Dam Removal: A Type of River Restoration?

Kathryn Brennan: Environmental Science

Ashley Busold: Geology and Biology

Mallary Rocheleau: Animal Science

The life of a salmon is unlike many other fish species, for it must swim through thousands of miles of ocean in search of the river where it was born.  The mission is to bring about the next generation in the same place that this salmon, along with many others, were produced. After the many months of migrating through the ocean, with only a few miles left to travel against the river current, there is a road block in the form of a giant wall. The salmon will search up and down looking for a way past without success, but never give up because to reproduce is the top priority. However, salmon can only stay in one place for a certain amount of time before either needing food, or being food. At this roadblock, all of the salmon that have travelled here are being wiped out by eager fisherman or in competition for food.  With no hope of ushering in the next generation, this species is on the path to extinction.

Salmon are anadromous and spend most of their life in the ocean, except when they need to breed, and return back to their birth stream or river. They then either return to the ocean or become a key food source for nearby predators such as bears, raccoons, and eagles. The entire surrounding ecosystem obtain nutrients from these fish. When dams are in place, salmon and other migrating species are unable to spawn and their populations decrease. This decrease results in less food for predators and the surrounding plants lose a key source of nutrients. Grable, the author of a magazine article titled “After the Flood”, talks about the beauty of the newly restored vegetation following a dam removal in southern Oregon:

Three years ago this area — where the Gold Ray Dam once stood — resembled a drained mud wallow… the willows are a positive sign. Streamside natives like willow and cottonwood keep invasive species at bay. The new vegetation catches sediments, dissipates energy during high-water events, and provides a home for birds and insects. (Grable, 2014, para. 1)

The picture that she paints of life rising out of destruction is one of pure wonder. A dam removal is extremely beneficial for every aspect of the environment and dams cause serious ecosystem downfalls (Grable, 2014).

However difficult the life of a salmon, and many other organisms in these rivers, humans do not want to consider the impact of dams on the environment. Many people do not realize such a large disruption of water flow greatly impacts the organisms living there. They would rather think about their personal gains, even if these dams are sometimes also a costly danger to nearby communities. Many dams outlive their design and become unsafe. The U.S. Army Corps of Engineers conducted an Inventory of Dams and found 85% of the 83,000 dams in the United States will “near the end of their design life” (Bennett, et. al, 2013, p. 26) by the year 2020 (Bennett, et. al, 2013). The American Association of State Dam Safety Officials (ASDSO) also collected data from 2001 to 2007 and found high hazard dams in need of repair increased from 488 to 1826. They also found an increase of 1348 to 4095 for total deficient dams. The ASDSO also reports “dam failures … increased more than 175% from 2000 to 2006” (Wildman, 2013, p. 8). Dams which need repair, and dams actually in the process of repair possess a significant and increasing gap between them. These dams put growing downstream urban areas at risk. These dams are also costly to repair and maintain. In 2009, the ASDO estimated the total cost to repair United States dams at $50 billion and estimated an additional $16 billion for repair in potential high-hazard dams (Wildman, 2013). However, despite the fact many dams outlive their purpose and harm native fish populations, some of the public and the government are still unwilling to remove them.

When the government decided in the 1980s on relicensing two hydroelectric dams in Washington’s Elwha River, the idea of removing the dams seemed ludicrous (Gowan, Stephenson & Shabman, 2005). The Federal Energy Regulatory Commission (FERC) became displeased about the idea for relicensing to a dam removal project once the old license expired. In a journal article titled “The Role of Ecosystem Valuation in Environmental Decision Making: Hydropower Relicensing and Dam Removal on the Elwha River”, the authors state:

When relicensing proceedings for the two dams began in earnest in the mid 1980’s, removing the dams to restore wild anadromous fish runs was not considered… FERC begins relicensing proceedings by developing and evaluating mitigation alternatives assuming the dams remain in place. (Gowan et. al., 2005, sec. 2.2)

The government was displeased with losing two good sources of hydroelectricity, and at the time completely ignored the dam removal strategy. They felt as though keeping the dams and doing slight mitigation might, at least, show some ecosystem benefits.

However, other studies suggest a better alternative than keeping all of the dams running while doing minor remediation on the Elwha River. One example comes from an article titled “A Multi Objective Optimization Model for Dam Removal: An Example Trading off Salmon Passage with Hydropower and Water Storage in the Willamette Basin”. The study analyzes all dams in the basin and if removal of specific dams will create any negative impacts on hydropower. The Willamette Basin is located in Oregon, and the basin’s close proximity to the Elwha dams allows for comparison. A model created in the study shows promising results. The authors of the article state “There were twelve dams removed in the solution in which 52% of the upstream drainage is reconnected and only 1.6% of hydropower and water storage is lost” (Kuby, Fagan, ReVelle & Graf, 2005, “Basic Results”). Much of the main river system reconnected and resembles its previous natural state, with barely any production of power lost. When comparing this study to the Elwha River, it shows a decent tradeoff with only removing one of the dams while keeping the other. An additional article shows optimistic results through a study model also located on the West Coast. The California Central Valley, whose river flow reaches to the Sacramento-San Joaquin Bay Delta and the ocean, contains multiple dam systems. These dams disrupt anadromous salmon species from getting back and forth from freshwater and saltwater. One proposal, to only remove some dams, makes it easier for fish to travel and reproduce while also still creating hydropower. Also, the article titled “Optimizing the Dammed” studies the costs and benefits of salmon populations versus hydropower when removing certain dams in the valley. Null, Medellin-Azuara, Escriva-Bou, Lent, and Lund (2014) conclude the removal of certain dams creates little impact on the decrease of hydropower, and would also improve migrating conditions for anadromous species. With support from this study being considered, removal of only one dam on the Elwha River may lead to similar outcomes.

Unfortunately, many people believe these salmon can get by a dam with a little help. Some common techniques to help them across are fish screens, catch and haul, and fish ladders. Some believe these techniques are a simple solution to the competition between restoring fish habitat and keeping dams. In the article “The Role of Ecosystem Valuation in Environmental Decision Making”, the Fisheries Assistance Office (FAO) submitted a report which recommends trap and haul facilities to physically remove fish from the lower dam and bring them above the dam, and fish screens help fish above the dam get below the dam (Gowan et al., 2005, sec. 2.2). To many people, these seem like reasonable suggestions to help salmon populations rebound.

On the contrary, these practices to help fish cross dams are not as helpful as studies suggest. A study conducted through the article “The Role of Ecosystem Valuation in Environmental Decision Making”, shows the implementation of fish passage aids were not helpful to most species of salmon. Gowan et al. (2005) talk about a solution through a study the James River Corporation did on salmon passage aids. The corporation concluded only three species of salmon, the chinook, coho, and steelhead, were able to increase their populations with the help of nets and catch and haul. The others were not physically capable even with the aids (Gowan et al., 2005, sec 2.2). The study shows only a portion of salmon species are able to reproduce, and many of them cannot survive to do so. Another study, from the article “Upstream Passage Problems for Wild Atlantic Salmon (Salmo salar L.) in a Regulated River and its Effects on the Population”, tagged all anadromous salmon species coming from the Atlantic going up the River Umealvin in Sweden to breed. The study hoped to find an increase in fish getting over the dam through the implementation of fish ladders versus without the ladders. However, the results show otherwise. The authors state, “the probability of wild salmon successfully migrating through the regulated part of the River Umealvin from the estuary is low, with average losses of 70% of potential salmon spawners” (Lundqvist, Rivinoja, Leonardsson & McKinnell, 2008, “Discussion”). With only 30% of survival for spawning, the salmon populations are sure to decrease even with the help of ladders. Hatcheries are often used as another solution for endangered anadromous fish species such as salmon who lost their habitat, but they can cause a lot of problems for fish as well. These problems include lowered fitness, susceptibility to disease, and altered run timing (Null et al., 2014).

Along with their effects on anadromous fish, dams also alter the physical features of a river. When dams are built, the flow, turbidity, depth, and temperature of the river changes above and below the dam. One example was observed in the Colorado River. After multiple dams were built, the river showed slower flow and less turbidity. Sometimes no flow is seen at all in lower locations and in fact, the river no longer reaches the Pacific Ocean. This impacts native fish populations negatively. Some native fish only thrive and reproduce in fast flowing and turbid water. Sudden changes in flow, turbidity, and depth cause stress and increase the risk of illness in the native fish, and prevent them from successfully spawning (Courtenay & Robins, 1989). When dams release water for power production, they create an unnatural timing of flow change and alter the temperature of the water in a short amount of time. The released water flows out from the bottom of the reservoir behind the dam, which contains colder, low oxygenated water due to depth. These sudden changes also put stress on native fish and often cause confusion with emergence and growth cues because these changes in flow and temperature usually only occur during certain seasons (Bednarek, 2001).

Dams also create habitat for non-native and invasive species. Large reservoirs often form behind dams, sometimes this is done purposely to create a source of drinking water or recreational areas for communities. Invasive fish species thrive in these reservoir ecosystems and are often introduced, either accidentally or purposely, for recreation such as fishing. This can be devastating for native fish. These non-native fish compete with native fish for food and resources, and consume their eggs and juveniles (Courtenay & Robins, 1989). However, Bednarek (2001) mentions that when the Woolen Mills Dam in Wisconsin was removed, large populations of invasive species like carp who prefer slow moving water decreased in number, and native species returned. Removing some of these dams may help reverse much of the damage caused by dams built there in the first place.

In order to fix the mentioned problems caused by dams, some being faulty and outdated, we propose removing dams that meet certain criteria, based on several different research methods.  Kibler, Tullos, and Kondolf (2011) go into detail about how proper research is needed before and after a dam is removed.  The different study methods are variations of the Before-After (BA) model.  The three types of BA research methods are called Before-After- Control-Impact (BACI), Before-After-Control-Intervention-Paired Sampling (BACIPS), and Multiple-Before-After-Control-Impact (MBACI).  The authors mention these methods as “one of the simplest diachronous approaches for dam removal studies … widely used to describe channel and ecological changes following restoration” (Kibler et al., 2011, p. 969).  These methods are utilized to determine if the ecosystem is restored by removing the dam along with whether or not the benefits are worth the time and effort.

One model used to “remove dams systematically and assess [eco] system response” (Null et al., 2014, p. 123) is the CALifornia Value Integrated Network (CALVIN).  Null et al. (2014) performed 19 model runs using CALVIN to model how removing different dams in California would affect the surrounding area.  The authors describe the model runs as “not … imply[ing] that removing all dams is worthwhile, but rather to prioritize for potential removal [of] those that have low economic benefit and large gains in upstream fish habitat” (Null et. al., 2014, p. 123).   Null et al. (2014) also found that the increasing warm and dry climate in California creates more water storage space and “removing dams to increase habitat for anadromous species may be increasingly feasible in the future” (p. 130).

Zheng and Hobbs (2013) use the Multiobjective Portfolio Analysis (MPA) approach to assess dam safety near urban areas and the overall financial investment of the dam(s). This second model takes into consideration the risk factors to the stakeholders, ecological benefits, and the financial burden of fixing the dam versus removing it. Some dams exist where the cost of maintenance is higher than the cost of removal (Zheng & Hobbs, 2013, p.65).  Zheng and Hobbs (2013) use previously studied dam portfolios and apply the MPA to them.  These study portfolios encompasses several watersheds and the dams in them and does further data analysis to find the safety risks involved with the dams to add to the already collected data relating to the ecological benefits and the overall cost of removing the dam(s) (Zheng & Hobbs, 2013, p. 66).

By using said research methods to evaluate and remove certain dams, migrating fish will return to their breeding grounds and river ecosystems will return to their natural state. Rivers will slowly return to their natural depth, flow, and turbidity. Native fish will gain access to the rest of the river and its tributaries. Restoring the river will cause invasive and non-native species that only thrive in reservoir systems to drop in population; additionally, native plants and trees will also return to the riverbanks and surrounding area. As native fish populations and plant life grow upstream and downstream, the entire surrounding ecosystem, and humans, will all benefit.


Bednarek, A.T. (2001). Undamming rivers: A review of the ecological impacts of dam removal. Environmental Management, 27(6), 803-814. doi: 10.1007/s002670010189

Bennett, S.J., Dunbar, J.A., Rhoton, F.E., Allen, P.M., Bigham, J.M., …, & Wren, D.G. (2013). Assessing sedimentation issues within aging flood-control reservoirs. In J.V. De Graff & J.E. Evans (Eds.), The challenges of dam removal and river restoration (pp. 25-44). Boulder, CO: The Geological Society of America, Inc.

Courtenay Jr., W.R. & Robins, C.R. (1989). Fish Introductions: Good management, mismanagement, or no management?. Aquatic Sciences, 1(1), 159-172. Retrieved from http://www.nativefishlab.net/library/textpdf/17630.pdf

Gowan, C., Stephenson, K., Leonard, S. (2005). The role of ecosystem valuation in environmental decision making: Hydropower relicensing and dam removal on the Elwha River. Ecological Economics, 56 (4), 508-523. doi:10.1016/j.ecolecon.2005.03.018.

Grable, J. (2014). After the flood. Earth Island Journal, 28(4). Retrieved from http://www.earthisland.org/journal/index.php/eij/article/after_the_flood/

Kibler, K. M., Tullos, D. D., & Kondolf, G. M. (2011). Learning from dam removal monitoring: Challenges to selecting experimental design and establishing significance of outcomes. River Research and Applications, 27(8), 967-975. doi: 10.1002/rra.1415

Kuby, M.J., Fagan, W.F., ReVelle, C.S., & Graf, W.L., (2005). A multiobjective optimization model for dam removal: an example trading off salmon passage with hydropower water storage in the Willamette Basin. Advances in Water Resources, 28 (8), 845-855. doi:10.1016/j.advwatres.2004.12.015.

Lundqvist, H., Rivinoja, P.,Leonardsson,K., & McKinnell, S. (2008). Upstream passage problems for wild Atlantic salmon (Salmo salar L.) in a regulated river and its effects on the population. Hydrobiologia, (602), 111-127. doi:10.1007/s10750-008-9282-7

Null, S., Medellin-Azuara, J., Escriva-Bou, A., Lent, M. & Lund, J. (2014). Optimizing the dammed: Water supply losses and fish habitat gains from dam removal in California. Journal of Environmental Management, 136, 121-131. doi: 10.1016/j.jenvman.2014.01.024.

Wildman, L. (2013). Dam removal: A history of decision points. In J.V. De Graff & J.E. Evans (Eds.), The challenges of dam removal and river restoration (pp. 1-10). Boulder, CO: The Geological Society of America, Inc.

Zheng, P.Q. & Hobbs, B.F. (2013). Multiobjective portfolio analysis of dam removals addressing dam safety, fish populations, and cost. Journal of Water Resources Planning and Management, 139(1), 65-75. doi: 10.1061/(ASCE)WR.1943-5452.0000209