Multi-criteria and multi-stakeholder assessment of

Land Use Policy 42 (2015) 131–140

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

Multi-criteria and multi-stakeholder assessment of cropping systems for a result-oriented water quality preservation action programme Clémence Ravier a,b,∗ , Lorène Prost c , Marie-Hélène Jeuffroy a,b , Alexander Wezel d , Laurette Paravano e , Raymond Reau a,b a

INRA, UMR211 Agronomie, BP 01, 78850 Thiverval-Grignon, France AgroParisTech, UMR211 Agronomie, BP 01, 78850 Thiverval-Grignon, France c INRA, UR1326 Sciences en Société, 77420 Champs-sur-Marne, France d Department of Agroecology and Environment, ISARA Lyon (Member of the University of Lyon), 23 rue Jean Baldassini, 69364 Lyon Cedex 07, France e Chambred’Agriculture de l’Yonne, 14 bis rue Guynemer, BP 50289, 89005 Auxerre Cedex, France b

a r t i c l e

i n f o

Article history: Received 14 August 2013 Received in revised form 30 May 2014 Accepted 14 July 2014 Keywords: Multi-criteria analysis Local stakeholders Water quality programme Cropping systems design Drinking water catchment

a b s t r a c t Water quality preservation programmes as currently proposed by public institutions are questionable with regards to efficient territorial development, yet necessary in catchment areas, and for the improvement of water quality. We provide a method based on a multi-criteria and multi-stakeholder analysis to assess cropping systems designed with farmers in a vulnerable drinking water catchment. Individual interviews with various stakeholders involved in the catchment improvement programme allowed us to gather a diversity of points of view on their preferences concerning various criteria describing cropping system sustainability and economic, social and environmental aspects. Five groups of stakeholders with different preferences were identified to define five scenarios of sustainability preferences. To support a result-oriented approach, achievable goals to improve water quality and contribute to sustainable development were chosen together with stakeholders. Then cropping systems designed with local farmers were assessed using the five scenarios of stakeholders’ preferences to open discussions on the implementation of alternative cropping systems within the drinking water catchment. The method was able to identify some cropping systems that, although very diverse, might assure the required drinking water quality, and were judged as theoretically highly sustainable by all the stakeholder groups. © 2014 Elsevier Ltd. All rights reserved.

Introduction Farming is responsible for diffuse water pollution, such as by nitrates and pesticides (Hénin, 1980; Panno and Kelly, 2004; Aubertot et al., 2005; Worrall et al., 2009). The European Water Framework Directive (WFD) states that all EU countries must reduce water pollution (Howarth, 2011). Member states have to delineate vulnerable zones within drinking water catchment areas and implement measures to conform to the WFD objective for water quality protection for the future (European Parliament, 2000). Farmers within these vulnerable zones are encouraged to implement practices that have less harmful environmental impacts (Laurent and Ruelland, 2011), and can be supported by public financial incentives to voluntarily adopt agro-environmental

∗ Corresponding author at: Avenue Lucien Brétignières, 78850 Thiverval-Grignon, France. Tel.: +33 1 30 81 52 95. E-mail address: [email protected] (C. Ravier). http://dx.doi.org/10.1016/j.landusepol.2014.07.006 0264-8377/© 2014 Elsevier Ltd. All rights reserved.

measures (AEMs). These measures are mostly action-oriented, as they rely on the implementation of recommended management practices. AEMs consist of a list of objectives concerning practices such as reduction of nitrogen fertilizer and pesticide use, guaranteeing a reward regardless of the results on water quality (Gerowitt and Bertke, 2003). However, these action-oriented measures, even when they are funded, are generally not sufficient to change farmers’ practices in the long term and may fail to achieve the desired objectives (Matzdorf and Lorenz, 2010). They are not adopted widely enough by farmers (Kuhfuss et al., 2012) but also, in some cases, they are not effective enough to achieve the required drinking water quality (Barnes et al., 2009; Howarth, 2011), meaning that alternative approaches must be sought for building efficient action programmes promoting new cropping systems appropriate to the local context (Burton and Schwarz, 2013; Hasund, 2013). In this respect, “result-oriented” measures are an interesting alternative as the reward is based on the result. Based on an obligation to produce results independently from the management practices, they have already proven their efficiency (Gerowitt

132

C. Ravier et al. / Land Use Policy 42 (2015) 131–140

and Bertke, 2003; Matzdorf and Lorenz, 2010; Sabatier et al., 2012; De Sainte Marie, 2014; Barataud et al., 2014). Result-oriented measures are probably relevant for water pollution issues since they allow a wider range of management strategies to be developed and implemented while relaxing constraints on management practices (Darradi et al., 2012; Sabatier et al., 2012). In addition, they seem to favour farmers’ participation in building the water quality programmes and taking management initiatives more suited to local conditions and specific socio-economic contexts (Benoît and Kockmann, 2008; De Girolamo and Lo Porto, 2012). The WFD has introduced some elements intended to shift water governance towards integrated water resources management, requiring action to be based on public participation (De Stefano, 2010). To encourage result-oriented programmes and long-term territorial organization of agriculture to restrict water pollution, local stakeholders should be involved in the processes of changing farming practices (Schwarz et al., 2008; Garin and Barraqué, 2012). In fact, the consequences of changing practices or land use in catchment areas may affect a larger public than just the agricultural sector. Hence, various stakeholders playing various roles in these zones should be considered when designing new practices and building alternatives (Kerselaers et al., 2013). Yet, the question remains as to the most efficient way to involve the large variety of stakeholders with different knowledge, expectations and objectives regarding agricultural activities and their impacts on socio-economics and the environment. The many stakeholders in a drinking water catchment may propose various solutions and have conflicting interests (ParraLópez et al., 2008), which can complicate the process of reaching a consensus on acceptable alternatives (Kerselaers et al., 2013). Also, these alternatives should be assessed as to their potential effects on water quality with simple, reliable, and easily monitored indicators, specified for local contexts (Melland et al., 2012; Hasund, 2013), and whether they suit the diverse objectives regarding the development of agricultural activities within a vulnerable area (De Stefano, 2010). In view of this diversity, a multi-stakeholder and multi-criteria assessment may be a way to address the complexity of the decision-making process for a particular region by identifying proposals able to satisfy the diversity of stakeholders’ preferences (Arnette et al., 2010; Luyet et al., 2012). MASC® (Multi-attribute Assessment of the Sustainability of Cropping systems) is a tool able to carry out sustainability assessment of cropping systems, and is thus well suited to rank them for their estimated sustainability (Sadok et al., 2008, 2009). As the weights of the various criteria defining sustainability can be changed, this tool can be adapted to the various stakeholders’ needs and requirements in a drinking water catchment. The aim of this study was to analyze the possibility of identifying cropping systems at a water catchment scale, able to satisfy the differing requirements of the various stakeholders involved. We first identify the diversity of these requirements for the chosen catchment area. Then we assess the proposed cropping systems for the catchment area for their impact on water quality and sustainability. Finally we discuss the proposed approach to elicit stakeholders’ priorities and assess alternatives to support decision makers within the water catchment area.

Materials and methods The case study: the drinking water catchment of Brienon-sur-Armançon, France Brienon-sur-Armançon is located in the province of Burgundy, north-eastern France. Brienon’s drinking water catchment is one of the 507 priority catchments for water management in France. The 1700 ha of agricultural land in the drinking water catchment

are currently dominated by annual field crops with very little livestock and only 2% of grassland (Agreste, 2010). Crop successions are mostly based on winter crops grown in short rotations, such as oilseed rape/cereals/cereals (Paravano, 2010). The water quality, analyzed monthly in the catchment, is characterized by high nitrate concentrations, sometimes up to the legal threshold of 50 mg l−1 , and pesticide residues whose concentrations occasionally exceed the legal maximum threshold (Duchenes, 2010). Water quality issues concern many stakeholders in the catchment: fifty-eight farmers, three agricultural cooperatives, four local municipalities, citizens, a regional agricultural chamber, technical institutes, and one water services provider whose role is to ensure that water policies are implemented in the catchment area. Farmers in the catchment suggested an alternative to action-oriented agroenvironmental measures, and proposed to take action by a means of a results-oriented programme. A steering group composed of local farmers, representatives from municipalities, technical institutes, chambers of agriculture, regional agencies, and the water services provider sought to devise a plan to assess a diversity of cropping systems regarding the need for water quality improvement and for sustainable development. The cropping systems to be assessed Two categories of cropping system were assessed: the current cropping systems, and alternative cropping systems designed to be adapted to the local water quality issues and to the local context. Current cropping systems Current practices are viewed as a baseline to assess farmers’ proposals for alternative cropping systems in a local and water quality context. To investigate farmers’ current practices, 18 interviews with farmers were carried out in the drinking water catchment (Paravano, 2010). The areas cultivated by these 18 farmers represent about 60% of the cultivated catchment area. From these interviews, eight cropping systems, considered as representative of the current situation, could be distinguished (A category in Table 1). They are based on three-year crop rotations (only winter crops) with high rates of nitrogen fertilization. We differentiated the crop rotations for two levels of pest control: intensive weed and pest control (A0), and intensive weed and integrated pest control (A1). This distinction allowed us to acknowledge the existing diversity of practices. Alternative cropping systems The second category (B) refers to the proposal of alternative cropping systems (Table 1). These were designed during a workshop involving farmers, agronomists, and technical advisors, according to the method proposed by Reau et al. (2012), with the aim of improving water quality in the catchment. Decreasing the intensity of pesticide use was expected with longer and more diversified rotations. Alternative cropping systems are based on 5-year crop rotations including a spring crop. We tripled the crop rotations with three different potential spring crops: pea (p), barley (b), or sunflower (s). To integrate the variability of farmers’ practices for the alternative cropping systems, we distinguished intensive weed and integrated pest control (B1), and integrated weed and pest control (B2). Reduction of nitrogen leaching was expected due to (i) the systematic use of cover crops before each spring crop, and after winter oilseed rape and pea, and (ii) lower fertilization rates based on integration of higher nitrogen soil mineralization due to the presence of a catch crop, in the predictive balance-sheet method (Hébert, 1969) and to the introduction of a legume crop in the rotation.


133

Table 1 Main cropping system categories assessed with multi-criteria evaluation based on current and alternative practices and crop rotations. A0, A1, B1 and B2 correspond to categories of cropping systems integrating diversity of farmers’ practices related to fertilization and pest management. Current practices A0 Crop rotation

Oilseed rape/winter wheat/winter barley

Cover crop

None

Fertilization rate

High fertilization rate (from 180 to 200 kg N ha−1 ) depending on crop (oilseed rape, wheat or barley) and soil type (deep or shallow)

Pest management

Intensive (pesticides and herbicides are always used to prevent yield loss)

Alternative practices A1

B1

B2

Oilseed rape/winter wheat/spring crop/winter wheat/winter barley After oilseed rape; before spring crop Medium fertilization rate (from 150 to 180 kg N ha−1 ) depending on crop (oilseed rape, wheat or barley (winter and spring)) and soil type, no fertilizer is applied on peas and less than 30 kg N ha−1 on sunflower Intensive weed control and integrated pest control (herbicides are always used as in the intensive management, but fungicides and insecticides are reduced according to the lower pest pressure due to the longer crop rotations)

Variants of soil and soil management practices In addition, to take account of the two main soil types of the fields in the catchment area, each cropping system category was duplicated and adapted for deep (A0, A1, B1, B2) and shallow (A 0, A 1, B 1, B 2) soils. Tillage practices can affect both management practices and strategies and water quality. Consequently, farmers designed each category of cropping system for both traditional ploughing (t) with soil inversion, and reduced tillage (r) without soil inversion. Thus, considering the spring crops (3), the diversity of pest management practices (2), of soil tillage (2) and of soil types (2), we described and assessed thirty-two cropping systems, of which eight characterized the current situation, and twenty-four were alternatives for the water quality programme. Multi-criteria and multi-stakeholder assessment of the sustainability of the cropping systems Description of the MASC® tool MASC® is a multi-criteria assessment tool that allows the assessment of cropping system sustainability on economic, social and environmental criteria, and that can be used as a basis for decision making (Sadok et al., 2009). It has a hierarchical structure that breaks down sustainability into aggregated criteria (Bohanec et al., 2008; Bergez, 2013). The three main dimensions of sustainability (environmental, economic and social) are distinguished, and each of them is progressively divided into basic criteria. The nature of the indicator and the ability to aggregate criteria make it possible to assess the sustainability of current cropping systems or alternatives before field implementation. The tool was evaluated by Sadok et al. (2009), and gave consistent results on several cropping systems. The economic dimension of MASC® consists of four basic criteria: profitability, independence, efficiency (combined to assess economic autonomy), and specific investment in materials needed. The social dimension is assessed through aggregation of contributions to employment, health risks, and complexity of implementation, the last being estimated with three basic criteria: physical constraints, number of crops, and number of specific operations. Finally the environmental dimension is assessed as the aggregation of environmental quality (which results from the aggregation of (i) water pollution risk, estimated with the four basic criteria, (ii)

Integrated (pest and weed management is partly controlled by the crop rotation and cultural practices, so pesticide and herbicide use is avoided as much as possible)

air pollution risk with three basic criteria, and (iii) soil quality with four basic criteria), abiotic resource conservation (which considers water, energy and phosphorus conservation), and biodiversity conservation (based on five basic criteria). Basic criteria are directly estimated either from field measurements, local experts’ assessment, or from agronomic models. They can for instance be characterized, for each cropping system, by using CRITER® software dedicated to calculating MASC® criteria. To homogenize qualitative and quantitative criteria, all criteria are then expressed as qualitative values on a scale from “very low” to “very high”. Threshold values separating these classes are either proposed in MASC’s user manual (Craheix et al., 2012), established from comparison with local data, or from the distribution of the range of results obtained for each cropping system assessed. Complete details about the definition of each basic criterion, the way of calculating it or estimating it (whether it be qualitative or quantitative) and the way of aggregating it with the others can be found in Sadok et al. (2009) and in the user manual giving in detail the different formula dedicated to each basic criterion (Craheix et al., 2012). The multi-criteria assessment, which aggregates MASC® elementary criteria until reaching overall sustainability, is defined by the weights allocated to each criterion in the hierarchical structure. We assessed the cropping systems with a double focus. First, we isolated two criteria related to water quality (nitrate leaching and pesticide leaching) to assess the potential of the cropping systems to meet the requirements that the steering group established for water quality. Second, we characterized each cropping system’s overall contribution to social, economic and environmental sustainability based on the multi-criteria and multi-stakeholders’ assessment. Water quality: assessing the efficiency of the cropping systems in terms of targeted goals at the catchment scale Water analysis carried in the catchment showed that the water quality of the catchment did not meet regulations as nitrate and pesticide content were above national safety norms. The steering group thus agreed on “nitrate leaching” (INO3 ) and “pesticide leaching to ground water” (I-Phygw ) as two relevant MASC indicators to assess water pollution risks at the catchment scale. These two indicators can be calculated for each cropping system proposal with CRITER® based on the Indigo® method (Bockstaller et al., 2008,

134


2009; Sadok et al., 2009). INO3 is the total estimated amount of nitrate leached during crop growth and after crop harvested. During crop growth, N leaching resulted from N fertilizer that is not absorbed by the crop. The amount of N leached is proportional to N fertilizer rate and depends on crop N uptake capacity, water balance at the moment of fertilization and rainfall risk after fertilization. After crop harvest, leaching results from over fertilization (if the total N rate applied exceed the rate recommended for the crop) and soil N mineralization potential (adjusted according to intercropping practices). Concerning pesticides, Iphygw is calculated considering practices, pesticides potential leaching, and toxicity of the active substances used. Based on an empirical reasoning, the steering group decided that averaged over the catchment, nitrogen leaching should not be more than 30 kg ha−1 year−1 . In fact, at the beginning of the work and according from the surveys carried out, nitrogen loss within the catchment was about 60 kg ha−1 and nitrate concentration in water around 50 mg l−1 . The stakeholders thus decided to target a nitrate concentration in water of 37 mg l−1 within about 10 years. The value of INO3 = 30 kg ha−1 year−1 was considered by the steering group ambitious enough to reach this target. Concerning Iphygw , in accordance with the threshold proposed by the designers of the indicator, the steering group decided that Iphygw should target a value of at least a grade of 8 on the 0 (maximum) to 10 (minimum) scale of pesticide leaching risk (Bockstaller et al., 1997). As these indicators are calculated for a cropping system they had to be upscaled to the catchment area. Therefore, we applied the values of INO3 and Iphygw calculated for each of the 32 systems to the area represented by each of these systems in the catchment (Mary et al., 1997). With this aim, we assumed that (i) the current distribution of cropping systems over the two main soil types, and (ii) the proportion of tillage practices in the catchment area, are fixed. As it is not possible to forecast the cropping systems that could be adopted by farmers, we also assumed that (iii) they will all choose the same spring crop, and (iv) that they will all adopt similar pest management types. Based on these assumptions, we then characterized the pollution risk at the catchment scale. Sustainability from various stakeholders’ points of view: individual interviews Fourteen interviews of 1–3 h were held with 21 stakeholders involved in the catchment protection programme. These interviews had a double purpose: to collect their own definition of cropping system sustainability from a diversity of stakeholders, and to clarify stakeholders’ perception of water quality (although the latter is not treated in this paper). We interviewed stakeholders with different roles and interests in the catchment: (i) farmers (F), who are responsible for part of the water pollution but at the same time who are all inhabitants of the area and so directly concerned by drinking water quality; (ii) municipality authorities (M), as they are representative of local authorities, most of them have activities involving environmental concerns or directly related to water quality issues; (iii) cooperative advisors (C), who are actors involved in agricultural activities; and (iv) the responsible environmental agency (EA) working specifically on water quality and responsible for ensuring that AEMs are well applied in the catchment. Small groups of 2 or 3 stakeholders and individual interviews were preferred to collective workshops in order to gather spontaneous and diverse opinions. In addition, individual meetings enabled us to grasp more clearly stakeholders’ opinions on what is a good water quality in the catchment. The interviews on cropping system sustainability definition consisted of collecting the weights attributed to each MASC® criterion by each stakeholder. At each node of the MASC® tree (Fig. 1), the sum of the weights assigned to the upper-level criteria is equal to %, e.g. Surface Water counts for 39% and ground water for 61% on the pesticide loss node (Fig. 1). Stakeholders allocated percentages

to the different criteria at each node. This step resulted in 21 sets of criteria weights (21 different MASC® trees) to assess cropping system sustainability.

Modelling stakeholders’ preferences to build scenarios To assess the overall contribution of the cropping systems to sustainability, we aimed at limiting the number of MASC® trees to be used. We thus decided to group stakeholders according to similarity in weighting criteria in the MASC® tree. In order to make consistent groups with similar perceptions of sustainability we chose a clustering method (Arnette et al., 2010). Statistical tests were performed on the 25 MASC® criteria whose weight differed from one stakeholder to another. A global weight was allocated to each criterion. It is the product of the percentages allocated to upper-level criteria of the branch. Statistical analyses were carried out with R statistical package (R Development Core Team, 2008). We first performed a Principal Component Analysis to define the most relevant factorial plans and retained only the most significant relations between variables and individuals within the data frame. We kept six axes of the PCA that explained 80% of the variance and carried out the hierarchical clustering on the principal component on these axes. For each tree resulting from the clustering analyses, each criterion was given the average of the weights given by each stakeholder constituting the group. We thus considered each tree as a scenario of preferences representative of the various stakeholders as they were grouped by the clustering method.

Results Making sustainability scenarios out of the diversity of stakeholders’ points of views Clustering automatically grouped the 21 MASC® trees designed from the interviews in groups that are significantly different from each other’s. We proposed to divide the hierarchical clustering into five groups, reducing the intra-group variance to 40% and the inertia gain to almost 60% (Table 2). Thus, it was statistically relevant and allowed a satisfactory number of groups to represent diversity of preferences (Fig. 2). Regarding the results of the clustering, the five groups can be characterized with the first fourth axes (Table 3). The clustering also provided the criteria weights on the four axes. On axis 1, criteria of biodiversity belonging to the environmental branch (fungicides, pesticides and herbicides) and criteria belonging to the social branch (system complexity) are negatively correlated to criteria of resource preservation and pest management related to water quality (data not shown). Environmental criteria like water quality and nitrates are the most correlated criteria to axis 2 while the economic dimension (profitability) is negatively correlated. On axis 3, economic autonomy is negatively correlated to investment in specific equipment; and criteria related to social dimension are negatively correlated to axis 4 while environment and soil quality are positively correlated. By analysing the correlation between the groups and the axes we can thus interpret the logic of each group. For instance, Table 3 shows that G1 is negatively correlated to axis 2: this group is thus characterized by priorities to economic criteria. Similarly, G2 (which is negatively correlated to axis 1) can be described as prioritizing resource preservation and pest management, G3 as prioritizing biodiversity and economic autonomy, G4 as prioritizing water issues and social dimension, and G5 as prioritizing environmental dimension and investment in material. This analysis thus resulted in defining five MASC trees representing five scenarios of


135

Fig. 1. Description of the MASC model, allowing ex-ante assessment of cropping systems (from Sadok et al., 2009). Numerical values in the tree represent weighting, it is the local weight of criterion in %. At each node the sum of the weights assigned to the upper-level criteria is equal to 100%, e.g. Surface Water counts for 39% and ground water for 61% on the pesticide losses node. The aggregation process goes along a branch - from basic criteria until the end of the branch devoted to the assessment of one of the three dimensions of sustainability i.e. Economic, Social and Environmental sustainability - according to the weight of each criterion to aggregate. The global weight represents the weight of a basic or aggregated criterion in the overall sustainability regarding it local weight and the weight of upper level criteria e.g. Surface Water counts for 39% on the pesticide losses node, which counts for 33% on the Water Pollution Risks, which counts for 35% on Environmental Quality, which counts 35% on Environmental Sustainability, which counts for 33% on the Overall Sustainability, which finally means that Surface Water counts for 0.5% of the Overall Sustainability.

preferences of the stakeholders, to be used for the assessment of the cropping systems. We thus built five MASC® trees corresponding to the five clusters. Table 4 shows the criteria’s weights for the five scenarios proposed. Most of the criteria used to define groups are

discriminant. For instance, we proposed G1 as prioritizing economic criteria. In Table 4 we can see that the average weight of the criteria related to economy are relatively high, compared with the other trees, and this especially for profitability. It also shows that this group is giving relatively high importance to profitability

Table 2 Inertia gain when increasing the number of clusters.

No of clusters Intra group variability (%) Inertia gain (%) Cumulative inertia gain (%)

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

79

62

51

41

34

28

24

20

16

13

10

7.5

5.7

4.3

3.1

2.4

1.7

1.0

0.37 0

21.1 16.5 10.9 10.2 7.0

5.9

4.8

3.9

3.8

3.2

3.0

2.1

1.8

1.3

1.3

0.72 0.70 0.68 0.62 0.37

21.1 37.6 48.5 58.7 65.7 71.6 76.4 80.3 84.1 87.3 90.7 92.8 94.6 95.9 97.2 97.9 98.6 99.3 99.9 100

Table 3 Description of the axes of the hierarchical clustering statistical analysis (R software).

G1 G2 G3 G4 G5

Axis

v test

Mean in category

Overall mean

sd in category

Overall sd

p value

2 1 1 3 2 4 4 3

−2.3 −3.4 2.3 2.5 2.3 −2.9 2.5 −2.3

−2.0 −5.1 2.0 2.8 1.4 −1.3 1.4 −1.3

−4.8e−11 2.2e−17 2.2e−17 2.2e−17 −4.8e−11 9.5e−11 9.5e−11 −4.8e−11

1.0 0.9 1.2 0.9 1.3 1.2 0.6 0.6

1.9 2.2 2.1 2.2 1.9 1.4 1.4 1.5

0.02 0.00 0.02 0.02 0.02 0.00 0.01 0.02

136


grouped together, meaning that the actors belonging to a given socio-professional category can have different perceptions of sustainability. By contrast, stakeholders that come from different professional categories can have similar perceptions of cropping systems’ sustainability. For instance, stakeholders from environmental agencies (EA) are grouped with stakeholders from municipalities and landowners. Nonetheless, the first three groups in which we find all the farmers and stakeholders from cooperatives correspond to scenarios that focus on the economic dimension, reduction of production cost and biodiversity conservation. Those scenarios emerged from stakeholders involved in agriculture as an economic activity. At the other extreme there are stakeholders for whom the preference is not economic activity with agriculture, but environmental protection and resource preservation. Those groups correspond to scenarios that focused on water issues, environment, employment, and that included the economic dimension only through the equipment investment to change practices. Selection of cropping systems for the water quality preservation programme

Fig. 2. Representation of the five groups (G1, G2, G3, G4, G5) found with the hierarchical clustering from R software. Individuals are labelled regarding professional origin, (M, municipalities; C, cooperatives; F, farmers; EA, environmental agencies; O, landowners). Table 4 Global weights of the criteria used to assess cropping system sustainability for the five scenarios resulting from the hierarchical clustering analysis.

Economy Profitability Autonomy Equipment Social Complexity Employment Toxicity Environment Env. quality Air quality Soil quality Water quality Pest Nitrates Phosphates Resources Energy Phosphorus Biodiversity Insecticides Fungicides Herbicides Species diversity

G1

G2

G3

G4

G5

0.47 0.24 0.11 0.12 0.30 0.06 0.05 0.11 0.23 0.09 0.03 0.03 0.03 0.02 0.01 0.00 0.06 0.05 0.01 0.08 0.02 0.01 0.01 0.04

0.60 0.21 0.18 0.14 0.00 0.00 0.00 0.23 0.40 0.23 0.07 0.09 0.07 0.05 0.02 0.00 0.18 0.11 0.07 0.00 0.00 0.00 0.00 0.00

0.46 0.20 0.19 0.07 0.26 0.14 0.00 0.05 0.28 0.12 0.02 0.07 0.06 0.01 0.01 0.01 0.03 0.02 0.01 0.13 0.03 0.02 0.02 0.05

0.30 0.09 0.15 0.06 0.32 0.06 0.07 0.10 0.38 0.16 0.05 0.05 0.03 0.03 0.02 0.01 0.12 0.07 0.05 0.10 0.03 0.02 0.02 0.03

0.43 0.19 0.08 0.17 0.22 0.05 0.02 0.10 0.35 0.15 0.03 0.05 0.06 0.03 0.02 0.01 0.07 0.04 0.03 0.13 0.02 0.02 0.02 0.06

Most representative criteria for each group are identified in bold.

regarding the weight attributed to other criteria. Scenario 1, established from Group 1, relied mainly to the economic dimension of sustainability. Group 2 with their scenario underlined the will to reduce the use of inputs such as energy, non-renewable resources and pesticides. Scenario 3 focused on biodiversity preservation, social dimension and economic autonomy. Scenario 4 emphasized water issues and the development of employment in agriculture. Scenario 5 gave priority to environment and soil quality with practices that do not rely on equipment investment. If we look further into those five scenarios to consider the stakeholders that are in those five groups, it firstly appears that the preferences are not necessarily bound to stated affiliations of participants (Fig. 2). For instance farmers (F) are not all

As explained in section “Description of the MASC® tool”, the cropping systems were assessed with a double focus: firstly by considering the two criteria related to water quality (INO3 for nitrate leaching and Iphygw for pesticide leaching) to assess the potential of the cropping systems to meet the requirements that the steering group established for water quality, and secondly by considering their overall contribution to social, economic, and environmental sustainability based on the multi-criteria and multistakeholders’ assessment. Water quality targeted at the cropping system for the entire catchment land use The values of IPhygw (ha−1 year−1 ) and INO3 (ha−1 year−1 ) calculated for current (A) cropping systems and extrapolated to the catchment scale according to the current distribution of land use, were confirmed to be always below the targeted value of 8 and above the limit of 30 kg ha−1 year−1 (Table 5) respectively. The current systems (A) thus did not achieve the objectives assigned to water quality at the catchment scale. Although some current cropping systems (A0.t, A1.t) had an IPhygw above 8, they did not satisfy the nitrate criteria. In contrast, most alternative cropping systems (B) achieved the objectives assigned to nitrate leaching with an INO3 below the threshold value. Individually, some alternative cropping systems are above the limit of 30 kg of nitrate leached per year and per hectare (e.g. those managed with reduced tillage on shallow soils), however the upscaling hypothesis leads to mix them with those of the same category but below the threshold (e.g. those managed with reduced tillage on deep soils) and finally, it results in nitrates losses at the catchment scale below the threshold. Concerning the objectives assigned to pesticide losses, at the catchment scale, only category B2 reached an IPhygw above the threshold value of 8. However, some alternative cropping systems from category B1 (e.g. those managed with ploughing on deep soil) were also satisfactory. Regarding our upscaling hypothesis, only cropping systems from category B2 could satisfy the objectives assigned to the catchment; however, those results show that it would also be acceptable to choose some alternatives cropping system to B1. Multi-criteria and multi-stakeholders’ assessment of the contribution of the cropping systems to sustainability The multi-criteria and multi-stakeholder assessment was performed for all cropping systems, whether or not they matched requirements on pesticide and nitrate at the catchment scale. The assessment compared the cropping systems proposed and designed by the farmers with the five scenarios established from


137

Table 5 Pesticide (I-Phygw ) and nitrate (INO3 ) losses estimated for each current cropping system (A) and alternative cropping system (B) including pea (p) barley (b), or sunflower (s) with three different pest management (0, 1 or 2, see Table 1). Results are extrapolated from field scale to the catchment scale in respect of the distribution of surface managed with ploughing (t) and reduced tillage (r) for two soil types (deep and shallow) Catchment’s land use is 437 ha with deep soils within 216 ha managed with traditional ploughing and 221 ha with reduced tillage. 247 ha are shallow soils within 207 ha managed with traditional ploughing and 40 ha managed with reduced tillage. Category

Current systems

Cropping system Soil type Pest management

Alternative systems with pea

7.1 38

7 38

5.7 50

221

207

40

A1.r A 1.t A 1.r B1pt B1pr B 1pt B 1pr Deep Shallow Shallow Deep Deep Shallow Shallow 1 1 1 1 1 1 1

B2pt B2pr B 2pt B 2pr Deep Deep Shallow Shallow 2 2 2 2

8.2 31

7.1 38

7 38

6 50

8.3 19

7.7 28

7 22

5.8 35

8.3 19

8.9 28

7.2 22

8 35

216 7.3 36

221

207

40

216 7.6 24

221

207

40

216 8.1 24

221

207

40

A0.t A0.r A 0.t A 0.r A1.t Deep Deep Shallow Shallow Deep 0 0 0 0 1

Field scale I-Phygw (ha−1 year−1 ) 8.2 INO3 (kg ha−1 year−1 ) 31 Catchment scale 216 Surface (ha) I-Phygw (ha−1 year−1 ) 7.3 INO3 (kg ha−1 year−1 ) 36

Category

Alternative systems with barley

Cropping system Soil type Pest management

B1bt Deep 1

B1br Deep 1

B 1bt Shallow 1

B 1br Shallow 1

B2bt Deep 2

B2br DEEP 2

B 2bt Shallow 2

B 2br Shallow 2

B1st Deep 1

B1sr Deep 1

B 1st Shallow 1

B 1sr Shallow 1

B2st Deep 2

B2sr Deep 2

B 2st Shallow 2

B 2sr Shallow 2

Field scale I-Phygw (ha−1 year−1 ) INO3 (kg ha−1 year−1 )

8.6 13

7.5 29

7.8 16

5.5 35

8.6 13

8.7 29

7.7 16

7.5 35

8.4 15

7.9 25

7.5 19

6.4 31

8.5 15

9.1 25

7.7 19

8.1 31

Catchment scale Surface (ha) I-Phygw (ha−1 year−1 ) INO3 (kg ha−1 year−1 )

216 7.82 20

221

207

40

216 8.3 20

221

207

40

216 7.9 20

221

207

40

216 8.4 20

221

207

40

Alternative systems with sunflower

I-Phygw : indicator of risk of water pollution by pesticides; INO3 indicator of risk of nitrate loss to water (Bockstaller et al., 2008).

Table 6 Assessment by the six groups of stakeholders of the overall sustainability for all cropping systems (attaining or not the water quality objectives). Results are distinguished for current (A) and alternative (B) cropping systems. Within those categories, results are distinguished for deep and shallow soils and for tillage practice (t = traditional tillage, r = reduced tillage). For alternative cropping systems (B), the different types of spring crop are distinguished (b = barley, s = sunflower, p = pea).

Category: current (A)/alternative (B) – Soil: deep/shallow (’) – Pest management: intensive weed and pest (0)/intensive weed and integrated pest (1)/integrated (2) – If spring crop: pea(p) /sunflower(s) /barley(b) – Tillage: traditional plough (t)/reduce tillage (r). ++ very high sustainability; + high; − low; −− very low (none of the cropping system assessed was judged very low).

138


the five clusters of stakeholders’ preferences. The contribution of the various cropping systems to sustainability showed variability according to the five scenarios (Table 6). This variability can be structured by distinguishing three types of cropping system: those that are never judged highly sustainable by any group of stakeholders, those called ‘specific’ that may be judged highly sustainable by some groups but not by others, and those called ‘robust’ that are judged highly sustainable by all the participants. The results of the multi-criteria and multi-stakeholder assessment showed that most of the current cropping systems (A0r on deep soil and all current cropping systems on shallow soils) were never judged highly sustainable. Some current cropping systems on deep soil (A0t, A1t and A1r) are highly sustainable for the three groups that include stakeholders involved in agriculture as an economic activity but they are not satisfactory for the others groups (Table 6). Regarding the stakeholders for whom agriculture is not an economic activity, those cropping systems were flawed on the environmental dimension (and especially on resources preservation and biodiversity of the branch, data not shown). We have shown that alternative cropping systems with level 1 of pest management did not reach the water quality requirement for IPhygw at the catchment scale (Table 5). Among them, some, such as B 1r(p) , B 1r(s) , B 1r(b) , are never judged highly sustainable (Table 6). And some, such as B1t(b) , B1r(b) , B1r(s) , B1r(p) , B 1t(s) and B 1r(p) , are judged highly sustainable for groups whose economic activity is linked with agriculture. As a consequence, even though the systems do not satisfy the objectives assigned to water quality, some farmers might adopt them. And some others such as B1t(s) , B1t(p) and B 1t(p) are robust, i.e., judged highly sustainable in all the scenarios, but they do not achieve the IPhygw target at the catchment scale. As shown in Table 5, alternative cropping systems with level 2 of pest management are achieving nitrate and pesticide objectives, however they are specific, i.e. not always highly satisfactory for all groups (Table 6). For instance, all cropping systems including spring barley are highly satisfactory for groups of agricultural stakeholders (Groups 1, 2, 3) but only medium-satisfactory for groups that favour “environment and water quality” (Groups 4, 5). Regarding basic performance, those cropping systems were unfavourable for criteria of crop diversity and energy consumption (data not shown). On the one hand, the crop rotation includes three cereal crops (two winter wheats and a spring barley) while the criterion “crop diversity” considers diversification in terms of crop diversity but not in diversity of practices. On the other hand, barley requires more fertiliser than peas or sunflowers; consequently energy consumption is higher for those cropping systems. Considering the relative importance of nitrate in water, environment and/or biodiversity for the groups of environmental and water quality stakeholders, the overall sustainability of cropping systems with spring barley is never very high. Another example of a specific cropping system is B2r(p) , which includes peas with reduced tillage. Regarding the performance of basic criteria, it requires intense pest management, which is harmful to biodiversity conservation (data not shown). Combined with the relative importance of this criterion for the groups of environmental and water quality stakeholders, it decreased the overall sustainability of those cropping systems. The assessment also provided robust cropping systems. Although the various stakeholders apparently had divergent definitions of sustainability (expressed through the five scenarios of the MASC® tree), we found that cropping systems including either peas or sunflower (spring crops) are favoured as proposals for the water quality programme (Table 6). We found four alternatives on deep soil: two with traditional ploughing (B2t(p) , B2t(s) ) and two with reduced tillage (B2r(p) , B2r(s) ), plus, two on shallow soils with traditional ploughing (B 2t(p) , B 2t(s) ). We consider those systems likely to find support among stakeholders as possible alternatives

for the water quality action programme. As their sustainability was judged very high through the five scenarios of criteria hierarchy, they are consistent solutions to balance the interests of the various stakeholders. Discussion MASC® to elicit preferences in a multi-stakeholder environment Our results are helpful in deciding how to elicit multistakeholder priorities related to environmental management of agricultural activities. First, the results show that professional affiliation is not a good indicator of these priorities, although most programmes involving various stakeholders’ decision-making processes assume this to be the case (e.g. Reed, 2008, Strager and Rosenberger, 2006). Like Arnette et al. (2010), we found that grouping stakeholders according to their stated affiliation may even sometimes mask their diversity of preferences. Second, in contrast to Saarikoski et al. (2013), we used average preferences to build different scenarios, and not to build one scenario based on a compromise between stakeholders’ priorities. We could show that our scenarios were contrasted and represented the range of different views and perspectives on the water quality programme. These scenarios did not conceal the fact that stakeholders had clearly different roles in the catchment area and supported divergent, or even contradictory, expectations about agricultural activities within a drinking water catchment. Third, grouping stakeholders according to their weighting of MASC® criteria showed that our assessment model was sensitive to the different groups’ weightings, especially between scenarios prioritizing either economic (groups of agricultural stakeholders) or environmental (groups of environment and water quality stakeholders) criteria. The elicitation based on MASC criteria thus demonstrated that priorities for sustainable development are disputed, depending on the degree of involvement in agricultural activities. Finally, by eliciting stakeholders’ priorities for alternative cropping systems through weighting MASC criteria, we obtained both information concerning stakeholders’ priorities related to their involvement in the programme and the framework to assess the alternative cropping systems in view of various priorities. However, we may question whether the use of such a method would be possible for a larger public, as individual face-to-face interviews were needed to elicit the weights of a wide range of criteria. Sustainability and flexibility for a result-oriented programme The results obtained for water quality criteria highlighted the need to move to new practices more adapted to environmental issues. We showed that new alternative cropping systems proposed (B) could potentially improve water quality, as some cropping systems designed with farmers were fundamentally suited to water quality objectives assigned for the result-oriented water quality programme. In addition, they were quite varied, providing farmers with a range of choices. The collective nature of our approach, based on participation rather than consensus-building, improves flexibility in terms of possible alternatives: we did not arrive at a single proposal for acceptable alternatives but a diversity of solutions acceptable for a local water quality programme. Compared to situations that aim to reach a consensual solution (Hajkowicz, 2008), the wide range of proposals suitable for the water quality programme can be seen as a means to achieve greater farmer participation, which would lead to improve the environmental effect of the AEMs (Mettepenningen et al., 2013). As soon as a proposal can be assessed by various means, flexibility of choice of alternative cropping systems to achieve


the objectives assigned for the water quality programme can be seen as relevant means to implement efficient result-oriented measures. This flexibility can also be enlarged if the ‘specific’ cropping systems that reach the water quality requirement are improved to become acceptable to all stakeholders (Hasund, 2013). In fact, the analysis of MASC® results yielded elements to identify sources of disagreement about some alternative cropping systems between groups and a better understanding of what makes these cropping systems ‘specific’ suggests interesting ways to improve them and make them robust. For instance, knowing that reduced tillage is widespread within the catchment (about 30% of the area) and having shown that the majority of alternative cropping systems using this practice were not very highly sustainable, it is interesting to propose some ways to redesign them. Therefore, this analysis can be a starting point to redesign some cropping systems on the basis of an initial proposition, with consideration of stakeholders’ perception of sustainability. And as a matter of fact, in our case study, this analysis allowed the farmers to provide the steering group with some changes in their cropping systems that had proven to be efficient. Their argumentation, based on the results of the scenarios, was judged convincing by the steering group and the local action plan was written from farmers’ proposals. Following the same idea of favouring the flexibility of choice and a process of continuous improvement of proposals for the catchment area, we should discuss our hypothesis of a generalized shift in practices on the whole catchment area, which in practice is never the case. Even if the results show that it seems very difficult to build an efficient water quality programme if farmers continue with current cropping systems, we should assume that a part of the catchment area may remain farmed using current practices. Indeed, many current cropping systems (A) were identified as quite or even very sustainable in the view of agricultural stakeholders. This can be due to a bias in the method to assess cropping system that is too indulgent or permissive, but we should accept that farmers may not be willing to change their practices if they are satisfied with the current cropping system. As the water quality objective is defined at the catchment scale and not applicable to individual farmers (as opposed to current AEMs), it may be possible to accept some flexibility in land use changes. As a consequence, it could be relevant to test finer assumptions of distributions of cropping systems for each soil type in the catchment area and assess their performance in terms of I-Phygw and INO3 to provide stakeholders with more nuanced proposals. In any case, these conclusions confirm, in line with Schwarz et al. (2008) for instance, that result-oriented action programmes, as opposed to those based on an obligation to adhere to certain procedures, are an interesting way to provide flexibility in the means available to the actors, notably the farmers’ management practices. This gives interesting leads to discuss the way the current environmental measures are devised. Mostly based on “obligation of means” (Burton and Schwarz, 2013), they in fact do not promote the flexibility of result-oriented schemes, although they have proven to be an efficient way to enhance farmers’ participation in environmental improvement (Hasund, 2013). However, one should recognize that result-oriented programmes may be more complex to implement and control. These approaches involve rethinking the content and nature of ‘water action programmes’ in drinking water catchment areas and to adapt them to the result-oriented principle. They also require a different organization and new competencies and roles (Klerkx et al., 2010) to support farmers in specific ways (Matzdorf and Lorenz, 2010), and of the collective dynamics. In fact, our method builds on a prevailing role of the steering group to take time to discuss the objectives, assess the proposals, and encourage flexibility in implementation of result-oriented programmes.

139

Conclusion The Water Directive Framework requires farmers to change their practices in vulnerable zones to achieve good water quality. We propose in this article an effective way of characterizing cropping systems and assessing the appropriateness of local solutions to solve environmental issues considered as important by a wide range of stakeholders. The assessment was managed with regard to various opinions to facilitate the decision-making process for the improvement of a current water quality programme in vulnerable zones. Its originality lies in the method proposed, which assesses cropping systems with a view to supporting collective agreement for a result-oriented water quality programme with flexibility of alternative cropping systems to be implemented in the catchment area. Even though preferences differed as to a hierarchy of criteria to assess sustainability, we found cropping systems meeting all stakeholders’ requirements regarding both water quality and contribution to sustainable development. As we did not seek a consensus among stakeholders, the results, even the divergent ones, although not satisfying the conflicting aims of stakeholders, can help to revive discussions to build a water quality programme suited to the local context. This method helps to build a favourable collective working environment, able to constructively discuss a wide range of proposals by predicting their probable results with a view to continuous improvement. Acknowledgments We thank all the stakeholders that participated in interviews, workshops and meetings. We thank the National Research Association (ANR-08-STRA-12-02) for funding this work as part of the “Crop production, public politics and environment” project and the National Agency for Water and Aquatic Environment in funding this work as part of the “Accompanying shift in agricultural activities in drinking water area” project (12-5-2). We thank Alan Scaife for reviewing the English. The package for multi-criteria assessment is available at: https://wiki.inra.fr/wiki/deximasc/Main/WebHome. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.landusepol. 2014.07.006. References Agreste, 2010. Recensement Agricole 2010, Exploitation par orientation technico économique (OTEX), Bourgogne. Arnette, A., Zobel, C., Bosch, D., Pease, J., Metcalfe, T., 2010. Stakeholder ranking of watershed goals with the vector analytic hierarchy process: effects of participant grouping scenarios. Environ. Model. Softw. 25, 1459–1469. Aubertot, J.N., Doré, T., Ennaifar, S., Ferré, F., Fourbet, J.F., Schneider, O., 2005. Integrated crop management requires to better take into account cropping systems in epidemiological models. In: Proceedings of the 9th International Workshop on Plant Disease Epidemiology, 11–15 April, Landerneau, France. Barataud, F., Aubry, C., Wezel, A., Mundler, P., Fleury, P., 2014. Management of water catchment areas in cooperation with agriculture. Experiences from Germany and France. Land Use Policy 36, 585–594. Barnes, A.P., Willock, J., Hall, C., Toma, L., 2009. Farmer perspectives and practices regarding water pollution control programmes in Scotland. Agric. Water Manag. 96, 1715–1722. Benoît, M., Kockmann, F., 2008. L’organisation des systèmes de culture dans les bassins d’alimentation de captage: innovations, retours d’expériences et leçons à tirer. Ingénieries 54, 19–32. Bergez, J.E., 2013. Using a genetic algorithm to define worst-best and best-worst options of a DEXi-type model: Application to the MASC® model of croppingsystem sustainability. Comput. Electron. Agric. 90, 93–98. Bockstaller, C., Girardin, P., van der Werf, H.M.G., 1997. Use of agroecological indicators for the evaluation of farming systems. Eur. J. Agron. 7, 261–270.

140


Bockstaller, C., Guichard, L., Makowski, D., Aveline, A., Girardin, P., Plantureux, S., 2008. Agri-environmental indicators to assess cropping and farming systems. A review. Agron. Sustain. Dev. 28, 139–149. Bockstaller, C., Guichard, L., Keichinger, O., Girardin, P., Galan, M.B., Gaillard, G., 2009. Comparison of methods to assess the sustainability of agricultural systems. A review. Agron. Sustain. Dev. 29, 223–235. Bohanec, M., Messéan, A., Scatasta, S., Angevin, F., Griffiths, B., Krogh, P.H., Zˇ nidarˇsiˇc, M., Dˇzeroski, S., 2008. A qualitative multi-attribute model for economic and ecological assessment of genetically modified crops. Ecol. Model. 215, 247–261. Burton, R.J.F., Schwarz, G., 2013. Result-oriented agri-environmental schemes in Europe and their potential for promoting behavioural change. Land Use Policy 30, 628–641. Craheix, D., Angevin, F., Bergez, J.E., Bockstaller, C., Colomb, B., Guichard, L., Reau, R., Dore, T., 2012. MASC® 2.0, un outil d’évaluation multicritères pour estimer la contribution des systèmes de culture au développement durable. Innov. Agron., 35–48. Darradi, Y., Saur, E., Laplana, R., Lescot, J.M., Kuentz, V., Meyer, B.C., 2012. Optimizing the environmental performance of agricultural activities: a case study in La Boulouze watershed. Ecol. Indic. 22, 27–37. De Girolamo, A.M., Lo Porto, A., 2012. Land use scenario development as a tool for watershed management within the Rio Mannu Basin. Land Use Policy 29, 691–701. De Sainte Marie, C., 2014. Rethinking agri-environmental schemes. A result-oriented approach to the management of species-rich grasslands in France. J. Environ. Plan. Manag. 57 (5), 704–719. De Stefano, L., 2010. Facing the water framework directive challenges: a baseline of stakeholder participation in the European Union. J. Environ. Manag. 91, 1332–1340. Duchenes, V., 2010. Etude “B.A.C” Diagnostic de territoire. Chambre d’Agriculture de l’Yonne, 14 bis rue Guynemer – 89015 Auxerre Cedex. European Parliament, Council of the European Union, 2000. Directive 2000/60/CE of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy. Garin, P., Barraqué, B., 2012. Why are there so few cooperative agreements between farmers and water services in france? Water policies and the problem of land use rights. Irrig. Drain. 61, 95–105. Gerowitt, B., Bertke, E., Hespelt et, S.-K., Tute, C., 2003. Towards multifunctional agriculture – weeds as ecological goods? Weed Res. 43, 227–235. Hajkowicz, S.A., 2008. Supporting multi-stakeholder environmental decisions. J. Environ. Manag. 88, 607–614. Hasund, K.P., 2013. Indicator-based agri-environmental payments: a payment-byresult model for public goods with a Swedish application. Land Use Policy 30, 223–233. Hébert, J., 1969. La fumure azotée du blé tendre d’hiver. Bull. Tech. d’Information 244, 755–766. Hénin, S., 1980. Rapport du groupe de travail; activités agricoles et qualité des eaux. Ministère de l’Agriculture, Ministère de l’environnement et de la qualité de vie. Howarth, W., 2011. Diffuse water pollution and diffuse environmental laws tackling diffuse water pollution in England, Report by the Comptroller and Auditor General, HC 186 Session 2010–2011, 6 July 2010. J. Environ. Law 23, 129–141. Kerselaers, E., Rogge, E., Vanempten, E., Lauwers, L., Van Huylenbroeck, G., 2013. Changing land use in the countryside: stakeholders’ perception of the ongoing rural planning process in Flanders. Land Use Policy 32, 197–206. Klerkx, L., Aarts, N., Leeuwis, C., 2010. Adaptive management in agricultural in innovation systems: the interactions between innovation networks and their environment. Agric. Syst. 103, 90–400. Kuhfuss, L., Jacquet, F., Préget, R., Thoyer, S., 2012. Le dispositif des MAEt: une fausse bonne idée? Rev. d’Etude Agric. Environ. 93 (4), 395–411.

Laurent, F., Ruelland, D., 2011. Assessing impacts of alternative land use and agricultural practices on nitrate pollution at the catchment scale. J. Hydrol. 409, 440–450. Luyet, V., Schlaepfer, R., Parlange, M.B., Buttler, A., 2012. A framework to implement stakeholder participation in environmental projects. J. Environ. Manag. 111, 213–219. Mary, B., Beaudoin, N., Benoit, M., 1997. Prévention de la pollution nitrique à l’échelle du bassin d’alimentation en eau. In: Lemaire, G.J., Nicolardot, B. (Eds.), Maîtrise de l’azote dans les agroecosystèmes. Ed. Quae, Paris, pp. 289–312. Matzdorf, B., Lorenz, J., 2010. How cost-effective are result-oriented agrienvironmental measures?—an empirical analysis in Germany. Land Use Policy 27, 535–544. Melland, A.R., Mellander, P.-E., Murphy, P.N.C., Wall, D.P., Mechan, S., Shine, O., Shortle, G., Jordan, P., 2012. Stream water quality in intensive cereal cropping catchments with regulated nutrient management. Environ. Sci. Policy 24, 58–70. Mettepenningen, E., Vandermeulen, V., Delaet, K., Van Huylenbroeck, G., Wailes, E.J., 2013. Investigating the influence of the institutional organization of agrienvironmental schemes on scheme adoption. Land Use Policy 33, 20–30. Panno, S., Kelly, W., 2004. Nitrate and herbicide loading in two groundwater basins of Illinois’ sinkhole plain. J. Hydrol. 290, 229–242. Paravano, L., 2010. Etude “B.A.C” Diagnostic des pratiques. Chambre d’Agriculture de l’Yonne, 14 bis rue Guynemer – 89015 Auxerre Cedex. Parra-López, C., Calatrava-Requena, J., de-Haro-Giménez, T., 2008. A systemic comparative assessment of the multifunctional performance of alternative olive systems in Spain within an AHP-extended framework. Ecol. Econ. 64, 820–834. R Development Core Team, 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3900051-07-0, Available at: http://www.R-project.org (verified 28.11.08). Reau, R., Monnot, L.A., Schaub, A., Munier-Jolain, N., Pambou, I., Bockstaller, C., Cariolle, M., Chabert, A., Dumans, P., 2012. Les ateliers de conception de systèmes de culture pour construire, évaluer et identifier des prototypes prometteurs. Innov. Agron. 20, 5–33, http://www7.inra.fr/ciag/revue/volume 20 juillet 2012 Reed, M.S., 2008. Stakeholder participation for environmental management: a literature review. Biol. Conserv. 141, 2417–2431. Saarikoski, H., Mustajoki, J., Marttunen, M., 2013. Participatory multi-criteria assessment as ‘opening up’ vs. ‘closing down’ of policy discourses: a case of old-growth forest conflict in Finnish Upper Lapland. Land Use Policy 32, 329–336. Sabatier, R., Doyen, L., Tichit, M., 2012. Action versus result-oriented schemes in a grassland agroecosystem: a dynamic modelling approach. PLoS ONE 7, e33257. Sadok, W., Angevin, F., Bergez, J.E., Bockstaller, C., Colomb, B., Guichard, L., Reau, R., Doré, T., 2008. Ex ante assessment of the sustainability of alternative cropping systems: implications for using multi-criteria decision-aid methods. A review. Agron. Sustain. Dev. 28, 163–174. Sadok, W., Angevin, F., Bergez, J.E., Bockstaller, C., Colomb, B., Guichard, L., Reau, R., Messean, A., Dore, T., 2009. MASC® , a qualitative multi-attribute decision model for ex ante assessment of the sustainability of cropping systems. Agron. Sustain. Dev. 29, 447–461. Schwarz, G., Moxey, A., McCracken, D., Huband, S., Cummins, R., 2008. An analysis of the potential effectiveness of a Payment-by-Results approach to the delivery of environmental public goods and services supplied by Agri-Environment Schemes. Report to the Land Use Policy Group, UK. Macaulay Institute, Pareto Consulting and Scottish Agricultural College, 108 pp. Strager, M.P., Rosenberger, R.S., 2006. Incorporating stakeholder preferences for land conservation: weights and measures in spatial MCA. Ecol. Econ. 57, 627–639. Worrall, F., Spencer, E., Burt, T.P., 2009. The effectiveness of nitrate vulnerable zones for limiting surface water nitrate concentrations. J. Hydrol. 370, 21–28.