What Price Salmon?

Credit: OAR/National Undersea Research Program (NURP) With the year-end release of a final decision on how to proceed toward saving wild Northwest salmon from extinction, the Clinton Administration left implementation of its long-awaited plan to the incoming Republicans. For years, researchers have struggled under a glare of media exposure to resolve a central issue: should four hydroelectric dams in Washington be removed to help save the fish? The conclusion is no, not yet, but a scientific divide remains over how close extinction is and over the potential effectiveness of various alternatives to dam removal that the plan recommends. President George W. Bush campaigned strongly against decommissioning the dams, while pledging support for other protection efforts. Anti-dam activists haven't ruled out legal action, as both sides of the debate continue to assert that the "best science" supports them. For researchers, the salmon story provides a multifaceted study in the difficulties of dealing with scientific uncertainties amid crosscurrents of political, economic, and sociological pressure. The story begins in central Idaho's Salmon River. Each spring and summer for perhaps 100 centuries, young chinook salmon have emerged from eggs in their gravel nests to begin a peripatetic life of five or six years. They will travel about 600 miles to the Pacific Ocean before eventually returning to their spawning grounds. One adult fish can spawn 5,000 offspring, but as is similar for many prolific organisms, death from predators and other causes will take all but a tiny percentage. Reared in fresh water, the juvenile fish, or smolt, migrate down the fast-flowing Salmon, which meets the Snake River at the Idaho-Washington border, which in turn is the largest tributary to the mighty Columbia. Along the way, the animal undergoes smoltification, a physiological change that enables it to survive in salt water. Emerging from the estuary, it inhabits the ocean for one to three years before beginning its famed upstream struggle home. The adult fish dies after spawning, its carcass becoming an important contributor to organisms up and down the ecosystem scale. Nowadays, the numbers of wild (as opposed to hatchery-raised) adults that return to their place of birth are so small that a dozen genetically distinct salmonid populations throughout the Columbia Basin are in danger of extirpation, or localized extinction. These populations, called Evolutionarily Significant Units (ESUs), become nonviable when they reach numbers below which loss of genetic diversity--and consequently, of adaptational fitness to the local environment--becomes a virtual certainty. The hatchery fish don't provide such advantages. In recent years, researchers have implanted tags to track the movements of salmon, but they are untraceable once they reach the ocean, and the numbers of tagged fish that make it back to spawn are often too small to provide strong statistical data. Scientists use what empirical and historical evidence is available, but researchers still must rely heavily on computer models to estimate extinction risks and to make projections about the effectiveness of different population management options. The assumptions that go into these models affect the results achieved, and thus are at the crux of the debate over how best to help the salmon not only survive, but recover their numbers enough for safe harvesting by recreational, commercial, and ceremonial Native American fishers. Mike Gold, Courtesy Idaho Deptartment of Fish and GameMerrold Gold and the 54-lb. chinook salmon (the state record) he caught in 1956 on the Salmon River near Chalis, Idaho Science at Odds The fish's abundance in the Columbia Basin began decreasing in the 1870s, but especially for spring/summer chinook; that decline started accelerating a few years after the last of eight federally owned hydroelectric dams was completed along the Columbia and Snake Rivers in 1975. For years, controversy has swirled around the final four dams to be built, along a roughly 140-mile stretch in southeastern Washington. If these four are the main culprits of accelerating decline--because Idaho fish can't withstand the doubling of the dams--they could be breached. That is, their dirt portions could be removed to create a free-flowing channel around the concrete sections that house the navigational locks, spillways, and power plants. But are they the main culprits? The science is divided. Many studies and reports have been funded by government agencies with vested interests in the dams, while conservationist groups also have commissioned numerous reports. Not surprisingly, results of studies funded by these two camps tend to fall in line with the pro-dam or anti-dam stances of their respective backers. Other papers have been produced by independent scientific groups, which usually also receive government financial support. Taken together, the studies leave little doubt that breaching would help Snake River salmon and steelhead, but parties differ over how effective various nonbreaching measures might be. In particular, there is much contention over mass transport of fish by truck and barge, from above the farthest upriver of the eight dams to below the farthest downriver one. This practice--which has been going on for years, beginning experimentally more than two decades ago--is variously thought to be the fundamental answer to the problem, worthy of increased effort, and an abject failure that actually contributes to fish mortality. Again, such conclusions depend to a significant degree on models. The political and economic considerations in the choice of breaching or barging are many. Breaching would incapacitate the dams, which not only produce about 5 percent of the Northwest's electricity, but enable commercial barge transport, provide recreational opportunities, and contribute to flood control and irrigation. Among arguments against dam removal are its expense and virtual irreversibility, concern that breaching would create serious new problems for fish, and the impact on the regional economy of defunct dams. Pro-breaching advocates cite the ecological, cultural, and economic importance to the region of saving the salmon, the perceived lack of time left to do so, and their doubts that the science supports the potential effectiveness of other restoration methods. Against this backdrop, the Federal Caucus, a group of nine agencies with natural resource responsibilities under the Endagered Species Act (ESA), issued the Basin-Wide Salmon Recovery Strategy on Dec. 21. Among those nine organizations, three are legally designated as the "action agencies" for Columbia River fish recovery, because their actions may affect the species or its habitat. Those three are: Bonneville Power Administration (BPA), which markets hydroelectricity from federal dams; the U.S. Army Corps of Engineers, which owns and operates 12 federal dams; and the US Bureau of Reclamation, which owns and operates the only two other federal dams. The trio also consulted on a second salmon document released Dec. 21, a Biological Opinion (BiOp) issued by the Seattle-based Northwest regional office of another Federal Caucus member, the National Marine Fisheries Service (NMFS).1 Delay Recommended The Strategy and the BiOp recommend that a decision on breaching be deferred for at least three years, while other mitigation actions are implemented. These measures involve the so-called "4 Hs": hydropower, habitat, hatcheries, and harvest. The plan calls for minimum river flows during crucial parts of the year, more spill over the dams, and upgrading of fish-passage facilities and spawning habitats. The almost 100 regional hatcheries releasing about 150 million smolts annually will be more closely regulated, and conservation hatcheries will be developed for endangered species, to help dampen disease spread by hatchery fish and population "swamping" of wild stocks. The current severe restrictions on harvest for commercial, sport, and tribal use will continue. The plan also includes performance standards to determine whether these nonbreaching methods are working. If not, permission to remove the dams could be requested of Congress by 2003, with breaching potentially beginning by 2005. There is widespread agreement that no one action represents a "silver bullet." Meanwhile, the time remaining for any combination of recovery methods to work has been variously estimated, depending on the population being considered and the models. Based on NMFS modeling, the BiOp gives as much as 100 years before many of the ESUs will have greater than a 50 percent chance of extinction. Conservationists point to estimates of roughly two decades, and in some cases, much less. Once started, partial removal of the dams would require at least five years to complete. The Strategy and BiOp argue that, despite scientific uncertainties, strong action must be taken right away. Yet, the NMFS extinction model uses a highly conservative threshold of one returning adult spawner per year in an ESU to define extinction. Other models use extinction thresholds of 15 to 300 spawning fish, depending on the population being studied. Michelle M. McClure, Columbia River salmon science coordinator for NMFS at the Seattle-based Northwest Fisheries Science Center, admits that one returning spawner is an extreme value, but she says the goal was to do a comparative study for all endangered species in the Columbia Basin. This requires standardization, in order to set management priorities. "A lot of people have motivations to have a particular scientific result, and the problem with the models is that you can make a whole lot of assumptions that can end up with just about any of those results," she says. Pointing out that the NMFS mandate with regard to salmon is to help recover the endangered species, she implies NMFS has been caught between the opposing camps. "We've been criticized for being both too optimistic and too pessimistic in our models." Gretchen R. Oosterhout, a systems scientist based in Eagle Point, Ore., has prepared several reports for the Portland-based conservation group Trout Unlimited that are critical of NMFS methods. Her reports charge that the NMFS model for assessing extinction risks is too simplistic, not validated, ignores trends and uncertainty in the data, and consistently errs on the side of optimism. Similarly, she and other scientists from various state and federal organizations issued a technical review last year that slams the NMFS population model for Snake River spring/summer chinook. Among their main concerns are that the model's estimates of survival rates at different life cycle stages contradict empirical data, and that it ignores the phenomenon of delayed or extra mortality, which is death occurring in a later life stage than its cause. "No one has ever been able to resolve the problem of delayed mortality," says Oosterhout. She explains that it exists in dammed rivers around the world, but she knows of no precedent for anadromous fish runs being restored without removing dams. Much controversy has surrounded a parameter called "D," which is the differential survival rate of fish transported around the dams compared to those that migrate in-river. Oosterhout and others believe that this emphasis is misplaced. What is most important is not how many young fish get past the dams safely, but how many adults return to their upriver spawning sites. Known as the smolt-to-adult return rate (SAR), this parameter is the number of returning adults divided by the number of outmigrating juveniles. Charles E. Petrosky, a fish biologist at the Idaho Department of Fish and Game (IDFG) in Boise, says SARs in the 1960s for spring/summer chinook were between 3 and 7 percent, but those figures have hovered between 2 percent and less than 1 percent in recent years. He attributes this to a combination of dam effects and poor ocean conditions. Courtesy Idaho Department of Fish and Game Chinook salmon on spawning grounds in Idaho Death Threats What is happening to outmigrating fish after the dams? About 60 percent of their estuarine habitats, where they undergo smoltification, have been diked and drained. Disproportionate numbers of predators, such as Caspian terns, cormorants, bass, and seals feast on them. Grazing, logging, and farming have contributed to river degradation. Oceanographic events, such as those in El Niño years, are thought to affect the food chain and the salmons' ocean survival. Such factors are why understanding delayed mortality is so important. For example, fish stunned by dam passage can then become easy meals for predators. The salmon also can be weakened by gas-bubble poisoning from dam operations, or high temperatures in normally flowing water that has been slowed by dams, or diseases and immune system deficiencies incurred during the process of transporting the fish around dams. However, if ocean conditions account for most of the delayed mortality, then moving fish past the dams is the extent of human intervention. McClure says the greatest mortality happens during the salmon's freshwater residence and in its later transition to salt water. This is why the BiOp recommends habitat improvements. Conservationists argue that upriver tributaries in Idaho already are as close to pristine as they'll ever get, and that transporting fish in a few days to the estuary rather than the weeks required for in-river migration adversely affects their smoltification. The pro-dam side says only four ESUs in Idaho would be affected at all by breaching, which leaves eight ESUs without help. Anti-dam advocates point out that almost half of all the Columbia Basin's spring/summer chinook historically have come from Idaho spawning grounds, and approximately the same is true of steelhead. Pro-dam people say removal would release tons of sediment, damaging spawning habitats. Their opponents retort that most of it will wash to sea, and the dams already have ruined much habitat for salmon that spawn in the river's mainstem. Aside from killing many young fish that pass through dams, the magnitude of their impact on salmon populations is undetermined, McClure contends. If you hypothesize that the only mortality from dams is due to passage, and if you show that fish can be barged around dams with no ill effects, then you should do things like improve habitat and reduce predation before breaching. Under that scenario, "Whether or not you want to have fish in barges on the river is a societal question, not a scientific question." What Will Work? Contemporary efforts to protect salmon can be traced back to 1980, when Congress enacted a law that gave authority to the Northwest Power Planning Council (NPPC) to protect fish and wildlife affected by Columbia Basin hydropower. Nevertheless, Snake River coho salmon were extinct by 1984, and in November of 1991, the river's sockeye salmon were the first to be listed under the ESA, followed in 1992 by fall chinook and spring/summer chinook, then by chum salmon and various basinwide chinook and steelhead populations over the last few years. In 1993, a BiOp from NMFS determined that hydroelectric operations posed "no jeopardy" to salmon, but this was successfully challenged in the Federal District Court of Oregon the following year by IDFG. Judge Malcolm F. Marsh ruled, "The situation literally cries out for a major overhaul. Instead of looking for what can be done to protect the species from jeopardy, NMFS and the action agencies have narrowly focused their attention on what the establishment is capable of handling with minimal disruption." NMFS issued a revised BiOp in 1995, which increased flows and other measures to improve water quality and temperature. It also proposed another such report in about five years, after the completion of more study. That led to creation of the Plan for Analyzing and Testing Hypotheses (PATH). Deliberately representative of diverse interests, PATH included more than 30 scientists from NMFS, the U.S. Fish and Wildlife Service, universities, state fish and wildlife services, Native American organizations, NPPC, the Corps, and BPA. Over the next few years, the PATH investigators assessed various scenarios for saving the salmon, and a Scientific Review Panel comprised of four experts from outside the region reviewed several papers produced by PATH and issued a formal Weight of Evidence report. The latter made no recommendations, but it used structured decision-making techniques to analyze three hypothetical management options. It concluded that the most effective recovery strategy involved breaching. After the Weight of Evidence and final 1998 PATH report2 were released, more than 200 scientists--mostly fisheries biologists in Idaho, Oregon, Washington, Montana, and Alaska--signed a letter to then-President Bill Clinton. Advocating abandonment of "the failed practices of the past," it advised that "we move quickly to restore the normative river conditions under which these fish evolved." A second letter was released three days before the final Strategy appeared in December. It explicitly called for breaching within a few years if other measures don't work, a stance that the final Strategy and BiOp have adopted. Yet, in assessing salmonid population trends and the impact of various actions on those trends, the BiOp relies primarily on the simplified models that Oosterhout and others deplore, but that NMFS defends as more expedient for establishing priorities across the basin than were the PATH methods and findings. At public hearings arranged last year by the Federal Caucus in 15 towns and cities across four Northwest states, thousands of citizens appeared, many of them waiting hours for a chance to speak. More than 9,000 gave testimony, and 60,000 written comments poured in. "The message was clear," read a statement on the Caucus Web site. "The people and governments of the region will make sacrifices to save the fish, but they want the burden to be shared and they want actions that work."3 The question remains: What actually will work? Ed Bowles, anadromous fish manager for IDFG in Boise, comments on uncertainties about the ecosystem, soft mathematics, and the inability to test hypotheses. "You'll always have, at best, 20 percent black, 20 percent white, and 60 percent gray," he says. "What we have in the Columbia Basin situation is a lot of people hired to exploit the gray, because as long as you have uncertainty, you'll default to the status quo." University of Washington zoology professor emeritus Robert T. Paine is chairman of a six-member scientific panel recently set up to guide technical work on recovery of 26 ESA-listed salmon and steelhead populations on the West Coast. The research will be conducted by perhaps 10 teams NMFS currently is appointing in California, Washington, and Oregon. "My fascination is whether, in fact, NMFS has the spine to overcome the deeply held political convictions of all the parties," Paine comments. "Whether they can overpower, or bully, or use the courts in the sense of trying to help salmon, to change these deeply held convictions of other people, I don't know." Steve Bunk (sbunk@uswest.net) is a contributing editor for The Scientist. References 1. The Basin-Wide Salmon Recovery Strategy and the Biological Opinion are at www.nwr.noaa.gov/1hydrop/hydroweb/fedrec.htm. 2. D.R. Marmorek et al., eds., "Plan for analyzing and testing hypotheses (PATH) final report for fiscal year 1998," Dec. 16, 1998, online at: www.efw.bpa.gov/PATH. 3. The Federal Caucus Web address for Pacific salmon preservation is: www.SalmonRecovery.gov.

What Price Salmon?

Interested in reading more?

Become a Member of