Sediments Threaten Salmon

In the year 2009, the Glines Canyon Dam and the Elwha Dam are scheduled to start being removed from the Elwha River, located in the Olympic Peninsula of Washington State. The sediment that has built up behind these dams

threatens the overall geomorphic condition of the river if released downstream. The Elwha Dam was built in the early 1900’s, 4.9 miles from the mouth of the river. The construction of this dam was finished in 1914. Glines Canyon Dam was completed 8.5 miles further upstream in 1927. Both of these dams impound reservoirs. The Elwha Dam forms the Lake Aldwell reservoir, and the Glines Canyon Dam forms the Lake Mills reservoir. (See Figure 1) The Glines Canyon Dam stands at 210 feet tall. This will be the tallest dam ever removed in the United States to date. Before the dams were built, the Elwha River was used my natives and produced about 380,000 migrating salmon and trout. The construction of Elwha Dam eliminated about ninety-three percent of the Elwha River habitat for these migratory fish, and thus began a very steep decline in the native populations of all ten runs of Elwha salmon and sea-run trout. The 1855 treaties between the Northwest tribes and the United States government guarantee the tribes the right to fish in their accustomed places indefinitely, but the disappearance of the Elwha salmon runs has made this impossible for the Lower Klallam Elwha tribes.

More than 300 dams have been removed in the United States in the last 20 years, but the Elwha River dams are the first to be acquired by the federal government primarily for the purpose of decommissioning, removal and restoring wild salmon. The Elwha River Ecosystem and Fisheries Restoration Act of 1992 authorized the Secretary of the Interior to acquire and remove the Elwha and Glines Canyon dams on the Elwha River to fully restore the ecosystem and native fisheries. The National Park Service completed two Environmental Impact Statements. EIS-1, (Environmental Impact Statement 1) found that both dams, the Glinds Canyon Dam as well as the Elwha Dam, must be removed to meet the goals of the Elwha Act. EIS-2 recommends allowing sediments accumulated within the reservoirs to naturally erode downstream. When the Glines Canyon Dam is to be removed the sediment, which is built up behind the dam and then is released downstream, threatens the ability for salmon to get upstream and spawn naturally.

Figure 1:

Figure 1 map shows the locations of the Glines Canyon Dam, Elwha Dam, and the reservoirs resulting from them in the Olympic National Park.

The fundamental geomorphic change associated with a dam’s presence on or removal from a river is the alteration of the longitudinal profile of the river. Dams create a long, flat-water surface marked by an abrupt drop in elevation at the dam. After a dam is removed, water levels and channel positions more closely resemble the original morphology of the river, and the sediments that had been stored behind the dam are sculpted by the subsequent river flow. This adjustment to a new longitudinal profile can cause major changes in the distributions of aquatic organisms, like salmon and sea-run cutthroat trout.1
One of the major environmental challenges of removing high dams is the height of sediments behind the reservoir. This is less of a concern with low-head dams or dams in wide valleys, because the vertical relief of the low sediment deposits does not create as much potential for abrupt vertical erosion. The elevation of natural floodplains in most rivers is a small, from less than 1% up to 10% of the width of the bankfull river channel. A bankfull is a hydrological measure that generally indicates the height or stage of water that just fills the channel. After a high dam, like the Glines canyon Dam, has been constructed, deposits of sediment upstream of the dam may exceed the relative dimensions of floodplain and bankfull channels found in natural river networks. The removal of a dam with deep sediment deposits may create high, unstable terraces that are accessible to flood waters at the upstream end of the reservoir that existed before the dam’s removal but perched far above the channel at the downstream end. The potential for episodic flood erosion of these high terraces and incision of lateral channels into the terraces complicates the restoration of the river and its floodplain after dam removal. The volume of sediments associated with dams can have major geomorphic and biological consequences for downstream reaches. Removing a dam can release large volumes of sediment to downstream reaches over short periods of time and creates easily eroded floodplains. The timing of sediment release and the downstream extent of sediment deposition are difficult to predict, leading to a high degree of uncertainty about ecological effects. Subsequent erosion of sediment deposits behind the dam results in frequent and complex channel change within the reach upstream of the dam. All this sediment greatly threatens biological life, specifically survival of salmon in the Elwha River.2

The life cycle of a salmon is quite interesting. Adult male and female salmon spawn together in gravel beds of rivers and streams. Using rapid fanning movements of her tail, the female salmon digs out a gravel nest. This nest is called a redd. The male fertilizes the eggs as the female deposits them. The female protects the redd for one to two weeks or as long as she is able before it dies. The fertilized eggs, or embryos, hatch and develop into alevins. An alevin is a newly hatched fish in the larval stage, one which has not yet emerged from the redd. As tiny alevins, they continue to live in the redd. Their food comes from a nutrient rich yolk sac attached to their undersides. The freshly hatched alevins mature into fry. This occurs as the yolk sac is absorbed into the alevin’s body, and the alevin emerges from the redd. Once they become fry, they develop vertical bars called parr marks on their sides that help them remain camouflaged from predators. The parr marks last from a few months to years, depending on the specific species of salmon.3

After a period of feeding and growth in freshwater, the fry begin migrating downstream towards the ocean. The fry become smolts as they migrate downstream towards the ocean. Smolts undergo physiological changes that allow them to adapt to the saltwater conditions in the ocean. One of the most distinctive physical changes is the transformation from a brown color with stripes to a more silvery color that serves as camouflage in the ocean environment. The smolts grow to a fully developed adult salmon as they continue their journey to the ocean habitat. Depending on the species, salmon will spend from one to five years in the ocean and travel hundreds to thousands of miles before returning to the stream where they were hatched. Some fish stray to nearby rivers, colonizing new areas and replenishing weak populations. These adult salmon become spawners themselves. Most salmon spawn only once during their lifetime and die within a week or so of spawning. One very important aspect of these salmon dying is the nutrients from their decomposing bodies help to fertilize the stream. Yet some steelhead and sea-run cutthroat trout can spawn more than once. After these adult fish, called kelts, spawn they migrate back to a saltwater environment.4

Excess sediment present can significantly affect the productivity of a salmon or trout stream.5 In a healthy stream, young salmon and trout hide in the interstitial spaces between cobbles and boulders to avoid predators. In streams that get extremely cold in winter, young steelhead may actually burrow into the streambed and spend the winter in flowing water down within the gravel. The area of the stream where flowing water extends down into the gravel is also extremely important for aquatic invertebrates; which supplies most of the food for young salmon, steelhead and sea-run cutthroat trout. If fine sediment is clogging interstitial spaces between streambed gravel, juvenile salmonids lose their important source of cover and food.

Salmon, steelhead and coastal cutthroat trout are also very vulnerable to sediment pollution because they build their nests in the stream bottom. The eggs, buried one to three feet deep in the gravel redd, rely on a steady flow of clean cold water to deliver oxygen and remove waste products. In coastal streams the eggs usually hatch in about thirty days, depending on the water temperature. Eggs hatch into alevin and remain in the gravel another thirty days or so, living on the nutrients in their yolk sacs. As they develop into fry, the yolk is used up and the fry must emerge through spaces in the gravel to take up life in the stream. During the sixty-day period when eggs and alevin are in the gravel, major shifts of the stream bottom can cause them to die.6

Tappel and Bjornn demonstrated that increased fine sediment in spawning gravels caused decreased survival and emergence of salmonid eggs and alevin.7 (See Figure 2) Nawa and Frissel found fines less than 0.85 mm to have the highest impact on salmonid spawning success.8 Particles of less than 6.4 mm are recognized as having the potential to infiltrate their redds; forming a layer in the stream gravels which sometimes prevents emergence of the fry.9 Kondolf, in a review of this, found that when fines that were less than 6.4 mm exceeded thirty percent, it reduced salmonid emergence and survival by about fifty percent.10

Figure 2:

Figure 2 graph shows that as the percentage of fine-grained sediment in streambeds increases, development of salmon eggs into emergent fry decreases.
[Source: North American Journal of Fisheries Management 3:132]

Studies conducted in actual redds in Olympic Peninsula streams in Washington State found that if there was more than thirteen percent fine sediment, less than 0.85 mm, intruded into the redd then almost no Steelhead or Coho salmon eggs survived.11 It was also noted that fine sediment levels inside and outside Coho salmon redds varied substantially. Fines less than 1.0 mm can sometimes average seven percent inside redds, yet can average thirteen percent outside them and with no inside redd measurement in excess of thirteen percent.

Salmon do have the ability to substantially lower fine sediment in the redd pocket during redd construction. However, if fine sediment levels in the streambed outside the redd are high, there is a potential for fines to intrude into already constructed redds during high flows and high turbidity.12 Because the redd is a depression in the streambed, it creates the Venturi effect, drawing water down into the gravel. The Venturi effect is a special case of fluid or airflow through a tube or pipe with a constriction in it. The fluid must speed up in the restriction, reducing its pressure and producing a partial vacuum. This effect is named after Giovanni Battista Venturi. Fine sediment in suspension during storms may be sucked down into the redd.

Tappel and Bjornn noted that pore space and permeability were key variables in the quality of salmonid spawning gravel, and suggested using the central tendency of particles as a standard.13 They assumed that, because of varying head diameters, Coho salmon have less success emerging as fry than Steelhead or sea-run Cutthroat when fine sediment levels in redds are high. Chapman suggested that measuring permeability itself might be a quicker, and more cost effective method of measuring sediment impacts on salmonids. Chapman concluded that measuring permeability was defined as the variability in spawning gravel quality with better resolution, and also at lower cost than substrate composition analysis.14 The relationship between permeability and salmonid egg survival is not as well known.

Suspended sediment in the water column causes turbidity. Nawa and Frissel found that turbidities as low as twenty-five nephelometric turbidity units (ntu’s) caused a reduction in juvenile steelhead and Coho growth.15 High turbidity during winter impacts the feeding ability of juvenile salmon, steelhead and cutthroat trout. The longer the duration of high turbidity the more damages to fish and other aquatic organisms.16 Measurement of turbidity were taken in excess of twenty-five ntu’s for weeks at a time in Freshwater Creek, located in Humboldt County California, in the winter of 1999.17

Coho and Chinook salmon do not have the leaping ability of Steelhead and are confined to low gradient reaches. These reaches were formerly the most productive spawning and rearing areas, with an abundant supply of good gravel and large wood. High bed load transport can bury low gradient reaches, making them much simpler and less productive salmonid habitat. These formerly productive low gradient reaches become wide and shallow and recovery of fish habitat may take a long time, perhaps decades.18 Lisle noted that recovery of streams with high gradient precedes much more rapidly following large flood events.19

Loss of pool volume has dramatic effects of salmon populations. During the year they spend in freshwater, Coho salmon prefer deep pools that form around large pieces of wood. High sediment transport can fill pools and cause reduction or loss of essential salmonid juvenile rearing habitat.20 Nawa and Frissel noted that optimal Coho habitat is comprised of pools of at least one meter deep, and found that yearling and older steelhead juveniles needed pools at least three feet deep for successful rearing.21

When both the Elwha and the Glinds Canyon Dams are removed, the river will flow freely for the first time is about one hundred years. That is a lot of time that sediments such as silt, sand, and gravels have had to build up behind these dams. In order to determine if river restoration is successful, it is important to have scientific information as to the geomorphic condition of the river before dam removal occurs. The USGS (United States Geological Survey) is developing suspended sediment monitoring systems, studying how sediment may be redistributed, and assessing how dam removal will affect the ecosystem of the Elwha valley. This will greatly help resource managers understand the effects of dam removal on sediment transport, watershed ecology, and aquatic habitat like salmon populations. This information will be essential for examining improvements in salmon habitat that may come with the removal of these dams on the Elwha River.

My hope is that they use the Dredge and Slurry alternative, which is removing fine-grained sediment prior to dam removal by using suction dredges, and sending the slurry to a different location like the Strait of Juan de Fuca. In my opinion, this is the method that needs to be used when considering river morphology. The sediment that is built up behind the Glines Canyon Dam poses a huge threat on salmon habitat and survival. In order to ensure salmon survival after these dams are removed, the Dredge and Slurry alternative seems to be the answer to me. But we will just have to wait and see what the “experts” decide to do. A lot of time and research still needs to be done before the destruction of these dams begin. If the Elwha dam removals succeed, they will provide a really strong example of what a powerful restoration tool dam removal can be.


Behnke, Robert J. 2002. Trout and Salmon of North America. Chanticleer Press Inc. pp 2-10.
Chapman. D.W. 1988. Critical Review of Variables Used to Define Effects of Fines in Redds of Large Salmonids. Transactions of the American Fisheries Society. 117: 1-21.
Gilbert, Francis A. 2003. Restoring the Flow: Undamming of America. Blockwells Pub Inc. pp 32-47
Kondolf, G.M. 2000. Assessing Salmonid Spawning Gravel Quality. Transactions of the American Fisheries Society. 129:262-281.
Lisle, T.E., and J. Lewis. 1992. Effects of sediment transport on survival of Salmonid embryos in a natural stream: A simulation approach. Canadian Journal of Fisheries and Aquatic Science. 49: 2337-23344.
McHenry, M.L., D.C. Morrill and E. Currence. 1994. Spawning Gravel Quality, Watershed Characteristics and Early Life History Survival of Coho Salmon and Steelhead in Five North Olympic Peninsula Watersheds. Lower Elwha S'Klallam Tribe, Port Angeles, WA. & Makah Tribe, Neah Bay, WA. Washington State Department of Ecology.
Nawa, R.K. and C.A. Frissell. 1993. Measuring scour and fill of gravel stream beds with scour chains and sliding bead monitors. North American Journal of Fisheries Management. 13: 634-639.
Newcombe, C.P. and D.D. MacDonald. 1991. Effects of Suspended Sediments on Aquatic Ecosystems. North American Journal of Fisheries Management. 11: 72-82.
Tappel, P.D., and T.C. Bjornn. 1983. Methods of relating size of spawning gravel to salmonid embryo survival. North American Journal of Fisheries Management 3:123-135.

Related Essays Geology