59808 v1 A REVIEW OF SELECTED HYDROLOGY TOPICS TO SUPPORT BANK OPERATIONS HEF Technical Report 1 ­ June 2010 Technical Report 2010 HEF HYDROLOGY EXPERT FACILITY AN EXPERT SUPPORT TEAM (EST) OF THE WATER PARTNERSHIP PROGRAM (WPP) Acknowledgments This document was prepared by Luis E. García and Gabrielle Puz under the direction of Abel Mejia, former Manager of the Water Anchor. The authors thank the workshop presenters, and discussants, namely, Torkil Jønch Clausen, Janusz Kindler, Juan B. Valdés, Bob Meade, Jeffrey Richey, Peter Droogers, Ignacio Rodriguez-Iturbe, Kenneth Strzepek, Pedro Restrepo, Curt Barrett, Peter J. Kolsky, Stephen Foster, Douglas C. Olson, Nagaraja Rao Harshadeep, Winston Yu, Ronald N. Hoffer, Grant Milne, Bekele Debele Negewo, Julia Bucknall, Paola Agostini, for their valuable contributions. The collaboration and participation in the workshop of NOAA, USACE, SWAT, and GW-MATE is gratefully acknowledged. Diego Rodriguez, Arno Boersma, Azamat Tashev, Meike van Ginneken, Immaculate Bampadde, Parivash Mehrdadi, Zsuzsanna P. Baldiviezo, Sanna- Leena Anneli Rautanen, Eileen Burke, Hywon Cha Kim, Doreen Kirabo, Eric Dickson, Benedicte Marie Cecile Augeard, contributed to the organization and realization of the workshop. The workshop and the publication of this report were made possible by the financial support provided by the Bank Netherlands Water Partnership Program (BNWPP) and the Water Partnership Program (WPP). Special thanks to Ashok Subramanian and the peer reviewers Matthijs Schuring, Rita Cestti and Grant Milne, who provided valuable guidance and suggestions for this report. Disclaimer The opinions expressed herein are those of the authors and do not necessarily represent the official position of the World Bank Group. The data provided in the tables and figures are not official Bank data. Permission is granted to reproduce this publication in whole or in part for noncommercial purposes only and with proper attribution to the authors, the Water Anchor and the Bank. B A Review of Selected Hydrology Topics to Support Bank Operations Papers from the Workshop Hydrologic Analysis to Inform Bank Policies and Projects: Bridging the Gap November 24­25, 2008 Washington, DC Washington, DC, June 30, 2010 ii Table of Contents Introduction ...................................................................................................................................................................3 Key Messages ..............................................................................................................................................................7 Papers ......................................................................................................................................................................... 11 1 Application of Integrated Approaches in Water Resources Management Beyond the Conceptual Phase, Torkil Jønch-Clausen ............................................................................................ 13 2 Water Demand Management, Janusz Kindler ............................................................................................ 35 3 Managing Droughts and Floods under Changing Environments, Juan B. Valdés ............................. 51 4 Sediment Transport and Deposition in Rivers: The Case for Non-Stationarity, Robert H. Meade ............................................................................................................................................... 69 5 Land-Ocean Interactions: Human, Freshwater, Coastal, and Ocean Interactions under Changing Environments, Jeffrey E. Richey.................................................................................................. 77 6 Managing the Real Water Consumer: Evapotranspiration, Peter Droogers ........................................ 99 7 Addressing The Links Between Hydrology and Watershed Climate, Soil and Vegetation, Ignacio Rodriguez-Iturbe ...............................................................................................................................111 Comments of World Bank Discussants ............................................................................................................119 Annexes in CD I. Workshop Program II. Bio-Briefs of Presenters III. Presentations 1 2 Introduction The World Bank's 2004 Water Resources Sector water resources management (IWRM) was embraced by Strategy focused on the need for both water resources the Bank's 1993 Water Resources Management Policy management and development in dealing with growth Paper and the ensuing 2004 Water Resources Sector and poverty alleviation. Planning and design of new Strategy, which gave strategic directions for World Bank hydraulic infrastructure for water supply and sanitation, engagement in water. Climate changes, as well as changing food production, hydropower generation, flood protection, conditions in river basins and watersheds, demand the ecosystem restoration or other such purposes require development and implementation of new integrated water dealing with all elements in the interaction among land, resources approaches while addressing the reality of low water, vegetation, human intervention and climate variability capacity and unfavorable local conditions. In these new and change, with an emphasis on the end-user. They approaches, the focus is on reaching the end user and also require the simultaneous consideration of technical, achieving results in the field, stressing the importance of economic, institutional (governance), political, financial, multisector and multistakeholder participation to address environmental and social factors, as called for in the Bank's the often conflicting interests of major water uses in a basin 1993 Water Resources Management Policy. or watershed management context. Although IWRM is a widely accepted concept, it has encountered hurdles in To provide high-level insight on the key hydrology issues implementation that go beyond the conceptual phase and involved, a group of world class experts gathered at a require additional analysis and discussion. workshop held at World Bank Headquarters in November 2008. The workshop was organized by the Hydrology The three-part paper by Torkil Jønch-Clausen Expert Facility (HEF) of the Water Anchor. The presenters focuses on integrated water resources discussed advancements in key hydrologic topics that were management (IWRM) "beyond the conceptual selected for their relevance to Bank operations. The focus phase," moving away from the rhetoric to results was on potential implications for the Bank's development "on the ground." Part 1 of this paper provides a assistance on water projects, programs and policies. brief status update of IWRM and how its principles are actually being applied around the world. In Part A wide spectrum of topics was presented for discussion 2, recent work by the author in Orissa, India, is in a workshop that was more exploratory than analytical. used to illustrate how a "roadmap for IWRM" can, Its main purpose was to identify the interest of the Bank's in fact, help a state (or country) define the relevant water community for those topics that could jointly be small steps leading towards improved water moved forward by following actions aimed at further resources development and management. In Part 3 dissemination and development of specific knowledge the author provides some brief personal reflections products. on how IWRM has evolved in the World Bank. The remainder of this introduction provides a brief Water scarcity and the degradation of water quality (both description of the topics and lays forth the reasons for surface and groundwater) are reaching alarming proportions selecting each of them for discussion at the workshop. in many parts of the world. It is often better, in economic and environmental terms, to manage existing supplies than to develop new ones. Demand management on a national Integrated Water Resources Management and basin level, as well as in urban areas, has become an important component of policy making and water resources This topic was selected because of its importance in planning and management. This topic was selected to current World Bank support for the role that water plays underline the importance of undertaking adequate demand in the development process. The concept of integrated management analyses in Bank water projects that focus on 3 growth-oriented sustainable poverty alleviation, especially The selection of this topic follows the need to maintain a in water scarce areas. It is also important because of the focus on data, even if basic hydrologic assumptions such as challenges that need to be faced in this area. Increased stationarity, are under scrutiny. The topic was also selected scarcity and costs highlight the importance of increased to stress the importance of not losing sight of the need efficiency. Challenges differ at different scales and for different to account for climate variability for the planning, design, users, and while water demand analysis has advanced and operation of water resources projects in the short and considerably over the last decades, there are still many medium term, along with climate change in the long term. unresolved issues and limitations requiring more research. In his paper, Juan Valdés discusses efforts to In his paper, Janusz Kindler synthesizes the state characterize floods and droughts and refers to of the art in this area and reflects on how it can be the state of the practice in dealing with extremes applied to water sector policies and water projects. based on instrumental records and the hypothesis The author presents the fundamentals of water of stationarity. He also discusses the importance of demand analysis and modeling approaches, and then maintaining data collection under non-stationarity discusses the demands of individual water users, conditions and using paleoclimatic data to increase such as households, industrial plants and irrigation the instrumental record length, the use of climate systems. Special attention is given to the relatively projections from the most recent runs of the new group of hydro-economic models and their role global circulation models (GCMs) for hydrologic in water demand analyses. The paper closes with a applications, and the implications of climate few comments on the role of water demand analysis, variability and change on the management of water its limitations, and further research needs. resources systems. Climate Variability and Change Hydrologic Interactions The latest reports issued by the Intergovernmental Panel There are many hydrologic interactions. However, three on Climate Change (IPCC) indicate that climate change were selected for this discussion because of their will intensify the hydrological cycle, making hydrologic importance to the Bank's water-related portfolio. These extremes like floods and droughts more frequent and of interactions are erosion and sedimentation, land-water higher magnitude. As a result of their limited resources and interactions, and management of evapotranspiration. stream-flow regulation capacity, developing countries will be disproportionately affected by the projected increased Erosion and Sedimentation. This issue was selected variability of precipitation and frequency of extreme events. for discussion during the workshop because it is among the major water and soil degradation problems in many Most of the existing hydraulic infrastructure was designed large river basins throughout the world. As more and using parameters and analytical tools established some more terrestrial materials are delivered to freshwater and decades ago, under different climatic and geographical coastal ecosystems, the interactions and conditions of conditions from those prevailing today or expected to upstream erosion and downstream sedimentation acquire prevail in the future. Hydraulic infrastructure design renewed importance. The Bank's 2004 Water Resources assumes stationary hydrological processes. However, if Sector Strategy called for renewed attention to hydraulic climate change follows projected scenarios, the stationary infrastructure built to alleviate poverty because of its assumption might no longer be valid. More erratic weather complementarities with other environmental, institutional is projected, including extreme variability of precipitation. and social measures within an IWRM approach. Proposals Experts in the field are still debating whether climate change for the construction of reservoirs and dams are expected requires different planning criteria for dealing with climate to increase as climate change increases water scarcity variability, or whether climate change simply represents one and variability. The economic life of the infrastructure will more factor to consider in the usual analysis. be affected by the sediment load they trap, while at the 4 same time, downstream impacts of changing sedimentation rainfed agriculture accounts for 60 percent of agricultural patterns can be considerable. output. In arid and semiarid regions, water management is crucial for agricultural production. As new methodologies The paper by Robert H. Meade focuses on the for efficient water use have acquired new relevance, ET significance of stationarity for sediment studies in has been shown to be a key component of the hydrological rivers and how this concept has worked against balance. A focus on the management of evapotranspiration monitoring programs by implying that predictive is required to understand water-related issues and improve models could substitute for actual measurements. water management. The concept of ET management requires The author uses examples from the Colorado, innovative tools such as remote sensing, which many believe is Amazon and other major rivers to discuss the the only tool currently available to monitor ET over large areas. difference between intrinsic non-stationarity and non-stationarity of measurement and purpose. Peter Droogers discusses the importance of focusing on evapotranspiration as the Land-water Interactions. These were specifically dominant water consumer. His paper discusses chosen to highlight their significance and the need for methodologies to support policy makers and water paying increased attention to integrated approaches. These administrators to manage evapotranspiration. interactions are particularly important because there are few It also provides practical examples from China instances of Bank engagement in the coastal zones. The and Egypt. Droogers advocates the inclusion of volume and quality of water at any given point of a stream, a combination of remote sensing and simulation lake or aquifer is a function of the precipitation regime as models in policy support tools and introduces the well as the geophysical characteristics of the catchments concept of scenario-based modeling. and the land-water interactions occurring in them, from the water divide to the ocean. These conditions are also changing continuously. Human intervention in watersheds Associated Changes in Climate and and river basins is now so widespread that human activity Land Use must be considered part of the hydrological cycle and taken into account in a comprehensive manner as water moves This topic was chosen to discuss the links between climate from its source to its sink in the coastal areas. change and land use and how the potential effects of these changes interact and reflect on the water resources of a The paper by Jeffrey Richey presents a region and its ecosystem, as well as how to quantify these "systems-level" overview of the key processes changes. Rapid changes in land use and access to water and transitions, from land to rivers to oceans and resources have been posited to be the greatest challenges their marine fate. It summarizes the types of issues facing many regions of the world. Global climate change confronted in coastal-focused topic areas. The models also predict an increase in total rainfall and rainfall author also discusses World Bank environment and variability in some regions and decreases in others. Local water projects as a means of identifying existing climatic observations can provide reasonable predictions projects and their requirements. He also presents of the impacts of changes and feedbacks on the natural a case study of the Mekong River basin, as an system. Information derived from these models and from example of a full suite of land-to-ocean issues. hydrologic and meteorological time-series will always be Finally, Richey's paper advances the concept of a needed, but it is not sufficient. Watershed geomorphology, "virtual river/coastal basin," driven by a "dynamic vegetation cover, soil and land use are of increasing information framework," as a means to provide a importance in water resources management because they convergence of cross-sector information. are dynamically linked to climate. Management of Evapotranspiration (ET). The bulk of In his paper, Ignacio Rodríguez-Iturbe posits the world's agricultural production (82 percent) continues to that accounting only for changes in mean be rainfed (as opposed to irrigated). In developing countries, responses to climatic variability is not sufficient 5 for a realistic evaluation of the impact of climate types of ecohydrological consequences especially change on ecosystems. Changes in the dynamics in aspects related to soil erosion. His paper also of less frequent and stronger rainfall events will focuses on those challenges where ecohydrology have larger consequences for the assimilation will contribute decisively to the understanding process and survival of vegetation. An increase in needed to ameliorate and manage these effects. the intensity of rainfall events also leads to other 6 Key Messages Workshop presentations and discussions yielded · There are other issues related to efficiency that go several key messages about integrated water resources beyond water charges. For example, improving irrigation management, climate variability and change, hydrologic efficiency does not necessarily mean that water is going interactions and associated changes in climate and land to be saved in the basin because when water is diverted use, which are discussed below. and transported, a portion of it recharges groundwater. Also, if efficiency is improved, the return flows decrease and the amount that is consumed increases. Integrated Water Resources · More attention needs to be paid to flood prevention in Management IWRM. Flood management is largely an afterthought in the countries with which the Bank works. Mostly, atten- · IWRM requires facing many challenges. There is a need tion has been given to mitigation measures once a flood to integrate across different spatial scales, all the way has occurred. But that is much more expensive than from river basin to the national level. In addition, there is preventive approaches. a need to integrate across sectors and in terms of the · Integrated water resources management links top-down environment and other cross-sector activities, which are and bottom-up approaches. Top-down approaches (that much more difficult to handle. It is also necessary to in- is, laws and policies, regulations, standards) are applied tegrate across institutional functions and responsibilities, mostly in the developing world. However, bottom-up which is probably the most challenging of all. approaches require increased attention. Work actually · There is also a need to balance the old challenges of starts in the field with the water users and uses, regulat- growth, poverty, and other development issues with new ing discharges and managing water at the very local level. emerging challenges posed by global climate change. · To improve the management of water resources in any This rebalancing must take into account global economic given river basin, there is a need to slowly move away links as well as increasing financing needs and fixed (if from the project-by-project approach and actually apply not decreasing) financial resources. and achieve the integrated water sources manage- · The issue of monitoring water use cannot be overempha- ment concepts. This implies the improved planning and sized. The balance between the supply of and demand for management of water use, improved construction and water requires the quantification of water availability and operation of infrastructure, improved reliability of extreme water use: How much is actually being extracted from a event forecasting, and so on. given source and how much is the intake for each and all · In order to move towards using modern systems and ap- of the different uses in a given watershed or river basin. proaches, decision support tools and systems should be However, developing countries usually lack monitoring or employed that rely on new and innovative ways of doing measurement capabilities. They also lack a system of wa- things (something that was not possible just a few years ter rights and an inventory and characterization of waste- ago). This will require a special and sustained effort water discharges. It is not possible to manage what is not within the Bank environment and is perhaps where the measured; this is a most basic part of water resources major challenge lies. management that developing countries need to address. · The demand management approach embedded in the concept of IWRM implies that water use has to be mea- Climate Variability and Change sured to price it on a volumetric basis in order to ensure that it is being charged and that users are paying for · Hydrologists have not spoken out about climate change that use. Otherwise, water charges are not going to be a as much as professionals in other fields. This may drive factor in water use efficiency and there is not going to be the Bank in some potentially unexpected ways, particu- any change in how much water is wasted. larly if not enough attention is given to the effects of cli- 7 mate change on water resources. In this realm, the focus ate time and effort should be devoted during project should be on precipitation and stream flow, rather than preparation for data collection and analysis. At least a temperature. Shifts in precipitation will result in shorter year of data should be collected regardless of the proj- but perhaps more intense periods of rainfall, an issue ect or whether it relates to hydrology, sediment or other that needs to be looked at in more detail. relevant topics. Also, sufficient time should be available · The study of climate variability and change is rife with un- at this stage of project preparation for analysts to look at certainty. Thus, the use of uncertainty analysis should be the trends provided by the information gathered. encouraged to help evaluate policy and project options. · More thought should be given to the selection and In addition, models can also be a useful tool. A modeling relevance for analysis of extreme events. The impacts system that meets user needs from the basin down to of highly improbable but possible events should be duly the sub-watershed or micro watershed level is required. considered in order to identify the latest techniques to However, the availability and use of models should not deal with them. mean that measurement and monitoring of key variables · The most important advances in the area of evapotrans- should cease. Quite to the contrary, the uncertainties piration are not the reexamination of the hydrologic cycle surrounding non-stationarity require that monitoring be or efforts to educate policy makers so they begin to think continued and accurate records be kept for as long as more like hydrologists. Indeed, the most important tools possible and by any means possible. in evapotranspiration are satellite technologies and other · Models that operate at the basin/sub-basin level are need- remote sensing products that many Bank clients can use. ed to establish a broader context in the area of watershed In terms of continuity and accuracy, the better approach management. These models could help guide watershed might be a judicious mix of on-the-ground information and projects, which tend to operate more at a sub-watershed images from satellites and other remote sensing platforms. and micro-watershed level. Nevertheless, no single model · There has been a tendency to focus on environmental can do it all. A common language is needed for seamless flows to ensure that enough water is reaching coastal communication among different models so that all contrib- wetlands and mangrove forests. However, the issue of ute to obtaining a better idea of the bigger picture. the quality and chemical constituents of that water need · Models are useful even when their results are not entirely to be further examined. Additional areas that require at- correct because they facilitate communication. Yet, to tention include watershed characteristics and sediment take full advantage of this feature, transparency in the yields, as well as how those sediment yields may be application of models needs to be improved. Explana- affected by changes in basin development or land use. tions about how to use the models are particularly · Projects are often prepared by government agencies important, as are means to communicate with users of without the benefit of all the expertise that may exist all types, including those who may be less sophisticated within their own jurisdictions. Agencies should be en- in the area of modeling. Models should be easier for couraged to avail themselves of this expertise. Specific non-experts to understand. A step in this direction is to efforts should be made to ensure that stakeholders are improve the ability to visualize results in graphical form. engaged with the project at an early stage. Although · Rising awareness of the effects of climate variability and many developing countries may have strong legal or change on water resources is not only important for practi- policy frameworks, they often lack the data, capacity, and tioners or policy makers. The media and society as a whole institutions to move projects forward. This is also an area also need to become more informed about this topic. that requires strengthening. Finally, project managers should consider bringing in experts in other disciplines. Hydrologic Interactions Associated Changes in Climate and · Hydrologic interactions are complex and often difficult Land Use to take into account in a comprehensive way. In many of the places where the Bank provides assistance, data · The availability of water at a given site in a watershed or and data collection remain challenging issues. Appropri- river basin depends on the amount of precipitation and 8 of the response of the watershed or basin. However, · Models and model results are needed to improve un- stream flows may sometimes change because conditions derstanding and raise awareness about the interactions other than precipitation, such as land use, have changed. among water, land, vegetation, and evapotranspiration, Land use is an important factor that needs to be taken particularly with respect to climate change. into consideration. · The temporal distribution of hydrological variables is as · Among other things, land use affects vegetation cover and, important as their spatial distribution. The main factor to as a result, the dynamic relationship among climate, land consider in the practical application of climate change use and vegetation. It is important to address forest health conditions to water resources management and use, to and changes in forest cover, as well as changes in crop floods, to ecosystem response and so on, is not neces- selection by farmers. However, just as important is the sarily how the mean value of each variable has changed management of water resources and future water needs because the mean may have remained the same. It may be focusing on how changes in precipitation and stream more important to analyze the rate of change during differ- flow affect the water balance, and also on the interrelated ent seasons throughout the year, the number of rainy days changes in vegetation cover, soil moisture, and so on. and the amount of precipitation during each rainy day. 9 Papers This section contains the papers that were submitted and presented at the November 2008 workshop, as prepared by the authors. Power point presentations and the presenters' biographical information are included in an accompanying CD. All views expressed are those of the authors. 1. Application of Integrated Approaches in Water Resources Management Beyond the Conceptual Phase Torkil Jønch-Clausen Water Policy Adviser, DHI Group s Ab tract Introduction The title of this workshop emphasizes "beyond the The world is facing serious water challenges, driven mainly conceptual phase." At this point in time that is very by population and economic growth. Water is essential in appropriate. Integrated water resources management achieving the Millennium Development Goals to reduce (IWRM) has been a key concept driving the rhetoric, poverty, hunger, diseases, and environmental degradation. policies and strategies of countries, development banks, Increased investments in water are fundamental to donor agencies and many international organizations for continued economic growth, and job creation, in both more than a decade. However, because it appears to be rich and poor countries. In addition, many countries, but difficult to show results in the field, questions have been particularly developing countries with poor capacity and raised about the concept itself. infrastructure, are vulnerable to the impact of climate change. Addressing their needs requires investment in Part 1 of this paper briefly describes the current (November appropriate adaptation measures. 2008) status of IWRM: Why was this concept developed and taken to heart by so many; what (in brief) the author In an increasing number of countries water scarcity and understands by IWRM, and how IWRM principles are deteriorating water quality have been or will soon become actually being applied throughout the world. Part 2 relies on critical factors limiting national economic development, recent work by the author in Orissa, India, to illustrate how a expansion of food production and/or provision of basic "roadmap for IWRM" can help a state (or country) define the health and hygiene services to the population. Extreme relevant small steps necessary to improve water resources events such as floods and droughts add to this challenge. development and management. In Part 3, the author The current concerns about adaptation to climate change provides some brief personal reflections on how IWRM has call for integrated approaches in all countries. evolved in the World Bank. Improved water governance can be achieved through This paper builds on recent work by the author in integrated water resources management. At the national collaboration with DHI colleagues in Denmark,1 and the level, IWRM provides a basis for balancing the different Asian Development Bank's Technical Assistance to the demands on a country's water resources. Investments State of Orissa.2 This support is gratefully acknowledged. in water infrastructure, water allocation decisions, and 1. A paper entitled "IWRM in Action" was drafted in 2008 by Jan Hassing, Niels Ipsen, Torkil Jønch Clausen and Henrik Larsen, of DHI Water Environment Health and the UNEP-DHI Centre for Water and Environment, in Hørsholm, Denmark, as a special contribution to the World Water Development Report. Passages from this draft paper are quoted. 2. Orissa Integrated Irrigated Agriculture and Water Management Investment Program (OIIAWMIP), Technical Assistance for Integrated Water Resources Management (IWRM) in Orissa, supported by the Asian Development Bank and carried out in cooperation with Prof. B. Das. 13 water management actions and policies have an impact increasing welfare for large parts of the population, have on a country's achievements in multiple ways. IWRM is implied changes towards more water consuming diets that an approach that can capitalize on the opportunities for add to current agricultural water requirements. In other synergies and help reconcile difficult trade-offs in the words, the current energy and food crises are intricately achievement of these goals. Hence, a lot of the "integration" related to water. Moreover, other sectors are placing in IWRM takes place at the basin scale, whether at the local increasing demands on water resources. This includes catchment or aquifer, or at the multi-state or multi-country the growing industrial and mining sectors of developing river basin. countries, as well as tourism and navigation. This added pressure is being placed on the already stressed and Like any other reform, IWRM is a process that could take vulnerable water resources, raising the question of several decades. In France, the process was started whether the ecosystems on which biodiversity and the with the establishment of basin agencies in 1968. Other livelihoods of millions of poor people depend can be important milestones were a revised water law in 1992, sustained. Protecting the environment calls for massive which was amended to comply with the European Union's investments in wastewater treatment and pollution Water Framework Directive in 2003. In Spain, maturing of reduction, along with maintenance of environmental flows the process has lasted close to 80 years. Other countries, for downstream ecosystems. such as India, have started to respond to these kinds of challenges in a number of ways. A National Water Policy was adopted in India in 2002 (updating the first version of Water Resources and the Millennium 1987) that is strongly inspired by IWRM. In Orissa, India, Development Goals the process of reforming and developing the water sector started in 2008. This case illustrates the realities of trying to The main challenges for developing countries are "introduce IWRM on the ground." addressing such basic societal issues as poverty, hunger, education, gender equality, health, and environmental sustainability, as expressed in the Millennium Development Part 1 ­ Integrated Approaches in Water Goals (MDGs). Adequate water availability and quality, and Resources Management: Theory and thus prudent water resources management, are important Practice contributions to achieving these goals. Water Resources and Key Global Water is basic to food production, and hence clearly a Development Issues factor in reducing poverty and hunger. The productivity of irrigated agriculture is particularly dependent on rational In addition to satisfying basic human needs for domestic and wise water resources management. Moreover, activities water supplies and basic sanitation (particularly in surrounding agricultural production help create jobs developing countries), increasing and acute water for the poor. Water related diseases (such as diarrheal challenges in the world relate to energy, food and diseases, malaria, bilharzias, and cholera) are among the the environment. Increased investments in water most common causes of death in developing countries, infrastructure and management are required to sustain and the poorest segments of the population often bear this development, especially in the face of considerable the heaviest burden, not least the women who carry the increases in electricity consumption3 for both domestic daily responsibility for the health of their families. Natural and industrial needs. A new focus on the production of resources and ecosystems face increasing degradation as biofuels, and the significant acceleration in hydropower a result of unsustainable exploitation, often for short-term production are also part of the solution. In an effort to gains. Degraded ecosystems can no longer retain their meet the main food challenge of reducing hunger among productivity and provide essential goods and services to the poorest people of the world, economic growth and sustain biodiversity and livelihoods. 3. Annual growth in Asia has been between 5 and 8 percent. 14 Countries Experience Serious Water IWRM is a comprehensive approach to the development Governance Issues and management of water, addressing its management both as a resource and, in establishing the framework The recognition of the need to redress the effects of weak for the provision of water services, as a political process water governance structures has convinced many countries that involves conflicts of interest that must be mediated. that a new water management framework is needed. Other Effective water governance is crucial for the implementation common critical issues are listed below. of IWRM. · Awareness of water issues, as well as the political prior- The concept of IWRM was already recognized in Agenda ity given to them, are limited. 21 (the UN Conference on Environment and Development, · Institutions are often rooted in a centralized culture with which took place in Rio in 1992) and is to a large extent supply-driven management and fragmented and sub- based on the four Dublin Principles developed earlier that sector approaches to water management. Few water year.4 Inspired by the Dublin Principles, the World Bank managers view water holistically. developed its Water Resources Management Policy in · Local governments lack the capacity to manage the differ- 1993. Ten year later, in 2002, the Plan of Implementation ent demands and pressures placed on water resources. adopted at the World Summit on Sustainable Development · Inappropriate pricing structures, and hence limited cost in Johannesburg (WSSD) called for countries to "develop recovery, result in the inefficient operation and mainte- Integrated Water Resources Management and Water nance of water systems, as well as in the misallocation Efficiency Plans by 2005." These plans were to be and loss of water. milestones in cyclic and long-term national water strategy · Investments in the water sector are low and do not get processes, and progress in their development and sufficient attention in national budgeting discussions. implementation has been measured regularly at the World · Information and data to support sound water manage- Water Forums (the most recent for this paper took place in ment are generally lacking. Mexico in 2006) and by the Commission for Sustainable · Economic, social, and environmental criteria for the ap- Development (CSD, most recently in 2008). proval of policies, plans, and projects are most often few and inadequate. IWRM is not a scientific theory, which needs to be proved or disproved by scholars, but a simple framework to understand water in its larger economic, political and Towards Integrated Water Resources societal contexts. IWRM has proven to be a flexible Management approach to water management that is adaptable to diverse local and national contexts. This requires policy makers to Improved water governance can be achieved through the make judgments about which set of suggestions and reform integrated management of water resources. The Global measures, management tools or institutional arrangements Water Partnership (2000) defines it as: are most appropriate given the particular cultural, social, political, economic and environmental circumstances they A process that promotes the coordinated are facing. development and management of water, land, and related resources in order to maximize the resultant One of the great strengths of IWRM and its principles economic and social welfare in an equitable and concepts is that it has given the water community a manner without compromising the sustainability of common language, which is applicable over a wide range of vital ecosystems. levels from the local to the national and regional. This allows 4. The UN Conference on Water and Development, which was held in Dublin in January 1992, produced the four Dublin Principles: (1) the holistic principle; (2) the participatory principle; (3) the gender principle; (4) and the economic principle. These provided an important mindset for water resources development and management. The World Bank's 1993 Policy Paper redefined these principles to three: the ecological, the institutional and the instrument principles. 15 Figure 1. The Three Pillars of Integrated an exchange of knowledge and lessons learned across Water Resources Management borders, across regions and at the local level. It also makes it possible for decision makers and managers to agree and IWRM Components monitor policies and targets for improving the management of water resources. Economic Environmental Principles Equity Efficiency Sustainability Management Enabling Institutional IWRM Processes Focus on the Critical Instruments Environment Framework Water Resources Issues of Any Country · Assessment · Policies · Central ­ Structure · Information · Legislation Local River · Allocation · Basin Public The role of IWRM and the shape it takes will vary depending instruments ­ Private on each particular country's stage of development. The implementation of IWRM in developing countries, countries in transition and developed countries will differ Balance "water for livelihood" and widely, as will the specific benefits that each will derive. "water as a resource" For developing countries, water resources management may be seen as a factor in addressing poverty, hunger, health problems, and environmental degradation, including the particular challenge of providing better livelihoods for concepts and useful approaches through specific tools women. Countries in transition may see IWRM as a rational ("good practices"), as well as relevant references and case approach to improving management of the resource to studies of IWRM experience. promote and sustain continued economic development. Developed countries may find valuable inspiration in IWRM processes and may choose to design their own variety, as in Roles of the Actors the case of the EU Water Framework Directive. Governments play a key role in the implementation of IWRM For all countries the current concerns about adaptation to as regulators and controllers in the water sector with its climate change call for integrated approaches.5 associated infrastructure. Private actors may be involved in the provision of water services. Governments need to promote improvements in the public sector, regulate private The Three Pillars of IWRM sector involvement, and make decisions about market mechanisms. Governments working with civil society must Implementing an IWRM process is basically a question raise awareness of the importance of improved water of getting the "three pillars" right. That is, (1) moving resources management among policy makers and the towards an enabling environment of appropriate policies, general public. Dialogues will take place among the many strategies, and legislation for the sustainable development stakeholders involved, including government, civil society and management of water resources; (2) putting in place and the private sector. Good water governance requires the institutional framework through which to implement the the involvement of all relevant national (and if appropriate policies, strategies and legislation; and (3) setting up the also regional/trans-boundary) stakeholders in the dialogue management instruments required by these institutions during the development and implementation of an IWRM to do their job. The three pillars are illustrated in Figure 1 framework that acknowledges the needs and rights of all below. The Global Water Partnership (GWP) has developed stakeholders, including poor and vulnerable populations a tool box to expand upon this framework and illustrate who depend on water for their livelihoods. 5. Global Water Partnership notes that "if mitigation is about energy, adaptation is about water," and that adaptation to climate change calls for IWRM approaches (www.gwpforum.org). The IPCC's 4th Assessment Report from 2007 (Working Group 2) in its chapter on freshwater states that "it can be expected that the paradigm of Integrated Water Resources Management will be increasingly followed around the world...which will move water, as a resource and a habitat, into the centre of policy making. This is likely to decrease the vulnerability of freshwater systems to climate change." 16 Figure 2. IWRM and Its Relationship to Cross-sector and Multistakeholder Subsectors Integration A critically important element of IWRM is the integration Cross-sectoral integration of various sectoral views and interests in the development and implementation of the IWRM framework (see Figure 2). · Enabling Water Water Water Water Integration should take place within: environment for for for for · Institutions people food nature other · Management uses · The natural system, with its critical importance for re- tools source availability and quality, and · The human system, which fundamentally determines Source: Global Water Partnership (2000). the resource use, waste production and pollution of Note: In its 2004 Water Resources Sector Strategy, the World Bank has ad- opted this figure, but added additional sectors and redefined the three pillars the resource, and which must also set the development of IWRM to four: Institutional Framework, Development and Management of priorities and control the associated infrastructure. Infrastructure, Management Instruments, and Political Economy of Water. For instance, within the natural system it concerns the integration of land and water management, as well as ensuring dialogue and the sharing of interests among the interests related to surface and groundwater, and upstream multiple stakeholders involved in and affected by water and downstream water, recognizing the full hydrologic cycle resources development and management. with respect to both quantity and quality. Recognizing that most of the important decisions affecting The Water Basin Is the Basic Planning and water resources are actually made in other sectors Management Unit (agriculture, energy, and so on), integration within the human system relates, in particular, to the cross-sector Water follows its own boundaries: the river or lake basin, or integration of policies and strategies among all relevant the groundwater aquifer. Analyses and discussions of water stakeholders in the decision-making processes. It is about allocation between the needs of users and ecosystems mainstreaming water in the national economy, rather than make sense only when addressed at the basin level. looking at water as a separate "sector." Formal mechanisms Hence, a lot of the "integration" in IWRM takes place at and means of cooperation and information exchange the basin (whether at the local catchment or aquifer) or at need to be established to secure the coordination of the multi-state or multi-country river basin. Many countries water management efforts across water related sectors understood this and organized their water management at and throughout entire water basins. These coordination the basin level many years ago.6 Other countries are only mechanisms should be created at the highest political level now setting up river and lake basin management structures. and put in place in all relevant levels of water management. The EU Water Framework Directive has made basin level management law for an entire region. It is equally important that IWRM should build on and be consistent with existing government policies and national The daily competition for water happens at the local level, or sectoral development plans and/or budgets and be and needs to be addressed in the context of the river consistent with these. The links between IWRM and national basin. Competition for water changes over time and as a and sectoral plans and processes must be clearly understood result of development, and the agricultural sector is often and taken into account during the planning stage. the prime stage where this competition takes place. In many countries of the developing world there is a strong Water is everybody's business, and although governments tradition of "agriculture takes all" (often at negligible to zero have a key role to play, good IWRM is a question of cost). Economic growth and the emerging prominence 6. The Spanish river basin management structure recently celebrated its 75th anniversary. The first Mekong River basin structures were established in the 1950s. 17 of other economic sectors are challenging that tradition. shrinkage of the Aral Sea (both in volume and surface area) Typical conflicts requiring dialogue, trade-offs, and good and a worsening of its ecological status at the downstream management to balance water uses in a river basin are: end of the basins. The importance of cooperation across international boundaries is also exemplified by the Mekong · Conflicts between irrigated agriculture and a growing River Commission, through which the four lower riparian industrial sector. countries (Laos, Cambodia, Vietnam, and Thailand) attempt to · Conflicts between irrigated agriculture and burgeoning coordinate the development of the basin. An IWRM strategy cities. is guiding the preparation of a basin development plan. · Conflicts among hydropower, irrigation, and flood control interests in the planning and operation of dams and reservoirs. IWRM Is a Never-ending Process · Ensuring that sufficient water is available for environmental needs, including for vulnerable ecosystems and the people As shown in Figure 3, IWRM is a cyclical process; that is, whose livelihood depends on them (environmental flows). circumstances and priorities change over time and require continued adjustments and development. The cycle starts Upstream/downstream issues are particularly important in with planning processes and continues into implementation IWRM at the basin level. Drainage from agricultural land of the frameworks and action plans, and monitoring of has increased salinity almost tenfold in the Colorado River/ progress. IWRM plans are just one step or milestone in the Rio Grande in recent years, resulting in production losses process of improved water resources management. in the billions of dollars (not accounting for the decline in ecosystem functions). Another example is the Syr Darya and Amu Darya river basins in Central Asia, where large-scale Changing the Way of Thinking cotton growing, farming, and domestic use in downstream areas, and power generation in upstream areas compete IWRM processes are now established or being established for access to water. Poor management has led to the in major parts of the world. At the 4th World Water Forum Figure 3. The Integrated Water Resources Management Cycle Establish Status Build Commitment to Reform · Water resource issues · Political will Monitor & Evaluate · Progress towards · Awareness Progress IWRM framework · Multistakeholder · Recent international dialogue · Indicators of progress developments toward IWRM and water infrastructure development Analyse Gaps framework · WR management functions required · Management potentials and Implement Frameworks constraints · IWRM framework · Framework for water infrastructure Build Commitment Prepare Strategy and development to Actions Action Plan · Build capacity · Political adoption · Enabling environment · Stakeholder · Institutional roles acceptance · Management · Raise funds instruments · Links to national policies 18 in Mexico (2006) it was reported that out of 95 countries water demand management were among the areas with the surveyed, 74 percent either had IWRM plans/strategies lowest scores. in place or had initiated a process for formulating them. Many of these are not "just water plans" The IWRM The results of the survey showed an expected pattern. plans for Malawi and Zambia, for example, flow directly Experience tells us that countries focus first on creating from the national economic development plans, and the enabling environment for reforming the water sector, were prepared jointly with the ministries responsible for including developing adequate policies, laws, and economic planning. The table in Annex 1 lists 42 countries regulations. New policies and laws pave the way for new that by 2008 had adopted integrated water resources institutions, establish institutional roles, and help develop management and explicitly used the term in their official the capacities for carrying out these roles. Once institutions documents. While the existence of these documents alone are in place, new management instruments and capabilities is not proof that IWRM is working in these countries, they can be developed to implement IWRM. are essential for helping to create and support an enabling environment for water reform. Another important indicator of progress towards IWRM is the number of countries whose water legislation includes The UNEP-DHI Centre on Water and Environment IWRM principles (see Figure 5). (previously known as the UCC-Water) carried out a survey in 2007 (with the support of Danida and UNEP) The figure shows that stakeholder participation is the to gain a more detailed understanding of the extent to highest scoring indicator, followed by user pays, and river which government institutions have adopted IWRM. basin management. These results (as well as other survey The survey covered 58 countries in Africa, Central Asia, results) indicate that many countries acknowledge the South East Asia, Latin America, and the Caribbean. The usefulness of IWRM principles and associated management respondents were typically senior government officials. approaches, and use them as guidance to advance water One of the indicators used was the institutional capacity resources management. However, they also show that for maintaining various functions basic to IWRM, including capacity is not always commensurate with the priorities set policy formulation, water allocation, and water demand out in legislation. The high priority given in legislation to the management. The results of the survey are shown in user pays principle is not reflected in the capacity to ensure Figure 4. It appears that the best developed capacities (that that water resources management costs are recovered. is, highest scores) are in policy formulation, drafting of laws, and cooperation on shared watercourses. Cost recovery of Since 2002 the UN organizations dealing with water (UN- expenses for water resources management and capacity for Water) have coordinated and monitored global progress Figure 4. Institutional Capacity for Carrying Out Various IWRM-inspired Functions Policy formulation Cost recovery of WRM 2 Drafting laws/regulations Facilitating water demand management 1.5 Environ. assessments Aquatic ecosystems 1 Coop. intern. shared watercourses 0.5 Conflict mediation Information and data 0 Socio-economic assessments Monitoring water availability Pollution loads Water res. assessments Ambient WQ Water allocation Planning resource use etc. Water use The ratings that the respondents were asked to give were: 0 = function not established, 1 = function has many gaps in quality and coverage, 2 = function has some gaps in quality and coverage, 3 = function operates at realistic goal levels. 19 Figure 5. Inclusion of IWRM Principles in Respondent's Water Law Public hearings by law? 40 Private sector involvement? 35 Participation of stakeholders in water management? 30 25 20 Water use efficiency? 15 River basin management? 10 5 0 Separation of water management and service provision? Lowest appropriate level? Role of women in water management in law? Finance contribution by users to water management? User pays principle? Polluter pays principle? Yes The scale shows number of water laws where the item has been included. towards IWRM and Water Efficiency Plans. At the CSD-16 floods causing unnecessarily high losses of lives and in 2008 UN-Water reported on the latest status of IWRM property because of poor preparedness; or simply greater planning work.7 social pressure caused by poor management of increasingly scarce water. Climate change may add to the list. Implementation Takes Time and Requires In France, the process was started with the establishment Trade-offs of basin agencies in 1968. The country's water law was revised in 1992 and amended again in 2003 to bring it into Implementing the IWRM concept has many challenges. line with the EU's Water Framework Directive. In Spain, it Some of these challenges revolve around integration and has taken almost 80 years for the process to mature. As the degree to which it can be achieved given that water expected, weak institutional capacity in developing countries resources are used by many sectors and many institutions has yielded slower progress towards IWRM. Other factors are engaged in water management. Thus, the first step is also limit the speed with which developing countries are coordination, but many real world factors, such as need, able to implement integrated water resources management. funding, resources, human capacity, institutional barriers, There is a high degree of informality in the water sector of and many others, establish the operational limits and many developing countries, particularly in rural areas, where determine how far integration can be taken. people rely on self-supply and local water institutions are based on customary laws and practices. Moreover, the Like any other reform, IWRM is a process that could take ability of national governments to enforce regulations is several decades before all its most important principles are limited and laws, prices, and policies often fail to function implemented. While change may happen gradually, a trigger as intended. In contrast, the water sector in developed is often required. Examples of triggering events include countries is more formalized and a large proportion is under a severe drought in Zimbabwe resulting in a 15 percent direct regulatory supervision. The chance of success for decline in GDP in one year; a major chemical spill on the IWRM at the national level goes hand in hand with the Rhine River, which led to trans-boundary cooperation; development of national governance structures. At the anoxic conditions resulting in dead lobsters in Danish local level, IWRM principles will still guide water resources coastal waters in 1987, which led to the development of a management, but initiatives and actions are often taken by succession of major national Aquatic Action Plans; severe the communities on their own accord. 7. This status reporting is based on surveys made by UN-DESA (2007), UCC-Water (2007), GWP (2006) and AfDB (2007). 20 Implementation of IWRM requires tough trade-offs among Box 1. Management Issues at sometimes conflicting objectives. Changing a water law Increasing Levels of Water Scarcity: typically involves changing an indirect power balance The Turn of the Water Screw in relation to water among different interest groups.8 Changing water allocation in order to achieve a better The crucial scarcity in dealing with water may not overall societal use of the resource will typically yield both be the scarcity of the natural resource--water--but winners and losers; that is, users who get more and users the scarcity of social resources needed to adapt to who get less water. In some countries (as, for example, the water scarcity. The significance of this is made clear large "irrigation countries" of Asia), large water users are by considering how water managers can deal with influential but often use water inefficiently. In such cases, increasing scarcity over time. implementation of IWRM may require delicate, time- consuming. and difficult negotiations and trade-offs, as well (1) At the first turn of the water screw, the remedy as a shift in the mindset of farmers. is to get more water. This goal is predominantly accomplished by water storage and transfer in time and space. Top-down and Bottom-up Processes Go Hand-in-hand (2) At the second turn the effort is redirected towards efficiency measures, predominantly end use efficiency. It has been shown that central influence on the management The goal is to get more benefit per drop. of water resources tends to be dictated by the degree of scarcity (either because of lack of water as such or because (3) The last turn of the water screw is reallocation the water available is of poor quality). It may not seem to of water rights. This requires profound changes in matter how one deals with a resource that is plentiful and of national policies, since achieving allocation efficiency good quality vis-à-vis its demand and environmental needs. could mean the withdrawal of water rights of irrigation In economic terms one might say that water has no or little schemes that generate a low value per unit of water. value (little or zero opportunity cost). However, as scarcity The food needed by growing populations will then increases so does competition and, as a result, the value have to be imported and paid for by the industry and of water rises. Governments need to regulate water use to services sectors. This will require large-scale social ensure that it is not wasted on low value uses, but it is used restructuring and entails risks of tension and conflicts, instead in the ways that yield the greatest social benefits. within countries and between sectors and population Until that happens, local initiatives and ownership are groups with different stakes in the new socioeconomic essential in the majority of cases where water demands are environment. relatively small. Large-scale use will continue to be under central regulation. Based on FAO (2000) and Yevjevich (1995) as cited by Snellen and Schrevel (2004). As alluded to in the box above, bottom-up water resources management processes can thrive in situations where there is abundant water relative to demands. Once that is no longer the case, the enabling environment and institutional changes called for in IWRM will have to be necessitated reallocation measures. The response was a put into practice. The clear specification of institutional "compulsory licensing" process. Existing water rights were roles will describe how authority and responsibility are to revoked and previous owners need to reapply for their be shared among local levels, basin levels, and the central allocation. In addition, water licenses were made time- government. In 2000, South Africa reached a degree of bound and land ownership and water licenses were no water scarcity (and inequality in access to water) that longer connected. 8. Water laws are not developed overnight. Vietnam's current water law took 22 drafts over 8 years and now needs to be revised. 21 Conclusion to Part 1 Part 2: The Orissa Case: Implementing IWRM in Small Steps The world has come a long way in recognizing the need for a new approach to water resources management since the Recent work by the author in Orissa, India (2008), Dublin Principles and the Rio Summit in 1992. This new illustrates the realities of trying to introduce IWRM "on approach was articulated in the 1990s as "IWRM," and the ground." Orissa is in the process of reforming and adopted ten years later by the World Summit (WSSD) in developing its water sector with significant assistance Johannesburg in a call for all countries to develop "IWRM from the World Bank and the Asian Development Bank and water efficiency plans." This has been a long process (ADB), among other institutions. Both banks are providing of building popular awareness and political will to start irrigation development support in the context of river basin developing and managing water as a resource that is management, each focusing on a pilot river basin (the World fundamental to economic growth, poverty reduction, and Bank in the Mahanadi River basin and ADB in Baitarani social equity, as well as environmental sustainability. The River basin). ADB support has an additional component evidence is clear. Countries, particularly the developing of introducing IWRM in the state. Working with a national countries, are adopting IWRM principles in their policies, specialist, the author has undertaken a preliminary ADB strategies, and legal framework for water resources technical assistance project to propose a "roadmap for management, while trying to change water management IWRM" in Orissa. Follow-up technical assistance was practices accordingly. The statistics and examples shown being formulated in close collaboration between the state above and in the Annex attest to that. government, the World Bank, and ADB to continue this activity. Unfortunately, however, the drivers of this development at the international level are impatient, and in spite of rhetoric A quick overview of the why, what, and how of the proposed about "small steps," "picking the low hanging fruits first," and IWRM roadmap for Orissa is presented below. so on, are demanding here-and-now evidence of the direct link between the IWRM concept and visible impacts on the ground.9 As a consequence, some major international Water Challenges Facing Orissa actors now shy away from referring to IWRM.10 This is unfortunate. After years of building up an understanding Orissa must adapt to increasing water challenges. about this concept, and with most developing countries Although the state is, on average, well endowed with water explicitly referring to IWRM in their policies, strategies, and resources (3,300 m3/yr/cap, well above the UN "stress laws, we are left to wonder why those who championed limit" of 1,700 m3/yr/cap), the average does not reflect the these sensible principles are now shying away from them strong seasonality of water (almost 80 percent of rainfall (the same question could be asked about the concept occurs in the three monsoon months), nor does it reflect sustainable development). anticipated declines in per capita water availability because of population growth, increased economic development The foregoing, together with the case study presented demands, and increased water consumption in upstream below, attempt to show that the principles of IWRM are as states leading to decreasing inflows to Orissa (upstream valid as ever, not least in the current age of concerns about sources currently accounts for about 30 percent of total climate change. However, additional patience and small water availability). The combination of these factors will lead steps in the right direction are required. Although particular to an estimated reduction in water availability of 30 percent events, or serious water stress and scarcity can speed up by 2050 (to approximately 2,200 m3/yr/cap, a low level in the process, evidence is accumulating that small steps are, a monsoon climate). At the same time, some basins suffer in fact, being taken throughout the world. from poor water quality because of inadequate treatment of 9. In October 2008, the author visited a Vietnamese province facing serious water problems and devoted to improved water management. However, as they expressed it, "we have 1.2 water professionals in the provincial administration per 1 million inhabitants in the province." So much for ambitions about speed! 10. The Fifth World Water Forum held in Istanbul in 2009 had 6 themes and 22 topics, none of which mention IWRM explicitly. In previous WWF in The Hague, Kyoto and Mexico, IWRM was the major theme, and the one attracting most contributions. 22 Figure 6. Maps of Orissa and Baitarani Basin municipal and industrial effluents, leading to environmental and industry, have already raised demand, and mining degradation. In addition, serious water logging attributable activities have increased surface and groundwater pollution. to lack of drainage creates problems in the lower part of most basins.11 The environment as well as the valuable coastal ecosystems in lower Baitarani/Brahmani and Mahanadi have a Climate change may further accelerate water problems in serious stake in these developments and depend on the Orissa. The 4th Assessment report of the International Panel maintenance of critical flows (environmental flows) and on Climate Change (IPCC, 2007) predicts a "projected pollution control. decrease in winter precipitation on the Indian subcontinent," and "intense rain occurring during few days, which implies increased frequency of flooding during monsoons." In the IWRM in India coastal areas these impacts will be compounded by sea level rise. India has responded to these challenges in a number of ways. A National Water Policy was adopted in 2002 This general situation is becoming very competitive as (updating the 1987 version), which draws strongly demand for water from a variety of users and sectors from IWRM principles, and makes explicit reference to increases, particularly in the dry season. While agriculture multi-sectoral, multi-disciplinary and participatory water (mainly irrigation) currently accounts for 93 percent of all management. In line with IWRM approaches, it also calls water use in the state, and domestic and industrial use for water resources development and management to be account for just 4 percent and 3 percent, respectively, planned for a "hydrological unit such as a drainage basin this pattern will change. Increased urbanization will lead to as a whole or for a sub-basin." Several Indian states have increased domestic and commercial water needs, and water responded to the National Water Policy by developing state demand for industrial uses will continue to increase rapidly. water plans, and also by moving towards the establishment Moreover, demand is also increasing in other economic of appropriate river basin organizations (RBOs). One sectors, such as fisheries, energy, and recreation and state, Maharashtra, has even created a quasi-judicial Water tourism. Fast economic growth, particularly in commerce Resources Regulatory Authority to determine, regulate, and 11. This information is drawn from the Orissa State Water Plan of 2004, which in turn was developed on the basis of 11 river basin plans. While figures for the present situation are credible, the projected figures appear to be, at most, indicative. 23 enforce the distribution of water entitlements, along with a representatives from different sectors. A resolution on water tariff system for various categories of water use. this was issued by the Orissa state government (OSG) in February 2007, specifying the functions of the RBOs Water in India is basically a state matter. While the central and the composition of the council and board. government plays an important role in setting overall policy directions, controlling and approving major water projects, With a growing population and economy, as well as and addressing interstate water issues, IWRM needs to be prospects of a general decrease in water resources caused developed at the state level. by upstream development and climate change, Orissa can no longer rely on a traditional fragmented water resources Legislation concerning water remains fragmented, and most management system. Key issues to be addressed through states rely on old irrigation acts and pieces of legislation IWRM include: in a range of other acts. No comprehensive "water acts," per se, exist in India. Although a Groundwater Model Act · Drinking water and domestic use (human and animal has been proposed by the central government, few states consumption). This includes water supply for rural and have adopted it, leaving groundwater largely uncontrolled. urban areas, and basic water use for livestock. The "Modern" legislation to support IWRM approaches needs to sector consumes relatively little water on average, but be developed. rapid urbanization will increase urban water demand significantly (a threefold increase by 2040 is estimated). Orissa has taken the lead in developing an IWRM process. Plans include reducing current water losses of 40 to 50 A multi-sector state water resources board was formed percent in urban distribution systems to a target of 15 in 1993, chaired by the chief secretary and including ten percent in the future. principal secretaries as members; a state water policy · Ecologic requirements. These are currently treated as was first formulated in 1994; and a state water plan was a fixed proportion (roughly estimated at 30 percent of developed in 2004. This plan calls for "a legal framework natural flows) for environmental flows to the ecosys- to ensure that water-related matters are appropriately dealt tems (as per the 2004 State Water Plan), including the with in the context of IWRM." Building on these, a revised two Ramsar sites in Orissa. Knowledge about these Orissa State Water Policy was adopted in March 2007 requirements (including sustaining the livelihoods of that reiterates the importance of IWRM and river basin poor people who depend on these ecosystems) and management by, for example, calling for "macro-level multi- approaches to address them through environmental flow sector river basin plans" to be "ground-truthed through the methods are currently low. river basin organizations." The Orissa Department of Water · Irrigation (and drainage), agriculture and other related Resources (DOWR) is taking the steps listed below to activities, including fisheries. Irrigation is currently gen- better respond to the requirements of IWRM. erally taking place during the monsoon season without water scarcity (apart from some dry spells in "bad · DOWR is creating an IWRM directorate headed by an years"), but the demand for water for irrigation in non- engineer-in-chief, and including the Orissa Water Plan- monsoon seasons (Rabi) still represents the dominant ning Organization headed by a chief engineer. Although water demand compared with other sectors. Improved still in the same department, this may be seen as an management of the irrigation and drainage systems, important first step towards separating water resources water conservation, and improving the efficiency and management from service provision functions. Potentially, productivity of water used for irrigation (and associ- it is also an important first step to create a multi-disciplin- ated drainage) are high priorities and hold potential for ary environment within the DOWR, with professionals accommodating an increase in agricultural production deputed form other line departments to address broader without adversely affecting other use sectors. Lack of water issues. proper drainage systems and overuse of water for ir- · The Department is also setting up and piloting river basin rigation cause serious water logging problems and soil organizations operating through a council made up of salinization in some areas, including the Mahanadi and key stakeholders, and a board that includes government Brahmani systems. 24 · Hydropower. Hydropower presently accounts for 55 per- required to treat their wastewater, but enforcement is cent of the energy production in the state. The desired uncertain. Apart from a few places, groundwater pollu- hydro-thermal mix is 60 percent hydropower to 40 per- tion is not (yet) a problem. cent thermal. In the future, more than 90 percent of the · Groundwater management. While surface water use thermal power may be exported to other states with little is controlled by the state, groundwater use is largely benefit for Orissa. The view of the sector is, therefore, uncontrolled and represents an important source of that power production should not compromise require- "self-supply." Groundwater accounts for the majority of ments for flood control and water use requirements for domestic use, and some 14 percent of irrigation use, but other sectors. only 25 percent of the state's total potential groundwater · Industries, including agroindustries, mining and com- is being used. However, overdrafts cause localized low- merce. The state expects industry in Orissa to grow ering of water tables. The state has considered devel- from the present 19 percent of state GDP to 30 percent oping a groundwater act, but given the limited scale of to 35 percent of state GDP over the next 10 years. overdraft problems, it has decided not to do so yet. No Industrial water use is the single most important "com- mechanisms for managing the conjunctive use of surface peting use" to agriculture in Orissa, and conflicts with water and groundwater exist. the irrigation sector are already occurring. Industries and · Extreme events such as intense rainfall, floods and mines, as well as urban areas, are serious contributors to droughts visit the state regularly and cause human and surface and groundwater pollution. economic losses. While infrastructure such as dams · Navigation and other uses such as recreation and tourism. and levees may mitigate some of the effects of floods, a number of "softer," non-structural management solutions In addition to satisfying the demands of these primary water are called for to increase preparedness and reduce users a number of water management issues call for IWRM losses. approaches, such as: · Watershed management in upstream catchments holds the key to local land and water management with both · Inter-state water sharing. Orissa receives some 30 local and basin benefits. Land use management and percent of its water from upstream states. With increas- practices (including forest management) are key to ing development and water abstraction in these states, controlling run-off and upper basin recharge. Clearly, discussions about water sharing arrangements are this calls for the integration of land, forest, and water becoming increasingly serious. management. · Storage capacity. Orissa is subject to strong seasonal- · Salinity intrusion in coastal river reaches and deltas ity of rainfall, losing the majority of the monsoon flows cause problems that, in many cases, can be addressed to the sea. In spite of this, Orissa has only 45 m3/cap only in a basin context. Adequate regulatory measures in storage capacity, against some 200 m3/cap for India are required to stop the overexploitation of coastal (and 5,000 m3/cap for the United States and Australia). aquifers. Obviously, additional surface water storage capabili- ties will help increase the availability of water during the dry season and reduce water losses. However, before Key Constraints to Developing IWRM in engaging in large-scale dam building for this purpose Orissa other, more cost efficient options should be considered first. These options include improved management of the In many ways, introducing IWRM is paramount to existing surface water dam and canal systems; improv- demanding a paradigm shift in water management, from ing drainage to bring presently waterlogged lands into relying solely on a traditional top-down supply oriented production; building small storage tanks, ponds, and approach (building new infrastructure to meet demand) rainwater harvesting structures in the upper catchments; to combining this with a bottom-up demand management and recharging the groundwater aquifers. approach that builds on extensive user participation. Such · Water quality management. Very little urban wastewa- a fundamental change takes time, and will only come about ter is currently being treated in Orissa. Industries are through small steps in the right direction. 25 As explained below, there are several constraints to the challenges involved in managing water resources and introduction of IWRM in Orissa. providing services in tribal areas · The state lacks the capacity, at all levels, to address · Except for extreme events (floods, droughts), stakehold- IWRM issues. This is the case within the water sector ers lack awareness and information about the long- itself, as well as in other sectors and in line ministries, term water situation and the opportunities to change it. in district administrations, blocks and Gram Panchayats Unless information is made publicly available and easily and Pani Panchayats. accessible, an informed dialogue among stakeholders · Finally, there is a lack of reliable and comprehensive (the general public, government officials, politicians, civil data to support informed decisions. society, and so on) cannot take place. · Irrigation is the dominant water user and farmers are the most important stakeholders. Changing the "Irriga- Priority IWRM Issues: First Step tion Department" to the "Water Department" does not change that perception overnight. It takes time to change Orissa has taken important steps towards introducing the culture and staff composition of the DOWR so that a IWRM at the policy and resolution level, but this is not better balance can be struck in the attention given to all a sufficient condition for action. The challenge now is to water users and issues. translate these intentions into actions. While the DOWR · There is a history of compartmentalized administrative remains focused on irrigation, the multisector dialogue functioning that has led to a lack of coordination and dia- among water dependent sectors has yet to be put into logue mechanisms between sectors and users. At the lo- operation, and stakeholders have not yet been properly cal level, the water users associations (Pani Panchyats) incorporated into the planning and decision processes. provide a mechanism for dialogue among farmers, but similar mechanisms do not exist for other sectors/users. A new water resources management approach system · There is a general political and administrative wariness that follows IWRM principles needs to be developed. about involving non-official and non-political stakeholders In the prevailing climate of limited awareness and and user representatives in decision-making processes. capacity for IWRM not all aspects of it can be addressed There is also a reluctance to share power with them. simultaneously. Prioritization--first steps first--is required. · Surface water and groundwater management are sepa- The two single most important IWRM functions that needed rated, as is responsibility for managing water quantity to be introduced in Orissa: (DOWR) and water quality (Pollution Board). · Supply management rather than demand management · Informed dialogues across sectors of the government, is the dominant paradigm. The traditional response to and among stakeholders at all levels. meeting demand is to plan and build new infrastructure · Creation of an accepted system of water allocation and to increase supply, rather than look for ways of adjusting tariffs, backed by legislation and institutional change. demand. · Orissa lacks of water entitlements and tariff systems to For the dialogues to be informed and for changes to be ensure proper water allocation and cost recovery. Al- accepted, immediate efforts are required to create: though the state water policy stipulates the full recovery of operation and maintenance costs from users, a long · awareness about water issues and a participatory frame- tradition of underpriced water for irrigation may be very work to address them, difficult to change. Similarly, pricing mechanisms for · a hydrological and environmental knowledge base to industrial water use needs urgent attention. support this, and · There is a large degree of self-supply in the informal · capacity building at all levels, and among all stakeholder sector; that is, people access surface water and ground- groups. water sources directly through private tube wells or from ponds, for example. These activities remain outside the The IWRM Roadmap for Orissa, described below, proposes reach of the government. Related to this are the special 12 actions to deliver on these priorities. 26 Informed Dialogues Across Government The State Level Sectors and Stakeholders Currently, the institutional structure for integrated water resources management in Orissa is limited to the State Putting IWRM processes into action requires initiatives at Water Resources Board (SWRB), which is not very active. four levels: (1) the local level, from the villages to the RBOs; The state is also exploring the possibility of setting up a (2) the basin/RBO level; (3) the state level; and (4) the regulatory authority. While the institutional set-up of other interstate/union level. states cannot easily be translated to Orissa because of differing geography and economic conditions, the The Local Level experience of other states can provide inspiration and The local level in water resources management goes all serve as a starting point.12 State authorities will have to the way from the villages through the Gram Panchayats, identify the structures best suited to the water conditions blocks and districts/Zilla Parishads to the river basin/RBO and challenges of Orissa. In addition, the appropriate state level. Only the irrigation sector has developed structures level structure for IWRM, and the enabling legislation will for stakeholder participation in water management at this need to be discussed and designed in close coordination level. However, since water is an important factor in overall with the development of the RBO (and lower) structures economic and social development, issues concerning water to ensure that bottom-up and top-down processes form a are also being addressed in other sectors. The potential coherent whole. for using these forums to promote the principles of IWRM, including discussions of allocation among users, needs to Central Government and Interstate Levels be explored. While water is a state matter and the state government will play the main role in the development of IWRM in The Basin Level Orissa, the central government also plays a role in state The function and composition of river basin organizations water resources management through the Department of as stipulated in the DOWR's February 2007 resolution Water Resources and the Central Water Commission, as are as follows: "The objective of RBO is to ensure IWRM well as a number of acts (such as the Environmental Act). in the basin. RBO is an organization of all stakeholders In addition, projects receiving financial support from the department in the basin and will bring coordination of central government as well as the environmental impact their activities with a view to resolving conflicts and assessments of major projects require central government avoid duplication among them" According to the 2007 approval. Orissa will also have to consider establishing and State Water Policy "the (river basin) plans prepared by possibly institutionalizing agreements with upstream states the OWPO will be ground-truthed through RBOs" and in order to come to water sharing arrangements. "placed for approval of the State level Water Resource Board" The interaction between the local level and the river basin level remains a serious challenge. In An Accepted System of Water Allocation particular, it remains unclear how the views and voices and Charging of stakeholders at lower levels will be represented in the basin level council, and how all water users, as well as One of the most important and difficult to address those affected by water, get to the table. Pilot projects constraints is that of the lack of properly functioning water for the implementation of river basin organizations are allocation and charging systems. There is still far to go in currently being undertaken in Rushikulya (DOWR), Orissa to achieve the stated policy goal of full recovery Mahanadi (with World Bank support) and Baitarani (with of operation and maintenance (O&M) costs. In irrigated ADB support). Developing these RBOs from theory to agriculture, which is the main water user in the state, practice while respecting the diversity of each basin and cost recovery in surface water systems (canals) remains its problems, requires extensive dialogues and stakeholder low. During the wet (Kharif) season, users are charged consultations. some 250 rupees per hectare (class i), which is only 12. An example is the Maharashtra Water Resources Regulatory Authority, which operates jointly with a state water council and water board. 27 about 30 percent of actual O&M costs. While charges It appears that there actually is a willingness and ability during the dry (Rabi) season differ according to the crop, to pay more realistic water tariffs; however, this would they are also much lower than actual costs. In contrast, require a change of the society's mindset from reli- farmers who depend on lift irrigation (from groundwater ance on the government as provider to taking individual or surface water) do pay their actual energy costs. For responsibility. The establishment of realistic prices that example, growers of high value crops pay between 5,000 are closer to the full-cost price should be within reach in and 10,000 rupees per hectare during the Rabi season, urban areas, particularly since the willingness to pay of indicating that there is, in fact, a willingness and ability to commercial and industrial users substantially exceeds pay full costs. the full cost. · Creating incentives, through an improved water alloca- Water in Orissa's urban areas is also underpriced, resulting tion and charging system, to improve water use efficien- in poor and unreliable service and a public financial deficit. cy, mainly in agriculture (through volumetric pricing), but This need not be so. Examples abound of Indian cities also in urban systems. where full cost recovery is taking place, and even some that are making a profit. Cost recovery need not be a burden for Several different models to make the improvements the urban poor who can be supplied at less than full cost necessary in the allocation of water may be conceived. through cross subsidies from commercial and industrial There is a spectrum of approaches running from a users. "minimum" to a "maximum" approach, as described below. The current allocation system for surface water can best be characterized as an evolving status quo. That is, the · Under a so-called minimum solution, the state govern- current system respects existing traditional rights while ment would take steps to improve the existing system "new water" from dams and other such water infrastructure by gradually raising the prices of surface water to projects is being allocated on a project-by-project basis. comply with the SWP. At the same time, appropriate The extent to which such allocations are based on dialogue structures would be created among sectors considerations of the actual cost of providing water to new and with stakeholder to help improve the decision- users, and their willingness to cover those costs, are not making process. clear. · A maximum solution would involve taking bold steps towards the implementation of a modern system of No groundwater system currently exists and there is a high water allocation and charging. Attempts would be made degree of self-supply from this source for both domestic to put in place mutually attractive sharing arrangements and agricultural use. (win-win) among industries, urban areas and irrigated agriculture. A system could be envisaged by which An improved system of water allocation and charging in water is allocated through entitlements to various user Orissa has two main requirements, as discussed below. groups, which are then allowed to negotiate and trade among themselves. The state government could look · Moving towards financial sustainability by improving cost to the legislation that created the Maharashtra Water recovery. The goals set in the State Water Policy (SWP) Resource Regulatory Authority (2005), which makes include "participation of the beneficiaries in the capital provision for the establishment of such a system, as an cost in suitable proportions," "differential water rates for example of how one could be created in Orissa. Such different categories of uses," and "cost of operation and new arrangements will require extensive consultation management will be fully recovered from the beneficia- and the dissemination of information to raise awareness ries." The SWP further requires the application of the in rural communities. It needs to be explained to local principle of "polluters must pay" to meet the expenses residents and policy makers that when prices reflect wa- of maintaining water quality. As described above, in ter's true economic value in an accurate and transparent order for this to happen, the prices paid by farmers who manner, everyone benefits. irrigate their fields must be raised to more realistic levels. 28 A Roadmap for IWRM in Orissa: Priority Actions to Create an Enabling The Purpose Environment The purpose of the IWRM Roadmap is to launch a stepwise · Action 1: Develop a revised State Water Plan (or an process to introduce and implement IWRM in Orissa IWRM Plan) by using the IWRM Roadmap to establish a by proposing a set of actions that can be realistically prioritized plan with costs and timelines to implement the implemented within the proposed time frame. An important state's water policy. assumption is that external assistance will be provided to · Action 2: Review the legal framework for water resourc- the Orissa state government in areas where there is no or es management in order to develop recommendations very limited capacity. for updating and harmonizing water related legislation as well as develop a Water Act for Orissa. Overall Criteria for the Roadmap Priority Actions Addressing Institutional The point of departure is the State Water Policy Development adopted in 2007. This implies moving from policy goals to the establishment of a timeline of prioritized actions · Action 3: Review existing institutional mechanisms (operationalizing the water policy). Addressing the key for water resources management at the highest state issues and priorities of the state as reflected in its water level. Revitalize the state's Water Resources Board policy should be considered essential. However, there are and consider the creation of a state water resources other key issues and constraints that are not specifically council. included in the policy document and have to be addressed · Action 4: Consider the creation of a regulatory author- as the proposed actions unfold. ity for water resources whose mandate would include establishing and enforcing the allocation of water entitle- Figure 7 illustrates the approach for the development of the ments for various categories of use, as well as establish- IWRM Roadmap, showing that it is strongly linked to the ing a system of water tariffs, including criteria for water state's water policy but also includes additional links to key charging at the basin and state level. issues and constraints that may not be part of the policy. · Action 5: Develop the river basin organization structure All the proposed actions are referenced against relevant as stipulated in the state water policy. The initial focus articles in the State Water Policy. should be on creating operational forums for intersector dialogue and stakeholder participation at the basin level and below (district, block, village), starting with selected Figure 7. The Basis for Developing the pilot basins. IWRM Roadmap · Action 6: Develop capacity within the DOWR to respond to the requirements of implementing the IWRM Road- map, initially through an interdepartmental and multidisci- Key issues Key constraints plinary IWRM Directorate and the Orissa Water Planning Organization (OWPO). · Action 7: Develop institutional and human capacity to respond to the requirements of implementing the state State water policy water policy through an IWRM approach, at the state, basin, and local levels. · Action 8: Develop a multi-stakeholder Orissa Water Partnership. IWRM roadmap 29 Priority Actions Addressing Management Earth Summit that took place in 1992. The policy paper is Instruments strongly inspired by the Dublin Principles, which the Bank summarizes into three: the ecological, the institutional, · Action 9: Develop an awareness, advocacy, and educa- and the instrument principles. A review by the Bank's tion program for IWRM. Operations Evaluation Department (undertaken some 10 · Action 10: Develop a water allocation system based on years later) concluded that "while the 1993 Policy Paper a system of water rights/entitlements associated with dif- remained relevant and appropriate, the major challenge was ferential water rates/tariffs for different uses, as well as a the developing of context-specific, prioritized, sequenced, system of pollution charges. realistic and patient approaches to implementation (World · Action 11: Develop a hydrological information system Bank 2004)." for collecting, processing, archiving, and disseminating water-related data. The Bank' s Water Resources Sector Strategy of 2004 · Action12: Develop a system to address the environmen- reflects IWRM thinking, and includes the "comb" developed tal flow requirements for the state's ecosystems. by the Global Water Partnership to illustrate IWRM. However, it mentions IWRM only by stating that "the main management challenge is not a vision of integrated water Concluding Remarks on Part 2 resources management but a pragmatic but principled approach." It goes on to say further that "there is some This case illustrates the relevance of the basic principles concern that this sequenced and prioritized approach of IWRM to Orissa. Of particular importance are the means abandoning the idea of integrated water resources governance dimensions in IWRM; namely, emphasis on management, which was a core principle of the 1993 multisector and multi-stakeholder dialogue and coordination Policy Paper. This is not the idea. As noted earlier, even at all levels, and the focus on the river basin as the natural the world's most developed countries are a long way from unit for managing water and balancing its uses. Both integrated water resources management, and progress dimensions are essential to address current water problems has been slow and incremental. The goal of the Strategy in the state. The state of Orissa has realized this and has is not to dismiss the goal of integrated water resources adopted IWRM principles in its policies and strategies, management, but to define practical, implementable and, inspired by similar decisions by the government of India, and therefore, sequenced and prioritized actions that can lead by encouraging examples from other Indian states. to that end." However, Orissa faces the challenge of implementing these Bank statements in the Water Resources Sector strategy policies and strategies in a situation of severely limited echo fully the views and experiences of the author as financial and human resources, and against the backdrop described in Parts 1 and 2 of this paper, although it is of old traditions where water development is synonymous regrettable that the Bank apparently has become hesitant with irrigation development. Moreover, the state has a to refer explicitly to IWRM. As argued, when a large relatively large informal sector that remains outside of the number of the Bank's client countries explicitly adopt this government's reach. A pragmatic approach with small steps terminology (after being strongly encouraged to do so by in the right direction is the only realistic way forward. the international community, including the Bank, in the 1990s), why not continue to use this common language while patiently focusing on the small steps? A shift in Bank Part 3: Reflections on the World Bank rhetoric took place in the early 2000s with renewed focus and Integrated Water Resources on the rising tide theory ("a rising tide raises all boats") and Management on Bank support for the development of water infrastructure, as well as an expressed willingness to take the necessary In 1993, the World Bank issued a Water Resources risks. As the Bank refocused on the development aspect, Management Policy Paper that reflected the broad global the softer governance aspects, including the IWRM consensus forged at the Dublin Conference and Rio terminology, apparently had to take second place. That is 30 unnecessary. Nothing in the IWRM concept contradicts the Global Water Partnership. 2000. Integrated Water need for infrastructure development;13 the argument is not Resources Management, GWP Technical Committee about if but how to develop water infrastructure. This may (TEC) Background Paper No. 4. sound like academic splitting hairs, but when confronted Hassing, Jan, Niels Ipsen, Torkil-Jønch Clausen, and with leaders in the developing countries who are committed Henrik Larsen. 2009. "Integrated Water Resources to their own IWRM-inspired policies and strategies, Management in Action". Jointly prepared by DHI terminology sometimes does matter. Water Policy and the UNEP-DHI Centre for Water and Environment. United Nations World Water Assessment The Bank's 1993 Policy Paper and 2004 Sector Strategy Programme, Side Publications Series, Dialogue Paper. are well-articulated documents, which, in their arguments Paris: UNESCO. and recommendations, constitute a strong support for Intergovernmental Panel on Climate Change (IPCC). 2007. improving water resource development and management in Climate Change 2007: Synthesis Report. Contribution accordance with the principles of IWRM. The challenge in of Working Groups I, II and III to the Fourth Assessment practical implementation is two-fold. At the policy level and Report of the Intergovernmental Panel on Climate in the dialogue with top decision makers, the challenge is Change [Core Writing Team, Pachauri, R.K and to discuss the role of water in all sectors of the economy Reisinger, A. (eds.)]. IPCC, Geneva, Switzerland. (notably agriculture and energy), not just within the "water Chapter on freshwater. sector" as is most often the case. This applies not only to Orissa, India (2004). Orissa State Water Plan. the Bank's dialogue with country stakeholders, but also UN Conference on Water and Development. 1992. Dublin internally among the Bank's staff and managers. At the Principles. Dublin: January. basin/local level, a major challenge is to build the bridge Various surveys made by UN-DESA (2007), UCC-Water between IWRM ideals and principles on the one hand, (2007), GWP (2006) and AfDB (2007). and the realities of old traditions, little awareness, little World Bank. 1993. Water Resources Management Policy political will, and low capacity, on the other. At this level the Paper. Washington, DC. serious issues of competing demands for water need to World Bank. 2004. Water Resources Sector Strategy. be addressed urgently, often in the absence of necessary Washington, DC. data and information, in the absence water allocation and regulation mechanisms, and in the absence of well-defined dialogue mechanisms among stakeholders to address them. Annex ­ Examples of countries having At both levels no quick fixes are possible, and Bank staff adopted IWRM as a key concept, and managers are challenged to be patient and take small Roadmaps, Strategies, Policies, Laws, steps as expressed in the policy and strategy documents. Plans etc. with Explicit Reference to "IWRM" References The table below provides examples of some 40 countries that have found IWRM a useful framework for management Asian Development Bank. 2008.). Orissa Integrated of water resources and have included it as a pivotal Irrigated Agriculture and Water Management concept. The concept has been included in key government Investment Program (OIIAWMIP), Technical Assistance documents that guide and regulate the use, conservation, for Integrated Water Resources Management (IWRM) and protection of a nation's water resources and in Orissa. Manila. implementation at local level. The table is not exhaustive. 13. In fact, the Global Water Partnership defines the "M" in IWRM as "development and management" (Global Water Partnership 2000). 31 Country Evidence of the continued adoption and explicit use of IWRM Algeria · National Plan for Water ­ Ministry of Water Resources (2003) · National Water Law ­ Government of Algeria (2005) · Action Plan for implementation of an IWRM Framework ­ Ministry of Water Resources (draft 2006­7] Angola · IWRM and Water Efficiency Roadmap ­ Ministry of Water & Energy (draft 2007) Argentina · IWRM Roadmap ­ Sub-secretariat of Water Resources (2007) Armenia · Water Code ­ Government of Armenia (2002) · National Water Policy ­ Government of Armenia (2005) · National Water Programme ­ Government of Armenia (draft 200:) Botswana · IWRM Strategy and Action Plan ­ Ministry of Minerals, Energy and Water Resources (2006) Brazil · National Water Policy (Law No 9433) ­ Government of Brazil (1997) · National Water Resources Plan ­ Ministry of Environment [SRH/MMA), National Water Council · (CNRH) & National Water Agency (ANA) (2007) Burkina Faso · Decree No. 2003-220: Action Plan for IWRM in Burkina Faso (PAGIRE) ­ Ministry of Agriculture, Hydraulics & Fishing Resources (2003) · Burkina Faso Water Vision ­ Ministry of Agriculture, Hydraulics & Fishing Resources (2000) · Water Law No. 002-2001 ­ Government of Burkina Faso (2001) Cambodia · Integrated Water Resources Management (IWRM 2005) and Roadmap of Cambodia Resources Management and Conservation (2006) · Water Law ­ Royal Government of Cambodia (Sept. 2006) China · China Water Law ­ Government of China 2002 Colombia · National Development Plan 2006­10 ­ National Planting Department (2006) Costa Rica · National Strategy for Integrated Water Resources Management ­ Government of Costa Rica (2006) · National IWRM Action Plan ­ Government of Costa Rica (2006) · National Water Law (No. 14585) ­ Government of Costa Rica (draft 2006) Cote d'Ivoire · IWRM Roadmap 2007­2015 ­ Ministry of Environment Water & Forestry (2007) Egypt · National Water Resources Plan ­ Ministry of Water Resources and Irrigation (2004) Eritrea · Integrated Water Resource Management and Water Efficiency Plan (IWRM/WE) ­ Ministry of Land Water & Environment (draft 2007) Ghana · IWRM Component Support programme (2004­2008) ­ Water Resources Commission (2004) · Water Resources Policy ­ Water Resources Commission (draft 2005) Grenada · Simultaneous preparation of IWRM Roadmap and National Water Policy ­ Water Policy Steering Honduras · IWRM Action Plan ­ Honduras Water Platform (2006) Indonesia · National Water Law No 7/2004 ­ Government of Indonesia (2004) · IWRM Roadmap ­ Directorate General Water Resources of Ministry of Public Works (2006) India · National IWRM Committee ­ Government of India 1999 Kazakhstan · IWRM National Roadmap including proposed project outlines ­ Speed-up of the IWRM 2005 objectives imple- mentation in Central Asia ­ Government of Kazakhstan (2006) Kenya · Water Act 2002 - Government of Kenya (2002) · National Water Policy on Water Resources Management and Development (Sessional Paper No. 1 of 1999) ­ Ministry of Water and Irrigation · Integrated Water Resources Management and Water Efficiency Plan for Kenya ­ Ministry of Water and Irrigation (draft 2007) Kyrgyzstan · IWRM National Roadmap including proposed project outlines ­ Speed-up of the IWRM 2005 objectives imple- mentation in Central Asia ­ Government of Kyrgyzstan (2006) Lao PDR · Policy on Water and Water Resources ­ Government of Lao PDR (draft 2000) · The Law on Water and Water Resources ­ Government of Lao PDR (1996) · IWRM National Roadmap ­ Water Resources Coordination Committee Secretariat (2206) (continued) 32 Country Evidence of the continued adoption and explicit use of IWRM (continued) Lesotho · Roadmap to completing integrated Water resources management and water efficiency planning in Lesotho ­ Ministry of Natural Resources, Water Commission (April 2007) Liberia · Liberia IWRM Roadmap ­ Ministry of lands, Mines and Energy (draft 2007) · National Water Policy ­ Ministry of Lands, Mines and Energy (draft 2007) Malawi · National Water Policy ­ Ministry of Irrigation and Water Development (2005) · Water Resources Act No. 15 of 1999 with later amendments. Government of Malawi · Integrated Water Resources Management/Water Efficiency (IWRM/WE) Plan for Malawi ­ Ministry Irrigation and Water Development (draft 2007) Malaysia · 9th Malaysia Plan- Economic Planting Unit ­ Prime Minister's Department (2005) · National Study for the Effective Implementation of IWRM in Malaysia ­ Ministry of Natural Resources and Envi- ronment (2006) · Our Vision for Water in the 21st Century ­ Ministry of Natural Resources and Environment (2000) Mauritania · IWRM Action Plan ­ National Council for Water (2007) · National Development Policy for Water & Efficiency ­ Ministries of Water, Energy & Environment (1998) · National Water Act (Article 3) ­ Government of Mauritania (2005) Morocco · Master Plans of Integrated Water Resources Development for Rivet Basin ­ Ministry of Land, Water and Envi- ronment (2001) · National Water Plan ­ Ministry of Land, Water and Environment (2006) · Decree no 2-05-1594 ­ Development and Revision of Master Plans & National Plans for Integrated Water Resources Management ­ Government of Morocco Mozambique · National Water Resource Strategy ­ Department of Water Affairs & Forestry (2004) · IWRM Plan ­ National Directorate of Water Affairs (draft 2007) Namibia · National Water Policy White Paper ­ Government of Namibia (2000) · Water Resources Management Act ­ Government of Namibia (2004) · Integrated Water Resources Management Strategy and Action Plan ­ Ministry of Agriculture, Water and Rural Development (2006) Nicaragua · General law on National Waters ­ Government of Nicaragua (2007) · Environmental Action Plan ­ Ministry of Environment (1994) · IWRM Action Plan ­ Ministry of Environment (1996) Philippines · Medium Term Philippine Development Plan (2004-2-10) ­ Government of Philippines (2004) · Clean Water Act ­ Government of Philippines (2004) · Integrated Water Resources Management (IWRM) Plan Framework ­ National Water Resources Board (2007) Swaziland · Water Policy ­ Ministry of Natural Resources and Energy (draft 2007) · IWRM and Water Efficiency Plan ­ Water Resources Branch (draft 2007) · Water Act (2003) ­ Government of Swaziland Tajikistan · IWRM National Roadmap Including Proposed Project Outlines: Speed-up of the IWRM 2005 Objectives Imple- mentation in Central Asia ­ Government of Tajikistan (2006) Tanzania · National Water Sector Development Programme 2006­2025 ­ Ministry of Water (2006) · IWRM Strategy and Action Plan ­ Ministry of Water (2004) · National Water Policy-Ministry of Water (2002) · National Water Law based on revised Water Act no. 42 of 1974 ­ Government of Tanzania (draft 2007) Thailand · National Water Law/Code ­ Government of Thailand (draft 2007) · National Water Policy ­ Ministry of Natural Resources and Environment (2000) · IWRM National Roadmap ­ Department of Water Resources (2007) Togo · National Water Policy ­ Directorate of Water and Sewerage (draft 2007) · National Water Law ­ Directorate of Water and Sewerage (draft 2007) · IWRM Roadmap ­ Directorate of Water and Sewerage (draft 2007) Turkmenistan · IWRM National Roadmap Including Proposed Project Outlines ­ Speed-up of the IWRM 2005 Objectives Imple- mentation in Central Asia ­ Government of Turkmenistan (2006) (continued) 33 Country Evidence of the continued adoption and explicit use of IWRM (continued) Uganda · A National Water Policy -Ministry of Water, Lands and Environment (1999) · National Water Action Plan ­ Water Resources Management Department (1994) · Water Resources Management Reform Strategy ­ Water Resources Management Department (2005) · National Water Quality Management Strategy ­ Ministry of Water and Environment (2006) Uzbekistan · IWRM National Roadmap Including Proposed Project Outlines ­ Speed-up of the IWRM 2005 Objectives Imple- mentation in Central Asia ­ Government of Uzbekistan (2006) Zambia · IWRM and Water Efficiency Plan ­ Ministry of Energy and Water Development (2008) · The Revised National Water Policy ­ Ministry of Energy and Taler Development (2007) · Water Resources Management Bill ­ Ministry of Energy and Water Development (draft 2007) · National Development Plan ­ Ministry of Energy and Water Development (2007) 34 2. Water Demand Management Janusz Kindler Warsaw University of Technology, Poland Abstract variables as government policies, population levels and distribution, energy use and costs, per capita disposable The purpose of this paper is to synthesize the state of income, technological development, consumer habits the art in the area of water demand management and to and lifestyles, and the prices of water withdrawals and reflect on how this knowledge can be applied to policies wastewater disposal. Developing relations between those and projects in the water sector. The paper first presents variables and using them to estimate water demands some fundamentals of water demand analysis and modeling under different climatic, social, and economic conditions approaches, and then discusses the water demands of requires analytical approaches. This paper describes some individual water-use activities, such as household, industrial of these approaches and shows how they can be used to plants and irrigation systems. The results of empirical analyze various demands for water within the framework studies show that, in most cases, residential water demand of integrated water resources management (IWRM). To is inelastic, that is, it does not respond to price changes. manage water demands more effectively, a balanced set The aggregated water-use activities are represented in of different demand measures should be sought that will the paper by urban agglomerations. In discussing different both harness the efficiency of market forces and strengthen analytic tools, special attention is given to the IWR-MAIN the capacity of governments to carry out their essential Water Demand Analysis model. Water demand management regulatory roles. at the level of the river basin is seen as part of the water resources allocation problem involving both the supply In many parts of the world water shortages and the and demand sides of the water balance equation. Special degradation of water quality, both surface and groundwater, attention is given to the relatively new group of hydro- have reached alarming proportions. There are several economic models and their role in water demand analyses. obstacles to environmentally sound, economically efficient At the national level, practically all studies and statistics and socially responsible water management that cause such refer to national water use rather than water demand. The situations. Although various technical and management paper closes with a few comments on the role of water measures may be used to increase the availability of the demand analysis, its limitations, and further research needs. resource, the most important thing is to manage existing supplies better. Emphasis should be placed on the efficient use of existing supplies, as well as their conservation, Introduction recycling, and reuse. Important questions about water demand arise whenever It has been already more than 30 years since Sewell and decisions have to be made regarding water resources Roueche (1974) underlined that one of the most critical management. Typically, these questions are about how challenges is to shift from the more or less traditional much water shall be used today and in the future, where supply-oriented extensive approach to water resources it will be needed, and what purposes it will serve. The management to a demand-oriented intensive approach. actual demands always depend on such time related The importance of the demand side of water resources 35 management is perhaps more crucial and obvious today The paper begins with a discussion of some fundamentals than in the past. As stressed by Gilbert White (2006): "The concerning the analysis of water demands. Next, it provides old paradigm of designing the cheapest reliable supply with a brief overview of the methodological framework of demand little attention to demand determinants, pricing structures, analysis and modeling approaches. The next four sections and financing policies is no longer suitable." In the IWRM of the paper correspond to the different levels of water framework, a broad range of demand management demand analysis shown in Figure 1: (1) the individual water- measures must be considered. use activities, (2) the aggregated water-use activities, (3) the river basin planning level, and (4) the national level. The purpose of this paper is to synthesize and summarize the state of the art in the area of water demand First, the demands of individual water-use activities are management and to reflect on how this knowledge discussed (households, industrial plants, and agricultural can be applied to water sector policies and projects. farms). Next, the demands of individual water-use activities Because water demand management, by definition, is an are combined into the demands of aggregated activities, interdisciplinary task, the readers of this paper may include illustrated by urban water demand management and persons with diverse professional backgrounds. For that forecasting. The discussion that then follows concerns the reason, as well as because of the length of this paper, the river basin and the national perspective in water demand treatment of the theoretical issues is kept on a fairly general analysis. The paper ends with few thoughts on the role of level. However, references are made to basic textbooks and water demand analysis, its limitations, and further research other publications where the interested reader may find needs. more in-depth information. It is recognized throughout the paper that any attempt to influence and improve demand analysis methods in water resources management requires Fundamentals of Water Demand the reader to pay careful attention to the institutional, Analysis administrative, legal, and economic constraints under which water demand decisions are being made. The alternative A useful way to begin the discussion of the fundamentals approaches to water demand analysis described in this of water demand analysis is by examining the definitions paper should always be interpreted in the context of these of relevant terms, especially water use, demand and case-by-case varying constraints. requirement, the dimensions of water demand, the demand Figure 1. The Different Levels of Water Demand Analysis National water Nation demand Regional River basin 1 River basin n River basin N water demand Water demand of aggregated Municipality water-use activities Chemical plants Households Crop production farms Water demand of individual water-use activities Steel plants Industrial plants Livestock production farms Thermal power stations Schools Crop or livestock processing plants Others Others Others Source: Kindler and Russell, 1984 36 elasticity dimensions of water demand, and demand substitution and adjustment that can be seized upon as the management measures. cost of water to users goes up. Water uses can be categorized as follows: (1) intake uses, At this point it seems reasonable to look at the notion of (2) onsite uses, and (3) in stream uses. Intake uses include "the price of water" and to ask where it comes from. In water for household, agricultural, and industrial purposes; principle, there are two ways of establishing water prices. that is, water uses that actually remove water from its One is through the interaction of supply and demand in an source. Onsite uses consist mainly of water consumed by open market. However, there are not that many examples of natural vegetation, rainfed agriculture, and evaporation from "water markets." The second option is setting water prices surface and groundwater bodies, swamps, and wildlife. In by means of administrative decisions. This underlies most principle, the onsite uses deplete water supplies before of the water pricing schemes in existence. The questions of they reach surface and groundwater resources. In stream primary importance in the latter case are how the price is to uses include water for maintenance of aquatic ecosystems be administered, how high it should be, and to what extent (especially wetlands), navigation, hydroelectric power it should be varied in time and space. The ultimate purpose generation, recreation, and some fish and wildlife uses. of managing the demand for water is to ensure that a given supply is allocated as close as possible to its "optimal" Demand is a general concept used by economists to use pattern (Winpenny 1994). For the optimal allocation of denote the willingness of consumers or users to purchase resources, the price that water users pay for their marginal goods, services or inputs to production processes, since units of water withdrawal, consumptive use, and wastewater that willingness varies with the price of the thing being disposal services should reflect the marginal costs of purchased. The demand function conforms to the negative supplying these units. Although this theoretical ideal is relationship between price and quantity demanded, difficult to achieve, demand management measures can all other factors affecting demand being equal (ceteris help to move the allocation of water closer to it. In addition, paribus assumption). It is known that, in addition to price, it should be remembered that the demand for water is not water demand is affected by several other variables, only a function of its price. including consumer income, the prices of substitute and complementary factors, and environmental parameters. The elasticity of demand with respect to price P is the Therefore, a general functional relationship between the percentage by which the quantity demanded Q changes quantity of water demanded q and k explanatory variables for a one percent change in price. For example, if the x1, x2, ..., xk, one of them being water price, is q = f (x1, x2, ..., price elasticity of demand for water is ­0.5, this means xk). When speaking of the demand for inputs to production that a 1 percent increase in the price of water will result processes (for example, industrial or agricultural water in a 0.5 percent decrease in water demand, with all other demand), it is called "derived" demand because the demand demand generating factors held constant. The demand for water is derived from the demand for the final output of for a commodity having no close substitutes is likely to the production process. be inelastic; that is, the absolute value of the PE (price elasticity) is < 1.0. If PE = 0 (perfect price inelasticity) a Requirement refers to water use that does not obey the given commodity or service is a requirement. In other words, demand rule. In other words, the same quantity of water the more easily available substitutes are, the greater is the is purchased and/or used regardless of the price (if any). elasticity of demand. It is obvious that there do exist minimum requirements for many things in life that are unresponsive to price. But The concept of elasticity can be used in relation to any one in agriculture and industry, the true "requirements" are of the demand-determining variables (price and income usually only a small part of observed water use, and are are generally considered the most important). However, a almost never what large water projects are built to supply. distinction should be made between short- and long-term Therefore, to treat all existing and future water uses as elasticities of water demand. As a rule, long-term demand is requirements is to ignore important possible ways of more elastic than short-term demand because longer time 37 periods allow for more opportunities to adjust, and thus stochastic. Some of the common econometric models are present more options for substitution. Box-Jenkins, ARIMA, and mulivariate regression models. Although this paper primarily concerns economic demand The modeling process usually proceeds by a series of management measures, it is important to mention also iterations through the following steps: that there are several other measures of importance. As shown in Table 1, they include educational, technical, (1) Choosing the model structure (specifying the model), regulatory, administrative/restrictive, and operational that is, selecting variables and hypothesizing structural control measures. relationships, including whether or not simultaneous determination is involved for two (or more) variables; All these measures have proved to be effective in reducing (2) Choosing functional forms; (3) Estimating model water demands and maximizing the benefits provided by parameters; (4) Verifying and validating the model, and the existing water infrastructure. They also free water for (5) Using the model. other uses and reduce environmental degradation. Efforts to reduce water demands can, therefore, contribute directly to In the case of programming models, analysis of demand for the development goals of many countries, especially those some inputs must be based on some knowledge of input- that are chronically short of water or the capital to invest in output relationships, although this is usually incomplete, water resources development. especially for production involving several inputs. Consequently, the extent to which input-output relationships (production functions) can be estimated depends on the The Modeling Approaches overall knowledge of the specific production processes and on data availability (for example see Kindler: 1988). There are two broad approaches to water demand modeling: econometric and programming. The theoretical The programming approach requires knowledge of what is framework of economic theories of demand for water as an going on within and among many unit production processes. input and consumer demand for water (Hanemann 1998) But for any given process one can be content with little underlies most of the models built and applied. observed data, because they can be calculated on the basis of the principles and rules of production practice. In general, an econometric model specifies the statistical These calculations take into account how each unit process relationship that is believed to hold among the various would operate under different assumptions about, for economic quantities pertaining to a particular economic example, boiler efficiencies, pressures and temperatures of phenomenon under study. An econometric model can be reaction, pump types, and so forth. Some of the necessary derived from a deterministic economic model by allowing steps for developing a programming water demand model for uncertainty, or from an economic model that itself is include the development of material and energy balances Table 1. Categories of Demand Management Measures No. Category Demand Management Measures 1 Economic (incentive-based) Pricing structures and levels; subsidies, taxes and tax credits; rebates and buy-backs; low- interest loans; fines for regulatory non-compliance 2 Technical Dual plumbing systems handling two qualities of water and toilets; water saving irrigation equipment 3 Educational Education, information provision and conservation campaigns; water audits 4 Regulatory Use and consumption regulations; building, plumbing and landscaping codes 5 Administrative/restrictive Rationing, moral suasion, and voluntary use reduction 6 Operational control Leakage detection and alleviation; pressure control and sewer infiltration control Source: Herrington, 2006; World Bank, 1994. 38 for unit processes, the specification of factor inputs and the reasons for this might be that in most high-income their costs, and the calculation of wastewater residuals. countries, residential water and sewerage bills are a small The principal question is to determine whether the savings part of household budgets. The conclusion that the price in costs of intake water and wastewater discharge justify of water is relatively ineffective in reducing water demand the increased costs associated with making the change. is sometimes strengthened by policy makers who argue Finding an answer to such questions is helped by the that water pricing ignores the equity implications of water application of mathematical programming; it involves the allocation decisions. Instead of pricing mechanisms they use of optimization techniques such as linear, non-linear, tend to favor targeted use restrictions to provide a basic or mixed-integer programming. The objective function is level of water service at lower costs. usually either to minimize the total cost of production or to maximize the net benefits of the system. The constraints An excellent review of econometric residential water take care of production requirements, availability of various demand models developed by numerous authors in the last input resources, and interrelations within and among the unit 30­40 years (mostly in the United States) is presented in production processes. the monograph by Renzetti (2002). He underlines that the primary purpose of these models is to "characterize the specific nature of the relationship between the observed Residential (Household) Water quantity of residential water use and the explanatory Demands variables suggested by economic theory." Among the principal issues analyzed is the determination of the price(s) The issues concerning residential water demands in the of water. The difficulties in modeling include specification high-income countries and in the developing countries of the usually complex price-quantity function for residential must be discussed separately because of differing water demand. A more fundamental problem is the lack of socioeconomic conditions among these groups of data on many potential demand explanatory variables and on countries. It must be recognized right from the outset, certain characteristics of residential water, such as reliability that interrupted (intermittent) water supply is the norm in and pressure. This confirms the problem raised by many most developing countries and that this is one of the most water policy makers and water utility managers; namely, common methods of controlling water use (Vairavamoorthy that information to determine the performance of price and and Mansoor 2006). In fact, intermittent water supply is a non-price water demand management measures in their control necessity and not a designed demand management communities is generally insufficient. Some of the doubts measure. There are many serious shortcomings associated mentioned above are confirmed by Renzetti (2002) who with intermittent water supply and, wherever possible, the concludes that "residential water demands with the possible goal should be full-day continuous supply. exception of outdoor water use in summer months are price and income inelastic." High-Income Countries The above conclusion is confirmed by most of the studies published after 2002. For example, Martinez Espineira The debates about residential water demand focus on both (2005) used unit root tests to find that time-series of the efficiency of different demand management measures residential water use and other variables affecting that and their equity implications. There are several analysts use are non-stationary. The application of the model who suggest that higher residential water prices are not developed in this study used monthly residential water use necessarily the best measures for reducing demand, data from the city of Seville (Spain). The price elasticities and that non-price measures (such as consumption of demand were estimated at about ­0.1 in the short run regulations), which place direct controls on water use, and ­0.5 in the long run. The author concludes: "...these are more viable means of reducing residential water estimates confirm that residential water is inelastic to price, demand. These opinions rely on the empirical research but not perfectly." A similar conclusion was reached by discussed below, which indicates that the elasticities Bithas and Chrysostomos (2006) in a study concerning of residential water demand are generally low. One of residential water demand in the city of Athens (Greece). 39 Water prices and income were used as the main demand block. Other disadvantages of IBT as well as alternative explaining variables in that study, which concluded that tariff scheme proposals are discussed by Liu et al. (2003). while increasing income would induce a drastic increase in water demand, economic instruments have little potential to influence water use. Water Demands of Industrial Plants It is worth quoting a comment made by Renzetti (2002) Almost all industrial plants use water as an input to their on the influence of climate change on residential water production activities, however, the purposes for which water demands. He observes that, generally, "increases in is used vary. Water can be used as part of the product, temperature or evapotranspiration rates lead to higher it can be used to convey the product from one stage of residential water demands while increases in precipitation production to another, it can used for washing and cleaning decrease demands." throughout the plant, and it can be used as the principal heat removal medium. Econometric and programming approaches are both used when modeling industrial water Developing Countries demand relationships. When looking at demand for residential water in developing In the 1960s, Bower, Lof, Kneese, and Russell at Resources countries it is important to distinguish among three principal for the Future in Washington, D.C., started developing categories of consumers: high-income, middle-income, and programming models to analyze water use and wastewater low-income (Vairavamoorthy and Mansoor 2006). In the disposal flows in the chemical, petroleum, pulp and paper, high-income category, applicable demand management and metal processing industries. About 10 years later, these measures include in-house retrofitting and outdoor water models were extended further by Calloway, Schwartz, and saving options (garden, swimming pool). Water pricing Thompson (1974). In the 1980s, they were once again measures must, however, be combined with extensive put on the agenda of the International Institute for Applied awareness-raising campaigns given that increased cost Systems Analysis (Kindler and Russell 1984). tends not to lead to water savings among the rich. The most effective demand management option when addressing The basic concept in these models is that of a unit process consumption by middle-income households is water pricing. characterized by fixed proportions between inputs and Increasing block tariff rates and undertaking an effective outputs, which are usually called "technical coefficients." In awareness-raising strategy seem to work for this group industrial production, one may identify such unit processes of consumers. Finally, low-income consumers, who rarely as, for example, water use, water treatment, coal supply have individual connections to piped water, are simply the and transportation, air pollution emissions, and so forth. beneficiaries of demand management measures introduced In general, if one process differs from another in the for the first two categories of water consumers. type, proportions, or timing of inputs, they are treated as separate unit processes in the model. A unit process may The most common water charging policy is the increased be operated on a smaller or larger scale and various unit block tariff (IBT), where high volume residential consumers processes may be operated simultaneously, each one at subsidize relatively poorer ones. The first block is charged the most appropriate level for its purpose. The choice of the at a low rate and it covers basic human needs only. The best combination of unit processes (how to define "best" is next block of so-called "normal" consumers is charged another issue), replaces the traditional engineering choice of a price that fully recovers maintenance costs and the the best combination of inputs and outputs. depreciation of capital assets. Finally, "luxury" consumers pay substantially more for water and thus subsidize the first The advantages of programming models are significant. block of users. Despite the subsidy, the IBT system has Above all, they are future-oriented and permit an analysis of some disadvantages for the poor. The practice of several water demands in hypothetical situations for which there are poor families sharing a single water connection implies that no statistical records. Most of the criticism of these models their total combined water usage exceeds the lowest IBT has been because of the purely economic character of the 40 objective function (usually maximization of net benefits) The demand for crops is determined by crop prices and when it is known that industrial enterprises have other quantities. Several inputs are required to produce the motives in addition to profit maximization. desired amount of crops, water being one of them. Hence, the demands for these inputs are derived demands; that In his overview of industrial water demand studies, Renzetti is, they stem from the specified levels of the primary set (2002) indicated that econometric models are used less of outputs. This is, of course, an idealized picture of what frequently than programming models. He quoted Stone actually happens even in predominantly market economies and Whittington (1984), who point out that econometric since it omits many adjustment mechanisms. For example, (statistical) estimates of industrial water demand if water becomes scarce, agronomists can be expected to relationships are difficult mostly because of the small switch to crop varieties that are more resistant to droughts. sample sizes. Some data are often missing even in the Many other non-market forces also affect the determination case of plants that are included in the sample. In addition, of prices and quantities. These include price supports or problems commonly arise with respect to the simultaneous restrictions in the size of the plot of land planted with a determination of the price of water and the quantity used. particular crop. The current situation with regarding water use and demand In the agricultural sector, the efficiency of water use is studies is well characterized by Renzetti (2002) in his generally low. There are, however, several methods of concluding remarks: increasing efficiency in irrigation. If incentives are in place, including pricing of irrigation water [see for example Tsur ...The relationship between water intake, and Dinar (1997) and Tsur (2000)], farmers are motivated recirculation and discharge on the one hand and the to adopt water-saving irrigation technologies. In principle, prices of other inputs is a particularly understudied these technologies rely on the frequent application of small area. Second, internal water recirculation appears amounts of water as directly as possible to the roots of to be the primary means for firms to reduce water the plants. Reducing the pollution loads of water used by intake. Once again, however, it is not clear whether industries and in urban areas would also enable the reuse the primary motivation for adopting recirculation of some of that wastewater in irrigation. The practical is to save on expenditures related to water intake, implementation of these solutions is not easy because of wastewater discharge, energy, raw materials or human health issues, but there are large potential benefits some combination of all of these. Third, both from the use of wastewater for irrigation. programming and econometric methods of modeling industrial water use ... are best seen as Agriculture will remain the dominant user of water at complementary rather than competing approaches. the global level. In many countries, in particular those They have different data requirements and highlight situated in the arid and semi-arid regions of the world, this different features of industrial water use. dependency can be expected to intensify. The contribution of irrigated agriculture to food production is substantial but, in the future, the rate of growth will be lower than Agricultural Water Demands in the past. Addressing the food and poverty crises in developing countries will require a new emphasis on small- The amount of water involved in agriculture is significant. In scale water management in rainfed agriculture involving several countries the agricultural sector uses more water the redirection of water policy and new investments in that than all other economic sectors combined. Worldwide, area. Possible water management improvements in rainfed most of this water is provided by rainfall stored in the soil agriculture are discussed in depth by Rosengrant et al. profile, and only 15 percent is provided through irrigation. (2002) and IWMI (2007). The amount of irrigation varies from country to country and region to region, depending mostly on climate conditions This paper concentrates on the water demands of irrigated and on the degree of development of the irrigation agriculture only. The analysis of these demands can be infrastructure. facilitated by modeling, and both the econometric and 41 programming models are applied. In the first of these two as equipment maintenance, labor, and so forth. The problem approaches, production functions are usually estimated is to find the set of decision variables that maximizes the to capture the relationships between crop yields and the objective function subject to constraints concerning available land, rain, solar energy, irrigation water, and irrigation resources. By varying parametrically the cost of irrigation technology variables. However, the complexities of water, the model can be used to derive a demand function agricultural production often cannot be adequately captured for that specific input to the production process. in the form of a production function. Many factors other than the price of water can change the quantity of water Renzetti (2002) discusses several agricultural production demanded; their mutual relationships and possibilities for programming as well as econometric models directly substitution cannot be explicitly expressed. Thus, it may identifying agricultural water demand relationships. He often be more reasonable to model agricultural production makes the point that "programming and econometric and related processes (the programming approach), and approaches to modeling agricultural water use are best derive approximate demand functions from them directly. In seen as complements rather than substitutes." Programming Figure 2, for example, a simple agricultural system of "one models describe better technological details under different type crop--one type animal" is presented. behavioral assumptions and they more easily incorporate information regarding physical processes, among them As illustrated in Figure 2, this system encompasses the climate change. Econometric models have the ability to process of producing wheat, its processing and product directly estimate price elasticities, as well as establishing marketing, including alternative uses of some products for the statistical significance of different explanatory demand feeding livestock and livestock processing. The first and variables. most important step in modeling is to define the objective function, which might be, for example, to maximize net Renzetti (2002), quoting the influential work of Caswell and benefits from agricultural production. The next step is to Zilberman, states that: identify the decision variables. The cost associated with each variable can be subdivided into two categories: fixed costs, ...adoption of new irrigation technologies is including capital investment depreciated over time; and positively related to output and water prices while variable costs that include resources (not concerned with negatively related to soil quality. Subsequent capital investments) and the cost of various activities such work, however, appears to demonstrate that the Figure 2. One Type Crop, One Type Animal Agricultural System MARKETING Wheat imports Product demands, reserves and exports Input resources Land Seeds Grain LIVESTOCK PRODUCTION CROP CROP PROCESSING Labor PRODUCTION Flour PROCESSING LIVESTOCK Machinery Wheat Milk (CATTLE) Non-irrigated Grain Cheese Fertilizers wheat Straw Bran Meat Irrigated Pesticides Straw wheat Irrigation water Potable water Manure Salinity Wastewater ENVIRONMENT Source: Gouevsky and Maidment 1984. 42 strength of these relationships may not always be range of activities that may be grouped into residential, strong. In particular, for a given allocation of crops industrial, commercial, and public uses, as well as system to available acreage, irrigation water demands are losses and unaccounted for water. Water use and demand price inelastic. The demand for irrigation water analysis in the residential and industrial sectors have becomes more responsive to price changes, been discussed above. Commercial customers are very however, once it is assumed that farmers are free heterogeneous in their water use. They include food and to alter their output mix. beverage services, commerce conducted from offices, shops, hotels, and so forth. Water is used for cleaning, cooling, In Europe, the implementation of a cost recovery approach sanitation, and landscaping. Public water use includes a in water resources management and water pricing policies wide range of activities, such as street cleaning, watering is promoted by the European Union (EU) Water Framework municipal parks, use in hospitals, government services, Directive, which is currently being implemented in all EU public toilets, public swimming pools, and other similar member countries. A recent publication by Iglesias and public services. Lost and unaccounted for water includes, Blanco (2008) on the role of water pricing policies in irrigated above all, leakages from mains and the distribution systems. agriculture is worth mentioning. The authors developed an These losses, expressed as a percentage of annual water innovative mathematical programming model to evaluate the production, vary substantially. For example, on average, impact of cost recovery in a large number of irrigation districts 12.3 percent of annual production in the United States in Spain. The proposed model allows farmers' behavior to be was reported as water losses in 2005 (Billings and Jones simulated under different water pricing scenarios, taking into 2007). Water losses in the cities of developing countries are account the possibilities of adopting new production patterns reaching extreme levels of up to 40 percent to 60 percent of and new irrigation technologies under changing environments, the water supply (Vairavamoorthy and Mansoor 2006). including climate change. It should be recognized that each urban water sector has distinct quality requirements, usage processes, disposal Urban Water Demands methods, and jurisdictional oversight and responsibilities. As a result, there is no cohesive approach to planning and A characteristic of the 21st century is that a growing policy that can formulate consistent and effective responses majority of the world's population lives in urban centers. to urban water issues, in particular protracted droughts. Urban water usage has increased steadily, reflecting Programs are designed to modify the level and/or timing of more concentrated populations and intensified economic demands for water by encouraging changes in consumer activities in urban areas. Although the volume of water behavior through the appropriate water and sewerage dedicated to urban use is less than that used by agriculture disposal pricing systems, among other options. and other sectors, its social and economic importance is enormous. In addition, urban water use has high embedded The significant diversity of water uses in the urban energy content. commercial and public water use sectors means that there are not many studies available on the application During the coming decades, urban areas, especially those of economic water demand measures (price incentives) in developing countries, will experience the most rapid rates in these sectors. Renzetti (2002) mentions only three of population growth. As a result, urban residents must be studies on commercial water use: a mail survey of provided with a variety of water services, including water commercial establishments in Miami (USA); estimation supply, sewerage collection and treatment, and wastewater of aggregate water demand equations for the publicly- disposal. This is why the management of water demand in supplied commercial sector in a US urban area (where urban areas deserves particular attention. price elasticities varied from ­0.141 to ­0.360); and a similar econometric study based on commercial water Urban water usage is unique in its fragmentation, in terms use data from 16 communities in Ohio (USA) showing of both physical use of the resource and the institutional short- and long-run elasticities of ­0.234 and ­0.918, structures that govern that use. Water use covers a wide respectively. 43 Boland (1998) analyses the state of the art in the it was upgraded several times under the sponsorship of application of different tools in urban water use analysis and many institutions, mostly the Institute for Water Resources forecasting. He comments on the simple bivariate models, of the US Army Corps of Engineers. The currently per capita requirements method, unit use coefficient available IWR-MAIN model, Version 6, includes a benefit- method, multivariate methods with and without economic cost module for analyzing water demand management explanatory variables, and econometric demand models. measures. It also includes an up-to-date database on Concerning the econometric demand models he offers the residential water use, nonresidential water use, and end- following opinion: use parameters. The model IWR-MAIN software package is designed for "(1) translating demographic, housing, and ...So far, econometric models are available business statistics (for cities, counties or service areas) mostly for residential water use and detached into estimates of existing water demands; and (2) using single-family dwellings. Such models have been projections of populations, housing, and employment highly successful, and are able to predict water to derive baseline forecasts of water use. The forecast use under a wide range of circumstances... module disaggregates total urban water use into spatial, Examples of econometric analysis of demand are temporal, and sectoral components". Figure 3 summarizes ... the residential sector model(s) of IWR-MAIN. the inputs and outputs of the IWR-MAIN Water Demand Unfortunately, few models of this kind have been Analysis Software, whose complete description is developed for multi-family residential buildings, or presented in IWR-MAIN (1996). for nonresidential uses. Further research is needed before this approach can be used to forecast urban Several water agencies across the United States have water use in all sectors. applied this model to forecast water demand. To what extent this model, especially its knowledge base, can be The IWR-MAIN Water Demand Analysis Software, Version used outside of the United States is not entirely clear to the 6, is presented by Opitz et al. (1998). The original version author of this paper. However, there are some examples of that model was developed at the end of the 1960s and similar to the IWR-MAIN application for analysis of demand- Figure 3. IWR-MAIN Inputs and Outputs WATER DEMAND FORECAST Average daily Low-use season High-use season Maximum-day Housing Sewer contribution Employment Price WATER SAVINGS Income Housing density Passive conservation IWR-MAIN Active savings Weather Industrial productivity Model and Price impacts Knowledge Base Emergency savings Plumbing code Efficient end-uses BENEFITS AND COSTS Conservation programs Net present value Drought restrictions Benefit-cost ratio Discounted payback Utility cost structure Levelized cost Capacity needs Life-cycle revenue impact for External costs Utility Participants Ratepayers Community Society Source: Opitz et al. 1998. 44 driven water policies in Volos, Greece (Kolokytha and When considering the river basin scale of water demand Mylopoulos 2004). management, it is necessary to "shift the focus of analysis away from the estimation of water demands towards the To close this section, some comments should be made use of that information" (Renzetti 2002). That information on forecasting urban water demands. Billings and Jones should be used within the framework of river basin models (2007) present four groups of forecasting methods: designed to incorporate interactions between water supply (1) judgment-based subjective methods, (2) extrapolation, and the demands of different water users and uses, and (3) multivariate regression, and (4) nonparametric methods assess the significance of these interactions from the (e.g. neural networks and fuzzy logic). Their book is very perspective of water demand policy. much based on US practice, but the methodology is applicable worldwide. Together with the book by Baumann, A number of schematics have been proposed to represent Boland and Hanemann (1998), these two works provide the river basin models. They have been available to assist a most complete discussion and evaluation of urban water resources policy, planning, and management for water demand management, planning, and forecasting decades. In principle, such models are built following a methodologies. process that consists of several interrelated and highly iterative steps which, in principle, are similar from model to model. The modeling process is illustrated in Figure 4 by Water Demand Management at the using the scheme proposed by Rodrigo et al. (1995) for River Basin and National Levels integrated river basin planning. So far, this paper has discussed water demand management issues with reference to individual (residential, industrial, and agricultural) and aggregated (urban) water use. The situation is different and more complex as we move Figure 4. The Integrated River Basin up to the higher levels of aggregation. Many individuals Modeling Process and aggregated users interact at these levels to the extent that they share sources of water and sinks for wastewater Existing Demographic trends, Existing & disposal. In addition, the intake water users interact with water planned economic indicators and instream (non-extractive) uses, including water required supplies climate data conservation for the maintenance of aquatic ecosystems, navigation, hydroelectric power generation, recreation, and fish and Water demand wildlife uses. Those uses have not been discussed in this forecasts paper because none of them is priced directly, but it should Supply-side Demand-side planning planning be recognized that each of them has a specific value. Even Supply reliability in the absence of market-clearing prices, there are a number evaluation of ways to estimate the value of water in alternative uses. But this discussion, which is of fundamental importance for Marginal-cost Rate design the efficient allocation of resources, is beyond the scope evaluation of this paper. For an in-depth treatment of this subject, the interested reader is referred to the book by Gibbons Additional Additional (1986). Here, it should only be stated that at the level of water Integrated water water resources plan the river basin, water demand issues become part of the supplies conservation resource allocation problem, involving both the supply and demand sides of the water equation. Within the framework Financial plan of integrated water resources management, a wide range of supply and demand management options must be considered in the scale of the entire basin. Source: Rodrigo et al. 1995 45 The modeling steps are discussed by Beecher (1996). The hydro-economic models help water managers design, They begin with the identification of demographic trends, operate, and expand water resource systems efficiently economic indicators, and climatic data for evaluating and in accord with explicitly represented societal values existing and future water supply and demand alternatives. and priorities. The cross-fertilization of engineering and This step includes an evaluation of water supply reliability, economics allows more realistic representations in rate design, and analysis of marginal costs and benefits mathematical models of how water is managed in practice within the framework of demand-side planning. Building and how management could be improved. Hydro-economic the integrated water resources plan involves determining models are distinguished by a solution-oriented and its principal objectives, and developing specific criteria for integrated approach. The central idea of these models evaluating feasible alternatives. Selection of the alternative is that water demands are not fixed requirements, but that best satisfies the plan's objectives is enhanced by the functions where different quantities of water at different development of a financial plan. times have varying total and marginal values. In this approach water management is driven by the economic The river basin models in question are close to the value of water in addition to other requirements or regional models discussed by Renzetti (2002), employing priorities. Economic concepts used include: economic econometrically estimated demand equations, and a water demand, value of environmental services, consumer computable general equilibrium approach or programming surplus, willingness-to-pay, and supply-side economics. techniques. Examples of such models presented by this Hydro-economic models are built with diverse aims, author include one of regional water use that combines formulations, levels of integration, spatial and temporal linear programming models of irrigation water demand scales, and solution techniques. Policy insights and with an input-output model (I-O) of the economy of one management practices revealed by the application of these of the US states in order to explore the relationship models promote integrated water resources management. between water use and economic conditions. The linear Hydro-economic models go well beyond minimizing programming model provides estimates of the marginal costs and maximizing profits; they provide a common value of irrigation water in alternative applications. Taking framework through which the value of all water services into account future increases in water costs, the model can be considered and used to direct system planning and allows the estimation of future changes in water use, crop operation (Haron et al. 2008). patterns and farm income for the entire state. Another example, as an alternative to the I-O model, involves Several examples of the basin-wide applications of hydro- construction of an econometric model to consider the economic models are already available. For example, Guan impact of a doubling of irrigation water prices in the region. and Hubacek (2008) present an integrated hydro-economic Computable general equilibrium models are sometimes accounting and analytic framework developed for water applied to assess water demand policies and gain a more resources management in North China, based on I-O complete representation of the regional economy. modeling combined with a mass-balanced hydrological model. Another study (Gurluk and Ward 2008) reports In this context, a new generation of hydro-economic development of a hydro-economic model in Turkey to study river basin models seems to be especially attractive. problems resulting from increasing demands for water in the These models take account of the fact that economic Nilufer River basin, where important agricultural, tourist, and issues and processes are becoming increasingly industrial activities are located. The model is solved using integrated with more traditional engineering and the GAMS system and, according to the authors, could be hydrologic models of water resources management applied to other river basins worldwide. Hydro-economic (Heinz et al. 2007). Combining economic management models are also applied in Morocco (M'Barek et al. 2004). concepts and performance indicators with an Changing climatic, economic, and social conditions have engineering, hydrologic, and nature conservation major impacts on the availability of water resources and understanding of a water resources system can provide rural poverty in developing countries like Morocco. The results and insights more directly relevant for demand integrated model of the Draa valley is based on the hydro- management decisions and policies. economic river basin model developed at the International 46 Food Policy Research Institute (IFPRI). The paper by use under different physical, economic, and institutional Maneta et al. (2007) reports on the preliminary results of conditions. This calls for the initiation of new data collection research that aims to develop a detailed hydro-economic programs. In addition, one of the missing elements of great model for assessing the effects of alternative surface and significance for water demand relationships is the quality of groundwater policies in the Buriti Vermelho sub-catchment water withdrawn. area of the São Francisco River basin in Brazil. A spatially explicit, farm-level, mathematical programming model has As pointed out in this paper, the demand side of water been developed. The model is capable of accommodating a management deserves more attention than it has been given broad array of farm sizes and characteristics to predict the so far. To this end, models capable of producing demand effects of alternative water policies and neighbors' water functions for residential, industrial, agricultural, and urban use patterns on agricultural production. water are useful additions to the tools of water management analysts. During the last few decades the methodologies The next level of demand aggregation is the nation as used for estimating demand relationships have been a whole. However, at this level, practically all available improved significantly. It should be made clear, however, studies and statistics refer to national water use rather that these models still raise a good number of questions, than water demand. It should be recognized that according many of which remain unresolved. The field is still open to a to water availability and local conditions, water prices vast amount of research and investigation. and wastewater disposal charges will vary substantially and, except for very small nations, there is no such thing as a uniform "nationwide" price of water. Most previous References aggregated national water use forecasts significantly overstated actual water use. Baumann, Duane D., John J. Boland and W. Michael Hanemann. 1997. Urban Water Demand Management and Planning. New York: McGraw-Hill, Inc. Conclusions Beecher, J. A. 1996. Avoided Cost: an Essential Concept for Integrated Resources Planning. Water Resources When water is readily available, and thus relatively Update, No. 104. inexpensive, attempting to use it efficiently may not be a Billings, R. Bruce and Clive V. Jones. 2007. Forecasting critical issue. However, it does becomes a critical issue as Urban Water Demand, American Water Works water demands from different economic sector increase, as Association, Second Edition. the costs of supplying water increase, and as the availability Bithas, K. and S. Chrysostomos. 2006. Estimating of the resource decreases. Urban Residential Water Demand Determinants and Forecasting Water Demand for Athens Metropolitan The purpose of this paper was to contribute to gain a Area, 2000­2010. South-Eastern Europe Journal of better understanding of the current state of practice in the Economics 1, pp. 47­59. area of water demand management, including alternative Boland, J. 1998. Forecasting Urban Water Use: Theory and methodologies, techniques, and applications. The paper Principles. In Baumann, Boland and Hanemann, Urban explored various dimensions of water demand analysis, and Water Demand Management and Planning. New York: showed that economic theories provide a good conceptual McGraw-Hill, Inc. framework for modeling water demands and identifying Calloway, J. A., A. K. Schwartz, and R. G. Thompson. 1974. variables that influence those demands, one of them being Industrial Economic Model of Water. Water Resources water price. Research 10:650­8. Gibbons, D. C. 1986. The Economic Value of Water. There are two broad approaches to water demand Washington, D.C.: Resources for the Future. modeling: econometric and programming. The modeling Gouevsky, Ilia V. and David R. Maidment. 1984. Agricultural task, however, is difficult. One of the key problems is Water Demands. In J. Kindler and C. S. Russell (eds.), the scarcity of adequate databases concerning water Modeling Water Demands. New York: Academic Press. 47 Guan, D. and K. Hubacek. 2007. Assessment of Regional M'Barek, R., T. Schmidt and M. Koll. 2004. Hydro-economic Trade and Virtual Water Flows in China. Ecological Modeling in Southeast Morocco. Rural Poverty Economics, 61, 159­170. Reduction through Research for Development. Gurluk, Serkan and Frank A. Ward. 2008. Hydro-economic Deutscher Tropentag, Berlin, October 2004. Modeling for Water Resources Management: the Maneta, M., M. Torres, R. Howitt, S. A. Vosti, W. W. Nilufer Basin, Turkey. Third International Conference on Wallender and L. Bassoi. 2007. A Detailed Hydro- Water Observation and Information System for Decision economic Model for Assessing the Effects of Surface Support, 27­31 May, 2008, Ohrid, Macedonia. Water and Groundwater Policies: A Demonstration Hanemann, W. Michael. 1998. Determinants of Urban Model from Brazil. Annual Meeting of the Agricultural Water Use. In Baumann, Boland and Hanemann, Urban and Applied Economics Association. Portland, July. Water Demand Management and Planning. New York: Martinez-Espineira, R. 2005. An Estimation of Residential McGraw-Hill, Inc. Water Demand Using Co-Integration and Error Haron, J. J., M. Pulido-Velásquez, D. E. Rosenberg, J. Correction Techniques. MPRA Paper No. 615 and J.Medellín-Azuara, J.R. Lund. 2008. Poster at the 13th Journal of Applied Economics v:10. IWRA World Congress, Montpellier, France, 1­4 Opitz, E. M., J. F. Langowski, B. Dziegielewski, N. A. September. Hanna-Somers, J. S. Willett, and R. J. Hauer. 1998. Heinz I., M. Pulido-Velázquez, J. R. Lund, and J. Andreu. Forecasting Urban Water Use: Models an Application. 2007. Hydro-economic Modeling in River Basin In Bauman, Boland and Hanemann (eds). Urban Management: Implications and Applications for Water Demand Management and Planning. New York: European Water Framework Directive. Water McGraw-Hill, Inc. Resources Management, Vol. 21, No. 7, July. Renzetti, S. 2002. The Economics of Water Demands. Herrington, Paul R. 2006. The Economics of Water Demand INSERT City: Kluwer Academc Publishers. Management. In David Butler and Fayyaz Ali Memon Rodrigo, D., T. Blair, and B. Thomas. 1995. Integrated (eds.), Water Demand Management. London: IWA Resources Planning and Reliabilty Analysis: A Case Publishing. Study of the Metropolitan Water District of Southern Iglesias, E. and M. Blanco. 2008. New Directions in Water California (unpublished), mentioned in Baumann, Duane Resources Management: The Role of Water Pricing D., John J. Boland and W. Michael Hanemann (eds.). Policies. Water Resources Research, Vol. 44, 1­11. Urban Water Demand Management and Planning. International Water Management Institute (IWMI). 2007., New York: McGraw-Hill, Inc. Comprehensive Assessment of Water Management Rosegrant, Mark, Ximing Cai, Sarah Cline, and Naoko in Agriculture. London: Earthscan, and Colombo: Nakagawa. 2002. The Role of Rainfed Agriculture in the International Water Management Institute. Future of Global Food Production. EPTD Discussion IWR­MAIN. 1996. Version 6, Water Demand Analysis Paper No. 90, International Food Policy Research Software. User's Manual and System Description. Institute, Washington, D.C. Kindler, J, C. S. Russell (eds.). 1984. Modeling Water Sewell, W. R. D., and L. Roueche. 1974. The Potential Demands. New York: Academic Press. Impact of Peak Load Pricing on Urban Water Demands, Kindler, J. 1988. Modeling Derived Demand for Irrigation Victoria, B.C., A Case Study in Priorities in Water Water, Agricultural Water Management, 13, 403­410. Management, University of Victoria, B.C. Canada. Kolokytha, E. G. and N. A. Mylopoulos. 2004. Evaluating Stone, J. C. and D. Whittington. 1984. Industrial Water Demand Management Aspects of Urban Water Policy. Demands. In J. Kindler and C. S. Russell (eds.), The City of Volos Case, Greece. Water, Air & Soil Modeling Water Demands. New York: Academic Press. Pollution, Vol. 4, Nos. 4­5, October. Tsur, Y. 2000. Water Regulation Via Pricing: The Role of Liu, J., H. H. G.Sanije, and J. Xu. 2003. Water as an Implementation Costs and Asymmetric Information. In Economic Good and Water Tariff Design Comparison A. Dinar (ed.), The Political Economy of Water Pricing Between IBT-con and IRT-cap. Physics and Chemistry Reforms. New York: Oxford University Press. pp. of the Earth 28, 209­217. 105­120. 48 Tsur, Y. and A. Dinar. 1997. On the Relative Efficiency of Water Demand Management and Planning. New York: Alternative Methods for Pricing Irrigation Water and McGraw-Hill, Inc. Their Implementation. World Bank Econ. Rev., 11, Winpenny, J. 1994. Managing Water as an Economic 243­262. Resource. Development Policies Studies. London: Vairavamoorthy, K and M. A. Mohamed Mansoor. 2006. Routledge and Overseas Development Institute. Demand Management in Developing Countries. In World Bank. 1994, Water Resources Management, A David Butler and Fayyaz Ali Memon (eds.), Water World Bank Policy Paper, Washington, D.C. Demand Management. London: IWA Publishing. World Bank Water Research Team. 1993. The Demand White, Gilbert F. 2006. Foreword. In Duane D. Baumann, for Water in Rural Areas: Determinants and Policy John J. Boland and W. Michael Hanemann, Urban Implications. World Bank Research Observer 8 (1). 49 3. Managing Droughts and Floods under Changing Environments Juan B. Valdés1 SAHRA2 and the University of Arizona Abstract The consequences of these dramatic changes are of great importance, affecting agriculture, energy, the water supply, According to reports from the Intergovernmental Panel wildlife, and nearly every aspect of human societies. on Climate Change (IPCC), climate change will intensify the hydrologic cycle, making extreme events like floods From a hydrology and water resources management and droughts more frequent and intense. This paper perspective, changes in the variability of hydroclimatic discusses efforts to characterize floods and droughts, factors are even more important than changes in their mean the use of paleoclimatic data to increase the instrumental values. The IPCC reports indicate that climate change will record length, the use of climate projections from the most result in an intensification of variation in the hydrologic cycle recent runs of the global circulation models (GCMs) for (see Figure 1). hydrologic applications, and the implication of climate variability and change in the management of water Climate projections, for instance, show that droughts will resources systems. be more frequent and more intense in southern Europe, particularly in the Iberian Peninsula, a region of high economic activity (Figure 2). Introduction This paper reviews some critical aspects for the discussion The IPCC's 2007 assessment (IPCC-AR4 2007) indicates of how to approach the management of water resources that anthropogenic climate change will result in significant under extreme hydrologic events in a changing climate. challenges for water resources systems. On the one hand, higher mean temperatures, more frequent, longer-lasting heat waves, and increased summer dryness in most parts of Characterization and Forecasting the northern, middle, and high latitudes will raise the risk of droughts. On the other hand, there will also be an increased In order to devise appropriate water resources management chance of intense precipitation and flooding caused by the plans, hydrologic extremes--floods and droughts--need greater water holding capacity of a warmer atmosphere (Frei to be first characterized in their magnitude and recurrence. et al. 1998). Consequently, precipitation events will tend Floods tend to be easier to characterize and there is to be more intense with longer drier periods in between. ample scientific literature on determining their magnitude 1. Julio Cañón, Francina Domínguez and Aleix Serrat-Capdevila of SAHRA and Javier González from the University of Castilla La Mancha (Spain) provided significant input to the writing of this article. Mary Black of SAHRA provided significant editing that greatly improved this manuscript. All contributions are greatly acknowledged by the author. 2. National Science Foundation Science and Technology Center for Sustainability of Semi-Arid Hydrology and Riparian Areas (SAHRA). 51 Figure 1. Variations in the Hydrologic Cycle Return period (years) Return period (years) 1 2 5 10 20 50 100 200 5000 1000 1 2 5 10 20 50 100 200 5000 1000 600 600 Higher mean Higher mean 500 500 Discharge (1000 cfs) Discharge (1000 cfs) 400 400 300 Lower mean 300 Lower mean 200 200 100 100 0 0 1 0.5 0.2 0.1 0.05 0.02 0.01 0.005 0.002 0.001 1 0.5 0.2 0.1 0.05 0.02 0.01 0.005 0.002 0.001 Probability of exceedance Probability of exceedance Source: SAHRA Note: The scenario at left shows a shift in the mean above or below current values. The scenario at right shows minimums decreasing and maximums increasing, in accordance with IPCC projections. Figure 2. Change in the Recurrence of 100-year Droughts Source: IPCC 2007. Note: Based on comparisons between climate and water use in 1961 to 1990 and simulations for the 2020s and 2070s (based on the ECHAM4 and HadCM3 GCMs, the IS92a emissions scenario and a business-as-usual water-use scenario). Values calculated with the model WaterGAP 2.1 (Lehner et al. 2005b). 52 and frequency. This is not the case for droughts, whose proposed an alternative measure, the Drought Frequency characterization will be emphasized here. Index (DFI). The DFI is a stochastic index that characterizes a persistent deviation of a variable (that is, precipitation) Characterization of Droughts toward the lower tail of its probability density function. The result is a retrospective, sequential measure of the Droughts are difficult to typify because they have multiple persistence of low values evaluated by their mean frequency statistical signatures (that is, magnitude, duration, frequency, of occurrence. Therefore, the index offers an integrated intensity/peak). In addition, they tend to extend over large measure of the severity and duration of a drought in each areas, and long records are needed to appropriately time step relative to its probability of occurrence, expressed characterize them. According to the context, droughts may as a mean return period. be considered meteorological, hydrological, or agricultural hazards. Huschke (1959) provides a definition of drought Figure 3 summarizes the procedure to calculate the DFI for that captures most of the aspects of these extreme events. a particular hydroclimatic signal X (for example, precipitation, He states that draughts are "A period of abnormally dry streamflow, or soil moisture), which realization produces a weather sufficiently prolonged for the lack of water to cause time series X0, X1,..., Xi. Since droughts are often described serious hydrologic imbalance in the affected area." not as isolated realizations but as sequences of persistent low values, the algorithm looks for the sequence of length Several indices are used to describe droughts. The most W, that ends at the evaluated time step and produces the widely used are the Palmer Drought Severity Index (PDSI) largest extreme persistent function (EPF). This sequence (Palmer 1965) and the Standardized Precipitation Index represents the worst drought situation that has taken place (SPI) (McKee et al. 1993). The strengths and limitations up to that moment, based on the chosen criteria. of these indices have been reported in the literature (for example Alley 1984). To address the multiple characteristics The EPF is solved through numerical integration of a drought in a single index, González and Valdés (2006) procedures. After locating the sequence that produces the Figure 3. Schematic Representation of DFI Calculation Procedure 1) Look for the sequence that produces 4) Estimate mean return period (DFI) for current EPF sequence the largest EPF up to time step t [the worst drought up to time t]. P DFI t t 3) Cumulative function of persistence of extreme wi = analyzed sequence conditions (CEPF), i.e., p (EPFi Continental United States. J. Climate, Vol. 12, no. 7, Joseph, R., and S. Nigam. 2006. ENSO Evolution and 1145­1162. Teleconnections in IPCCs Twentieth-century Climate Daly, C., R. P. Neilson, and D. L. Phillips. 1994. A Statistical- Simulations: Realistic Representation? J. Climate, 19: topographic Model for Mapping Climatological 4360­4377. Precipitation Over Mountainous Terrain, J. Appl. Kidson, J. W., and C. S. Thompson. 1998. A Comparison of Meteor., 33: 140­158. Statistical and Model-based Downscaling Techniques Dominguez, F., J. Cañon, and J. Valdés. 2008. IPCC-AR4 for Estimating Local Climate Variations, J. Climate, Climate Simulations for the Southwestern US: The 11(4): 735­753. Importance of Future ENSO Projections, submitted to Kim, T., J. B. Valdés, B. Nijssen, and D. Roncayolo. 2006. Climatic Chang. Quantification of Linkages Between Large-Scale Fowler, H., S. Blenkinsop, and S. Tebaldi. 2007. Linking Climatic Patterns and Precipitation in the Colorado Climate Change Modeling to Impacts Studies: Recent River Basin. Journal of Hydrology, 321, 137­186. Advances in Downscaling Techniques for Hydrological Knowles, N., M.D. Dettinger, D. R. Cayan, 2006: Trends in Modeling, Int. J. Climatol, 27: 1547­1578. Snowfall versus Rainfall in the Western United States, Frei, C., C. Schär, D. Lüthi, and H.C. Davies. 1998. Heavy J. Climate, 19 (18): 4545­4559. Precipitation Processes in a Warmer Climate, Geophys. Lansey, Kevin J., 2008. Personal communication. Dept. of Res. Lett., 25(9): 1431­1434. Civil Engineering, University of Arizona. Tucson, AZ. Giorgi, F., and L. Mearns, 2002: Calculation of Average, Lehner, B., P. Doll, J. Alcamo, T. Henrichs and F. Kaspar, Uncertainty Range, and Reliability of Regional Climate 2006: Estimating the Impact of Global Change on from AOGCM Simulations Via the "Reliability Ensemble Flood and Drought Risks in Europe, Climatic Change, Averaging" (REA) Method, J. Climate, 15: 1141­1158. 75(3):273­299. 67 Lins, H.F. and D. Wolock, 2008. Personal communication. Systems: Application to the Amistad and Falcon U.S. Dept. of the Interior, U.S. Geological Survey. System (Lower Rio Grande/RioBravo). Water Reston, VA. Resources Management, August 20. Liu, Z., J. B. Valdés, and D. Entekhabi. 1998. Merging and Serrat-Capdevila, A., J. B. Valdés, J. Gonzalez, K. Baird, Error Analysis of Regional Hydrometeorologic Anomaly L. J. Mata, and T. Maddock. 2007. Modeling Climate Forecasts Conditioned on Climate Precursors, Water Change Impacts­and Uncertainty--on the Hydrology Resour. Res., 34(8): 1959­1969. of a Riparian System: The San Pedro Basin (Arizona/ Magistro J., and M. Lo. 2001. Historical and Human Sonora). J. Hydrol., 347: 48­66, 2007. Dimensions of Climate Variability and Water Resource Serrat-Capdevila, A., J. B. Valdés. 2008. From GCM Constraint in the Senegal River Valley. Clim Res., 19: Projections to DSS Models: A Holistic Assessment 133­147. of Climate Change Impacts in the San Pedro Basin McKee, T. B., N. J. Doesken, and J. Kleist. 1993. The (Arizona/Sonora). (To be submitted to the Journal of Relationship of Drought Frequency and Duration to Hydrology). Time Scales, Preprints, 8th Conference on Applied Sheppard, C. R. C., M. Spalding, C. Bradshaw, and S. Climatology, Jan. 17­22, Anaheim, California, pp. Wilson. 2002. Erosion vs. Recovery of Coral Reefs 179­184. After 1998 El Niño: Chagos Reefs, Indian Ocean. Meko, D. M., M. D. Therrell, C. H. Baisan, and M. K. Hughes. Ambio 31, 40­48. 2001. Sacramento River Flow Reconstructed to A.D. Trenberth, K. E. 1997. The Definition of El Niño. Bull. Amer. 869 from Tree Rings. J. Amer. Water Resour. Assoc., Met. Soc., 78: 2771­2777. 37(4): 1029­1039. VanOldenborgh, G. J., S. Y. Philips, and M. Collins. 2005. Milly, P., K. Dunne, and A. Vecchia. 2005. Global Pattern El Niño in a Changing Climate: A Multi Model Study. of Trends in Streamflow and Water Availability in a Ocean Science, 1, 81­95. Changing Climate. Nature, 438, 347­350. Wilby, R. L., L. E. Hay, W. J. Gutowski (Jr.), R. W. Arritt, E. S. Milly, P. C. D., J. Betancourt, M. Falkenmark, R. M. Takle, Z. Pan, G. H. Leavesley, and M. P. Clark. 2000. Hirsch, Z.W. Kundzewicz, D. P. Lettenmaier, R. J. Hydrologic Responses to Dynamically and Statistically Stouffer. 2008. Stationarity Is Dead: Whither Water Downscaled General Circulation Model Output. Management? Science Vol. 319 1, February. Geophys. Res. Lett., 27: 1199­1202. Palmer, W. C. 1965. Meteorological Drought. Research Wilby, R. L., and T. M. L. Wigley. 1997. Downscaling General Paper No. 45, U. S. Department of Commerce, Circulation Model Output: A Review of Methods and Weather Bureau, Washington, D. C. Limitations. Progress Phys. Geog., 21: 530­548. Philip, S. Y. and G. J. van Oldenborgh. 2006. Wood, A. W., E. Maurer, A. Kumar, and D. Lettenmaier. Shifts in ENSO Coupling Processes Under 2002. Long Range Experimental Hydrologic Global Warming. Geophys. Res. Lett., 33, doi: Forecasting for the Eastern U.S. J. Geophys. Res., 10.1029/2006GL026196. 107(D20):4429. Seo, Dong-Jun, 2006. Personal communication. Hydrologic Xu, C. Y. 1999. From GCMs to River Flow: A Review Research Laboratory, National Weather Service/NOAA. of Downscaling Methods and Hydrologic Modeling Silver Spring, MD. Approaches. Progress in Physical Geography, 23(2), Serrat-Capdevila, A and J. B. Valdés. 2006. An Alternative 229­249. Approach to the Operation of Multinational Reservoir 68 4. Sediment Transport and Deposition in Rivers: The Case for Non-Stationarity Robert H. Meade1 Abstract otherwise, was only waiting to be discovered has led to frustration on the part of working scientists and, perhaps This paper focuses on the significance of stationarity more tragic, to the neglect and decay of comprehensive for sediment studies in rivers and how this concept has monitoring programs. worked against monitoring programs by implying that predictive models could substitute for measurements. A The late Carl Nordin, who mentored several members of differentiation between intrinsic non-stationarity and non- the HEF Expert Panel, used to emphasize that hydrology stationarity of measurement and purpose is discussed, is a historical science. That is, that the foundations of our based on examples from the Colorado, the Amazon and science are the historical data that have been collected other rivers. routinely and carefully over long time periods at selected observation points around the world, and that the quality As a conclusion it welcomes non-stationarity and calls for of our science is grounded in the quality of our historical more continuity in the methods, not only with the hope data. During the last several decades, however, the that the lessons learned from one project might be applied presumption of stationarity has been part of the thinking with more quantitative assurance to the next, but that the that has led water managers and research administrators collection of reliable data continues after the design and to neglect the traditional routine collection of data. It has construction of projects, in order to better evaluate their also led them to urge working scientists to concentrate impacts and consequences fifty years from now. their efforts on (1) finding faster and cheaper ways of getting the minimal amount of data needed to fulfill a perceived specific need, and (2) devising a workable Introduction model that can be expanded and improved until the day when we no longer will have to spend so much time and The "news" that stationarity is dead (Milly et al. 2008) money on monitoring. may come as a relief to many fluvial sedimentologists. Stationarity, as a concept and as an underlying assumption This discussion will follow the concept of non-stationarity of predictive hydrologic assessment, has been of only of fluvial sedimentation in three directions: (1) the intrinsic limited use in studies and assessments of riverine sediment. non-stationarity of river sediment, (2) non-stationarity of And the presumption, by managers and supervisors, that measurement, and (3) non-stationarity of investigative a workable model of riverine sedimentation, stochastic or purpose. 1. Opinions expressed are those of the author, and do not necessarily reflect official policies and attitudes of the U.S. Geological Survey, with which the author (now Emeritus) has enjoyed a long (53-year) and productive association. 69 Stationarity in Riverine Sedimentation transport models have been devised, and an enormous scientific literature has accumulated. These studies have Before plunging into the discussions of non-stationarity, been of practical value in understanding the morphology let us begin with instances and examples where the and stability of river channels (most of which are floored assumption of stationarity has been useful in sediment by sand or gravel), and in dealing with associated practical studies. It has been useful in two places: (1) long-term problems such as scouring around bridge piers. But records in un-engineered rivers that flow through stable (or because most of the load of sediment carried by rivers at least, stabilized) landscapes, and (2) sand transport, for consists of finer and more cohesive materials (silt, clay, which the movement of particles can be predicted from the organic particles, and organic aggregates) these studies physics of fluid flow. have been of limited use to engineers and managers who need reliable predictions of total sediment fluxes. Long-term records of sediment transport are not commonly available in un-engineered rivers that drain stabilized landscapes because the expense of data collection Intrinsic Non-stationarity of Riverine is difficult to support in streams that are not seen as Sedimentation problematical. The few examples we do have of such data--based on sufficiently intensive samplings (daily) Most of the short-term non-stationarity that is visible in river during a sufficiently long period of record (decades)-- sediment records has been caused by humans (Meade 1969 demonstrate (in temperate humid areas, at least) such and 1996; Syvitski et al. 2005). In the first place, sediment stationarity generalities as (1) 90 percent of an average records tend to be concentrated in river basins where annual sediment load is transported in only 10 percent sediment is viewed as an actual or potential problem. So it is of the time, and (2) during the infrequent high-intensity almost inevitable that the records we do have will be strongly event (such as a hurricane-induced flood), the river can be biased toward non-stationarity. The underlying dilemma of expected to transport more sediment in a few days than it such records is this: If the record is of long-enough duration had transported during the previous several years (Meade to provide clear insights into the scope of short-term et al. 1990, Table 1 and Figure 3). More widely used (seasonal, year-to-year) variations in sediment transport, then (mainly because their construction requires fewer years of its integrity is likely to have been compromised by long-term representative sediment data) are sediment-rating curves, influences on sedimentation such as changes in land use in in which suspended sediment (either as concentration, in the drainage basin, or the construction of dams and other milligrams per liter, or as sediment discharge, in tons per engineering works in the river channel. day) is plotted against water discharge (in cubic meters per second) (Meade 1982, Figure 2; Meade et al. 1990, The effects of changing land uses (deforestation, agriculture, Figs. 2 and 4; Nordin et al. 1994, p. 252). This conventional mining, urbanization) usually begin as years-to-decades procedure allows an investigator to synthesize a long-term increases in riverine sedimentation that eventually taper off record of sediment discharge by combining a short record more gradually at longer time scales of decades to centuries. of sediment observations with a much longer record of A specific example from the Piedmont of Maryland (USA) is water-discharge measurements. So long as the sediment the diagram first published by Wolman (1967, p. 368), which regimes retain some reasonable semblance of stationarity, shows how, between the years 1800 and 2000, sediment sediment-rating curves can be useful and practical yields increased as the original forests were converted predictors of sediment loads in rivers. to croplands, decreased again as farms were abandoned and the lands reverted to woodlands and pasture, then Lured by the certainties of Newtonian physics and the spiked abruptly as the lands were disrupted by the intensive predictability of the effects of fluid forces on non-cohesive construction of suburban housing, and decreased again as sediment particles, many fluvial sedimentologists have roads were paved and lawns were planted. restricted their efforts to the study of the transport and deposition of sands and gravels. Many experimental studies More immediate changes usually result from the have been made (usually in laboratory flumes), many construction of engineering works. Because most river 70 engineering works are placed in channels or on adjacent patterns of scour and deposition in the lower 700 riparian lands, and because they are specifically designed kilometers of main channel and proximal floodplain that can to alter the existing hydraulics of flow, then one should be related to different operation routines at Sanmenxia Dam expect them to produce the most pronounced effects on and Reservoir (Zhao et al. 1987). Fifty-year-long records of river sediment transport and deposition. Dams, large and declining sediment discharges in the lower Yangtze River small, produce the most abrupt effects; and dams are have been analyzed to discriminate the effects of dam ubiquitous on six of the seven continents (that is, excluding construction and reforestation in the Yangtze basin from only Antarctica). Vörösmarty et al. (2003) estimate (1) that the effects of progressive climate change (Xu et al. 2007). more than 40 percent of global riverine water discharge is In the United States, the U.S. Bureau of Reclamation interrupted locally by large reservoirs, (2) that approximately record of sediment discharge in the Colorado River near 45,000 dam-impounded reservoirs trap 25­30 percent of the border with Mexico (Yuma, Arizona 1911­1979) the total sediment being carried seaward by the world's shows a large variation in annual fluxes (between 100 and rivers and streams, and (3) that some 800,000 smaller 300 million tons per year) before 1930, and the abrupt impoundments worldwide have an "additional but unknown decrease that followed the closure in 1933 of the Hoover impact." Since these impacts (of dams as well as those of Dam, 500 kilometers upriver (Meade and Parker 1985, other engineering activities in rivers) have been incurred Figure 29; Meade et al. 1990, Figure 12). Concerning over many decades, we can find only a very few instances the Rio Grande of the southwestern United States and in which the data have been sufficient to document the long- northeastern Mexico, records collected at six stations over term effects on sediment loads. several decades by the U.S. Geological Survey and the International Boundary and Water Commission show the The data that show these impacts usually are presented in downriver changes that followed the closures of four dams: three ways: (1) as paired maps that compare river sediment Elephant Butte in 1915, Falcon in 1953, Amistad in 1969, loads before and after engineering works were installed, and Cochiti in 1974 (Meade and Parker 1985, Figure 28). (2) as sediment-rating curves (before-and-after graphs of And a half century of consistent historical record at stations river sediment concentrations or tonnages versus water along the Mississippi River has provided insights into the discharge at fixed stations downriver of dams and other impacts of extensive river engineering, beginning with the engineering works, or constructed from measurements closure of major dams during the 1950s on the Missouri made upstream and downstream of major impoundments), River (the principal source of sediment to the Mississippi) and (3) as historical time series, taken from records of and continuing through the completion of other works such sediment monitoring stations that have been operated as river-training structures and bank revetments (Meade consistently for periods measurable in decades. Examples and Moody 2008, 2009). of the paired-map form portrayals of sediment discharges are those showing the impacts of dams on rivers of the southeastern United States (Meade and Parker 1985, Non-stationarity of Measurement Figure 30; Meade et al. 1990, Figure 14) and those showing pre-engineering and post-engineering sediment The same dilemma mentioned above--that any sediment discharges in the Mississippi River system (Meade 1995, record long enough to provide insights into the ranges Figure 6A). Examples of before-and-after sediment-rating of short-term variations is likely to be long enough to curves are those for the Roanoke River of North Carolina include the effects of long-term changes that reflect (Meade 1982, Figure 10) and the lower Mississippi River non-stationarity--is also applicable to the techniques (Meade and Moody 2008 Figure 6, 2009 Figure 6). and strategies of the measurement of riverine sediment. Sampling equipment and techniques for the collection of Examples of historical time series are fairly rare, and most of sediment data have evolved over decades, and sometimes the long-term (multi-decadal) sets of continuous historical the changes have been applied with revolutionary data on riverine sediment loads have been collected in suddenness. Likewise, the strategies for computing such either China or the United States. Multi-decadal records of things as total annual sediment loads from the collected and data from China's Yellow River portray different longitudinal analyzed data have changed over the years. Non-stationarity 71 is introduced when such changes are enacted without of the Orinoco and Amazon rivers (Meade 1985; Richey et proper calibrations between the old and the new. al. 1986) was later used, with equal success, in a 5-year study of sediment-borne contaminants in the Mississippi A Case in Point: The Colorado River at the River (Meade and Stevens 1990; Meade et al. 1995). More Grand Canyon recently, the Federal Interagency Sedimentation Project has developed collapsible-bag samplers of more streamlined During the mid-1940s, following nearly 20 years of regular design that have been tested in a laboratory flume for their sediment sampling using the Colorado River Sampler (a sampling characteristics (Davis 2001 and 2006; McGregor vertical bottle that was opened on the river bottom and 2006). These samplers probably represent the optimal quickly drawn up to the river surface), the equipment choices for present and future studies of suspended was changed to a depth-integrating sampler with a sediment in large rivers. horizontally-aligned isokinetic nozzle that admitted water and suspended sediment at ambient velocities. Subsequent These are perilous times in the history of sediment sampling perusals of the ensuing sediment records led to the in rivers. The time-tested methods of sampling (isokinetic observation that the mid-1940s were the beginning years depth-integrating and point-integrating sampling, and the of a drastic reduction (by half) of sediment discharges related field processing and laboratory analysis of sampled in the Colorado River, which investigators attributed to materials) have become prohibitively (in the eyes of water improvements in rangeland grazing practices (Hadley 1974) managers) expensive and time consuming, and the search and to regional climate change (Graf et al. 1991). These is on for cheaper surrogate methods. Surrogate methods interpretations went largely unchallenged because (1) a include devices that operate on such principles as those of perfunctory calibration study had been made at the time the bulk optics (turbidity), laser optics, pressure difference, and samplers were changed and (2) the mid-1940s were the acoustic backscatter (Gray et al. 2002; Gray and Gartner beginning years of a prolonged drought in the southwestern 2009). The Acoustic Doppler Current Profiler (ADCP) has United States. However, a more thorough calibration also been applied to the estimation of suspended sediment study in the Colorado River (using the old Colorado (Filizola and Guyot 2004), but this application cannot yet be River Sampler and a more recent isokinetic sampler) has considered quantitative because the necessary calibrations confirmed that most of the mid-1940s "reduction" in the between acoustic backscatter and suspended sediment suspended-sediment discharge of the Colorado River concentration have not been made (Dinehart and Burau was merely an artifact of the change in sediment samplers 2005; Gamaro 2008). (Topping et al. 1996). Any project that adopts any of these surrogate methods After the general adoption in the United States of isokinetic should realize that the adoption process entails a samplers in the late 1940s and early 1950s, sediment responsibility for a thorough calibration with the older and sampling techniques settled into routines that allowed more established methods. Moreover, these calibrations for the collection, over periods of several decades, of need to be continued over the years. All rivers are different procedurally-consistent data sets. Manuals were produced (Schumm 2005); therefore, even if the first investigator to not only for field methods, but also for laboratory procedures apply a new surrogate method has done a calibration, we and for computational methods (Guy 1969; Porterfield cannot assume that all subsequent investigators do not also 1972; Guy and Norman 1982; Edwards and Glysson 1999; have to undertake one. Suspended sediment particles can see also Carvalho 2008). be expected to differ in their properties (such as grain size, surface area, optical reflectance, acoustical reflectance, Sampler technology continued to improve to meet newly aggregation state) from one river to the next (or from one perceived needs. For collecting suspended sediment in season to the next in the same river). These are the very large rivers, collapsible-bag samplers were developed to properties that the surrogate methods use as measures avoid the air-pressure-compensation difficulties in using of sediment concentration. Furthermore, the presence or standard samplers at great river depths. An experimental absence of organic aggregates and organic detritus (such model that was used successfully in comprehensive studies as twigs and leaves) can complicate the calibration process 72 to a significant degree. One should anticipate that every parameter has been shifted from mass per unit time river will have to be calibrated anew, and that separate (sediment discharge) to mass per unit volume (sediment calibrations may well be needed at different sampling sites concentration). The advantage of concentration over and at different seasons on the same river. discharge is that it is more readily measured and more easily enforced, and therefore of more immediate interest In the long term, it may make more economic sense to to the Environmental Protection Agency, which considers continue to collect samples and make direct measurements suspended sediment to be a major pollutant on par with of suspended sediment, rather than to avidly pursue each nutrients such as nitrate. Many of the recent assessments newly introduced "magic bullet." The known uncertainties view suspended sediment through this lens (Gray et al. of reliable direct measurements are preferable to the much 2000; Langland et al. 2001; Blevins 2006; Sprague et al. greater uncertainties that surrogate methods cannot avoid. 2007). Diametrically at odds with those who consider sediment Non-stationarity of Purpose as a pollutant are those who see sediment as a necessary resource in the maintenance and restoration of riparian There is no firm consensus regarding whether sediment and coastal floodplains and wetlands. Exchanges of is a liability or an asset in river systems. Depending on sediment between river channels and their floodplains can one's outlook and purpose, sediment may be viewed be highly significant in un-engineered rivers. In a 1,500 as a potential liability in the design of reservoirs, in the kilometer reach of the Brazilian Amazon, for example, the maintenance of channels for navigation, as an unfortunate quantity of sediment transferred between the channel result of poor soil conservation, as a conveyor of adsorbed and the floodplain exceeds the quantity of sediment pollutants, and as a threat to the habitats of aquatic transported out of the reach by the channel itself (Dunne species. Likewise, one may consider sediment to be an et al. 1998; Meade 2007). Much of the decline in the asset because it is the foundation material of which rivers area of the coastal wetlands of the Mississippi River construct their channels and floodplains. Sediment also delta in Louisiana has been attributed to the decline in transfers useful nutrients and soil onto riparian agricultural the delivery of sediment by the river (Blum and Roberts lands, sequesters adsorbed contaminants, and restores 2009). The data needed for assessments such as riparian and coastal wetlands. Consequently, as the times these range from traditional measurements of sediment change, so do the lenses through which riparian societies tonnages to remote sensing analyses of the aerial extents view riverine sediment. of the wetlands involved. Major programs in the monitoring of riverine sediment began in the United States in conjunction with the design of major Conclusions dams. Elephant Butte Dam on the Rio Grande and Hoover Dam on the Colorado River were among the first. After Milly et al. (2008) noted that "In a non-stationary world, the Second World War, massive data collection programs continuity of observations is critical." As investigators in were undertaken on the Rio Grande and the Missouri River. fluvial sedimentology, we have little influence on continuity The emphasis was on data for reservoir design, and the of societal purpose or even on the course of major events in basic question was: How many years will we be able to the control, maintenance, and restoration of rivers. But we operate this reservoir before the river is able to fill it with can strive for more continuity in our methods, not only with sediment? Data required for making this assessment were the hope that the lessons we learn from one project might the measured tonnages of transported sediment and the be applied with more quantitative assurance to the next, calculated volumes that they would occupy once they were but that we might continue to collect reliable data after the deposited. design and construction of projects so as to better evaluate their consequences. The rivers that the World Bank builds In more recent decades, emphasis has shifted to the role dams across this year may be the same rivers it is asked to of suspended sediment in water quality, and the relevant help restore fifty years from now. 73 References Engenharia de Sedimentos, Campo Grande, Brazil, 2­8 November 2008. Blevins, D.W. 2006. The Response of Suspended Graf, J.B., Webb, R.H., and Hereford, R., 1991, Relation Sediment, Turbidity, and Velocity to Historical of sediment load and flood-plain formation to climatic Alterations of the Missouri River. U.S. Geological variability, Paria River drainage basin, Utah and Arizona: Survey Circular 1301. http://pubs.usgs.gov/ Geological Society of America Bulletin, v. 103, p. circ/2006/1301 1405­1415. Blum, M.D., and Roberts, H.H., 2009, Drowning of the Gray, J.R., and Gartner, J.W., 2009, Technological advances Mississippi Delta due to insufficient sediment supply in suspended-sediment surrogate monitoring. Water and global sea-level rise: Nature GeoScience, v. 2, p. Resources Research, Vol. 45, W00D29, Special issue 488­491. on Advances in Monitoring in Hydrology. Carvalho, N. O. 2008. Hidrosedimentologia Prática (2d Gray, J.R., Glysson, G.D., and Mueller, D.S., 2002, edition). Rio de Janeiro: Editora Interciência. Comparability and accuracy of fluvial-sediment data--a Davis, B.E. 2001. The US D-96: An Isokinetic Suspended- view from the U.S. Geological Survey: American Society sediment/Water-quality Collapsible-bag Sampler. [U.S.] of Civil Engineers Specialty Conference, Hydraulic Federal Interagency Sedimentation Project Report PP. Measurements & Experimental Methods, Estes Park, http://fisp.wes.army.mil/Report%20PP,%20US%20 Colorado, July 28-August 1, 2002, Proceedings. http:// D-96%20011114%20.pdf water.usgs.gov/osw/techniques/asce.pdf Davis, B. 2006. FISP's Suite of Federally Approved Gray, J.R., Glysson, G.D., Turcios, L.M., and Schwarz, Suspended-Sediment/Water-Quality Collapsible- G.E., 2000, Comparability of suspended-sediment Bag Samplers. [U.S.] Federal Interagency concentration and total suspended solids data: U.S. Sedimentation Conference, 8th, Reno, Nevada, Geological Survey Water-Resources Investigations April 2­6, 2006, Proceedings http://pubs.usgs.gov/ Report 00­4191, 14 p. http://pubs.usgs.gov/wri/ misc_reports/FISC_1947­2006/pdf/1st-7thFISCs- wri004191/ CD/8thFISC/8thFISC.pdf Guy, H.P., 1969, Laboratory theory and methods for Dinehart, R. L., and J. R. Burau. 2005. Repeated surveys by sediment analysis: U.S. Geological Survey Techniques acoustic Doppler current profiler for flow and sediment of Water-Resources Investigations, book 5, chapter dynamics in a tidal river. Journal of Hydrology, v. 314, C1, 58 p. p. 1­21. Guy, H.P., and Norman, V.W., 1982, Field methods for Dunne, T., Mertes, L.A.K., Meade, R.H., Richey, J.E., and measurement of fluvial sediment: U.S. Geological Forsberg, B.R., 1998, Exchanges of sediment between Survey Techniques of Water-Resources Investigations, the flood plain and channel of the Amazon River in book 3, chapter C2, 59 p. Brazil: Geological Society of America Bulletin, p. 110, Hadley, R.F., 1974, Sediment yield and land use in p. 450­467. southwest United States, in Effects of Man on the Edwards, T.K., and Glysson, G.D., 1999, Field methods for Interface of the Hydrological Cycle with the Physical measurement of fluvial sediment (2nd edition): U.S. Environment: International Association of Hydrological Geological Survey Techniques of Water-Resources Sciences Publication 113, p. 96­98. Investigations, book 3, chapter C2, 89 p. http://pubs. Langland, M.J., Edwards, R.E., Sprague, L.A., and Yochum, usgs.gov/twri/twri3-c2/ S., 2001, Summary of trends and status analysis for Filizola, N., and Guyot, J.L., 2004, The use of Doppler flow, nutrients, and sediments at selected nontidal sites, technology for suspended sediment discharge Chesapeake Bay basin, 1985­1999: U.S. Geological determination in the River Amazon: Hydrological Survey Open-File Report 01­73, 49 p. http://pa.water. Sciences Journal, v. 49, p. 143­153. usgs.gov/reports/ofr01­73.pdf Gamaro, P.E.M., 2008, Verdades e mitos sobre medição de McGregor, J., 2006, The US D-99: An isokinetic depth- sedimento em suspensão com medidores acústicos integrating collapsible-bag suspended-sediment Doppler: Proceedings of VIII Encontro Nacional de sampler: [U.S.] Federal Interagency Sedimentation 74 Project Report RR, 19 p. http://fisp.wes.army.mil/ Meade, R.H., and Stevens, H.H., Jr., 1990, Strategies Report%20RR.pdf and equipment for sampling suspended sediment Meade, R.H., 1969, Errors in using modern stream-load and associated toxic chemicals in large rivers--with data to estimate natural rates of denudation: Geological emphasis on the Mississippi River: The Science of the Society of America Bulletin, v. 80, p. 1265­1274. Total Environment, v. 97/98, p. 125­135. Meade, R.H., 1982, Sources, sinks, and storage of river Meade, R.H., Yuzyk, T.R., and Day, T.J., 1990, Movement sediment in the Atlantic drainage of the United States: and storage of sediment in rivers of the United States Journal of Geology, v. 90, p. 235­252. and Canada, in Wolman, M.G., and Riggs, H.C., Meade, R.H., 1985, Suspended sediment in the Amazon editors, Surface Water Hydrology: Boulder, Colorado, River and its tributaries in Brazil during 1982­1984: Geological Society of America, The Geology of North U.S. Geological Survey Open-File Report 85­492. America, v. O-1, p. 255­280. http://pubs.er.usgs.gov/djvu/OFR/1985/ofr_85_492. Milly, P.C.D., Betancourt, J., Falkenmark, M., Hirsch, R.M., djvu Kundzewicz, Z.W., Lettenmaier, D.P., and Stouffer, Meade, R.H., 1995, Setting: Geology, hydrology, sediments, R.J., 2008, Stationarity is dead: Whither water and engineering of the Mississippi River, in Meade, management?: Science, v. 319, p. 573­574. R.H., editor, Contaminants in the Mississippi River, Nordin, C.F., Mejía, A., and Delgado, C., 1994, Sediment 1987­92: U.S. Geological Survey Circular 1133, p. studies of the Orinoco River, Venezuela, in Schumm, 12­28. http://water.usgs.gov/pubs/circ/circ1133 S.A., and Winkley, B.R., editors, The Variability of Large Meade, R.H., 1996, River-sediment inputs to major deltas, Rivers: New York, American Society of Civil Engineers, in Milliman, J.D., and Haq, B.U., editors, Sea-Level Rise p. 243­265. and Coastal Subsidence: Causes, Consequences, and Porterfield, G., 1972, Computation of fluvial-sediment Strategies: Dordrecht, Kluwer Academic Publishers, p. discharge: U.S. Geological Survey Techniques of 63­85. Water-Resources Investigations, book 3, chapter C3. Meade, R.H., 2007, Transcontinental moving and storage: http://pubs.usgs.gov/twri/twri3-c3/ the Orinoco and Amazon Rivers transfer the Andes Richey, J.E., Meade, R.H., Salati, E., Devol, A.H., Nordin, to the Atlantic, in Gupta, A., editor, Large Rivers: C.F., Jr., and dos Santos, U., 1986, Water discharge Geomorphology and Management: Chichester, John and suspended sediment concentrations in the Amazon Wiley & Sons, p. 45­63. River: 1982­1984: Water Resources Research, v. 22, Meade, R.H., and Moody, J.A., 2008, Changes in the p. 756­764. discharge of sediment through the Missouri-Mississippi Schumm, S.A., 2005, River Variability and Complexity: River system, 1940­2007: Proceedings of VIII Cambridge University Press. Encontro Nacional de Engenharia de Sedimentos, Sprague, L.A., Clark, M.L., Rus, D.L., Zelt, R.B., Flynn, J.L., Campo Grande, Brazil, 2­8 November 2008. and Davis, J.V., 2007, Nutrient and suspended-sediment ------. 2009, Causes for the decline of suspended-sediment trends in the Missouri River basin, 1993­2003: U.S. discharge in the Mississippi River system, 1940­2007: Geological Survey Scientific Investigations Report Hydrological Processes, v. 24, p. 35­49. http://water. 2006­5231. http://pubs.usgs.gov/sir/2006/5231/ usgs.gov/nrp/proj.bib/Publications/2010/meade_ Syvitski, J.P.M., Vörösmarty, C.J., Kettner, A.J., and Green, moody_2010.pdf P., 2005, Impact of humans on the flux of terrestrial Meade, R.H., Moody, J.A., and Stevens, H.H., Jr., 1995, sediment to the global coastal ocean: Science, v.308, Sampling the big rivers, in Meade, R.H., editor, p. 376­380. http://instaar.colorado.edu/~kettner/ Contaminants in the Mississippi River, 1987­1992: publication_folder/Science2005.pdf U.S. Geological Survey Circular 1133, p. 41­51 Topping, D.J., Parker, R.S., Nelson, J.M., and Bennett, J.P., Meade, R.H., and Parker, R.S., 1985, Sediment in rivers of the 1996, The apparent mid-20th century load decreases United States, in National Water Summary 1984: U.S. in the Colorado River basin: An investigation of the Geological Survey Water-Supply Paper 2275, p. 49­60. Colorado River Sampler: Geological Society of http://pubs.er.usgs.gov/usgspubs/wsp/wsp2275 America Abstracts with Programs, v. 29, no. 7, p. 261. 75 Vörösmarty, C.J., Meybeck, M., Fekete, B., Sharma, K., discharges from the Yangtze River (Changjiang), Green, P., and Syvitski, J.P.M., 2003, Anthropogenic 1950­2005, in Gupta, A., editor, Large Rivers: sediment retention: major global impact from registered Geomorphology and Management: Chichester, John river impoundments: Global and Planetary Change, v. Wiley & Sons, p. 609­626. 39, p. 169­190. Zhao, Y-a., Pan, X-d., Fan, Z-y., and Han, S-f., 1987, Wolman, M.G., 1967, A cycle of sedimentation and erosion Sedimentation in the lower reaches of the Yellow River in urban river channels: Geografiska Annaler, v. 49A, p. and its basic laws, in Brush, L.M., Wolman, M.G., and 385­395. Huang, B.W., editors, Taming the Yellow River: Silt and Xu, K-h., Milliman, J.D., Yang, Z-s., and Xu, H., 2007, Climatic Floods: Dordrecht, Kluwer Academic Publishers, p. and anthropogenic impacts on water and sediment 477­516. 76 5. Land-Ocean Interactions: Human, Freshwater, Coastal, and Ocean Interactions under Changing Environments Jeffrey E. Richey School of Oceanography, University of Washington Abstract a role. Floods and droughts have an impact on biodiversity, freshwater resources, agriculture, and livelihoods. While the Particularly complex and pressing challenges exist at the development of hydropower provides much-needed energy, interface between the upland terrestrial and freshwater it also alters the flow regime and sediment transport of rivers. realm, and the evolving ocean. Deciphering how signals Climate change affects all aspects of the system, bringing propagate downstream to interact with changing coastal changes in temperature and rainfall regimes, and reducing dynamics is a multifaceted task. This paper presents snow cover. Global economic development and food a "systems-level" overview of the key processes and shortages also have an impact on river basins and coastal transitions, from land to rivers to oceans and their marine zones and are a growing concern. International efforts must fate. It also summarizes the types of issues confronted in be made to predict and mitigate potential changes in climate. coastal-focused topic areas. As climate evolves, management options cover a range of issues, from bringing safe water to local villages for the rural The paper comments on a portfolio of World Bank poor, to adaptation strategies for large infrastructure. projects in the environment and water arena, as a means of identifying what existing projects are, and what their Particularly complex and pressing challenges exist at the requirements might be. In addition, it presents a case study interface between the upland terrestrial and freshwater of the Mekong River basin, as an example of a full suite of areas, and the evolving ocean. Deciphering how signals land-to-ocean issues that must be addressed. propagate downstream to interact with changing coastal dynamics is a multifaceted task. The mission is further Finally, it advances the concept of a virtual river/coastal basin, complicated by man's pervasive alteration to the natural driven by a dynamic information framework, as a means to system. Rising energy demands are met with ambitious provide a convergence of cross-sector information. The paper hydraulic projects that change the timing and volume of ends with a summary of lessons learned. sediment and water discharged to the sea. Reductions in the supply of sediment to the coastal zone and concurrent changes in ocean conditions create synergies that have From Land to Ocean negative impacts on coastlines. Coastal erosion and flooding are projected to accompany rises in sea levels and increased River basins and their downstream coastal zones are storm frequency and intensity. Coastal interactions (disasters facing a series of challenges critical to their future. These associated with sea level rise, increased frequency and challenges are centered on the availability and distribution of intensity of storms, saltwater intrusion and salinization water. Floods and droughts, the development of hydropower, of aquifers) are getting more attention. Nutrients and climate change, and global economic development, all play contaminants from upriver deposited in coastal areas as well 77 Figure 1. Schematic of the Major as salt water intrusion (exacerbated by the depletion of near Reservoirs and Pathways in Fluvial shore aquifers) can change the chemical environment along Systems the coasts. The scale of management issues facing countries in the coastal zone are multinational in nature, and will have Atmosphere to be looked at in a new way. Soils Further exacerbating the situation is the surprisingly sparse understanding of what is involved because it lies at the Dams Streams Rivers Coastal boundaries between more traditional disciplines and Zone more traditional geographies. For example, it is rare for oceanography programs to fund near-shore research since Riparian Floodplains most of them take place in the open ocean on board large ships. Similarly, terrestrial/freshwater ecology programs rarely approach issues relating to salty water. Overall, this Source: Adapted from Richey (2004). Note: Inputs from land occur directly or pass through the riparian zone. intersection is poorly understood and the paucity of data and Streams coalesce to form larger rivers that exchange with their floodplain. Rivers can pass directly to the coastal zone, or be retained behind dams. Dot- complexity in processes creates significant challenges. An ted lines indicate exchange with the atmosphere, grounded arrows indicate immediate challenge is to incorporate the best understanding sinks, arrows within the boxes indicate internal transformations. of the dynamics involved in changing environmental conditions in these sectors into World Bank policies and projects. This needs to be done in a cross-sector manner and As will be discussed in more detail later, the most common in a way that optimizes a multi-stakeholder return. estimations of the magnitude of these fluxes found in the literature are 0.4 petagrams (a petagram is 1,015 g) of carbon per year (PgC y-1) for total organic carbon (evenly A Template for Land to Ocean Fluxes divided between particulate and dissolved organic phases), and 0.4 PgC y-1 for dissolved inorganic carbon. While This section provides a summary of land to ocean fluxes these bulk fluxes are small components of the global carbon and processes as a way to establish a template to cycle, they are significant compared to the net oceanic evaluate specific regions (which, in turn, depend on their uptake of anthropogenic carbon dioxide (Sarmiento and connections to other parts of the system). The analysis is Sundquist 1992) and to the interhemispheric transport expressed using carbon as the currency, reflective of both of carbon in the oceans (Aumont et al. 2001). But these the fundamental role carbon plays in establishing ecosystem estimates contain very considerable uncertainty. Each term dynamics, and as the (eventual) basis for carbon trading is briefly evaluated below (see Figure 2). options. The analysis that follows draws substantially on Richey (2004, 2005). Mobilization from Land to Water and Riparian Zones Fluvial systems integrate hydrological and biogeochemical cycles, over scales from small streams to regional and, The fluxes from land to rivers are generally inferred directly ultimately, to continental basins (Figure 1). The transfer from the fluxes out of a basin, especially at a global scale. of organic matter from the land to the oceans via fluvial Although there is considerable truth to this for dissolved systems is a key link in the global carbon cycle because species (especially conservative ones), it is less true for it represents the main pathway for the preservation of particulate species, especially with human intervention. terrigenous production in modern environments (Ittekot and Hawke 1990; Degens et al. 1991; Hedges et al. 1992). The modern terrestrial sediment cycle is not in equilibrium Hence, the role of rivers in the global carbon cycle is most (Stallard 1998). Meade et al. (1990) estimated that typically expressed as the fluvial export of total organic agricultural land use typically accelerates erosion ten- to and dissolved inorganic carbon from land to the ocean (for one-hundred-fold, via both fluvial and Aeolian processes. example, Likens et al. 1981). Multiple reports in the literature support this conclusion. 78 Figure 2. Uncertainty Scenarios in the Fluxes of Carbon through Fluvial Systems Relative to Atmospheric CO2 Conventional CW + Continent CW+CS+ CW+CS+ POC,DOC) Transient Wisdom Sediment (POC,DOC) +Outgassing 1.3 1.8 2.6 1.3 Cont Outgas ? ? 2x 2x 2.5x 2.5x Net ­0.6 ­1.1 ­1.6 ­0.2 ­0.3 Pg C y­1 ­2 ­1 0 1 2 Source: Adapted from Richey (2004). Note: Including atmosphere to rivers (River Atm, as sum of other fluxes), continental sedimentation (Cont Sed, alluvial, reservoirs), outgassing from rivers to atmosphere (Outgas Atm), export of DIC to the sea (DIC sea, as 50% of total DIC, to represent just atmospheric weathering component), DOC export to the sea (DOC sea), export of POC to the sea (POC sea), and by difference the net exchange with the atmosphere ( Net Atm). For the purposes here, it is assumed that the export fluxes to the sea constitute a sink, with no return to the atmosphere, on an immediate decadal time scale. Particulate inorganic C is not included (as it is abiotic and does not interact with the other pools). Units are Pg C y-1 (scale at bottom lines up with the axis of each graph at 0). Scenarios include: CW (conventional wisdom on DOC and POC export, no continental sedimentation or outgassing), + Cont Sed (adding continental sedimentation to CW), +POC,DOC (adding the higher values for POC and DOC fluxes discussed in the text to +Cont Sed), +Outgassing (adding outgassing to the +POC,DOC), and Transient (inferring anthropogenic transients, based on the +Outgassing scenario). With the maturation of farmlands worldwide, and with the is not passive; significant transformations occur along the development of better soil conservation practices, it is way. Rivers exchange with their floodplains (depending on probable that human-induced erosion is less than it was how canalized and diked a river is). The movement of POC several decades ago. Overall, however, there has been a is, of course, directly linked to the movement of suspended significant anthropogenic increase in the mobilization of sediments. Sediments are deposited and remobilized sediments (and associated particulate organic carbon or multiple times and over long timescales. In the Amazon, POC) through fluvial processes. The global estimates of for example, Dunne et al. (1998) computed that as much the quantities, however, vary dramatically. Stallard (1998) sediment was being recycled within a reach as was leaving poses a range of scenarios, from 24 to 64 Pg y-1 of bulk it. Presumably, a significant amount of the erosion-excess sediments (from 0.4 to 1.2 Pg y-l of POC). Smith et al. sediment discussed in the previous section makes it some (2001) estimate that as much as 200 Pg y-1 of sediment is distance downstream but is then slowed and retained within moving, resulting in about 1.4 PgC y-1. the alluvial floodplains. Where does this material go? Does it all go downstream via An additional process--the mineralization to pCO2--within big rivers, ultimately to the ocean, or is it stored inland? Stallard flowing water significantly affects organic matter (OM). Most (1998) argues that between 0 and 40 Pg y-1 of sediments river and floodplain environments maintain pCO2 levels that are stored as colluvium and alluvium and never make it are supersaturated with respect to the atmosphere. High downstream. Using a different approach, Smith et al. (2001) partial pressures of CO2 translate to large gas evasion estimate that about 1 PgC y-l of POC is stored this way. fluxes from water to atmosphere. Early measurements in the Amazon suggested that global CO2 efflux (fluvial export plus Within-River Transport and Reaction respiration) from the world's rivers could be on the order Processes of 1.0 PgC y-1. Recent measurements of temperate rivers lead to estimates of global river-to-atmosphere (outgassing) Within-river transport processes carry these eroded fluxes of ­0.3 PgC y-1, which is nearly equivalent to riverine materials downstream through the river network. Transport total organic carbon (TOC) or dissolved inorganic carbon 79 (DIC) export (Cole and Caraco 2001). Richey et al. (2002) more detailed analysis of the coterminous United States to computed that outgassing from the Amazon alone was an estimate of about 10 PgC y-1 worldwide (versus 13 PgC about 0.5 PgC y-1. Assuming that the fluxes computed for y-1 efflux to the oceans), for a storage of about 0.2 PgC the Amazon are representative of the fluvial environments y-1 (which he includes as part of his overall calculation of of lowland humid tropical forests in general, surface water continental sedimentation). CO2 evasion in the tropics would be on the order of roughly 0.9 PgC y-1 (three times larger than previous estimates of Export to the Coastal Zone global evasion). Factoring in recent Amazon results, a global flux of at least 1 PgC y-1 directly from river systems to the The conventional wisdom is that the flux of particulate atmosphere is likely. and dissolved organic matter are each about 0.2 PgC y-1 and dissolved inorganic carbon (DIC) is 0.4 PgC y-1 (for Pre-aging and degradation may alter significantly the example, Schlesinger and Melack 1981; Degens 1982; structure, distribution, and quantity of terrestrial organic Meybeck 1982, 1991; Ittekot 1988; Ittekkot and Laane matter before its delivery to the oceans. As noted by Ludwig 1991; Ludwig et al. 1996; Ver at al. 1999). That these (2001), the organic matter that runs from rivers into the analyses converge is not terribly surprising. They are all sea is not necessarily identical to the OM upstream in river based on much of the same (very sparse) field data and catchments. Cole and Caraco (2001) observe that the use variations of the same statistically based interpolation apparent high rate of decomposition of terrestrial organic schemes. Let us evaluate these numbers. Because direct matter in rivers may resolve the enigma of why organic measurements are few, POC flux estimations are typically matter that leaves the land does not accumulate in the a product of the flux of total suspended sediments (TSS) ocean (Hedges et al. 1997). Overall, this sequence of and the estimated weight-percent organic carbon (w%C) processes suggests that the OM that is being respired is associated with the sediment (because the bulk of POC is translocated in space and time from its points of origin, such organic carbon sorbed to mineral grains). The first problem that, over long times and large spatial scales, the modern is an adequate resolution of the TSS flux. Data on TSS aquatic environment may be connected with terrestrial are frequently poor and of unknown quality. Many reported conditions of another time. data are surface samples, and the depth integrations necessary to accurately characterize sediment flux are Input to Reservoirs on the order of two to three times higher. Additionally, much sediment moves during episodic storm events, Reservoir construction and the accompanying fragmentation when measurements are almost never made. Finally, most in the flow of the world's large rivers have had a tremendous measurements of both water flow and chemistry are made impact on the hydrologic cycle and the fate of dissolved and some distance from the actual mouth (in the Amazon particulate material. Starting about 50 years ago, large dams for example, the last regular sampling station, Óbidos, were seen as a solution to water resource issues, including is over 700 km from the sea, with an island the size of flood control, hydroelectric power generation, and irrigation. Connecticut). Overall, estimates of sediment/POC inputs Now, there are more than 40,000 large dams worldwide to the ocean should be considered to vary by a factor of (World Commission on Dams 2000). This has resulted at least 2.5, particularly in Oceania, Southeast Asia, and in a substantial distortion of freshwater runoff from the South Asia (Figure 3). continents, raising the "age" of discharge through channels from a mean of between 16 and 26 days to nearly 60 days As summarized by Vorosmarty et al. (2003), estimates of (Vorosmarty et al. 1997). Whereas erosion has clearly total suspended sediment transport to the oceans have increased the mobilization of sediment off the land, the ranged from 9 PgC y-1 to more than 58 PgC y-1, with proliferation of dams has acted to retain those sediments. more recent studies converging around 15 to 20 PgC y-1. Vorosmarty et al. (2003) estimate that the aggregate impact These estimates are generally based on extrapolations of all registered impoundments is on the order of 4 to 5 of existing data, which are weighted to the large rivers PgC y-1 of suspended sediments (of the 15 to 20 PgC y-1 of passive margins and temperate regions. Milliman and total that he references). Stallard (1998) extrapolates from a Syvitski (1992) called attention to the much higher yield 80 Figure 3. Uncertainties in Sediment/POC Loading 350 300 250 200 Tg/y 150 100 50 0 Oceania E Asia SE Asia NE S America W S America SE S America W N America At N America EG N America EA At Europe W Africa E Africa Australia "Traditional" : 200 Area-loading : 350 Calc. Yield : 800 "Best Guess: ~ 500" Note: By geographic region, with emphasis on Oceania and SE Asia, as calculated by 3 different methods. rates from steep mountainous environments (without Marine Fate directly computing a global total). More recently, Milliman et al. (1999) estimated that the total sediment flux from Long-term preservation of terrestrially derived organic the East Indies alone (the islands of Borneo, Java, New matter in the oceans occurs largely within sediments Guinea, Sulawesi, Sumatra, and Timor, which represent that accumulate along continental margins. Organic about 2 percent of the global land mass) is about 4 PgC carbon within these sediments is thought to be preserved y-1, or 20 to 25 percent of the current global values. This largely because it is adsorbed to mineral grains (Keil et type of environment (steep relief, draining directly to the al. 1994; Mayer 1994; Bishop et al. 1992). Hedges and oceans) is found elsewhere in the world, so the results are Keil (1997) estimated that carbon preservation along not likely to be unique. Data from Taiwan support these continental margins over the Holocene was split roughly high levels, with isotopic analyses of the carbon showing evenly between sediments accumulating within the delta that a significant part of the flux is human-driven (Kao and or sedimentary plume of rivers and non-deltaic sediments Liu 2002). accumulating outside the direct influence of major rivers (but within range of multiple smaller systems). To obtain particular organic matter flux estimates, these values (and their uncertainties) must be multiplied by The storage efficiency of deltaic and non-deltaic systems region-specific carbon values. The total uncertainties in both is different. The amount of organic carbon in non-deltaic sediment and carbon must then be propagated. To account continental shelf sediments falls in a narrow range (0.5­1.1 for this range, POC flux can be computed as an ensemble milligrams of carbon per square meter [mg C m2] of based on different combinations of weight-percent organic mineral surface), and typically more than 90 percent of the carbon and total suspended sediment fluxes, resulting in a preserved organic matter is adsorbed to mineral surfaces. range of 0.3 PgC y-1 to 0.8 PgC y-1, with a "more likely" Deltaic sediments are distinctly different, containing only level of about 0.5 PgC y-1 (depending on the assumptions a fraction of the organic carbon (by weight) found in used). Therefore, it is possible that the common estimate other margin sediments. Suspended sediments from the of 0.2 PgC y-1 is low and that the overall value lies in the Amazon River, for example, have loadings (-0.67 mg C m2) range of 0.2 to 0.5 PgC y-1. that are three times higher than the corresponding deltaic 81 sediments (Keil et al. 1997), so more than two-thirds of the components of inland water projects are water supply (32 terrestrial particulate organic load delivered to the Amazon percent of all projects in this category), and irrigation and delta is lost from the mineral matrix and is not preserved. drainage (24 percent). The number of coastal and marine The Mississippi, Yellow, and other river/delta systems also projects is small and is dominated by the "miscellaneous" show extensive loss of terrestrial organic matter. Thus, many category, which includes a broad array of projects. Marine deltaic systems bury only a small fraction of the potential projects are relatively evenly split into projects dealing with organic load normally sorbed to mineral particles, with the fisheries, marine, and coral reefs. Further insight into the 41 balance presumably desorbed or mineralized (and entering coastal projects can be obtained by re-filtering the sector the dissolved inorganic matter pool). analysis. Graph (d) shows that 34 percent of projects focus on biodiversity, 32 percent focus on pollution, 29 percent The organic matter lost by mineralization and not buried is on integrated coastal management (ICM), and 5 percent on one of the factors in maintaining the historical perspective reconstruction. that marginal seas are net heterotrophic (Chen 2004). But Chen (2004) reviews more recent evidence, based on direct measurements of pCO2 (again, showing the Region-Specific Evaluations critical importance of actual field measurements of key parameters!) and comes to the conclusion that these seas How does this portfolio relate to how land-ocean are net autotrophic, driven primarily by nutrients delivered connections function, as both natural and managed via upwelling (with enhanced nutrients delivered by rivers systems? leading to eutrophication constituting only a minor source), and net consumers of atmospheric CO2. The overall As shown earlier in Figure 4, World Bank projects and implication of this sequence of processes is that much of interests cover a wide range of sectors. They include the anthropogenically mobilized riverborne organic matter projects in coastal subsidence and sea level rise, coastal (and perhaps the naturally mobilized OM) is liable to remain estuaries and wetlands, carbon in coastal wetlands, coral in the marine environment over timescales longer than the reefs, and hydropower. current increase of atmospheric carbon dioxide. Coastal Subsidence and Sea Level Rise The World Bank Environmental/Water This section discusses impacts to deltas related to Resources Project Portfolio human activities. Deltas respond to both landward and seaward pressures (Figure 5). Construction of levees and The avowed purpose of the Bank's Hydrology Expert Facility the alteration of natural dispersal processes decrease (HEF) is to bring (new) expertise to bear on World Bank sediment input to the delta plain, while eustatic sea level projects in the overall water-related portfolio. To help focus and erosive storm activity continues to rise with warming this effort, it is useful to briefly examine the World Bank ocean temperatures. The result is that the world's deltaic portfolio of projects (Figure 4). coastlines are extremely vulnerable to anthropogenic change. While there are many examples, the following The projects that can be considered water and water illustrate the nature of the problem. resources (W&WR) are shown in graph (a) of Figure 4 and represent the majority category, with 22 percent of the The Mississippi River delta is a prime example of a total projects. The same graph shows that the ecosystems heavily impacted dispersal system. Alteration of source and biodiversity category accounts for 18 percent of all and dispersal processes is exacerbated by oil and gas projects. This category establishes the conditions for inputs extraction, and groundwater off-take. The upriver-supplied to the water systems. Graphs (b) and (c) break down the nutrient load promotes the formation of a "dead zone." In water and water resources category into its constituent addition, the fate of river-borne organic carbon is being projects by World Bank sector for inland waters (Graph altered where the Mississippi delta has grown to the shelf b) and coastal and marine waters (Graph c). The main break, allowing direct deposition to deeper water and 82 Figure 4. World Bank Project Portfolio: Environment and Water Resources (a) World Bank Project Portfolio (b) W&WR: Inland Waters (#=2427) 600 160 120 400 80 200 40 0 0 W&WR Inf E&B Mis ENH Soc C&A WS I&D Mis S&S GW HP Fld GAF Drt W&WR Water & water resources ENH Energy (non hydropower) WS Water supply HP Hydropower Inf Infrastructure Soc Social I&D Irrigation & drainage FL Flood E&B Ecosystems & biodiversity C&A Climate & atmosphere Mis Miscellaneous GAF General agriculture, fishing Mis Miscellaneous S&S Sewer & sanitation and forestry GW General water, sanitation Drt Drought and flood protection (c) W&WR: Coastal and Marine Waters (d) Coastal and Marine Waters by Theme 20 R 15 B 10 P 5 0 ICM Mis GAF S&S GW Fsh Mar Ref Coastal: Marine: Mis Miscellaneous Fsh Fisheries B: Biodiversity GAF General agriculture, fishing and forestry Mar Marine ICM: Integrated coastal management S&S Sewer & sanitation Ref Coral reefs P: Pollution GW General water, sanitation and flood protection R: Reconstruction Source: http://www.worldbank.org/projects Note: (a) The number of active, proposed, and pipelined projects to do with environment and natural resource management since about 1984 are broken down by primary categories. The water and water resources category is broken down into (b) Inland Waters sectors (sector as defined by the data source) and (c) Coastal and marine waters. The coastal and marine category can be recombined (d) into theme areas. Figure 5. The Mekong Delta promoting submarine landslides. A significant issue is the increased vulnerability of coastal communities to storms (lessons from Katrina and Rita) (Day et al. 2007). Hydrologic/sediment changes on the Nile River have driven the Nile Delta into a destruction phase over the last 150 years (Stanley and Warne 1998). The High Aswan Dam and Reservoir have trapped almost one hundred percent of sediment delivery to the estuary, and drastically altered the hydrography. Effects include accelerated coastal erosion and straightening of the shoreline, reduction in wetland size, increased landward incursion of saline groundwater, and buildup of salt and pollutants to toxic levels in the wetlands and delta plain. Moreover, seasonal floods capable of flushing agricultural products/pollutants created by Egypt's expanding population are being reduced or eliminated. 83 The demands of expanding populations present multiple The root causes of these threats include: challenges in balancing a return to any semblance of natural conditions and the advantages inherent in them. · Land uses and land tenure, such as the transformation of Restoration efforts aim to re-establish dynamic interactions, the Umfolozi swamps for improved agricultural produc- with emphasis on reconnecting the river to the deltaic plain. tion, which disrupt terrestrial and wetland processes; Science must guide restoration, which will provide insights · Poverty; and into coasts facing climate change in times of resource · Weak institutional environment. scarcity. Integrating the delivery of sediment and discharge of freshwater to the delta with large-scale hydrology models to The issue is well-phrased, in this excerpt from a GEF make better predictions about coastal erosion, subsidence, project, on the iSimangaliso Wetland Park, in South Africa. groundwater salinity intrusion, and other forces at play would help set a rigorous template for decision making. "The challenge faced by the Wetlands Authority is therefore to respond to the twin imperatives of Coastal Estuaries and Wetlands conservation and development in a manner that aligns with the shift in national (and global) priorities Estuaries are depositional environments that are often from a strong focus on conservation-in-isolation dominated by fine-grained sediments. Sedimentation is to a new approach that integrates biodiversity promoted by the existence of estuarine turbidity maximum conservation with regional development." (ETM), a zone of convergence at the mouth of a river. Salinity effects enhance flocculation and the increase settling Carbon in Coastal Wetlands rate. Contaminants such as trace metals, polychlorinated biphenyls (PCBs), pesticides, and polyclyclic aromatic Beyond the role of coastal wetlands in fisheries, agriculture, hydrocarbons (PAHs) are adsorbed to the surface of particles and coastal protection, there is the substantial, but tricky, and settle out of the water column in the estuary. Benthic role of wetlands in carbon storage (mitigation of CO2 communities are adversely affected by the toxic sediments. emissions), as well as in adaptation (M. Hatziolos, pers. Sediment quality guidelines (SQGs) have been established comm.). Mangrove forests appear to provide a double through experiments in US estuaries (lead by Edward Long), dividend with respect to mitigation and adaptation in and have been validated abroad (McCready et al. 2006). addressing climate change at the local level. But a potential glitch with respect to natural carbon capture and storage There are many examples throughout the world of specific is working out the carbon cycle under different conditions coastal estuaries and wetlands where changes in both of mangrove and wetland (including mudflats) disturbance. upstream hydrology and marine-side forces have had an The net carbon storage appears to be very closely related to impact on the region, and become subject to remedial hydrology and exposure of soils, as well as methane release. actions. For example, the iSimagaliso Wetland Park, on the If progress is to be made on possible carbon credits and east coast of South Africa, is suffering under the impact of a offsets through mangrove reforestation or protection, then series of factors. Immediate threats include: good measures of net carbon storage or emissions under these different conditions are needed. · Degradation of the iSimagaliso Wetland ecosystem be- cause of closure of the mouth of the St. Lucia estuary; This stresses the importance of maintaining the hydrology · The presence of commercially viable mineral deposits in intact (or at least ensuring environmental flows) to support the coastal dune cordon; healthy mangroves so that the increasingly important · Large-scale commercial afforestation in endemic grass- carbon storage service is maintained. Similar concerns lands and water catchments on the park's fringes; and with mudflats (which are apparently even greater natural · Spread of invasive alien plants that are threatening the stores of carbon than peatlands) revolve around dredging, highly productive communities growing in moist environ- filling, and building over these carbon reservoirs. These key ments, particularly on the alluvial floodplains along the ecosystem services, which are not adequately valued or coast line and in the valleys of the Lubombo Mountains. acknowledged by decision makers, are being lost. 84 Coral Reefs The boundary of the Coral Triangle region coincides with the most productive region in the world in terms of sediment Coral reefs represent an intersection between changing discharge (Figure 3). According to Milliman and Syvitsky ocean conditions, immediate population pressures (1992), this part of the world accounts for 50 percent of (overfishing, destructive fishing), and impacts from land. For the global sediment flux to the ocean, but only about 3 example, the Coral Triangle covers all or parts of Indonesia percent of the land area. This is because of the role of small, (Central and Eastern), East Timor, The Philippines, Malaysia mountainous rivers with highly erodible rock, combined with (part of Borneo), Papua New Guinea, and the Solomon high population pressures. The ability to attribute changes in Islands. Sometimes referred to as the "Amazon of the Seas," the landscape production of sediments to loadings on reefs it is the epicenter of marine life abundance and diversity would help develop suitable management practices. An on the planet. While the area covers only 2 percent of the emerging class of coupled hydrology/sediment models is a world's oceans, it contains more than 75 percent of all step in that direction (Figure 6). known coral species, more than 30 percent of the world's coral reefs, nearly 40 percent of coral reef fish species, and For example, Kimbe Bay (Papua New Guinea) is home the greatest extent of mangrove forests anywhere in the to at least 860 species of reef fish and 350 species of world. Regional-scale gradients exist in reef biodiversity, with hard coral, making it one of the world's richest marine decreasing diversity with distance from the Indo-Australian environments. This unique area is under threat from logging archipelago. Bellwood and Hughes (2001) best explain and development, destructive fishing, and rapid population this variation with large-scale patterns in the availability of growth. The Derawan Islands (Indonesia) feature some shallow-water habitat. The challenge now is to identify the of the most significant green turtle nesting beaches in relation between taxonomic composition, species richness, Southeast Asia and a unique saltwater lake with four and ecosystem function in reef systems. Low-diversity endemic, stingless jellyfish species. The area's reefs are regions are particularly sensitive to anthropogenic impacts, extremely diverse because of the influence of the Berau and underscore the need for "integrated management at River on the coastal waters, illustrating the sensitive link multinational scales." between land and sea in some places. Figure 6. A Coupled Hydrology-Sediment Model, DHSVM 3.0 Source: Doten et al. (2006). Note: Based on computing the probability of mass wasting and surface erosion. 85 This region is the focus of the emerging Coral Triangle a confluence of issues. Transnational agencies play Initiative (CTI) on Coral Reefs, Fisheries, and Food Security important roles in mediating among competing interests. that aims to bring together six governments in a multilateral A key player in the Mekong basin is the Mekong River partnership to conserve the extraordinary marine life in the Commission (MRC), which is based in Vientiane, Laos. region. In December 2007, government representatives from The mandate of the MRC includes current and future environment and fisheries ministries in Indonesia, Malaysia, water resource management of the riparian countries Papua New Guinea, The Philippines, Solomon Islands, (Laos, Thailand, Cambodia, and Vietnam) of the lower and Timor-Leste met to agree upon a way forward for the Mekong, according to the terms of the Agreement on CTI. After the meeting, President Yudhoyono of Indonesia the Cooperation for the Sustainable Development of the launched the CTI. The GEF saw the CTI as one of the most Mekong River Basin (April 5, 1995). Institutions such as important initiatives in its history expecting to see at least the MRC need to be able to work with regional political $25 million focused on the program. realities, and also harness the most "complete" science in order to inform decision makers. The guiding principles agreed to by the Coral Triangle governments illustrate the complexity of objectives that The Region require cross-sector approaches. These principles are also relevant to other such projects. The principles are: The Mekong is a large, diverse transboundary river basin. It has the world's 8th largest discharge (ca. 0.47 km3/yr), · Support people-centered biodiversity conservation, sus- 12th largest length (ca. 4,800 km), and 21st largest tainable development, poverty reduction, and equitable drainage area (ca. 795,000 km2) (Figure 7). The Upper benefit sharing. Mekong basin covers an area of 189,000 km2 in China, · Be based on solid science. Burma, and the northern part of Laos. This area has a · Be centered on quantitative goals and timetables ad- mountainous terrain with elevations ranging between 400 opted by governments at the highest political levels. · Recognize the transboundary nature of some important marine natural resources and communities. Figure 7. The Mekong River Basin · Be inclusive and engage multiple stakeholders. Hydropower Pressed by growing demands for clean(er) energy throughout the world, the hiatus in dam building is ending with a gathering "hydropower renaissance." While not yet quantified, the consequences will be considerable. A more detailed discussion is provided below in the discussion of the Mekong River case study. The Mekong River Basin: A Case Study Transboundary river basins, where a river passes through several countries, pose particularly vexing problems in water resource allocation. These problems encompass not only water, but also fisheries production, sediment transport, and navigation. The 6-country Mekong River basin is a very important example of this class of issues. Emerging conditions in the Mekong River basin represent 86 and 5,000 m, and provides about 16 percent of the annual pronounced gaps in data records. Trends that have been flow to the Lower Mekong basin (which encompasses "perceived" could also be because of channel scouring 606,000 km2). The Northern and Eastern Highlands, with or silting at gauge locations, defective gauge operation, elevations of up to about 2,800 m, are the wettest regions poorly developed rating curves, and/or undetected trends in in the basin. In contrast, the Khorat Plateau (in northeastern precipitation. Thailand), is a dry region with intense evapotranspiration. The main feature of the lower Mekong is Cambodia's Tonle The recent application of basin-wide models is providing Sap (or "Great Lake") a complex and important ecosystem key insights into the functioning of the Mekong basin. As driven by an annual flood pulse. The lake's fisheries, which will be discussed below, an important aspect of model are critically dependent on the subtleties of the flow regime, application is not only the computed (relative to observed) are important for their biodiversity as well as a critical food flows, but that model development itself "forces" data source, providing 60 to 80 percent of the fish protein to the integration. Takeuchi et al. (2008) report on a series of region. Finally, the river passes through the delta in Vietnam, model applications. Costa-Cabral et al. (2007) used the VIC and discharges to the South China Sea. model to provide a detailed analysis of the interactions of landscape structure and use, climate, and water movement. Population growth and socioeconomic development in Costa-Cabral et al. (in prep.) analyze the potential the Mekong River basin in the second half of the 20th consequences of land use change, dams, and climates. A century and into the 21st century has been accompanied provocative result of this work is that the lack of increase in by unprecedented changes in land cover and land use. streamflow in northeastern Thailand, which would have been All Mekong regions were affected, although to a different expected but is not being observed, could be the result of extent, depending on environment, population growth, the use of bunded paddies in which collected rainwater, socioeconomic development, and each country's style added irrigation water, or both, is prevented from running off of transition to a market economy. The irrigated area has the paddy and eventually infiltrates or evaporates, returns an expanded greatly with the construction of large reservoirs amount of water to the atmosphere that surpasses the large for irrigation and power production. During this same evapotranspiration losses from the original forest. Hence, period, the Mekong experienced floods (that caused great a decline in the region's runoff ratio has accompanied the loss of life and material damage), as well as crop-damaging expansion of agriculture. droughts. These hydrologic disasters have been attributed to man-made changes, primarily deforestation (forests are The MRC is developing its own model environment, perceived as streamflow moderators and precipitation following the Decision Support Framework (DSF; not attractors). Rising crop damage from droughts has also published) based on the SWAT, ISIS, and IQQM models, been blamed on deforestation by some, while other analysts and is looking at including the VIC model in this portfolio. contend that croplands lower evapotranspiration relatively The experiences in the development and application of more than forests, and that this should lead to increased, these models call attention to the importance of matching not reduced, streamflows. Alternatively, low dry-seasons models, applications, and capabilities. have been attributed to the Chinese dams. Upcoming Issues for the Mekong Conflicting opinions and lack of scientific evidence on streamflow trends hinder policy making and international Political stabilization and increased global energy and agreements, and exacerbate conflicting interests between market demands have intensified pressures on the Mekong. countries and stakeholders, as well as between the goals of conservation and development. Agricultural expansion and dramatic deforestation in northeast Thailand have resulted in a decreased recurrence Initial Applications of System Models of low streamflows. However, in an apparent paradox, crop yields are increasingly vulnerable to precipitation shortages, A significant problem is lack of regional data, particularly and drought has become a major issue in this region. The discharge and rainfall. Decades of strife have led to Thai government has funded the construction of thousands 87 Figure 8. Map of Projected Mainstem of small reservoirs on individual farms to help mitigate and Tributary Dams the problem. The expansion of agriculture in northeast Thailand towards less favorable lands has increased crop vulnerability to climate change. It may be possible to counteract this with a technological response, such as rice varieties that are more resistant to drought. Irrigation may continue to expand and intensify in all countries, even though it faces institutional as well as natural resources constraints (MRC 2003). The future of the low-yield, labor intensive rainfed rice cultivation may be in decline. The cumulative impacts of all upstream events converge with changes along the coast itself as well as with changing marine conditions. Vietnam is particularly at risk of sea level rise (for example, Dasgupta et al. 2007). Direct impacts along the coast include conversion of mangrove forests to aquaculture, particularly shrimp farming (Tong et al. 2004), which has become susceptible to viral infections and salinity intrusion. Reservoir impacts on the flow of the Mekong are currently relatively limited. The two existing reservoirs in the Chinese Mekong (Manwan and Dachaoshan) have limited regulation potential. The Pak Mun dam is a run-of-the-river dam, fed by the Mun-Chi river system in Thailand. The Ubol Ratana (also called Nam Pong) dam, located in the Pong tributary of the Chi River in Thailand has been in immediate and acute impact would be physical barriers for operation since 1966 and is used for power generation, fish migration. irrigation, water supply (including for industry), and flood control. The Nam Theun II dams in Laos are under An additional consequence of reservoirs on the Mekong and construction. other tropical river basins is the potential for the production of greenhouse gases (GHG), especially methane. As noted A series of dams, currently in the planning stages will further by the report of the World Commission on Dams (2000), impact the Mekong River basin. The Chinese government hydropower cannot, a priori, be automatically assumed to be is planning the construction of a cascade of hydropower a cleaner technology than thermal alternatives with respect reservoirs along the upper Mekong, with a reported to GHG emissions. Case-by-case research is needed to (massive) 23 km3 of active storage beyond 2020. The make this claim. The organic carbon/gas dynamics of the so-called "Hydropower Renaissance" (sensu World Bank Mekong are very "active," fed by terrestrial inputs as well SDN Weeks, February 2008) includes the Mekong. On the as in situ production. The implication is that a cascade order of 100 dams are under discussion in the tributaries of of reservoirs could be expected to have a very significant Laos, and on the mainstream as far down as Cambodia. The GHG footprint. potential impact of even a subset of these dams would be very high. The "far-field" cumulative flow impacts, even if all Finally, the critical question that must be asked is: What were run-of-rivers dams, would be substantial by the Tonle are the cumulative impacts of land use change, reservoir Sap and the Mekong Delta, and on into the South China construction, and climate change (Figure 9)? The answer to Sea. Sediment trapping would significantly reduce the flux this question represents the ultimate cross-sector analysis, of sediments and associated nutrients downstream. Of not only for the Mekong, but across regions. 88 Figure 9. Synergistic, Cumulative Impacts from Land Use, Dams, and Climate Change for the Mekong (and Elsewhere) Source: Adapted from Costa-Cabral et al. (2007). A Foundation for Multi-sector disseminated. A baseline assessment of current and past Integration of Information environmental conditions (to establish both the extent and processes of change) of a basin provides the foundation As such, these targets represent a very complex set of from which to build. A baseline allows the analysis of future intersecting issues of scale, cross-sector science and scenarios as well as monitoring the evolution of key system technology, education, politics, and economics. One of the variables. most significant challenges for evaluating past performances and establishing the basis for future decisions is how to Establishing such a process is not a trivial task, for several undertake a quantitative analysis of the multiple complex reasons. First, the information required comes from multiple pathways and trade-offs involved in a policy project, from sources, from individual rain gauges to statistics on rice small farms to regional implications. A template for decision yield and fisheries. It also comes from multiple disciplines, makers to rigorously consider alternative scenarios could which presents problems even with communication play an important role in making complex environmental between specialists. Existing data are not always readily and economic decisions. This requires an accurate obtainable, sometimes for institutional reasons. New field understanding of linkages between water and multiple measurements, especially holistic and cross-boundaries, allocations, with the ability to carry out quantitative forecasts are challenging. Second, handling such diverse data and of the individual and combined impacts of demand. Once executing models is not straight-forward. There are very real that information is available, it would then be necessary to problems in converting data streams into useful information evaluate the trade-offs among sectors in order to establish that goes beyond a database. Third, perhaps most future policy interventions and financial investments. challenging is how to get the information into the hands of users, from specialists to local and regional decision To do this, information from multiple sources must converge, makers, and to the local farmers or fishermen. Finally, few, it must be organized and evaluated (preferably according if any, institutions in the world have sufficient in-house to organizing ecosystem principles), and it must be expertise to execute all parts of such a process. 89 Figure 10. A Schematic of a Virtual River The information and decision issues confronting a basin Basin, from Topography (Bottom) to are challenging but not unique. There is now broader Land Cover Attributes (Middle) to Political recognition of the need for more holistic views. Advances in Boundaries (Top). the science of how to analyze complex systems is evolving, as are sensors (on the ground and in orbit), and computers that facilitate the acquisition and processing of information. Knowledge about the process of organizing complex information (sometimes known as "cyber-informatics") is also evolving rapidly. In this spirit, it may be useful to think of a virtual river basin (VRB) as both a metaphor and a practical engine for organizing and processing the information and decision needs for a basin (Figure 10). A virtual river basin can be thought of as the common environment for the overall Each data "layer" is a "model" in its own right, of interest to diverse parties. information sources describing a basin, organized in a highly The "summation" provides not only within- but cross-sector integration. systematic fashion, to facilitate analyses, and to "visualize" outcomes. Information organized according to landscape principles (below) can serve multiple purposes, with Fundamental to these is a new class of hydrology models, specific targets for information identified and prioritized. The which can also be regarded as overall landscape models intersection of biophysical processes and environmental because of the processes (and data layers) they represent. stressors can be seen in a geospatially-explicit fashion. A key aspect of these models is that they are geospatially Careful attention must be paid to how the information is explicit, fully distributed, recognize the spatial heterogeneity organized, displayed, and distributed. of the watershed, and are process-based. Because these models can, and must, meld information from multiple Organizing Principle for the VRB: The sources, they can be functional in specific regions where Movement of Water Across and Down a local data are relatively sparse. River Network The Information Structure The theoretical structure for a VRB is to track the overall pathways and processes of water as it moves from the At the core of a virtual river basin is a dynamic information atmosphere to and through the landscape and down river framework (DIF) that can provide a consistent theoretical channels, through reservoirs and lakes, to the sea, on a basis and the overall capability of integrating across sectors. geospatially-explicit, multi-temporal basis (as described, "Dynamic" refers to the fact that the landscape is evolving; below). The knowledge necessary to track water includes that is, that we must look not only at the present, but also an understanding and mobilization of information for all at the past and, especially, the future. Information is not aspects of the landscape, including agriculture practices, static. "Information" means that more than just data needs land cover, topography, soils, fisheries, infrastructure, and to be considered; that is, what products must be developed human interactions. from the data? "Framework" means that an overall set of information must be logically arranged and communicated The robust framework for tracking water to be enacted within a flexible environment. The ability to interact with and for a VRB is the emergence of a new generation of earth communicate the results of a DIF is critical. system science, based on rapidly evolving capabilities for addressing global change issues. This involves use Essentially, a DIF is a numeric and quantitative "commons" of satellites, new generations of dynamic models, field that builds on the legacy of knowledge from experience, measurements focused by model requirements covering with the goal of harmonizing watershed function for wide areas, and, especially, a focus on "integrated systems." multiple users. The goal is to provide an instrument for a 90 (quantitative) analysis of complex interdependent problems. A Cyber Infrastructure The process of creating the model provides an integration of data from multiple sources (of interest to many). The The computational and data organizational issues framework provides a way to interpolate sparse data, as represented in executing the DIF are not trivial, but they well as the basis for cross-scale/upscaling analyses, and the are manageable. Figure 11 shows the sequence of issues foundation for building "scenarios." to be resolved, from the details of metadata and data storage, to facilitated access. It is useful to think in terms of The specific components of the DIF include: mobilizing the data from archives (and its attendant issues) to "data streams," which focus on specific outcomes, · Base data layers; as represented by the modules. The actual execution · Directed data layers, focused on synthetic objectives; of moving data from archives to something useful is · Geospatially-explicit, process-based, cross-sector expedited by including data services for processing the simulation models (requiring data from the directed data data into usable forms. Given the complexity of outcomes, layers). A modular structure allows ready swapping of experience has shown that attention to providing visually models (while focusing on getting work done); compelling data products is very important for effective · Facilitated input/output (including visualizations); communication, not only with decision and policy makers, · Decision support system and scenario testing capabilities. but also with the public at large. Underlying the technical details are the issues of dealing with (1) ownership of and The framework should be cross scale, allowing accurate access to primary data, (2) where systems reside (national, representation of large regions and far-field effects, while ministry, agency), (3) accessing and using core information being able to "zoom in" to a specific site of a project. While from multiple locations for inclusion in analysis, synthesis, flexibility is highly desirable, hence the term "framework," and outputs, and (4) the communication of scenarios and emphasis must be given to "getting the job done." likely outcomes. Figure 11. The "Cyber Infrastructure" to Support a DIF Including databases, data archives, data services, models, and "visualization servers". 91 A Prototype Virtual River Basin be used with or "coupled" to other models (for example, for climate or hydropower or carbon exchange with the As a means to start the discussion on developing a virtual atmosphere) and used to evaluate the impacts of land river basin, consider the conceptual framework shown use change, irrigation, dams, and climate change on the in Figure 12. The construct is that each module of the hydrologic cycle. The lake water balance (Module 4) is the framework represents internally consistent data and product of water inputs (from Module 3), outflows, and information, and that exchanges between each module bathymetry. occur sequentially. The information is derived from direct measurements and observations, and from modeling to The second set of modules addresses the production basis interpret that information. of the basin, building on its basic "physics". The Landscape Production (Module 5) represents primary production by The first set of modules establishes the basic structure land cover (including natural vegetation and agriculture), and dynamics of the basin. The drainage basin (Module and secondary production (including livestock), responding 1) establishes basic attributes of the landscape, including to the structure of the drainage basin, and climate forcing topography, soils, land use, and land cover. The climate (including changes in climate). Coupled to the hydrology forcing (Module 2) "drives" the landscape (including Tonle models, net ecosystem (carbon) production can be Sap) with precipitation, temperature, and winds. Climate calculated. Specific agriculture crops can be represented can be derived from surface observations (including at progressively finer resolution ("downscaling") with data telemetry back to a home base), satellites, and climate from multiple sources and models. The chemical loading models. The water movement (hydrology, Module 3) then (Module 6) is the input of chemicals (nutrients, toxics), as proceeds as the product of the climate acting across the product of hydrology and drainage basin properties. the templates of the landscape. Such models can then Lake water quality and net ecosystem production (NEP, Figure 12. A Schematic for the Execution of a River Basin/Coastal Dynamic Information Framework Observations/Measurements, Modeling Climate Forcing (2) Policy (10) Precip, Temp, Rn Water Movement (3) Informing the Local Met Records Soil Moist, ET, Runoff Decision Makers Regional Downscaling Stage/Discharge Regional Climate Models Water table (Local Hydrometer) Water use rules WB/WT (hydro) Model · Channel flow Economics (9) Drainage Basin (1) · Dams Value of assets "Scaled" structure · Irrigation DEM/Typography River Networks/riparian Soils (depth, texture, fertility) Landuse/Landcover Landscape Production (5) · Veg (Nat'l, Agriculture) class Chemical (6) Agriculture, (Industrial) · Veg functional attributes Dissolved, Sediment Crop yields · Biodiversity Direct river loadings Industrial resource · Infrastructure Hydrochem model Agriculture Models Landuse Change Models Fisheries (8) Lake Water Balance (4) Water Quality/NEP (7) Yield, Structure Water Level Chem, PPr (plankton, Catch Statistics Bathymetry Hyacinths) Biodiversity Stage Water chemistry Genetics Balance Model (Circulation) Hyacinth extent, production Recruit Model Nutrient/PPr Model 92 Module 7) is then driven by the loading and water balance. Bhutan for the project Distributed Hydrology Modeling The all-important fishery (Module 8) responds to external and DrukDIF Design and Development. The work was fishing pressure and NEP. presented to World Bank Water Week, in February 2007 (Quantitative Approaches to Optimizing Water, Land and Finally, the third set of modules addresses how economics Biodiversity Management) and the World Bank Sustainable and policy interact with the "biosphere." The economics Development Network (SDN) in February 2008 (Watershed (Module 9) represents the economic consequences and and Basin Management­Integrated Approaches across the feedbacks of the use of ecosystem goods and services. SDN Practice). Policy (Module 10) represents the legislative intersection with the management of the basin, including polices from land tenure decisions to specific, nominally informed, Lessons Learned and Future Directions legislation. It would be ideal to implement integrated water resources The concepts are equally relevant to progressively fined management along the continuum from land to ocean, as scales, down to individual projects. The construct allows a systematic process for the sustainable development, upscaling as well as relating how an individual project allocation and monitoring of water resource use in the or locale is "nested" in a larger region. The execution context of social, economic, and environmental objectives. of an architecture such as the one sketched out here At its simplest, integrated water resources management provides a framework for identifying specific field sampling is a logical and intuitively appealing concept. Its basis is requirements, from climate stations to suspended sediments that the many different uses of finite water resources are to economics of resources. The framework can then serve interdependent. High irrigation demands and polluted as the organizing structure for the activities of the Mekong drainage flows from agriculture mean less freshwater basin, including providing a basis for development of for drinking or industrial use; contaminated municipal management scenarios. A basin baseline can be executed and industrial wastewater pollutes rivers and threatens as organizing and analyzing the information required to bring ecosystems; if water has to be left in a river to protect each module "to life." fisheries and ecosystems, less can be diverted to grow crops; and so on. There are many more examples of the Applications to Existing World Bank basic theme that, in a rapidly changing environment, Projects unregulated use of scarce water resources is wasteful and inherently unsustainable. The VRB/DIF construct is not an esoteric, theoretical exercise. It is a construct that is not only realistic at this Relative to such goals, the analysis of the World Bank point, but practical. It is currently being applied to, and project portfolio makes several points. The projects deal developed from, emerging World Bank/GEF projects. The most directly with immediate services to be provided model and information framework was used to establish (water supply, irrigation, etc). The projects dealing with the baseline for the GEF­Zambezi Valley Market-Led the consequences of (sudden) change, such as floods or Smallholder Development Project: Baseline Data on Land droughts, are considerably fewer. While clearly the sectors Use, Biodiversity, and Hydrology. Through consultation within each one of the major categories are highly related with the Lake Victoria Basin Commission (LVBC), and to each other, in an ecosystem/water cycle sense, there the national teams for Kenya, Uganda, and Tanzania, the was surprisingly little overlap between them. Overall, this framework elements for a Lake Victoria basin Dynamic suggests the need for enhanced multi-sector cross-over Information Framework were created for the proposed and integration, and the need to pay more attention to the IAD Lake Victoria Environmental Management Plan 2. emerging ideas of "ecosystem goods and services." VIC is the core model for the ongoing World Bank/GEF project on the China 3H Basin project Mainstreaming A template for how to undertake integration and consider Adaptation to Climate Change into Water Resources services is provided by the broad-brush analysis of land- Management and Rural Development. It is being setup in ocean fluxes, summarizing the net transport of dissolved and 93 particulate materials from land to and through fluvial systems · Evaluate the restoration progress with explicit, quantita- to the sea. While the general patterns of fluxes are clear, tive criteria. there are (perhaps surprisingly) large uncertainties in the · Maintain long-term monitoring of restoration outcomes. magnitudes of specific fluxes, which could ultimately impact the ability to quantitatively determine possible outcomes What is the best way to incorporate the necessary of management actions. Part of the problem is scarcity of technologies to achieve these goals? Market-based reliable measurement campaigns. A substantial investment incentive systems provide rewards in the hope of promoting in improved measurements systems is needed. sustainable land and water stewardship in catchments and basins. They generally work on the concept that enhanced This template provides a basis for region-specific analyses resources management in upper catchments results in both and case studies, from hydropower to sea-level rise. productivity increases and ecosystem services that can Sediments coming off hill slopes impact coral reefs as well benefit stakeholders in the lower catchments and coastal as streams. The Mekong case study shows how inter- regions. In most incentive-based systems, the beneficiaries connected apparently separate sectors are. Decisions are charged an appropriate amount that is then equitably must consider the simultaneous and multiple interactions shared among the land users in the upper catchment. To of land use, reservoirs, and climate change. In many coastal be successful, the volume and quality of water flows and regions, these effects then propagate on local coastal associated benefits (for example, vegetation biomass and change, exacerbated by sea level rise. Combined with soil cover, reduced erosion, and added food and fiber the issues of greenhouse gases from reservoirs, mudflats production) provided by good land and natural resources and deltas, the land-to-ocean carbon cycle should be stewardship must be identified and reliably quantified. considered as part of global carbon trading. Creating an appropriate decision-making framework and The problem is, how does the development community deal institutional support structure that can be accessed by all with such complex, cross-over issues? New opportunities stakeholders is a critical step in the process. Key to being are emerging for nature-based adaptation in the coastal able to execute objectives is to be able to acquire, integrate, zone and new investments to secure coastal ecosystem and process the multiple sources of information required services through better management, restoration, good to do this. The Virtual River Basin/DIF concepts advanced governance, and so forth, as cost-effective adaptation here represent significant and practical advances towards strategies and alternatives to hard engineering (including providing such a framework. The capabilities now being flexible/adaptive infrastructure, source control) solutions in provided through earth system sciences, with its use of some cases. Incorporation of the concept of "ecosystem geospatial information from satellites combined with ground goods and services" should be a key part of the agenda. measurements, internet-accessible databases, and dynamic These become cost-effective ways of addressing global process-based models provide a new generation of tools. change issues, including building resilience into linked The capabilities for advanced visualization not only make it natural-human coastal ecosystems, and accommodating easier for the practitioner to understand his/her own results, future conditions (whatever they may be). but to convey them to a much broader audience, including decision makers. Choi (2004) suggests the following five steps: It must be made clear that the capabilities to do this are · Set realistic and dynamic goals for future environments, now eminently feasible and tractable. Perhaps the main rather than static goals based on the past. issue is to evaluate how best to overcome institutional · Assume multiple possible trajectories acknowledging constraints to adapting to new directions. The resource the unpredictable nature of ecological communities and agencies and ministries of host governments are obviously ecosystems. important. The role of transboundary organizations, such · Take an ecosystem or landscape approach (instead of an as the Mekong River Commission and Nile Basin Initiative, ad hoc approach) for both function and structure. could be enhanced. 94 Working as partners with current Work Bank Staff, the Degens, E. T. 1982. Riverine carbon: An overview. In Hydrology Expert Facility (HEF) is timely and particularly Transport of carbon and minerals in major world well-suited to act as a catalyst in moving such an agenda rivers, Part 1, edited by E. T. Degens. SCOPE/ forward. Applications could cover a diverse portfolio of UNEP Sonderbd. Mitteilungen aus dem Geologisch- Bank projects, from regions of melting glaciers to coral reefs Palaontologischen Institut der Universitat Hamburg 52, affected by sediments. An important starting point would be 1­12. to simply insure that the basics of hydrology are understood Degens, E. T., S. Kempe, and J. E. Richey. 1991. and applied (for example, that trees don't "produce" water). Biogeochemistry of major world rivers. New York: John Then, the more advanced technologies discussed above, Wiley and Sons. and represented by the HEF members, could be brought to Dasgupta, S., B. Laplante, C. Meisner, D. Wheeler, and J. bear on specific projects. Yan. 2007. The Impact of Sea Level Rise on Developing Countries: A Comparative Analysis World Bank Policy Research Working Paper 4136, February 2007 References Doten, C., L.C. Bowling, J.S. Lanini, E. Maurer, and D.P. Lettenmaier (2006). A spatially distributed model for the Aumont, O., J. C. Orr, P. Monfray, W. Ludwig, P. Amiotte- dynamic prediction of sediment erosion and transport Suchet, and J.-L. Probst. 2001. Riverine-driven in mountainous. Water Res Research. 42, W04417, interhemispheric transport of carbon. Global doi:10.1029/2004WR003829, Biogeochemical Cycles 15:393­406. Dunne, T, L. A. K. Mertes, R. H. Meade, J. E. Richey, and B. Bellwood, D.R., Hughes, T.P., 2001. Regional-scale R. Forsberg. 1998. Exchanges of sediment between assembly rules and biodiversity of coral reefs. Science, the flood plain and channel of the Amazon River in 292, 1532­1534 Brazil. Geology Society of America Bulletin 110 (4): Bishop, A. N., A. T. Kearsley, and R. L. Patience. 1992. 450­467. Analysis of sedimentary organic materials by scanning Hedges, J. I., and R. G. Keil. 1995. Sedimentary organic electron microscopy: The application of backscattered matter preservation: an assessment and speculative electron imagery and light element X-ray microanalysis. synthesis. Marine Chemistry 49:81­115. Organic Geochemistry 18:431­446. Hedges, J. I., J. R. Ertel, P. D. Quay, P. M. Grootes, J. E. Chen, C-T. A. 2004. Exchanges of carbon in the coastal Richey, A. H. Devol, G. W. Far-well, F. W. Schmitt, and seas. In: C. B. Field and M. R. Raupach (eds.) The E. Salati. 1986. Organic carbon-14 in the Amazon River Global Carbon Cycle: Integrating Humans, Climate, system. Science 231:1129­1131. and the Natural World. Island Press, Washington, D.C Hedges, J. I., P. G. Hatcher, J. R. Ertel, and K. J. Meyers- 526p. Schulte. 1992. A comparison of dissolved humic Choi, Y.D., 2004. Theories for ecological restoration in substances from seawater with Amazon River changing environment: towards "futuristic" restoration. counterparts by i3C¬NMR spectrometry. Geochimica Ecol. Restor. 19, 75­81 et Cosmochimica Acta 56:1753­1757. Cole, J. J., and N. F. Caraco. 2001. Carbon in catchments: Hedges, J. I., G. L. Cowie, J. E. Richey, P. D. Quay, R. Connecting terrestrial carbon losses with aquatic Benner, and M. Strom. 1994. Origins and processing metabolism. Marine and Freshwater Research 52 (1): of organic matter in the Amazon River as indicated 101­110. by carbohydrates and amino acids. Limnology and Costa-Cabral, M. C., J. E. Richey, G. Goteti, D. P. Oceangraphy 39 (4): 743­761. Lettenmaier, C. Feldkötter, and A. Snidvongs (2007). Hedges, J. I., R. G. Keil, and R Benner. 1997. What happens Landscape structure and use, climate, and water to land-derived organic matter in the ocean? Organic movement in the Mekong River Basin. Hydrological Geochemistry 27:195­212. Processes, vol 22, No. 12, pp1731­1746. Ittekkot, V., and B. Hawke. 1990. The terrestrial link in the Day, J.W. et al. 2007. Delta: Lessons from Hurricanes removal of organic carbon in the sea. Pp. 318­325 in Katrina and Rita. Science, 315, 1670­1684 Facets of modern biogeochemistry, edited by V. Ittekkot, 95 S.Kempe, W. Michaelis, and A. Spitzy. Heidelberg: 282:401­425. 1991. C, N, P, and S in rivers: From Springer-Verlag. sources to global inputs. In Interactions of C,N, P and Ittekkot, V., and R. W. P. M. Laane. 1991. Fate of S. Biogeochemical rycles, edited by R. Wollast, E. T. riverine particulate organic matter. Pp. 233­243 in MacKenzie, and L. Chou. Berlin: Springer Verlag. Biogeochemistry of major world rivers, edited by E. T. Milliman, J. D., and J. P. M. Syvitski. 1992. Geomorphic Degens, S. Kempe, and J. E. Richey. New York: John tectonic control of sediment discharge to the ocean: Wiley and Sons. The importance of small mountainous rivers. Journal of Jones, J. B., E. H. Stanley, and P. J. Mulholland. 2003. Long- Geology100 (5): 525­544. term decline in carbon dioxide supersaturation in rivers Milliman, J. D., K. L. Farnswortha, and C. S. Albertin. 1999. across the contiguous United States. Geophysical Flux and fate of fluvial sediments leaving large islands Research Letters 30:2-1-2-4. in the East Indies. Journal of Sea Research 41:97­107. Kao, S. J., and K.-K. Liu. 2002. Exacerbation of erosion Raymond, P. A., and J. E. Bauer. 2001. Riverine export induced by human perturbation in a typical Oceania of aged terrestrial organic matter to the North Atlantic watershed: Insight from 45 years of hydrological Ocean. Nature 409:497­500. records from the Lanyang-Hsi River, northeastern Raymond, P. A., and J. J. Cole. 2003. Increase in the export Taiwan. Global Biogeochemical Cycles 16:1­7. Keil, of alkalinity from North America's largest river. Science MG., D. B. Montlucon, F. G. Prahl, and J. I. Hedges. 301:88­91. 1994. Sorptive preservation of labile organic matter in Richey, J. E., J. I. Hedges, A. H. Devol, P. D. Quay, R. marine sediments. Nature 370:549­552. Victoria, L. Martinelli, and B. R. Forsberg. 1990. Keil, R. G., L. M. Mayer, P. D. Quay, J. E. Richey, and J. I. Biogeochemistry of carbon in the Amazon River. Hedges. 1997. Losses of organic matter from riverine Limnology and Oceanography 35:352­371. particles in deltas. Geochimica et Cosmochimica Acta Richey, J. E., J. M. Melack, A. A. K. Aufdenkampe, V. 61 (7): 1507­1511 M. Ballester, and L. Hess. 2002. Outgassing from Likens, G., F. Mackenzie, J. Richey, J. Sedell, and K. Amazonian rivers and wetlands as a large tropical Turekian, eds. 1981. Flux of organic carbon by rivers to source of atmospheric CO2. Nature 416:617­620. the oceans. National Research Council/DOE Conf- Richey, J.E. 2004. Pathways of Atmospheric CO2 through 009140.Washington,U.S. Department of Energy. Fluvial Systems. Pp. 329­340. In: C. B. Field and M. R. Ludwig, W. 2001. The age of river carbon. Nature Raupach (eds.) The Global Carbon Cycle: Integrating 409:466­467. Humans, Climate, and the Natural World. Island Press, Ludwig, W, J. L. Probst, and S. Kempe. 1996. Predicting the Washington, D.C 526p. oceanic input of organic carbon by continental erosion. Richey, J.E. 2005. Global River Carbon Biogeochemistry. Global Biogochemical Cycles 10 (1): 23­41. (in. M. Anderson, Ed.) Encyclopedia of Hydrological McCready, S., Birch, G.F., Long, E., Spyrakis, G., Greely, Sciences Part 15. Global Hydrology. John Wiley & C.R., 2006. Predictive abilities of numerical sediment Sons, Ltd DOI:10.1002/0470848944.hsa191 quality guidelines in Sydney Harbour, Australia, and Richey, J.E., and E. Fernandes. (2007). Towards Integrated vicinity. Environmental International, 32, 638­649 Regional Models of Transboundary River Basins in Mayer, L. M. 1994. Adsorptive control of organic carbon Southeast Asia: Lessons Launched from Waters and accumulation in continental shelf sediments. Watersheds. Journal of Contemporary Water Research Geochimica et Cosmochimica Acta 58:1271­1284. & Education. 136: 28­36. Meade, R. H., T. R. Yuzyk, and T. J. Day. 1990. Movement Sarmiento, J. L., and E. T. Sundquist. 1992. Revised budget and storage of sediment in rivers of the United States for the oceanic uptake of anthropogenic carbon-dioxide. and Canada. Pp. 255­280 in Surface water hydrology, Nature 356 (6370): 589­593. edited by M. G. Wolman and H. C. Riggs. Boulder, CO: Schlesinger, W. H., and J. M. Melack. 1981. Transport of The Geology of North America, Geological Society of organic carbon in the world's rivers. Tellus 33:172­187. America. Smith, S. V., W. H. Renwick, R. W. Buddemeier, and C. Meybeck, M. 1982. Carbon, nitrogen and phosphorus J. Crossland. 2001. Budgets of soil erosion and transport by world rivers. American Journal of Science deposition for sediments and sedimentary organic 96 carbon across the conterminous United States. Global Ver, L. M., E T. Mackenzie, and A. Lerman. 1999. Biogeochemical Cycles 15 (3): 697­707. Biogeochemical responses of the car-bon cycle to Stallard, R. F. 1998. Terrestrial sedimentation and the natural and human perturbations: Past, present and carbon cycle: Coupling weathering and erosion to future. American Jour¬nal of Science 299:762­801. carbon burial. Global Biogeochemical Cycles 12:231­ Vorosmarty, C. J., K. Sharma, B. Fekete, A. H. Copeland, J. 257. Stanley, D.J., and Warne, A.G., 1998. Nile Delta Holden, J. Marble, and J. A. Lough. 1997. The storage in its destruction phase. Journal of Coastal Research, and aging of continental runoff in large reservoir 14(3), 794­825 systems of the world. Ambio 26:210­219. Takeuchi, K., H. Ishidaira, H. Sawada, and T. Masumoto. Vorosmarty, C. J., M. Meybeck, B. Fekete, K. Sharma, P. 2008. Special Issue: Studies on the Mekong River Green, and J. Syvitski. 2003. Anthropogenic sediment Basin--Modelling of Hydrology and Water Resources. retention: Major global impact from registered river Hydrological Processes 22. impound¬ments. Global and Planetary Change Tong, P., Y. Auda, J. Populus, M. Aizpuru, A. al Habshi, and 39:169­190. F. Blasco. 2004. Assessment from space of mangroves World Commission on Dams. 2000. Dams and evolution in the Mekong Delta, in relation to extensive development: A new framework for decision-making. shrimp farming. Int. J. Remote Sensing 25: 4795­4812. London: Earthscan. 97 6. Managing the Real Water Consumer: Evapotranspiration Peter Droogers Walter Immerzeel Future Water, The Netherlands Abstract to water and irrigation is a major determinant of land productivity and the stability of yields. While water related problems are diverse and location specific, water shortage is frequently the most pressing However, in sub-Saharan Africa, only 4 percent of the area issue in many developing countries. A central challenge in production is under irrigation, compared with 39 percent for the next decades is the increasing international and in South Asia and 29 percent in East Asia. Investments inter-sector competition for scarce water, in the context of in improving the productivity of water in agriculture growing demand for food and uncertain impacts of climate are becoming increasingly critical. Climate change change. and reduced glacial runoff are raising uncertainties in agriculture at the same that growing water scarcity and the This paper discusses the role of agriculture as one of the rising costs of large-scale irrigation schemes are creating main causes of water-related problems, as well as the opportunities for enhancing productivity that should be importance of evapotranspiration as the dominant water explored. consumer. Experiences in China and Egypt serve as examples for a discussion of methodologies to support Agriculture, and more specifically irrigated agriculture, is policy makers and water administrators and assist them in often regarded as one of the main causes of water related managing evapotranspiration. problems. The 2008 World Development Report claims: "Agriculture is by far the largest user of water, contributing Several models, ranging from those that are completely to water scarcity." The very same report also concludes physically-based to conceptual allocation models, are that "Without irrigation, the increases in yields and output discussed as policy support tools, some of which may that have fed the world's growing population and stabilized prove to be too complex for practical applications. The food production would not have been possible." In general, paper advocates the inclusion of a combination of remote irrigated land productivity is more than double that of rainfed sensing and simulation models in policy support tools and land and evapotranspiration is the main consumer of water introduces the concept of scenario-based modeling as a (see Figure 1). better alternative to support policy makers. However, increasing complexity, and insufficient knowledge and tools to evaluate the consequences of alternative Introduction interventions constrain the ability of policy makers and planners to make appropriate decisions. Furthermore, Water to sustain food production plays a key role in efforts important misconceptions often underlie strategies to reach the Millennium Development Goals. Access proposed to address these problems. 99 Figure 1. Global Water Use for Land policies, especially in irrigation science (Allen et al. 2005; Seckler et al. 2002; Molden 2007; Perry 2008; Droogers et Baseflow Evapotranspiration 2% 54% al. 2000). For example, irrigation science has traditionally focused Direct runoff 36% on improving "efficiency" while completely ignoring what happens with the "non-efficient" water. In many cases Open water this "non-efficient" water is reused by downstream users, evaporation pumped from the groundwater, serves to reduce salt 8% intrusion, or contributes to wetlands. It is quite common Source: Shiklomanov 1999 that a substantial amount of these "losses" is beneficial to the poorest in a region. From a discussion of these Evapotranspiration efficiency concepts, Perry (2007) showed that the following conclusions may be drawn: This section discusses the concept of evapotranspiration and the tools available to policy makers. · high efficiency reflects low losses; · losses are a non-recoverable waste of resources; Concepts · reductions in ``losses'' will mean that more of the input is available for alternative uses; A persistent misconception is that irrigated agriculture is the · high efficiency is ``good.'' main consumer of water (Figure 2). This misconception is mainly based on a combination of ambiguous terminology The concept of "irrigation in the basin" has been promoted and undefined domains. Regarding terminology, it is often and partly put into practice over the last decade to unclear what is meant by "consumers," "users," "efficiencies," overcome the misconceptions that arise from considering "losses," and other such terms. This has led to confusing only the irrigation domain (Seckler 1996; Kite and Droogers Figure 2. Global Water Use Rainfall (thousands of cubic kilometers per year) Bioenergy 110 forest products 100% grazing lands Green Water Blue Water biodiversity Rivers Soil Wetlands Crops livestock Crops livestock Landscape 56% moisture Lakes aquaculture from rain Groundwater Rainfed agriculture Irrigated 4.5% agriculture Water storage aquatic 0.6% 1.4% biodiversity fisheries Open water evaporation Cities and 1.3% industries 0.1% Ocean 36% Green Water Blue Water Source: Molden 2007. 100 1999). This line of thinking is also reflected in the first of of total ET, especially at the time of crop emergence when eight recommendations in the Comprehensive Assessment leaf cover is very limited. Crop transpiration is, in fact, the of Water Management in Agriculture (Molden 2007): only term that can be considered as a productive use, since it supports vegetation growth. However, it is important to Change the way we think about water and note that less than one percent of the transpired water is agriculture. Thinking differently about water is actually retained by the vegetation. Carbon dioxide is the essential for achieving our triple goal of ensuring only carbon source for plants and in order to obtain it, plants food security, reducing poverty, and conserving have to open their stomata. Water diffuses outwards during ecosystems. Instead of a narrow focus on rivers this process, and it could be claimed that plants have to and groundwater, view rain as the ultimate source transpire water to obtain the required carbon. In addition, of water that can be managed. (Policy action #1). plants might also transpire some water to maintain their internal temperature. The basic concept put forward in this policy action is that regardless of the policies that are put into place, Ignoring ET and simply reducing water diversions almost the ultimate restriction is always the total rainfall in a always results in a reduction in return flow back to the basin (provided that no inter-basin transfer occurs). resource. Therefore, the quantity of net consumption Acknowledging that rainfall is the only source of water, it by an irrigation system may be largely unchanged by a could be claimed that there is effectively only one ultimate conservation program. To effectively create "new" water consumer of water: evapotranspiration. In other words, in in a regional context, unless directly upstream of a salt the same way that rain can be regarded as the ultimate sink, a conservation program must in some way reduce source of water on the supply side of the hydrological ET or improve return flow quality, and not simply reduce equation, it could said that evapotranspiration is the only diversions. Reduction of crop ET will almost always reduce term on the consumer side. crop yields, unless evaporation from the soil is reduced without reducing plant transpiration. This simple fact has tremendous impact on policies. In situations where the non-evaporated components of In fact, the performance of an irrigated area can only be irrigation diversions return to the fresh water resource for evaluated by examining the irrigation water when it leaves reuse by others, conservation programs may not stretch the defined boundaries of interest. The applied irrigation water supplies or "save" water in the region, especially water can be placed into five categories (Clemmens and in the long term. Water conservation programs should Allen 2005): fundamentally be evaluated against the general principle that the only real loss of water from an irrigation project is 1. Water consumed by the crop within the area under con- by the process of evaporation from open water surfaces, sideration for beneficial purposes. evaporation from soil and wet foliage, transpiration from 2. Water consumed within the area under consideration but vegetation, and flows into saline sinks. In fact, one should not beneficially. go back to the fundamental hydrologic concepts that were 3. Water that leaves the boundaries of the area under already recognized by the early Greek philosophers, and consideration, but is recovered and reused by the same mathematically underpinned in the 18th century by Bernoulli party or by a "downstream" party. and Chezy, among others (Hubart 2008). 4. Water that leaves the boundaries of the area under con- sideration, but is either not recovered or not reusable. The term evapotranspiration (ET) relates to three 5. Water that is in storage within the area under consider- components: (1) interception evaporation, (2) soil ation. evaporation, and (3) crop transpiration. The interception evaporation for agricultural crops is often around 10 percent In practice, much emphasis in irrigation engineering has of total ET, while for forests this can range as high as 80 been on category 3 using the concept of efficiencies, percent to 90 percent, depending on prevailing climate while categories 2 and 4 are those that deserve greater conditions. Soil evaporation can be a substantial amount recognition by policy makers and water managers. 101 A similar approach based on the diversion of water Policy Support Tools allocations was advocated by Perry (2007), who stated that all water that enters a certain domain (irrigation, From the previous section it is clear that a focus on ET is streamflow, and rainfall) can be classified into one of four not only justified, but also required to understand water- terms: beneficial consumption, non-beneficial consumption, related issues and improve water management. The concept recoverable fraction, and non-recoverable fraction. of ET management requires innovative and policy-oriented supporting tools. Figure 3 provides a conceptual framework 1. Beneficial consumption is water evaporated or trans- highlighting that a clear distinction should be made between pired for the intended purpose; for example, evaporation understanding and monitoring the past and the current from a cooling tower, or transpiration from an irrigated situation on the one hand, and pro-active planning using crop. modeling tools, on the other hand. 2. Non-beneficial consumption is water evaporated or transpired for purposes other than the intended use; In terms of monitoring ET, special emphasis should be for example, evaporation from water surfaces, riparian placed on remote sensing. One could safely claim that vegetation, or waterlogged land. remote sensing is the only tool available nowadays to 3. Recoverable fraction is water that can be captured and monitor ET over large areas. reused; for example, flows to drains that return to the river system, percolation from irrigated fields to aquifers, Over the last decades, various ET algorithms have or return flows from sewerage systems. been developed to make use of remote sensing data 4. Non-recoverable fraction is water that is lost to further acquired by sensors on airborne and satellite platforms. use; for example, flows to saline groundwater sinks, The reported estimation accuracy of various methods deep aquifers that are not economically exploitable, or varied from 67 percent to 97 percent for daily ET, and flows to the sea. greater than 94 percent for seasonal ET, indicating that these methods have the potential to estimate regional ET Based on these discussions, it is clear that only by accurately (Gowda et al. 2008). Only in the last decade considering the basic concepts of hydrology and continuity have these tools made the transition from research to of mass can proper intervention options be explored. When application. In particular, the SEBAL approach, introduced water is scarce, key areas of attention would be to reduce in 1998 (Bastiaanssen et al. 1998), and some successors non-beneficial consumption, and to reduce non-recoverable (SEBS: Su 2002; METRIC: Allen et al. 2007), have flows to the extent that proper hydrological analysis shows been influential in promoting acceptance of these remote that no unintended consequences of such reductions occur. sensing approaches into operational and strategic decision Based on this conclusion, it is essential that all terms of the support systems. water balance should be known. All policy should be based on comparing different options (interventions) for the future, and requires appropriate planning tools in the form of simulation models (Droogers Figure 3. The Concept of Using Policy and Kite 1999). Over the last decades, models have Oriented Supporting Tools been used successfully to support policy making by first improving the understanding of processes, and then · Remote sensing by conducting scenario analyses. The main reason for Past Understand past water Trend resources · Observations the success of models in promoting the understanding · Analysis Understand current water · Statistics of processes is that they can provide output over an resources Today unlimited time-scale, at an unlimited spatial resolution, ? and for sub-processes that are difficult to observe (for Future Options for future · Models example, Droogers and Bastiaanssen 2002). The most · Technical important benefit of applying models, however, is their · Socioeconomic · Policy oriented use to explore different scenarios. These scenarios can 102 Figure 4. Spatial and Physical Detail of capture aspects of the water management system that Hydrological Models cannot directly be influenced, such as population growth and climate change (Droogers and Aerts 2005). These Continent model outputs are often referred to as projections. In Podium contrast, management scenarios or interventions can be STREAM simulated where water managers and policy makers can Basin SLURP Spatial scale WSBM make decisions that will have a direct impact. Examples of SWAT the latter are changes in reservoir operation rules, water System WEAP allocation among sectors, investment in infrastructure Future View such as water treatment or desalinization plants, and SWAP Field agricultural/irrigation practices. WaterMod High Low A huge number of hydrological models exist, and applications are growing rapidly. The number of pages on the Internet including "hydrological model" is over 300,000 Although much information is available on agricultural (Google, November 2008). Using the same search engine water allocation to individual fields, information on real with "water resources model" results in 13 million pages. water consumption (actual ET) is lacking. Moreover, water Therefore, a critical question for hydrological model studies consumption at the basin scale is essentially unknown. The is related to the selection of the most appropriate model. aim of the GEF World Bank project "Hai Basin Integrated One of the most important issues to consider is the spatial Water and Environment Management Project" is to manage scale to be incorporated in the study and how much ET to restore groundwater levels and maintain outflow to the physical detail needs to be included. Figure 4 illustrates Bohai Sea (Bastiaanssen et al. 2008). the negative correlation between the physical detail of a model and the spatial scale of the application. This figure In this project, ET from the Hai basin is calculated also indicates the position of commonly used models in this using remote sensing measurements. Based on these continuum. observations, allocation plans for each county are under development. Future scenarios to reduce evapotranspiration are being explored by using various modeling tools. A Examples typical example of some of the policy-supporting tools is the basin-wide water consumption map shown in Figure 5. Several projects started over the last years take ET into This map has been aggregated per county and is currently consideration as a key component of the overall objective to used to define water quotas. An innovative aspect is that improve water management. Three of these projects will be these quotas will not be based on allocations, but on real summarized in the following sections. They are China's Hai water consumption (actual ET). A major advantage of basin, Egypt's Nile basin, and a hypothetical basin derived this approach is that allocations that yield return flows to from a real situation in northern Africa. downstream counties are not considered as consumption. China's Hai Basin Various modeling tools have been set up to support county The Hai basin in the People's Republic of China water managers in the development of plans to reduce ET. is experiencing groundwater overdraft, resulting in Figure 6 shows an example of exploring the impact of an dropping groundwater levels and water shortages. The intervention. This example shows the impacts on ET and water balance shows a non-sustainable situation, with groundwater of reducing irrigation by 50 percent. more water leaving the basin than entering it. Outflow from rivers in the Hai basin barely reach the Bohai The Hai basin project was ongoing in 2008, but the uptake Sea, and most of the water leaves the area through of the concept of ET management is impressive. Chinese evapotranspiration. policy makers and water managers have developed their 103 Figure 5: Actual Annual Evapotranspiration (2002) for the Hai Basin own remote sensing applications and suite of models to much water is actually used in contrast to the amount of focus on real consumption rather than on allocations. water that is allocated? The cornerstone of the analysis was remotely-sensed ET estimates of the Nile (Bastiaanssen et Egypt's Nile Basin al. 2003; Noordman and Pelgrum 2004). Debates on the actual water balance of the Egyptian part of the Figure 7 shows the actual ET over the entire Nile Nile basin have persisted over decades. The political sensitivity Basin in Egypt for one particular year (2007). By using of the Nile Water Agreements of 1959 has made it virtually comparable information from other years, the long-term impossible to obtain realistic numbers on actual consumption. actual ET for irrigated lands is estimated at 32 km3 y-1, The agreed 55.5 km3 entitlement is often equated to the total while ET from non-irrigated areas (mainly from seepage) amount of water consumed. However, expansion of irrigated is about 8 km3 y-1. Actual water allocations over the areas, large amounts of uncommitted flows to the sea, and last decade, as recorded at Aswan, are higher than the water savings attempts have made the situation even more 55.5 km3 entitlement, and are on average 68 km3 per confusing. The main problem is that no information at all on real year. Table 1 shows water balances for the entire Nile water consumption (actual ET) has been available. basin based on these figures and including some other data sources. The study showed that focusing on real A recent study (Droogers et al. 2008b) combined various water consumption, based on unbiased non-political completed studies focusing on the main question: How estimates from remote sensing, provides decision makers 104 Figure 6: Scenario Analysis Applicable to Counties in the Hai Basin, China Impact of Reducing Irrigation by 50 Percent (right) Compared to the Current Situation (left) on ET (top) and Groundwater (bottom) with the necessary information to discuss the Nile water is that the model is able to mimic reality. Moreover, in many resources. cases relative model accuracy (comparing model baseline with model scenario) is much higher than the actual Scenario-Based Modeling accuracy (comparing model to observations) (for example, Bormann 2005; Droogers et al. 2008a). As indicated earlier in this paper, various modeling tools exist ranging from completely physically-based models to Droogers and Perry (2008) demonstrate concepts of conceptual allocation models. Policy makers require models scenario analysis for a hypothetical basin, derived from that have a focus on scenario analyses, rather than models a real situation in Northern Africa. The hypothetical that are too complex to use for practical applications. There basin comprises four catchment areas and two irrigation are too many modeling studies where the final conclusion systems, one upstream and one downstream in the basin 105 Figure 7. Actual Evapotranspiration for 2007 in the Nile Delta Based on Remote Sensing Table 1. Estimated Water Balances improve the efficiency of the irrigation systems. The latter in the Nile Basin in Egypt (For a are based on observations that the efficiency, defined as Representative Year Under Current the amount of water allocated to a system divided by the Conditions) uptake of plants, is approximately 50 percent. Based on this number, it was concluded that a huge amount of water In (km3) Out (km3) could be saved. Outflow Aswan 68.0 ET irrigation 32.0 However, a first basin-wide analysis showed that by far Rainfall 0.5 ET other 8.0 the major consumers of water in the basin are forests and Industry/domestic 1.0 natural vegetation. Actual evapotranspiration from irrigated ET seepage 2.3 crops is about 20 percent of overall ET in the basin. Since Outflow to sea (rest) 25.2 managing ET from forests and natural vegetation is difficult, Total 68.5 Total 68.5 the focus here remains on irrigated agriculture. Note that managing non-irrigated water consumption has been under debate for reforestation projects, as in many cases these (Figure 8). Groundwater tables in the basin are dropping new forests consume more water by ET compared to the at alarming rates and interventions are discussed to original vegetation (Calder 1999). 106 Figure 8. Hypothetical Basin Including the Four Catchment Areas Considering only the irrigation sector, it is important to · Evapotranspiration should be considered as the main evaluate the different locations of the two irrigation systems consumer of water, in the same way as rainfall is re- in the basin. Irri01 is located upstream and outflow of garded as the only source of water. this system might be reused downstream, while outflow · Irrigation should always be considered in a location- of the downstream system is lost from the basin. The specific (basin) context. water balance of the two systems is depicted in Figure 9, · Remote sensing data can support policy making by indicating that about 50 percent of the incoming water evaluating current and past water consumption (ET). (irrigation and rainfall) is consumed by ET. In terms of water · Simulation modeling supports policy making by evaluat- saving programs, it is important to recognize three different ing different scenarios (interventions). outflow components: · beneficial outflow: crop transpiration Figure 9. Water Balance of the Two · non-beneficial outflow: soil evaporation, drainage (down- Irrigation Systems stream) · reusable outflow: percolation (upstream), drainage 125 Upstream Downstream (upstream) 100 Precipitation 75 By estimating these three terms, different interventions for 50 Relative Flows (%) the upstream and the downstream irrigation systems can be Irrigation 25 assessed to obtain the real water saving. 0 ­25 Transpiration The Way Forward ­50 Soil evaporation ­75 Groundwater recharge The main message conveyed in this paper is summarized by ­100 Runoff the following four points: ­125 107 Figure 11. Remote Sensing of Actual In practice this means that projects should include an ET, Rio Grande, New Mexico, June 16, evaluation of the full hydrological cycle considering the 2003 appropriate domain. The preferred domain in this respect is not the irrigation system but a hydrological (sub) basin. In cases where the entire basin is not considered, one should understand the upstream and downstream interactions of the domain under study. Policy supporting tools should include a combination of remote sensing and simulation models. A somewhat unexplored subject is the role that remote sensing information can play in calibrating models (Immerzeel and Droogers 2008). Currently, model development has progressed to the extent that further development is hardly required for practical applications; the main challenges are in obtaining the data and information necessary as inputs Source: Hong and Hendrickx, 2003 to these models (Immerzeel et al. 2008). Typical examples of remote sensing products that have emerged recently to the benefit of user groups include (1) actual rainfall provided Figure 12. Typical Example of by the TRMM satellite, (2) actual ET information available GRACE Results Showing Changes in on a near real-time basis, and (3) changes in groundwater Groundwater for the Mississippi Basin, observed from space using the GRACE satellite (Figures July 2005 10, 11, and 12). This information is essential to obtain realistic model outputs that can be used to explore the impact of interventions. A typical example of such an approach is the ongoing IFAD project in Kenya on Green Water Credits (Dent and Kauffman Figure 10. Satellite-estimated Precipitation (TRMM) 22­28 October 2008 Source: Rodell et al., 2006 2008). By combining remotely sensed information and modeling tools, a much better understanding of the impact of certain interventions on all water related issues, including erosion, can be obtained (Figures 13 and 14). Finally, the phrase by Lord Kelvin "To measure is to know" can be Source: http://trmm.gsfc.nasa.gov expanded to "To measure ET is to know where to act." 108 Figure 13. Scenario Analyses for the Tana Basin, Kenya. Spatial Variation of Increases in Actual Crop Transpiration Under the Enhance Water Productivity Scenario Figure 14. A Comparison of Three Water Allen, R.G., A. J. Clemmens, L. S. Willardson. 2005. Agro- Management Scenarios in the Tana hydrology and irrigation efficiency. ICID Session on Basin, Kenya Agrohydrology and Efficiency. Bastiaanssen W.G.M., M. Menenti, R.A. Feddes, A.A. 80 Holtslang. 1998. A remote sensing surface energy 60 balance algorithm for land (SEBAL): 1. Formulation. J Hydrol 212­213:198­212 40 Bastiaanssen, W., E. Noordman, H. Pelgrum. 2003. Change (%) 20 Monitoring summer crops under changing irrigation 0 practices: A Remote Sensing Study in the North- ­20 western Nile Delta for the Irrigation Improvement Project 1995­2002. Report WaterWatch. ­40 Bastiaanssen, W.G.M., W. Bingfang, D. Olson, J. Liping. ­60 2008. Real water savings and ET reduction: the Hai ­80 Basin paradigm, Water Front, World Water Council, Crop Soil Groundwater Surface Sediment Transpiration Evaporation Recharge Runoff Loss Stockholm (mm/y) (mm/y) (mm/y) (mm/y) (ton/ha/y) Bormann, H. 2005. Evaluation of hydrological models Contour Strips Mulch Ridges for scenario analyses: signal-to-noise-ratio between scenario effects and model uncertainty. Advances in Geosciences, 5: 43­48. References Calder, I.R. 1999. The Blue Revolution, Land Use and Integrated Water Resources management. Earthscan, Allen R.G., M. Tasumi, R. Trezza. 2007. Satellite-based London, ISBN 1 85383 634 6. energy balance for mapping evapotranspiration with Clemmens, A.J., R.G. Allen. 2005. Impact of agricultural internalized calibration (METRIC)-Model. ASCE J water conservation on water availability. Proceedings of Irrigation Drainage Eng. Vol. 133, No. 4, 380­394 the EWRI World Water and Environmental Resources 109 Congress 2005: Impacts of Global Climate Change. evapotranspiration. Journal of Hydrology 349: May 15­19, 2005, Anchorage, Alaska, USA, 14 p. 411­424. Dent, D., S. Kauffman. 2008. Green Water Credits: proof of Immerzeel, W.W., A. Gaur, S.J. Zwart. 2008, Integrating concepts. Report 7, ISRIC, Wageningen. remote sensing and a process-based hydrological model Droogers, P., A. Van Loon, W. Immerzeel. 2008a. to evaluate water use and productivity in a south Indian Quantifying the impact of model inaccuracy in climate catchment. Agricultural Water Management 95: 11­24. change impact assessment studies using an agro- Kite, G.W., P. Droogers. 1999. Irrigation modeling in a basin hydrological model. Hydrology and Earth System context. Water Resources Development 15: 43­54. Sciences 12: 1­10. Molden, D. (Ed). 2007. Water for Food, Water for Life: A Droogers, P., C. Perry. 2008. Scenario Based Water Comprehensive Assessment of Water Management Resources Model to Support Policy Making. in Agriculture. London: Earthscan, and Colombo: FutureWater report (in press). International Water Management Institute. Droogers, P., C. Perry, W. Immerzeel. 2008b. Application of Noordman, E., H. Pelgrum. 2004. Monitoring winter crops remote sensing in national water plans: Demonstration under changing irrigation practices: A Remote Sensing cases for Egypt, Saudi-Arabia and Tunisia. FutureWater Study in the North-western Nile Delta for the Irrigation report (in press). Improvement Project 1997/98­2002/03. Report Droogers, P., J. Aerts. 2005. Adaptation strategies to WaterWatch. climate change and climate variability: a comparative Perry, C. 2007. Efficient Irrigation; Inefficient study between seven contrasting river basins. Physics Communication; Flawed Recommendations. Irrigation and Chemistry of the Earth 30: 339­346. and Drainage 56: 367­378 Droogers, P., W.G.M. Bastiaanssen. 2002. Irrigation Perry, C. 2008. Efficient irrigation; inefficient performance using hydrological and remote sensing communications; flawed recommendations? ICID/UEA modeling. Journal of Irrigation and Drainage Engineering international seminar--Towards a political ecology of 128: 11­18. irrigation & water use efficiency and productivity. 6th Droogers, P., G. Kite, H. Murray-Rust. 2000. Use of simulation November 2008. East Anglia. UK models to evaluate irrigation performance including water Rodell, M., J. Chen, H. Kato, J. Famiglietti, J. Nigro, C. productivity. Irrigation Science 19: 139­145. Wilson. 2006. Be consistent with use of brackets Droogers, P., G. Kite, 1999. Water productivity from for years. Estimating ground water storage changes integrated basin modeling. Irrigation and Drainage in the Mississippi River basin (USA) using GRACE. Systems 13: 275­290. Hydrogeology Journal doi:10.1007/s10040-006-0103-7 Gowda, P.H., J.L. Chavez, P.D. Colaizzi, S.R. Evett, T.A. Seckler, David. 1996. The new era of water resources Howell, J.A. Tolk. 2008. ET mapping for agricultural management: From "dry" to "wet" water savings. IWMI water management: present status and challenges. Irrig Research Report 1. Colombo, Sri Lanka: International Sci 26:223­237 Water Management Institute. Hong, S.H., J.M.H. Hendrickx. 2003. Spatio-temporal Seckler, D., D. Molden, R. Sakhivadivel. 2002. The concept distributions of evapotranspiration and root zone soil of efficiency in water-resources management and moisture in the Middle Rio Grand basin. Presentation at policy. Appendix A. A Note on Transpiration. 2002. In J. 3rd SAHRA meeting in Tucson, AZ. W. Kijne, R. Barker and D. Molden, Water Productivity Hubbart, Jason (Lead Author); Jim Kundell (Topic Editor). in Agriculture: Limits and Opportunities. CAB 2008. "History of hydrology." In: Encyclopedia of International, Wallingford, UK. Earth. Eds. Cutler J. Cleveland (Washington, D.C.: Shiklomanov, I.A. 1999. State Hydrological Institute (SHI, Environmental Information Coalition, National Council St. Petersburg) and United Nations Educational, for Science and the Environment). Su, Z. 2002. The surface energy balance system (SEBS) Immerzeel, W.W., P. Droogers. 2008. Calibration of a for estimation of turbulent fluxes. Hydrol Earth Syst Sci distributed hydrological model based on satellite 6:85­99 110 7. Addressing the Links between Hydrology and Watershed Climate, Soil, and Vegetation Ignacio Rodríguez-Iturbe Department of Civil and Environmental Engineering, Princeton University Abstract their livelihoods. Many more depend on forest product industries, leather goods industries, cotton Among the greatest challenges facing sustainable and woolen textile industries, and food processing development are those derived from climate change. These industries for their jobs. challenges are varied and also qualitatively different in character. In carrying out a realistic evaluation of the impact A strategy for eradicating poverty will not succeed of climate change on ecosystems, it is not sufficient to if an economy's environmental support systems are merely account for changes in mean responses to climatic collapsing. If croplands are eroding and harvests variability. Changes in the dynamics of less frequent and are shrinking, if water tables are falling and wells stronger rainfall events will have larger consequences for are going dry, if rangelands are turning to desert the assimilation process and survival of vegetation. An and livestock are dying, if fisheries are collapsing, increase in the intensity of rainfall events also leads to other if forests are shrinking, and if rising temperatures type of ecohydrological consequences especially related are scorching crops, a poverty-eradication to soil erosion. As a result, farming activities are either program--no matter how carefully crafted and well dramatically reduced or supplanted by pastoral subsistence, implemented--will not succeed." constraining sustainable development. This paper focuses on those challenges where ecohydrology will contribute in It has been well documented (by Diamond [2005], among a most decisive manner to the necessary understanding for others) that the fundamental reasons that earlier civilizations their amelioration and management. declined were tied to the environment rather than directly to the economy. Ecohydrology plays a key role in sustainable development through environmental stewardship. Introduction Ecohydrology is the science that studies the hydrologic To begin, it is worth quoting Brown (2006) at length dynamics responsible for ecological patterns and processes because his words convey a clear picture of the reason (Rodriguez-Iturbe 2000). As such, it takes place at the why ecohydrologic factors are a central part of regional and frontiers of environmental sciences where the historically global sustainable development. distinct disciplines of biology and physical sciences converge. According to Hedin et al. (2002)," [T]his "The health of an economy cannot be separated disciplinary convergence will, over the next several decades, from that of its natural support systems. More transform our understanding of basic processes that control than half the world's people depend directly on the stability and sustainability of natural environmental croplands, rangelands, forests, and fisheries for systems. The ensuing findings will have extraordinary 111 implications for our abilities to predict and manage how where average precipitation changes only very little are humans impact the health of ecosystems across local, likely to experience serious impacts resulting from changes regional, and global scales. Such knowledge is a critical in the dynamics of the precipitation regime. In 2001, the component of a safe, sustainable, and prosperous future." Intergovernamental Panel on Climate Change reported an increase in precipitation (rainfall and snowfall) of between 5 percent and 10 percent across most mid and high latitudes Ecohydrologic Implications of Climate of the Northern Hemisphere. Although there are regions Change that will experience decreased precipitation, the important point is that extremes will intensify. In other words, droughts It is universally accepted that the world has become will become longer and more oppressive and storms will warmer during the last 150 years. Moreover, there is ample be more frequent and intense. Some spectacular events evidence that global temperature fluctuations are correlated of this type have already been widely reported, including with the concentration of carbon dioxide in the atmosphere. devastating hurricanes like Katrina or droughts like those Carbon dioxide and other gases absorb radiation in the afflicting Sudan. While changes in many other ecosystems infrared spectrum causing a greenhouse effect. Since the occur more subtly, their consequences are no less serious. light from the sun contains energy in all wavelengths and There is growing evidence that predicted changes in rainfall the radiant heat from the earth is mainly in the infrared regime because of climate change will reduce the primary spectrum, more energy is kept than is left out, leading to an productivity of ecosystems and induce shifts in community increase of the atmospheric temperature. This increase in composition as well as loss of biodiversity. In water temperature takes place at the global scale and presents controlled ecosystems in particular, hydroclimatic variability large spatial fluctuations which then, directly and indirectly, together with soil and plant characteristics produce the soil have enormous ecohydrological impacts. moisture dynamics that largely control vegetation conditions. The most important characteristics of hydroclimatic This paper groups the ecohydrologic impacts of climate variability are changes in temperature and in the frequency change in two large categories that mainly relate to either and intensity of precipitation events. Plant productivity is precipitation or streamflow dynamics (obviously, there are largely controlled by the pulsing and unpredictable nature strong linkages and correlations between them). Thus, of soil moisture dynamics, which is itself a result of the the impact of temperature changes and the associated characteristic of the precipitation input and the transpiration fluctuations in precipitation and streamflow in space and of the vegetation (Rodríguez-Iturbe and Porporato 2004). time are responsible for a very large number of different types of changes that directly or indirectly affect the It is crucial to understand that accounting only for changes sustainability of natural ecosystems. This paper focuses only in mean responses to climatic variability is not sufficient on the most important ecohydrologic changes. for a realistic evaluation of the impact of climate change in ecosystems. It is necessary to account for changes Precipitation Dynamics and Ecosystem in the stochasticity of the hydrologic forcing and its Response possible alterations in terms of frequency and amount of precipitation (Porporato, Daly and Rodríguez-Iturbe 2004). Henson (2006) noted that "[A]lthough it's natural to think Such alterations are responsible for modifying soil moisture of temperature first when we think of global warming, the dynamics (that is, intensity, duration, and frequency) of impact of climate change on precipitation may be even more periods of water stress and impaired plant assimilation important in the long run for many places and many people." (Rodríguez-Iturbe and Porporato 2004). An increase in atmospheric carbon dioxide may alone contribute to The most important impacts arise because of the highly accelerate photosynthesis, but the accompanying effects spatially varying fluctuations in rainfall as well as the of stomata being stimulated to close, the increase in temporal changes that take place in the seasonality of plant respiration, the costs in water transpired, and the precipitation. This means that some places will become higher release of carbon by the bacteria and fungi in the wetter and others will become drier, but even those places soil, are also highly detrimental to the ecosystem. The 112 matter of water is especially important since a depletion reduction in the frequency of rainfall events (Knapp et al. in soil moisture induces a reduction of the plant's water 2002, Porporato et al. 2004). potential. This can, in turn, cause dehydratation, turgor loss, xylem cavitation, stomatal closure, and a reduction of The scenario described in Figure 1 may become reality photosynthesis (see, for example, Nilsen and Orcutt 1998). in many areas of the world as a consequence of climate Soil moisture deficits result from the full dynamics, where change. Studies being carried out as part of the IPCC's the infiltration of water depends on the soil and precipitation 2007 assessment confirm that many parts of the world characteristics as well as on the transpiration from the show an increase in the fraction of rainfall and snowfall that plant. Soil moisture is thus cause and consequence of falls in the wettest 5 percent of all days with precipitation. A plant transpiration. "Even maintaining the same total rainfall, helpful manner to quantify such changes for ecohydrological an increase in the intensity of rainfall events, concomitant purposes is to estimate the rate of occurrences of days with a reduction in their frequency, will affect soil moisture with precipitation during different seasons, as well as the dynamics and plant conditions in a manner that depends on mean depth of precipitation per day during wet days. This is the soil and plant physiological characteristics at the site" especially useful for the period of the growing season. (Porporato, Daly and Rodríguez-Iturbe 2004). Figure 2 shows some results from Franz et al. (2008) for a Figure 1 shows a comparison of the experimental rainfall station with the longest period of daily rainfall data in results of Knapp et al. (2002) with the theoretical results the Upper Ewaso Ngiro River basin in Kenya. The seasons obtained by Porporato et al. (2004) for the mean daily analyzed correspond to those of the "long rains," which goes carbon assimilation rate as a function of the frequency of from the beginning of March to the end of May, and to the rainfall events for a constant total amount of precipitation "short rains" that extend from October to December. In both during a growing season. The analysis corresponds to cases, there is a statistically significant trend in the increase the response of a messic grassland to ambient rainfall of the mean depth of rainfall during wet days, as well as a pattern versus an artificially increased variability. There is decreasing trend in the rate of occurrence of rainy days. a 20 percent decrease in net assimilation for the altered rainfall conditions when total rainfall was the same but Caylor (2003) shows the mean value and coefficient of concentrated in fewer events. The analysis also shows variation for the mean annual rainfall along the Kalahari that in such a grassland ecosystem the impact on transect in Africa, jointly with the structure of the tree carbon assimilation of a decrease in total rainfall is more vegetation found along the transect (see Figure 3). It is clear pronounced when the decrease is accompanied by a that the very strong gradients in rainfall are accompanied Figure 1. Rainfall Dynamics Has Dramatic Impact on Net Assimilation p(x) 3 Altered Ambient 24 2 1 (µmolm­2s­1) x 20 0.15 0.30 0.45 0.50 p(x) 3 16 Altered Ambient 2 1 12 x 0.15 0.30 0.45 0.50 0.05 0.10 0.15 0.20 (d­1) Source: Porporato et al. (2004) 113 Figure 2. Jacobson Farm; 68 Years of Rainfall Data Early season Summer rains 0.8 0.4 Lambda (day­1) Lambda (day­1) 0.6 0.3 0.4 0.2 0.2 0.1 0 0 1920 1940 1960 1980 2000 2020 1920 1940 1960 1980 2000 2020 Year on record Year on record Long rains Short rains 0.8 0.4 Lambda (day­1) Lambda (day­1) 0.6 0.3 0.4 0.2 0.2 0.1 0 0 1920 1940 1960 1980 2000 2020 1920 1940 1960 1980 2000 2020 Year on record Year on record Early season Summer rains 0.8 0.4 0.6 0.3 Alpha (mm) Alpha (mm) 0.4 0.2 0.2 0.1 0 0 1920 1940 1960 1980 2000 2020 1920 1940 1960 1980 2000 2020 Year on record Year on record Long rains Short rains 0.8 0.4 0.6 0.3 Alpha (mm) Alpha (mm) 0.4 0.2 0.2 0.1 0 0 1920 1940 1960 1980 2000 2020 1920 1940 1960 1980 2000 2020 Year on record Year on record Source: Franz et al (2008) by interannual fluctuations around the average values, the daily rainfall events, are key controls for the type of which are much stronger for the drier areas of the regional vegetation of the different subregions, as well as for their landscape. These interannual fluctuations, accompanied spatial structure. As mentioned before and described with the changes described above in the dynamics of in Figure 1, changes in the dynamics regarding less 114 Figure 3. Annual Rainfall and Vegetation Characteristics along the Kalahari Transect Mean annual rainfall (mm) 1200 0.6 900 CV annual rainfall 0.4 600 0.2 300 0 16 18 20 22 24 26 28 Local species richness (LSR) Source: Caylor (2003) frequent and stronger rainfall events will have even larger reduction of the recycling sources of moisture from inland consequences for the assimilation process and survival regions to the atmosphere. These effects have been amply of vegetation when accompanied by a decrease in overall documented. An important example is the Yangtze River precipitation. basin, where the flood control services of trees have been evaluated to be worth much more than their value as lumber. An increase in the intensity of the rainfall events also leads Another example is recycling of the evapotranspiration to other types of ecohydrological consequences, particularly of the Amazonian forest, which constitutes an important in relation to soil erosion. For example, a report from the fraction of the total rainfall over the inland part of this Food and Agricultural Organization (FAO 2002) as quoted enormous river basin. in Brown (2006) found that: "Agriculture in Lesotho faces a catastrophic future, crop production is declining and Fire is another specific aspect of great ecohydrological could cease altogether over large tracts of the country if significance in relation to the impacts of climate change. steps are not taken to reverse soil erosion, degradation, Its frequency and intensity will also be greatly affected by and the decline in soil fertility." Brown (2006) goes on to climate change dynamics. note that "[W]hether the land is in northern Syria, Lesotho, or elsewhere, the health of the people living on it cannot be Hydrological Controls of Biodiversity and separated from the health of the land itself. A large share Impact of Climate Change of the world's 852 million hungry people lives on land with soils worn thin by erosion". Biodiversity is crucially affected by hydrologic conditions both in savannas and in river basin ecosystems. Soil erosion leads to loss of vegetation and desertification. Farming activities are either dramatically reduced or Muneepeerakul et al. (2008) have recently developed frequently supplanted by pastoral subsistence, which a very simple neutral model that is able to predict the provides feedback for further vegetation loss. Thus, main biodiversity features of both vegetation and fish sustainable development becomes out of reach. communities in river basins. Figure 4 shows part of their results for the case of fish biodiversity in the Mississippi- Soil is the medium in which plants grow and, in turn, Missouri river system (MMRS). plants protect the soil from erosion. Deforestation and soil erosion commonly go together; their impacts are multiple The local species richness (LSR) as well as the frequency ranging from the occurrence of large disastrous floods to distribution of LSR are extremely well reproduced by a 115 Figure 4. Data and Model Results of Fish Biodiversity in the Mississippi-Missouri River System Diversity profile Diversity histogram 100 100 Local Species Richness 80 80 Frequency 60 60 40 40 20 20 0 0 30 25 20 15 10 5 0 0 50 100 150 Upstream Outlet Topological distance to the outlet Local species richness (LSR) Species range Diversity 10 3 0.8 Jaccard's similarity index 0.7 0.6 102 0.5 Range 0.4 0.3 10 1 0.2 0.1 100 0 0 100 200 300 400 500 0 10 20 30 40 Rank Topological distance between DTA pairs Source: Muneepeerakul et al. (2008) model with only four parameters. Moreover, the rank- vegetation the habitat capacity is controlled by the amount occupancy curve (where the occupancy is given by of green water on the DTA. the number of direct tributary areas (DTAs) where the species is present) is also very well reproduced by the The impacts of climate change and man-made alterations model. Muneepeerakul et al. (2008) measure the between on the habitat capacity and/or network connectivity can community diversity (or how diversity changes spatially) be directly studied in the model of Muneepeerakul et al. through the Jaccard's similarity index (JSI), which the model (2008). Also, this type of approach allows the identification (again) reproduces very well with respect to that found in of crucially important subregions where changes of the the data. previous type will bring the most impacting changes in the biodiversity of the system. This identification will then allow Similar results are being presently obtained with an for the optimal organization and planning of conservation extensive analysis of species of vegetation existing in the campaigns. MMRS. These results are of great relevance for the study of the possible impacts of climate change on the biodiversity characteristics of ecosystems. Thus, the controlling Ecohydrological Footprints variables in the results of Munepeerakul et al. (2008) for the case of freshwater fish diversity are: (1) the habitat capacity D'Odorico et al. (2008) have recently proposed the of the different DTAs (which are a direct function of the concept of ecohydrological footprints to quantitatively freshwater runoff arising from the DTA), and (2) the network account for the human impacts on ecosystems resulting connectivity in the river system. Similarly, in the case of from anthropogenic disturbances of hydrologic processes. 116 The ecohydrological footprint will measure the change in a Patterns in the Upper Ewaso Ngiro River Basin in specific ecosystem function or service caused as result of Kenya, in review, November 2008. human intervention on hydrologic drivers. Hedin,L., Chadwick,O., Schimel,J., and M.Torn, Linking Ecology, Biology, and Geoscience, Report to the An example of ecohydrological footprints are the changes National Science Foundation, 2002. in carbon sequestration resulting from the changes in Henson, R., The Rough Guide to Climate Change, Penguin soil/water balance, which in turn result from land use or Group,341p.,2006. drainage projects. Another example is the changes in fish or IPCC (2001). Climate Change 2001: The Scientific vegetation biodiversity arising from the reduction of habitat Basis. Intergovernmental Panel on Climate Change. capacity ensuing from the decrease of direct contributing Cambridge University Press. runoff in different regions of a river basin (D'Odorico et Knapp,A.K.,P.A.Fay,J.M.Blair,S.L.Collins,M.D.Smith,J. al. 2008). These changes can be quantitatively measured D.,Carlisle,C.W.Harper,B.T.Danner,M.S.Lett,and via models like the one developed by Muneepeerakul et al J.K.McCarron. Rainfall Variability,Carbon Cycling,and (2008). Plant Species Diversity in a Messic Grassland,Scien ce,298,2202­2205, 2002. Muneepeerakul,R., E.Bertuzzo,H Lynch,W.Fagan, References A.Rinaldo,and I.Rodriguez-Iturbe, Neutral Metacommunity Models Predict Fish Diversity Patterns Brown,L.,R., Plan B 2.0,W.W. Norton, 365p.,2006 in Mississippi-Missouri Basin,Nature,453,220­224, Caylor,K.,Structure and Function of Kalahari Transect 2008 Vegetation,Ph.D. Thesis,University of Virginia,2003 Porporato, A.,Daly E., and I.Rodriguez-Iturbe, Soil water Diamond, J., Collapse: How Societies Choose to Fail or balance and ecosystem response to climate change, Succeed, Viking Press,575 p.,2005. American Naturalist, 164 (5),625­632, 2004. D'Odorico, P., Laio.,F., Ridolfi,L., Porporato,A., Rinaldo,A., Nilsen,E.T., and D.M.Orcutt, Physiology of Plants Under and I.Rodriguez-Iturbe, Ecohydrological Footprints, in Stress; Abiotic Factors, New York,John Wiley,1998. review, November 2008. Rodriguez-Iturbe, I., Ecohydrology: a Hydrologic Perspective United Nations Food and Agriculture Organization of Climate-Soil-Vegetation Dynamics, Water Resources (FAO),"FAO/WFP Crop and Food Assessment Mission Research,36,3­9, 2000. to Lesotho Special Report". FAO, 2002. www.fao.org Rodriguez-Iturbe,I., and A. Porporato, Ecohydrology of Franz,T., Caylor,K., Nordbotten, J., Celia, M., and I.Rodriguez- Water Controlled Ecosystems, Cambridge University Iturbe, Soil Moisture, Water Stress,and Vegetation Press,442p., 2004. 117 118 Comments of World Bank Discussants The comments made by invited World Bank discussants extracted and how much intake is accounted for by different are summarized in this section. These comments reflect the uses. However, water use is seldom monitored or measured. personal experience of the various discussants about the November 2008 workshop topics and presentations, and Within that context and in terms of water use, building and are included as part of the selected hydrology topics review. operating infrastructure is also a very important part of water resources management that faces many problems. For example, good plans are usually lacking, operation is not Integrated Water Resources optimized, and multi-purpose operations for flood control, Management hydropower, water supply, and irrigation are also lacking. The problems facing infrastructure operation are in need of Integrated water resources management (IWRM) is like a improvement and must be addressed. nice warm place where everybody would like to go, much like sustainable development and its different aspects. It The third challenge to IWRM has to do with the question of is something that no one can disagree with, since no one pollution discharge, something that is not handled very well. dare say that "disintegrated" water resources management Little is usually known about the location of discharge points is desirable. However, some challenges can be identified or how much pollutants they are discharging. Integrated in the main messages that came across from all the water resource management requires that all the discharges presentations. be inventoried, characterized, and controlled. First, IWRM needs to stop being a buzz word (still the case Finally, there is a need to balance new challenges with in many places). Integrated water resources management old ones using modern approaches, (especially those that needs to be placed into operation and implemented. have been shown to work in other countries), and those However, this in itself is challenging because IWRM approaches should be institutionalized. It may be in terms implementation faces hurdles of information, institutions, and of managing existing climate variability or climate change capacity. In addition, the effort requires integration across challenges or just managing development challenges. different spatial scales (all the way form a particular use, to a This would include basic management challenges of river basin or to the national level), across uses and sectors different types. This could be accomplished with more (especially in cross-sector themes like the environment), cross-learning-type of efforts, so that cross-fertilization and and across different institutions (which is probably the most learning across different regions becomes a reality. challenging). Second, there is a need for measurement. It is time to Climate Variability and Change move from estimation to the use of the information that is available. Moreover, without a system of water rights The key words in the presentation about hydrology for in place and without appropriate measurements it is very water systems development and management under climate difficult to manage water resources in terms of water use. change were risk and uncertainty, linked to probability and This challenge needs addressing. hydro-economics. Uncertainty analysis is not used enough to help evaluate policy and project options and to understand There is another part of water resources management that vulnerability to extremes vis-à-vis climate change. has not been mentioned; namely, water scarcity. Water scarcity refers to the balance between the supply of and But the one issue that was highlighted in the demand for water, as well as water availability and use. presentations is the idea of non-stationarity. Especially Water use refers to how much water is actually being useful is the typology of intrinsic non-stationarity, but also 119 the non-stationarity issues having to do with measurement are based on legal or regulatory requirements. In many and purpose. developing countries, this is a similar situation where a very strong legal or policy framework exists. However, the data Regarding measurement non-stationary, it is clear that in are lacking, and so are the capacity, and the institutions to many of the places where the Bank works, data are often an drive their implementation forward. A very important point issue and how they are collected remains a challenge. While was made; namely, that hydro information cuts across decades-worth of streamflow data have been collected, sciences, economics, policy, and regulation. discontinuous jumps because of measurement unit errors (rather than because of change in hydrological processes) A modeling system is required that meets the needs of user are still common. There are data inconsistencies as well as of different scales, from a basin down to a sub-watershed or problems with how it is collected; in many places, the quality micro-watershed. However, it should be borne in mind that of the data is extremely low. But, the historical record (or no single model can do it all. whatever synthetic flow series are generated stochastically from the historical record) is all that is available. Sedimentation is another issue that is worth exploring further. Sedimentation is good and bad; the question is: Non-stationarity of purpose is more frequent in environments good and bad for whom? Clearly, sedimentation is bad for where political leadership changes very frequently. a reservoir operator, but for a farmer in a very productive Institutional interests change, and in that environment it is floodplain, sedimentation is good. So there is a need to difficult to narrow on the data to focus on future scenarios balance the interests of these different stakeholders and or converge on the futures of interest, when these may uses of sedimentation. change frequently. The Bank's low involvement in coastal environments is an In this sense it may be worthwhile to seek opportunities issue. Attention is mainly focused on environmental flows for longer term institution building. Opportunities to build and making sure there is enough water going down to relationships with experts in the region and to work with coastal wetlands and mangrove forests, but the issue of bilateral and other regional organizations should be sought the water's quality and chemical constituents needs to be in order to gain a broader perspective from people and explored further. institutions with a longer term stake in a particular country or region. Finally, the biggest innovation on the issue of evapotranspiration may be the existence of all the Last but not least, it pays to be humble. As knowledge technologies from satellites and other remote sensing increases, it must be recognized that Bank clients have equipment for Bank clients to use. There is a real different perspectives about the Bank's engagement, opportunity to leapfrog knowledge instead of having to whether as partners in sector work or potential lenders or spend huge amounts on developing a database from even competitors. scratch. Maybe some judicious mix of ground truth data plus satellite images would be the better approach in terms not only of continuity, but also of accuracy and of the types of Hydrologic Interaction decisions that many of the clients face. Hydrologic interactions are complex. The Bank has shown a general interest in learning more about those interactions Associated Changes in Climate and as well as the various models that can assist in watershed Land Use management and planning for practitioners, for Bank clients, and for other stakeholders. The very good presentation on ecohydrology, climate, and ecosystem response made three very pertinent points. It was very interesting that one of the presentations began One was that the focus of the effects of climate change in by outlining some of the driving forces for modeling that water resources should be on precipitation and streamflow, 120 rather than temperature. Another pertinent point was the on a daily basis. There is a need to do more to find need to have a better understanding of how precipitation ways to operationalize climate change consequences could change over time, and the impacts this could have on water resources at appropriate spatial and temporal on biodiversity. Finally, the presentation noted that it is resolutions. important to look at shifts in precipitation that will result in shorter but perhaps more intense periods of rainfall. This is Selecting the right model has become a task in itself given very relevant to regions that have short, intense monsoon the large number of such models in the field. No single seasons. In watershed management it is important to model can solve all water resources problems. Model capture the surface water and deal with groundwater to selection should be dictated by the problems at hand, support people over the dry season. availability of data and capacity of model users. The type of problems faced should lead to selection of the most Talking about climate change that is going to affect appropriate models. Similarly, availability of data to calibrate the livelihood of people 20 or 30 years from today on the model and calibrate model parameters, and also the an average basis seems like a big assumption in areas capacity of end users at the end of model development that are not even able to use what they are receiving should help zero in on most appropriate models. 121 HEF Hydrology Expert Facility Hydrology and Water Resources Advisory Service What we do · Expertpanel: High level 6 member panel for advisory role and application of cutting edge tech- HEF assists in addressing complex hydrology and water nology and approaches resource management problems by providing short-term · Disseminationandlearning: Technical events expertise on-demand in collaboration with thematic groups · Publications: Publications include Mission Briefs and HEF Notes How we do it Support is focused on specific issues, situations or prob- lems in connection with the different stages of the Bank Sampleareasofsupport project cycle · Integrated water resources development and management, including hydropower · Watershed management modeling HEF services · Hydro meteorological risk management · Water quality, wastewater disposal alternatives and design of underwater outfalls · Operationssupport: Expert advice for short assignments from a roster of more than 150 hydrol- ogists/water resources experts How to request HEF support CONTACTUS Task Team Leaders/members from the regions submit a short form describing the general characteristics of the assignment Gabrielle Puz and its contribution to Bank business development gpuz@worldbank.org + 1 (202) 473-9973 HEF in action: Some examples WaterQuality Luis Ernesto García Study of wastewater disposal alternatives in Rio de la Plata, lgarcia@worldbank.org Argentina; review and design of underwater outfalls in Baku + 1 (202) 458-1897 Sea, Azerbaijan and Lake Titicaca, Bolivia. RiskManagement Susanne Scheierling Hydrologic analysis for regional and urban flood management sscheierling@worldbank.org in Ghana and Jakarta, Indonesia; strengthening of real-time + 1 (202) 473-7276 hydrology forecast capabilities in Albania and Moldova. Hydropower Hydropower downstream impact analysis in India; water-ener- gy linkage analysis in Central Asia. THEWORLDBANK http://intranet.worldbank.org/WBSITE/INTRANET/SECTORS/INTWAT/0,,contentMDK:21907421~menuPK:5317783~pa gePK:210082~piPK:254376~theSitePK:4602115,00.html 122 123 1818 H Street, NW Washington, DC 202.123.4567 www.worldbank.org/water/wpp