Climate Change 2001:
Working Group II: Impacts, Adaptation and Vulnerability
Other reports in this collection Impacts and Adaptation Options

The impacts of hydrological changes and options for effective adaptation will depend on the nature of the hydrological changes, the amount of buffering provided by natural and artificial storage capacity, the nature of demands on the resource, and the effectiveness with which institutions in place balance competing demands. Continued increases in demand are likely for the multiple services provided by North American water resources, encompassing out-of-stream and instream uses. Adaptations to the effects of climate change therefore will occur in conjunction with adjustments to changes in the level and characteristics of water demands.

A major comparative study of six U.S. water resource systems found that although climate change could have significant and often adverse impacts on system performance, adjustments to reservoir operations could be made, allowing those impacts to be smaller than the underlying hydrological changes (Lettenmaier et al., 1999). The authors also found that "The effects of anticipated demand growth and other plausible future operational considerations…would about equal or exceed the effects of climate change over system planning horizons" (Lettenmaier, et al., 1999).

The importance of future changes in demand is echoed by Frederick and Schwarz (1999). They found that even in the absence of climate change, the annual cost of supplying water to meet projected increases in U.S. water demands would be about US$13.8 billion higher in 2030 than in 1995, based on the current trend of increasingly using conservation to balance growing demands with supplies. This trend has been driven by "…the high costs of developing new supplies, environmental concerns, a growing appreciation for the values of instream flows and efforts to improve water quality" (Frederick and Schwarz, 1999). Using projected runoff changes calculated by Wolock and McCabe (2000) for 18 water resource regions, the study found that runoff increases projected by the HadCM2 model tended to reduce the cost of balancing future water supplies and demands, whereas the reduced flows projected by the CCC model resulted in large cost increases. With the CCC model projections, the study found that under an "efficient management" scenario, annual water costs could increase "…by nearly US$105 billion, about US$308 per person. Much higher cost increases would result from policies to maintain relatively high streamflow levels under such dry conditions" (Frederick and Schwarz, 1999).

This finding—that institutions can have significant impacts on the costs arising from reduced water availability—is supported by a study that assessed the possible consequences of a severe sustained drought on the Colorado River system (Lord et al., 1995). The study authors found that the "Law of the River," as currently interpreted and implemented, would leave sensitive biological resources, hydropower generation, recreational values, and Upper Colorado basin water users vulnerable to damages despite extraordinary engineering attempts to drought-proof the river. The study also found that certain proposed institutional and system operating changes could considerably alter the level and incidence of the damages (Booker, 1995; Kenney, 1995; Lord et al., 1995; Sangoyomi and Harding, 1995). For example, reallocation from low- to high-valued uses (e.g., through intrastate or interstate water marketing) and changes in reservoir storage policies to hold water in the Upper Colorado basin to reduce evaporative losses could reduce consumptive use damages by more than 90% (Booker, 1995).

Such "what if" analyses can foster creative thinking about adaptation options. Future adaptation strategies also are likely to build on current creative solutions to meeting water supply challenges. For example, when increases in agricultural and sewerage runoff in the Catskill/Delaware watershed region threatened the quality of New York City's water supplies, the city was faced with a choice of either building an artificial filtration plant or taking action to protect and restore the natural purification capacity of the watershed's ecosystem. The capital cost of the filtration plant would have been US$6-8 billion, plus annual operating costs of US$300 million. The city found that for a fraction of that cost it could enter into agreements with landowners in the watershed to adopt land-use practices that would adequately protect water quality, making the filtration plant unnecessary (PCAST, 1998; Platt et al., 2000).

Adjustments to the effects of climate change as well as to evolving demands for out-of-stream and instream water uses may entail impacts on the distribution of water-use benefits. The question of who will bear the costs of any reduction in water availability depends on the nature and ownership of existing water rights, how those rights are measured, and how they may be modified by other policies. In the United States, federal legislation—including the Endangered Species Act and the Clean Water Act—has regulated or constrained development of many new water projects and could be applied to modify existing water diversions (WWPRAC, 1998). Native American (aboriginal) communities often possess reserved water rights that, despite their high priority, have never been developed or clearly quantified. Substantial litigation and negotiation efforts have focused on clarifying these rights. Any significant change in water availability may heighten tensions between these communities and neighboring water users (Echohawk and Chambers, 1991). Under riparian water law and permit systems, governmental authorities may have substantial discretion in regulating water uses under drought conditions (Abrams, 1990; Flood, 1990; Sherk, 1990; Dellapenna, 1991; Scott and Coustalin, 1995). Under the prior appropriation system of water law, which is followed in western Canada and most states in the western portion of the United States, the risk of a shortfall generally is inversely proportional to the seniority of the right. Many junior rightholders, however, have access to water stored in reservoirs, which improves their security of supply.

Several studies have argued that improving the functioning of water markets could help to create the kind of flexibility needed to respond to uncertain changes in future water availability (e.g., Trelease, 1977; Tarlock, 1991; OTA, 1993; Miller et al., 1997). If water supplies decline in particular locations or seasons, water markets could soften the impacts by moving water from lower to higher valued uses. In the western United States, where irrigation now accounts for more than 80% of consumptive water use, water market activity is likely to continue the current trend of movement of water out of irrigated agriculture to accommodate other water uses (WWPRAC, 1998). However, water markets are not likely to adequately protect instream flows and sensitive biological resources unless public agencies are given budgets to buy or rent water to protect those values (Wilkinson, 1989; NRC, 1992). In addition, water rights have to be clearly defined for water markets to function properly. Even in the western United States, where water markets have developed rapidly in recent years (Saliba and Bush, 1987; NRC, 1992; Colby, 1996; Yoskowitz, 1999), market development has been hampered by the fact that water rights often are not well documented (e.g., Costello and Kole, 1985; Gould, 1988; Colby, 1998). To avoid adverse impacts on other water users, water authorities typically review each transfer proposal. The entire process often entails substantial transaction costs, and these costs differ significantly across jurisdictions (Saliba and Bush, 1987; MacDonnell, 1990; NRC, 1992). This suggests that administrative practices and other institutional factors affect the efficiency of the water transfer process and that administrative reforms to reduce transaction costs could improve adaptability to the effects of climate change.

Finally, permanent water transfers are not particularly effective in promoting flexible adaptation to climatic variability or uncertain climate changes. Small, temporary transfers through water banks such as California's Emergency Drought Water Banks would provide greater flexibility (see Section

Although well-functioning water markets may ameliorate the socioeconomic impacts of reduced water availability, they cannot completely eliminate the adverse impacts of a drying scenario. Using models that assume optimal allocation of water across all sectors (which implicitly assumes perfect, cost-free water markets), Hurd et al. (1998) estimated the water resources-related costs of climate change scenarios for four major river basins in the United States. They found that losses from moderate and severe drying scenarios tended to be greater in the semi-arid western United States than in the eastern states. From this analysis, they project that national water-related welfare losses arising from a temperature increase of 2.5°C coupled with a 7% increase in precipitation would be on the order of US$9.5 billion. A scenario with a temperature increase of 5°C coupled with no change in precipitation would result in losses of US$43 billion. Overall, they found that the economic costs probably would be dominated by impacts on water quality and instream nonconsumptive water uses, especially hydroelectricity generation. This result hinges on their assumption that irrigators—who currently dominate ownership of water in the western United States—would benefit by selling some of their water to other sectors. More realistic assumptions regarding the costs of transferring water through water markets would have produced larger estimated losses and a different distribution of losses across sectors.

Water rights are not always defined in terms that would permit the transfer of a specific amount of water from one use to another, particularly in eastern U.S. and Canadian jurisdictions that still follow a traditional riparian system of water law (Tarlock, 1989; Flood, 1990; Scott and Coustalin, 1995). In such areas, accommodation of competing demands and changing supplies will require local and regional planning and coordination. For example, if the levels and outflows of the Great Lakes-St. Lawrence system decline, as most available scenarios suggest, adaptation options are likely to raise sensitive interjurisdictional allocation issues and may require infrastructure investments that will increase the need for cooperative planning and management of the basin's water resources (Bruce et al., 2000).

Other reports in this collection