Four teenagers, one of them with his hair shaved into a Mohawk, pile out of a Honda and scramble down from the gravel access road to the riverbank. They’re here to see the salmon. One of them lobs a rock at the mass of circling dark shapes in the green water below a broken basalt cliff. The fish, thick and muscular, dodge out of the way of the rock, but continue circling, undaunted. They’ve seen worse in their run from the ocean: sharks, sea lions, Deadline Falls, more than a few fishing hooks. More rocks fall—plunk, plunk, plunk—and the surface of the pool is broken with conflicting ripples that reflect brightly on the basalt cliff. The fish circle, killing time.

This is the end of the journey for the North Umpqua River’s spring chinook salmon. A few hundred yards upriver, Soda Springs Dam has inundated and blocked access to the historic spawning beds of the salmon, and they must make do with pockets of gravel tucked in hollows of the river’s bedrock below the dam. When they spawn in September, many of these fish will crowd into the small patch of gravel that forms a riffle above the pool where they circle. There, backs out of the water, the fish will jostle for a place to lay and fertilize eggs. There won’t be enough space, and some of the beds will be destroyed by other salmon following the same instinct. The resources here are painfully finite, the population of salmon confined by what development has left.

One of the kids mentions the dam just upstream; it was in the news recently. They decided not to take it out, he says, disappointed, to a man standing on the bank watching the scene. The man wears a checked shirt and a baseball cap. He is a U.S. Forest Service scientist, an expert on this watershed and, until a couple of years ago, an integral part of a team of people working to restore the North Umpqua River to functional, if not historic, health. But the kid doesn’t know all that, he’s just making casual conversation, and the scientist doesn’t let on.

The teenagers pile back into the car and drive off. The scientist turns to one of the three other Forest Service scientists with him at the river’s edge and excitedly reports the exchange. The kid knew about the issue, and the scientist is a little thrilled that the benefits of dam removal may be more deeply seated in the local consciousness than he realized.

A small victory, but for scientists brought into the decision-making process at the outset and then dropped from it when their findings didn’t fit with political expectations, the kid’s frustration was an affirmation that their work hadn’t been completely forgotten. Still, the Forest Service scientists’ work to make the project meet the requirements of the Northwest Forest Plan had been largely ignored by their superiors. Their suggestions to restore connectivity to a 440-square-mile watershed harnessed almost exclusively for hydropower while reducing the project’s power generation by less than 10 percent had been abandoned. They were galled when told by the power company that there wasn’t enough science to support the benefits they projected and that as a replacement, a suite of untested fixes would replace most of what the Umpqua scientists had suggested.

Above all, dam removal, once the Forest Service’s top priority for rehabilitating the watershed, would set a precedent that the company didn’t want to make. It’s not a big dam, one of eight in a small hydropower project in southwestern Oregon, with hundreds of hydropower dams on Forest Service land, interpreting the Northwest Forest Plan to mean that the dam had to go might cripple the relicensing of many other projects across the country. The Forest Service capitulated. The scientists fumed.

In early summer, newspaper headlines across Oregon announced a settlement for relicensing the eight dams and canals, powerhouses and diversions in the North Umpqua River watershed. PacifiCorp, the project’s owner, and state and federal agencies had agreed, the newspapers said, to terms that would protect the environment and allow profitable power generation. Soda Springs Dam would stay, but there were plenty of other environmental benefits to the project that would make it better than before. Better, however, is relative—a “deception of scale,” says Jeff Dose, one of the Umpqua scientists—and those settlement terms will be tested in court.

In exploring the effect of hydropower on the North Umpqua River, those Umpqua National Forest scientists opened a time capsule to another forest management era. Hydropower dams are licensed by the Federal Energy Regulatory Commission for thirty to fifty years, and many of the hydropower projects built during the post–World War II boom of dam construction are or will be up for relicensing in the next decade. The dams were built in an era before the National Environmental Policy Act, before the Clean Water Act, before the National Wild and Scenic Rivers Act. Forest management meant diversifying a portfolio of natural resource extraction, and hydropower was a natural fit. There are 198 FERC-licensed hydropower projects on national forests around the country.

The Forest Service has a responsibility under the Federal Power Act to review the impact of FERC-licensed projects on its land and try to bring them up to today’s environmental standards. It’s a balance between the benefit of an emissions-free power source and the environmental degradation caused by replumbing an entire mountain watershed. In the case of the North Umpqua, the forest is managed under the Northwest Forest Plan, the comprehensive management plan brokered by President Clinton in the early 1990s as a result of threatened spotted owl populations. To many it’s a logging plan, but its rules are broader, addressing the condition of entire watersheds and the movement of all species through them. In national forests such as the Umpqua, logging has had a drastic and measurable effect on ecosystem health. So, too, has hydropower.

Steep, rugged country laced with rivers and streams offers three key elements to the dam builder: water, gravity and natural geographic and geologic features that can be used to build an efficient plug. In national forests, the engineers of two generations ago found a bonanza of potential hydropower sites. In the case of the North Umpqua project, small dams were built where broad alluvial plains of cobble and gravel necked down to narrow canyon walls.

Those alluvial plains also were key to the health of the ecosystem. The North Umpqua follows a course of bedrock and at times the river narrows to pass turbulently and powerfully through vertical canyon walls. Where the valley widens, the river slows and becomes shallow. Small rocks and gravel washed down from high in the Cascade Mountains collect. Logs hang up in the shallows and are partially buried by the ever moving cobble. Currents scoop deep, slow pockets behind these logs. Salmon and steelhead use these alluvial plains to spawn, fanning the small rocks with their tails to move them just so before depositing their eggs. Upstream of the anadromous fish’s natural range, the alluvial plains slowed the rushing water of the steep valley and spread it out, forming broad wetlands, vibrant with amphibian, mammal and bird life.

Connectivity gets the Umpqua scientists talking. It’s a concept common to each of their specialties. Mikeal Jones, a hydrologist, studies the mechanics of how water moves through the landscape. Dose and Craig Burns, fish biologists, have spent their careers understanding how fish use that water. Cindy Barkhurst, a wildlife biologist, studies how and why animals need to move through the landscape.

Or why they can’t move through the landscape. The disconnected North Umpqua watershed turns the conversation animated. Dose, tall and white-haired, uses his resonant voice to deliver tightly worded assessments of the situation. He’s practicing for a lecture at Southern Oregon University about just this issue. “If you’re going to restore a river, you can’t have a dam there,” he says. The massive federal plan to dam the West delivered on its anthropocentric promise of the mid–twentieth century, he says, and “that was really the start of the end.” That hatcheries, fish ladders and screens—what Dose derisively calls “techno-fixes”—haven’t worked to maintain the health of a disrupted ecosystem is no surprise. “People who know nothing about the ecology of the river are making the decisions,” Dose says.

Barkhurst steps quickly across a narrow dirt-covered bridge above a concrete canal. The water rushes silently below. The canal holds most of the flow of the North Umpqua, sluicing it on a shallow angle from dam to powerhouse. Open canals like this run more than thirty miles through the North Umpqua watershed. Each is built to the same specifications: six feet deep with smooth walls lined with gunite, a material that keeps the concrete from crumbling in the constant rush of water. The flow is deceivingly fast, as there is little in the canal to give perspective. The water has nothing to crash against, no roaring display of energy like the chaos of a whitewater rapid. There are no pools, nothing to slow the descent. It’s as frictionless as a river can get, nearly pure momentum, a river turned into a simple physics problem. Nature streamlined. The water is North Umpqua green and cold and eventually it will return loudly to the bedrock and imperfections of the natural river course, the wild and scenic part, downstream.

A small backhoe bucket and arm are permanently mounted above a screen in midcanal to make cleaning more efficient. The canals kill an average of eleven deer each year and an occasional elk, according to PacifiCorp. Small animals aren’t counted, aren’t found. There’s almost no escape once an animal—or person for that matter—falls in the canal. Barkhurst mentions a couple of people who drowned in the canals in recent memory, but from a public safety aspect, the canals still aren’t much of a danger. They’re in remote parts of the national forest out of the way of much recreation activity and are paralleled by rough roads (the road along the Lemolo Canal is nicknamed “the Burma Road”). But the threat is apparent, standing on the edge of the swift, unforgiving current, a gut feeling that says, “Don’t cross.”

The story of the disconnected North Umpqua isn’t just about salmon and steelhead, Barkhurst cautions. As a terrestrial biologist who works the uplands of a world-famous steelhead fishery, she’s had to make a case for hundreds of other species of the North Umpqua under the shadow of the famous fish. Connectivity works on several levels, she says, and this is what the Northwest Forest Plan mandates for the Umpqua National Forest. The forest needs to be hospitable for movement, riparian reserves must be maintained as wildlife corridors, and stepping stones of untrammeled forest, called late-successional reserves, must be maintained throughout the landscape. The license grants PacifiCorp use of more than 2,000 acres—about three square miles—of the Umpqua National Forest, but those are critical acres for the 440-square-mile watershed above the project. Dams prevent movement of fish and the redistribution of rocks and logs and make the former wetlands inhospitable to many species because of fluctuating water levels. The canals divide the terrestrial ecosystem like swift-flowing moats between the biologic oasis of the river and the forest.

The canals parallel the river within the mandated riparian reserve of the Northwest Forest Plan. This reserve, 300 feet on each side of the river or a distance equal to the length of two mature Douglas-fir trees laid end to end, is untouchable under the plan. Federal law requires forest managers to make development in places such as the riparian reserve comply with the forest plan as soon as is practicable. With the FERC license for the North Umpqua project under review, this is the window in which to make the hydropower project comply with the Northwest Forest Plan.

The settlement agreement brokered this past spring provides for more bridges across the canal. Barkhurst says wildlife bridges create bottlenecks in connectivity that some animals may exploit but others may avoid. The best solution for connectivity on land would be covering the canals. PacifiCorp says that doing so would cost $1 million per mile. “We’re so far below what would be minimally acceptable [if the project were constructed today],” Barkhurst says, “it’s mind boggling.”

They keep trying new things, addressing symptoms instead of acknowledging there’s a real problem,” Burns says. After twenty years as a fish biologist for state and federal agencies, he left public service to farm hazelnuts, but he remains passionate about the shortcomings of the North Umpqua project. Burns tends to listen for a long time to what people have to say, and then he bookends conversations with a succinct delivery.

The scientists tick off PacifiCorp proposals about how to make the North Umpqua a healthier watershed. There’s been talk of using helicopters to drop boulders in the river to help gravel collect and a proposal for wider bridges over the canals. Fish ladders would help salmon move upriver, and screens would keep young salmon out of the turbines. The power company trucks in gravel to fortify the spawning bed below Soda Springs Dam. There’s talk of placing logs in the streambed and of provisions for money for the hatchery below the project at Rock Creek. Techno-fixes one and all, the scientists say. But Burns, in his way, had summed it up a few days before in a newspaper editorial.

“This is akin to taking aspirin for headaches caused by an improper eyeglass prescription,” Burns wrote.

The North Umpqua project is an example of what can go wrong with relicensing, the hydropower industry says. The project’s FERC license was granted in 1947 and expired in 1997, and the power company has been working under extensions of that license ever since. The company has commissioned environmental studies and completed a multivolume application—at a cost of $41 million—and still, a decade after starting the process, PacifiCorp doesn’t have a new FERC license. The money and the time could have been spent on fixing problems in the watershed rather than jumping through bureaucratic hoops, hydropower lobbyists say. The industry is trying to persuade Congress to streamline the FERC relicensing process, giving the agency full authority over projects. In the next fifteen years, more than a fourth of FERC hydropower licenses, those post–World War II projects, will expire, and the industry and FERC are trying to relax what could become an expensive bureaucratic bottleneck.

The commission isn’t organized properly to have complete control over relicensing, the General Accounting Office, the investigative arm of Congress, said in a report released this spring. A few weeks later, FERC countered with its own report describing how efficient a streamlined licensing process would be. The relicensing process is “far more

complex, time consuming and costly today than it was when the Commission issued the approximately 1,000 original hydropower licenses thirty to fifty years ago,” the GAO report states. But since 1986, FERC has had to give “equal consideration” to other uses of the watershed besides hydropower, among them environmental rehabilitation. The slate of 1960s and 1970s environmental protection laws require FERC to assess the environmental impact of projects that it grants licenses to, but FERC has neither the will nor the expertise to assess environmental impacts and compel licensees to fix problems. So the task is taken up by other federal agencies with interests in the projects, agencies such as the Forest Service with experts like Barkhurst, Burns, Dose and Jones. This is part of the process the hydropower industry is trying to streamline.

It costs PacifiCorp less than one cent per kilowatt-hour to produce electricity at the North Umpqua project, according to Stanley Vejtasa, a power industry consultant. The company sold that power for a minimum of two cents per kilowatt-hour before May 2000 for a projected annual profit of about $19 million, Vejtasa’s calculations show. Since then, prices have risen substantially, with annual profits from the North Umpqua increasing to the $70 million to $90 million range, Vejtasa projects, based on power prices at the California-Oregon border from October 2000. Removal of Soda Springs Dam would reduce power output on the project by 7 percent, and the North Umpqua project represents less than 1.5 percent of PacifiCorp’s total electric capacity.

The conflict isn’t about the millions of dollars that have and will be spent on the North Umpqua project relicensing. It’s about the future millions that PacifiCorp and other hydropower licensees will have to pay at hundreds of other projects in the coming decades. Tim O’Connor, director of hydro operations for PacifiCorp, wrote a letter to former Umpqua ranger Jim Wieman in 1999 when the power company walked out of negotiations over the issue of dam removal. The Forest Service at that point had been sticking to dam removal as a condition of the new license, saying the connectivity conditions of the Northwest Forest Plan required it. PacifiCorp saw that as a dangerous precedent. “If the [Northwest Forest Plan] is interpreted to require the removal of Soda Springs Dam, it could be used to justify the removal of other hydroelectric projects on or adjacent to national forest lands,” O’Connor wrote. It could be a big problem for the industry.

Negotiations resumed and decision-making authority moved to higher-ranking federal officials. Political pressure increased. The Forest Service’s resolve to remove the dam weakened, waned and disappeared. Science slowly was replaced by politics, Dose says, but the story of the North Umpqua license isn’t over. This case will more than likely end up in the courts when an environmental group sues over NEPA violations. Dose will testify, so in a way he and the other Umpqua scientists likely haven’t spoken their last about the project.

“The reason science isn’t saving salmon is because science isn’t being used to save salmon,” Dose says.