Translating Ecosystem Functions to the Value of Ecosystem Services: Case Studies
Valuing ecosystem services requires the integration of ecology and economics. Ecology is needed to comprehend ecosystem structure and functions and how these functions change with different conditions. Both ecology and economics are required to translate ecosystem functions into the production of ecosystem goods and services. Economics is needed to comprehend how ecosystem goods and services translate into value (i.e., benefits for people; see also Figure 1-3). The two preceding chapters discuss much of the relevant ecological and economic literature. Chapter 3 focuses on the relevant ecological literature on aquatic and related terrestrial ecosystem functions and services, while Chapter 4 focuses on the economic literature on nonmarket valuation methods useful for valuing ecosystem goods and services. In this chapter, the focus is on the integration of ecology and economics necessary for valuing ecosystem services for aquatic and related terrestrial ecosystems. More specifically, a series of case studies is reviewed (including those taken from the eastern and western United States; see Chapter 1 and Box ES-1 for further information), ranging from studies of the value of single ecosystem services, to multiple ecosystem services, to ambitious studies that attempt to value all services provided by ecosystems. An extensive discussion of implications and lessons learned from these case studies is provided and precedes the chapter summary.
Development of the concept of ecosystem services is relatively recent. Only in the last decade have ecologists and economists begun to define ecosystem services and attempted to measure the value of these services (see for example, Balvanera et al., 2001; Chichilnisky and Heal, 1998; Constanza et al., 1997; Daily, 1997; Daily et al., 2000; Heal, 2000a,b; Pritchard et al., 2000; Wilson and Carpenter, 1999). There is a much longer history of natural resource managers and economists evaluating “goods” produced by ecosystems (e.g., forest products, fish production, agricultural production). For example, in 1926, Percy Viosca, Jr., a fisheries biologist, estimated that the value of conserving wetlands in Louisiana for fishing, trapping, and collecting activities was $20 million annually (Vileisis, 1997). In the 1960s and early 1970s, pioneering work by Krutilla (1967), Hammack and Brown (1974), and Krutilla and Fisher (1975), among others, greatly expanded the set of “goods and services” generated by natural systems considered by economists to be of value to humans (e.g.,
clean air, clean water, recreation, ecotourism). Economic geographers and regional scientists (e.g., Isard et al., 1969) examined spatial relationships among natural and socioeconomic systems. Recent work on ecosystem services has broadened the set of goods and services studied to include water purification, nutrient retention, and flood control, among other things. It has also emphasized the importance of understanding natural processes within ecosystems (e.g., primary and secondary productivity, carbon and nutrient cycling, energy flow) in order to understand the production of ecosystem services. Yet, as discussed throughout this report, for the most part, the importance of these natural processes in producing ecosystem services on which people depend has remained largely invisible to decision-makers and the general public. For most ecosystem services, there are no markets and no readily observable prices, and most people are unaware of their economic value. All too often it is the case that the value of ecosystem services becomes apparent only after such services are diminished or lost, which occurs once the natural processes supporting the production of these services have been sufficiently degraded. For example, the economic importance of protecting coastal marshes that serve as breeding grounds for fish may become apparent only after commercial fish harvests decline. By then, it may be difficult or impossible to repair the damage and restore the production of such services.
Although there has been great progress in ecology in understanding ecosystem processes and functions, and in economics in developing and applying nonmarket valuation techniques for their subsequent valuation, at present there often remains a gap between the two. There has been mutual recognition among at least some ecologists and some economists that addressing issues such as conserving ecosystems and biodiversity requires the input of both disciplines to be successful (Daily et al., 2000; Holmes et al., 2004; Kinzig et al., 2000; Loomis et al., 2000; Turner et al., 2003). Yet there are few existing examples of studies that have successfully translated knowledge of ecosystems into a form in which economic valuation can be applied in a meaningful way (Polasky, 2002). Several factors contribute to this ongoing lack of integration. First, some ecologists and economists have held vastly different views on the current state of the world and the direction in which it is headed (see, for example, Tierney, 1990, who chronicles the debates between a noted ecologist and economist [Paul Ehrlich and Julian Simon]). Second, ecology and economics are separate disciplines, one in natural science and the other in social science. Traditionally, the academic organization and reward structure for scientists make collaboration across disciplinary boundaries difficult even when the desire to do so exists. Third, as noted previously, the concept of ecosystem services and attempts to value them are still relatively new. Building the necessary working relationships and integrating methods across disciplines will take time.
Some useful integrated studies of the value of aquatic and related terrestrial ecosystem goods and services are starting to emerge. The following section reviews several such studies and the types of evaluation methods used. This review begins with situations in which the focus is on valuing a single ecosys-
tem service. Typically in these cases, the service is well defined, there is reasonably good ecological understanding of how the service is produced, and there is reasonably good economic understanding of how to value the service. Even when valuing a single ecosystem service however, there can be significant uncertainty about either the production of the ecosystem service, the value of the ecosystem service, or both. Next reviewed are attempts to value multiple ecosystem services. Because ecosystems produce a range of services that are frequently closely connected, it is often difficult to discuss the valuation of a single service in isolation. However, valuing multiple ecosystem services typically multiplies the difficulty of valuing a single ecosystem service. Last to be reviewed are analyses that attempt to encompass all services produced by an ecosystem. Such cases can arise with natural resource damage assessment, where a dollar value estimate of total damages is required, or with ecosystem restoration efforts. Such efforts will typically face large gaps in understanding and information in both ecology and economics.
Proceeding from single services to entire ecosystems illustrates the range of circumstances and methods for valuing ecosystem goods and services. In some cases, it may be possible to generate relatively precise estimates of value. In other cases, all that may be possible is a rough categorization (e.g., “a lot” versus “a little”). Whether there is sufficient information for the valuation of ecosystem services to be of use in environmental decision-making depends on the circumstances and the policy question or decision at hand (see Chapters 2 and 6 for further information). In a few instances, a rough estimate may be sufficient to decide that one option is preferable to another. Tougher decisions will typically require more refined understanding of the issues at stake. This progression from situations with relatively complete to relatively incomplete information also demonstrates what gaps in knowledge may exist and the consequences of those gaps. Part of the value of going through an ecosystem services evaluation is to identify the gaps in existing information to show what types of research are needed.
MAPPING ECOSYSTEM FUNCTIONS TO THE VALUE OF ECOSYSTEM SERVICES: CASE STUDIES
Despite recent efforts of ecologists and economists to resolve many types of challenges to successfully estimating the value of ecosystem services, the number of well-studied and quantified cases studies remains relatively low. The following section reviews cases studies that have attempted to value ecosystem services in the context of aquatic ecosystems. These examples illustrate different levels of information and insights that have been gained thus far from the combined approaches of ecology and economics.
Valuing a Single Ecosystem Service
This review begins with studies of the value of ecosystem services using examples that attempt to value a single ecosystem service. These cases provide the best examples of both well-defined and quantifiable ecosystem services and of services that are amenable to application of economic valuation methodologies. The best-known example of a policy decision hinging on the value of a single ecosystem service involves the provision of clean drinking water for New York City, which is reviewed first. Other examples include cases where ecosystems provide habitat for harvested fish or game species and cases where they provide flood control.
In all of the cases reviewed in this section, the ecosystem service is well-defined, although there may be some scientific uncertainty surrounding quantification of the amount of the service provided. In some cases, adequate methods for valuing the single ecosystem service exist. Further, for some cases, such as the New York City example below, information about a single ecosystem service may prove sufficient to support rational environmental decision-making. In other cases, this will not be so, and further work to assess a more complete set of ecosystem services will be necessary. Under no circumstances, however, should the value of a single ecosystem service be confused with the value of the entire ecosystem, which has far more than a single dimension. Unless it is kept clearly in mind that valuing a single ecosystem service represents only a partial valuation of the natural processes in an ecosystem, such single service valuation exercises may provide a false signal of the total economic value of the natural processes in an ecosystem.
Providing Clean Drinking Water: The Catskill Mountains and New York City’s Watershed
One of the best-studied water supply systems in the world is the one that provides drinking water for more than 9 million people in the New York City metropolitan area (Ashendorff et al., 1997; NRC, 2000a; Schneiderman, 2000). New York City’s water supply includes three large reservoir systems (Croton, Catskill, and Delaware) that contain 19 reservoirs and 3 controlled lakes. This system, including all tributaries, encompasses a total area of 5,000 km2 with a reservoir capacity of 2.2 × 109 m3. This complex array of natural watersheds requires a wide range of management to sustain the water quality supplied to the reservoirs and aqueducts. Historically, these watersheds have supplied high-quality water with little contamination. However, increased housing developments with onsite septic systems, combined with nonpoint sources of pollution such as runoff from roads and agriculture, have posed threats to water quality. Further significant deterioration of water quality would force the U.S. Environmental Protection Agency (EPA) to require New York City build a water filtra-
tion system1 to ensure that drinking water delivered to consumers would meet federal drinking water standards. By 1996, New York City faced a choice: it could either build water filtration system or protect its watersheds to ensure high-quality drinking water.
The cost of building a new, larger filtration system necessary to meet water quality standards was estimated to lie in the range of $2 billion to $6 billion. Moreover, the city estimated that it would spend $300 million annually to operate the new filtration plant. Together, the costs of building and operating the filtration system were estimated to be in the range of $6 billion to $8 billion (Chichilnisky and Heal, 1998).
Instead of investing in a water filtration facility, New York City opted to invest more in protecting watersheds. Maintaining water quality in the face of increased human population densities in the watershed required increased protection of riparian buffer zones along rivers and around reservoirs. These zones help to regulate nonpoint sources of nutrients and pesticides from stormwater runoff, septic tanks, and agricultural sources. In 1997 the city received “filtration avoidance status” from the EPA by promising to upgrade watershed protection. The 1997 Watershed Memorandum Agreement resulted from negotiations among the State of New York, New York City, the EPA, municipalities within the watershed, and five regional environmental groups. The agreement provided a framework for compliance with water quality standards and contained plans for land acquisition through mutual consent, watershed regulations, environmental education workshops, and partnership programs with community groups. For example, a farmer-led Watershed Agricultural Council provides programs for the approximately 350 dairy and livestock farms in the watershed to minimize nutrient input from agricultural runoff (Ashendorff et al., 1997).
Under this agreement, New York City is obligated to spend $250 million during a 10-year period to purchase lands within the watershed (up to 141,645 hectares). In this part of the overall response, the New York City Department of Environmental Protection land acquisition program purchases undeveloped land from willing sellers rather than relying on condemnation and the power of eminent domain. Property rights to develop land in the watershed rests in the hands of local landowners. In some cases these rights are regulated by local ordinances. New York City’s 1953 Watershed Rules and Regulations give the city some authority over watershed development to limit water pollution. Decades-old resentment remains among some residents of upstate watersheds because earlier land acquisitions to build the reservoirs displaced entire communities. Moreover, recent concerns about security of the reservoirs have also polarized residents whose road access has been limited. Exactly what legal rights New York City has and what legal rights local municipalities, and local landowners
have to make decisions is not fully resolved. The long-term costs of riverbank protection, upkeep of sewage treatment plants by municipalities and overall maintenance costs of this approach remain uncertain.
On the other hand, a series of regulations prohibiting certain types of development in certain places (e.g., areas in close proximity to watercourses, reservoirs, reservoir stems, controlled lakes, wetlands) was agreed upon. The city together with the Catskill Watershed Corporation developed a comprehensive geographical information system to track land uses and to analyze runoff and storm flows resulting from precipitation. Runoff is sensitive to connections among stream network, and to the amount of impervious surface in the watershed (e.g., roads, buildings, driveways, parking lots), which results in increased peak flows that can cause flooding and bank erosion (Arnold and Gibbons, 1996; Gergel et al., 2002). To minimize these effects, new construction of impervious surfaces within 300 feet of a reservoir, rivers, or wetland is prohibited. Road construction within 100 feet of a perennial stream and 50 feet of an intermittent stream is also prohibited. Septic system fields cannot be located within 100 ft of a wetland or watercourse or 300 feet of a reservoir because these onsite sewage treatment and disposal systems do not work effectively in saturated soil. Septic fields also interfere with the natural nutrient processing in floodplains, wetlands, and riparian buffer zones along streams. Funds are available to subsidize upgrades of local wastewater treatment plants and septic systems throughout the watershed. There are 38 wastewater treatment plants in the watershed that are not owned by New York City. Overall, New York City projected that it would invest $1 billion to $1.5 billion in protecting and restoring natural ecosystem processes in the watershed (Ashendorff et al., 1998; Chichilnisky and Heal, 1998; Foran et al., 2000; NRC, 2000a). Incentives for landowners to improve riparian protection through conservation easements and educational outreach efforts were combined with management of state-owned lands to minimize erosion and protect riparian buffers.
In this case, it was not necessary to value all or part of the services of the Catskill watershed; it was merely necessary to establish that protecting and restoring the ecological integrity of the watershed to provide clean drinking water was less costly than replacing this ecosystem service with a new water filtration plant. As discussed in Chapter 4, Shabman and Batie (1978) suggest that a replacement cost approach can provide a “proxy” valuation estimation for an ecological service if the alternative considered provides the same service, the alternative compared is the least-cost alternative, and there is substantial evidence that the service would be demanded by society if it were provided by that least-cost alternative. In the Catskill case the proposed filtration plant would provide very similar services (more on this below). Of course, the city will have to provide clean water somehow. So these conditions are met and the cost of replacing the provision of clean drinking provided by the watershed with a filtration plant, less the cost of protecting and restoring the watershed, can be thought of as a measure of the ecosystem service value to New York City as a water purification tool. If, however, demand side management can reduce demand for water
at less cost than it costs to provide the water via the filtration plant, then demand side management costs would provide the relevant avoided costs. Both methods—natural processes in watersheds and a water filtration plant—are capable of providing clean water that meets drinking water standards.
This case also appears to provide clear environmental policy direction. For New York City, it is likely to be far less costly to provide safe drinking water by protecting watersheds, thereby maintaining natural processes, than to build and operate a filtration plant. Further, protecting watersheds to provide clean water also enhances provision of other ecosystem services (e.g., open space for recreation, habitat for aquatic and terrestrial species, aesthetics). As discussed throughout this report, such ecosystem services are arguably far harder to value economically. Since these values add to the value of protecting watersheds for the provision of clean water, which is the preferred option even without consideration of these additional values, it is not necessary to establish a value for these services for policy purposes. Thus, protecting watersheds can be justified on the basis of the provision of clean drinking water alone.
Despite the appearance of being a textbook case for valuing a single ecosystem service, several issues make the answer to ecosystem valuation less obvious than at first glance. The replacement cost approach assumes that the same service will be provided under either alternative. In reality, it is unlikely that watershed protection and filtration will provide identical levels of water quality and reliability over time because engineered systems can fail—especially during storms when heavy flows overwhelm the system. Likewise, natural watersheds can also vary in their effectiveness in response to severe storm flows or other disturbances (Ashendorff et al., 1997). Managed watersheds can require some maintenance costs to sustain ecosystem services such as clean up of accidental spills or fish kills to prevent pollution or control of invasive species such as zebra mussels (Covich et al., 2004; Giller et al., 2004). Both engineered and ecosystem approaches are vulnerable but they differ in the types of uncertainty associated with each investment.
New York City’s watershed investment plan includes several maintenance costs such as thorough, multistaged monitoring of water quality and disease surveillance that triggers active management and localized water treatment. Baseline data on water quality and biodiversity of stream organisms in the watershed (e.g., aquatic insects) are being collected by the Stroud Water Research Center (2001) annually to determine if the city’s recent management efforts are effective. By reducing the risk of contaminants from various sources, the city can minimize use of disinfectants at the final water treatment stages. Reducing chemical use saves money directly and may also have health benefits since chlorination can produce halogenated disinfection by-products (e.g., chloroform, trihalomethane) in drinking water, especially in ecosystems with high levels of organic matter (Symanski et al., 2004; Villanueva et al., 2001; Zhang and Minear, 2002). Some of these by-products may be carcinogens. On the other hand, filtration may provide higher-quality drinking water because chlorination
is not completely effective in killing pathogens, particularly when there are high levels of suspended materials (Schoenen, 2002).
Despite the regulations and the comprehensive framework contained in the city’s watershed protection plan, considerable uncertainties exist about whether the plan can sustain high quality water supplies over the longer-term. Enforcement of the regulations and monitoring the rapid rate of suburban growth constitute a major challenge, and these development pressures in the area may increase the opportunity costs of watershed protection. Construction in the headwaters of streams, permitted under the plan, may result in increased runoff rates and erosion. Filling tributary channels with sediments can take place incrementally, with each step occurring at a small scale. In addition, numerous small-scale changes may transform the watershed in detrimental ways over time without sufficient oversight and long-term planning. The U.S. Army Corps of Engineers (USACE) has authority under Section 404 of the Clean Water Act to review permits. However, without site-by-site reviews of small projects (less than four hectares), allowable incremental alterations can have significant cumulative effects on small streams. Decreased stream density (stream length per drainage basin area) would occur if natural stream channels were replaced by pipes and paved over for development, resulting in loss of the essential ecological processes of organic matter breakdown and sediment retention (Meyer and Wallace, 2001; Paul and Meyer, 2001).
Additional uncertainties might impact decision-making, besides the adequacy of protection in the watersheds. Model uncertainty that arises from imperfect understanding of ecosystem function and the translation to ecosystem services is a major issue for most ecosystem valuation studies. In this case, there is model uncertainty because the hydrologic modeling used for determining water supplies is affected by the definition of spatial and temporal boundaries. For example, other municipalities in New York and New Jersey use water from the Catskills. Changes in water diversions from the Catskill Mountains can affect outflows to the Delaware River and modify salinities in the lower sections of the river used by Philadelphia (Frei et al., 2002). Given the additional uncertainties of future regional droughts, floods, and extreme temperatures, as well as acid rain and nitrogen deposition from atmospheric sources, planners must consider the range of intrinsic natural variability in decision-making. Planners can cope with aspects of model and parameter uncertainty by carefully monitoring land uses in the basin and incorporating environmental data into any new regulations that might be required. A long series of studies on nutrient budgets and acid deposition provides some essential baseline information for the Catskills (e.g., Frei et al., 2002; Lovett et al., 2000; Murdoch and Stoddard, 1992, 1993; Stoddard, 1994). Other locations may lack sufficient information, and thus, considerable sources of uncertainty will limit the analysis of complete replacement costs.
In this case, the provision of clean drinking water supplies through the protection of natural processes in watersheds rather than through the human-engineered solution of building and operating a water filtration system offers an
estimate of the value of restoring an ecosystem service that provides clear advice to a policy decision. Replacement costs for natural processes in watersheds providing clean drinking water are estimated to be in the neighborhood of $6 billion to $8 billion, which is far higher than estimates of the cost necessary to protect the watersheds. Because the policy question is relatively specific (i.e., whether to build a filtration plant or to protect watersheds), currently available economic methods of ecosystem service valuation are sufficient.
Even in this example however, obtaining a precise estimate of the value of the provision of clean water through watershed conservation is probably not possible given existing knowledge. First, it is not clear that the two approaches, filtration and watershed protection, provide the same level of water quality and reliability. There are numerous dimensions to the provision of clean drinking water, such as the concentrations of various trace chemicals, carcinogens, and suspended solids, natural variance of water quality, and the adequacy of supply. It is unlikely that the two approaches will deliver water that is identical in all of these dimensions under all conditions. Second, there is no guarantee that protecting watersheds will continue to be successful. Increased development pressure on lands outside the riparian buffer zones or inadequate enforcement may require building a filtration system at some point in the future. If the watershed protection plans prove to be insufficient in the future, the investments in protection will still likely reduce future costs of building filtration plants because the quality of the water to be treated will be enhanced through these land-use programs.
Finally, it should be emphasized that (1) the value of providing clean drinking water is only a partial measure of the value of ecosystem services provided by the watershed, and (2) replacement cost is rarely a good measure of the value of an ecosystem service. Even if water quality benefits alone did not justify watershed protection, such a finding would not justify abandoning efforts at watershed protection. To make that decision would require a broader effort to measure the value of the wider set of ecosystem services produced by Catskill watersheds. It is less clear that estimates to answer this broader question are sufficiently precise to provide policy-relevant answers (see Chapters 2 and 6 for more on framing). Replacement cost methods can be used as a measure of the value of ecosystem services only when there are alternative ways to provide the same service and when the service will be demanded if provided by the least cost alternative. Replacement cost does not constitute an estimate of value of the service to society; it represents the value of having the ability to produce the service through an ecosystem rather than through an alternative method.
Other Surface Water Examples
Other cities have used similar strategies to invest in maintaining the ecological integrity of their watersheds as a means of providing high quality drinking water that meets all federal, state, and local standards. Boston, Seattle, San
Francisco, and Greenville, South Carolina, are other examples where the value of ecosystem services could be estimated using a replacement cost approach for building and operating water treatment plants that are roughly equivalent in the quality of drinking water supplied (NRC, 2000a). The costs of producing safe drinking water were traditionally derived from production cost estimates associated with engineering treatments. Filtration plants were built to remove organic materials, and then some form of chemical purification was used to control microorganisms. Engineers generally considered natural ecosystems such as rivers and lakes mostly from the viewpoint of volumes, transport systems, resident times, dilution, and natural “reoxygenation.” In other words, they viewed many natural ecosystems as large pipes rather than as complex habitats for a diverse biota. Yet even viewed strictly through the lens of water supply systems, protecting natural processes within ecosystems may be superior to engineering solutions, and such a result may be sufficient for decision-making purposes. Replacement cost estimates for provision of clean drinking water, however, provide an estimate of just one source of value and should not be confused with the complete value of ecosystem services provided by watersheds. Further, as discussed in Chapter 4, replacement cost is a valid approach to economic valuation only in highly restricted circumstances—namely, that there are multiple ways to achieve the same end and the benefits exceed the costs of providing this end.
Provision of Drinking Water from Groundwater: San Antonio, Texas
In contrast with the Catskill case, there has been a lack of studies to date on the economic value of the Edwards Aquifer (see also Box 3-5) that supplies drinking water to San Antonio as well as water for irrigation and other uses. Groundwater supplies approximately half of America’s drinking water (EPA, 1999). It is relied on heavily in some parts of the arid West where surface waters are scarce. The long-term supply of groundwater is a concern in some of these areas (Howe, 2002; Winter, 2001). For example, depletion of the Ogallala Aquifer is creating great uncertainties about future water supplies throughout a large region of the central United States (Glennon, 2002; Opie, 1993). Similarly, depletion of groundwater aquifers in the Middle Rio Grande Basin is creating uncertainty about the future supply of drinking water for Albuquerque, New Mexico (NRC, 1997, 2000b). Aquifers generally provide high quality drinking water, but pollution lowers water quality in some areas, such as the Cape Cod Aquifer where there are threats from sewage and toxic substances leaching into groundwater from the Massachusetts Military Reservation (Barber, 1994; Morganwalp and Buxton, 1999).
The long-term sustainability of groundwater depends on matching extraction with recharge (Sanford, 2002). It is often difficult to predict the timing and rate of recharge because of complications of local geology, time lags, and climate uncertainties. Recharge of the porous karstic limestone that characterizes
the Edward Aquifer occurs primarily during wet years when precipitation infiltrates deeply into the soils and underlying rock (Abbott, 1975). Drought conditions have complex effects on lowering recharge rates while simultaneously tending to increase the demand for water. The greatest source of uncertainty about groundwater recharge is the range of natural interannual variability in precipitation and land-use changes. Increasing demands from a growing population and the difficulty in predicting climate change raise questions about the adequacy of groundwater supplies in arid regions (Grimm et al., 1997; Hurd et al., 1999; Meyer et al., 1999; Murdoch et al., 2000).
Aquifer depletion has both economic and ecological consequences. The costs for deeper drilling and pumping increase as groundwater is depleted. Removal of water in the underground area may cause collapse of the overlying substrata. These collapses decrease future storage capacity below ground and may cause damage on the surface as areas subside, buckle, or collapse. In some areas, depleted groundwater may cause the intrusion of low-quality water from other aquifers or from marine-derived salt or brackish waters that could not readily be restored for freshwater storage and use.
Depletion of groundwater supplies creates uncertainty and generally is offset by supplies from surface waters. An interesting exception is San Antonio (the ninth largest city in the United States) that relies primarily on groundwater for its source of municipal water. An outbreak of cholera in 1866 from polluted surface waters prompted the City of San Antonio to switch to groundwater from the Edwards Aquifer. The aquifer is estimated to contain up to 250 million acrefeet of water with a drainage area covering approximately 8,000 square miles. The average annual recharge is estimated at approximately 600,000 acre-feet of water (Merrifield , 2000). Given this large supply, the Edwards Aquifer plays a major role in the economy of San Antonio and south-central Texas (Glennon, 2002). In some parts of this region, clean, free-flowing springs and artesian wells provide drinking water without the cost of pumping and with minimal treatment. San Antonio built its first pumping station in 1878. The U.S. Geological Survey (USGS) has monitored aquifer recharge rates since 1915 and water quality monitoring began in 1930. In 1970 the Edwards Aquifer was designated a “sole source aquifer” by the EPA under the Safe Drinking Water Act. Currently, more than 1.7 million people rely on the Edwards Aquifer for water. Industrial and agricultural demands on the Edwards Aquifer have increased, and the city has planned for new reservoir storage as part of its water supply several times over the last two decades. As the demand for water in the area has grown, concerns have arisen over both the quantity and the quality of groundwater available (Wimberley, 2001).
Depletion also raises the specter that adequate supply will not be available for future demand at any price. The $3.5 billion-a-year tourist industry in San Antonio is centered on the city’s River Walk, which relies primarily on recycled groundwater (Glennon, 2002). Uncertainties over the long-term availability of water make long-term planning problematic and threaten long-term investments. For example, aquaculture companies (e.g., Living Waters Artesian Springs,
Ltd.) expanded their catfish operations in March 1991, but subsequently closed in November 1991 because of concerns over pumping rates and the impaired water quality of return flows (i.e., high concentrations of dissolved nutrients) to surface- and groundwaters associated with the Edwards Aquifer.
Groundwater storage is critical in most aquatic ecosystems to provide persistence spring and stream habitats during dry seasons or during drought. Several springs (Comal, San Antonio, San Pedro) in the area began to dry up following a seven-year drought in the 1950s. Chen et al. (2001) used a climate change model to estimate the regional loss of welfare at $2.2 million to $6.8 million per year from prolonged drought. They estimated groundwater recharge based on historic data for recharge rates as influenced by precipitation and temperature. These researchers forecasted municipal and irrigation demand for five scenarios, including current condition and four different levels of climate change. Estimates of demand elasticity were based on models and methods used in other studies of arid regions. Given the projected reductions in available water, it would be necessary to protect endangered species in springs and groundwater at an additional reduction of 9 to 20 percent in pumping that would add $0.5 million to $2 million in costs.
The economic value of organisms living in groundwater and in springs, wetlands, and downstream surface flows supplied by groundwater is difficult to estimate. However, their value is generally assumed to be high because of their many functional roles in maintaining clean water as well as their existence values. For example, many diverse microbial communities and a wide range of invertebrate and vertebrate species live in groundwater, springs, and streams (Covich, 1993; Gibert et al., 1994; Jones and Mulholland, 2000). Their main functions are breaking down and recycling organic matter that forms the base of a complex food web (Covich et al., 1999, 2004). Depletion of groundwater aquifers results in possible loss of habitat for endemic species protected by state and federal regulations. For example, the Edwards Aquifer-Comal Springs ecosystem provides critical habitat for several endangered and threatened species, including salamanders (the Texas blind salamander and San Marcos Spring salamander), fish (the San Marcos gambusia and fountain darter), and Texas wild rice (Glennon, 2002; Sharp and Banner, 2000). In all, 91 species and subspecies of other organisms are endemic in this aquifer and its associated springs (Bowles and Arsuffi, 1993; Culver et al., 2000, 2003; Longley, 1986).
Most studies predicting groundwater supply focus on usable water quantities given drought frequencies and recharge. Land use is also important because it influences demand as well as runoff and recharge. As a result of water shortages in San Antonio, regulations controlling development were issued beginning in 1970. These regulations included rules for limiting economic development within the recharge zone. As noted previously, economic development often increases the extent of impervious surfaces that, in turn, cause more rapid runoff and loss of infiltration during and after precipitation events. Studies indicate that when impervious cover exceeds 15 percent of the surface of a watershed,
there are adverse impacts on surface water quality and subsurface water recharge (e.g., Veni, 1999).
The quality of groundwater is also an issue. Increasing concerns about water pollution of the Edwards Aquifer led former (now deceased) Congressman Henry B. Gonzalez of San Antonio to propose the Gonzalez Amendment to the Safe Drinking Water Act of 1974. The amendment dealt with protection of sole source aquifers used for water supplies (Wimberley, 2001). Leachate from landfills, leaking petroleum storage tanks, and pesticides all pose contamination threats that could render groundwater unusable. In 1987, a regional committee was formed to determine how the aquifer could be further protected. Henry Cisneros, then mayor of San Antonio, chaired the committee and proposed a plan that limited total withdrawals and called for a reservoir construction program (the Applewhite Reservoir was proposed but ultimately not approved).
A severe drought in 1990 and above-average pumping combined to dry up two of the aquifer’s major springs (Merrifield, 2000). In 1993, the Sierra Club sued the state under the Endangered Species Act for failure to guarantee a minimum flow of 100 cubic feet per second to Comal and San Marcos Springs (Sierra Club vs. Lujan, 1993 W.L. 151353). The State and the U.S. Fish and Wildlife Service entered into an agreement to resolve this conflict. The Texas legislature created the Edwards Aquifer Authority to control pumping and reallocate water through market mechanisms (McCarl et al., 1999; Schiable et al., 1999). This approach reallocated water from lower economic uses (such as agricultural irrigation) to higher-valued uses (such as domestic and industrial water supplies and environmental and recreational uses). In 2000, the Edwards Aquifer Authority decided to ban the use of any type of sprinkler in the eight-county region whenever flow at Comal Springs declined to 150 cubic feet per second (cfs) or less. In September 2002, the USGS reported that the flow had declined to 145 cfs and the ban went into effect.
Groundwater is a renewable resource that provides both extractive use value and in situ value. In situ value refers to the value created by having a stock of groundwater in the aquifer. Extraction of groundwater generates current extractive use value but can result in lower in situ value if extractions rates exceed aquifer recharge rates. Efficient use of groundwater requires extraction only when extractive use value per unit exceeds in situ value per unit of groundwater. Most economic analyses, such as those discussed above, have focused on extractive use values because these are most readily quantified. Extractive use values include the value of water for municipal and agricultural uses as well as recreation.
Characterizing the in situ value of groundwater is more difficult. Aquifer depletion imposes direct economic costs on water users by increasing pumping costs. Depletion can also impose costs through a loss of ecosystem services, such as processing of organic matter by diverse microbes and invertebrates, providing possible dilution of some types of surface-originating contaminants, and sustaining populations of rare and endangered species that are often restricted to very local habitats (Culver et al., 2000). Further, depleting the stock of ground-
water means that less water is available for use, or for maintenance of ecosystem services in the future. With uncertain recharge because future precipitation is uncertain, there is an insurance value from maintaining adequate groundwater stocks. Maintaining adequate stocks helps avoid shortages during drought years, prevents land subsidence, and provides late summer supplies of water to springs and streams for sustaining fisheries and wildlife and for recreational uses (NRC, 1997). Estimating in situ values of groundwater requires a dynamic model that incorporates expected recharge rates, pumping costs, and demand through time. Dynamic renewable resource models of groundwater with uncertain recharge exist and could provide a basis upon which to estimate in situ values (Burt, 1964; Provencher, 1993; Provencher and Burt, 1994; Rubio and Casino, 1993; Tsur and Zemel, 1994), though uncertainties about local hydrology would make it difficult to know the correct model specification (model uncertainty).
The construction of water transfer pipelines and additional surface storage reservoirs in San Antonio is under consideration along with conjunctive storage (pumping water into subsurface storage associated with aquifers). Although surface water can substitute for groundwater for extractive uses, surface water and groundwater do not contribute to the same ecosystem functions nor do they provide the same set of ecosystem services. At present, alternatives to continued reliance on groundwater are on hold because city voters rejected development of the proposed Applewhite Reservoir as an alternative water source.
Dependence on a sole source aquifer leaves communities subject to the risk that they will not have adequate water supply if it is depleted or polluted. As population and economic activity continue to increase in the San Antonio area, it seems unlikely that the Edwards Aquifer will be sufficient to meet future demand for water. Attempts to purchase water from surrounding counties and to build more storage have been under consideration for decades but have not yet materialized. While the establishment of a water market will help reallocate a fixed amount of water to high-value uses, it does not guarantee that adequate supply will be available (Merrifield and Collinge, 1999). Weighing the benefits of extractive use of groundwater versus the value of water in situ for insurance against future drought and for maintaining natural ecosystem functions and the survival of endangered species poses difficult questions. Uncertainties about potential climate change, local hydrology, and the likely future value of ecosystem services, such as provision of drinking water and habitat necessary for the survival of endangered species, complicate the task of informing decision-makers about trade-offs between current extractive use value and in situ value of groundwater. Predictions about likely future aquifer recharge and water demand, as well as evidence about the value of other ecosystem services, such as habitat provision for endangered species, all would help in guiding decisions.
Valuation of Fish Production Provided by Coastal Wetlands and Estuaries
Coastal wetlands (e.g., seagrass meadows, marshes, mangrove forests) are increasingly recognized as providing economically valuable ecosystem services. One of the most important services provided by coastal wetlands is the provision of important habitat for many species of commercially harvested fish, crustaceans, and mollusks (Beck et al., 2001). Given their high diversity and productivity, coastal wetlands are often referred to as nurseries (Boesch and Turner, 1984; NRC, 1995).
The economic value of coastal wetlands as breeding and nursery grounds can be estimated using a production function approach (see Chapter 4 and Appendix C). In economic terms, a coastal wetland is like a production facility or factory that transforms inputs (nutrients, energy) into valuable outputs (fish, crustaceans, and mollusks). The production function approach applied to fisheries requires being able to estimate the increased quantities of various marketable species produced when coastal wetlands are preserved. Then, the value of the coastal wetland as breeding and nursery grounds can be estimated by calculating the increase in consumer and producer surplus due to the increased production. Barbier (2000) provides a review of production function approaches to economically valuing the ecological function of coastal wetlands as breeding and nursery grounds.
Estimates of value of coastal wetlands for fisheries production have ranged widely. For example, Barbier and Strand (1998) estimated that conversion of one square kilometer of mangrove in Campeche, Mexico, to other than natural uses reduced the value of annual shrimp harvest by more than $150,000 for 1980 to 1981. Such a large value argues for protecting the mangroves even when ignoring the value of other ecosystem services. On the other hand, Swallow (1994) found that loss of normal-quality wetlands reduced fishery values by an estimated $2.77 per hectare, or $277 per square kilometer. Swallow concluded that protecting normal-quality wetlands is not justified because the economic value of increased value of shrimp production is less than the value of agricultural development. Basing such a conclusion on the economic value of a single ecosystem service, however, is premature; only when the value of all ecosystem services provided by the wetland is less than the value of agricultural development can such a conclusion be justified.
A major difficulty with the production function approach in the context of coastal wetlands and fisheries is the complex nature of the ecological relationships involved. Subtle changes in nutrient cycles, water temperatures and currents, and fluctuations in the populations of predators and prey, all can have a large influence on the number of fish that reach adulthood. Large variations in fish populations occur even with no apparent change in physical conditions.
The production function models of wetlands and fisheries employed by economists to date have assumed simple ecological relationships that ignore most of this complexity. Starting with Lynne et al. (1981), these models assume
that the productivity of the systems is a simple nonlinear function of the area of coastal wetlands. Static production function models assume that productivity increases with the natural logarithm of area (Bell, 1989, 1997; Farber and Costanza, 1987; Lynne et al., 1981), or that the natural logarithm of productivity increases with the natural logarithm of area (Ellis and Fisher, 1987; Freeman, 1991). Dynamic production function models (Barbier and Strand, 1998) include effects of population stock size as well as area of coastal wetlands. Increasing coastal wetland area shifts the natural population growth function up (stock-recruit function) that defines population in one period as a function of the population in the previous period. However, both the static and the dynamic production function models do not account for other important environmental factors such as the aforementioned nutrient cycling, temperature, or currents, nor do they attempt to account for stochasticity in ecological conditions or in species populations. While these models are suggestive of increased fisheries productivity from wetlands, more work is needed before quantitative estimates of the value of increased productivity can stand up to critical review. An ongoing challenge will be to discern realistic ecological relationships between structure and function of coastal wetland ecosystems and fisheries productivity amid the complex and seemingly chaotic fluctuations in fishery stocks.
How fisheries are managed also influences estimates of value (Freeman, 1991). An optimally managed fishery typically generates far higher economic returns than does an open-access fishery. For example and as noted previously, Barbier and Strand (1998) estimated that the annual value of a square kilometer of mangrove was more than $150,000 in 1980 to 1981, but dropped to less than $90,000 in 1989 to 1990 when overfishing had depleted stocks, resulting in lower harvests. In addition, market prices, which depend on consumer preferences as well as production from other ecosystems, will affect estimates of value.
For commercially marketed outputs, well understood methods can be used to estimate the change in consumer plus producer surplus from a change in available resource stock. The major difficulty in applying the production function approach is the great uncertainty typically present in understanding the link between structure and function of coastal wetlands and productivity of fisheries. Complexity of ecosystems, chance events, and natural variability of populations all make it difficult to discern the input-output relationships that are necessary for estimating a production function. Assumptions about fisheries management and market conditions will also influence estimates of economic value.
Provision of Flood Control Services by Floodplain Wetlands
Flood control is an important ecosystem service provided by riverine and coastal floodplains. Floodplains absorb excess water during floods that otherwise might inundate and damage developed areas. In addition to providing flood control, floodplain ecosystems provide critical resources for plant and
animal communities. Despite their importance, humans have attempted to replace or supplement natural flood control services provided by floodplains by building flood control structures (e.g., dams, reservoirs, levees, floodwalls). The magnitude of flood control infrastructure development is evidenced by the fact that as a result of the Mississippi River flood of 1927—which inundated 5.26 million hectares and forced 700,000 persons to relocate—Congress authorized $325 million for flood control works on the Lower Mississippi River, which at that time was the largest public works expenditure in U.S. history (Hey and Philippi, 1995; Wright, 2000). In fact, during the height of the flood control movement spanning 1936 to 1951, Congress spent more than $11 billion for flood control projects (Wright, 2000). Although development of this regionally engineered infrastructure has protected some areas of the United States from flood damage, it has also served to promote floodplain development. Such development ultimately exacerbates levels of flood damage during large precipitation events. Furthermore, flood control structures have often given farmers and city dwellers a false sense of protection.
In principle, flood control services provided by floodplain ecosystems can be clearly defined and quantified. They are an input into production of a valuable service, namely reducing the probability of damage from floods. In this sense, floodplain ecosystems perform a role in of flood control similar to that of coastal wetlands in fishery production—one valuation method is to estimate how changes in the ecosystem lead to changes in production of the service in question and then to value the change in the service. The simplest method for economically valuing floodplain ecosystems in providing flood control is to multiply estimates of the change in probability of floods of various magnitudes with and without floodplain conservation by the estimate of damage that floods of various magnitudes would cause. This method is essentially what insurance companies routinely do in assessing risks.
A complication in assessing flood control is that measures to prevent floods or ameliorate the damage may cause changes in human behavior. For example, if the risk of building in a floodplain is lowered, there is less reason to avoid floodplain development. Further, if those building in the floodplain do not have to pay full costs for damages from floods (e.g., they are provided with subsidized flood insurance or with disaster payments that reimburse damages from floods), then one might expect excessive development in floodplains. Insurance companies are no stranger to this phenomenon, which has been referred to as a “moral hazard.” Conducting an assessment of the value of flood control services depends on assumptions about patterns of development and infrastructure. Assuming that existing buildings and infrastructure are fixed and immovable will result in a different answer than an approach that factors in a behavioral response. While doing the latter is more realistic, it is also more difficult.
Another complication in evaluating wetlands and floodplains in providing flood control is that the value of this service also depends on human-engineered infrastructure in the form of dikes, levees, or flood control dams. Floodplain ecosystems and dams are alternative ways to prevent floods, similar to water-
sheds as alternatives to filtration plants to produce clean water. Information relevant to the value of floodplains in providing flood control is given by avoided costs of human engineered flood control through dikes, levees, or flood control dams. For example, the USACE opted to purchase 3,440 hectares of floodplain wetlands in the upper portion of the Charles River watershed in Massachusetts. By protecting this land, the Corps estimated that almost 62 million cubic meters of water could be stored on the floodplain—similar to the capacity of a proposed dam. Purchase of the development rights to these floodplain wetlands cost $10 million, which was one-tenth of the $100 million estimated for the dam and levee project originally proposed (American Rivers, 1997; Faber, 1996). This natural wetlands flood control system was able to deal with large floods during 1979 and 1982. For a discussion of replacement cost as a method to estimate the economic value of an ecosystem service see the discussion of the Catskill watershed above.
The Napa River Flood Protection Project in California provides another example that includes both structural and nonstructural flood protection approaches. These range from residential and commercial development relocation, to road reconstruction and bridge removal, along with floodplain reconstruction of 80 hectares of seasonal wetlands, intertidal mudflats, and emergent marshlands. The $155 million cost of the project is a fraction of the estimated $1.6 billion that would have to be spent by Napa County to repair flood damage over the next 100 years if the project is not implemented. The project is projected to save the community $20 million annually (USACE, 1999).
Although much anecdotal information exists regarding how flood damage is related to alterations of natural floodplains and subsequent development in high flood risk areas, determining what percentage of total flood damage costs can be attributed to wetland drainage and floodplain alterations is difficult. For example, in the Upper Mississippi River basin, a strong relationship was found between flood damage and wetland destruction; areas having fewer wetlands due to drainage generally suffered greater flood damages. Likewise, in the Puget Lowlands in Washington State, water discharge events (with a recurrence interval of 10 years prior to urbanization) increased in frequency (to a recurrence interval of 1 to 4 years) after urbanization, with the increase in probability of flooding proportional to the degree of urbanization (Moscrip and Montgomery, 1997).
Wetlands and floodplains generate other services that benefit the public, such as wastewater reclamation and reuse, pollution abatement, aquifer recharge, and recreation. One study that attempted to estimate values for a range of ecosystem services in monetary terms is a study of the multipurpose Salt Creek Greenway in Illinois (Illinois Department of Conservation, 1993; USACE, 1978). The sum of the natural values of floodplain land, other than for flood control, was estimated at $8,177 per acre. The estimated value of regional floodwater storage was $52,340 per acre (Forest Preserve District of Cook County Illinois, 1988). Combining these estimates provides an estimated total value of preserved floodplain land of $60,517 per acre. Such high values indi-
cate that preserving floodplain ecosystems was the best use of such land, far outstripping its value in agriculture or development. Demonstrating the magnitude of these values in a clear and convincing fashion would encourage sensible land use decisions that include the preservation of floodplains where their value is high (Scheaffer et al., 2002).
In general, the value of an ecosystem service will vary with its level of provision. For example, the preservation of an additional acre of floodplain wetlands will tend to be quite valuable when only a few acres of wetlands have been similarly preserved and the probability of flooding is high. In contrast, the value of preserving an additional acre of wetlands will tend to be smaller when many acres of wetlands have already been preserved and the probability of flooding is low. Estimates such as those provided in the preceding paragraph are stated in a way that makes it seem as if the value of an additional acre of floodplain wetlands is constant. Indeed, estimates of marginal changes are sometimes derived by equating them with the average value per unit over a large change. When marginal values are not constant however, this will result in biased estimates of marginal value.
Reasonably good information to estimating the value of floodplain ecosystems in providing flood control, at least in some cases. Hydrologic models can be used to estimate the amount of water that a floodplain ecosystem can absorb during a flood. Economic values from lowering the risk of damages from floods can be estimated with reasonable precision and, in fact, are calculated by government agencies and private insurance companies on a regular basis. Trying to incorporate changes in human behavior or investments in flood control infrastructure are complications that can affect valuation estimates. As with the other cases of estimating the value of single ecosystem services, such estimates should not be confused with estimates of the value of the ecosystem itself, which would require estimates of a range of ecosystem services.
Studies that focus on economically valuing a single ecosystem service show promise of delivering results that can inform important environmental policy decisions. In some cases, the valuation exercise is clearly defined, there is sufficient natural science understanding and information available, and well-supported economic valuation methods can be applied to generate reliable estimates of value. The provision of drinking water for New York City by protecting watersheds in the Catskills is an example in which evidence of the cost of replacing an ecosystem service informed decision-making. In other cases, the valuation of ecosystem services has not advanced far enough to provide clear and compelling evidence for formulating policies that are likely to be accepted by competing interests. Although some information is available, more work is necessary before reasonably precise estimates of the value of in situ groundwater can be made in the case of the Edwards Aquifer. The impacts of drought and
legal issues regarding endangered species and rights to groundwater make such economic valuation efforts quite complex. Similarly, while providing useful information, studies on the value of coastal wetlands for fishery production are in need of further refinement before a high degree of confidence can be attached to estimates of economic value. Even where there is reasonably good information and valuation methods are available, details about ecological functions, the dynamics of ecosystems, human institutions, and human behavior can make estimation of economic value a difficult task. However, the limited scope of valuing a single ecosystem service allows researchers to address many of these complications.
One danger inherent in the economic valuation of a single ecosystem service is mistaking this value for the value of the entire ecosystem. Ecosystems produce a wide range of services and the value of a single service will necessarily represent only a partial valuation of the entire ecosystem. Sometimes this partial valuation is enough for purposes of decision-making, as in the New York City example. Other times, as in the case of Swallow’s (1994) integrated ecological-economic analysis of the impacts of wetlands conversion on coastal shrimp nursery habitat in North Carolina, it will not be enough. Although that particular study provides a reliable estimate of the economic costs of wetlands conversion in terms of loss of key hydrological function and consequent effects on shrimp nursery habitat, other important ecosystem services provided by wetlands were not considered or addressed. Thus, there is a danger that the study could be used to advocate too much conversion of wetlands with the concomitant loss of a multitude of ecosystem services.
Valuing Multiple Ecosystem Services
This section reviews three examples that estimate the economic value of multiple services from an ecosystem. As discussed throughout this report, ecosystems provide a wide range of services. Because of the interconnection of processes within an ecosystem, it may be difficult to isolate and study the production of one ecosystem service without simultaneously considering other services. Further, production of some ecosystem services may be in conflict with provision of others. In such cases, providing clear policy advice requires the simultaneous estimation of multiple ecosystem values. Expanding the range of ecosystem services covered brings the resulting estimates of economic value closer to providing an accurate estimate of the value of all ecosystem services. Nevertheless, these studies, although more comprehensive than single ecosystem service studies, still represent only partial estimates of the complete economic value of services generated by an ecosystem.
Fish Production, Irrigation Waters, Navigation, Flood Control, and Clean Drinking Water: The Columbia River Basin
The Columbia River basin is the fourth largest in North America, covering large portions of the States of Idaho, Oregon, and Washington and the Canadian province of British Columbia. The Columbia River provides a wide range of ecosystem services including hydroelectric power, water supply for municipalities and industries, irrigation for agriculture, transportation, recreation, fish production, and diverse aesthetic values. The basin is highly developed and contains a large number of dams, including 18 on the mainstem of the Columbia and Snake Rivers; most of the large dams are multipurpose (i.e., hydroelectric power generation, flood control, irrigation, recreation, municipal and industrial water supply). Besides hydroelectric power generation, a major economic benefit of the dams is storage of snowmelt runoff and diversion of water for irrigated crops during the growing season. Navigation is also enhanced by maintenance of sufficient river depths. The dams along the Lower Columbia and Snake Rivers allow barge transportation to Lewiston, Idaho, making it a port with access to the ocean despite being located 465 river miles inland.
However, the dams along the Snake River and the mainstem of the Columbia River have been at the center of a major controversy. On the one hand, dams provide a range of economic benefits as listed above; on the other hand, dams are blamed, at least in part, for declines of Columbia and Snake River salmon stocks. One study estimated that the number of wild adult salmon returning to the Columbia River was less than 10 percent of the presettlement numbers of 8 million to 10 million (NRC, 1996). Several fish stocks are listed on the federal threatened and endangered species list including: spring- and summer-run chinook, fall-run chinook, sockeye, steelhead, and bull trout in the Snake River; spring-run chinook, steelhead, and bull trout in the Upper Columbia; steelhead and bull trout in the Mid-Columbia; and chinook, chum, steelhead, and bull trout in the Lower Columbia. The dams have fundamentally changed the ecology of the river, altering it from free-flowing to a chain of reservoirs linked by rivers that impact both downstream migration of juvenile fish and upstream migration of spawning adults (Deriso et al., 2001; NRC, 1996; Schaller et al., 1999). These dams have also closed-off access to 55 percent of the drainage area and 31 percent of the stream miles of original salmon and steelhead habitat in the Columbia River basin (NRC, 1996).
However, dams are not thought to be the only reason for the decline in the wild salmon population in the Columbia River basin. Urban development, industry, agriculture, grazing, mining, forestry, the large-scale introduction of hatchery fish, fish harvesting, ocean conditions, and climate change are also implicated. Forestry and grazing practices that result in reduced streamside vegetation can increase water temperatures above beneficial levels for salmon (Beschta, 1997; Beschta et al., 1987; Platts, 1991; Rishel, 1982). In fact, failure to attain stream temperature standards is the most prevalent water quality violation in the Pacific Northwest (Wu et al., 2003). Water withdrawals for irrigation
reduce instream flow and water diversions without screens lead to loss of juvenile fish (Jaeger and Mikesell, 2002; NRC, 1996). Removal of woody debris, changes in water velocity, and erosion causing increased siltation of streams also negatively impact salmon populations (Hicks et al., 1991; NRC, 1996). Furthermore, ocean and climate conditions influence salmon populations, including decade-long changes in ocean conditions that affect currents and upwelling in the Pacific Northwest (Hare et al., 1999; Nickelson, 1986); interannual variability in precipitation influenced by El Niño-Southern Oscillation and other periodic climate shifts (Hamlet and Lettenmaier, 1999a,b; Miles et al., 2000); and long-term climate change (Beamish and Mahnken, 2001; Beamish et al., 1999; Pulwarty and Redmond, 1997).
Decision-making about fisheries management, land management, and the operation of the hydroelectric dams involves calculations of the effect on salmon populations and on other valued ecosystem services. The effects of various alternative management actions on salmon stocks and on electricity generation, irrigated agriculture, navigation, and other economic activities have been analyzed in a number of ecological and economic studies (NRC, 2004). Debates on whether to remove hydroelectric dams on the Lower Snake River focused attention on the costs and benefits of dam removal. Several recent ecological and economic studies analyze the effects of the removal of dams (Budy et al., 2002; Grant, 2001; Gregory et al., 2002; Kareiva et al., 2000; Levin and Tolimieri, 2001; Poff and Hart, 2002; Schaller et al., 1999). The benefits of restoring migratory routes for fish to upper headwaters are widely appreciated. The costs of removing sediments that accumulate in reservoirs by dredging or by allowing sediments to be washed downstream and alter spawning substrates (by infilling gravels with fine mud) are difficult to quantify but often significant. Furthermore, elimination of some dams that currently form barriers to fish migration (preventing non-native species from moving upstream and displacing native fish species) may be important costs, not benefits, in some rivers. The USACE estimated that forgone economic benefits that would occur with the removal of four dams on the Lower Snake River would be $267 million annually (USACE, 2002), though Pernin et al. (2002) derived far lower estimates of forgone benefits from dam removal. At present, there is no consensus on how costly dam removal would be or on how effective such actions would be for salmon recovery throughout the Columbia River Basin.
Studies have been undertaken of the costs and benefits of enhancing river flows or restoring more natural patterns of flow, such as allowing more spring flooding to remove fine sediments to enhance spawning conditions (Adams et al., 1993; Fisher et al., 1991; Jaeger and Mikesell, 2002; Johnson and Adams, 1988; Moore et al., 1994, 2000; Naiman et al., 2002; Paulsen and Hinrichsen, 2002; Paulsen and Wernstedt, 1994; Wernstedt and Paulsen, 1995). Some of these studies include integrated ecological and economic models that build from biological models of fish populations to economic models of the valuation (Adams et al., 1993; Johnson and Adams, 1988; Paulsen and Wernstedt, 1995; Wernstedt and Paulsen, 1995). Studies by Johnson and Adams (1988) and Ad-
ams et al. (1993) estimated the value of increased flows in the John Day River in Oregon for recreational steelhead fishing. These researchers estimated changes in fish population by combining a hydrologic and a biological model. They then combined this estimate using contingent valuation methods to derive an estimate of value for an increased fish population.
Economic studies that focus strictly on valuing recreational or sportfishing in the Pacific Northwest include Olsen et al. (1991) and Cameron et al. (1996); though other studies have valued salmon fishing in Alaska (Layman et al., 1996) and central California (Huppert, 1989). Valuation estimates vary depending on the location of the study and the methodology employed. Other studies have focused on costs of providing increased streamflows (Aillery et al., 1999; Jaeger and Mikesell, 2002; Moore et al., 1994, 2000). Jaeger and Mikesell (2002) noted that the costs of augmenting streamflows to increase the survival of native fish in the Pacific Northwest are likely to be “modest” (between $1 and $10 per capita per year within the region). Studies have also evaluated the costs and benefits of modifying habitat condition (Loomis, 1988; Wu et al., 2000) and decreasing stream temperatures (Wu et al., 2003). Another area of research is on the cost-effectiveness of fish hatcheries that were initially built to offset losses of migratory fish after dam construction (Congleton et al., 2000; Levin et al., 2001; Lichatowich, 1999; Meffe, 1992). Populations of hatchery-reared fish are known to have different genetic composition and behaviors than wild populations of the same species, and in some cases, these hatchery-reared fish may compete with or breed with wild populations thereby diminishing the stocks of those populations best adapted for long-term survival in the wild (Fisher et al., 1991).
Efforts to rebuild salmon stocks have been going on for several decades. The Pacific Northwest Electric Power Planning and Conservation Act of 1980 created the Northwest Power Planning Council to create a plan “to protect, mitigate and enhance fish and wildlife, including related spawning ground and habitat, on the Columbia River and its tributaries while assuring the Pacific Northwest an adequate, efficient, economical and reliable power supply.” Despite legal authority and expenditures of more than $3 billion to date (Northwest Power Planning Council, 2001), salmon populations have not recovered.
In part, this failure is due to the lack of scientific understanding about what measures are likely to be effective in restoring salmon: “The list of central topics that we know too little about is surprisingly long. The topics include, for example, the survival of young fish between dams compared with their survival as they pass through and over dams; the relationship of survival of young fish to the flow rates of water in rivers; the effects on survival of various management practices including logging, grazing, irrigation, agriculture, and use of hatcheries, the influence of ocean conditions…” (NRC, 1996). Such pervasive uncertainty has led to calls for increased research effort to reduce critical uncertainties (NRC, 1996) and for adaptive management (Lee, 1993, 1999; Walters, 1986). Several studies have analyzed the value of reducing uncertainty by learning or better forecasting ability (Costello et al., 1998; Hamlet et al., 2002; Paulsen and
Hinrichsen, 2002). At present, managers face a difficult challenge in making decisions under uncertainty (see also Chapter 6). Sometimes decisions cannot wait for science to provide clear evidence, but decision-making without clear evidence allows the management policies to be attacked as excessively risky. Such policies impose potentially high costs on certain sectors of society while lacking an adequate basis of scientific support to show that they will be either biologically effective or efficient (cost-effective). The fact that some consequences are irreversible (e.g., extinction) raises the stakes further.
Questions such as how to recover salmon populations and how to protect or restore other ecosystem services in the Columbia River basin have been, and likely will continue to be, contentious issues. The costs of recovery efforts for salmon are high, already topping several billion dollars (Northwest Power Planning Council, 2001). Changing the fisheries management, regional land use, or operation of dams could lead to fundamental changes in the functioning of the ecosystem, with consequent effects on the production of multiple ecosystem services, ranging from hydroelectric power generation to the existence value of salmon. At present, there are large gaps in the scientific understanding of how such changes would impact important elements of the ecosystem, particularly salmon populations. Even if those scientific controversies were resolved, difficult valuation questions would remain. Furthermore, estimating the existence value and spiritual value of salmon with currently available economic valuation methods is controversial (some would argue economic methods cannot fully capture such values; see Chapter 2 for further information). The large and uncertain costs and benefits of alternative proposals, which will fall disproportionately on different groups within society, amplify the difficulty of decision-making. The political nature of this controversy will make it a difficult arena for ecosystem valuation to be viewed as rational, objective, and conclusive. Despite these challenges, it is important to try to impart good information to such debates.
Upstream Versus Downstream Water Use: Losses in Downstream Economic Benefits as a Result of Upstream Diversion from Dams
The development of the Hadejia-Jama’are floodplain in northern Nigeria is one of many examples worldwide where water diversion upstream (associated with dams) is negatively affecting economic activities downstream. Supporters of dams and water diversion projects typically point to the economic benefits created by such projects but often fail to consider costs imposed elsewhere. In this particular case, economists and hydrologists worked together to estimate both upstream benefits and downstream costs (Acharya and Barbier, 2000, 2001; Barbier, 2003; Barbier and Thompson, 1998). These studies are among the few integrated case studies to assess the impact of upstream water allocation on wa-
ter availability and groundwater recharge downstream and to value the effects on irrigated agriculture and potable water supplies downstream.
Barbier and Thompson (1998) combined economic and hydrological analysis to compare the benefits of upstream diversion with losses of downstream floodplain benefits in terms of agriculture, fishing, and fuel wood. They found that fully implementing all existing and planned upstream irrigation projects results in losses of approximately $20 million (1989-1990 U.S. dollars) versus the case with no irrigation upstream. Full implementation of upstream irrigation project generated estimated benefits of approximately $3 million, while floodplain losses were estimated to be around $23 million. Acharya and Barbier (2000, 2001) analyzed impacts of a one meter drop in groundwater from lower water recharge in the floodplain on dry season agriculture and rural domestic water use in villages. They estimated annual losses of $1.2 million in irrigated dry season agriculture and $4.8 million in domestic water consumption for rural households. These analyses strongly suggest that expansion of existing irrigation schemes within the river basin is not economically desirable (Barbier, 2003).
In a very different setting, Berrens et al. (1998) reported similar conclusions about upstream diversions of water. The purpose of this study was to analyze the costs of imposing minimum instream flow regulations in the Colorado River to protect endangered fish species. However, instead of costs they found that imposing instream flow restrictions generated overall positive net benefits because it allowed more water to be used further downstream where it would be put to higher-valued uses.
Cumulative alterations in hydrologic connections in the landscape exert major environmental and economic effects at different spatial scales (e.g., Pringle, 2001). In the last decade, ecologists have begun to identify and quantify the substantial environmental consequences of dams on local, regional, and even global scales (e.g., McCully, 2002; Pringle et al., 2000). However, relatively few integrated studies have evaluated economic consequences from hydrologic modifications and the resultant changes in provision of ecosystem services. Even at local scales, studies are conspicuously lacking that attempt to quantify the economic costs to downstream human activities from upstream water diversions such as those associated with dams. In many cases, damage assessments are attempted decades after a dam is completed so research is dependent on historical records to recall or reconstruct wetland environments and associated economic activities that once existed. For example, researchers are dependent on midden piles (i.e., a collection of biotic materials that can provide a paleoenvironmental history of an area) to assess the extent of shellfish production near the mouth of the Colorado River before dams diverted virtually all of its flow.
Fully evaluating the consequences of many projects, such as dams and water diversions, requires assessment of the change in value of ecosystem services that may play out at different spatial scales. Some of the consequences may occur far removed from the site of the project, such as consequences to downstream environments (floodplains, deltas, etc.). As the case studies of the
Hadejia-Jama’are floodplain illustrate, a full accounting of downstream consequences can generate a different perspective of whether a project generates positive or negative net benefits.
Other well-known examples, such as water use in the Colorado River, the hypoxic zone in the Gulf of Mexico caused by high nitrogen runoff from Mississippi River drainage, and the drying of the Aral Sea due to upstream diversion of water, further illustrate the importance of considerations of downstream consequences. Ecosystem processes are often spatially linked, especially in aquatic ecosystems (see Chapter 3 for further information). Full accounting for the consequences of these actions on the value of ecosystem services requires understanding these spatial links and undertaking integrated studies at suitably large spatial scales to fully address important effects.
Food Production, Recreational Fishing, and Provision of Drinking Water from Lakes: Lake Mendota, Wisconsin
In many ecosystems it is difficult to isolate the economic value of a single good or service because of the complex connections among species and ecosystem functions. For example, food production such as a largemouth bass may seem obvious as an economic “good” derived from a lake ecosystem. Similarly, the recreational value of fishing may be measured by economic analysis as another good. However, much of an ecosystem’s productivity may not produce a harvestable yield of interest to human consumers (algae or other aquatic plants). Furthermore, the type of fish (largemouth bass, lake trout, or carp) may also vary in value as products for either food or recreation. Although productivity is a fundamental measure of ecosystem functioning (see Box 3-1), it is different from what economists would typically use to evaluate human uses of ecosystem function. Generally, ecologists measure units of energy required for a species maintenance (respiration) and the energy converted to live matter (biomass) per unit area per unit time as the total productivity, whereas economists focus on harvestable amounts of certain desirable species as the valuable yield or one type of good produced by the ecosystem. Breakdown of dead organic matter through decomposition by microorganisms might be deemed an ecosystem service that maintains clean water in the lake, but its economic value is difficult to isolate from the recycling of nutrients needed for the productivity of plants and animals. Clean drinking water, food production, and recreation are all products of a lake ecosystem, but it is not easy to measure each one separately or to resolve conflicting views on which one is more or less important if trade-offs in management decisions are required. Removing excessive nutrients from a lake will improve drinking water quality (up to some point), but the resulting effect on fish production requires careful study of the entire food web.
Lake Mendota, located on the edge of the campus of the University of Wisconsin, Madison, is probably the most thoroughly studied medium-sized lake (>4,000 hectares) in the world (e.g., Brock, 1985; Kitchell, 1992; Lathrop et al.,
1998, 2002). In the early 1980s, the combined decline of walleye populations and recreational fishing, together with concerns over unpredictable outbreaks of noxious and sometimes toxic Cyanobacteria (blue-green algae) in the lake, led to a joint research effort that demonstrated that water quality and food web management could be successfully integrated. This research effort focused on the following issues: (1) trade-offs between increased stocking for walleye and northern pike fishing or managing for bass or perch (distinctly different goods for different groups of people); (2) effects of increased water clarity following removal of algae by grazing zooplankton on deep light penetration that can result in increased growth of submerged aquatic plants2; and (3) effects of improved water quality (clear water with lower concentrations of dissolved nutrients) that may reduce fish productivity and result in lower recreational fishing harvest levels. Finding the right balance of the production of various ecosystem goods and services is challenging, especially since what happens in the lake ecosystem depends on management decisions for the surrounding land as well. Inflowing waters from agricultural sources and municipal sewage treatment plants can provide excessive nutrients without appropriate land and municipal wastewater management. Conventional management approaches often focus on one sector at a time. However, management to address the problems of one sector may increase problems in other sectors if important interconnections are ignored. Successful management requires understanding the linkages between sectors and may require interdisciplinary teams to address complex multisector issues.
Economic analyses of ecosystem services of Lake Mendota (Stumborg et al., 2001) and similar lake ecosystems have considered costs and benefits of managing eutrophication relative to recreation, real estate values, drinking water quality, and other site-specific attributes (Boyle et al., 1999; Brock and de Zeeuw, 2002; Carpenter et al., 1999; D’Arge and Shogren, 1998; Wilson and Carpenter, 1999). These studies illustrate the unique aspects of Lake Mendota that constrain benefits transfer of results to other lakes. They also highlight the considerable uncertainties in lake management. Significant sources of uncertainties are related to high levels of temporal variability in lake ecosystem dynamics, surrounding land-use changes, and hydrological variables. For example, regional droughts greatly reduce inflows, increase residence times of nutrients, and often decrease transport of suspended sediments that affect water quality by altering turbidity and light regimes, as well as influencing nutrient input, transport, and cycling (Kitchell, 1992). Land clearing for development generally increases peak flows of runoff, increases bank erosion of tributaries that drain into lakes, and greatly increases turbidity. Thus, despite intensive programs to remove nutrients from point sources such as sewage treatment plants, continued input of nutrients from diffuse, nonpoint sources (e.g., fertilizers from
agricultural runoff, soil erosion, septic tanks) remains a major challenge in many watersheds (NRC, 2000).
Aquatic ecologists manipulated fish and zooplankton species to regulate algal production and restore clear water to lakes. Some lakes were covered with green scum and characterized by fish kills resulting from deoxygenation during warm-water periods in late summer. Ecologists learned that successive, small increments of phosphorus additions to lakes were critical to eutrophication in many situations. The ratio of phosphorus to nitrogen was also found to alter the species composition of the planktonic algae. Low values of phosphorus led to the dominance of lake waters by green algae that were readily consumed by grazing zooplankton and fish. Incremental nutrient additions caused lakes to flip from one state (clean water) to another (green, turbid water) that altered ecosystem services and lowered real estate values of surrounding property (Carpenter et al., 1999; Wilson and Carpenter, 1999).
Although harvesting fish was known to remove nutrients, especially phosphorus, and to alter pathways of food webs to minimize algal blooms, the effects of large-scale applications of this approach to managing water quality in Lake Mendota and other lakes remained unknown until a number of field experiments and models were completed (DeMelo et al., 1992; Gulati et al., 1990; Kitchell, 1992; Reed-Anderson et al., 2000). The concept of removing some dissolved nutrients from the open waters by optimizing their incorporation into green algae that is later consumed by zooplankton, and then by juvenile fish, was widely understood to work in small ponds but was not often tested in lake ecosystems. Excretion of nutrients by grazers and predators can increase nutrient turnover and productivity, but understanding and stabilizing the balance of different consumer species in food webs remains complex. Lake management efforts use a combination of biomanipulation of food webs (Shapiro, 1990), diversions of some tributaries that have high nutrient loadings, and nutrient removal technologies that focuses on point sources. This combined management approach provides an opportunity to examine trade-offs between alternative investments in water pollution control and recreational fisheries management.
As the case studies in this section illustrate, aquatic ecosystems produce multiple services, many of which are closely interconnected. These interconnections often make it difficult to analyze one service in isolation. For example, a dam that diverts water from a river or increases nutrient input to a lake may alter ecosystem structure and function in fundamental ways, thereby causing changes in the production of a range of ecosystem goods and services. Thus, increasing the number of services to be economically valued necessarily increases the complexity of the valuation exercise and will likely increase the set of specialized skills and experience needed. Deriving a unified assessment of economic value requires integrating disciplinary skills. This integration becomes increas-
ingly difficult both on an intellectual level and on a practical level as the number of services is increased. The interconnection of ecosystem services may take place on a spatial or temporal scale, as well. As the Hadejia-Jama’are floodplain example illustrates, there are links between the provision of ecosystem services at upstream and downstream sites. Finally, it will often be the case that there are trade-offs among the production of different services. For example, reduced nutrient input into a lake may increase recreational values by decreasing algal blooms and turbidity, but it may also lower total fish productivity. Building a dam will change a section of free-flowing river into a lake, which may result in a decrease in the population of some fish species (e.g., salmon) and in opportunities for river recreation (e.g., canoeing, kayaking, whitewater rafting) while increasing populations of lake-adapted fish species and lake-based recreation (e.g., sailing, waterskiing). Trade-offs among ecosystem services increase the likelihood of sociopolitical debates because different groups are likely to place different relative values on different services. Natural variation, such as interannual differences in flood and drought frequencies and intensities, further complicates issues associated with reaching agreement on trade-offs among different ecosystem services. Although economic valuation of multiple ecosystem services is more difficult than valuation of a single ecosystem service, interconnections among services may make it necessary to expand the scope of the analysis.
This section reviews three cases that in some sense attempt to cover the economic value of all ecosystem services either for a single ecosystem or, more ambitiously, for the entire planet. The policy context of these three sets of studies is quite different. The first case study in this section reviews valuation studies done for the purpose of natural resource damage assessment for the Exxon Valdez oil spill. The second case, concerning the Florida Everglades, reviews studies that support what is probably the most expensive attempt at ecosystem restoration undertaken to date. The final case study by Costanza et al. (1997) represents the most ambitious attempt at valuation of ecosystem services to date. Its scope is nothing less than the value of ecosystem services for the entire planet (i.e., “the value of everything”).
Exxon Valdez Oil Spill
In March 1989, the Exxon Valdez oil tanker spilled 38,000 metric tons of crude oil (about one-fifth of its total cargo) into Prince William Sound in south-central Alaska. This accident inflicted large-scale environmental damage. Approximately 2,100 km of shoreline were impacted, with 300 km heavily or moderately impacted and 1,800 km lightly or very lightly oiled. Much of this coastline consists of gravel beaches into which oil penetrated to depths as great as one
meter. The carcasses of more than 35,000 birds and 1,000 sea otters were found after the spill, but this is considered to be a small fraction of the actual death toll since most carcasses sink. The best estimates are that the spill caused the deaths of 250,000 seabirds, 2,800 sea otters, 300 harbor seals, 250 bald eagles, up to 22 killer whales, and billions of salmon and herring eggs. While lingering injuries continue to plague some species, others appear to have recovered. Knowledge of the fate of the 38,000 metric tons of oil lost by the Exxon Valdez is imprecise; however, it is estimated that 30-40 percent evaporated, 10-25 percent was recovered, and the rest remained in the marine environment for some period of time (Shaw, 1992).
Following the accident, both private groups and governments sued Exxon for damages caused by the oil spill. Commercial fish interests pursued their own damages under federal and state law because they had a direct economic stake in the resource. Federal, state, and tribal governments serve as the legal trustees for public resources. The State of Alaska and the federal government sued for damages to public natural resources. Damage to public resources included lost recreational opportunities, diminished passive use values, and diminished use by Native peoples.
To prepare for possible trial in these cases, private parties, the State of Alaska, the federal government, and Exxon commissioned research bearing on the question of damages caused by the oil spill. Recognized researchers in a number of fields were recruited to undertake this research. The research was conducted for the purposes of litigation and took place in a highly charged atmosphere with billions of dollars of potential liability on the line. It was subject to intense scrutiny and generated heated debates over methods and results, particularly about validity and reliability of nonuse values estimated using contingent valuation methods. Although the State of Alaska and the federal government settled with Exxon over damages to public resources in 1991, debates about the validity and reliability of contingent valuation estimates of nonuse values raised by the affair continued. Some analysts extended these critiques to applications of contingent valuation to estimate use values. A conference sponsored by Exxon held in 1992 presented research papers that were quite critical of contingent valuation estimates of nonuse values (these papers were subsequently published in Hausman, 1993). In response to the ongoing controversy over the use of contingent valuation in natural resource damage assessment, the National Oceanic and Atmospheric Administration (NOAA) convened a blue-ribbon panel to assess the validity of contingent valuation applications to nonuse values, resulting in a widely cited NOAA panel report (NOAA, 1993).
Researchers used a variety of valuation techniques to assess the dollar value of damage from the Exxon Valdez oil spill to an array of public resources. Economic studies were conducted on recreational fishing losses (using a travel-cost model), impacts on tourism, replacement costs of birds and mammals, and a contingent valuation study of lost passive nonuse values. Studies of sportfishing activity and tourism indicators (i.e., vacation planning, visitor spending, canceled bookings) all indicated decreases in recreation and tourism activity. A
major study using contingent valuation was undertaken to estimate losses in (nonuse) values from the oil spill for people who did not visit or directly use the resources of Prince William Sound. There were also studies of lost value from commercial fishing. Commercial fishing losses, although part of the economic measure of damage to the ecosystem, were not part of the public resource injuries. Recreational fishing losses were counted as part of the public resource injuries.
Recreational fishing losses were estimated by two different teams, one representing Exxon and one representing the State of Alaska. Both teams used a random utility travel-cost model to estimate forgone use values but they arrived at estimates that differed by an order of magnitude. Hausman et al. (1992, 1993) estimated losses at $2.6 million to $3.2 million in the first year after the oil spill (1989) depending on the specific model used. This damage estimate would be expected to decline in future years as salmon stocks recovered from the spill. Carson and Hanemann (1992) estimated losses as high as $50 million per year. These differences occurred largely because Hausman et al. (1992, 1993) assumed 16,000 fewer recreational trips per year while Carson and Hanemann assumed 180,000 fewer trips. Hausman et al. (1992, 1993) also estimated lost recreational use values for hunting and hiking or viewing as well as a gain in recreational use value for pleasure boating (due to more trips taken to observe the aftermath of the spill). In total, they estimated “lost interim use values” due to the oil spill of $3.8 million in 1989.
An extensive contingent valuation study (Carson et al., 1994) estimated a loss of $2.8 billion in passive nonuse values by people who did not use or anticipated using Prince William Sound in the future. That estimate was derived from a national in-person survey that asked respondents about their willingness to pay to prevent the ecological harm of an oil spill of the magnitude of the Exxon Valdez. The survey found that median household willingness to pay to avoid similar injury to the marine ecosystem of the Prince William Sound region was $31 per household—which results in a value of $2.8 billion when summed across all households in the United States. However, it can be argued that this estimate was conservative and that the value of the ecological damage was far higher. For example, the persons surveyed were informed that ecological damages included 75,000 to 150,000 seabirds, 580 sea otters, and 100 harbor seals, compared to best estimates of 250,000 seabirds, 2,800 sea otters, and 300 harbor seals. Survey respondents were also told that no long-term damage would occur to the ecosystem and that wildlife populations would return to previous densities within three to five years. In addition, willingness to pay was used as the measure of damages, rather than willingness to accept (compensation) estimates, which typically are higher (Hanemann, 1991; see also Chapters 2 and 4). On the other hand, Hausman et al. (1993) were quite skeptical of estimates of nonuse values of several billion dollars when their estimate of use value was only several million dollars.
The replacement costs study identified a per-unit replacement cost of various seabirds and mammals, as well as eagles (Brown, 1992). For example, the
market price or the costs of relocating otters vary from $1,500 to $50,000 per otter. Replacement costs cannot be added to the public and private losses noted above, however, because these are expenditures to restore both the ecological services of the ecosystem and the aspects of these services enjoyed by humans (e.g., viewing wildlife and fishing).
A market model was used to evaluate private economic losses to commercial fisheries. Cohen (1995) estimated that the upper bound of the accident’s first-year social costs was $108 million. Second-year effects may have been as high as $47 million. Although estimates of economic losses to commercial fisheries are typically far less controversial than estimates of nonmarket values, there remain a number of sources of uncertainty. Cohen (1995) was not able to fully consider the numerous sources of variability inherent in the marine environment that may have contributed to harvest volume impacts but were provisionally attributed to the oil spill. In addition, efforts to distinguish effects of the oil spill on the value of harvest from other potential influences were hindered by inadequacies in economic data on supply responses of other U.S. commercial fisheries and the Japanese commercial fish market (Cohen, 1995). The analysis did not attempt to analyze economic harm to other components of south-central Alaska’s regional economy (e.g., fish processing and service sectors) or the extent to which the oil spill contributed to changes in the overall economic climate in south-central Alaska (Cohen, 1995).
Natural resource damage assessments require accurate assessment of the dollar value of damages to ecological resources. However, difficulties in understanding ecosystems, the production of services, and the values of those services are likely to lead to imprecise estimates. A precise determination of the damages caused by the Exxon Valdez oil spill is constrained by the dynamic interaction of numerous biological and economic variables (Cohen, 1995; Paine et al., 1996; Shaw, 1992). It is difficult to measure the full impact of the oil spill, to predict the time path of ecosystem recovery, and the extent of recovery that will ultimately occur. Furthermore, it is difficult to disentangle the effects of the oil spill from other environmental changes. Therefore, some unavoidable uncertainty will remain in attempts to quantify the link between the oil spill and changes in the provision of ecosystem services valued by humans. On top of this, valuing changes in the ecosystem involves both use values and passive nonuse values, the latter being notoriously difficult to estimate with much precision. However, even valuing damages to marketed commodities (e.g., the value of lost commercial fishing), where traditional uncontroversial market methods were used, proved difficult and a source of disagreement. Although studies of the value of ecosystem services can generate useful information, the degree of imprecision of the resulting estimates of values leaves plenty of room for arguments in court in natural resource damage assessment cases.
Restoration of the Florida Everglades
The Comprehensive Everglades Restoration Plan (CERP) is a framework (see also Box 3-6) and guide to restore the water resources of central and south Florida including the Everglades. This plan covers an area of 18,000 square miles and is predicted to take more than 30 years to implement. It is designed to regulate the quality, quantity, and distribution of water flows (CERP, 2001). The Florida Everglades ecosystem is one of the most endangered wetland complexes in the United States. More than one-half of the original marshes contained in this highly productive and diverse ecosystem have been drained. The remaining area is dissected by 2,253 kilometers of canals that transport water loaded heavily with nutrients from fertilizer and waste runoff from urban and agricultural lands. The Everglades provides habitat for 14 endangered or threatened species including the Florida panther (Felis concolor coryi), wood stork (Mycteria americana), and Florida Everglades snail kite (Rostrhamus sociabilis plumbeus).
The hydrologic connectivity (Pringle, 2003) between many different ecosystems within the Everglades makes quantifying the changes in ecosystem services due to restoration an extremely complex issue. The Everglades provide recharge water for aquifers across the state. Water flow through the Everglades also affects the salinity and biological integrity of connecting marine waters of Florida Bay. The effects of hydrologic alterations on these interconnected ecosystems are still subject to dispute. These and related issues have served as the basis of several previous National Research Council reports (e.g., NRC, 2002a,b). For example, the effectiveness of regional aquifer storage and recovery3 as a component of the CERP Plan is limited (NRC, 2002a). While aquifer storage and recovery have many advantages, disadvantages include low recharge and recovery rates relative to surface storage. Likewise, ecological impacts of altered hydrologic flow scenarios into Florida Bay also require more study (NRC, 2002b).
The Florida Everglades includes 4 national parks and preserves, 13 national wildlife refuges, 2 national marine sanctuaries, 17 state parks, 10 state aquatic preserves, and 5 wildlife management areas. Everglades National Park was created in 1947 to protect the approximately 20 percent of the remaining wetlands and is thus a vestige of the original Everglades ecosystem (which once included what is presently the Everglades Agricultural Area, the Water Conservation Area, and western portions of coastal urban areas). Large-scale drainage efforts over the last several decades have led to rapid agricultural, commercial, and residential growth (Englehardt, 1998) to the extent that native flora and fauna of the Everglades and adjacent interconnecting systems are imperiled. Efforts to
restore hydrologic function (i.e., flows) to the region are complicated by the magnitude and extent of human modification of the landscape.
Waters of the Kissimmee River flow south into Lake Okeechobee (the second-largest freshwater lake in the United States) and then into agricultural fields through an extensive system of flood control canals and reservoirs. Eventually the waters flow into the Everglades and into mangrove forests and estuaries on the Atlantic and Gulf Coasts. The Kissimmee was once a broad (1-2 miles wide), 103-mile-long river that meandered through an extensive network of floodplain wetlands (20,000 hectares). The ecosystem provided habitat for more than 300 fish and wildlife species, including resident and over-wintering waterfowl, a diverse wading bird community, and 13 game fish species. Channelization of the Kissimmee and drainage of approximately two-thirds of the floodplain wetlands were undertaken in the 1960s by the U.S. Army Corps of Engineers to improve flood protection and to provide drainage for agriculture. This has damaged the river-floodplain ecosystem, resulting in a 92 percent reduction in over-wintering waterfowl and negative effects on the native fish community (Englehardt, 1998). Moreover, agricultural drainage waters contain elevated phosphorus concentrations and have caused enrichment of Lake Okeechobee and the Everglades. Algal blooms have resulted in dramatic reductions in dissolved oxygen which has led to the death of many aquatic species; for example, nesting bird populations have decreased by 90 percent over the past 60 years.
One aspect of the CERP is to reestablish historic geomorphic and hydrologic conditions so that the Kissimmee River will once again be connected with its floodplain. This is being accomplished by back-filling the central portion of the dredged flood control canal (mainstem Kissimmee) and reestablishing side channels and backwaters (Toth, 1996). The restoration effort is also attempting to reduce phosphorus levels in the ecosystem by constructing stormwater treatment areas (large constructed wetlands). Other efforts to restore the Everglades include increasing water flows through the region, mimicking historic flow patterns, cleaning up polluted waters (e.g., Guardo et al., 1995), and purchasing private lands to protect them from development.
The economic valuation of restoration alternatives for the Everglades involves many challenges, primarily due to the complexity of the ecological systems (Davis and Ogden, 1994; Englehardt, 1998; Toth, 1996). Although restoration efforts promise to increase habitat for a wide variety of species, it is difficult to predict how different species will respond to changes in water quantity and quality. For example, ongoing restoration of the Everglades is dependent on numerous computer models to understand ecosystem processes, test alternatives, and evaluate restoration performance (Sklar et al., 2001). Landscape models used for restoration include hydrologic models, transition probability models, gradient models, distributional mosaic models, and individual-based models. When several landscape models are combined, they have the potential to contribute to water management and policymaking for Everglades restoration (Sklar et al. 2001); however, they have shortcomings based on their inherent assumptions and lack of important information. Although this is one of the most stud-
ied ecosystems in the world, much additional ecological knowledge is necessary (Kiker et al., 2001) to improve existing models and develop new ones. Curnutt et al. (2000) developed spatially-explicit species index models to predict how a number of species and species groups (e.g., cape seaside sparrow, snail kite, a species group model of long-legged wading birds) would respond to different hydrological restoration management alternatives. While no one scenario was beneficial to all species, the model allowed assessment of relative species responses to alternative water management scenarios.
Englehardt (1998) evaluated ecological benefits and impacts of proposed and alternative restoration plans in monetary terms. Current plans for restoration involve discharge of phosphorus-enriched water from artificial wetlands (stormwater treatment areas) to relatively pristine Everglades marshes for 3-10 years, risking conversion of the ecosystem to a eutrophic cattail marsh. Uncertain benefits and impacts were analyzed probabilistically, following principles of net present value analysis. This analysis indicated that alternative “bypass plans” would avoid the loss of up to 1,200 hectares of sawgrass marsh at a cost that is probabilistically justified by the value of the ecosystem preserved. This type of analysis can help clarify trade-offs but is complicated by the reality that restoration alternatives may have competing ecological benefits and losses over time. Again, there is also often a lack of scientific understanding and agreement (Englehardt, 1988).
Aillery et al. (2001) provide an analysis of trade-offs between restoration and agricultural economic returns to the Everglades Agricultural Area under alternative water retention targets. They developed a model linking economic and physical systems (including agricultural production, soil loss, and water retention). Effects of water retention scenarios, such as groundwater retention and surface water storage development, on production returns and agricultural resource use were estimated. Not surprisingly, the results suggest that small increases in water retention can be achieved with minimal losses in agricultural income, while agricultural returns decline more significantly with higher water retention targets.
To date there have been no attempts at a comprehensive economic valuation of the Everglades restoration efforts. Given the hydrological, ecological, and economic complexities of South Florida, a complete accounting of values is unlikely anytime in the near future. However, advances in our understanding of hydrological, ecological, and economic relationships could be of great help in guiding future restoration efforts. Such data can be useful in comparing the net benefits of alternative management policies even if an overall estimate of ecosystem values remains elusive.
The Value of Everything: Multiple Services in Multiple Ecosystems
In an ambitious and controversial paper, Costanza et al. (1997) attempted to estimate the total economic value of the services provided by all ecosystems on earth. The paper received a great deal of attention, not all of it favorable. A follow-up briefing article in Nature the following year stated that “The paper was a box-office success but was panned by the critics” (Nature, 1998).
In the paper, Costanza et al. (1997) estimated values for 17 ecosystem services4 from 16 ecosystem types including wetlands, forests, grasslands, estuaries, and other marine and terrestrial ecosystems. To derive estimates of the economic value of ecosystem services, Costanza et al. (1997) began with existing estimates of the productivity of a hectare for each ecosystem type for each service and a willingness to pay estimate for the service. Multiplying these estimates generated a per hectare value of the ecosystem service for each ecosystem type. They then aggregated across all services to establish a value per hectare for each ecosystem type. Finally, they multiplied this per-hectare value by the number of hectares of each ecosystem type and summed across ecosystem types to derive the total value of ecosystem services. For the bottom line, they estimated that the annual value of ecosystem services for the earth ranged from $16 trillion to $54 trillion, with a mean estimate of $33 trillion. This value was notably higher than the value of global GDP (gross domestic product) at the time ($18 trillion).
Critics have pointed out a number of serious flaws that lead to conclusions that the estimate has little scientific merit (e.g., Bockstael et al., 2000; Toman, 1998) while some attacked the approach as a meaningless exercise. If the question is the value of the life support system of the planet, there can be only one of two answers depending upon whether a willingness to pay or a willingness to accept approach is used. Willingness to pay should be bounded by global ability to pay (i.e., global GDP, or $18 trillion). If willingness to accept is used, then as Toman (1998) concludes, $33 trillion is “a serious underestimate of infinity.”
Other criticisms focused on problems with the methods and assumptions used in the paper. The paper itself has a long list of “sources of error, limitations and caveat” (Costanza et al., 1997). Obviously, there will be large data gaps in any such exercise. In addition, aggregation issues pose particular trouble in this study. According to Bockstael et al. (2000),
…Simple multiplication of a physical quantity by ‘unit value’ (derived from a case study that estimated the economic value for a specific resource) is a serious error. Small changes in an ecosystem’s services do not adequately characterize, with simple multipliers, the loss of a global ecosystem service.
Values estimated at one scale cannot be expanded by a convenient physical index of area, such as hectares, to another scale; nor can two separate value estimates, derived in different contexts, simply be added together.
A similar aggregation problem occurs in ecology, “A linear aggregation rule treats each change as if it could be made independent of the other constituent elements. In doing so, it assumes independence within and across the ecosystems being considered, and it ignores the possible effects of feedback cycles” (Bockstael et al., 2000). The approach used by Costanza et al. (1997) also assumes that ecosystem service production is “scale-free” in the sense that provision per unit area is constant no matter how big or small the ecosystem under consideration. Other papers (see also Chapter 3) have since stressed the importance of more focused analysis that matches the scale of analysis for ecosystem valuation to the scale of management questions (Balmford et al., 2002; Daily et al., 2000).
However, even some harsh critics of the paper have concluded that it served a useful role in getting more attention on the values of ecosystem services. One prominent economist said the paper was “a recklessly heroic attempt to do something futile” but that it was “very useful—it stirred things up a lot.” (Nature, 1998)
In one sense, attempting to economically value all ecosystem services can be viewed as the correct approach to take because it offers a complete accounting. It would certainly be advantageous to have evidence on all benefits and costs prior to decision-making because anything less will be partial and incomplete and risks giving incorrect advice to decision-makers. Yet trying to attain the “value of everything” through a complete and reliable accounting of all ecosystem services cannot be done with current understanding and methods and is unlikely to be accomplished anytime soon. Problems arise because knowledge of the translation from ecosystem function to ecosystem services is often incomplete as is the translation from services to values. For studies of the value of a single ecosystem service, and to some extent for studies of the value of multiple ecosystem services, attention can be directed toward services that are easier and relatively straightforward to value, such as the economic value of reducing the likelihood of flood damage or providing clean drinking water without filtration. In the case of the Exxon Valdez and the Florida Everglades restoration however, many of the important values are linked to the existence of species or the existence of the ecosystem itself in something akin to its original (pre-human-altered) condition. Valuing such services presents difficult challenges even when ecological knowledge is relatively complete. In addition, aggregation issues can cause problems in comprehensive approaches to ecosystem service valuation, particularly when scaling up the valuation exercise to cover multiple ecosystems.
IMPLICATIONS AND LESSONS LEARNED
This chapter has reviewed a number of applications of ecosystem valuation ranging from economic valuation of a single ecosystem service to attempts to value all services for an ecosystem and even for the entire planet. The valuation of ecosystem services is still relatively new and requires the integration of ecology and other natural sciences with economics. Such integration is not easy to accomplish. Still, examples of approaches and interdisciplinary studies that provide such integration indicate successful beginnings. Some of the lessons emerging from the case studies reviewed in the previous sections are discussed below.
Extent of Ecological and Economic Information for Valuing Ecosystem Services
As examples in this chapter have shown, the ability to generate useful information about the value of ecosystem services varies widely across cases. For some policy questions, enough is known about ecosystem service valuation to help in decision-making. A good example is the value of providing drinking water for New York City by protecting watersheds in the Catskills rather than building a more costly filtration system. As other examples make clear, knowledge and information may not yet be sufficient at present to estimate the value of ecosystem services with enough precision to answer policy-relevant questions.
The inability to generate sufficiently precise and reliable estimates of ecosystem values for purposes of informing decision-making may arise from any combination of the following three reasons: (1) there may be insufficient ecological knowledge or information to estimate the quantity of ecosystem services produced or to estimate how ecosystem service production would change under alternative scenarios; (2) existing economic methods may be unable to generate reliable and uncontroversial estimates of value for the provision of various levels of ecosystem services; and (3) there may be a lack of integration of ecological and economic analysis.
Much of the difficulty in generating reliable estimates of the value of ecosystem services derives from the fact that ecosystems are complex and dynamic and our understanding of them is typically incomplete or flawed. Learning how such ecosystems evolve and change as inputs to the system change can be a slow process (perhaps not even as fast at the system itself is changing). The example of the Everglades and the difficulty in designing a restoration plan aptly illustrate problems inherent in attempting to understand and manage aquatic ecosystems because the links from ecosystem condition and function to the production of goods and services may be hard to decipher. Other examples reviewed include fish production in coastal wetlands and salmon production in the Columbia River, where changes in ocean currents, flow of nutrient, water temperature,
precipitation patterns, disease prevalence, predator and prey populations, and other factors can impact fish populations. Although an increase in fish population from one year to the next could be related to a beneficial change in management strategy, it may also be due to changes in ocean conditions or other causes. In other cases, it is not necessary to understand the entire ecosystem in order to be able to estimate the production of an ecosystem service of interest with reasonable precision, such as the degree of flood control provided by wetlands. However, without adequate ecological understanding of ecosystem structure and function, it will not be possible to predict the level of some ecosystem services provided or the way provision levels may change under alternative management options.
Other difficulties arise because some ecosystem services are notoriously difficult to value. As stated previously, it is clear that people place value on such things as the continued existence of species, wilderness, beautiful scenery, and restoring ecosystems to a pre-human-altered condition. Ignoring such values, essentially assigning a value of zero to them, is clearly incorrect. What value should be assigned, however, is often far from clear and subject to debate. Estimating existence values and other nonconsumptive or nonuse values is among the most difficult challenges in environmental economics. For entire ecosystem valuation efforts, such as the Exxon Valdez case or the Everglades restoration, estimating such values cannot be avoided because they may account for a significant fraction of total economic value. Although the development and application of nonmarket valuation approaches have advanced significantly over the past two decades (see Chapter 4), there remains controversy, both within the economics profession and outside it, regarding the reliability of economic valuation methodologies (contingent valuation in particular) for environmental goods and services. For some ecosystem services such as valuing commercial fish harvests or the reduction of flood damage, the valuation exercise is more straightforward and uncontroversial. Difficulties may remain in knowing the level of services provided (e.g., how many fish are produced by coastal wetlands) or in obtaining relevant data (e.g., costs of fish harvesting), but there is relatively little disagreement about the utility of existing valuation methodology. One method, however, deserves particular mention and caution.
Using replacement or avoided cost to value an ecosystem service is justified under a restricted set of circumstances—namely, when there are alternative ways of providing the same service and the value of the service exceeds the cost of providing it, such as the provision of drinking water for New York City by increasing the protection of watersheds in the Catskills. However, this approach is sometimes applied when these conditions do not hold, thereby generating numbers that may bear no relation to the actual economic value of ecosystem services. For example, tallying up the large sum of money necessary to restore Prince William Sound to something close to its pre-spill condition does not necessarily imply that the economic value for services provided by the ecosystem is anywhere close to this cost.
Even when ecologists understand a system reasonably well and economists can apply widely accepted valuation methods, an effort at valuing ecosystem services may still fail if ecologists and economists fail to integrate their approaches. Unless the correct questions are asked at the outset, ecological information may not be of particular use for generating estimates of the production of ecosystem services in a useful form for economists to apply valuation methods. For their part, economists may apply valuation methodologies to cases that are not built on solid ecological grounding. It is important for ecologists and economists to talk at the outset of the valuation exercise to design a unified approach. Although it is easy enough to state or even recommend that ecologists and economists need to work together on integrated studies, accomplishing such integration is often difficult because of institutional constraints and reward structures that are largely disciplinary-based. Advances in interdisciplinary efforts may be risky or professionally unrewarding, especially for junior faculty members. It is important to overcome some of the institutional barriers that prevent ready and effective collaboration between ecologists and economists. Explicitly interdisciplinary programs, such as Dynamics of Coupled Natural and Human Systems as part of the Biocomplexity in the Environment Program5 at the National Science Foundation (NSF), represent a move in the right direction. Expanding “Schools of the Environment” at universities, where faculty from different disciplines interact routinely in addressing environmental issues, is another way to help overcome disciplinary barriers.
As discussed throughout this report, the adequacy of information in providing estimates of the economic value of ecosystem services that are policy relevant depends in large part on what policy question is asked. If the relevant policy question (or questions) can be answered by a relatively narrow evaluation of ecosystem services, the value of ecosystem services can likely be estimated with a relatively high degree of confidence with existing methods. For example, it is possible to answer questions about whether to conserve watersheds to provide clean water is worthwhile, as in the Catskills, or to conserve floodplains for flood control, as in the Salt Creek Greenway in Illinois. However, if the questions were reframed to identify the complete value of the conservation of watersheds or floodplains, there is insufficient information available on which to generate a reliable and credible answer. The issue of the effect of framing in terms of the policy context is also discussed in Chapters 2 and 6.
The NSF Dynamics of Coupled Natural and Human Systems Program emphasizes quantitative understanding of short- and long-term dynamics of natural capital, including how humans value and influence ecosystem services and natural resources, and considering uncertainty, resilience, and vulnerability in complex environmental systems. Further information is available on-line at http://www.nsf.gov/od/lpa/news/publicat/nsf0203/cross/pma.html.
Scope of Coverage, Spatial and Temporal Scale
Aquatic ecosystems produce a broad range of ecosystem services. Typically, however, ecological and economic information suitable for estimating reasonably precise values for ecosystem services exists for only a relatively narrow range of services. Lack of natural science (often ecological) information or understanding, or imprecision of valuation estimates for certain services, limits the ability to obtain precise estimates of economic value over the entire range of services provided by an ecosystem. In addition, there is considerable variation in ecosystem structure and function across space and time. As a consequence, the value of services from a particular ecosystem at a particular time may not necessarily be a good predictor of the economic value of services for other ecosystems or even the same ecosystem at a different time. Such ecosystem idiosyncrasies make benefits transfer problematic (see Chapter 4 for a discussion of benefits transfer). For these reasons, measures of the economic value of ecosystems services will continue to be partial and incomplete, at least for the foreseeable future. Some limit on the scope and scale of analysis is inevitable, but just where to set the boundaries for analysis is an important question.
The difficulty in obtaining estimates of economic value for the full range of ecosystem services presents analysts with a problematic trade-off. While relatively precise estimates of the value of ecosystem services may be derived for a fairly narrow set of services, an ecosystem valuation study that analyzes only a partial list of services may be insufficient for policy purposes. For example, suppose a proposed development would destroy a wetland. If relatively uncontroversial estimates of ecosystem service value such as flood reduction and increased fishery production do not exceed the value of development, it may be necessary to estimate values for a wider array of ecosystem services to inform the decision. However, when there are large uncertainties associated with estimates of value of these other ecosystem services, even collecting information on a wider set of ecosystem service values may not yield a clear recommendation about whether it is better to protect the wetland or allow development.
A second difficulty with limiting the scope of coverage of an ecosystem valuation study is the interconnection of processes within an ecosystem. Changing the inflow of nutrients into a lake will change ecosystem function and result in changes in fish productivity, recreational opportunities, and other ecosystem services. When there is a conflict between the provision of different ecosystem services—for example, hydroelectric power generation and fish production—the analysis should include the potentially conflicting ecosystem services if it is to be of use in policy decisions. Further, there may be cascading effects in which changes in one part of an ecosystem can ripple through the ecosystem, causing additional effects that may be difficult to foresee. For example, removal of a top predator may cause an increase in small predators, and changes in the herbivore prey base, with consequent changes in vegetation. It may be difficult to predict a priori how ecosystem functions and services will change when a predator is removed.
The preceding paragraphs strongly favor a more complete scope of coverage and a systems approach to valuing ecosystem services. However, expanding the scope of services covered by the analysis not only increases the workload and range of expertise necessary to design and conduct the analysis, but it will also likely to force analysts to estimate values for services whose production is poorly understood or for which valuation methods may generate imprecise estimates. There are no case studies that include a broad range of ecosystem services for which the value of these services can be estimated within a narrow range with much confidence.
In addition to questions about the scope of services studied, analysts will face difficult issues about the proper spatial and temporal scales. Spatial heterogeneity also limits the utility of benefits transfer, in which the estimates of value generated for one ecosystem are applied to other ecosystems. On the other hand, analyzing every ecosystem in detail can be prohibitively expensive and time consuming. In generating estimates of the economic value of ecosystem services across larger spatial scales, some method of extrapolation may be unavoidable, but such extrapolations bear careful scrutiny.
Interconnections in the production of ecosystem services across whatever spatial boundaries are chosen are virtually inevitable. A real danger of being too narrow in spatial scale is that important linkages in the production of ecosystem services or in the value of those services will be ignored. For example, focusing on upstream benefits from dams in the case of the Hadejia-Jama’are floodplain in northern Nigeria, while ignoring downstream losses, would give an incorrect assessment of the net benefits of dams and water diversions. Besides obvious physical interconnections, other types of interconnections may create important linkages in the production of ecosystem services. One mechanism that creates important interconnections across ecosystems occurs when multiple conditions contribute to the level of service provided. For example, protecting the summer habitat for neotropical migrant birds may be for naught if their winter habitat is destroyed. Protecting coastal wetlands in Louisiana as fish breeding grounds will be more or less valuable depending on the level of nitrogen export from Mississippi River drainage and the extent of the hypoxic zone. Another interconnection may occur with the existence of ecological thresholds and cumulative effects (as discussed in Chapter 3). Stress may be tolerated with little damage to an ecosystem service until a threshold is reached, at which point system function might change drastically, giving rise to a large change in ecosystem services. A classic example is the change in a shallow lake from oligotrophic to eutrophic conditions. A study of the consequences of increased nutrient export from a single stream into a lake may show that there is no change in economic value of the ecosystem services produced by the lake. However, the cumulative effects of increasing nutrient export from all streams into the lake could be sufficient to trigger a regime shift, causing a large change in the value of ecosystem services.
There may be interconnections between ecosystem services on the valuation side even when no biophysical connections exist between ecosystems. The
marginal value of an ecosystem service typically depends on the quantity of service supplied rather than being constant (e.g., demand curves generally slope downward). So, for example, a collapse in fish harvest in one ecosystem will tend to increase the economic value of fishery production from other ecosystems. In all valuation studies, some assumption must be made about the level of related ecosystem services produced elsewhere. In addition, the value of particular ecosystem services may also be a function of the level of provision of other ecosystem services or other human-produced services. In other words, there may be important complementarity or substitutability among services.
Most existing valuation techniques used by economists work well for valuing marginal changes but may be more problematic for valuing larger changes. Market price is an accurate signal of the marginal change in value for a small change in the quantity of a marketed good. However, to estimate the change in value from a nonmarginal change in quantity requires information about how price changes with quantity (i.e., the shape of the demand curve), information that may not be readily available. There are similar difficulties for nonmarketed services. For example, it is difficult obtain values for nonmarginal changes in hedonic studies (see Chapter 4). Changes in ecosystem structure and function, and hence in the provision of ecosystem services, however, may require nonmarginal valuation, such as with regime shifts (e.g., oligotrophic to eutrophic conditions in lakes) or large-scale disturbances. For nonmarginal changes, it is not valid simply to multiply the change in provision of the ecosystem service by an estimate of the marginal value of the service under current conditions to derive an estimate of the total change in economic value. Estimates of changes in total value must account for changes in marginal values as conditions change. Failure to take this fact into account can lead to serious errors—for example, in claiming that diamonds are of greater value than water, based on the fact that the price of diamonds (which are scarce) is high while the price of water (which is not scarce in many places) is low.
Because of biological or physical connections and the dependence of marginal value on conditions, great care must be exercised when estimates of value derived at one scale of analysis are applied at a different scale. Typically, there are no simple rules for aggregating values from small scales to larger scales. Some of the most pointed criticisms of the Costanza et al. (1997) study involved aggregation issues.
The temporal scale to be considered also presents challenges to the economic valuation of ecosystem services. Just as ignoring downstream effects in a spatial sense generates an incorrect assessment of net benefits, ignoring the future costs or benefits of decisions will result in an incorrect assessment of the present value of net benefits. For example, ignoring the loss of future benefits when stocks of groundwater are depleted or when the population of a commercially valuable species such as salmon declines will not provide adequate signals of the value of conserving such resources. The difficult issue of comparing present and future values arises when the consequences of a decision impact not only present but also future conditions. A common approach in economic stud-
ies is to discount future values. However, there is concern about discounting, especially for decisions having long-term consequences that will have repercussions for decades, centuries, or even longer (see Chapters 2 and 6 for further information). Assessing future consequences necessarily introduces uncertainty into the valuation of ecosystem services. Numerous events that affect ecosystems (e.g., disease outbreaks, fire patterns, weather) and human systems (e.g., innovation, changes in preferences, political change) cannot be predicted in advance. Knowing that ecosystem conditions may change or that values may shift places a premium on the ability to learn and adapt through time and to avoid outcomes with irreversible consequences (or consequences that can be reversed only at great expense). Adaptive management (see Chapter 6) and avoiding difficult-to-reverse decisions prior to reducing uncertainty arose in the context of managing salmon in the Columbia River basin.
The estimate of value of ecosystem services typically depends on a number of current conditions both in the ecosystem itself and in other interconnected systems, many of which are not explicitly stated. A change in fundamental underlying conditions, such as with climate change or an invasive species, may result in large changes in the estimated value of ecosystem services.
Finally, although there is great danger that studies will be partial and incomplete, as discussed in this section, there is also the possibility that the economic value of some ecosystem services will be counted more than once. When value is attributed to coastal wetlands as an input to fishery production, it cannot also be attributed to increased fishery production as an output. Unless studies are carefully designed and executed, such “double-counting” issues may arise.
SUMMARY: CONCLUSIONS AND RECOMMENDATIONS
This chapter has reviewed a series of case studies that value ecosystem services from aquatic and related terrestrial ecosystems, with a focus on their integration of ecology and economics. The case studies varied from those valuing a single ecosystem service, to multiple ecosystem services, to ambitious attempts to value all services from an ecosystem and even the entire planet. Many of the topics and issues addressed in this chapter directly respond to the committee’s statement of task (see Box ES-1). An extensive summary of implications and lessons learned from these reviews is provided in the previous section, and no attempt is made to resummarize that section here.
Based on the case studies reviewed in this chapter and the various implications and lessons learned, the committee makes the following specific conclusions regarding efforts to improve the valuation of ecosystem services:
Studies that focus on valuing a single ecosystem service show promise of delivering results that can inform important policy decisions. In no instance, however, should the value of a single ecosystem service be confused with the value of the entire ecosystem, which has far more than a single dimension.
Unless it is understood clearly that valuing a single ecosystem service represents only a partial valuation of the natural processes in an ecosystem, such single service valuation exercises may provide a false signal of total value.
Even when the goal of a valuation exercise is focused on a single ecosystem service, a workable understanding of the functioning of large parts or possibly the entire ecosystem may be required.
Although valuation of multiple ecosystem services is more difficult than valuation of a single ecosystem service, interconnections among services may make it necessary to expand the scope of the analysis.
Ecosystem processes are often spatially linked, especially in aquatic ecosystems. Full accounting of the consequences of actions on the value of ecosystem services requires understanding these spatial links and undertaking integrated studies at suitably large spatial scales to fully cover important effects. In generating estimates of the value of ecosystem services across larger spatial scales, extrapolation may be unavoidable but should be applied with careful scrutiny.
The value of ecosystem services depends on underlying conditions. Ecosystem valuation studies should clearly present assumptions about underlying ecosystem and market conditions and how estimates of value could change with changes in these underlying conditions.
Building on these preceding conclusions, the committee provides the following recommendations:
There is no perfect answer to questions about the proper scale and scope of analysis in ecosystem services valuation. Decisions about the scope and scale of analysis should be dictated by a clearly defined policy question.
Estimates of value should be placed in context. Assumptions about conditions in ecosystems outside the ecosystem of interest should be clearly specified. Assumptions about human behavior and institutions should be clearly specified.
Concerted efforts should be made to overcome existing institutional barriers that prevent ready and effective collaboration among ecologists and economists regarding the valuation of aquatic and related terrestrial ecosystem services. Furthermore, existing and future interdisciplinary programs aimed at integrated environmental analysis should be encouraged and supported.
Abbott, P.L. 1975. On the hydrology of the Edwards Limestone, south-central Texas. Journal of Hydrology 24(3/4):251-269.
Acharya, G., and E.B. Barbier. 2000. Valuing groundwater recharge through agricultural production in the Hadejia-Jama’are wetlands in Northern Nigeria. Agricultural Economics 22: 247-259.
Acharya, G., and E.B. Barbier. 2001. Using domestic water analysis to value groundwater recharge in the Hadejia’Jama’are floodplain in northern Nigeria. American Journal of Agricultural Economics 84(2):415-26.
Adams, R.M., R.P. Berrens, A. Cerda, H.W. Li, and P.C. Klingeman. 1993. Developing a bioeconomic model for riverine management: Case of the John Day River, Oregon. Rivers 4:213-226.
Aillery, M., M.R. Moore, M. Weinberg, G. Schaible, and N. Gollehan. 1999. Salmon recovery in the Columbia River basin: Analysis of measures affecting agriculture. Marine Resources Economics 14:15-40.
Aillery, M., R. Shoemaker, and M. Caswell. 2001. Agriculture and ecosystem restoration in South Florida. American Journal of Agricultural Economics 83(1):183-195.
American Rivers. 1997. Protecting wetlands along the Charles River. Available on-line at http://www.amrivers.org/floodcase.html#protecting. Accessed June 14, 2004.
Arnold, C.L. and C.J. Gibbons. 1996. Impervious surface coverage: The emergence of a key environmental indicator. American Planners Association Journal 62: 243-258.
Ashendorff, A., M.A. Principe, A. Seely, J. LaDuca, L. Beckhardt, W. Faber, and J. Mantus. 1997. Watershed protection for New York City’s supply. Journal of American Water Works Association 89(3):75-88.
Balmford, A., A. Bruner, P. Cooper, R. Costanza, S. Farber, R.E. Green, M. Jenkins, P. Jefferiss, V. Jessamy, J. Madden, K. Munro, N. Myers, S. Naeem, J. Paavola, M. Rayment, S. Rosendo, J. Roughgarden, K. Trumper, and R.K. Turner. 2002. Economic reasons for saving wild nature. Science 297:950-953.
Balvanera, P., G.C. Daily, P.R. Ehrlich, T.H. Ricketts, S.A. Bailey, S. Kark, C. Kremen, and H. Pereira. 2001. Conserving biodiversity and ecosystem services. Science 291:2047.
Barber, L.B. II. 1994. Sorption of chlorobenzenes to Cape Cod aquifer sediments. Environmental Science and Technology 28(5):890-897.
Barbier, E.B. 2000. Valuing the environment as input: Review of applications to mangrove-fishery linkages. Ecological Economics 35:47-61.
Barbier, E.B. 2003. Upstream dams and downstream water allocation: The case of the Hadejia-Jama’are Floodplain, Northern Nigeria. Water Resources Research 39(11):1311-1319.
Barbier, E.B. and I. Strand. 1998. Valuing mangrove-fishery linkages: A case study of Campeche, Mexico. Environmental and Resource Economics 12:151-166.
Barbier, E. B., and J.R. Thompson. 1998. The value of water: Floodplain versus large-scale irrigation benefits in northern Nigeria. Ambio 27:434-440.
Beamish, R.J., and C. Mahnken. 2001. A critical size and period hypothesis to explain natural regulation of salmon abundance and the linkage to climate and climate change. Progress in Oceanography 49:423-437SI.
Beamish, R.J., D.J. Noakes, G.A. McFarlane, L. Klyashtorin, V.V. Ivanov, and V. Kuraschov. 1999. The regime concept and natural trends in the production of Pacific salmon. Canadian Journal of Fisheries and Aquatic Sciences 56:516-526.
Beck, M., K.L. Heck, K.W. Able, D. L. Childers, D.B. Eggleston, B.M. Gillanders, B. Halpern, C.G. Hays, K. Hoshino, T.J. Minello, R.J. Orth, P.F. Sheridan, and M.P. Weinstein. 2001. The identification, conservation and management of estuarine and marine nurseries for fishes and invertebrates. BioScience 51:633-641.
Bell, F.W. 1989. Application of Wetland Valuation Theory to Florida Fisheries. Report No. 95, Florida Sea Grant Program. Tallahassee, Fla.: Florida State University.
Bell, F.W. 1997. The economic value of saltwater marsh supporting marine recreational fishing the southeastern United States. Ecological Economics 21:243-254.
Berrens, R., D. Brookshire, M. McKee and C. Schmidt. 1998. Implementing the safe minimum standard approach: Two case studies from the U.S. Endangered Species Act. Land Economics 74(2):147-161.
Beschta, R. 1997. Riparian shade and stream temperature: An alternative perspective. Rangelands 19(2):25-28.
Beschta, R., R.E. Bilby, G.W. Brown, L.B. Holtby, and T.D. Hofstra. 1987. Stream temperature and aquatic habitat: Fisheries and forest interactions. In Streamside Management: Forestry and Fishery Interactions E.O. Salo and T.W. Cundy (eds.). Seattle, Wash.: Institute of Forest Resources, University of Washington.
Bockstael, N.E., A.M. Freeman, R.J. Kopp, P.R. Portney, and V.K. Smith. 2000. On measuring economic values for nature. Environmental Science and Technology 34(8):1384-1389.
Boesch, D.F., and R.E. Turner. 1984. Dependence of fishery species on salt marshes: The role of food and refuge. Estuaries 7:460-68.
Bowles, D.E. and T.L. Arsuffi. 1993. Karst aquatic ecosystems of the Edwards Plateau region of Central Texas, USA—A consideration of their importance, threat to their existence, and efforts for their conservation. Aquatic Conservation 3:317-329.
Boyle, K.J., P.J. Poor, and L.O. Taylor. 1999. Estimating the demand for protecting freshwater lakes from eutrophication. American Journal of Agricultural Economics 81:1118-1122.
Brock, T.D. 1985. A Eutrophic Lake: Lake Mendota, Wisconsin. New York: Springer-Verlag.
Brock, W.A., and A. deZeeuw. 2002. The repeated lake game. Economics Letters 76:109-114.
Brown, G., Jr. 1992. Replacement Costs of Birds and Mammals. Report for the State of Alaska. Available on-line at http://www.evostc.state.ak.us/pdf/econ4.pdf. Accessed January 14, 2004.
Budy, P., G.P. Thiede, N. Bouwes, G.E. Petrosky, and H. Schaller. 2002. Evidence linking delayed mortality of Snake River salmon to their earlier hydrosystem experience. North American Journal of Fisheries Management 22:35-51.
Burt, O. 1964. Optimal resource use over time with an application to groundwater. Management Science 11:80-93.
Cameron, T.A., W.D. Shaw, S.E. Ragland, J.M. Callaway, and S. Keefe. 1996. Using actual and contingent behavior data with different levels of time aggregation to model recreational demand. Journal of Agricultural and Resource Economics 21(10):130-149.
Carpenter, S.R., D. Ludwig, and W.A. Brock. 1999. Management and eutrophication for lakes subject to potentially irreversible change. Ecological Applications 9:751-771.
Carson, R.T. and W.M. Hanemann. 1992. A Preliminary Economic Evaluation of Recreational Fishing Losses Related to the Exxon Valdez Oil Spill. A report to the Attorney General of the State of Alaska. Available on-line at http://www.evostc.state.ak.us/pdf/econ1.pdf. Accessed October 6, 2004.
Carson, R.T., R.C. Mitchell, W.M. Hanemann, R.J. Kopp, S. Presser, and P.A. Ruud. 1994. Contingent valuation and lost passive use: Damages from the Exxon Valdez. Discussion Paper 94-18. Washington, D.C.: Resources for the Future.
CERP (Comprehensive Everglades Restoration Plan). 2001. Project Management Plan: Florida Bay and Florida Keys Feasibility Study. Available on-line at http://www.evergladesplan.org/pm/program/program_docs/pmp_study_florida/cerp_fb_fk.pdf. Accessed June 14, 2004.
Chen, C.C., D. Gillig, and B.A. McCarl. 2001. Effects of climatic change on a water dependent regional economy: A study of the Texas Edwards Aquifer. Climatic Change 49:397-409.
Chichilnisky, G., and G. Heal. 1998. Economic returns from the biosphere. Nature 391:629-630.
Cohen, M.J. 1995. Technological disasters and natural resource damage assessment: An evaluation of the Exxon Valdez oil spill. Land Economics 71:65-82.
Congleton, J.L., W.J. LaVoie, C.B. Schreck, and L.E. Davis. 2000. Stress indices in migrating juvenile Chinook salmon and steelhead of wild and hatchery origin before and after barge transportation. Transactions of the American Fisheries Society 129:946-961.
Costanza, R., R. d’Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R.V. O’Neil, J. Paruelo, R.G. Raskin, P. Sutton, and M. van den Belt. 1997. The value of the world’s ecosystem services and natural capital. Nature 387:253-260.
Costello, C., R. Adams, and S. Polasky. 1998. The value of El Niño forecasts in the management of salmon: A stochastic dynamic assessment. American Journal of Agricultural Economics 80:765-777.
Covich, A.P. 1993. Water and ecosystems. Pp. 40-50 in Water in Crisis: A Guide to the World’s Fresh Water Resources, P.H. Gleick (ed.). Oxford, U.K.: Oxford University Press.
Covich, A.P., K.C. Ewel, R.O. Hall, P.G. Giller, D. Merritt, and W. Goedkoop. 2004. Ecosystem services provided by freshwater benthos. Pp.45-72 in Sustaining Biodiversity and Ecosystem Services in Soils and Sediments, D. Wall (ed.). Washington, D.C.: Island Press.
Covich, A.P., M.A. Palmer, and T.A. Crowl. 1999. The role of benthic invertebrate species in freshwater ecosystems. BioScience 49:119-127.
Culver, D.C., L.L. Master, M.C. Christman, and H.H. Hobbs. 2000. Obligate cave fauna of the 48 contiguous United States. Conservation Biology 14:386-401.
Culver, D.C., M.C. Christman,W.R. Elliott, H.H. Hobbs, and J.R. Reddell. 2003. The North American obligate cave fauna: Regional patterns. Biodviersity and Conservation 12: 441-468.
Curnutt, J.L., J. Comiskey, M.P. Nott, and L.J. Gross. 2000. Landscape-based spatially explicit species index models for everglades restoration. Ecological Applications 10(6):1849-1860.
Daily, G.C. (ed.) 1997. Nature’s Services: Societal Dependence on Natural Ecosystems. Washington, D.C.: Island Press.
Daily, G.C., T. Soderqvist, S. Aniyar, K. Arrow, P. Dasgupta, P.R. Ehrlich, C. Folke, A. Jansson, B.O. Jansson, N. Kautsky, S. Levin, J. Lubchenco, K.G. Maler, D. Simpson, D. Starrett, D. Tilman, and B. Walker. 2000. Ecology—The value of nature and the nature of value. Science 289:395-396.
D’Arge, R.C., and J. Shogren. 1989. Okoboji experiment: Comparing non-market valuation techniques in an unusually well-defined market for water quality. Ecological Economics 1:251-259.
Davis, S., and J. Ogden. 1994. Everglades: The Ecosystem and Its Restoration. Delray Beach, Fla.: St Lucie Press.
DeMelo, R., R. Fraqnce, and D.J. McQueen. 1992. Biomanipulation: Hit or myth? Limnology and Oceanography 37:192-207.
Deriso, R.B., D.R. Marmorek, and I.J. Parnell. 2001. Retrospective patterns of differential mortality and common year-effects experienced by spring and summer Chinook
salmon (Oncorhunchus tshawytscha) of the Columbia River. Canadian Journal of Fisheries and Aquatic Science 58:2419-2430.
Ellis, G.M., and A.C. Fisher. 1987. Valuing the environment as input. Journal of Environmental Management 25:149-156.
Englehardt, J.D. 1998. Ecological and economic risk analysis of Everglades: Phase I restoration alternatives. Risk Analysis 18:755-771.
EPA (U.S. Environmental Protection Agency). 1999. Safe Drinking Water Act, Section 1429, Ground Water Report to Congress. EPA-816-R-99-016. Available on-line at http://www.epa.gov/safewater/gwr/finalgw.pdf. Accessed October 20, 2004.
Faber, S. 1996. On Borrowed Land: Public Policies for Floodplains. Cambridge, Mass.: Lincoln Institute of Land Policy.
Farber, S., and R. Costanza. 1987. The economic value of wetlands systems. Journal of Environmental Management 24:41-51.
Fisher, A.C., W.M. Hanemann, and A. Keeler. 1991. Integrating fishery and water resource management: A biological model of a California salmon fishery. Journal of Environmental Economics and Management 20:234-261.
Foran, J., T. Brosnan, M. Connor, J. Delfino, J. DePinto, K. Dickson, H. Humphrey, V. Novotny, R. Smith, M. Sobsey, and S. Stehman. 2000. A framework for comprehensive, integrated, water monitoring in New York City. Environmental Monitoring and Assessment 62:147-167.
Forest Preserve District of Cook County Illinois. 1988. An evaluation of floodwater storage. River Forest, Ill.: Forest Preserve District of Cook County.
Freeman, A.M. 1991. Valuing environmental resources under alternative management regimes. Ecological Economics 3:247-256.
Frei, A., R.L. Armstrong, M.P. Clark, and M.C. Serreze. 2002. Catskill Mountain water resources: Vulnerability, hydroclimatology, and climate-change sensitivity. Annals of the Association of American Geographers 92:203-224.
Gergel, S.E., M.G. Turner, J.R. Miller, J.M. Melack, and E.H. Stanley. 2002. Landscape indicators of human impacts to riverine systems. Aquatic Sciences 64:118-128.
Gibert, J., D.L. Danielopol, and J.A. Stanford (eds.). 1994. Groundwater ecology. San Diego, Calif.: Academic Press.
Giller, P.S., A.P. Covich, K.C. Ewel, R.O. Hall, Jr., and D.M. Merritt. 2004. Vulnerability and management of ecological services in freshwater systems. Pp. 137-159 in Sustaining Biodiversity and Ecosystem Services in Soils and Sediments, D. Wall (ed.). Washington, D.C.: Island Press.
Glennon, R. 2002. Water Follies: Groundwater Pumping and the Fate of America’s Fresh Waters. Washington, D.C.: Island Press.
Grant, G. 2001. Dam removal: Panacea or pandora for rivers? Hydrological Processes 15:1531-1532.
Gregory, S., H. Li, and J. Li. 2002. The conceptual basis for ecological responses to dam removal. BioScience 52:713-723.
Grimm, N.B., A. Chacon, C.N. Dahm, S.W. Hostetler, O.T. Lind, P.L. Starkweather, and W.W. Wurtsbaugh. 1997. Sensitivity of aquatic ecosystems to climatic and anthropogenic changes: The basin and range, American Southwest, and Mexico. Hydrological Processes 11:1023-1041.
Guardo, M., L. Fink, T.D. Fontaine, S. Newman, M. Chimney, R. Bearzotti, and G. Goforth. 1995. Large-scale constructed wetlands for nutrient removal from stormwater runoff: An Everglades restoration project. Environmental Management 19(6): 879-889.
Gulati, R.D., E.H.R.R. Lammens, M.L. Meijer, and E. Donk (eds.). 1990. Biomanipulation-Tool for Water Management. Belgium: Kluwer Academic Publishers.
Hamlet, A.F., and D.P. Lettenmaier. 1999a. Columbia River streamflow forecasting based on ENSO and PDO climate signals. Journal of Water Resources Planning and Management 125:333-341.
Hamlet, A.F., and D.P. Lettenmaier. 1999b. Effects of climate change on hydrology and water resources in the Columbia River Basin. Journal of the American Water Resources Association 35:1597-1623.
Hamlet, A.F., D. Huppert, and D.P. Lettenmaier. 2002. Economic value of long-lead streamflow forecasts for Columbia River hydropower. Journal of Water Resources Planning and Management 128:91-101.
Hammack, J., and G.M. Brown, Jr. 1974. Waterfowl and Wetlands: Toward Bioeconomic Analysis. Baltimore, Md.: Johns Hopkins University Press-Resources for the Future.
Hanemann, W.M. 1991. Willingness to pay and willingness to accept: How much can they differ? American Economic Review 18:635-647.
Hare, S.R., N.J. Mantua, and R.C. Francis. 1999. Inverse production regimes: Alaska and West Coast Pacific salmon. Fisheries 24:6-14.
Hausman, J.A. (ed.). 1993. Contingent Valuation: A Critical Assessment. Amsterdam: North-Holland.
Hausman, J.A., G.K. Leonard, and D. McFadden. 1995. A utility-consistent combined discrete choice and count data model: Assessing recreational use losses due to natural resource injury. Journal of Public Economics 56:130.
Hausman, J.A., G.K. Leonard, and D. McFadden. 1993. Assessing use value losses caused by natural resource injury. Pp. 341-363 in Contingent Valuation: A Critical Assessment, J.A. Hausman (ed.). Amsterdam: North-Holland.
Heal, G.M. 2000a. Valuing ecosystem services. Ecosystems 3:24-30.
Heal, G.M. 2000b. Nature and The Marketplace: Capturing the Value of Ecosystem Services. Washington, D.C.: Island Press.
Hey, D.L., and N.S. Philippi. 1995. Proceedings of the scientific assessment and strategy team workshop on hydrology, floodplain ecology and hydraulics. Pp. 47-52 in Science for Floodplain Management into the 21st Century, G.E. Freeman, and A.G. Frazier (eds.). Volume 5. Washington, D.C.: Government Printing Office.
Hicks, B.J., R.L. Beschta, and R.D. Hart. 1991. Long-term changes in streamflow following logging in Western Oregon and associated fishery implications. Water Resources Bulletin 27:217-226.
Holmes, T.P., J.C. Bergstrom, E. Huszar, S.B. Kask, and F. Orr. 2004. Contingent valuation, net marginal benefits, and the scale of riparian ecosystem restoration. Ecological Economics 49:19-30.
Howe, C.W. 2002. Policy issues and institutional impediments in the management of groundwater: Lessons from case studies. Environment and Development Economics 7:625-641.
Huppert, D. 1989. Measuring the value of fish to anglers: Application to central California anadromous species. Marine Resource Economics 6:89-107.
Hurd, B., N. Leary, R. Jones, and J. Smith. 1999. Relative regional vulnerability of water resources to climate change. Journal of the American Water Resources Association 35:1399-1409.
Illinois Department of Conservation. 1993. The Salt Creek Greenway Plan. Springfield, Ill.: Illinois Department of Conservation.
Isard, W., K. Bassett, C. Chogull, J. Furtado, R. Izumita, J. Kissin, E. Romanoff, R. Seyfarth, and R. Tatlock. 1969. Linkage of socio-economic and ecologic systems. Ekistics 28:28-34.
Jaeger, W.K., and R. Mikesell. 2002. Increasing streamflow to sustain salmon and other native fish in the Pacific Northwest. Contemporary Economic Policy 20:366-380.
Johnson, N.S., and R.M. Adams. 1988. Benefits of increased streamflow: The case of the John Day River steelhead fishery. Water Resources Research 24(11):1839-1846.
Jones, J.B. and P.J. Mulholland (eds.). 2000. Stream and Ground Waters. San Diego, Calif.: Academic Press.
Kareiva, P., M. Marvier, and M. McClure. 2000. Recovery and management options for spring/summer chinook salmon in the Columbia River basin. Science 290: 977-979.
Kiker, C.F., J.W. Milon, and A.W. Hodges. 2001. Adaptive learning for science-based policy: The Everglades restoration. Ecological Economics 37(3):403-416.
Kinzig, A. (and 50 co-authors). 2000. Nature and Society: An Imperative for Integrated Environmental Research. A report from an NSF sponsored workshop held in June 2000. Available on-line at http://lsweb.la.asu.edu/akinzig/NSFReport.pdf. Accessed December 10, 2003.
Kitchell, J.F. (ed.). 1992. Food Web Management. A Case Study of Lake Mendota. New York: Springer-Verlag.
Krutilla, J.V. 1967. Conservation reconsidered. American Economic Review 57:777-786
Krutilla, J.V., and A.C. Fisher. 1975. The Economics of Natural Environments: Studies in the Valuation of Commodity and Amenity Resources. Baltimore, Md.: Johns Hopkins University Press-Resources for the Future.
Lathrop, R.C., S.R. Carpenter, C.A. Stow, P.A. Soranno, and J.C. Panuska. 1998. Phosphorus loading reductions needed to control blue-green algal blooms in Lake Mendota. Canadian Journal of Fisheries and Aquatic Sciences 55:1169-1178.
Lathrop, R.C., B.M. Johnson, T.B. Johnson, M.T. Vogelsang, S.R. Carpenter, T.R. Hrabik, J.F. Kitchell, J.J. Magnuson, L.G. Rudstam, and R.S. Stewart. 2002. Stocking piscivores to improve fishing and water clarity: A synthesis of the Lake Mendota biomanipulation project. Freshwater Biology 47:2410-2424.
Layman, C., J. Boyce, and K. Criddle. 1996. Economic evaluation of chinook salmon sport fishery of the Gulkana River, Alaska, under current and alternative management plans. Land Economics 72:113-128.
Lee, K.N. 1993. Compass and Gyroscope: Integrating Science and Politics for the Environment. Covelo, Calif.: Island Press.
Lee, K.N. 1999. Appraising adaptive management. Conservation Ecology 3(2):3. Available on-line at http://www.consecol.org/vol3/iss2/art3. Accessed December 10, 2003.
Levin, P.S., and N. Tolimieri. 2001. Differences in the impacts of dams on the dynamics of salmon populations. Animal Conservation 4:291-299.
Levin, P.S., R.W. Zabel, and J.G. Williams. 2001. The road to extinction is paved with good intentions: Negative association of fish hatcheries with threatened salmon. Proceedings of the Royal Society of London Series B-Biological Sciences 268:1153-1158.
Lichatowich, J. 1999. Salmon Without Rivers. Washington, D.C.: Island Press.
Longley, G. 1986. The biota of the Edwards Aquifer and the implications for paleozoogeography. In The Balcones Escarpment, P.L. Abbott and C.M. Woodruff, Jr. (eds.). Available on-line at http://www.lib.utexas.edu/geo/BalconesEscarpment/BalconesEscarpment.html.
Loomis, J.B. 1988. The bioeconomic effects of timber harvesting on recreational and commercial salmon and steelhead fishing: A case study of the Siuslaw National Forest. Marine Resource Economics 5:43-60.
Loomis, J., P. Kent, L. Strange, K. Fausch, A. Covich. 2000. Measuring the total economic value of restoring ecosystem services in an impaired river basin: Results from a contingent valuation survey. Ecological Economics 33:103-117.
Lovett, G.M., K.C. Weathers, and W.V. Sobczak. 2000. Nitrogen saturation and retention in forested watersheds of the Catskill Mountains, New York. Ecological Applications 10:73-84.
Lynne, G.D., P. Conroy, and F.J. Prochaska. 1981. Economic value of marsh areas for marine production processes. Journal of Environmental Economics and Management 8:175-186.
McCarl, B.A., C.R. Dillon, K.O. Keplinger, and R.L. Williams. 1999. Limiting pumping form the Edwards Aquifer: An economic investigation of proposal, water markets and spring flow guarantees. Water Resources Research 35:1257-1268.
McCully, P. 2002. Silenced Rivers: The Ecology and Politics of Large Dams. 2nd edition. London: Zed.
Meffe, G.K. 1992. Techno-arrogance and halfway technologies: Salmon hatcheries on the Pacific coast of North America. Conservation Biology 6:350-354.
Merrifield, J. 2000. Goundwater resources: The transition from capture to allocation. Policy Studies Review 17(1):105-124.
Merrifield, J., and R. Collinge. 1999. Efficient water pricing policies as an appropriate municipal revenue source. Public Works Management and Policy 4(2):119-130.
Meyer, J.L. and J.B. Wallace. 2001. Lost linkages in lotic ecology: Rediscovering small streams. Pp. 295-317 in Ecology: Achievement and Challenge, M. Press, N. Huntly, S. Levin (eds.). Boston, Mass.: Blackwell Science.
Meyer, J.L., M.J. Sale, P.J. Mulholland, and N.L. Poff. 1999. Impacts of climate change on aquatic ecosystem functioning and health. Journal of the American Water Resources Association 35:1373-1386.
Miles, E.L., A.K. Snover, A.F. Hamlet, B. Callahan, and D. Fluharty. 2000. Pacific Northwest regional assessment: The impacts of climate variability and climate change on the water resources of the Columbia River Basin. Journal of the American Water Resources Association 36:399-420.
Moore, M.R., N.R. Gollehon, and M.B. Carey. 1994. Multicrop production decisions in western irrigated agriculture—The role of water price. American Journal of Agricultural Economics 76:859-874.
Moore, M.R., N.R. Gollehon, and D.M. Hellerstein. 2000. Estimating producer’s surplus with the censored regression model: An application to producers affected by Columbia River Basin salmon recovery. Journal of Agricultural and Resource Economics 25:325-346.
Morganwalp, D.W., and H.T. Buxton (eds.). 1999. U.S. Geological Survey Toxic Substances Hydrology Program, March 8. Volume 3—Subsurface Contamination from Point Sources. USGS Water-Resources Investigations Report 99-4018C.
Moscript, A. L., and D. R. Montgomery. 1997. Urbanization, flood frequency, and salmon abundance in Puget Lowland streams. Journal of the American Water Resources Association 33:1289-1297.
Murdoch, P.S., and J.L. Stoddard. 1992. The role of nitrate in the acidification of streams in the Catskill Mountains of New York. Water Resources Research 28:2707-2720.
Murdoch, P.S., and J.L. Stoddard. 1993. Chemical characteristics and temporal trends in eight streams of the Catskills Mountains, New York. Water, Air, and Soil Pollution 67:257-280.
Murdoch, P.S., J.S. Baron, and T.L. Miller. 2000. Potential effects of climate change on surface-water quality in North America. Journal of the American Water Resources Association 36:347-366.
Naiman, R.J., S.E. Bunn, C. Nilsson, G.E. Petts, G. Pinay, and L. Thompson. 2002. Legitimizing fluvial ecosystems as users of water: An overview. Environmental Management 30:455-467.
Nature. 1998. Audacious bid to value the planet whips up a storm. 395:430.
Nickelson, T.E. 1986. Influence of upwelling, ocean temperature, and smolt abundance on marine survival of Coho Salmon in the Oregon production area. Canadian Journal of Fisheries and Aquatic Sciences 43:527-535.
NOAA (NOAA Panel on Contingent Valuation). 1993. Natural Resource Damage Assessment under the Oil Pollution Act of 1990. Federal Register 58(10):4601-4614.
Northwest Power Planning Council. 2001. Inaugural Annual Report of the Columbia Basin Fish and Wildlife Program, 1978-1999. Northwest Power Planning Council Document 2001-2. Portland, Ore.: Northwest Power Planning Council.
NRC (National Research Council). 1995. Understanding Marine Diversity: A Research Agenda for the Nation. Washington, D.C.: National Academy Press.
NRC. 1996. Upstream: Salmon and Society in the Pacific Northwest. Washington, D.C.: National Academy Press
NRC. 1997. Valuing Groundwater: Economic Concepts and Approaches. Washington, D.C.: National Academy Press
NRC. 2000a. Watershed Management for Potable Water Supply: Assessing the New York City Strategy. Washington, D.C.: National Academy Press.
NRC. 2000b. Investigating Groundwater Systems on Regional and National Scales. Washington, D.C.: National Academy Press.
NRC. 2002a. Regional Issues in Aquifer Storage and Recovery for Everglades Restoration. Washington, D.C.: National Academy Press.
NRC. 2002b. Florida Bay Research Programs and Their relation to the Comprehensive Everglades Restoration Plan. Washington, D.C.: National Academy Press.
NRC. 2004. Managing the Columbia River: Instream Flows, Water Withdrawals, and Salmon Survival. Washington, D.C.: The National Academies Press.
Olsen, D., J. Richards, and R.D. Scott. 1991. Existence and sport values for doubling the size of the Columbia River basin salmon and steelhead runs. Rivers 2(1):44-56.
Opie, J. 1993. Ogallala: Water for a Dry Land. Lincoln, Neb.: University of Nebraska Press.
Paine, R.T., J.L. Ruesink, A. Sun, E.L. Soulanille, M.J. Wonham, C.D.G. Harley, D.R. Brumbaugh and D.L. Secord. 1996. Trouble on oiled waters: Lessons from the Exxon Valdez oil spill. Annual Review of Ecology and Systematics 27:197-235.
Paul, M.J., and J.L. Meyer. 2001. Streams in the urban landscape. Annual Review of Ecology and Systematics 32:333-365.
Paulsen, C.M., and R.A. Hinrichsen. 2002. Experimental management for Snake River spring-summer chinook (Oncorhynchus tshawytscha): Trade-offs between conservation and learning for a threatened species. Canadian Journal of Fisheries and Aquatic Sciences 59:717-725.
Paulsen, C.M., and K. Wernstedt. 1995. Cost-effectiveness analysis for complex managed hydrosystems: An application to the Columbia River Basin. Journal of Environmental Economics and Management 28:388-400.
Pernin, C.G., M.A. Bernstein, A. Mejia, H. Shih, F. Reuter, and W. Steger. 2002. Generating Electric Power in the Pacific Northwest: Implications of Alternative Technologies. Santa Monica, Calif.: RAND.
Platts, W.S. 1991. Livestock grazing. In Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitat. American Fisheries Society Special Publiaction 19: 389-423.
Poff, N.L., and D.D. Hart. 2002. How dams vary and why it matters for the emerging science of dam removal. BioScience 52:659-668.
Polasky, S. (ed.). 2002. The Economics of Biodiversity Conservation. Aldershot, Hampshire, U.K.: Ashgate Publishing Limited.
Pringle, C.M. 2002. Hydrologic connectivity and the management of biological reserves: A global perspective. Ecological Applications 11:981-998.
Pringle, C.M. 2003. What is hydrologic connectivity and why is it ecologically important? Hydrological Processes 17:2685-2689.
Pringle, C.M., M. Freeman, and B. Freeman. 2000. Regional effects of hydrologic alterations on riverine macrobiota in the New World: Tropical-temperate comparisons. BioScience 50:807-823.
Pritchard, L., Jr., C. Folke, and L. Gunderson. 2000. Valuation of ecosystem services in institutional context. Ecosystems 3:36-40.
Provencher, B. 1993. A private property rights regime to replenish a groundwater aquifer. Land Economics 69(4):325-340.
Provencher, B., and O. Burt. 1994. A private property rights regime for the commons: The case for groundwater. American Journal of Agricultural Economics 76(4):875-888.
Pulwarty, R.S., and K.T. Redmond. 1997. Climate and salmon restoration in the Columbia River Basin: The role and usability of seasonal forecasts. Bulletin of the American Meteorological Society 78:381-397.
Pyne, R.D.G. 1995. Groundwater Recharge and Wells: A Guide to Aquifer Storage Recovery. Boca Raton, Fla.: Lewis Publishers.
Reed-Anderson, T., S.R. Carpenter, and R.C. Lathrop. 2000. Phosphorus flow in a watershed-lake ecosystem. Ecosystems 3:561-573.
Rishel, G.B., J.A. Lynch, and E.S. Corbett. 1982. Seasonal stream temperature changes following forestry harvest. Journal of Environmental Quality 11:112-116.
Rubio, S.J., and B. Casino. 1994. Competitive versus efficient extraction of a common property resource: The groundwater case. Journal of Economic Dynamics and Control 25:1117-1137.
Sanford, W. 2002. Recharge and groundwater models: An overview. Hydrogeology Journal 10:110-120.
Schaller, H.A., C.E. Petroky, and O.P. Langness. 1999. Constrasting patterns of productivity and survival rates for stream-type chinook salmon (Oncorhynchus tsawytscha) populations of the Snake and Columbia Rivers. Canadian Journal of Fisheries and Aquatic Sciences 56:1031-1045.
Scheaffer, J.R., J.D. Mullan, and N.B. Hinch. 2002. Encouraging wise use of floodplains with market-based incentives. Environment (January-February):33-43.
Schiable, G.D., B.A. McCarl, and R.D. Lacewell. 1999. The Edwards Aquifer water resource conflict: USDA farm programs resource use incentives. Water Resources Research 35: 3171-3183.
Schneiderman, J.S. 2000. From the Catskills to Canal Street: New York City’s water supply. Pp. 166-180 in The Earth Around Us, J.S. Schneiderman (ed.). Boulder, Colo.: Westview Press.
Schoenen, D. 2002. Role of disinfection in suppressing the spread of pathogens with drinking water: Possibilities and limitations. Water Research 3:3874-3888.
Shapiro, J. 1990. Biomanipulation: The next phase—Making it stable. Pp. 13-27 in Biomanipulation-Tool for Water Management, Gulati, R.D., E.H.R.R. Lammens, M.L. Meijer, and E. Donk (eds.). Belgium: Kluwer Academic Publishers.
Sharp, J.M, Jr., and J.L. Banner. 2000. The Edwards Aquifer: Water for thirsty Texans. Pp. 154-165 in The Earth Around Usk, J.S. Schneiderman (ed.). Boulder, Colo.: Westview Press.
Shaw, D.G. 1992. The Exxon Valdez oil-spill: Ecological and social consequences. Environmental Conservation 19:253-258.
Sklar, F.H., H.C. Fitz, Y. Wu, R. Van Zee, and C. McVoy. 2001. The design of ecological landscape models for Everglades restoration. Ecological Economics 37(3):379-401.
Stoddard, J.L. 1994. Long-term changes in watershed retention of nitrogen. Pp. 223-284 in Environmental Chemistry of Lakes and Reservoirs, L.A. Baker (ed.). Washington, D.C.: American Chemical Society.
Stroud Water Research Center. 2001. Water Quality Monitoring in the Source Water Areas for New York City: An Integrative Watershed Approach. Contribution No. 2001007. Avondale, Pa..
Stumborg, B.E., K.A. Baerenklau, and R.C. Bishop. 2001. Nonpoint source pollution and present values: A contingent valuation study of Lake Mendota. Review of Agricultural Economics 23:120-132.
Swallow, S.K. 1994. Renewable and non-renewable resource theory applied to coastal agriculture, forest, wetland and fishery linkages. Marine Resource Economics 9:291-310.
Symanski, E., D.A. Savitz, and P.C. Singer. 2004. Assessing spatial fluctuations, temporal variability, and measurement error in estimated levels of disinfection by-products in tap water: Implications for exposure assessment. Occupational and Environmental Medicine 61:65-72.
Tierney, J. 1990. Betting the planet. New York Times Magazine. December 2.
Toman, M.A. 1998. Why not calculate the value of the world’s ecosystem services and natural capital? Ecological Economics 25:57-60.
Toth, L.A. 1996. Restoring the hydrogeomorphology of the channelized Kissimmee River. Pp. 369-383 in River Channel Restoration: Guiding Principles for Sustainable Projects, A. Brookes and F. D. Shields Jr. (eds.). New York: John Wiley and Sons.
Tsur, Y., and A. Zemel. 1995. Uncertainty and irreversibility in groundwater resource management. Journal of Environmental Economics and Management 29(2):149-161.
Turner, R.K., J. Paavola, P. Cooper, S. Farber, V. Jessamy, and S. Georgiou. 2003. Valuing nature: Lessons learned and future research directions. Ecological Economics 46:493-510.
USACE (U.S. Army Corps of Engineers). 1978. Nonstructural Plan for the East Branch of the DuPage River. Chicago, Ill.: USACE.
USACE. 1999. A Citizen’s Guide to the City of Napa, Napa River and Napa Creek Flood Protection Project. Guidebook prepared by the U.S. Army Corps of Engineers and Napa County Flood Control and Water Conservation District. Available on-line at http://www.usace.army.mil. Accessed December 10, 2003.
USACE. 2002. Lower Snake River juvenile salmon migration feasibility study. Available on-line at http://www.nww.usace.army.mil/lsr/final_fseis/study_kit/studypage.htm. Accessed October 20, 2004.
Vileisis, A. 1997. Discovering the Unknown Landscape: A History of America’s Wetlands. Washington, D.C.: Island Press.
Villanueva, C.M., M. Kogevinas, J.O. Grimalt. 2001. Drinking water chlorination and adverse health effects: A review of epidemiological studies. Medicina Clinica 117:27-35.
Walters, C.J. 1986. Adaptive Management of Renewable Resources. New York: Macmillan.
Wernstedt, K., and C.M. Paulsen. 1995. Economic and biological analysis to aid system-planning for salmon recovery in the Columbia River Basin. Journal of Environmental Management 43:313-331.
Wilson, M.A., and S.R. Carpenter. 1999. Economic valuation of freshwater ecosystems services in the United States: 1971-1997. Ecological Applications 9:772-783.
Wimberley, L.A. 2001. Establishing “sole source” protection: The Edwards Aquifer and the Safe Drinking Water Act. Pp. 169-181 in On the Border: An Environmental History of San Antonio, C. Miller (ed.). Pittsburgh, Pa.: University of Pittsburgh Press.
Winter, T.C. 2001. The concept of hydrologic landscapes. Journal of the American Water Resources Association 37:335-349.
Wright, J.M. 2000. The Nation’s Responses to Flood Disasters: A Historical Account. Association of State Floodplain Managers Inc., Madison, Wis. Available on-line at http://www.floods.org/PDF/hist-fpm.pdf. Accessed December 10, 2003.
Wu, J., R.M. Adams, and W.G. Boggess. 2000. Cumulative effects and optimal targeting of conservation efforts: steelhead trout habitat enhancement in Oregon. American Journal of Agricultural Economics 82:400-413.
Wu, J., K. Sketon-Groth, W.G. Boggess, and R.M. Adams. 2003. Pacific salmon restoration: Trade-offs between economic efficiency and political acceptance. Contemporary Economic Policy 21(1):78-89.
Zhang, X., and R.A. Minear. 2002. Characteristics of high molecular weight disinfection by products resulting from chlorination of aquatic humic substances. Environmental Science and Technology 36:4033-4038.