Fish and Hydropower Vie for Columbia River Waters

In a controversial move this past summer, the Northwest Power Planning Council placed 44,000 miles of the region's streams and rivers off limits to future hydroelectric development. This is just the latest move in a massive biological restoration experiment taking place in the Pacific Northwest's Columbia River Basin. At the heart of the matter is an attempt to resolve an old conflict between two competing users of the river—hydropower and fish. More than a century of exploitation and development have reduced the numbers of salmon and steelhead trout produced in the Columbia Basin to a mere fraction of their original abundance. But efforts are under way to rebuild the fishery, and the ratepayers who purchase hydropower produced in the basin are picking up the tab.

Once one of the world's largest producers of fish, the Columbia River and its tributaries are now the world's leading producers of hydroelectricity among river systems of comparable size. Among the biggest dams in the area that have extensive environmental impacts are those owned by the federal government. The present system of 31 federal dams and powerhouses has an installed generating capacity of almost 20,000 megawatts of electricity combined with more than 20 million acre-feet of storage capacity (one acre-foot being the amount of water required to cover one acre with water one foot deep). Dams and powerhouses owned by public and private utilities make up the balance of the more than 130 dams with hydroelectric generating capacity lying within the Columbia Basin. This extensive system provides relatively cheap electricity to northwestern consumers but has changed dramatically the free-flowing nature of the river system. The change has had substantial effects on the fish and wildlife that inhabit the basin.

Especially affected by the hydroelectric system are anadromous fish such as Pacific salmon, steelhead trout, sturgeon, and shad. These fish spend much of their lives in the ocean, but must return to freshwater to spawn and reproduce. Hydroelectric dams can block or impede the natural migration of fish, or the dams may inundate natural spawning areas and render them unsuitable to salmon and other fish that require running water to reproduce successfully.

Scientists estimate that the current annual production of adult salmon and steelhead trout in the basin is about one fifth of the annual production during the early 1800s. Not all of the decline in production can be attributed to the hydroelectric system. Excessive fishing pressure, especially around the turn of the century, and a combination of other factors such as mining, timber harvesting, agriculture, and urbanization have had substantial impacts. However, the hydropower system remains a major deterrent to increased fish production. Why is this? What is being done about it? And who pays?

Early years

These questions can be answered only in the context of the basin's history. Hydropower development began in earnest in the Columbia Basin during the 1930s, partially fueled by the surge in public work projects during the Great Depression. The first major dam, Rock Island Dam, was built by Puget Sound Power and Light Company and was completed in 1933. The much larger federal Bonneville and Grand Coulee dams were completed in 1938 and 1941, respectively. It quickly became apparent that an agency would be needed to transmit and market the large amounts of hydroelectricity that would soon become available. To this end Congress passed the Bonneville Project Act in 1937, creating the Bonneville Power Administration (BPA).

Fortuitously, a large market for electricity developed quickly, primarily in the electrometallurgical industries that produced aluminum for aircraft construction during World War II. Electrical demand in the Pacific Northwest and concomitant hydropower development grew steadily after the war until the late seventies, when growth in demand tapered off.

Electricity shortages?

In the mid 1970s, serious misjudgments concerning the future supply and demand of electricity led to overly optimistic expectations for growth and demand. Regional forecasters predicted electricity shortages and the possibility of brownouts in the Northwest. As a result, the BPA notified customers in 1976 that existing power supplies would be insufficient to meet load demands beyond 1983.

The perception of impending shortages of electricity prompted regional utilities to begin an aggressive program to build large thermal generating plants to augment hydroelectricity. In the meantime, a slowdown in the region's economy, caused in part by depressed demand for Northwest aluminum, combined with lower fuel prices (which make alternative sources of energy more competitive) lessened the demand for electricity. The electricity shortage failed to materialize, and the need for new thermal generating plants disappeared at a time when utility revenues were falling and construction costs were escalating. This foray into nuclear plant construction precipitated the largest rate increase in the BPA's history and a massive default on public bonds.

A second impact of the anticipated, although unrealized, electricity shortage was an effort in Congress, led by delegates from the Northwest, to establish a planning authority that could ensure a safe, reliable, and efficient power supply in the future. This resulted in congressional passage of the Pacific Northwest Electric Power Planning and Conservation Act (referred to herein as the Regional Act) in 1980. While the main impetus for the Regional Act was a perceived failure of power planning, conservationists seized the opportunity to include special provisions for fish and wildlife within the language of the act. The Regional Act established a unique interstate compact, the Northwest Power Planning Council, which is composed of eight members—two members each from the states of Oregon, Washington, Idaho, and Montana, appointed by the governors of those states. The council has responsibility for developing programs for regional power planning and conservation of electricity and for planning measures that mitigate the effects on fish and wildlife of hydropower projects in the Columbia River Basin.

Having followed the procedures specified by the Regional Act, the council adopted its original Fish and Wildlife Program late in 1982. The program was amended in 1984 and 1987. The initial program contained a variety of mitigation measures, including new hatchery construction, improvements in existing hatcheries, restoration of degraded streams, and the installation of bypass facilities to guide migrating young salmon (smolts) around powerhouse turbines at mainstem dams. All of the dams below Chief Joseph Dam on the Columbia River and Hells Canyon Dam on the Snake River had been built with fishways that permit the upstream migration of adult fish (see photo), but few dams were equipped to permit safe downstream passage of smolts. Many smolts are captured at upstream dams and transported downstream via barges or trucks—avoiding intervening dams—but where transportation is not provided and at dams that lack bypass systems, downstream migrants must either pass through the turbines, where around 15 percent are killed at each dam, or over spillways.

A unique plan to increase survival of downstream migrants by increased streamflow to accelerate migration is a key feature of the Fish and Wildlife Program. Under this plan, federal project officers and regulators release 4.64 million acre-feet of water from storage, at the discretion of the fishery managers, between April 15 and June 15 each year. These releases augment flows that are normally provided for hydroelectric generation and navigation. The enhanced flows help carry the smolts to the ocean and are timed in such a way as to maximize the effect of the flows. But benefits gained by the smolts can be losses eventually borne by the ratepayers. In drier years, when natural flows fall below average, such as occurred in 1987, providing additional flow for fish can result in substantial losses in revenues to power producers because water released in the spring in excess of hydropower needs is not available for power production later in the year.

One highlight of the 1987 program amendments is the initiation of a systematic and comprehensive planning process. The purposes of this process are to identify additional measures that collectively would increase the numbers of adult salmon and steelhead trout available for harvest, and to establish production targets for each of 31 tributary streams that feed the Columbia and the Snake. In principle, this systemic approach differs from the earlier approach to mitigation planning in that subbasin improvements are to be integrated more carefully within a basinwide plan. Previously, mitigation measures had been planned and evaluated in relative isolation, based on their individual merit.

Clearly, while much activity already is under way to increase fish populations, current measures will not be sufficient to attain the council's goal of doubling the number of adult fish produced. The council recognizes that this interim goal is an ambitious one which, if feasible, will take considerable time and large amounts of funds to realize. For example, mitigation cost the BPA about $375 million for the period 1983 to 1986, a cost which included direct expenditures, forgone power revenues, and repayments to the U.S. treasury on behalf of other federal agencies. Moreover, as the Regional Act requires, program funds must be used in a cost-effective manner. As mitigation proceeds, incremental costs of further improvements are expected to increase, and more questions will arise about the economic cost, effectiveness, and justification of alternative measures and their system wide implications.

Developing an equitable and effective fish and wildlife program requires making hard choices. As a system, the Columbia River and its tributaries produce an abundance of goods and services: hydro-electricity, navigation, irrigation, recreation, fish, and others. The actual combination of goods and services produced annually is determined by how the system is managed—and any given mix of goods and services involves tradeoffs. For instance, water that is consumed in irrigation or released to enhance fish migration cannot be used later to produce electricity. In a similar fashion, monies spent to build fish hatcheries and enhance habitat cannot be used to develop new power resources or pay off past debts. Finding the mix of goods and services that maximizes benefits to society is the formidable challenge faced by the council, the BPA, and other agencies charged with the management of the region's natural resources.

The primary difficulty in trying to evaluate different combinations of goods and services is that the various entities cannot be measured with a common metric. This is not the old problem of comparing apples and oranges; it's worse. Apples and oranges can be sold or traded in a common market that determines their relative values. But how, for example, does one compare kilowatts to the prospect of losing salmon stocks that have inhabited upstream tributaries for millennia? How many kilowatts does one give up to maintain a given stock at a certain level? How many acres of irrigated land does one take out of cultivation, possibly disrupting a local farm economy, to ensure that sufficient water reaches a stream to support a viable fishery?

The short answer is that there is no best way to compare non marketable goods and services with each other or with those that have a well-defined market. Economists at Resources for the Future (RFF) and elsewhere have struggled with this dilemma for years. Historically, most approaches have been based on attempting to attach monetary values to each product and then comparing these values. However, such analyses inherently involve implicit value systems that are not well delineated, thus producing value estimates that inevitably have a great deal of uncertainty associated with them.

A consensus on values, and the value of consensus

To attempt to place values on all of the goods and services produced in the Columbia System would be a tremendous undertaking that would likely result in a long, loud, and acrimonious debate among the many interested parties over any estimates that might be produced. Perhaps for this reason the council and others in the region have apparently decided to skip the valuation exercise and jump right into the debate. For example, the aforementioned decision to place 44,000 miles of streams off limits to future hydropower development is not based on a comparison of the economic values of the hydropower potential of these areas with their recreational and fishery benefits. Instead, it comes from a perception that the people of the region implicitly value these streams in their natural state more than they would value any hydroelectricity that might be produced from them. As the mounds of public testimony presented to the council preceding this decision attest, this perception is not shared by everyone. But to its credit, the decision does reflect a consensus that was reached in an open public forum.

The ability to build a consensus among many parties with conflicting views has been one of the greatest strengths of the council. Power interests such as the BPA, the Corps of Engineers, and the region's utility companies often are at odds with the state, federal, and tribal fish and wildlife agencies. The council has been quite effective at pulling these parties together so that a common course of action can be agreed upon. But there can be instances in which a consensus decision does not serve the best interests of the region.

The most common situation in which a general accord can be less than optimal is when the decision is based on incomplete or inaccurate information. There are several ways that such a situation could arise in the Columbia River Basin. First, the system under consideration simply may be poorly understood. All parties may then share a common, but inaccurate, perception of the system and how it will respond to management actions. There is a large amount of uncertainty regarding many of the ecological processes at play within the basin. For example, no one knows the extent to which estuary and ocean conditions might limit the production potential of the basin.

A similar situation can occur when various parties withhold or misrepresent information, either for strategic reasons or unintentionally. The latter commonly occurs when different parties present large amounts of technical material that is relevant to program planning. The other parties, lacking familiarity with the information or knowledge of how it is generated, may draw incorrect or unwarranted conclusions from the presentations. This can be especially troublesome when information is presented from models that are given to the other parties as "black boxes" with little explanation of the assumptions incorporated into the models or the suitability of data used to run the models. Several models have been developed by various agencies to assist decision making in the region, but most of these are poorly documented; persons not intimately knowledgeable about each model have a difficult time judging either the credibility or usefulness of model results.

A common basis for decision making

The problem of informational integrity is exacerbated because all parties to a discussion tend to look with suspicion at information presented by others, regardless of the objectivity or accuracy of the information. To do otherwise would be less than responsible.

A partial solution to this problem is to provide a common knowledge base that can be used to support decision making. Such a knowledge base would contain information relevant to the subject under discussion and acceptable models of the system involved, and would be freely accessible to all. This approach enhances the decision-making process because it provides a common basis from which all parties can assess alternatives and reach a consensus.

Developing and implementing an integrated, comprehensive, and equitable fish and wildlife management plan for the Columbia River Basin will require balancing a multitude of economic, hydrologic, and biologic factors. Since 1984, Resources for the Future has contributed to this endeavor by working with the BPA to develop methodological tools and data suitable for estimating the productivity and cost implications of mitigation and enhancement alternatives. The purpose of this research is to support the goal of systematic and comprehensive planning by developing an analytical framework for comparing alternative management strategies and for searching for that combination of measures which is most cost-effective.

RFF has proposed the development of an integrated suite of analytical models that link together hydrologic, economic, and ecological processes. If developed and put to use, these models could actively support program planning in the Columbia Basin by becoming part of a joint-decision support system. The emphasis here is on decision support. Such a system would not usurp the decision-making role properly reserved by the council and other regional policymakers; rather, the analytical system should provide a realistic assessment of the economic costs involved, the level of fish production that might be expected, and a measure of the degree of uncertainty or risk involved in pursuing a particular mitigation strategy. Providing this information in a forum that is accessible to all should increase the odds of making the grand experiment in the Columbia an emphatic success.