Before the 1970s energy crisis led to conservation measures, the per capita consumption of electricity in the Pacific Northwest was three times the national average. That energy is largely hydropower. Today, the world's technology feeds off the Columbia. Amazon, Google, and Microsoft gobble up the river and take it away from salmon. Huge air-conditioned spaces with energy-hungry servers move email, power cell phones, and serve the internet. In writing this book on a computer, I am probably using the water of the far-away Columbia.
Two gillnetting tribes, the Swinomish and the Upper Skagit, unable to get the dam on the Baker River removed, found another way to save their fish. Their fish, Baker Lake sockeye, known for their dark red meat and bellies striped with white fat, are considered one of the best sockeye in the world. The fish spawn in a 600-foot-deep lake that is fed by the glaciers of Mount Baker. But the run—the last native Washington sockeye in Puget Sound—was slowly dying off, unable to pass a hydroelectric dam in the river.
By the year 2000 there were only ninety-eight fish left, and the tribes decided to act. They captured the fish struggling below the dam and trucked them to Baker Lake for natural spawning. Once the fry grew to smolts ready for the sea, they trucked them back below the dam. When this did not save enough salmon, they started a hatchery. In 2016 there were 56,000 Baker Lake sockeye. The tribes estimate the fish will become an important income source when they get to 120,000.
Trucking is one of several ideas that have been tried to get salmon past dams. There was also a device called a "fish cannon," a large water-pressured hose that shot fish over a dam. Not surprisingly, like fish passes and ladders, none of these newer solutions are as effective as an unobstructed river.
The Pacific Northwest, and most of the salmon world, has written laws requiring dams to have fish passes. Ladders aid returning adults, but exiting juveniles need more. For thirty years scientists have studied how dams could be made more passable for juveniles. Pass designs have been improving lately, built to resemble nature by constructing ramps from rocks or digging a passageway (as was done on the Penobscot). Scientists have found that the juveniles like to swim near the surface and so can escape in spillways at the top of a dam. Of course, like ladders, this leaves them vulnerable to predators—birds, fish, and mammals that learn to wait for them there.
In the 1960s the Army Corps of Engineers began using trucks, barges, and even airplanes as ways to transport fish around dams. Originally the approach was just viewed as experimental, but it became so popular with the public that it was permanently written into the Corps' operating budget. The Corps now has a substantial flotilla for moving smolts around dams, and as many as 30 million young salmon are hauled downstream every year. But results have not been positive. These smolts do very well getting to sea but many do not come back. Marine-return rates of barged or trucked salmon are far lower than for natural runs. The problem is not only survival but homing. Barged salmon do not seem to know where they are to spawn, and frequently stray. Barging on the Snake River has failed to rebuild that stock and, given the vagaries of both trucking and hatcheries, the long-term prospects for Baker Lake salmon are not clear. Human inventiveness keeps proving inadequate for replacing the natural order.
The growth of the renewable energy industry has weakened the argument that hydroelectric power is the only alternative to burning fossil fuels or resorting to nuclear power. But dam removal, even when the owners agree, is often unpopular with locals, who fear it will increase the cost of electricity or spell the end of local industry. In 1994 the Oregon Water Resources Commission, realizing that building irrigation pumps to replace the dam was cheaper than building fish ladders, voted to remove the Savage Rapids Dam on the Rogue River. They voted for removing it twice and the legislature, urged by the local citizenry, rejected it twice.
But dams are being removed. According to American Rivers, a river conservation group, more than 450 dams have been removed since 1999. In 2017 alone, 86 dams were removed. Most of these dams do not involve salmon runs, but a few do.
In Oregon the removal of the Beeson-Robison Diversion Dam on Wagner Creek, a tributary of the Rogue, is expected to help both juvenile passage of steelheads to the sea and their return for spawning. It is also hoped that it will improve the coho run.
Also on the Rogue, in 2010 the Gold Ray Dam was removed—both the original wooden 1904 structure and the 1941 thirty-eight-foot-tall concrete one. It was a hydroelectric dam, but once it was taken off the grid in 1972, it no longer served a purpose. Its removal opened up 333 miles of spawning grounds.
The Hood River, which flows into the Columbia in Oregon, had its Powerdale Dam removed in 2010. This was an obsolete 1923 hydroelectric dam, and its removal opened many miles to salmon passage.
In July 2017, also in Oregon, the Dillon Diversion Dam, a fifteen-foot-high structure built in 1915, was removed. The project is aimed at improving irrigation while also opening habitat for steelhead, kings, and coho, as well as Pacific lamprey. Pacific lamprey, also an anadromous species, is native to the Pacific coast from Mexico to Japan. In recent years this fish, important to the food and culture of Native Americans, especially in the Pacific Northwest, has been dramatically declining, in part because of dams.
In California the Albion River in Mendocino County, struggling since the lumber industry established itself there in the 1850s, has new hope since the ten-foot-high Glenbrook Gulch Dam was torn down to allow the passage of steelhead and coho.
In April 2016 the governors of California and Oregon, the US Secretary of the Interior, NOAA, and Pacific Power agreed to remove four hydroelectric dams on the Klamath River. As with most modern dam-removal agreements, the deal is complex. For the utility company, it avoids the costly obligation of building modern fish passes. But the deal also offers benefits to farmers and ranchers and specifies the undertaking of habitat restoration as well. If successful, this could restore what once was one of the greatest salmon and steelhead rivers of the Pacific Northwest.
One of the most celebrated dam removals is on Washington's Elwha River. Beginning in the Olympic Mountains and emptying into the Strait of Juan de Fuca between Washington and Canada, this river is only forty-five miles long. But with eight tributaries and a 321-mile watershed in mostly unspoiled wilderness, the Elwha once had an annual run of 8,000 king. It also had sizeable runs of sockeye, chum, pinks, and coho. The runs of pink salmon were an estimated 275,000 annually. Since 1911, when a hydroelectric dam was built across a narrow gorge near the mouth of the river, Elwha salmon of all species had nearly disappeared. The dam was in violation of an 1890 state law that required any dam blocking fish runs of any kind to provide adequate passage. But in turn-of-the-century Washington State, laws protecting fish were often ignored, especially if they interfered with entrepreneurs trying to make money. It was called economic development.
The Elwha Dam was completed in 1912 with no fish passage. In an extraordinary example of how hatcheries are used to avoid true conservation measures, hatcheries began to be built to compensate for the vanishing salmon. The legislature even rewrote the law to allow hatcheries as acceptable substitutes for fish passages. But the hatcheries all failed.
A second dam, also without fish passages, the Glines Canyon Dam, was completed eight miles farther upriver in 1927. It had a stylish curve and was generally considered a well-designed and attractive piece of engineering. The two dams produced enough electricity to power 15,000 average homes. But the Olympic Peninsula only had a protected National Park and one small tourist town, Port Angeles. In 1992 Congress authorized the Secretary of the Interior to acquire both dams and remove them if it was determined that this was necessary for the restoration of the river.
Clearly it was. But that only began a debate that lasted for nearly two decades. Legally, politically, and economically it is no small matter to tear down huge concrete structures that generate electricity such as the 108-foot-tall concrete Elwha and the 210-foot-high Glines Canyon. Dams this large had never been removed. The major opponent was one pulp mill, the dams' only remaining customer. (The energy had also been used for lumber mills, built over land sacred to the Klallam that damaged the countryside, but this activity had stopped.) The area was now mostly powered from the Bonneville Dam.
The $325 million to proceed was not approved until almost twenty years later, during the Obama presidency. But the dam removal was only the first step because the reservoir the dam created had destroyed gravel beds and filled the river bottom with sediment. In fact, removing the dam itself did not represent the majority of the cost of the project. A riverbed below the dam suitable for salmon spawning had to be rebuilt. The work on the Elwha River is the largest dam-removal project in American history, though the Klamath project could end up even bigger.
Once the dams came down in 2011, the formerly clear water turned mud brown. So much sand poured out of the rivers that a delta with a recreational beach has been created in the expanded coastline. Only months after the last of the concrete was removed, steelhead entered the river to spawn above the dams for the first time in a century. Since then king, coho, and chum have entered the river.
The problem is that the $325 million recovery program included the building of hatcheries. And so, a debate begins as to whether a hatchery will enhance, damage, or accomplish nothing for this river. Historically, hatcheries have usually not worked. In a few cases where habitat was restored and no stock spawned there anymore, hatcheries have brought back a salmon run. There are also rivers like the Mersey, where once the habitat was restored strays from the area repopulated the river. If successful, this would be the most natural and the most likely strategy for longterm success. But some people regard it as too much of a long shot and cannot shake off the notion that fish can be made.
And there are certain to be other problems. The State wants both commercial and recreational fishing, which are usually in conflict, and the Klallam tribe wants their fishing rights restored.
The Elwha demonstrates that to restore the Pacific Northwest, dam removal is only the beginning. The habitat must be restored, and that means rebuilding forests and riverbanks.
Hydroelectric dams were part of a strategy to develop human activity at the expense of nature. And this idea still has its supporters. Many people may not want to live in a world of dangerous bears and pesky insects. Species in California, Oregon, and Washington figure prominently on lists of endangered and declining insects. Many of these insects—such as stoneflies, which are vanishing along with wild riverbanks—are prime food for salmon and trout. It is not only the tall grasses and bushes of pesticide-free banks that need to be restored, but also the forests that surround, or used to surround, the rivers.
Sergei Korostelev, a leading fish biologist on Russia's Kamchatka Peninsula, recently visited Washington State and was shown rivers such as the Columbia, where Washington was hoping to restore king salmon stocks. "I was taken to a beautiful green area, but I noticed there were no mosquitoes, no insects. Kings live in a river for years and with no insects, there would be nothing to eat. Why would a king return to a river with no food?" he asked.
While the idea of habitat restoration is not nearly as old as habitat destruction, some projects have been pursued for more than a century. Ironically, one of the oldest is dam building: not concrete walls that block rivers but small dams with sluices that can protect salmon from unpredictable floods and droughts. In the Scottish Highlands such a dam was built in 1900 on the River Helmsdale and another, still in service, in the 1950s. The one in use on the River Thurso was built in 1907.
In British Columbia projects were undertaken to restore rivers that had silted over by uncovering and cleaning gravel. The Nooksack in Washington is a wide and wild river with strong currents, but salmon need the pools and calm pockets created by log jams that do not naturally exist anymore. Rivers like this constantly change, so a spawning salmon does not always manage to find the place of its birth, or discovers that its original nesting spot is no longer suitable for a redd. The salmon will find a new place if there are suitable spots. The river has had sockeye, chum, coho, and pink, but for the moment the focus is on bringing back king salmon.
There have been small projects for the past ten years, but now a huge habitat restoration project is being undertaking by the Nooksack tribe with Washington State funding. They are constructing side channels and Engineered Log Jams (ELJ), which create the pools and eddies once made by fallen trees. Deep holes are dug and thick logs sunk vertically in the riverbed with other logs chained to them. Every log jam engineer has a different log jam design for different locations. On one wide crook of the fast, gray, cold Nooksack, 3,000 logs were being placed. The State of Washington is spending millions of dollars building strong, stable, artificial log jams.