discussion Paper 20. Washington: International Food Policy Research Institute Rosenzweig, Cynthia; Kenneth M. Strzepek, David C. Major, Ana Iglesias, David N. Yates, Alyssa McCluskey, Daniel Hillel (2004). Water resources for agriculture in a changing climate: international case studies. In: Global Environmental Change 14: 345 – 360. Ruthenberg, H. (1980). Farming systems in the tropics. New York: Oxford University Press Sakthivadivel, R.; Charlotte de Fraiture; David D. Molden; Christopher Perry ; Wim Kloezen (1999). Indicators of Land and Water Productivity in Irrigated Agriculture. In: Water Resources Development 15.1/2: 161 – 179 Seckler, D. (1996). The New Era of Water Resources Management: From ‘Dry’ to ‘Wet’ Water Savings. Issues in Agriculture 8, April 1996. Consultative Group on International Agricultural Research. http://www.cgiar.org/publications/pub_issues.html Seckler, D.; D. Molden; R. Sakthivadivel (2002). The concept of efficiency in water resource management and policy. In: Water productivity in agriculture: Limits and opportunities for improvement, ed. J.W. Kijne. Wallingfort, UK: CABI SIWI-IWMI (2004). Water – More Nutrition per Drop. Stockholm International Water Institute. Stockholm Skaggs, R.K.; Z. Samani (2005). Farm size, irrigation practices, and on-farm irrigation efficiency. Irrigation and Drainage 54: 43 – 57 Turner, Neil C. (2004). Agronomic options for improving rainfall-use efficiency of crops in dryland farming systems. Journal of Experimental Botany, Vol. 55, No. 407: 2413–2425 van Dam, J.C.; R.S. Malik (eds) (2003): Water productivity of irrigated crops in Sirsa district, India Integration of remote sensing, crop and soil models and geographical information systems. Final Report of the WATPRO project Viets, F.G., Jr. (1962). Fertilizers and the efficient use of water. Advances in Agronomy 14:223-264 Wallace, Jim S.; Peter J. Gregory (2002). Water resources and their use in food production systems. Aquatic Sciences 64: 363 – 375 Wallace, J.S.; C.H. Batchelor (1997). Managing water resources for crop production. Philosophical Transactions of the Royal Society of London: Biological Sciences 352: 937–47 Warkentin, B.P. (1994). Protection of groundwater quality through efficient irrigation. In: Héctor Garduno, Felipe Arreguin-Cortés: Efficient Water Use. International Seminar on Efficient Water Use, Mexico City, October 21-25, 1991. Wichelns, Dennis (2002): An economic perspective on the potential gains from improvements in irrigation water management. In: Agricultural Water Management 52: 233 – 248 Yuan, Baozhong; Soichi Nishiyama, Yaohu Kang (2003). Effects of different irrigation regimes on the growth and yield of drip-irrigated potato. Agricultural Water Management 63: 153 – 167 Zhang, Yongqiang; Eloise Kendy, Qiang Yu, Changming Liu, Yanjun Shen, Hongyong Sun (2004). Effect of soil water deficit on evapotranspiration, crop yield, and water use efficiency in the North China Plain. Agricultural Water Management 64: 107 – 122 Zwart, Sander J.; Wim G.M. Bastiaanssen (2004). Review of measured crop water productivity values for irrigated wheat, rice, cotton and maize. In: Agricultural Water Management 69: 115 – 133

286



EFFECTIVENESS OF INDICATORS FOR SUSTAINABLE WATER USE IN AGRICULTURE G. Özerol Institute for Environmental System Research (USF) - University of Osnabrück Barbarastr. 12 - 49069 Osnabrück, Germany [email protected] SUMMARY – Water scarcity is a prominent issue about water sustainability in many countries, in particular in the Mediterranean Region. Given the fact that agriculture has a dominant share among the water user sectors and water scarcity is a threat to water sustainability, it is expected that sustainable use of water resources in the agricultural sector is crucial for dealing with the water scarcity problem. Accordingly, water use in agriculture is the topic under discussion. Being among the tools for assessing sustainability, indicators can foster actions that can contribute to implementing sustainability. As with all the indicators, the effectiveness of the indicators for water use in agriculture is desirable. By effectiveness it is understood that through monitoring and evaluation, indicators can have an effect on the behaviour of stakeholders, who develop, manage and use water resources, in order to engage them in collective action for sustainable water use. The results of a field study in Harran Plain, a region in the southeast Turkey, demonstrate that lack of stakeholder participation during the development of indicators for water use in agriculture is among the reasons for the resulting ineffectiveness of the indicators. It is also observed that lack of participation of stakeholders during the development of indicators has implications about the resulting outcomes, which indicate the lack of an integrated approach for managing water resources and a resulting collective action problem of unsustainable water use in the field study region. It is concluded that integrated approaches to water management, which adopt participation as a core principle and reflect on the economic, social, institutional and ecological dimensions of sustainability, can bring about not only effective indicators of sustainable water use, but also long-term achievements. Key words: sustainable water use, indicators, stakeholder participation, collective action.

INTRODUCTION Water scarcity is a prominent issue about water sustainability in many countries, in particular in the Mediterranean Region. Scarcity of water is related to availability and use - or consumption -. On the one hand, the availability is limited since total water supply available for human use cannot be increased substantially given the economic, ecological and social constraints on the development of non-conventional water resources. On the other hand, water consumption continuously increases due to several reasons, mainly including warmer climate and population rise. 287



There is a need to ponder on how to deal with water scarcity and ensure water sustainability through the integration of economic, social, ecological and institutional dimensions of sustainability. Integrated approaches, which consider these four dimensions, integrate the perspectives of related disciplines and respect the interests of all stakeholders, are essential. However this is not a simple issue given the existence of several factors, which put water resource under pressure. These factors mainly include population rise, increased economic activity, improved living standards, social inequity, lack of pollution control measures and management of water resources through sectoral approaches and top-down institutions (Gleick, 1995; Gardner-Outlaw and Engelman, 1997; GWP-TAC, 2000). Furthermore, the dominant approach to the management of water resources has been supplyoriented, meaning that increasing the water supplies is the priority issue, whereas the need for managing the corresponding demand is neglected. Within this context, integrated water resources management (IWRM) is relevant. IWRM is defined as follows: “a process, which promotes the co-ordinated development and management of water, land and related resources, in order to maximize the resultant economic and social welfare in an equitable manner without compromising the sustainability of vital ecosystems” (GWP-TAC, 2000:22). It is emphasized in the definition of IWRM that success in ecological, economic and social dimensions, depends on coordination of development and management activities, which constitutes the institutional dimension. Indeed the causes of the problems about water resources are attributed to “inefficient governance” and “increased competition for the finite resource” (GWP-TAC, 2000: 9). The inefficiency of governance can be due to disintegrated and uncoordinated approaches for water management, which rely on sectoral approaches and top-down institutions (Jaspers, 2003). Such institutions might not create a common understanding about the need for ensuring sustainability and neglect, or exclude, the knowledge and perspectives of stakeholders, who are not represented or involved in the existing institutions. The increased competition can be considered as the result of the allocation of available water resources to users, or user sectors, each of which claim their own right to use water. It is important to ensure the participation of stakeholders in efforts towards water sustainability, whatever the scale and the issue. In the general context of water policy implementation and water resources management, stakeholder participation is stated among the key requirements for success (Seppala, 2002; Mostert, 2003). IWRM also takes participation as an approach to be adopted for water management. It is suggested that all the relevant stakeholders should be a part of the decision making process, so as to create the balance of top-down and bottom-up institutions and to ensure that all the stakeholders are aware of the water sustainability as their common issue at stake (GWPTAC, 2000; Dungumaro and Madulu, 2003; Jasper, 2003). These premises require that the stakeholders have the capacity to participate and represent their interests, which calls for the need to build capacity for participation. Justifying the emphasis put on participation, there are various benefits expected from stakeholder participation in the management of resources. Firstly, participation enables an understanding of the impacts of the individuals’ actions on the current state of the resource that they use (Marshall, 1999). These impacts can be costs and benefits of resource use decisions both at the individual and collective level (Johnson, 1997). Through participation, the stakeholders can realise these costs and benefits, and having information about the impacts of their actions on the resource that they use, they are more likely to arrange their actions for the interest of the collective. Participation provides information about the gap between the institutions -or the rules- and the actions of the stakeholders (Johnson, 1997). Stakeholders can have a better understanding of the rules by obtaining information about them and they can become aware of and reduce the gap between their perception about the rules and the resulting costs and benefits of following or breaking the rules. They can also agree on and make commitments so as to act in compliance with the rules. Having information about the actions of other stakeholders would also be useful for the stakeholders to make sure that the commitments are kept by every stakeholder. In that respect, participation can be useful for making use of the experiences of stakeholders (Johnson, 1997; Marshall, 1999). Each participant can monitor his own actions as well as the actions of others, which creates an environment that binds the individuals to each other and to the outcomes of their actions. Such an environment

288



can enhance the perception of the stakeholders that the sanctions are assessed on those who break the rules (Ostrom et al., 1994). Participation has additional benefits in terms of enabling the decision making process to be at the same time a social learning process (Pahl-Wostl, 2002). This means that, through participation, a common understanding about the situation can be built; exchange of ideas among stakeholders, who might have diverse interests and knowledge, can be possible and the stakeholders can become more likely to act for the common goal of sustainability. The final, and very crucial, benefit that can be expected from participation is the ownership that it raises about the resource used by the stakeholders. Participation demonstrates to the stakeholders that each of them has a stake in the state of the resource that they use and their knowledge and perspectives are important for the betterment of the resource (Johnson, 1997; Marshall, 1999). Consequently, based on the added-value of the knowledge of stakeholders, the benefits of stakeholder participation can be summarised as follows: “Local knowledge is often valuable for devising rules, decision procedures and monitoring and sanctioning mechanisms that take equity as well as efficiency considerations into account, and therefore are likely to gain broad support from local citizens or resource users” (Baland and Platteau, 1996, cited in Marshall, 1999). In addition to the participation of stakeholders, the assessment of the progress towards or away from sustainability is a major issue for ensuring sustainability. Developed and used at different levels, sustainability indicators constitute a major tool for assessing sustainability. The emphasis put on indicators for the assessment of sustainability is attributed to the expected contribution of indicators to sustainability, which is possible through the monitoring and evaluation. Monitoring enables the quantification and communication of the information and in turn improves stakeholders’ knowledge about the system. It also establishes the basis for description of the system. Through monitoring, system characteristics and dynamics of the system can be better understood by the stakeholders. If the measured values of indicators are also evaluated by taking implications into consideration, then indicators can become tools to assess the success of the actions and policies for implementing sustainability and to improve their performance. With regard to the assessment of water sustainability, the importance of indicators should be acknowledged as well. In particular regarding the water scarcity problem, the lack of stakeholders’ awareness of the water scarcity problem is mentioned as a challenge for water resources and this ignorance is attributed to the perception that water will always be abundantly available (Abu-Zeid, 1998). Therefore, developing and using indicators that address the problem of water scarcity is relevant since they enable the assessment of the state of water resources and the impacts of the actions of stakeholders. Given the fact that the availability of water resources is limited and human population continuously rises, a competition among different user sectors is inevitable and agricultural sector might also be affected from the situation. Therefore agricultural sector can gain from developing and using indicators for water use, since it has the largest share among all water user sectors, especially in the Mediterranean region (Araus, 2004). Justifying the argument that indicators are useful for assessing water use in agriculture, in many countries indicators are developed and used within the agricultural sector, e.g., indicators for irrigation (Bos, 1997; Molden at al, 1998; Lorite at al., 2004; Kellett at al., 2005). Within the context of water use in agriculture, the effectiveness of the indicators is desirable. In this paper, the effectiveness of an indicator is defined the degree of adoption and utilisation of the indicator by the stakeholder in taking actions that contribute to sustainable water use in agriculture. It is argued that effectiveness of indicators requires stakeholder participation during indicator development. Through a participatory approach the knowledge and perspectives of the stakeholders can be incorporated when the indicators for sustainable water use in agriculture are developed and hence the indicators can be perceived as relevant and useful and hence their effectiveness can be improved. It is acknowledged that the indicator development requires the multiple concerns and, among others, stakeholder participation is considered as a principle during the development of indicators (Hardi and Zidan, 1997; Bell and Morse, 2004; McCool and Stankey, 2004).

289



MATERIALS AND METHODS With the aim of investigating the role of stakeholder participation in the effectiveness of indicators for sustainable water use, an analytical framework was developed and applied using empirical findings from a field study.

Analytical Framework The analytical framework was developed through the incorporation of the relationship of the stakeholder participation during indicator development and effectiveness of indicators to the theory on collective action for common-pool resources (Ostrom, 1990; Ostrom et al., 1994). After a review of the theory on collective action and its application to the use of common-pool resources, it was inferred that sustainable water use requires collective action, since water for agriculture is a common-pool resource that needs to be used collectively by all relevant stakeholders. Several relevant aspects of collective action for sustainable water use and the relationships among these aspects were examined. These aspects include the context for water use in agriculture, the stakeholders of water use in agriculture, the indicators used by the stakeholders, the institutions -or the rules- for water use in agriculture, the actions of the stakeholders in terms of following the rules, using water for agriculture, and developing and using indicators, and the outcomes of these actions. It is suggested that the actions of the individuals take place in three different levels, namely the operational, collective-choice and constitutional-choice level (Kiser and Ostrom, 1982 cited in Ostrom et al., 1994) Corresponding rules are devised and used for each level and for the interactions between the levels (Ostrom, 1990; Ostrom et al., 1994). The rules at the constitutional-choice level are about the governance and they indirectly affect the operational level actions since they define the way that the collective-choice rules are made and changed. Similarly collective-choice rules also affect the operational level actions indirectly. They are the rules about how the operational rules are made and by whom the operational rules can be made and changed. Collective-choice rules mainly include the rules about policy making and management. Finally the operational rules include the rules that affect the daily actions of the individuals about appropriation, provision, monitoring and enforcement. Given the scope of the research, the focus of the analysis of the empirical data is on the collective-choice and operational rules and the corresponding actions of the individuals. The constitutional-choice rules are discussed to the extent that there are findings about the rules and actions at this level. With regard to the position of indicators for sustainable water use in the framework, it is acknowledged that indicators can contribute to collective action, since they have the potential to influence the behaviour of the stakeholders in a way to engage in collective action. Such behaviour of stakeholders is associated with the adoption and utilisation of indicators by the stakeholders and indicators are considered effective to the extent that they fulfil this potential by affecting the behaviour of stakeholders, and in turn their actions. Based on the assumption that stakeholder participation during indicator development is among the reasons for the effectiveness of indicators, stakeholders and indicators were examined thoroughly in terms of their interactions and their relationship to collective action for sustainable water use.

Field Study The field study was carried out through individual interviews with the representatives of the stakeholders and through the review of written documents, which are extracted from different sources of information. Harran Plain was chosen in order to carry out the interviews with the stakeholders from regional and local level. Harran Plain is a region in Şanlıurfa province, located in the south-eastern part of Turkey and included in the GAP (Southeastern Anatolia Project). GAP is a regional development programme, aiming at, among others, increasing agricultural production, employment and income in the South-Eastern Anatolia Region of Turkey (Ünver, 1997). The region has a semi-arid climate with very low precipitation levels. Within the content of the GAP, investments have been made for water

290



development and as a result irrigated agriculture is practiced for 10 years in the region. Before that time, rain-fed agriculture was practiced. Currently, water for agriculture is abundant due to the water transferred from the Euphrates River through Şanlıurfa Tunnels and main irrigation canals. However the area of the irrigated land gradually increases through the completion of canals that carry water to the regions other than Harran Plain. This means that the same amount of water will be shared by higher number of users, which implies a potential for water scarcity when the water is used excessively. There are studies, which address the problem of excessive water use in the region, having also adverse effects on the soil quality and water table levels (Kendirli et al, 2005). Therefore the field study was fruitful in terms of eliciting the perception of stakeholders about water scarcity and indicators for water use in agriculture as tools to prevent the potential impacts of irrigated agriculture. The specific circumstances, under which the indicators for water use in agriculture are developed and used, were also investigated for Turkey in general, and for the field study region, in particular. It is expected that the use of indicators in terms of monitoring and evaluation, can be made on the basis of different water user sectors. However the development of indicators cover several related processes about water, i.e., planning, investment, management, etc., which are usually done for all sectors through several overlapping laws and regulations. Making the system more complicated, the responsibility and authority regarding different uses of water are distributed among different stakeholder organisations. Therefore it is meaningful to have an understanding of the national setting for water use without a narrow focus on agricultural water use. Stakeholders of water use in agriculture include the governmental organisations, environmental non-governmental organisations and water user organisations. The activities of each stakeholder and the distribution of responsibility and authority among the stakeholders at different levels and were explored. The two major stakeholders at regional level are State Hydraulics Office (DSI) and Water User Organisations (WUOs). DSI is the governmental organisation responsible for the planning, development and administration of water resources at national level. It has a general directorate and 26 regional directorates (DSI-RD), which function at regional level. Two officers of DSI-RD, which is responsible for Harran Plain, were interviewed. WUOs are the legal entities formed by the representative of farmers. They are responsible for the appropriation, i.e., distribution of water to the farmers and provision, i.e., operation, maintenance and repair of the irrigation facilities. Since the WUOs are active in all Turkey, with a total number of around 780, only the situation in Harran Plain is investigated and five out of the eleven WUOs were contacted in the region. In addition to the scheduled interviews, conversations were made with farmers and the relevant observations and impressions from these conversations are incorporated. Given the fact that the aim of interviews was to elicit the perception and the knowledge of the stakeholders, it was found appropriate to conduct semi-structured interviews. Most of the questions were prepared as open-ended questions so as to enable the interviewees to talk freely on the issue addressed by the questions. While most of the questions are the same for all the interviewees, one or two questions were included or excluded for several interviewees, according to the major tasks and specialisation fields of the interviewees.

Analysis of Findings The findings from the field study were utilised as empirical data for the application of the framework. For this purpose, the context, rules, actions and outcomes, which are related to water use in agriculture, were identified and analysed. Firstly, the examination of the context for water use in agriculture was made, which includes description of water for agriculture as a common-pool resource, the producers, providers and the appropriators of water for agriculture, and the social and cultural conditions of the region. Secondly, the rules about water use were identified and discussed. Utilising the empirical data, the rules at constitutional-choice, collective-choice and operational levels were identified. Given limited empirical data, a thorough analysis of the constitutional-choice rules could not be made and these rules were mentioned in order to give a general idea about the background for the collectivechoice and operational rules.

291



Thirdly, the outcomes of water use in agriculture were identified at the system level and the actions of the stakeholders, which might lead to these outcomes, were investigated. The analysis of outcomes is focused on the collective-choice and operational levels with the purpose of tracing how the interactions of the context, rules and actions might lead to the outcomes. From the analysis of context and the rules, several implications have been made. The resulting set of implications is classified under two groups; the implications of the context and the constitutional-choice rules and the implications of the collective-choice and the operational rules, most of which are related to the monitoring and enforcement of rules for allocation of water.

RESULTS The interactions among the context and the rules were investigated with the purpose of revealing how these interactions might lead to the actions and outcomes of water use in agriculture in the field study region. For this purpose, firstly, the outcomes observed at the regional level were identified. According to all interviewees from DSI-RD and the WUOs, there is “excessive water use” in the region, which results in the following outcomes: - rise in water table - degradation of soil quality; mainly in the form of salinisation and becoming barren - high quantity of water that goes to discharge The above outcomes are negative externalities of water withdrawals by water users. These externalities are experienced by all the water users in the region and they imply that the individuals do not act for the achievement of common welfare or reflect on the ecological impacts of their individual interest. Hence it can be stated that there is a collective action problem in the form of unsustainable water use. There is no evidence that DSI or the WUOs experience problems about the irrigation facilities. This situation might mean that the focus of the WUOs and DSI are more on the irrigation facilities. Therefore it can be inferred that the actions of the stakeholders do not lead to provision problems, but rather appropriation problems in the form of the outcomes above. It is argued that excessive water use can be a result of the context in the region, as well as the lack of monitoring and enforcement of the rules. The factors, which might lead to above outcomes, are identified and explained below.

Context and Constitutional-choice Rules Water as a new resource Water for irrigation is available since ten years. Before that time, there was no irrigated agriculture and almost no water at all. Hence, irrigation water is new for the farmers. Therefore it can be expected that the farmers are not experienced in using water for agriculture. Another expectation is the lack of awareness about the limited water availability and the need for using water efficiently. This might cause a distorted perception. Since water continuously flows in the canals, the farmers might have a perception that water is not abundant. Similarly the farmers, inexperienced in irrigation, might think that using more water brings about higher yield. Unlike the farmers, all the WUOs state that the water resources are not abundant and they try to enforce the farmers to use water efficiently. Hence, it can be expected that the farmers’ awareness of water availability and use might be a critical factor on their actions and resulting outcomes. Governmental support for cotton production The national agricultural policy, which supports cotton production, might also have an impact on the actions of the farmers. The governmental support on cotton production might induce the farmers to cultivate cotton with the expectation that they can sell what they raise and guarantee their income. Currently, the share of cotton is about 80-90% of the whole cultivated land in the region. This proportion had been foreseen as 25% in the GAP Master Plan (Ünver, 1997). The impact of this high proportion is a water demand much higher than expected, since cotton consumes more water than many other crops. The support of cotton production has an impact on the type of the crop cultivated

292



by the farmers. The individual interests of farmers about increasing their income and the outweighing short-term benefits can be the factors that lead to this choice of crop, which in turn affects substantially the quantity of water used by the farmers. Size and composition of the WUOs There is no rule about the limit to the number of farmers to be included in a WUO. This situation results in the fact that the WUOs are not homogeneous in terms of their size and the area of land that they manage. Having a high number of members can make it difficult to build continuous relationships and trust among the farmers. Furthermore, the existence of leasehold farmers increases the diversity of the members of the WUOs. The resulting heterogeneous and unstable composition of the WUO can lead to short-term relationships among the farmers and lack of a shared interest for a common future with other farmers in the WUO. It is also mentioned that the leasehold farmers might give less consideration to the externalities of water use in agriculture and take only the short-term benefits into account. Combined with the size of the WUOs, the existence of leasehold farmers can also lead to problems about monitoring the actions of farmers in terms of water use and enforcing the rules for allocation and provision. The farmers who withdraw water from the same canal even might not know each other. Organisation of the Water User Organisations The WUOs find their institutional capacity low in terms of the outputs of their activities. They attribute it again to the lack of a WUO law, as well as the lack of awareness among farmers, lack of coordination among higher level organisations, limited financial and administrative resource and lack of qualified staff. Furthermore, within the organisational structure of the WUOs, the secretary general is the only person who has to deal with the administrative, operational and legal issues. This situation makes secretary general the central person in the WUOs and in turn creates dependency on the side of other personnel as well as the member farmers of the WUO.

Monitoring and Enforcement of Collective-choice and Operational Rules Monitoring and Enforcement by DSI Monitoring of water use for agriculture at the regional level is carried by DSI-RD through monitoring the quantity of water allocated to each WUO and the water use per hectare for the region. However most of the WUOs indicate that there is no comparison of among the WUOs with respect to each other or to the regional level. Furthermore, each year, WUOs submit three documents to DSI, namely, monitoring and evaluation report, inspection report and water distribution plan, all of which have the same structure for all WUOs, meaning that they are not tailored according to specific conditions of the WUOs. Inspection report and water distribution plan are used for declaring respectively the provision and distribution activities. Monitoring and evaluation report is used in order to monitor and evaluate all activities of the WUOs. Since none of the abovementioned outcomes are related to provision problems, the use of inspection report is not analysed, whereas the issues about monitoring and evaluation report and water distribution plan are discussed in detail. Monitoring and evaluation report is a very comprehensive document conveying many types of data to DSI. It can be expected that the data in this document are evaluated by DSI and the performance of the WUOs are assessed in terms of their past vs. present outputs or against other WUOs. With regards to the uses of the monitoring and evaluation report, most of the WUOs indicate that they could not make use of it in terms of a contribution to their success. All the WUOs indicate that they receive no regular or formal feedback from DSI about the monitoring and evaluation report. This situation can imply that the monitoring and evaluation reports, which the WUOs send in the previous years, are not regularly evaluated by DSI. It is mentioned by all the interviewees from WUOs and DSI that DSI informally evaluates the performance of the WUOs, e.g., by talking about the outputs of the monitoring and evaluation reports during their meetings. However there is no evidence about a regular performance evaluation of the WUOs.

293



Absence of feedback from DSI implies a lack of common understanding about the uses of the monitoring and evaluation report as well as inspection report and water distribution plan. It is also a communication problem between the DSI and WUOs due to the one-way flow of information from WUOs to DSI. Some WUOs mention that they expect DSI to evaluate their performance. It can be expected that WUOs require a performance evaluation from DSI, since they share the water from the same canal and each of them would like the others to perform well. However the WUOs complain about their legal and organisational incompetence, which constitutes a barrier for them to take such actions. With regards to the water distribution plan, it is used for planning the allocation of water to the WUOs. It is mentioned that the demanded value on the water distribution plan are negotiated and it can be decreased by DSI. However, during the irrigation season, water is continuously available and the weekly demands of WUOs are met. Hence, even if DSI monitors the quantity of water allocated to each WUO, the WUOs do not make their allocation rules according to the quantity of water allocated to them. Monitoring and Enforcement by the WUOs As observed from the operational rules, most of the decisions made by the WUOs are based on two variables, namely, the area of the land irrigated and the type of the crop cultivated on the land. A water related variable is not used by the WUOs when they make decisions, nor they monitor the quantity of water used. Distribution of water to farmers and contribution of farmers to provision are made based on area of the land and the crop type. WUOs plan the distribution of water based on the area of the land to be irrigated and the type of the crop cultivated on the land. The amount of water to be distributed increases with increasing area of land and is also dependent on the type of the crop. According to the distribution plan, the water is allocated to each farmer for a predefined number of days. Thus the WUOs do not monitor the quantity of water withdrawn by each farmer; instead they rely on the duration of irrigation. Only one of the interviewed WUOs is an exception to this case. This WUO was founded in 2004. It is planned by the WUO that water is withdrawn through valves instead of siphons and irrigation is made with piped irrigation network and a closed system instead of open canals. It is stated that existence of valves also enables the metering of the quantity of water taken by each farmer. Currently, some of the fields have been equipped with the valves and it is mentioned that they will monitor the water use through the valves. It is also mentioned that this method decreases the amount of water that goes to discharge. With regards to the method for the determination of the irrigation fee to be paid by each farmer, most WUOs use the area of the irrigated land and the type of the crop cultivated as the two criteria. This implies that the irrigation fee is independent of the quantity of water actually used by the farmers. Indeed, most of the WUOs indicate that there is no metering of the water withdrawn by each farmer; instead the farmers take the water from the tertiary canals through the siphons, which withdraw the water from the canal. It can be expected that collecting the irrigation fee based on quantity is not adopted by the WUOs since its justification would more difficult, given that they do not monitor it. Finally, all the WUOs indicate that they are responsible for monitoring the level of water table in their region, but most of the do not make it. It is also mentioned by some of the WUOs that currently DSI monitors the level of water tables, but there is an impression that WUOs are not successful in terms of communicating these values with the farmers and convincing the farmers not to use excessive water. When the monitoring and enforcement actions of WUOs are considered as a whole, three factors, which might have an effect on the actions of the farmers, in terms of monitoring their own water use, are identified as follows: - determining the irrigation fee and water distribution according to crop type and irrigated area - using the irrigation duration as the unit of water distribution - not monitoring the quantity allocated to each farmer Monitoring by the Farmers

294



Following from the rules, which are based on the above three factors, it can be inferred that the farmers do not have an incentive to monitor how much water they use or what the impacts of their water use are. In that respect, the interactions between three variables, namely the type of the crop, quantity of water and duration of irrigation, and two main issues that result from these interactions are relevant about the actions of the farmers and the abovementioned outcomes. Firstly, despite the fact that WUOs assume 24 hours of irrigation per day, it is mentioned that some of the farmers do not apply “night-irrigation”; they irrigate the land during the day and let the water go to the discharge during evening and night. Hence, insufficiency of night irrigation is the main reason of the high quantity of water that goes to discharge. It is also indicated that quantity of discharged water should be as low as possible, and zero in the ideal, however, it is not possible with open canals and siphons. It can be expected that because the distribution of water is on the basis of time, duration of irrigation is critical for the farmers and they should adopt night irrigation. Indeed, it is mentioned that night irrigation has several advantages. Firstly, if it is applied as complementary to day-light irrigation, the duration that the water is withdrawn from the canal can be extended to 24 hours/day and irrigation can be finished earlier. Furthermore it is likely that the water allocated to the WUO is used more efficiently, since the amount of water that goes to discharge reduces significantly. Secondly, if it is applied either as a substitute (or complementary) to day-light irrigation, the water loss due to evaporation is decreased since the crop consumes the water, not (only) during the warmer hours of the day, but cooler hours of the evening and night. Thirdly, the consumption of water by most of the crops is easier when the crop is irrigated during evening or night. But still the WUOs cannot effectively implement night irrigation, implying that the farmers might not have enough motivation to apply it. Secondly, the quantity of water that they use is not a priority of farmers when they irrigate the land. The WUOs takes into account the type of the crop, since each crop needs different amount of water. However, the farmers do not monitor the quantity of water that they use; instead the relation between the type of the crop and the amount of water that the crop needs is reflected by a heuristic adopted by the farmers. It is mentioned that the farmers base their actions about water use according to the “total number of times” that they irrigate the crop. It is stated by a farmer that in the previous years, the farmers used to irrigate the cotton ten times; now they irrigate six or seven times. Even if they do not monitor their own water use, it could be expected that the farmers could have an interest in the water use at WUO level and follow the outcomes of the monitoring activities carried out by the WUO. However it is mentioned that the farmers do not have a concern about these monitoring activities. Consequently, it is observed from the actions of farmers that using the irrigation duration as the unit of water distribution and planning it according to crop type and irrigated areas are not reflected on the side of the farmers’ actions. This situation has several implications about the WUOs and farmers. These issues imply that the rules used by the WUOs do not relate the quantity of water, needed by the crops, to the duration of irrigation. Therefore the farmers might not adopt the night-irrigation and monitor the quantity of water at the farm or local level. Several secretary generals explain the reasons for the lack of monitoring by the farmers and the farmers’ awareness of the impact of their actions on the abovementioned outcomes using a common proverb. This implies that, since most of the outcomes are observed with a time lag after their actions, the farmers might not be considering what the outcomes will be. It can be expected that lack of knowledge, experience and awareness of water for agriculture, and the outweighing short-term benefits, such as earning income, can lead to these actions.

DISCUSSION Remember that the condition for the indicators to be effective is that they are adopted and utilised by the stakeholders in taking actions that contribute to sustainable water use in agriculture. Hence the actions of DSI-RD, WUOs and farmers, in terms of monitoring and evaluating the indicators, are of utmost importance in evaluating the effectiveness of the indicators. Analyses of findings address the following actions of DSI-RD, WUOs and farmers: - Neither DSI-RD nor WUOs monitor the quantity of water withdrawals by the farmers. - Farmers monitor neither the quantity of water that they use nor the impacts of irrigation on soil quality and water table level.

295



- WUOs do not monitor the realisation of their water distribution plan in terms of the allocation of the quantity of water according to it. - WUOs monitor many indicators, use them to prepare the monitoring and evaluation report and submit it to DSI annually, but they receive no regular feedback about it and most of the WUOs make little use of the monitoring and evaluation report in their success. - WUOs do not have an ownership of indicators that they monitor for the monitoring and evaluation report; several WUOs perceive monitoring and evaluation report as irrelevant to their activities and having no contribution to their performance. - DSI monitors indicators about water use by WUOs, but the performance of the WUOs is not evaluated explicitly and little communication occurs between DSI and WUOs about the outcome of the evaluation. Above findings suggest that the indicators are either not monitored at all, or monitored with a lack of evaluation of their value, implying the lack of utilisation of indicators by the stakeholders. Furthermore there is an impression that the adoption of the indicators by the stakeholders is very low, which is reflected by the lack of ownership about indicators and motivation to monitor and evaluate the indicators. All these findings are considered as strong evidences for the existence of ineffective indicators and it is concluded that the indicators for water use in agriculture are ineffective in the field study region. In order to verify whether there was lack of stakeholder participation during indicator development, the functioning of WUOs and the method, with which the indicators for water use in agriculture had been developed, are discussed. The role of the WUOs in creating a participatory context is considered important, since the WUOs act like a bridge between DSI and the farmers. It is acknowledged that, by carrying out the operation and maintenance of the irrigation facilities and by distributing the water on their own, WUOs constitute a useful collective-choice entity for the participation of farmers in the management of the irrigation system. However, participation of WUOs remains at operational level. There is no evidence from the empirical data that WUOs have participated during the design of many constitutional-choice rules that in turn affect their decisions at both collective-choice and operational level. For instance, with regards to the monitoring and evaluation of the activities of the WUOs, there are several collective-choice actions, in which the WUOs could be involved. Among others, contributing to the design of monitoring and evaluation report, tailoring it according to their context, communicating its content to their farmers and in turn adapting it by considering the changes in conditions would be forms of participation at collectivechoice level. However there is little evidence that WUOs have been involved in such actions. The methods that had been used for the development of indicators monitored by DSI and WUOs have direct implications about the participation of stakeholders during indicator development. At this point, a differentiation can be made with the formal and informal indicators, which imply the indicators used for monitoring and enforcement of formal and informal rules. The formal indicators mainly include those monitored for the monitoring and evaluation report. These indicators had been imposed by DSI with a unique format to be used by all the WUOs in Turkey. As explained above, there is no evidence that the WUOs or farmers have contributed to the design or preparation of the monitoring and evaluation report or adapted it to their specific conditions. With regards to the informal indicators, WUOs have the initiative to choose the method of distribution of water to the farmers. Accordingly the WUOs develop and use indicators, in particular for monitoring water distribution. However, the informal indicators that are used by the WUOs do not match with the formal indicators of DSI-RD. On the one hand, DSI-RD takes weekly water demands from the WUOs and monitors the quantity of water allocated to each WUO. On the other hand, the WUOs distribute the water to farmers on the basis of irrigation durations and number of irrigation turns, which are different indicators to monitor water distribution. The choice of WUOs to use durations and number of turns as indicators is their own decision, but not a result of the participation of DSI-RD or the farmers. Furthermore, the secretary general is the only person responsible for the fulfilment of administrative and technical tasks of the WUOs. This situation might imply lack of sharing of responsibility with other members of the WUO and lack of a participatory approach for the establishment, monitoring and enforcement of collective-choice rules devised and used by the WUOs.

296



Consequently, the development of neither formal nor informal indicators was made using a participatory approach and lack of a common understanding, about the monitoring and enforcement of collective-choice and operational rules, is observed, implications of which are experienced in the form of mismatching formal and informal indicators. Thus, it can be concluded for the field study region that there is a lack of stakeholder participation during the development of indicators for water use in agriculture. Having concluded that indicators for water use in agriculture are ineffective and there was lack of stakeholder participation during indicator development, it can now be discussed whether the latter is among the reasons of the former. For investigating this relationship, several factors are discussed based on the interactions among the context and the rules at different levels. Firstly, the constitutional-choice level decisions can affect the behaviour of the farmers. These decisions mainly include the governmental support for cotton production as part of the agricultural policy and the supply-oriented approach of DSI. Governmental support for cotton production is likely to effect the decision of the farmers in terms of their crop choices. Since the cultivation of a crop, which can be sold, is rational for the farmers, assuring their income, i.e., the short-term benefits of water use can become dominant for the farmers. Therefore short-term benefits of using water without monitoring its quantity or impacts might outweigh the concern for common interests, which require monitoring and evaluation. Furthermore the supply-oriented approach of DSI, through the increase of water supply for irrigation by constructing necessary infrastructure and diversion of water from the rivers is relevant. This approach can result in a stimulus, on the side of the WUOs and farmers, that water use in agriculture and its impacts are not a priority issue, since water is continuously available and irrigation is possible throughout the year, no matter what the indicator values are. Given also the lack of awareness about irrigation methods and potential impacts of irrigated agriculture, it becomes more likely that indicators are not monitored or evaluated by the WUOs and the farmers. Secondly, the socio-economic conditions of the region require that some farmers cultivate the land through leasing it temporarily. It is observed from the analysis of the findings that the existence of leasehold farmers both increases the diversity of the WUOs and makes it more likely that indicators are not monitored. The leasehold farmers might have different preferences than the owners, in particular about the sustainability of soil and water resources the region. Since the priority of a leasehold farmer can be to earn as much as income, without caring for the water use levels and impacts of irrigation, it can be possible that the leasehold farmers do not monitor and evaluate the indicators. Hence, the ineffectiveness of indicators can be attributed to the ownership patterns of the farms, too. Thirdly, the perceptions of the stakeholders about water for agriculture are important. The context for water use in agriculture in the region shows that water for agriculture is a new resource for the farmers. This situation makes it more likely that the farmers are inexperienced about irrigation methods and ignorant of the availability of water for agriculture and the impacts of irrigation. Despite the fact that both DSI-RD and the WUOs are aware of the limited availability of water, all the problematic outcomes at regional level are attributed to excessive water use. However, the transfer of irrigation water from the Euphrates River conceals the potential water scarcity problem, which would be experienced in the absence of the irrigation canals. This supply-oriented and top-down approach does not enforce DSI-RD and WUOs to allocate water without causing excessive use. In the absence of such enforcement, the awareness of DSI and WUOs are not put in practice and it is not communicated to the farmers, either. Under these conditions, the farmers become critical actors for water use; they are the people closest to water and their perceptions and actions have direct effect on the quantity of water used. Thus, existence of an action like excessive water use might indicate that the farmers do not perceive water as a scarce resource. Such distorted perception of the farmers might imply lack of farmers’ awareness about limited availability of water. Lack of farmers’ awareness is attributed to the fact that farmers the communication of farmers with DSI-RD and WUOs about limited availability of water is not sufficient and the farmers were not trained about irrigation methods and water use, i.e., they did not participate in the design of the rules, which include the monitoring and evaluation of the currently used indicators. This means that lack participation of farmers can also be among the reasons of ineffective indicators. Fourthly, the limited capacity of the WUOs to train the farmers for monitoring, to communicate the results of the monitoring activities and to involve the farmers in the management of the WUOs can also lead to a lack of motivation on the side of the farmers to monitor and evaluate their water use

297



behaviour. This reasoning applies to the relationship of DSI and the WUOs, too. There is little evidence that the WUOs are motivated to monitor and evaluate the indicators. This problem is associated with the lack of communication between DSI and the WUOs, especially in terms of the design of the allocation and provision rules, the evaluation of the performance of the WUOs and the feedbacks of the monitoring activities. Additionally, as mentioned above, the mismatch between the indicators monitored by DSI and the WUOs is a result of the lack of participation of the other during the development of own indicators. These communication and cooperation problems bring about lack of adoption and utilisation of the indicators. Thus, the lack of participation by DSI, WUOs and farmers to develop the indicators is a reason for ineffective indicators. The third and fourth factors imply that lack of stakeholder participation during the development of indicators is among the reasons for ineffective indicators. It is also acknowledged that the constitutional choice level rules, e.g., governmental support on cotton production, and the context related factors, e.g., the socio-economic conditions of the region and existence of leasehold farmers, also have an effect on the ineffectiveness of indicators. However it is expected that their impacts on the outcomes at regional level could be mitigated if the farmers would have been involved when the indicators had been developed. Through participation it could be more likely that the farmers adopt and utilise the indicators; since that would have had the opportunity to be aware of the scarcity of water resources, the impacts of their actions on the sustainability of soil and water resources, the trade-off between individual and common interest and the benefits of monitoring and evaluating the indicators. It is concluded from the above discussion that indicators for water use in agriculture are ineffective, there is lack of stakeholder participation during the development of indicators and lack of stakeholder participation during the development of indicators for water use in agriculture is among the reasons for having ineffective indicators in the end. Consequently, the overall situation in the region addresses a collective action problem due to excessive use of water and its adverse impacts on the soil quality and water table level. The previous barrier on agricultural production, i.e., the arid climate that causes water scarcity, has been overcome through bringing water from Euphrates River. However, excessive use of water indicates that the management of water resources, which is practiced since the construction of irrigation system ten years ago, has not adopted an integrated approach and the outcomes indicate a collective action problem of unsustainable water use. It is expected that an integrated approach would not only improve social and economic conditions by creating employment and income through increased agricultural production, but also build institutions, which enable participation of WUOs and farmers, and in turn ensure that the adverse ecological impacts of irrigated agriculture are prevented, or at least minimised.

CONCLUSIONS From the review of literature it was observed that indicators are developed and used as tools to support the efforts towards implementing sustainability and the participation of stakeholders during indicator development is considered among the major factors that bring about effective indicators. In this paper, an attempt was made in order to explore the relationship between the lack of stakeholder participation during the development of indicators for water use in agriculture and the ineffectiveness of the indicators. The empirical findings demonstrate that indicators for sustainable water use in agriculture are essential tools for dealing with collective action problems. When they are not monitored and evaluated by the stakeholders, the outcomes at system level are more likely to be the symptoms of a collective action problem, which is experienced as excessive water use in the field study region. Lack of stakeholder participation during the development of indicators proves to be among the reasons for the ineffectiveness of the indicators, since the stakeholders do not have an ownership about the indicators and lack a common understanding about the benefits of monitoring and evaluating them. Furthermore, the affects of the context and constitutional-choice rules are also to be acknowledged. In that respect, the socio-economic conditions in the region and decisions made regarding the national water, energy and agricultural policies affect the perceptions of the stakeholders to create a

298



priority for short-term benefits and can result in a lack of motivation to monitor and evaluate the indicators. The results presented in this paper demonstrate a static view. However the national context is rapidly changing due to several reasons. For instance, new institutional arrangements are foreseen at the constitutional-choice level, e.g., a water law and a WUO law (DPT, 2001). The priorities of all governmental organisations may also change according to the course of events in the EU accession process. The limited availability of empirical evidence is a reason for making conclusions only for Harran Plain, in which the interviews were carried out with the WUOs. At the national level, a more thorough analysis could be made if there were more empirical data available. For instance, data could be collected for more than one region and the results could be compared. If more data could be collected at the national level, the effects of the constitutional-choice rules could also be discussed in more detail. At this level, two issues were analysed, namely the irrigation projects in the region, which made water available for agriculture, and the effect of governmental support on farmers’ crop choice and in turn water use decisions. However, there is not enough evidence from the field study so as to discuss all the rules and actions at this level. This situation is due to relevance of many other stakeholders, who could not be included in this study, and national policies, which affect the constitutional-choice rules and actions about energy, agricultural production, and water planning and development. Given the scope and aims of this paper, it is hoped that the presented findings and discussions constitute a comprehensive picture of the situation for water use in agriculture.

REFERENCES Abu-Zeid, M.A., 1998, “Water and sustainable development: the vision for world water, life and the environment”, Water Policy, 1, 9-19. Araus, L. J., 2004, “The problems of sustainable water use in the Mediterranean and research requirements for agriculture“, Annals of Applied Biology, 144, 259-272. Bell, S., Morse, S., 2004, “Experiences with sustainability indicators and stakeholder participation: a case study relating to a ‘Blue Plan’ project in Malta”, Sustainable Development, 12(1), 1-14. Bos, M.G., 1997, “Performance indicators for irrigation and drainage”, Irrigation and Drainage Systems, 11, 119–137. DPT (State Planning Organisation), 2001, “Long-term strategy and eighth five–year development plan, 2001-2005”, Ankara, Turkey. Dungumaro, E.W., Madulu, N.F., 2003, “Public participation in integrated water resources management: the case of Tanzania”, Physics and Chemistry of the Earth, 28, 1009–1014. Gardner-Outlaw, T., Engelman, R., 1997, “Sustaining water, easing scarcity: A second update”, Population Action International, Washington DC. Gleick, P. H., 1995, “Human population and water: To the limits in the 21st century”, Pacific Institute for Studies in Development, Environment, and Security, Oakland, California. GWP-TAC (Global Water Partnership - Technical Advisory Committee), 2000, “Integrated Water Resources Management”, TAC Background Papers no.4, Global Water Partnership. Hardi, P., Zdan, T. (eds), 1997, “Assessing sustainable development: Principles in practice” International Institute for Sustainable Development, Winnipeg, Manitoba, Canada. Jaspers, F.G.W., 2003, “Institutional arrangements for integrated river basin management”, Water Policy, 5, 77-90. Johnson, C. A., 1997, “Public participation and sustainable development: Counting the costs and benefits”, TDRI Quarterly Review, 12(2), 25-32. Kellett, B. M., Bristow, K. L., Charlesworth, P. B. 2005, “Indicator Frameworks for Assessing Irrigation Sustainability”, Land and Water Technical Report No. 01/05, Commonwealth Scientific and Industrial Research Organisation, Australia. Kendirli, B., Cakmak, B. Ucar, Y., 2005, “Salinity in the Southeastern Anatolia Project (GAP), Turkey: Issues and options”, Irrigation and Drainage, 54, 115–122.

299



Lorite, I.J., Mateos, L., Fereres, E., 2004, “Evaluating irrigation performance in a Mediterranean environment: Model and general assessment of an irrigation scheme”, Irrigation Science, 23, 7784. Marshall, G.R., 1999, “Economics of incorporating public participation in efforts to redress degradation of agricultural land”, 6th Annual Conference of the New Zealand Agricultural and Resource Economics Society, Christchurch, New Zealand. McCool, S.F., Stankey, G.H., 2004, “Indicators of sustainability: Challenges and opportunities at the interface of science and policy”, Environmental Management, 33(3), 294-305. Molden, D. J., Sakthivadivel, R., Perry, C.J., de Fraiture, C., Kloezen, W. H., 1998, “Indicators for comparing performance of irrigated agricultural systems”, Research Report 20, Colombo, Sri Lanka: International Water Management Institute. Mostert, E., 2003, “The challenge of public participation”, Water Policy, 5, 179-197. Ostrom, E., 1990, “Governing the commons: The evolution of institutions for collective action”, Cambridge: Cambridge University Press. Ostrom, E., Gardner, R., Walker, J., 1994, “Rules, games and common-pool resources” Ann Arbor: The University of Michigan Press. Pahl-Wostl, C., 2002, “Participative and stakeholder-based policy design, analysis and evaluation processes”, Integrated Assessment, 3, 3-14. Seppala, O.T., 2002, “Effective water and sanitation policy reform implementation: need for systemic approach and stakeholder participation”, Water Policy, 4, 367-388. Ünver, H.O., 1997, “Southeastern Anatolia Project (GAP)”, Water Resources Development, 13(4), 453-483.

300



301



CIHEAM Centre International de Hautes Etudes Agronomiques Méditerranéennes International Centre for Advanced Mediterranean Agronomic Studies

Conseil d'Administration / Governing Board Président / Chairman: Mouïn HAMZÉ

Membres / Members Albanie / Albania: Sali METANI Algérie / Algeria: Mohamed Foued RACHEDI Egypte / Egypt: ABD EL-AZIM EL-TANTAWI BADAWI Espagne / Spain: Vicente FLORES REDONDO France : Philippe BARRÉ Grèce / Greece: Elefterios TJAMOS Italie / Italy: Giuliana TRISORIO LIUZ ZI

Liban / Lebanon: Hady RACHED Malte / Malta: Salvino BUSUTTIL Maroc / Marocco: Fouad GUESSOUS Portugal : José Manuel ABECASSIS EMPIS Tunisie / Tunisia: Abdelaziz MOUGOU Turquie / Turkey: Vedat MIRMAHMUTOGULLARI

Comité Scientifique Consultatif / Scientific Advisory Committee Président / Chairman Teodoro Massimo MIANO Dipartimento di Biologia e Chimica Agro-forestale e Ambientale (DIBCA) Università di Bari, ITALIA Membres / Members Velesin PEÇULI Faculty of Agriculture, Agri-environment and Ecology Department Agricultural University of Tirana, ALBANIE Foued CHEHAT Institut National Agronomique d’El Harrach El Harrach, ALGERIE Jacques BROSSIER Président du Centre INRA D ijon Dijon, FRANCE Mohamad TALAL FARRAN Agricultural Research and Education Center (AREC) American Univerity of Beirut, LEBANON George ATTARD Institut of Agriculture U niversity of Malta, MALTA

302



Options Méditerranéennes O M is a CIHEAM se rie s de voted to the d evelopme nt of M e dite rranean agriculture. Having appe ared in the fo rm of a pe riod ical fro m 1 97 0 to 1 976 , the title O M has b ee n give n to the "Etud es" se rie s from 19 81 to 198 9. To date , O M inclu des th re e se rie s: Séminaires

Méditerranéens (Ser. A), Etudes et Recherches(Ser. B) and Cahier OM.

OM est une collection du CIHEAM dédiée au développement de l’agriculture méditerranéenne. Publiée en forme de périodique du 1970 à 1976, le titre OM a été donné à la collec tion Etu des du 1981 au 1989. Actuellement, OM comprend trois collections: Sé min aires M éd ite rran éen s (Ser. A ), Etude s et Re ch erche s (Se r. B) et Cah ier O M .

I NTERNATI ONAL CENTRE FOR ADVANCED MEDITERRANEAN AGRONOMI C STUDIES

INTERNATIONAL CENTRE FOR ADVANCED MEDITERRANEAN AGRONOMIC

WASAMED (WAter SAving in MEDiterranean agriculture) is a Thematic Network funded by the European Commission (INCO-MED Programme). The main objective of WASAMED is to establish a platform for effective Mediterranean dialogue on water saving in agriculture, contributing to improved management of limited water resourcesand sustainabledevelopment intheMediterraneanRegion. Specific objectivesof the Project are: #To improveregional co-ordinationof present and futureactionsinwatersavi ng; #To establish a Mediterranean-wide convention to strengthen communication and sharing of experience among relevant researchers, decision and policy makers, and end-users; #To develop water saving research projectsand actionsthat meet with the needsand concernsarising fromthedifferent Mediterraneancontexts; #To facilitateaccessof different stakeholdersto aneasy-to-use knowledge-base; #To create a framework and seek consensus to assist regional planning and EUfunding inwater resourcesman agement for theMediterraneanRegion. WASAMED Thematic Network involves 42 partners from decision-policy making Institutions, Universities and End-Users associations of 16 countries: Algeria, Cyprus, Egypt, Germany, Greece, Italy, Jordan, Lebanon, Malta, Morocco, Palestine, Portugal, Spain, Syria, Tunisia, Turkey. The Workshop of Amman is the fourth of a seriesof five Workshopsplanned between 2003 and 2006, each of them addressing respectively different issues of Water Saving: Participatory Irrigation Management and Cultural Heritage (in Turkey), Irrigation Systems Performance (in Tunisia), Use of non-conventional waters (in Egypt), Water Use Efficiency and Water Productivity (in Jordan), Integration and Harmonisation of technical water saving options and policies (in Lebanon). A final International Conference in Bari (Italy) on the whole context of Water Saving perspectivesintheMediterraneanwill closetheproject inFebruary2007.

CENTRE INTERNATIONAL DE HAUTES ETUDES AGRONOMIQUES CENTRE I NTERNATI ONAL DE HAUTES ETUDES AGRONOMIQUES MEDI TERRANEENNES

€ 50,00

ISBN : 2-8 535 2-355 -1 ISSN : 10 16 -1 228

303