Adaptation for Regional Water Management

Jeroen Aerts1 and Peter Droogers2

institute for Environmental Studies, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands; 2FutureWater, Arnhem, The Netherlands


Dealing with climate change and climate variability is generally considered to be one of the largest challenges for the coming decades, on all geographical scales, across all economic sectors. Water managers see themselves confronted with a continuous stream of increasingly credible scientific information on the potential magnitude of climate change and climate variability and the vulnerability of water resources to its impacts. The urgency to take action is more apparent than ever, yet clear guidance on exactly how to respond to the challenge of climate change is lacking, especially at the river basin level. A consensus, however, is emerging around the viewpoint that it is not only necessary to mitigate climate change by reducing greenhouse gas emissions and enhancing carbon sinks, but also to adapt to the inevitability of climate change by preparing for impacts and reducing vulnerability (Pielke, 1998; Kane and Shogren, 2000).

The Intergovernmental Panel on Climate Change (IPCC) states that changes in precipitation are likely to have a major impact on the hydrological cycle and, subsequently, on the environment and on food production (Arnell et al., 2001). Currently 800 million people suffer from hunger, among them 200 million children under 5 years of age. It is estimated that by 2025 cereal production will have to increase by 38% to meet world food demands. Climate change (CC) is expected to amplify climate variability (CV) and hence the occurrence of extreme events such as floods and droughts. For example, it is expected that cereal production in sub-Saharan Africa will decrease, with 2-3% of this decrease being due to increased climate variability (FAO, 2003). Allocating more water to agriculture, however, is not always the answer to these problems, since water is used for other sectors, such as nature and hydropower.

The ADAPT project

This book is the result of the ADAPT project, which focused on developing regional adaptation strategies for water, food and the environment in river basins across the © CAB International 2004. Climate Change in Contrasting River Basins

Fig. 1.1. The seven basins included in the ADAPT project.

world. The projected impacts of CC on water resources underline the necessity of water managers to seek new and sustainable water allocation measures that address the potential impact of CC. Response or adaptation in water management to ensure food security under changing climatic conditions is a difficult process, where complex trade-offs have to be made across different policy objectives and hence requires input from all stakeholders involved in this process. Despite its complexity, adaptation increasingly receives attention in policy making as a complementary coping mechanism to mitigation. Adaptation is explicitly addressed in several policy documents, such as in article 10 of the Kyoto Protocol (UNFCCC, 1997), where 'parties are further committed to promote and facilitate adaptation and deploy adaptation technologies to address climate change'. Furthermore, the process of adaptation is not new. Throughout history people have adapted to changing or extreme climate conditions. Especially in the water sector, people can learn from adaptation experiences in the past (Tol et al., 1998).

As impacts primarily take place at the regional (river basin) or local scales, adaptations similarly should be sought at those scales (IPCC, 2001b). The basins for the ADAPT project were selected within dry and wet areas and located in developed and developing countries. The rationale for formulating this project was that exchanging knowledge on adaptation drawn from very different regions would accelerate scientific innovation in this relatively unexplored field. The seven basins selected are (see Fig. 1.1):

The main goal of the ADAPT project was to develop a generic methodology for river basins (called Adaptation Methodology for River Basins, AMR) that allows the development and assessment of adaptation strategies for alleviating food and environmental impacts (Fig. 1.2). The methodology puts stakeholders in a central role in the adaptation process and iteratively addresses the following steps.

1. Story lines Socio-economic and climate change scenarios

2. Water Resource System State Indicators (SI)

2. Water Resource System State Indicators (SI)

5. Effects and evaluation

3. Objectives, impacts and Decision Indicators (DI)

4. Adaptation strategies

Fig. 1.2. An overview of the adaptation framework for river basins (AMR) followed by all case study areas in the ADAPT project (Aerts et al2003).

5. Effects and evaluation

3. Objectives, impacts and Decision Indicators (DI)

4. Adaptation strategies

Fig. 1.2. An overview of the adaptation framework for river basins (AMR) followed by all case study areas in the ADAPT project (Aerts et al2003).

Derive storylines for each basin, such as socio-economic developments, and climate change and variability projections per basin.

Select a set of models at basin and field scale capable of simulating hydrology and food production.

Assess CC/CV impacts on regional water management by comparing future simulations with baseline references with respect to the environment and food security.

Define adaptation strategies for water managers to respond to climate change. Evaluate adaptation strategies.

Purpose of this book

This book can be seen as a handbook for regional water management to develop and evaluate adaptation strategies to climate change and climate variability. For this, the generic AMR methodology is first described in this chapter, and next explained for seven case studies. These basin case studies (Chapters 5-11) followed this common methodology and each of the framework steps is addressed in the different basin chapters of this book. Apart from the basin chapters, several supporting chapters provide in-depth supporting information. Chapter 2 discusses the use of climate change scenarios as provided by the IPCC and, more specifically, how these scenarios can be used for regional studies. Chapters 3 and 4 describe in more detail the possible consequences of CC and CV for food security and environmental quality, respectively. Chapter 12 integrates the findings of the basin studies and compares these findings with global trends in climate change related to food security. Finally, Chapter 13 provides a summary of the experiences encountered during this project and provides the reader with some key findings that should be addressed in new regional adaptation studies.

Climate Change and Water Resources

Strong scientific evidence indicates that the average temperature of the earth's surface is increasing due to greenhouse gas emissions. The average global temperature has increased by about 0.6° since the late 19th century (IPCC, 2001a). The latest IPCC scenarios project temperature rises of 1.4—5.8°C, and sea level rises of 9-99 cm by 2100 (Figs 1.3 and 1.4; IPCC, 2001a). Warming and precipitation are expected to vary considerably from region to region as the greatest increase in temperature is expected in the Northern Hemisphere. Changes in climate averages, coupled with changes in the frequency and intensity of extreme weather events, are likely to have major impacts on natural and human systems. Effects on the world's poor in tropical and sub-tropical areas are likely to be disproportionately large, in particular since their potential to adapt to such changes is low (Smit et al., 2001).

Changes in the cycling of water between land, ocean and atmosphere can have significant impacts across many sectors of the economy, society and environment. Consequently, there are many studies that focus on the effects of CC on the hydro-logical cycle and the related availability of water resources for human and environmental use. The majority of these studies have focused on changes in the water balance. Other, but fewer, studies focused on the impacts of CC on water resources in terms of the reliability of the water supply, the risk of flooding or on exploring possible adaptation strategies (Arnell et al., 2001; Kabat and van Schaik, 2003).

Hydrological impacts

Climate change is only one of the pressures facing the hydrological system and water resources. Other global changes, such as population growth, pollution, land use changes and land management, also have a profound impact on the hydrological cycle. In general, there is an increasing move towards sustainable water management and increasing concern for the impacts of global change on the water resources system. Recent initiatives to address these issues include, for example, the 'Dublin Statement' in 1992, which urges the sustainable use of water; and activities by the World Water Council, which led to a vision for a 'water secure world' (Cosgrove and Rijsberman, 2000) and a report that addresses the need for water managers to better prepare for change in climate (Kabat and van Schaik, 2003).

The IPCC has assessed the major recent studies into the effects of CC on hydrology in its Third Assessment Report (TAR) (Arnell et al., 2001). IPCC found that most hydrological studies on the effects of CC have concentrated on streamflow and runoff (streamflow is water within a river channel, whereas runoff is the amount of precipitation that does not evaporate). Changing patterns in runoff are consistent with those identified for precipitation. However, in large parts of eastern and northwestern Europe, Canada and California, a major shift in streamflow from spring to winter has been associated with a change in precipitation. It appears that in these areas

A1B A1T A1Fl A2 B1 B2

IS92e high 1

IS92c low J

Several models all SRES envelope

Model ensemble all SRES envelope

(TAR method)

Model ensemble all SRES envelope

(TAR method)

Bars show the range in 2100 produced by several models








Fig. 1.3. Global temperature projections, according to different scenarios. (From IPCC, 2001a.)







Bars show the range in 2100 produced by several models


Fig. 1.3. Global temperature projections, according to different scenarios. (From IPCC, 2001a.)


1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100


Fig. 1.4. Global sea-level rise projections according to different scenarios. (From IPCC, 2001a.)

1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100


Fig. 1.4. Global sea-level rise projections according to different scenarios. (From IPCC, 2001a.)

Fig. 1.5. A map of long-term average annual water resources (MAR) by basin, calculated by the WaterGAP2 model. (From Smakhtin et al., 2004.)

precipitation will fall as rain, rather than snow, in winter periods (e.g. Lettenmaier et al., 1999; Middelkoop et al., 2001). In colder regions, no significant changes have been observed. Figure 1.5 shows a global map of long-term average annual water resources per major river basin (Smakhtin et al., 2004).

Increased temperatures generally result in an increase in potential evapotranspiration. In dry regions, potential evapotranspiration is driven by energy and is not constrained by atmospheric moisture contents. In humid regions though, atmospheric moisture content is a major limitation to evapotranspiration. Studies show increases in evapotranspiration with increased temperatures. However, models using equations that do not consider all meteorological controls may be very misleading. Vegetation plays an important role in evaporation by intercepting precipitation and by determining the rate of transpiration. Higher CO2 concentrations may lead to increased water use efficiency (WUE, or water use per unit biomass), implying a reduction in transpiration. Higher CO2 concentrations may also be associated with increased plant growth, compensating for increased WUE - thus, plants may acclimatize to higher CO2 concentrations. The actual rate of evaporation is constrained by water availability.

Most increases in greenhouse gasses (GHGs) are associated with reduced soil moisture in the Northern Hemisphere summers (IPCC, 2001a). This is the result of higher winter and spring evaporation, caused by higher temperatures and reduced snow cover, and lower rainfall during the summer. It appeared that a lower waterholding capacity of the soil results in greater sensitivity to CC. Furthermore, increased winter rainfall in the Northern Hemisphere may result in increased groundwater recharge. However, increased temperatures may increase evaporation, which leads to longer periods of soil water deficits.

In general, however, it is difficult to identify trends in hydrological data (Fig. 1.6). First of all, records are short, and monitoring stations are continuing to be closed in many countries. An alternative is the use of remote sensing to assess runoff. Secondly,

Fig. 1.6. Annual precipitation trends between 1900 and 2000.

trends in the records of stream flow are obscured by interannual and decadal scale climate variability and non-climatic factors, such as land use change and various river management measures. Some general implications described by the IPCC (Arnell et al., 2001), however, are:

• in systems with large reservoirs, changes in resource capacity may be proportionally smaller than changes in riverflows;

• potential effects of CC must be considered in the context of changes in water management - CC changes may have little effect on the water resources as compared to changes in water management over a period of 20 years; and

• the implications of CC are likely to be the greatest in systems that currently are highly stressed.

Impacts on extreme events

Relatively few studies have examined CC effects on flooding frequencies. The main reason is that the generally available general circulation model (GCM) scenarios are monthly averages, which are not suitable in the study of small-scale, short duration events. A flood frequency study was conducted by Mirza (2002) in South Asia. Here, according to four GCM scenarios, the flood discharges in the Ganges Brahmaputra Meghna (GBM) basin could increase by 6-19%.

Droughts are even more difficult to define in quantitative terms compared to floods, since changes in water resources management have a relatively large effect on drought as well as climatic and hydrological inputs. The effects of drought are often expressed with water resources stress indicators. These include, for example, 'amount of water available per person' and 'ratio of volume withdrawn to volume available'. Projections show that 0.5 billion people could see increased water resources stress by 2020 as a result of CC. Case studies show that the impacts of different demands and operational assumptions by 2050 are greater than, or of similar magnitude to, the potential impacts of CC (e.g. Lettenmaier et al., 1999).

Regional differentiation in impacts

The impacts of CC on hydrology are usually estimated by defining scenarios for changes in climatic inputs through a hydrological model. These scenarios can be of two basic types: (i) synthetic scenarios, consisting of assumed changes in temperature and precipitation, using generally available datasets or a weather generator; or (ii) outputs from previously mentioned general circulation models (GCMs) are also increasingly used.

The use of GCM output data, however, suffers from three major problems. First, GCMs are often not able to simulate the current observed regional and local climate. Secondly, the spatial scale is too coarse to be directly used in hydrological models that simulate changes at a much lower spatial and temporal scale. Thirdly, different models project different changes in precipitation. The latter is mainly because changes in the hydrological cycle due to CC are more difficult to simulate than potential impacts through temperature changes because precipitation observations are less complete and the physical constraints are weaker (e.g. Dvorak, 1997; Allen and Ingram, 2002).

Regional precipitation patterns have been modelled by GCMs under different climate change scenarios (see also Chapter 2). The results and the analysis of cross-model consistency for two different emission scenarios (called A2 and B2) in regional precipitation change have been assessed by IPCC (2001a).

Climate Change and Water Use

Water demand is a synonym for human and environmental 'water requirements'. There are in-stream demands (no withdrawals, e.g. hydropower generation, navigation) and off-stream demands (withdrawals). Off-stream demands can be either consumptive (e.g. irrigation) or non-consumptive (water is returned to the river).

Agricultural use is the largest consumer of water around the world, accounting for 67% of all withdrawals and 79% of all water consumed (FAO, 2003). Municipal or domestic uses account for 9% of withdrawals. It is expected that water withdrawals would exceed 40% of annual water availability comparing 2025 and 1995 values (Alcamo and Heinrichs, 2001). The greatest rates are projected in developing countries, e.g. in Africa and the Middle East (without taking CC into account). Water withdrawals are expected to fall in developed countries because of water pricing, for example. Industrial water withdrawals account for 20% of all withdrawals. Without CC, these withdrawals will increase, and are concentrated largely in Asia, Latin America and Africa (IPCC, 2001b).

The expected effects on agricultural use are: (i) a change in field-level climate may alter the need for and timing of irrigation. Increased dryness may lead to increased demand, but demand could be reduced if soil moisture content rises at critical times. (ii) Higher CO2 concentrations would lower plant stomatal conductance, hence increase the WUE, but this may be offset to a large extent by increased plant growth.

An important impact of CC and CV is 'drought', when rainfall drops below the long-term average, and the largest regional reduction in cereal production is therefore expected in Africa (2-3%). At higher latitudes, increased temperatures can lengthen the growing season and ameliorate cold temperature effects on growth. In warmer mid-latitude environments, adverse effects could include increased pests and crop diseases, soil erosion, increased flooding, desertification and reduced water resources for irrigation (Fischer et al., 2001).

When studying the effects of CC on water for food security, related issues have to be considered. First, the effects due to CC are small compared to economic and technological growth. Secondly, it is expected that a rise in atmospheric CO2 can also be a positive factor in tree and crop growth and biomass production. It stimulates photosynthesis (the so-called CO2 fertilizer effect) and improves water use efficiency (Bazzaz and Sombroek, 1996).


Adaptation increasingly receives attention in policy making as a complementary coping mechanism to mitigation. Adaptation is explicitly addressed in several policy documents, such as the declarations of the UNFCCC (1992), where Article 4.1b states that 'parties are committed to formulate and implement national, and where appropriate, regional programmes containing measures to facilitate adequate adaptation to Climate change'. This has been strengthened in article 10 of the Kyoto Protocol (UNFCCC, 1997), where 'parties are further committed to promote and facilitate adaptation and deploy adaptation technologies to address Climate change'.

As stated in the Introduction, the process of adaptation is not new. Throughout history, people have adapted to changing or extreme climate conditions. Think of the development of both Dutch and Bengal villages on dykes and man-made hills in flood-prone areas, or the development of irrigation and reservoir networks in the desert areas of California and Iran. In these cases, people have adapted their way of living to extreme environmental conditions. Within the context of anthropogenic climate change, however, the question is how robust these adaptations are to future, unknown extreme conditions.

Different definitions of adaptation exist. Burton et al. (1998, p. 66) describe adaptation as 'responses to climate change that may be used to reduce vulnerability', where vulnerability is defined as susceptibility to harm or damage potential. Furthermore, adaptation considers such factors as the ability of a system to cope or absorb stress or impacts and to 'bounce back' or recover. Adaptation can also refer to actions designed to take advantage of new opportunities that may arise as a result of climate change. Pittock andJones (2000) define adaptation as a response to climate change that seeks to maintain viability by maximizing benefits and minimizing losses.

In order to study the development and evaluation of adaptation strategies for river basins, ADAPT will assess the necessary ingredients of an adaptation strategy and aim to thoroughly understand the system to which the adaptation strategy will be applied. Smit et al. (1999) specify a number of elements that need to be addressed in any scientific adaptation research. These elements, posed as questions, are: (i) who or what adapts? (ii) Adapts to what? (iii) How does adaptation occur? A possible fourth question relates to the quality of adaptation: How good is the adaptation?

Who or what adapts?

The first element refers to the system definition: who or what adapts? Is it an individual or a community, a region or a nation? Are we considering a species or an ecosystem? A system definition is important for studies where impacts are assessed, with and without adaptation and to determine its vulnerability.

Once the system has been defined, the next step is to characterize the system. These characteristics should allow for assessing the prospects of planned adaptation and adaptive capacity (the latter is also referred to as autonomous adaptation) of the system under survey (Smit et al., 1999). Important system characteristics are: sensitivity, vulnerability, susceptibility, coping range, critical levels, adaptive capacity, stability, robustness, resilience and flexibility (e.g. Klein and Tol, 1997; Smit et al., 1999; Reilly and Schimmelpfennig, 2000).

What is adaptation?

What is adaptation?

How good is the adaptation?


• Principles

Fig. 1.7. Main elements for studying adaptation strategies. (From Smit etal., 1999.)

The most broadly used terms for a system's characterization are: sensitivity, adaptive capacity (or adaptability) and vulnerability (Smit et al., 1999). Sensitivity can be defined as the degree to which a system is affected by or responsive to climate stimuli. Adaptive capacity is the extent to which sectors, regions and communities ('the system') are able to adapt to climate change impacts. Adaptive capacity reflects the notion that the existence of adaptation options does not necessarily mean that each vulnerable sector, region or community has access to these options or is in a position to implement them. In other words, it is the capacity to adapt rather than the availability of adaptation options that determines the degree of resilience to climate change (Smit et al., 2001). Options to increase adaptive capacity include: increasing wealth, scientific understanding, technology and flexibility (e.g. through developing early warning systems). Finally, the vulnerability of a given system or society is a function of its physical exposure to climate change effects and its ability to adapt to these conditions (Kelly and Adger, 2000; Smit et al., 2001).

The IPCC (Smit et al., 2001) further states that by assessing differences in vulnerability among regions and groups and by working to improve the adaptive capacity of those regions and groups, planned adaptation can contribute to equity considerations of sustainable development and may contribute to alleviating poverty in developing countries. Hence, vulnerability is a function of sensitivity and adaptive capacity and its concept is increasingly seen as key input for developing and evaluating adaptation strategies (IPCC, 1998).

Fig. 1.8. A shift of climate variable Xpc in terms of its mean and frequency distribution results in a future distribution Xcc. The figure shows how the current adaptation zones are insufficient in the future, since the risk zone (shaded zone) increase dramatically.

Values of climatic attribute (X)

Fig. 1.8. A shift of climate variable Xpc in terms of its mean and frequency distribution results in a future distribution Xcc. The figure shows how the current adaptation zones are insufficient in the future, since the risk zone (shaded zone) increase dramatically.

Adaptation to what?

The second element in adaptation studies refers to determining the climate stimuli relevant to the system and to relate climate stimuli to the sensitivity of the system. Climate conditions, such as temperature and precipitation, can be classified into three temporal categories: (i) long-term changes; (ii) inter-annual or decadal changes; and (iii) isolated extreme events. This notion is important, since current research emphasizes the importance of studying both gradual climate change and climate variability, including extreme events. Spatial characteristics of climate stimuli also play a role, as projected regional climate change can be very different from what is simulated for the globe (Smit, 1993; Tol, 1996).

The most important climate stimuli that influence the hydrological cycle of a river basin and hence the availability of water resources are temperature and precipitation. It is, therefore, important to derive some quantitative projections of expected temperature and precipitation changes in the future. From these projections, impacts or effects can be determined. It is also important to make a distinction between changes in the frequency of extreme events versus gradual climate changes. This can be illustrated with Fig. 1.8. It shows the frequency distribution of variable climate Xpc (e.g. precipitation) with adaptation ranges. That is, currently adaptation measures have been implemented to cope with extreme events within those ranges. However, a change in the mean and frequency of a climate variable will cause a shift in the risk zone, or the zone that is not covered through either autonomous or planned adaptation. The risk zone in the future is considerably larger than in the current situation (Smit et al., 1999; Kabat and van Schaik, 2003).

How does adaptation occur?

This element refers to the process of developing and implementing adaptation and the forms of adaptation. Figure 1.9 summarizes the general types of adaptation and some examples, which are differentiated according to timing (anticipatory versus reactive) and



Human Natural systems systems

Public Private


• Changes in length of 1 growing season

• Changes in ecosystem 1 composition

• Wetland migration

• Purchase of insurance

• Construction of house on 1 stilts

• Redesign of oil-rigs

• Changes in farm practices

• Changes in insurance 1 premiums

• Purchase of air-conditioning

• Early warning systems

• New building codes, 1 design standards

• Incentives for relocation

• Compensatory payments, 1 subsidies

• Enforcement of building 1 codes

• Beach nourishment

Fig. 1.9. Types of adaptation to climate change, including examples. (From Smit et al., 2001.)

human versus natural. Anticipatory adaptation is also referred to as preventive adaptation (thus implemented 'before the event'). Human-induced adaptation is also referred to as planned adaptation and can be differentiated according to the type of agents - private enterprises such as producers and industries versus public governmental organizations. Other characteristics of adaptation options include intent, duration, form and type (Klein and Maclver, 1999; Maclver and Dallmeier, 2000; Smit and Skinner, 2002).

Most adaptations are modifications to existing practices and public policy decision-making processes such as those already developed in the agriculture and water sector. Therefore, there is a need for a better understanding of the relation between potential adaptations and existing practices (Reilly and Schimmelpfennig, 2000). Kabat and van Schaik (2003) describe an extensive compendium of possible adaptation measures. These measures have formed the basis for developing adaptation strategies studied within this book.

How good is the adaptation?

Adaptation research warns that most impact studies do not consider adaptation as an integral part of their assessment. Without assessment of adaptations, these impact studies could well overestimate the potential negative effect of climate change (Burton et al., 1998). It is, therefore, suggested to include adaptation options iteratively within impact assessment studies by first identifying potential impacts without adaptation and next simulating impacts including adaptation. The potential set of adaptation options that results from this iterative process can finally be evaluated using a number of evaluation methods, e.g. sensitivity analysis, scenario analysis and multi-criteria analysis. Another evaluation technique is a cost-benefit analysis, but estimations of adaptation costs versus the losses without adaptation is still a relatively unexplored scientific field (Leary, 1999; Tol et al., 1998).

Challenges for adaptation research

Efforts are required to better understand the human response to climate change, as it is likely that socio-economic systems respond the most to extreme realizations of climate change (Yohe and Dowlatabadi, 1999). Therefore, adaptation research needs to consider socio-economic scenarios, although it may affect the uncertainty of the result. Also, it is important to distinguish between a prediction or estimation of the effect of adaptations and, on the other hand, a normative study that focuses on the evaluation of adaptations. The latter requires information from the impact study to derive a set of feasible adaptations and to evaluate this set of adaptations. Policy response studies that incorporate evaluation of adaptation options are relatively new.

With a greater focus on adaptation, the debate on the impacts of CC will not stand in the way of effective mitigation, as most adaptations make sense under any CC scenario (Pielke, 1998). Policy options proposed as adaptation measures to reduce negative impacts of climate change that would be justified even in the absence of climate change are referred to as 'no regret' measures (IPCC, 2001b).

As described above, costs of adaptation are rarely studied and even less is known about the benefits. Most studies focus on total damage costs, including adaptation, and not on avoided damages through adaptation (e.g. Zeidler, 1997). It is estimated that, globally, adaptation costs only comprise about 7-10% of the total damage costs (Tol et al., 1998). Another issue for further study is to estimate transition costs, since most adaptation and impact studies assume equilibrium now and in the future. However, climate continues to change, as do the impacted systems. The optimum level of adaptation minimizes the combined costs of adaptation and residual negative effects, with the most cost-effective steps taken first. Factors that affect adaptive capacity itself include: institutional capacity, wealth, planning time, scale, etc. (Tol et al., 1998).

Finally, more research on the timing of adaptation is required. In this respect, Burton et al. (1998) point out that adaptation in socio-economic sectors is easier when investments relate to activities with a shorter product cycle. For example, different cropping methods can be adjusted every year. But a forest has a life-cycle of decades. Dams are even costlier to reconstruct in order to meet new climate conditions.

Important to realize is that adaptation measures are not only related to climate change. They can be a response to other internal (= manageable) and external (= less manageable) stressors. Examples of these stressors that should be taken into account are: land use change, population growth, increased competition between sectors (urban, industry, agriculture, nature), power generation, transboundary water allocation, environmental concerns, etc. The term 'no regret strategy' is therefore often used, since implementing an adaptation may solve a problem due to climate change in the future. But if not, it may solve other even bigger problems.

Adaptation and agriculture

It is projected that the paramount issue in changes in precipitation will be the increase in extremes rather than a long-term change in average precipitation. Increasing the buffer capacity is therefore the appropriate adaptation measure, where buffer capa city should be considered in terms of increased water storage (reservoirs, soil water, ground water) but also increased economic (savings/loans) and food buffer capacity. Essential is that an increase in extremes includes an increase in successive years of dry or wet periods, which are very difficult to overcome for poor people. A poor farmer might overcome a 1-year drought followed by a normal year, but a period of 2 or more years of drought, even followed by a longer period of normality, will be catastrophic to this farmer.

During the last decade, research has focused on the impacts of climate change on agricultural production (Parry et al., 1999; Fischer et al., 2001; FAO, 2003). These are studies at the global scale and only recently have impact studies involved adaptation at farm level (Smit and Skinner, 2002). Reilly and Schimmelpfennig (1999) suggest evaluating impacts of CC on agriculture by considering vulnerability, where vulnerability in agriculture can be defined in terms of yield, farm profitability, regional economy and hunger. These studies also make clear that lower income populations and marginal agricultural regions, particularly the flood-prone and arid areas, are most vulnerable to CC.

Most adaptations are modifications to existing farming practices and public policy decision-making processes. Therefore, there is a need for a better understanding of the relationship between potential adaptations and existing practices (Smit and Skinner, 2002).

An Adaptation Framework for River Basins

In this research the focus is on adaptation strategies for regional water resources management, which are constrained by the hydro-geographical extensions of a watershed or river basin. The need for integrated basin-wide climate change and water resources studies has been addressed by several studies (Arnell et al., 1996, 2001; Strzepek et al., 1998; Lettenmaier et al., 1999; Kabat and van Schaik, 2003) for a variety of arguments. First, a regional hydrological cycle is bounded by its watershed and is therefore a more appropriate geographical entity than an administrative region or country. Secondly, upstream water-related activities, processes and adaptations have clear effects for downstream water availability. Thirdly, regional water resources management becomes increasingly important in policy making as, for instance, outlined in the EU water framework directive (EU, 2000).

Currently, no specific adaptation framework for river basins exists, although studies point to the relevance of such a framework. Stakhiv (1996) and Frederick (1997) suggest considering the need for adaptation in the water sector, but relevant institutions need to include the process of adaptation in the evaluation criteria that refer to the quality of water resources management. A challenging aspect for developing a generic adaptation framework for river basins is the huge difference in water resources characteristics, environmental controversies and socio-economic issues across river basins. Another issue relates to the above raised question of 'who or what adapts?' For smaller river basins, there is often a water board or a similar institute. For transboundary river basins, however, there is no basin manager or basin-wide institution with a mandate, since country or state borders often determine the jurisdiction of water management. Yet, from a water management perspective, a basin-wide approach still holds for developing and evaluating adaptation strategies, as discussed above.

Thus far, there is a rather ad hoc treatment of adaptation in impact assessments and there is too much focus on technical measures (Smit et al., 2000). Moreover, many impact assessment studies within water resources research lack the evaluation adaptations. This can be improved by including stakeholders in the development and evaluation of adaptation, such as outlined in the concept of Integrated Water Resources Management (IWRM).

The above issues are points of departure for developing a generic adaptation framework for river basins called 'AMR' (generic adaptation methodology for river basins). The basic postulates for developing AMR are:

1. That water management in a river basin has a central role in intervening in the water resources system, including adaptation. It is water management, either at basin level or at local level (e.g. a farmer) that implements new adaptations in order to cope with changed climatic conditions (Mendelsohn and Bennett, 1997).

2. The water resources system provides goods and services that are managed so that current and future values are optimized in relation to the objectives of the regional water management policy (Gilbert and Janssen, 1998; EGIS, 2000).

3. Goods and services are expressed as functions of the state of the water resources system, expressed in the terms 'water availability' and 'water quality' (e.g. Gilbert and Janssen, 1998).

4. Climate change is seen as an exogenous influence on the regional water resources system.

5. AMR must allow for evaluating potential adaptation strategies on the basis of a set of decision criteria or indicators that relate to goals and objectives of regional policies (Aerts and Heuvelink, 2002).

6. In order to identify all relevant indicators and to capture the potential adaptations, AMR should allow for active participation of stakeholders in an iterative development and evaluation.

7. AMR preferably builds on existing approaches, such as research by OECD (1993), EEA (1998), Smit et al. (1999), Wheaton and MacIver (1999), EGIS (2000) and Barker (2003).

AMR: a goal-based performance framework

The point of departure for developing AMR is to seek a structure that addresses both 'policy objectives' (formulated in regional water management plans, for example) and 'the physical state' of the water resources system. This is done in the following way.

The water resources system can be seen as a productive system that provides 'goods and services' for both humans and ecosystems. These goods and services in a river basin can be broadly classified into four categories: water for food, water for industry, water for nature and water for the human environment (e.g. Gilbert and Janssen, 1998; Kabat and van Schaik, 2003). It is assumed that these goods and services relate directly to the state of the water resources system, which itself can be quantified in both 'water availability' (or quantity) and 'water quality'.

It is the primary task of a water manager optimally to manage the water resources system by securing water quality and allocating water in response to demands for all uses. Hence, the definition of optimal management can be expressed as optimally to use the goods and services by:

• enhancing human welfare;

• enhancing food capacity and security;

• enhancing industrial capacity; and

• enhancing natural ecosystems quality.

These four objectives relate to most water-related management issues in any basin, although basins across the globe obviously differ in water resources characteristics, physical and social environments and therefore in the use of the available goods and services. Every water manager in a given basin has to deal with trade-offs between measures with respect to the four above-mentioned objectives. Each water manager has his or her own priorities in those objectives. For example, within the Rhine Basin in Western Europe, both security against floods and preserving industrial capacity in the form of the number of navigable days have priority above enhancing environmental quality. Water management measures in the Volta Basin, on the other hand, are more targeted to enhance food security through irrigation supply and to preserve hydropower generation.

By quantifying these objectives and priorities, potential measures (including adaptation) can be assessed on their performance. The link to the water resources system determines whether it is feasible to attain such objectives under a given set of adaptation measures. Figure 1.10 schematizes this approach and shows the four objectives of a water manager/policy maker on the left, while the state of the water resources system is presented on the right, expressed in the boxes 'quantity' and 'quality'.

State and decision indicators

In order to operationalize AMR, we need to define sets of indicators that reflect the four aforementioned objectives (decision indicators) and that represent the water resources system in detail (state indicators). Decision indicators allow for quantifying the performance of water management measures (including adaptation strategies) with respect to the goals (Fig. 1.10).

An indicator has to meet several criteria in order to make it operational: (i) an indicator has to be representative with respect to the goal it represents; (ii) it must be flexible to use and understandable for all stakeholders and users involved in using the framework; (iii) the data needed to measure an indicator must be available; and, finally, (iv) indicators must be generally comparable across the different basins and should preferably aggregate to an index (Cole et al., 1998). Indicators are discussed in detail in Chapters 3 and 4.

The identification of decision indicators (DI) may be done using a hierarchical tree, starting with the main objectives and going down to finally arrive at a set of measurable indicators. Figure 1.10 shows an example of such a tree. The four main objectives are presented on the left. From these objectives, a set of intermediate objectives


Human welfare

Food capacity

Intermediate objective

Quality of nature

Industrial capacity





—1 Water quality



Decision indicator

Access to drinking water Vector-borne diseases People at risk ha Floodplain forest -ha Upstream forest

- % Longitudinal freedom Wetland species at risk Fish species at risk

- Yield

- Total production

■ Tonnes of river fish

- Water consumption Farm income

■ Variation in yield

- Gross revenue

- Water productivity Variation in farm income kW produced

I-No. dams and barrages

State indicator State of WRS

Functional relationships

No. floods Droughts Annual discharge


Fertilizer -Etc.



- No. un-navigable days

Fig. 1.10. Decision tree with policy objectives and derived decision indicators on the left, and state indicators on the right (Aerts et al., 2003).

can be derived, which together address the intent of the objectives. From the intermediate objectives, decision indicators (sometimes called 'criteria') can be derived that allow for quantifying the attainment of water management measures (including adaptation) with respect to the objectives. Note that objectives have to be quantified as well. Thus the objective of 'enhancing human welfare' is, for example, characterized by: (i) '75% of all inhabitants of a certain river basin have to have access to safe drinking water'; and (ii) 'only 35,000 people are exposed to risks from floods'.

Furthermore, a set of state indicators (SI) can be defined that characterizes the state of the water resource system (WRS) of a river basin in terms of water quantity and water quality. These usually relate to quantifiable indicators as 'annual discharge', 'no. of droughts', 'BOD concentration', etc.

EGIS (2000, p. 11) states that 'the functional relations between SIs and DIs are usually expressed in terms of changes'. This is where both simulation models and the role of professionals and stakeholders is required. For example, a hydrological model is capable of simulating the state of the WRS in the form of calculating SIs such as discharges and number of droughts. A subsequent food production model may use these calculations as input for calculating DIs such as 'yield' and 'water productivity'.

Figure 1.10 shows the interlinkage between DIs and SIs. The set of indicators (both DIs and SIs) is derived from the ADAPT Project and obviously not a generic list applicable for all basins. The framework allows inserting new indicators that pertain to a particular case study.






Fig. 1.11. The DPSIR chain of cause-effect relationships (EEA, 1998).

Involving stakeholders

In order to involve stakeholders in the process of developing and evaluating adaptations, we need to provide these stakeholders with an assessment structure. For this, the DPSIR (Drivers, Pressures, State, Impacts and Response) approach is suggested. DPSIR allows the structuring of issues and problems in a basin and finally developing responsive adaptation strategies to cope with the impacts of external drivers such as climate change (OECD, 1993; EEA, 1998). The DPSIR approach was developed by OECD (1993) and later expanded by the European Environment Agency (EEA, 1998) for indicator-based reporting on the environment. The approach assumes cause-effect relationships between interacting components of social, economic and environmental systems (Fig. 1.11).

The DPSIR approach can be divided into five parts, which are here explained within the context of a water resources system of a river basin:

1. Driving forces, such as population growth, economic growth and climate change. They act upon the

2. Pressures, which are activities and/or pollutants resulting from the influence of the drivers. Most commonly, pressures are very much related to 'issues and problems' of the water resources system. These pressures cause a change in the

3. State of water resources system of a river basin, expressed in terms of (proxi-) indicators. Through quantifying a change, the

4. Impacts on the WRS can be determined. This may induce a

5. Response by a water manager in the form of policy measures or technical interventions.

Barker (2003) outlines several weak aspects within the DPSIR approach. The most important argument in this research is the fact that DPSIR does not address feedback mechanisms with respect to account for effects of mitigations. For instance, mitigation measures may alter the effect of climate drivers and hence change the way a response strategy will be implemented.

The weak points of DPSIR are addressed in AMR. First, the effects of mitigation strategies within regional studies can be neglected and will be strictly regarded as exogenous influences. Secondly, all driving forces are explicitly separated from other variables (the DIs and SIs) through treating them as exogenous influences. Thirdly, physical feedback mechanisms will not be considered, except for incorporating CO2 fertilization (Reilly and Schimmelpfennig, 1999). The framework, in addition, particularly addresses the involvement of stakeholders in an iterative process, and hence allows for policy feedbacks within the cause-effect chain. Finally, the integration of DPSIR and the goal-based framework allows for an evaluation of adaptation strategies through using the decision indicators as evaluation criteria. Its simplicity, though, is a strong point and DPSIR strongly 'steers' the response strategy towards the problem.

AMR in practice

Figure 1.2 shows the total AMR framework. The circle shows the participative approach where stakeholders in a river basin play a key role. They are confronted with: (i) exogenous influences such as population growth and climate change; (ii) how these impact the WRS; and (iii) how these impacts affect the goals and decision variables. Next, the framework allows for: (iv) deciding which potential adaptation strategies can be applied to respond to the impacts. And finally, (v) each of these strategies can be evaluated by measuring their performance against the pre-set goals.

The following example explains the use of AMR by going through the indicated boxes in Fig. 1.2. Consider a river basin that is potentially vulnerable to climate change. (1) Scenarios point to an increase in temperature and hence both increased evaporation and droughts in the basin. (2) Hydrological models have calculated the water availability according to both a Business As Usual (BAU) scenario and a CC scenario. (3) Next, food and wetland models use these water availability figures in order to calculate the effects of CC for two decision indicators: 'Rice production' in tonnes per hectare and 'Preserved wetlands' in hectares. First, an increase in rice production under CC (from 2000 to 2500 t/ha) is projected, but is much less than would be expected without CC (the target is 3500 t/ha). Secondly, the area of viable wetlands will decrease under CC from 150,000 to 100,000 ha, while the national wetland protection plan aims for a stable area of 150,000 ha of wetland in the future. (4) Stakeholders, including water managers, have selected two possible adaptations. One is targeted to alleviate impacts on rice production in the basin by increasing irrigation efficiency (A1, Table 1.1). The other adaptation measure (A2) aims at preserving downstream wetlands at the expense of irrigated rice production. (5) The previous food and wetlands models have calculated the effects of implementing these adaptations (Table 1.1). It appears in this example that A2 performs better than A1 when considering both indicators as equally important. Although A1 better achieves the goal formulated for 'rice production', it shows poorer performance on achieving 'wetland preservation'.

Note that, in practice, decision indicators are certainly not preferred equally and they have to be prioritized in order to finally use them in an evaluation of different management options. This process of prioritizing and evaluating using indicators (also referred to as 'criteria') is called 'Multi Criteria Analysis'.

Table 1.1. Example of measuring the performance of adaptation strategies (A1 and A2) against a set of two decision indicators (rice production and wetland preservation).




Future with

Future with

situation Objective

with CC

CC + A1

CC + A2

Slack 1:

Slack 2:

Slack 3:


% from

% from

% from





Rice production

2,000 3,500



3,000 14

2,800 20



150,000 150,000



80,000 47

120,000 20

preserved (ha)

In the previous example, the adaptation strategy A1 'increasing irrigation efficiency' is probably a set of individual measures. For instance, increasing irrigation efficiency can be achieved through better irrigation techniques, other crops, or new crop rotation schemes. Thus, an adaptation strategy is seen as a set of individual measures that through joint implementation can achieve the different objectives. We have seen in the same example that these are often conflicting objectives, as adaptation strategy A2 is favourable for preserving nature, while A1 is primarily targeted at food security. On many occasions, though, a water manager seeks to satisfy all objectives and hence must combine a variety of measures, which together form a sustainable and balanced adaptation strategy.

Concluding Remarks

In this chapter we have provided an extensive overview of planning for adaptation for water managers, with a special focus on climate change. Although some references have been provided, adaptation strategies are still in their early stages in comparison with other climate change related issues, such as mitigation and projection. We have provided a systematic approach, here referred to as AMR: Adaptation Methodology for River Basins. AMR aims to link policy objectives with the physical state of water resources in a river basin. The four focal areas are: water for drinking, water for food, water for industry and water for ecosystems, while the sets of 'decision indicators' and 'state indicators' complete the entire framework.

In the subsequent chapters of this book, and especially the seven chapters referring to individual basins, this approach will be demonstrated. Not every chapter will exactly copy the approach described here, but the key components can be found in each basin chapter. Readers may wish to pay special attention to the way the four focal areas of AMR (domestic, industry, food, ecosystems) are addressed in the different river basins.


Aerts, J.CJ.H. and Heuvelink, G.B.M. (2002) Using simulated annealing for resource allocation. International Journal of Geographical Information Science 16, 571-587.

Aerts,J.CJ.H., Lasage, R. and Droogers, P. (2003) A Framework for Evaluating Adaptation Strategies. Institute for Environmental Studies report R-03/08. Vrije Universiteit Amsterdam, The Netherlands.

Alcamo, J. and Heinrichs, T. (2001) Critical regions for water and vulnerability. Proceedings Global Change Open Science Conference 'Challenges of a Changing Earth', 10-13 July 2001, Amsterdam, The Netherlands.

Allen, M.R. and Ingram, WJ. (2002) Constraints on future changes in climate and the hydro-logic cycle. Nature 419, 224-232.

Arnell, N.W., Bates, B., Lang, H., Magnuson, JJ., Mulholland, P., Fisher, S., Liu, C., McKnight, D., Starosolszky, O., Taylor, M., Aquize, E., Arnott, S., Brakke, D., Braun, L., Chalise, S., Chen, C., Folt, C.L., Gafny, S., Hanaki, K., Hecky, R., Leavesly, G.H., Lins, H., Nemec, J., Ramasastri, K.S., Somlyôdy, L. and Stakhiv, E. (1996) Hydrology and freshwater ecology. In: Watson, R.T., Zinyowera, M.C. and Moss, R.H. (eds) Climate Change 1995 -Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses. Contribution of Working Group II to the Second Assessment of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, pp. 325-363.

Arnell, N.W.C., Liu, R., Compagnucci, L., Da Cunha, C., Howe, K., Hanaki, G., Mailu, I., Shiklomanov, A., Doll, P., Becker, A. and Zhang, J. (2001) Hydrology and water resources. In: McCarthy, JJ., Canziani, O., Leary, N.A., Dokken, D.J. and White, K.S. (eds) Climate Change 2001 - Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, pp. 191-233.

Barker, T. (2003) Representing global climate change, adaptation and mitigation. Global Environmental Change 13, 1-6.

Bazzaz, F. and Sombroek, W. (1996) Global Change and Agricultural Production. Food and Agriculture

Organization of the United Nations (FAO), John Wiley & Sons, New York.

Burton, I., Smith, J. and Lenhart, S. (1998) Adaptation to climate change: theory and assessment. In: Feenstra, J.F., Burton, I., Smith, J.B. and Tol, R.SJ. (eds) UNEP Handbook on Methods for Climate Change Impact Assessment and Adaptation Strategies. Institute for Environmental Studies, Vrije Universiteit, Amsterdam.

Cole, D.C., Eyles, J. and Gibson, B.L. (1998) Indicators of human health in ecosystems: what do we measure? The Science of the Total Environment 224, 201-213.

Cosgrove, WJ. and Rijsberman, F.R. (2000) World Water Vision - Making Water Everybody's Business. Earthscan Publications, London.

Dvorák, V, Hladny, J. and Kaspárek, L. (1997) Climate change hydrology and water resources impact and adaptation for selected river basins in the Czech Republic. Climatic Change 36, 93-106.

EEA (1998) Towards Environmental Pressure Indicators for the EU - First Edition. European Environment Agency, Copenhagen.

EGIS (2000) Blue Accounting - Introduction to a Methodology for Monitoring and Assessing the Functionality of the Water Resources System. EGIS technical note 15. Dhaka, Bangladesh.

EU (2000) Directive 2000/60/EC, EU Water Framework Directive. Brussels, Belgium.

FAO (2003) World Agriculture: Towards 2015/2030. An FAO Perspective. Bruinsma, J. (ed.) Earthscan Publications, London.

Fischer, G., Shah, M., van Velthuizen, H. and Nachtergaele, F.O. (2001) Executive Summary Report: Global Agro-ecological Assessment for Agriculture in the 21st Century. International Institute for Applied Systems Analysis, Laxenburg.

Frederick, K.D. (1997) Adapting to climate impacts on the supply and demand for water. Climatic Change 37, 141-156.

Gilbert, A.J. andJanssen, R. (1998) Use of environmental functions to communicate the values of a mangrove ecosystem under different management regimes. Ecological Economics 25, 323-346.

IPCC (1998) The Regional Impacts of Climate Change - An Assessment of Vulnerability. Watson, R.T.,

Zinyowera, M.C. and Moss R.H. (eds) Cambridge University Press, Cambridge.

IPCC (2001a) Climate Change 2001 - The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J.T., Ding, Y., Griggs, DJ., Noguer, M., van der Linden, PJ., Dai, X., Maskell, K. and Johnson, C.A. (eds) Cambridge University Press, Cambridge.

IPCC (2001b) Climate Change 2001 - Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. McCarthy, JJ., Canziani, O.F., Leary, N.A., Dokken, DJ. and White, K.S. (eds) Cambridge University Press, Cambridge.

Kabat, P. and van Schaik, H. (2003) Climate Changes the Water Rules: How Water Managers Can Cope with Today's Climate Variability and Tomorrow's Climate Change. Dialogue on Water and Climate. Printfine, Liverpool.

Kane, S. and Shogren, J. (2000) Linking adaptation and mitigation in climate change policy. Climatic Change 45, 75-101.

Kelly, P.M. and Adger, W.N. (2000) Theory and practice in assessing vulnerability to climate change and facilitating adaptation. Climatic Change 47, 325-352.

Klein, R.J.T. and Tol, R.SJ. (1997) Adaptation to Climate Change: Options and Technologies. An overview paper. Technical Paper FCCC/ TP/1997/3, United Nations Framework Convention on Climate Change Secretariat, Bonn, Germany.

Klein, R.J.T. and MacIver, D.C. (1999) Adaptation to climate variability and change: methodological issues. Mitigation and Adaptation Strategies for Global Change 4, 189-198.

Leary, N.A. (1999) A framework for benefit-cost analysis of adaptation to climate change and climate variability. Mitigation and Adaptation Strategies for Global Change 4, 307-318.

Lettenmaier, D.P., Wood, A.W., Palmer, R.N., Wood, E.F. and Stakhiv, E.Z. (1999) Water resources implications of global warming: a U.S. regional perspective. Climatic Change 43, 537-579.

MacIver, D.C. and Dallmeier, F. (2000) IPCC Workshop on adaptation to climate variability and change: adaptive management. Journal of Environmental Monitoring and Assessment 61, 1-8.

Mendelsohn, R. and Bennett, L.L. (1997) Global warming and water management: water allocation and project evaluation. Climatic Change 37, 271-290.

Middelkoop, H., Daamen, K., Gellens, D., Grabs, B., Kwadijk, J.CJ., Lang, H., Parmet, B.W.A.H., Schädler, B., Schulla, J. and Wilke, K. (2001) Impact of climate change on hydr

0 0

Post a comment