Hells Canyon (map), is the spectacular canyon of the Snake River — deepest in North America, averaging about a mile from rim to river — that forms the border between Idaho to the east and Oregon and Washington to the west. Much of the canyon is in public ownership. Hells Canyon National Recreation Area encompasses 652,488 acres, including 214,000 acres of wilderness. Hells Canyon itself is about 71 miles long and 10 miles wide, on average, and in places the distance from the tops of the highest peaks bordering the canyon to the river is nearly 8,000 feet.

In the 1950s and ‘60s, Hells Canyon was something of a battleground over efforts to build hydropower dams for economic development, on the one hand, and to protect wild spaces, fish, and wildlife, on the other. At the center of this public debate were questions about how to develop the hydropower potential of the substantial annual volume of the Snake River in Hells Canyon; how to preserve the scenic canyon for public recreation; and how to protect salmon and steelhead that passed through or spawned in the canyon. These fish had economic importance in the region and, for Indian tribes, a cultural importance.

Salmon and steelhead

The Pacific Fishery Management Council has estimated that prior to 1850, when Euro-American people began to settle in the Northwest, the Snake River Basin produced about 1.4 million Chinook, 200,000 coho and 150,000 sockeye annually. Steelhead numbers are harder to estimate, but one researcher estimated in 1970 that steelhead habitat existed throughout the Columbia Basin at a ratio of about 1.7 to 1 compared to coho habitat. Thus, the Snake Basin probably produced about 340,000 adult steelhead per year, an estimate based on the estimated abundance of coho. Based on the Pacific Council’s estimate of salmon and steelhead production in the entire Columbia Basin, this means that the Snake produced 41 percent of all Chinook salmon, 16 percent of all coho, 16 percent of all steelhead, and 23 percent of all sockeye. Expressed another way, Peter Hassemer, an Idaho Department of Fish and Game biologist, wrote in 1992 that before the Federal Columbia River Power System in the Snake and Columbia rivers was completed in the mid-1970s, the Snake River produced about 40 percent of the adult spring Chinook salmon, 45 percent of the adult summer Chinook salmon and 55 percent of the summer steelhead that returned from the ocean each year to spawn.

Salmon and steelhead spawned in numerous Snake River tributaries, and fall Chinook salmon spawned in the mainstem of the river including in Hells Canyon and farther upstream. The Clearwater, which enters the Snake at Lewiston, Idaho, downstream of Hells Canyon, and the Salmon in Idaho and Imnaha in Oregon, which enter the Snake within the canyon, supported strong populations of salmon and steelhead. There was an important Indian fishery at the mouth of Asotin Creek, just south of present-day Clarkston, Washington, and tribal fishers journeyed from the desert plains of southern Idaho to fish for salmon and steelhead in Snake River tributaries in the mountains of central Idaho. Salmon and steelhead spawned in the Snake River Basin as far upriver as Shoshone Falls, some 300 miles above Hells Canyon.

As non-Indian farmers and miners steadily populated the region in the mid to late 1800s, salmon and steelhead provided an important food source and, for many, a source of income. At the headwaters of the Salmon River, 6,000 feet and more above sea level, there was talk about a salmon cannery, and men sold salmon and steelhead to the mining camps. Over the mountains to the south, at Payette Lake, the headwaters of the Payette River, commercial fishers salted and packed as many as 75,000 sockeye per year in the 1870s. A resident of the area commented in the journal of the United States Fish Commission in 1895 that the runs, which were declining at the time, once had been so thick that he sometimes had to shoo the salmon away before his horse would cross the river.

By 1888 the commercial salmon catch at Payette Lake had declined substantially, perhaps as the result of overfishing in the lower Columbia River, which also was taking a toll on salmon and steelhead populations that spawned in other parts of the Columbia River Basin. Tributary dams, like Black Canyon Dam, completed in 1924 on the Payette, impacted many of Idaho’s salmon and steelhead populations. Black Canyon Dam, like many others built in that era, lacked fish-passage facilities and resulted in the loss of the only Sockeye salmon population that existed upstream of Hells Canyon. Other human activities in the spawning areas, such as mining and farming, also took a toll on water quality and spawning habitat.


Beginning in the 1930s, the hydropower potential of the Columbia River Basin began to be developed in earnest. Rock Island, completed in 1933, the first dam across the mainstem Columbia, was not a federal project, but subsequent big dams were — Bonneville (1938), Grand Coulee (1941), McNary (1953), and Chief Joseph (1955), for example. By the 1950s, though, the Eisenhower administration sought to reduce the role of the federal government in power projects. And so it was not surprising that a battle erupted over who should construct a big hydropower dam in Hells Canyon: the federal government or a private business. In 1949 the U.S. Army Corps of Engineers issued a water development plan for the Columbia River Basin that included a series of low, run-of-the-river dams in the lower Snake River and a high dam in Hells Canyon. The high dam would regulate river flows downstream to optimize hydropower generation at the lower dams, similar to the way Grand Coulee Dam on the Columbia regulates flows and power production at federal and non-federal run-of-the-river dams downstream.

Meanwhile in the 1940s, Idaho Power Company also saw the hydroelectric potential of the Snake River in Hells Canyon. In 1947, two years before the Corps issued its plan and one year before the federal Interior Department issued its own proposal for new dams in the Columbia River Basin, Idaho Power applied to the Federal Power Commission for a license to build the three-dam Hells Canyon Complex. The Commission, forerunner of today’s Federal Energy Regulatory Commission, issued a license to Idaho Power in August 1955. The first of the three dams, Brownlee, was completed in 1958.

Idaho Power’s eight-year journey through the licensing process did not cool the interest of others in competing to build the high dam. There were two competing proposals, and either would have impounded a reservoir stretching more than 50 miles upriver to Brownlee Dam.

One, called High Mountain Sheep Dam, was proposed by the Pacific Northwest Power Company, a consortium of four private power companies. Pacific Northwest Power filed its license application in March 1958. The dam, planned to be 670 feet tall, would have been built about one mile upstream from the mouth of the Salmon River. The reservoir would have been 58 miles long. The other, called Nez Perce Dam, was proposed by the Washington Public Power Supply System, an agency that represented a number of publicly owned electric utilities in that state (this is the same entity that took on the task of building five nuclear power plants in Washington in the 1970s. See Hydro-Thermal Power Program). Nez Perce Dam would have been even taller — 700 feet. It would have been built at a site just downstream from the mouth of the Salmon River, and thus would have wiped out anadromous fish runs to the Salmon River and its tributaries. Meanwhile, the federal Bureau of Reclamation and the Corps of Engineers argued that any new, high dam should be built by the federal government as part of a federal water development plan for the lower Snake and Columbia rivers.

After a lengthy review of the competing proposals, the Federal Power Commission issued a license for High Mountain Sheep Dam to Northwest Power in February 1964, despite objections by the federal Department of the Interior and citizen groups that formed in opposition to the dam. Litigation followed.

It was an unusual lawsuit — the Department of Interior suing the Federal Power Commission — but the Interior Department feared the environmental impacts of the dam, particularly on salmon and steelhead, and challenged the license on these grounds even though the Federal Power Commission noted in its order granting the license that “. . . Any power development would adversely affect the fish and wildlife resources of the area, and particularly the anadromous fish … [M]easures for the conservation of those resources could and would be taken, and . . . about $5,000,000 would be required for the fishery program.”

The license was upheld by an appeals court, but overturned in the U.S. Supreme Court in an historic decision written by Justice William O. Douglas. The Supreme Court ordered the Federal Power Commission to reconsider the application; Douglas interpreted the Federal Power Act, the authority for issuing the license, to require the consideration of alternatives to federal development — including no development. According to the opinion (Udall v. Federal Power Commission, 387 U.S. 428), Section 10 of the Federal Power Act provides that a project licensed by the Commission shall be “…such as in the judgment of the Commission will be best adapted to a comprehensive plan for improving or developing a waterway . . . and for other beneficial public uses, including recreational purposes.” Douglas wrote:

The objective of protecting ‘recreational uses’ means more than that the reservoir created by the dam will be the best one possible or practical from a recreational viewpoint. . . . The importance of salmon and steelhead in our outdoor life as well as in commerce is so great that there certainly comes a time when their destruction might necessitate a halt in so-called ‘improvement’ or ‘development’ of waterways. The destruction of anadromous fish in our western waterways is so notorious that we cannot believe that Congress through the present Act authorized their ultimate demise. . . The test is whether the project will be in the public interest. And that determination can be made only after an exploration of all issues. . . including future power demand and supply, alternate sources of power, the public interest in preserving reaches of wild rivers and wilderness areas, the preservation of anadromous fish for commercial and recreational purposes, and the protection of wildlife.

Douglas held that the commission also must consider “the need to destroy the river as a waterway, the desirability of its demise.”

It was the fight over High Mountain Sheep Dam that led Congress to create the Hells Canyon Wilderness within the larger Hells Canyon National Recreation Area, in 1975. The fight over the high dams in Hells Canyon was significant in the history of conservation, author Tim Palmer writes, because the Supreme Court addressed the damage the dam would cause to fish, wildlife and recreation in Hells Canyon and directed the Federal Power Commission to deny dam licenses when doing so would serve a larger public interest.

The Hells Canyon Complex

It is difficult to say today how many salmon spawned upstream of the Hells Canyon Complex historically, but fish counts at the Brownlee and Oxbow dams between 1958 and 1960 probably give some indication, although the numbers could have been higher 100 years earlier. At the time, the maximum counts were approximately 17,000 fall Chinook in 1958, 2,600 spring Chinook in 1960 and 4,500 steelhead in 1959 and 1960. Sockeye were extirpated with construction of Black Canyon Dam in 1924. There were no coho upstream of Hells Canyon historically.

The resolution of the High Mountain Sheep Dam litigation and the subsequent failure of the high-dam proposals cleared the way for Idaho Power to construct the other two dams in its three-dam Hells Canyon Complex. Brownlee, at Snake River mile 285, had been completed in 1958 and began generating power the following year. Oxbow, 12 miles downstream at Snake River mile 273, was completed in 1961, and Hells Canyon, 26 miles downstream from Oxbow at River Mile 247, was completed in 1967. None has fish passage, but fish passage was in the original plans.

Article 34 of the Hells Canyon Complex license required Idaho Power to make $250,000 available to study the fish runs and determine how to mitigate losses that would be caused by the dams. In response, state and federal fishery agencies reviewed all known means of mitigation, including passage, relocation of the runs, artificial propagation, and natural redistribution of fish in streams downstream of the complex. Of these, passage appeared most promising and also held out the promise of restoring salmon to their historic spawning areas.

In 1956, the year construction began at Brownlee, the Idaho Department of Fish and Game and the U.S. Department of the Interior sent letters to Idaho Power requesting fish passage facilities for both upstream- and downstream-migrating fish at the Hells Canyon Complex. The same year, the Secretary of the Interior sent a studies program and budget to the Federal Power Commission, and the next month the Commission authorized Idaho Power to spend the set-aside $250,000 on the studies.

Construction began immediately on temporary upstream adult passage facilities at Brownlee. A fish trap using an electric barrier to guide fish to the trap was completed in December 1957. This trap was soon replaced by an adult trap at the Oxbow Dam construction site downstream in May 1958.

While state and federal fish and wildlife agencies pressured Idaho Power and the Federal Power Commission to provide fish passage at the dams, the agencies also made clear that this policy position assumed that protective devices and procedures would be experimental. Unfortunately, the fast pace of dam construction was too fast for investigations that may have provided more information about the efficacy of fish passage. Other forms of mitigation remained under consideration, but these were problematic. For example, hatcheries were dismissed because the initial review of mitigation options revealed water temperature and siting problems in Hells Canyon that would make fish production difficult. Despite the fact that hatcheries had been operating in the Columbia River Basin for more than 70 years, including, for example, the successful hatcheries in the upper Columbia built as mitigation in the 1940s for Grand Coulee Dam, the Fish Commission of Oregon and the Oregon State Game Commission worried about whether hatcheries could maintain large runs of salmon and steelhead, particularly those in the Snake River. In comments prepared for the Hydroelectric Commission of Oregon in 1956 regarding permits for the Hells Canyon dams, the two agencies wrote that:

. . .maintenance of large runs of summer-run steelhead and spring Chinook salmon by means of artificial propagation is not feasible in light of present knowledge. This is due to the unique life history of these fish, which brings them into the rivers months before their spawning time and which requires a year’s residency in freshwater prior to their journey to the sea.

Hatcheries, if anything, were considered a fallback option. Artificial spawning and incubation channels also were considered, but again, these had proven problematic elsewhere. Channels had been built at McNary, Priest Rapids, Rocky Reach, and Wells dams on the Columbia. While similar channels had been successful in Canada, it was nothing more than speculation that they might work in Hells Canyon, and they had not been particularly successful to date in the Columbia. Idaho Power also considered whether the runs should be relocated, as was done with the upper Columbia runs affected by Grand Coulee. Translocation was an intriguing concept, but there was no way to know whether it would work. Biologists lacked critically important information on the life histories of the Snake River stocks and their ability to adapt to conditions in tributaries where they might be relocated.

So the focus of the Article 34 effort remained on fish passage. Events outside of Idaho Power’s control added to the complexity of the challenge. How to provide fish passage at the Hells Canyon Complex depended on whether either of the high dams proposed for construction downstream would be built. Given their height, it is likely that neither would have had fish passage facilities, and this would have changed the fish-passage effort at the Hells Canyon Complex. As early as 1951, Idaho Power was considering ladders for adult fish passage at the Hells Canyon Complex, and by 1954 the company was investigating whether fish ladders or fish elevators would work best. But if either of the high dams were built, fish ladders would be useless. By 1958, when Pacific Northwest Power filed its license application for the High Mountain Sheep Dam, Idaho Power had decided that best option for the Hells Canyon Complex would be trapping and hauling the fish as opposed to constructing fish ladders or fish elevators.

And that is what happened. Adult fish returning upriver would be trapped at the face of the first dam — fish traps moved downstream progressively as the three dams were completed — and then hauled to release points above the dams. In fact, adult spring Chinook and steelhead were trapped and hauled successfully to a release point about a mile above Brownlee from 1956 to 1964.

While passage of adult fish was successful, passage of juvenile fish migrating downstream was not. How and where to collect juvenile fish at Brownlee was a matter of intense discussions among Idaho Power and state and federal fishery agencies.

The Fish Commission of Oregon, for example, preferred that juvenile fish be collected at the surface in the Brownlee Dam forebay and that the dam’s turbine entrances be located 120 feet below the surface in order to prevent juvenile fish from passing through the turbines. This would have allowed the fish to be collected at the surface using outlet pumps, as it was believed that the fish would not dive 120 feet to go around the dam. However, because of a requirement by the Corps of Engineers for Brownlee Reservoir to provide flood control with a potential flood-control drawdown of 101 feet, the turbine outlets would have been required to be 220 feet deep. This would have required a different design to the dam — concrete as opposed to rock fill dam, and that design would have been more expensive. Idaho Power planned the turbine entrances to be only 16 feet below the surface at the minimum reservoir elevation. This design required a different means of collecting fish than originally envisioned by the fish management agencies.

The solution proposed by Idaho Power and ultimately adopted with some reservation by state and federal fishery agencies was to install a plastic mesh barrier net across the reservoir that would extend to a depth of 120 feet with another 20 feet of net extending horizontally upstream from the bottom of the net. The net would be suspended from pontoons and anchored to large concrete abutments on each bank a mile upstream from the dam. The barrier net would guide fish to collection points on the surface, and the fish then would be trucked to release sites downstream of the dam. As each successive dam was built, the fish could either be trucked to a point below the lowest dam or released into the pools between dams.

Idaho Power and the fish agencies explored various collection and transportation alternatives. Returning adult fish would be trapped at the lowest dam and released above the upstream dam or dams. Again, there were multiple options, including releases into creeks that flowed into the pool behind Oxbow Dam and releases into streams above Brownlee in both Oregon and Idaho. But the success of any option depended on the success of the juvenile collection system and, in particular, the enormous barrier net.

The net didn’t work as planned. It was installed in 1958, and within a year, state fish and wildlife personnel began to speculate publicly that it, and the associated juvenile fish-collection and trucking facilities, would fail. Periodic inspections by divers found the net sound in places but failing in others. Parts of the net hung in the water at odd angles as the river current pushed and pulled. In short, the net did not guide fish well. Many fish apparently passed under the net or through it. As a result, more fish went through the Brownlee turbines than had been anticipated. Juvenile fish survival, as measured by marking fish upstream of the dam and recovering them at the barrier net or at the dam, declined.

In hindsight, there were other problems for juvenile fish in Brownlee Reservoir that probably affected survival as much as the failing barrier net system. Brownlee Reservoir was large, slow-moving and, particularly in summer, warm. Juvenile fish apparently became disoriented and had difficulty finding their way through the reservoir. Wastes from sugar and potato processing plants upriver, fertilizers that ran off of fields into the reservoir, and algae blooms combined to reduce the oxygen content of the water. By August of most years, low levels of dissolved oxygen and warm surface water temperatures forced fish to live within a narrow band in the water column.

Faced with fish-collection and water quality problems, the fish agencies recommended developing a plan to mitigate the impacts of the complex by building hatcheries downstream from Hells Canyon Dam. In August 1960, Idaho Power first discussed abandoning the fish passage facilities and using hatcheries to compensate for the losses. The first hatchery was proposed for construction at Oxbow Dam. The previous month, July 1960, Milo Moore of the Washington Department of Fisheries reported the barrier net was not working and that the fish runs in Hells Canyon “are being badly mauled by inadequate facilities.” He proposed a program of hatchery production. Later, in November 1960, Moore reported: “the Brownlee-Oxbow facilities have failed. To continue putting a major portion of the anadromous fish runs above this project for study purposes is tragic.” The federal Interior Department did not agree, however, and advised against abandoning fish passage, but by 1962 it was apparent the barrier net concept never would work. Plans were under way to convert fall Chinook mitigation efforts to hatchery production and to relocate spring Chinook and steelhead stocks to tributaries downstream of Hells Canyon Dam.

Some interests continued to argue for fish passage, including the Idaho Wildlife Federation, which issued a statement to that effect in July 1963. Idaho Power responded with a letter that stated, in part, “Fisheries agency people are already agreed that fish passage of fall Chinook (the most abundant species in the Brownlee Dam complex area) should be discontinued in favor of below-dam artificial spawning and rearing facilities.” Two years later, in June 1965, the Oregon Fish Commission issued a statement suggesting that passage be re-established if a better means of collecting and passing the downstream migrants could be developed at Brownlee.

But the steady failure of fish passage facilities, and the pace of dam-building in Hells Canyon, supported the positions of Idaho Power and others who believed that artificial production would provide the surest, quickest and most effective mitigation.

In December 1963, the Federal Power Commission ordered Idaho Power to abandon the barrier net collection system at Brownlee and purchase land in the lower Rapid River, a tributary of the Salmon River, for a hatchery. By August 1965, the fish and wildlife agencies had decided in favor of hatcheries and expressed their decision in a document entitled, “Policy for the perpetuation and management of anadromous fish in Snake River drainage upstream of Salmon River.” The policy left open the option of re-establishing fish runs above the dams:

If determined by the agencies that the fish runs will be re-established, upriver production areas now blocked by existing dams or currently rendered unusable because of other water uses shall be made accessible and/or productive as early as possible through the provision of passage facilities, pollution abatement and releases of water from storage to augment quantities and improve qualities of the flows to satisfy migration, spawning, incubation and rearing requirements.

In January 1966, the Federal Power Commission ordered Idaho Power to limit its mitigation efforts to artificial production. It may seem remarkable to some people today that more effort didn’t go into fish passage in Hells Canyon, but the decision needs to be understood in the context of the time. Biologist Don Chapman, a longtime observer of fish policies in Idaho, wrote a history of the Hells Canyon complex of three dams for Idaho Power Company as part of its application to the Federal Energy Regulatory Commission to relicense the project. In his history, Chapman says the key issues for state and regional decision-makers in the late 1950s were development, energy for economic growth and the question of whether dams would be built by public or private interests. The two decades after World War II were “the last years relatively unfettered by legislation designed to protect natural, aquatic and terrestrial communities and habitats,” he writes. They were, in his opinion, “the last in the 200-year march toward manifest destiny in resource exploitation.”

At the time, ecological understanding was “relatively primitive,” Chapman wrote. Mechanistic scientific research had only begun, and knowledge of fish genetics, behavior, and ecology was limited, at best. “In short,” Chapman wrote, “the mid-1950s were not conducive to regulators being able to ignore the societal forces that insisted on development at a rapid pace.”

And the pace of development was rapid, indeed. The Federal Power Commission issued the permit for construction of Brownlee Dam in August 1955; from that date to closure of the spillway gates in May 1958, when the reservoir began to fill, was just 33 months. Even with the fast track, the scant knowledge about anadromous fish, and the available mitigation tools, would have made fish passage at the dams impractical. Eventually, Idaho Power built four hatcheries: Oxbow, Rapid River, Niagara Springs, and Pahsimeroi. In 1980, Idaho Power, the states of Idaho, Oregon, and Washington, and the National Marine Fisheries Service, signed a settlement agreement providing that the hatchery program together with agreed-upon fish-production numbers constituted full and complete mitigation for all numerical losses of salmon and steelhead caused by the construction and operation of the project.

In 2003, Idaho Power filed an application with the Federal Energy Regulatory Commission (FERC) to relicense the three dams. The existing license expired in 2005, and the company is seeking a new, 30-year license. The dams provide more than two-thirds of the hydroelectricity the company sells.

In anticipation of the relicensing, American Rivers and Idaho Rivers United filed a petition to compel FERC to initiate ESA consultation on the ongoing operation of the Hells Canyon Project in advance of relicensing. FERC delayed its response to the petition and the issue eventually ended up before the U.S. Court of Appeals, which instructed FERC to respond to the petition. FERC responded, and subsequently Idaho Power entered into an interim agreement with various parties, including American Rivers and Idaho Rivers United, on operations pending relicensing. Separate filings with FERC addressed fish-passage studies.

By 2006, FERC had completed a Draft Environmental Impact Statement for the Hells Canyon Complex relicensing. Public hearings were conducted in September. But then the effort essentially ground to a halt. 

In its license application, Idaho Power proposed to develop actions and measures to restore passage and habitat for bull trout, but not for salmon or steelhead, concluding that in most tributaries above the Hells Canyon complex, 1) other dams would preclude passage, and 2) land use practices have degraded habitat and water quality to the point of precluding successful reintroduction.

The dispute over fish passage did not go away, and in fact it foundered on opposing laws in Idaho and Oregon.  Because the dam is on the border, both states must approve its operation.  An Oregon law dating to 2001 requires fish passage at any dam that blocks passage, as Hells Canyon obviously does, and without a commitment to provide passage the state could not issuing a required water quality certification for a new license for the Complex. Idaho law requires approval by the Governor and Legislature before fish passage could be restored, and Idaho Power continued to oppose passage. 

Finally, after lengthy and complex negotiations between the two states and Idaho Power, an agreement was announced in December 2018 (Oregon DEQ news release, Idaho Power new release).  Under the agreement, Idaho Power will pay $20 million over 20 years to improve water quality in the Snake River below the Complex, conduct research, and boost the production of spring Chinook salmon at the company's Rapid River Hatchery from 3.2 million smolts per year to 4 million. In exchange, Oregon accepted that fish passage would not be required by the state Department of Environmental Quality. As a result, both states will issue water-quality certifications for the Complex, allowing the long relicensing effort to move forward. That will resolve the last administrative obstacle in the way of FERC issuing a new license with terms and conditions.