Using climate information for supporting climate change adaptation in ...

Climatic Change (2010) 103:537–554 DOI 10.1007/s10584-009-9771-3

Using climate information for supporting climate change adaptation in water resource management in South Africa Gina Ziervogel · Peter Johnston · Margaret Matthew · Pierre Mukheibir

Received: 6 October 2008 / Accepted: 30 October 2009 / Published online: 19 January 2010 © Springer Science+Business Media B.V. 2009

Abstract Water resources, and in particular run-off, are significantly affected by climate variability. At present, there are few examples of how the water management sector integrates information about changing intra-annual climate conditions in a systematic manner in developing countries. This paper, using the case study of Cape Town in the Western Cape, South Africa, identifies processes and products to facilitate increased uptake of seasonal climate forecasts among water resource managers. Results suggest that existing seasonal forecasts do not focus enough on specific users’ needs. In order to increase uptake, forecasts need to include information on the likely impact of precipitation variability on runoff and water availability. More opportunities are also needed for those with climate knowledge to interact with water resource managers, particularly in the developing country context where municipal managers’ capacity is strained. Although there are challenges that need to be overcome in using probabilistic climate information, seasonal forecast information tailored to the needs of water resource planners has the potential to support annual planning and is therefore a means of adapting to climate change.

G. Ziervogel (B) · P. Johnston · M. Matthew Department of Environmental and Geographical Science, University of Cape Town, Cape Town, South Africa e-mail: [email protected] P. Johnston e-mail: [email protected] M. Matthew e-mail: [email protected] P. Mukheibir Wannon Water, Warrnambool, Australia e-mail: [email protected]

538

Climatic Change (2010) 103:537–554

1 Introduction Climate change presents a significant concern for the availability, access and quality of water resources, particularly in Africa (Boko et al. 2007; Schulze 2005). Yet, managing water resources in a variable climate context is challenging. Studies that have explored how seasonal climate forecasts might be used in managing water resources suggest that there are many limitations to using this information (Johnston et al. 2007; Patt et al. 2007; Pulwarty and Redmond 1997; Ritchie et al. 2004; Yarnal et al. 2006). Although the water sector may have specific challenges, the use of seasonal forecasts to improve livelihoods has been evident in a range of cases including in the agricultural sector at both the subsistence and commercial scale (Archer 2003; Patt et al. 2007), for managing food relief and for malaria prevention (Hellmuth et al. 2007). Despite challenges in using probabilistic climate information, it is necessary to continue exploring how information about annual climate variability might help to manage water resources, particularly as responses to seasonal variability are interlinked with those that deal with projected climate change impacts (see Fig. 1). Research that identifies key vulnerabilities to climate variability and appropriate responses to manage this variability will facilitate the design and implementation of climate change adaptation options. Water resources are significantly impacted by climate variability particularly through the impact on run-off (Schulze 2005). Under projected climate change, spatial variability of rainfall is likely to increase in many, but not all, areas (Boko et al. 2007). The IPCC 2007 report (Christensen et al. 2007) suggests that changes in climate variability are likely to occur on a sub-regional scale, but are not detectable by global scale models. Finding effective responses to seasonal variability is critical and can be viewed as a strategy to deal with long term climate change, as the implementation of medium term responses to annual variability can contribute to the development of a suite of strategies to respond to climate change and facilitate the structural climate-society linkage needed to implement longer term strategies (Mukheibir 2007; Washington et al. 2006). Response on both medium and longer timescales are needed to address societal needs by meeting basic infrastructure requirements and local development goals, whilst making decisions in a risk context. In the context of water resources this is important when scarce resources need to be allocated to make adaptive adjustments that respond to climate variability and change, economic growth and alleviating poverty. Most studies that outline the challenges of using forecasts in the water sector have focused in the developed world (e.g. Callahan et al. 1999; Carbone and Dow 2005; Patt et al. 2007; Yarnal et al. 2006) but very few have explored the challenges in the developing world where water quality and availability link directly to poverty and well being. This paper focuses on the potential for seasonal climate forecasts to be used in the water sector in South Africa, and the potential for improving access and use of climate information within the context of the growing importance of addressing climate change. In southern Africa it is important to understand the impact of annual variability on water resources where climate change is expected to see changes in means and extremes (Midgley et al. 2005; Hewitson 2007). The Department of Water Affairs and Forestry (DWAF), projects that water demand in South African will have

Climatic Change (2010) 103:537–554 Fig. 1 Diagrammatic view of the relationships between seasonal variability, climate change and development (after Mukheibir 2007)

539 Seasonal variability Periodic impacts

Frequent impacts

Short-term water shortages/ flood damage

Short-term strategies

Climate change

Frequent water shortages/ flood damage

Selection of robust strategies

Long-term resilience

Meeting basic requirements & development goals

outstripped natural supply by 2025 (Mukheibir 2007). However, the water management sector has not reached a stage of integrating information about changing intraannual climate conditions in a systematic manner, which is a critical step needed in responding to longer term change. This is partly due to a lack of awareness of available products and in some cases products are not tailored to suit user needs. Farmers in South Africa for example, have tended to respond to seasonal variability rather than pre-empt decadal change in rainfall and temperature (Thomas et al. 2007). Seasonal forecasts and climate change scenarios present an opportunity to help manage future climate risks relevant to water resources. If stakeholders learn to use information about climate variability on the annual time-scale, they could be better equipped to adapt to a change in climate variability as experienced from a climate in transition (Washington et al. 2006). The challenge is to explore whether value-added seasonal forecasts would be of use in the water sector that has certain requirements in terms of assuring supply. Seasonal climate forecasts have been available in South Africa for over 10 years, yet they are of little value to the water management sector if they are not accessible and understood. This paper seeks to identify processes and products that would facilitate increased uptake of seasonal climate forecasts among water resource managers. It is important to identify the type of climate (and weather) information that water resource managers require and could benefit from, in order to match these requirements with existing and developing climate products. This paper presents some of the key issues in the literature related to seasonal forecasts and water resource management and then discusses water-related issues of relevance to South Africa and in particular the Western Cape. Within the Western Cape lies the Cape Metropolitan area (CMA), which includes the City of Cape Town (CCT). This area is used as a case study to explore the current and potential uptake of seasonal climate forecast information in the water sector. The methodology for the study is outlined before a discussion of the results, key findings and conclusions. 1.1 Climate variability, seasonal forecasts and water resource management Climate information can generally be divided into three types; short term (weather forecasts), medium term (seasonal climate forecasts) and long term (climate variabil-

540


ity and climate change projections). Each has specific intentions and envisaged uses, as summarised in Table 1 below. Different types of information are used in different ways within each sector depending on the kinds of priorities and decisions that need to be made. Short term climate information can be used both for daily management of water resources as well as managing extreme events such as floods. It is most likely that seasonal forecasts will be used for tactical decisions over the medium term that consider issues about water allocation and demand management. Then on the longer-term timescale, planning issues related to demand need to consider decadal variability and climate change. However, tactical decisions that are based on the medium-term scale are seen as a good way to respond to longer-term change (Washington et al. 2006). The application of seasonal forecasts, that provide information on the mediumscale of 3 to 9 months, has received wide attention in the literature in a number of fields (O’Brien and Vogel 2003; Johnston 2008; Patt et al. 2007; Stern and Easterling 1999). Seasonal forecasts can be used for annual planning to prepare for expected above or below normal conditions and can help to respond to annual climate variability. Most of the applications of forecasts have been within agriculture and has been used to inform issues such as timing or variety of crops to plant, investment in land and inputs as well as irrigation management (Ingram et al. 2002; Murphy et al. 2001; Stone and Meinke 2006). In regions that lack sufficient water resources for irrigation and therefore depend on rainfall for farming, prior knowledge of the likely pattern of precipitation could lead to substantial improvements in food security and in profits for larger-scale producers (Blench 1999; Hammer et al. 2001). Some of the limitations discussed regarding forecast use were related to a function of their validity and spatial resolution, as well as the amount of useful information provided. Although much research has focused on crop production, there have also been explorations to improve the utility of forecasts for livestock production.

Table 1 Types of forecasts and the decision making options associated with each forecast (adapted from Schulze 2005) Type of decision

Climate Long term (10–50 years)

Type of information

Decadal changes; climate change scenarios Supplying demand Reservoir safety Reservoir sizing Land management

Strategic

Tactical

Operational

Medium term (3–9 months)

Weather Medium term (7–30 days)

Short term (0–7 days)

Seasonal forecasts

Week–monthly weather forecasts

Daily forecasts; observations

Operating rules Water orders Water allocation Demand management Irrigation scheduling Field operations Flood warning


541

Livestock farmers are particularly concerned about managing drought and forecasts can support decisions around reducing herd size or buying excess fodder for example (e.g. Hudson and Vogel 2003). Seasonal forecasts have therefore been shown to be useful as an adaptive strategy to respond to climate variability, especially in determining the planting and harvesting times and responding to variability in market demand (Ziervogel et al. 2005). Patt et al. (2007) explore how seasonal forecasts have been used within Africa. They suggest that there are limited examples (for reasons mentioned above) of how seasonal forecasts have been effectively used (Patt et al. 2007) but present a number of examples of where forecasts have been integrated into operations. In terms of water management, seasonal catchment forecasts were used to model seasonal commitments for power generation and agriculture for the Manantali Dam in West Africa and inform actual dam operations (Axel and Céron 2007). One of the major challenges in using forecast information in the water sector is shifting from the traditional reliance upon historical records (that infer the probability that shortages and floods might occur given their frequency of occurrence in the past) to using forecasts that project future information with uncertainty about their reliability. Pulwarty and Redmond (1997) have offered eight reasons why water resource managers do not access and use the available information. Amongst these are lack of certainty, poor communication, lack of focus on users’ needs, insufficient trust, lack of resources, poor timing of forecasts and the lack of training among users. A number of studies have explored how seasonal forecasts might be used in water management contexts (Georgakakos and Yao 2001; Landman et al. 2001; Yarnal et al. 2006). One study in south-east Australia looked at the use of seasonal streamflow forecasts using ENSO and serial correlation in reservoir inflow to optimise water restriction rules (Chiew et al. 2003). This was undertaken in an urban settlement where seasonal forecasts of reservoir inflow were used to help make management decisions in two irrigation systems, although the benefits derived from using the forecasts were minimal. In the United States of America, studies have shown that although it might be expected for seasonal climate forecasts to be employed in water resource management, the evidence thereof is limited (Callahan et al. 1999; O’Connor et al. 2005; Rayner et al. 2005). A study in southern Carolina (Carbone and Dow 2005) suggested that obstacles to using forecasts among community water system managers include a lack of awareness of their existence, distrust of their accuracy, perceived irrelevance to management decisions and competition from other technological innovations. They also explored methods on how to customise long lead forecasts into secondary products that address issues of greater relevance to water resource managers particularly under drought conditions. O’Connor et al. (2005) found that forecast skill is not the key issue, as water managers who find forecasts reliable are not more likely to use them than managers who do not find them reliable. They suggest that the most important determinant of forecast use is risk perception. This tends to be linked to past experience, so that water managers who have had problems with weather events in the previous 5 years, such as flood emergencies, would be more likely to feel at risk and would therefore use weather and climate forecasts. Part of the challenge in ensuring maximum seasonal forecast usage, is addressing effective communication of the information (Hartmann et al. 2002; HendersonSellers 1998). As Everingham et al. (2002) suggest, good communication requires

542


an understanding of each component of the chain of communication that includes an understanding of the constraints to understanding and using the information. In order to improve communication and uptake, scientists need to take responsibility, alongside users, in collaborating with institutional intermediaries to ensure that reliable, effective and useful forecasts are produced (Hammer et al. 2001). Although water resources are directly and indirectly impacted by climate variability, seasonal climate information needs to be communicated in a way that is understood, relevant and prioritised by water resource managers in order for the information to be used. This first requires research into what the requirements are for specific water resource manager groups. 1.2 Climate information availability and uptake in South Africa A range of institutions in South Africa produce different types of forecasts, including meteorological services and university research groups. The South African Weather Service (SAWS) is the country’s national meteorological service, with a permanent representative at the World Meteorological Organisation (WMO). In terms of its WMO obligation, SAWS is the primary weather and climate information service provider in South Africa. SAWS used to e-mail the forecast to users and place it on their website, though at present only notifications about new forecasts appearing on the website are sent. The Climate Systems Analysis Group (CSAG) at the University of Cape Town also produces seasonal forecasts that are available on their website and distributed, by request, as an annotated report to selected farmers and water resource managers. Both institutions post their forecasts on the Global Forecasting Centre for South Africa (GFSCA) website, but this site is not intended for non-scientific user access. Thus, a potential user of a seasonal forecast would need to be aware that the forecast is located on the organisations’ websites in order to access it. Over the last 10 years significant research has been undertaken to produce reliable and accessible seasonal forecasts for general use in sectors such as agriculture, transport and water resource management (Johnston et al. 2004). The uptake of seasonal forecast information in South Africa has, however, not been widespread (Archer 2003; Klopper and Bartman 2003; Ziervogel et al. 2005). The use of weather and climate forecasts has been limited mostly to the 1–3 day weather forecasts that are easily accessible i.e., in the news media. 1.2.1 Climate change and water resources South Africa is generally viewed as a water-stressed country with an average annual rainfall of 500 mm, which is approximately 60% of the world average. The greater part of the interior and western part of the country is arid or semi-arid (DWAF 1994). Since rainfall displays strong seasonality, the natural availability of water across the country is variable. Climate change scenarios suggest that changes in seasonality and intensity of rainfall will impact the runoff and groundwater recharge and the storage of water in the soil, dams and reservoirs (Schulze 2005). Increased temperatures will increase evaporation and result in an increased transpiration rate from plants. Combined with land transformation (particularly land converted to agriculture) and denuded land surfaces, availability and accessibility of water could be further reduced. The effects of any decrease in precipitation may be amplified through the hydrological system, where both run-off and groundwater recharge, especially in


543

semi-arid and arid regions, will decrease at a much higher rate than the underlying decrease in precipitation with the magnitude of run-off being reduced by two to four times more than the reduction in the precipitation, for a 50–60 year historical period (OneWorld Sustainable Investments 2007). In a study of the Northern Cape (Mukheibir 2007), local responses to drought and future climate change impacts by local municipalities were explored. Two key obstacles and limitations to ensuring pre-emptive responses to drought at a local municipal level were identified in this study. The first was the limitations of local capacity. A large number of the smaller municipalities lack the necessary skills capacity to adequately plan for periods of water shortages. The second obstacle is the low financial resource base to cover the capital and running costs for adequate clean water supplies (Wall 2006). Running costs are mostly covered from local revenues, which, for the smaller and remote local municipalities, are insufficient to ensure water security at this level. This is further compounded by the fact that local governments compete with each other for nationally allocated funds for capital expenditure across all sectors. Without adequate measures to increase revenues and reduce costs, water systems of small towns experience serious financial crises that may result in their closure or over utilisation of a local source (Dziegielewski and Bik 2004). Whilst these capacity deficiencies relate to responses to drought impacts, they are significant in that they also need to be overcome when developing systems to ensure the usage of forecast information, which will in turn ensure adequate responses to dry periods. 1.2.2 Water resource management in Western Cape, South Africa The Cape Metropolitan area (CMA) in the Western Cape province in the south west of South Africa, has an urban and peri-urban population of near three million. It depends chiefly on dams in the mountains of the south-western Cape for domestic and industrial water supply. The surrounding farmlands, the smaller towns of the region, and the extensive tourist and holiday regions both north and east of the city’s urban area also require water emanating from similar sources. The region experiences a Mediterranean-type climate with rainfall confined almost entirely to the cool winter months between April and October. Dry, hot summers, fairly frequent droughts (3 years in ten) together with significant population growth, (over 3% pa) and increased seasonal demand from tourism place a severe strain on the city’s water resources (City of Cape Town 2006). Water management in the Western Cape includes potable water capture, storage and supply, sewage and stormwater. These are described further in Table 2. 2 Methodology Research was undertaken between 2005 and 2008 for a project funded by the Water Research Commission (WRC) that aimed to improve the understanding of water resource managers’ use of seasonal forecasts and the potential for increasing uptake of forecasts. First an inventory of available climate information in South Africa was compiled, including short term weather and medium term climate information. Case studies of water resource managers’ understanding and use of climate information were then undertaken before developing an online toolkit of climate information for water resource managers.

544


Table 2 Characteristics of water management in the Western Cape Water management activities

Characteristics

Capture and storage

Supply of water from the dams around CMA is controlled by a carefully regulated system that depends on imminent rainfall, demand characteristics and individual dam levels Alterations and enlargements to dams are restricted; some municipalities are investigating sourcing water from the ground Problems of iron oxide content and aquifer depletion are some of the difficulties they are facing Each municipality is responsible for delivering the water to the residents and for recovering the costs The population of the urban areas, and thus the demand for water, is increasing, placing a higher stress on the supply and infrastructure The Free Basic Water Regulation ensures that poor households have a minimum amount of 25 l per person per day of water available free of charge Reticulation of water within the urban area requires robust, well-built infrastructure which requires upgrading and constant maintenance Maintenance and upkeep is susceptible to weather conditions, such as dry spells, heavy rain, flooding and high wind Storm water drains are designed to cope with peak rainfalls however, during times of heavy rainfall events, drains often overflow and cause flooding in low-lying areas. The main cause is often the lack of maintenance and clearing of debris Much of the flooding occurs on the low-lying Cape Flats area where high numbers of people live in informal housing and are significantly impacted by flooding that damages homes and possessions every year Sewage reticulation systems are overextended and outdated, and treatment works are under pressure due to increased loading Many smaller municipalities have conservancy tanks which need to be emptied on a regular basis The use of septic tanks has led to infiltration into groundwater and these have been discontinued in most of the susceptible areas When flooding occurs, the infiltration of storm water entering the sewage system puts an undue hydraulic load on the purification works Sewage system flooding can cause overflowing resulting in sewage entering storm water systems and the introduction of untreated sewage into river systems and coastal waters Many towns require costly but necessary sewage disposal and treatment upgrading. Budgetary and capacity shortages have led to the problem leading to crisis proportions during heavy rains

Supply

Disposal—storm water

Disposal—sewage

This paper focuses on the case study of the Cape Metropolitan area in the Western Cape region. This area was chosen because it is currently water stressed and climate change scenarios suggest that traditional winter rainfall patterns may be changing, with decreases in annual precipitation strongest in early and mid-winter linked to a shorter winter rainfall period (DEADP 2007). An added consideration is the population increase of around 3.2% per annum through urbanisation and natural population growth (Jaglin 2004) that is placing increasing pressure on water resources and the growing tourism industry.


545

The case study focused on the information that water resource managers and other related stakeholders presently use to make planning decisions, including the type of data and information currently received and used, and the interest and potential for using climate data or information, if not currently used. Firstly, indepth semi-structured interviews were undertaken with 18 key individuals, which produced detailed profiles of the users of information and their needs. Interviewees included representatives from DWAF (Department of Water Affairs), consultants involved in water resource issues, municipal water managers and a range of other stakeholders. The information from the interviews was used to create a map of linkages between stakeholders and different information sources. Questions included the type of climate information accessed, the source of access and format of the data. Interviewees were asked how easy it was to access the information and to rate the skill of the forecast. The next theme was how the information was used or applied to their work. Interviewees were also asked about any perceived shortcomings of the forecasts and whether there were other climate products which they desired. A follow-up workshop (with 22 participants) built on the information from the interviews and sought wider representation in order to verify interview findings. Focus groups were used to follow up on issues emerging from the workshop. Focus groups were held for different sectors (including water storage; urban water supply; water disposal and stormwater; nature conservation, biodiversity, agriculture and disaster management) to explore how specific water resource strategies might be better able to utilise seasonal forecasts. The workshops and focus groups sought to address four aims: 1. to determine the understanding managers have of climate variability and climate change 2. to identify the strengths and weaknesses in current seasonal forecast tools 3. to examine the type of climate variability information managers think would be relevant at an annual and longer-term scale 4. to enable water resource managers to consider how forecast and climate change information could be better used for decision making Using the information gained from the inventory, workshops and focus groups, a climate information toolkit was produced for water resource managers to use and to train other stakeholders about climate variability information and its potential applications and limitations. The toolkit included a website, a brochure and briefing notes. Participants in the interviews, workshop and focus groups and other interested parties were invited to a presentation to showcase the toolkit website. Brochures and briefing booklet were handed out. Feedback was requested through a feedback form on the website was used to make changes to the site some months later.

3 Results 3.1 Current availability and use of climate information in the City of Cape Town From the interviews it emerged that the City of Cape Town (CCT) bulk water managers use shorter term (daily to fortnightly) forecasts to manage the water in dams. Forecasts inform decisions on which dam(s) to draw on at any particular time

546


to supply Cape Town as well as to manage the transfers of water from one dam to another. Forecasts are also used to inform water managers’ planning decisions, where they discuss climate information with others (especially DWAF), and report concerns in water liaison meetings and to politicians. Seasonal forecasts were seen as less skilful, and thus less useful, than short-term forecasts. However, they believe that seasonal forecasts could be used in agriculture to decide which cash crops to plant, in nature conservation to draw up annual plans of operations e.g. trail maintenance, vegetation monitoring, control burns and in municipal water management deciding whether to implement water restrictions if a particularly dry summer is predicted. They suggested that long-term predictions such as climate change scenarios could inform their decisions about building new infrastructure or resource diversification. Seasonal forecasts are used mainly in drought management. For example, based on seasonal forecasts DWAF make tactical decisions on the level of water demand restrictions to impose and the volume of water transfers between catchments. Decisions are also made on water use planning and disaster management preparation, e.g. when floods are expected. They monitor seasonal forecasts, and, if a dry season is predicted, a meeting of the planning committee is called to discuss water restrictions and possible emergency measures. The committee includes water managers in the city and surrounding municipalities, DWAF, consultants and the irrigation board. Farmers and other groups are advised accordingly. In general it could be said that water managers in CCT used shorter-term weather forecasts more than medium-term forecasts. DWAF and some CCT water resource managers interviewed use a forecast and water supply assessment bulletin emailed approximately weekly by consultants to DWAF. The bulletin includes information on cumulative rainfall figures and dam levels, dam storage forecasts under different demand scenarios, suggestions for managing water in dams supplying Cape Town, as well as 7-day, 1 month and seasonal weather forecasts. The workshops explored the potential for using seasonal forecasts in the future, including improvement stakeholders would like to see. In general, participants felt that seasonal forecasts could inform planning and management of water resources, whether to apply water restrictions, emergency preparedness, early warning crop estimates, fire management, vegetation monitoring, planning outdoor work, dam and river bank inspections, environmental impact evaluations and whether to make an environmental flood release. However, there were a number of improvements to seasonal forecasts that managers would like to see, including: • • •

Improved forecast skill Information on the websites about the principal drivers behind seasonal forecasts in order to improve understanding and confidence in the forecasts Reference to normal (or mean) values, an indication of expectations, and skill levels shown

The type of climate variability information managers think would be relevant to their operations and planning (in addition to current available information) included: • •

Historical rainfall intensity data—used for flood prediction and design of stormwater infrastructure Rainfall intensity data—for water transfer planning, planning irrigation schedules, disaster planning and flood prediction.


• • • •

547

Change in temperature (weeks to month)—for algal bloom management, fire management, planning irrigation schedules Wind conditions—for fire management, planning irrigation schedules Radar—for intensity prediction, storm management, dam release, rescue Seasonal forecast—for agricultural crop management, deciding whether to plant cash crops, stock water provision, management of artificial groundwater recharge

Improvements in availability and access to this data could improve management decisions such as better assessment of the return period of a particular storm so that it could be used for risk mapping; rainfall intensity at a smaller time scale that could be used for water transfer planning, disaster planning and flood prediction; temperature change at a fine spatial resolution—for algal bloom management, fire management; chill unit predictions—for produce such as apples and grapes; better accuracy for seasonal forecasts—for dam construction and level management, corresponding runoff so decisions for water use can be made and integration of all the data in one place. 3.2 Sector-specific requirements The focus groups (see Table 3 below) provided insight on the specific information different sectors would find useful. This process helped to unpack the details of operational, tactical and strategic decisions made by the different sectors based on weather and climate information. 3.2.1 The water storage focus group The key issues identified by this group included the need for an annually issued forecast for the following 12 months. Decisions on water restrictions are made only once a year therefore a shorter forecast is not suitable. Seasonal forecasts could be used for predicting runoff to the dams and decisions about whether to have conjunctive use of surface and aquifer water.

Table 3 Table of the focus group attendees Attendees

City of Cape Town Engineering Consultant Wildlife and Environment Society Department of Agriculture University Freshwater Research Group Berg River Irrigation Board Cape Nature Conservation

Focus groups Water Urban water storage supply, disposal and stormwater X X

Nature, biodiversity and agriculture

X X

Disaster management X

X X X X X

548


Rainfall and temperature greatly influence water demand in the urban centres during the summer months, and this influences long-term planning. An early warning system of progressive or successive dry seasons would be useful. Climate change scenarios would be useful for planning what impact changes in temperature and precipitation would have on water resources. Climate change is expected to bring about a temperature increase in the long term, thus this group concluded that efforts should be made to minimise future and existing dam surface areas to decrease evaporation losses. If a change in rainfall patterns is to be expected then constructing new pipelines that could carry larger volumes during high rainfall intensities may need to be considered. However these would be an expensive investment, and would require strong confidence in the prediction, which currently planners do not have. 3.2.2 Urban water supply, water disposal and stormwater focus group A key concern for this focus group was the impact on the sewerage system. A significant increase in rainfall intensity leads to an increase in sewer inflow, which increases the volume of influent through the treatment plant and may disrupt treatment functioning. Infiltration may also lead to the overflow of sewage into the natural systems. Forecasts of seasonal rainfall intensity could help in developing responses. Treatment plants are sensitive to effluent concentrations and variations in dilution levels can affect their functioning. This makes them sensitive to rainfall changes. Climate variations, rather than population changes, have the greatest influence on the seasonal pattern of disposed water volume. A change in both the seasonal or long term climate is therefore important to consider for planning. Increases in temperature may enhance the functioning of the treatment as it may increase biological decomposing activities but this may also create stronger odours. Short-term weather information can be used to prepare for extreme events. If a flood warning is received from the weather service, little can be done in terms of mitigation, but disaster management preparation can be mobilised (e.g. staff placed on stand-by to unblock drains and to aid people in flooded informal settlements). Medium term information, such as seasonal forecasts, can be useful in terms of planning for a winter preparedness programme that would allocate human resource to sites that are susceptible to flooding. Drier conditions would have little effect on their design philosophy. If seasonal forecasts included projections of rainfall intensity they would be beneficial in helping managers prepare for dam overflows, especially during the end of winter. 3.2.3 Nature, biodiversity and agriculture focus group This group stated that users traditionally lacked trust in weather/climate information (due to a perceived poor track record) and therefore tended not to make major decisions based on such information. Users also often lack the necessary knowledge to use the information appropriately. In terms of water management, the group felt that seasonal forecasts would be most useful in determining the appropriate amount of water to release from dams for environmental flows in keeping with in-flow stream requirements. These forecasts are essential in determining rules for dam operation (e.g. amount of water to release).


549

Climate is an important variable in estuarine activities. Estuaries are often closed in dry conditions, requiring mechanical methods to re-open. Climate information, particularly medium term weather forecasts could be used in planning the re-opening of estuary mouths mechanically to link to run-off events. Changing seasonal conditions in the future would impact on the management requirements and ecological sensitivity of the estuaries. 3.2.4 Disaster management focus group These representatives agreed that the City of Cape Town (CCT) was interested in seasonal weather forecasts because informal settlements are often at risk during the rainy season. In winter they are particularly interested in high rainfall intensity because of the likelihood of flooding. The CCT Disaster Risk Management department is involved in an annual flood risk plan (winter preparation) that uses seasonal forecasts to infer the number of shacks that are susceptible to potential flooding. With this information they can plan preventative measures and inform the finance departments. This type of annual planning seems to have achieved better results in mitigating floods in recent years. Short term forecasts or predictions are potentially useful in a variety of ways to help prepare the disaster management department for a number of weather events. These include consecutive days of heavy rain that can causing flooding; wind information can be used for managing spring time onshore wind that can raise the tidal levels as well as plans for responding to Koeberg nuclear power station’s theoretical radioactive plumes. Fire management can also benefit from weather information on wind, humidity and temperature forecasts to help determine the spread of fire and its potential to continue burning. Climate change projections are very important in terms of the city’s planning for possible flooding scenarios in the future. If reliable climate change evidence is documented, then managers could start advising politicians regarding the impacts of changes in development, especially development along the coastline as well as influencing the design of new storm water infrastructures.

4 Key findings Climate information is of interest and relevance to a range of water resource management activities. Drawing on the empirical evidence, a number of key findings emerge that address the question of whether water resource managers are aware of seasonal forecasts, if they are using them and if there is potential for them to be disseminated and used more effectively. Finding 1: Current knowledge of seasonal forecasts is limited. Water resource managers in general have little understanding of seasonal forecasts. Although some awareness is evident, explanation and interpretation was needed to ensure the forecast figures and terminology were correctly understood. None of the stakeholders were aware of the amount of forecast information available. However, they were all eager to hear more information about the seasonal forecasts and how to interpret them.

550


Finding 2: There is little evidence of use of seasonal forecasts in the water sector. From the interviews and workshops it became apparent that seasonal forecasts were of interest to a wide range of stakeholders in the water management sector. Yet, although the forecasts are factored in to some decision-making frameworks there is little evidence of the application of seasonal forecasts in actual decisions. The available forecasts seem to be accessed by some respondents but there is no evidence of systematic use in decision-making. Water restriction planning was one example where seasonal forecasts were used in the City of Cape Town. The forecast was integrated into supply planning by forming part of a rolling probability study, which informed the City and DWAF officials about potential dam levels and usage demands. If the forecast suggested below normal rainfall when dam levels were already low, the City was more likely to consider restrictions. The most active current use of forecasts seems to focus on short to medium term weather information, such as dam withdrawal allocation decisions and irrigation scheduling. Finding 3: Utilisation of seasonal forecasts is hindered by the lack of knowledge about forecasts and limited interaction with climate scientists. The most common means of access to the seasonal forecasts has been the internet. Most users were not aware of the amount of information available. The nature of accessing the information via the internet limited interaction with the producers of the climate information and users were not able to build up their understanding of forecast products. Finding 4: Tailored products may enable better integration of seasonal forecast information in water resource planning. Water resource managers requested indicators relating to the accuracy of seasonal forecasts. They also suggested that more active dissemination, interpretation and liaison between forecasters and water managers would improve the use of the information. It was clear that there were definite opportunities that were not being utilised at the moment. Methods need to be developed that would enable historical data and forecast data to be used in planning. Such tailored products could enable water resource managers to integrate seasonal forecast information in their planning.

5 Conclusions and recommendations It is apparent that current forecast information is not meeting the expressed needs of water resource managers sufficiently. This is partly an issue of availability and partly one of accessibility, with users also expressing difficulty in interpreting or understanding seasonal forecasts. A different format for seasonal forecasts needs to be explored to make them more accessible to users. Water managers expressed the need for other information to assist with their decision making, such as rainfall intensity predictions (frequently mentioned), more specific locations for forecasts, and longer forecast lead times (9–12 months) for drought and restrictions planning.


551

Forecasts of different types were used for many operational, tactical and strategic decisions. Shorter term climate and weather information was used by many water managers to plan and manage water storage and supply. Some also used this information to schedule short term operational activities such as road maintenance. Short term forecasts have been used to help with managing dam levels, decisions on which dam to draw water from, and whether to move water between dams. Seasonal forecasts are used mostly when considering the adoption of water restrictions, the level of restrictions and the implementation timing as well as for planning response to flooding in the informal settlements. Access and use of seasonal forecasts was haphazard and random in many cases. Some users knew exactly what they wanted, were used to and trusted a certain source and subsequently used it for their purposes without being aware of alternatives. While the presentations made by team members during the project were appreciated, they are not a sustainable way of communicating with a growing group of potential users. For this reason a web-based toolkit was developed as an initial means for providing information on forecast availability and to explore potential as it serving as a long term outreach to water managers (www.c4w.org.za). Although the toolkit was a useful source that linked to a range of forecasts, it was not adopted as a regular source of information and so had limited success. Results of the case study showed that the activities of water resource managers would be enhanced by improved access to forecasts. This could be met by improving awareness of existing weather and climate information. Although current forecast information is used to a limited degree, once stakeholders became aware of the available information they are more inclined to use it (O’Connor et al. 2005). Additional information requested related to runoff, fire danger, river temperatures, intensity of rainfall within short time frames and extreme conditions. Despite the potential usefulness of forecast information, it was clear from the stakeholder interviews and larger group meetings that managers do not always use weather and climate forecasts very effectively or efficiently. The barriers to managers’ use of climate forecasts were similar to other studies (Callahan et al. 1999; Rayner et al. 2005). To realise the potential of climate forecasts for water resource managers, technical improvements to the forecast products and specific verification and skill evaluations are required as well as active communication of the information. Research has shown that users in other sectors have been keen to welcome seasonal forecasts into their decision making, but that it requires significant training, explanation and supplementary analysis (Johnston 2008; Patt et al. 2007; Rayner et al. 2005). Forecasts in their current form do not focus on specific users’ needs, but rather cater for general use. Although much useful information is available for immediate and practical use, if water resource managers are to gain the most possible benefit from forecasts they will need expert assistance in interpretation for their specific uses. It has been shown in the case of the City of Cape Town that this formula works, but requires resources such as the services of a private consulting company. Whether this is within the capabilities of other metropolitan and urban areas is debatable, but on the other hand, can policy and decision makers manage without it? In the case of smaller urban areas, the costs of such a service will have to be weighed up against the benefits that may accrue. Joint efforts between forecast producers and water resource managers are clearly necessary to develop and demonstrate climate forecast applications through two-way education and interaction (Rayner et al. 2005). This was shown to be highly effective

552


in Western Australia in 2003 (Power et al. 2005) and in the City of Cape Town case study. Effective communication of forecast information depends not only on the user education and awareness creation, but equally as much on the forecast producers and research funders who need to be aware of the changing and developing needs of users in the latter’s quest to maximise their advantage by minimising their weather-related risks. Responses to seasonal variability are directly linked with those that deal with projected climate change impacts. It is therefore important to strengthen responses that relate to seasonal change as these can be tested on an annual timescale and help deal with variability as opposed to long term decadal change. In addition, it is important that responses meet development needs at the same time as addressing climate adaptation needs (Ziervogel and Taylor 2008). The future success of forecast uptake and usefulness depends on how the lessons learnt in this and other research are converted into more skilful, better disseminated and more aptly targeted forecasts. It is necessary that more opportunities are provided for those with climate knowledge to interact with water resource managers. This will contribute to a more holistic strategy to adapt to climate variability and change in the water sector.

References Archer ERM (2003) Identifying underserved end-user groups in the provision of climate. Inf Bull Am Meteorol Soc 84(11):1525–1532 Axel J, Céron P (2007) Griffiths J (ed) Elements for life. Tudor Rose, London, pp 70–71 Blench R (1999) Seasonal climatic forecasting: who can use it and how should it be disseminated? Natural Resource Perspectives 47. Overseas Development Institute, London Boko M, Niang I, Nyong A, Vogel C, Githeko A, Medany M, Osman-Elasha B, Tabo R, Yanda P (2007) Africa climate change 2007: impacts, adaptation and vulnerability. In: Parry ML, Canziani OF, Palutikof JP, van der Linden PJ, Hanson CE (eds) Contribution of working group II to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, pp 433–467 Callahan BM, Miles EL, Fluharty DL (1999) Policy implications of climate forecasts for water resources management in the Pacific Northwest. Policy Sci 32(3):269–293 Carbone GJ, Dow K (2005) Water resource management and drought forecasts in South Carolina. J Am Water Resour Assoc 41(1):145–155 Chiew FHS, Zhou SL, McMahon TA (2003) Use of seasonal streamflow forecasts in water resources management. J Hydrol 270(1–2):135–144 Christensen JH, Hewitson B, Busuioc A, Chen A, Gao X, Held I, Jones R, Kolli RK, Kwon W-T, Laprise R, Magaña Rueda V, Mearns L, Menéndez CG, Räisänen J, Rinke A, Sarr A, Whetton P (2007) Regional climate projections. In: Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (eds) Climate change. The physical science basis. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge City of Cape Town (2006) Draft water services development plan for City of Cape Town 2006/07. Department of Environmental Affairs and Development Planning, Western Cape (DEA&DP) (2007) A climate change response strategy and action plan for the Western Cape. Department of Environmental Affairs and Development Planning, Western Cape (DEA&DP), Cape Town DWAF (Department of Water Affairs and Forestry) (1994) Water supply and sanitation policy— White Paper Water—an indivisible national asset. DWAF, Cape Town Dziegielewski B, Bik T (2004) Technical assistance needs and research priorities for small community water systems. J Contemp Water Res Edu 128:13–20 Everingham YL, Muchow RC, Stone RC, Inman-Bamber NG, Singels A, Bezuidenhout CN (2002) Enhanced risk management and decision-making capability across the sugarcane industry value chain based on seasonal climate forecasts. Agric Syst 74:459–477


553

Georgakakos AP, Yao H (2001) Value of climate forecasts for the management of Lake Lanier. Georgia Water Resources Institute, Atlanta Hammer GL, Hansen JW, Phillips JG, Mjelde JW, Hill H, Love A, Potgieter A (2001) Advances in application of climate prediction in agriculture. Agric Syst 70:515–553 Hartmann HC, Bales R, Sorooshian S (2002) Weather, climate and hydrologic forecasting for the US Southwest: a survey. Clim Res 21(3):239–258 Hellmuth M, Moorhead A, Thomson MC, Williams J (eds) (2007) Climate risk management in Africa: learning from practice. IRI, Columbia University, New York Henderson-Sellers A (1998) Climate whispers: media communication about climate change. Clim Change 40:421–456 Hewitson B (2007) Regional climate change scenarios for the Western Cape Province. Presentation to the Western Cape Government Climate Change Conference Hudson J, Vogel C (2003) The use of seasonal forecasts by livestock farmers in South Africa. In: O’Brien K, Vogel C (eds) Coping with climate variability: the use of seasonal forecasts in Southern Africa. Ashgate, Aldershot, pp 75–96 Ingram KT, Roncoli MC, Kirshen PH (2002) Opportunities and constraints for farmers of west Africa to use seasonal forecasts with Burkina Faso as a case study. Agric Syst 74:331–349 Jaglin S (2004) Water delivery and metropolitan institution building in Cape Town: the problems of urban integration. Urban Forum 15(3):231–253 Johnston PA (2008) The uptake and utility of seasonal forecasting products for commercial maize farmers in South Africa. Ph.D. thesis, University of Cape Town Johnston PA, Archer E, Vogel C, Bezuidenhout C, Tennant W, Kuschke R (2004) Review of seasonal forecasting in South Africa: producer to end-user. Clim Res 28(1):67–82 Johnston PA, Ziervogel G, Matthew M (2007) The uptake and usefulness of weather and climate information among water resource managers. Papers Appl Geog Conf 30:380–389 Klopper E, Bartman A (2003) Forecasts and commercial agriculture: a survey of user needs in South Africa. In: O’Brien K, Vogel C (eds) Coping with climate variability: the use of seasonal climate forecasts in Southern Africa. Ashgate, Hampshire, pp 170–182 Landman WA, Mason SJ, Tyson PD, Tennant WJ (2001) Statistical downscaling of GCM simulations to streamflow. J Hydrol 252:221–236 Midgley G, Chapman A, Hewitson B, Johnston P, De Wit M, Ziervogel G, Mukheibir P, Van Niekerk L, Tadross M, Van Wilgen B, Kgope B, Morant P, Theron A, Scholes R, Forsyth G (2005) Western Cape report: a status quo, vulnerability and adaptation assessment of the physical and socio-economic effects of climate. http://wwwercuctacza/publications/wcape_climate_ change_impacts_11pdf Mukheibir P (2007) Qualitative assessment of municipal water resource management strategies under climate impacts: the case of the Northern Cape, South Africa. Water SA 33(4), 575–581 Murphy SJ, Washington R, Downing TE, Martin RV, Ziervogel G, Preston A, Todd M, Butterfield R, Briden J (2001) Seasonal forecasting for climate hazards: prospects and responses. Nat Hazards 23:171–196 O’Brien KL, Vogel HC (eds) (2003) Coping with climate variability: the use of seasonal climate forecasts in Southern Africa. Ashgate, Aldershot O’Connor RE, Yarnal B, Dow C, Jocoy CL, Carbone GJ (2005) Feeling at risk matters: water managers and the decision to use forecasts. Risk Anal 25(5):1265–1275 OneWorld Sustainable Investments (2007) Western Cape climate change response strategy and action plan. Provincial Government of the Western Cape Department of Environmental Affairs, Cape Town Patt AG, Ogallo L, Hellmuth M (2007) Learning from ten years of climate outlook forums in Africa. Science 318:49–50 Power S, Sadler B, Nicholls N (2005) The influence of climate science on water management in Western Australia. Bull Am Meteorol Soc 86(6):839–844 Pulwarty RS, Redmond K (1997) Climate and salmon restoration in the Columbia River Basin: the role and usability of seasonal forecasts. Bull Am Meteorol Soc 78(3):381–397 Rayner S, Lach D, Ingram H (2005) Weather forecasts are for wimps: why water resource managers do not use climate forecasts. Clim Change 69(2–3):197–227 Ritchie JW, Zammit C, Beal D (2004) Can seasonal climate forecasting assist in catchment water management decision-making? A case study of the border rivers catchment in Australia. Agric Ecosyst Environ 104(3):553–565 Schulze RE (2005) Setting the scene: the current hydroclimatic “Landscape” in Southern Africa. In: Schulze RE (ed) Climate change and water resources in Southern Africa: studies on scenarios,

554


impacts, vulnerabilities and adaptation. Pretoria, Water Research Commission WRC Report 1430/1/05: 6: 83–94 Stern PC, Easterling WE (eds) (1999) Making climate forecasts matter. National Academy, Washington, DC Stone RC, Meinke H (2006) Weather, climate, and farmers: an overview. Meteorol Appl 13:7–20 Thomas DSG, Twyman C, Osbahr H, Hewitson B (2007) Adaptation to climate change and variability: farmer responses to intra-seasonal precipitation trends in South Africa. Clim Change 83:301–322 Wall K (2006) An investigation of the franchising option for water services operation in South Africa. Water SA 32(2):265–268 Washington R, Harrison M, Conway D, Black E, Challinor A, Grimes D, Jones R, Morse A, Kay G, Todd M (2006) African climate change: taking the shorter route. Bull Am Meteorol Soc 87:1355– 1366 Yarnal B, Heasley AL, O’Connor RE, Dow K, Jocoy CL (2006) The potential use of climate forecasts by community water system managers. Land Use and Water Resources Research 6. http://wwwluwrrcom/contentshtml Ziervogel G, Taylor A (2008) Feeling stressed: integrating climate adaptation with other priorities in South Africa. Environment 2(50):32–41 Ziervogel G, Bharwani S, Downing T (2005) Adapting to climate variability: pumpkins, people and policy. Nat Resour Forum 30(4):294–305