Skip to main content

Oregon Dam Removal: A Sustainable Choice

The decision to remove the Marmot Dam was not taken lightly. Great care and study was undertaken to understand the ramifications of how the Sandy river system would be affected. Because river systems play important roles in ecosystem services, it was important to make sure systems like fish habitat, nutrient cycling, or regulating flood waters would not be adversely affected by the removal of the dam.

By analyzing these systems, and following the cause and effect of changes in the system, we are able to more easily see the benefits that the system achieves from dam removal. This requires outlining the areas that can be changed by dam removal and assigning monetary values to those specific systems. When an accurate understanding of the value of the system is obtained, a management decision can be more easily justified by comparing those ecosystem service values to the cost of removing the dam. In the case of the Marmot Dam, the value of the ecosystem services far outweighed the cost of removing the dam.


The issues of the Marmot Dam were wholly created by human intervention of the river. The Marmot Dam, part of the greater Bull Run Hydroelectric Project, was built in 1913 and served to divert water into Lake Roslyn. Another dam held in Lake Roslyn with a powerhouse generating electricity (Fogarty, 2007). The growing need for electricity, as well as the lack of sustainable sources, led to the creation of hydropower projects like this one. These circumstances persuaded engineers to disregard the negative environmental consequences that came from dam construction. In fact, a dam was considered to be environmentally friendly, with few drawbacks. It is only due to modern developments in power generation, and regulations implemented to restore river ecology, that the cost benefit of removing the dam outweighs the costs of maintaining it.

In 1913 the Marmot Dam was built on the Sandy river in Oregon. The Marmot Dam has undoubtedly impacted the environment in a negative way. While in place, dams are known to dramatically change ecological systems. For example, dams have been shown to have a great deal of responsibility “for a loss of 80% of the salmon and steelhead population since the 1950s, 90% of delta smelt, 96% of Pacific Flyway wetlands, 89% of riparian woodlands and 95% of spawning habitat for spring-run salmon” (Downs, 2009). In addition, the dam created a large reservoir that changed the river environment upstream. Reservoir water can commonly become thermally stratified. The hypolimnion layer, which is usually deoxygenated and stagnant, can be taken up into the dam turbines and discharged. This can disrupt the ecosystems downstream (, 2016). After careful analysis by Portland General Electric, the company that owned the dam, the Marmot Dam was removed in 2007. It was determined that the cost of continued maintenance was going to be more than what the dam could provide (Fogarty, 2007). After the dam was removed, approximately 730,000 square meters of silt that had built up behind the dam was carried downstream by the river (Keith, 2012). These factors; the reservoir, the dam itself, and the downstream effects during and after the dam, all contribute to the negative impact the Marmot Dam has had on the Sandy River.

Because the cost of removing the dam was substantial, it is important to understand what the ecological gains were to the Sandy river, if any. In the case of a small dam like the Marmot, Downs (2009) tells us, “there can be economic benefit to water resource providers of removing the dam rather than continuing to pay for its maintenance. This is frequently the case for older, smaller, and privately owned dams” (p. 434). This can be done by evaluating the ecological services, and the changes that have occurred. These sorts of actions are important because “managers need to prioritize actions given their budgetary constraints … [and] environmental managers should be aware of trade-offs existing between different ecosystem services” (Acuña, 2013 p. 989). By evaluating the ecosystem services before the dam was removed, we will be able to establish a baseline of value that the river was able to provide. We will contrast this to an evaluation of the river system post dam removal, so a determination of whether removal was cost effective or not.

Let Your Voice Be Heard

System Description

Rivers play a vital role as a supporting service for our ecosystems. According to the journal Biogeochemistry, “river networks regulate nutrient export from the terrestrial landscape to receiving waters” (Reisinger, Tank, Rosi-Marshall, Hall, Baker, 2015). This tells us that rivers are helping to balance any excess of nutrients from one area, and moving them to another area. This “retention and transformation of matter and the cycling of nutrients … are among the most important ecosystem services provided by rivers” (Acuña et al., 2013). The unique features of the river environment and the distances they cover make rivers essential to nutrient flow throughout an entire watershed (Reisinger, Tank, Rosi-Marshall, Hall, Baker, 2015).

Rivers also act as a provisioning service and cultural service by serving as fish habitat. The dynamic nature of a river allows for the variety of environmental conditions that are needed for healthy fish populations. In the case of the Sandy river, the necessary conditions for wild salmon and trout. For instance, “[fallen trees] create pools and cover for aquatic organisms, stabilizes stream banks, and fosters habitat diversity … pools serve a variety of functions for many salmon species and life stages, particularly for juveniles” (Larsen, Kaufmann, Kincaid, Urquhart, 2004). These pools aid in upstream travel, resting locations, and spawning. Salmon and trout also prefer spawning in water with velocity, which make rivers the ideal habitat (Armstrong, Kemp, Kennedy, Ladle, Milner, 2003). Fish can be a source of food for people as well as recreation for sport fishing.

The Sandy river is a cultural service as well because of the natural beauty that is enjoyed by local residents and the recreation that occurs on and around the river. The Sandy river has roughly twenty miles of riverfront classified as scenic and recreational (, 2016). There are 8 separate public parks on the Sandy river; each getting regular traffic year round (, 2016). The sandy river also provides for kayaking, tube floating, and boating (, 2016).

The Sandy River system is also a regulating service. Rivers can mitigate flood conditions which can prevent ecosystem damage and property damage (Kumar, 2010). Rivers also regulate the ability of the soil to absorb nutrients (Kumar, 2010).

According to Amoros and Bornette (2002), an abiotic contributor is the distinct morphology which affects rate as well as distribution of nutrient processing within a river. The journal, Hydrobiologia, tells us that the “highest potential denitrification rates occur during warm water conditions in non-winter months” (Houser & Richardson, 2010). This identifies temperature as an abiotic contributor too. In addition, Houser and Richardson (2010) reports that “denitrification is the set of metabolic reactions performed by a large suite of facultatively anaerobic bacteria”, which would be biotic factors. Organic sediments can come from sources such as river algae, phytoplankton, and various plant material (Houser & Richardson, 2010). These would be classified as biotic components. According to Houser and Richardson (2010), regulation of phosphorus concentrations correlated to chlorophyll concentrations, while at other times phosphorus was regulated by the process of phosphorus equilibrium. Here we see that biotic and abiotic processes are occurring. The velocity of water has been found to be the dominant indicator for salmon spawning (Shirvell & Dungey, 1983). Salmon can spawn in a variety of depths, but have been found to prefer a water depth in excess of 20cm (Armstrong, Kemp, Kennedy, Ladle, & Milner, 2003). These both indicate the water in the river environment contributes heavily to where a salmon will reproduce; this is an abiotic component. According to a study of salmon substrate spawning preference, salmon prefer gravel that has a diameter that is at most 10% of the salmon length (Kondolf & Wolman, 1993). This is an abiotic factor. Salmon spawning habitat is also affected highly by what is referred to as “fine material”. Fine material is an abiotic component to this system. Water temperature is also an abiotic factor for this system. (Crisp, 1993). According to Utz et al. (2012), the primary diet of salmon consists of insects and insect larvae, which are biotic factors. Geography such as cliff edges, winding, river rapids, and distance from populated areas; factors such as air quality, number of sunny days per year, and temperature; these are abiotic. Biotic aspects include interesting wildlife such as deer, squirrels, birds; salmon and trout for fishing; and a lack of pests such as mosquitoes. Biotic features of value also include a variety of plant life such as trees, bushes, flowers and grasses. According to Kerr and Swaffield (2012), “high value environments had taller exotic trees shading the water’s edge.”

All of these biotic and abiotic factor work together to perform the mentioned ecosystem services. There are the abiotic factors of river morphology and temperature, which work with biotic factors such as plant and animal material, as well as anaerobic bacteria. Fluvial geomorphology contributes by determining where buildup of nutrients and materials occurs. Temperature can affect these processes due to greater river flow during warmer temperatures due to snow pack melt from tributaries. Temperature also affects the rate at which anaerobic bacteria are able to break down material. This process can be limited in lower temperatures. The influence of biotic materials such as river algae, phytoplankton, and various plant material contribute to an ever replenishing balance of material on which the chemical reactions vital to nutrient cycling can occur. Salmon will choose spawning areas based on the amount of silt in the gravel. This silt will be determined by upstream environment, and the water conditions that bring it down river. Biotic factors like tree cover for spawning sites will also be affected by the amount of water the river is providing. Less available water will lead to drier conditions that could result in fewer trees or bushes. This plant life can also be affected by the nutrient cycling which is occurring. The amount of nutrients that are able to make it into the soil will have a direct result in how productive the plant life is able to be. Salmon and trout also need adequate food sources in the form of insects and smaller fish. Mayflies and other fish species will eat plant life. This plant life is sustained by the nutrient cycling service which occurs in rivers. From these examples we can see that the provisioning and cultural service of fish habitat is dependent on the balance of other biotic and abiotic factors, even other ecosystem services, to maintain healthy habitat.

The policy decision to remove the dam leads to the dam being removed. The Dam being removed will have a balancing effect on the water depth throughout the system, balance water velocity throughout the system, change the river morphology, and balance the distribution of gravel and silt throughout the system. More consistent depth leads to increased recreational kayaking and tubing on the river, more areas for fish habitat where water level has lowered, and more area for algae to grow where water level has risen. Consistent velocity will increase kayaking and tubing, and contribute to an increase in oxygenated sediment throughout the system. Increased river morphology will also contribute to more oxygenated sediment. A balance of gravel and silt throughout the system will also lead to more oxygenated sediment, as well as increase areas for fish habitat. More river algae in the system will increase chlorophyll concentrations, food for insect populations, as well as organic sediments. Chlorophyll concentration will increase phosphorus regulation which is vital nutrient cycling. Oxygenated and organic sediments will increase the nitrification and denitrification in the system, which is vital nutrient cycling. Nutrient cycling itself is a return on investment, will increase soil health, leading to producing riparian forest. Healthy soils will also contribute to disease regulation, which is a return on investment. Riparian forest will increase natural beauty, produce more organic sediments, and provide shade. Shade will lower river temperature in places, providing more fish habitat. More fish will increase recreational fishing, providing a return on the initial investment. Natural beauty will increase sightseeing visitors, which is a return on investment.

Scroll to Continue


Each ecosystem service in the Sandy River system needs to be evaluated by different criteria. In the case of nutrient cycling, Costanza (1997) has evaluated lakes and river systems to contribute $8,498 per hectare per year. In a Canadian study, river disease regulating services were estimated overall, whether they were connected directly to current human water supplies or not. Value ranges from 15,319,798.36 US dollar to 26,440,998.58 US dollar, over 4,000,000 hectares (Kumar, 2010). This puts a dollar range per hectare at $3.8 - $6.6, I will use $5.2 as an average. Values of fish habitat can be determined on a per fish basis. A study of the Columbia river system places Spring Chinook at $177 per fish, Fall Chinook at $64 per fish, Coho Salmon at $64 per fish, and Steelhead at $128 per fish (Meyer-Zangri Associates, Inc., 1982). In addition, a study of the Sandy river determined averages of 28 Spring Chinook per mile, 28 Fall Chinook per mile, 11 Coho Salmon per mile, and 28 Steelhead per mile (Sandy River Basin Working Group, 2005). Cultural services have been estimated by the TEEB Valuation Database (2010) at $2089 per hectare of riparian buffer.

The area affected by the Marmot Dam removal varies by each service. The portion of river that was determined to have morphological changes totaled 17.9 miles, with significant changes occurring 2 miles downstream due to sediment flow, and 1.6 miles upstream from reservoir drain (Milstein, 2008). In addition to linear miles of river, PGE donated 607 hectares of land surrounding this portion of the Sandy river to the Bureau of Land Management (The Oregonian, 2007). For area affected by the Marmot Dam removal that will impact these services, I will include only the land within 100 meters of the affected river bank on each side. 1,609 meters per mile X 17.9 miles = 28,801 meters of riverbank X 100 meters from shore = 2,880,110 square meters per river side X 2 = 5,760,220 square meters of affected land / 10,000 square meters in a hectare = 576 hectares of land affected by Marmot Dam removal.




Per Unit Dollar



Nutrient Cycling

576 ha




Disease Regulation

576 ha




Provisioning Service Fish

17.9 miles



Riparian Forest

Cultural Service

607 ha




Nutrient Cycle value was determined based on hectares of land near river bank because nutrient cycling involves the river and surrounding areas. Disease Regulation was also determined by the hectares of land near river bank because disease regulation is affected by surrounding soils . Fish habitat value was determined based on linear miles of river affected. Lastly, cultural services were determined as hectares donated by PGE. Total yearly value of the portion of the Sandy river system affected by the Marmot dam removal totals $6,627,009.80.


The total yearly value of the ecological services provided by the Sandy river system is substantial. Determining this value is important because it tells us that the cost of removing the Marmot Dam was not only ecologically beneficial, but economically beneficial as well. PGE reports the total cost of the Bull Run restoration project, which includes the cost of removing the Marmot Dam, to be over $17 million (Fryburg, 2007). It has been over eight years since the removal of the Marmot Dam, and we can estimate, at $6,627,009 of ecological value being received each year, that the Sandy river system has added over $53 million in service values in that time.

Management and Sustainability

The Sandy continues to be an evolving ecosystem with many ecosystem services that intertwine with each other forming a web of interdependency and resilience. By removing the Marmot dam this system has been able to extend the reach upstream and downstream of ecological factors that play these vital roles. From an environmental standpoint, the resiliency of the system has only improved. Instead of having a buildup of silt and water in the upstream reservoir, the river is evenly distributed over the entire system. Instead of a blockade for salmon and trout, the entire habitat is open for them. Instead of a wide expanse of open water in the reservoir, the river is able to twist and meander creating more riparian forest. In economic terms, PGE no longer has to maintain the dam, and in addition, ecological services have been gained or bolstered. This economic benefit will certainly be a factor to deter future development projects in the area. This is especially true in the case of fish species that are endangered. This adds to the long term economic sustainability of protecting this environment. The social sustainability of the area has certainly improved as well. The Sandy River area affected by the Marmot Dam removal has only become more useful to people for recreation. Now that kayakers and tubers can travel more of the river, and hikers and fishers can make use of areas with more trails and fish, this area is more likely to have the attention and protection of policy makers.

The Sandy River and effects of the Marmot Dam removal are still under observation by various agencies and stakeholders. General river conditions are monitored by local municipalities such as the City of Sandy and the City of Troutdale, as well as park operators such as Oregon Parks and Rec and Metro. All of which operate recreational areas and perform general maintenance that serves the sustainability of their facilities. Water quality of the Sandy River is monitored by a partnership of the Portland Water Bureau and the US Forest Service, with Portland Water Bureau monitoring the water quality itself while the Forest Service will monitor compliance with laws (United States Department of Agriculture, 2007). Long term monitoring of the health of the Sandy River in regards to the Marmot Dam removal is shared by numerous stakeholders; Portland General Electric, Oregon Department of Fish and Wildlife, Mt. Hood National Forest, Forest Service Pacific Northwest Lab, U.S. Geological Survey, NOAA - National Marine Fisheries Service, Portland Water Bureau, Sandy River Watershed Basin Council, and the City of Portland. Together, these agencies will coordinate the monitoring of sediment load, stream structure, riparian vegetation, and biological diversity (United States Department of Agriculture).

All stakeholders have needed to agree upon an adaptive management strategy that will ensure the continued support of the Sandy River system. This is a long term plan that requires commitment from all parties involved. To properly assess the recovery of aquatic and riparian habitat in the Sandy River Basin input has been coordinated from the above mentioned stakeholders. Each will take part in the collection of data and ensuring compliance of policies. As stated, initial monitoring of sediment load, stream structure, riparian vegetation, and biological diversity has been implemented for the purpose of assessing what, if any, issues now exist. Designing the plan has centered around restoring fish habitat through restoration of the riparian forest in the upper basin. This was implemented by PGE after the basin was drained. Monitoring of fish redds is accomplished by the Oregon Department of Fish and Wildlife, and is considered a key indicator of overall normalization of the system (United States Department of Agriculture). Evaluation of implementation occurs at 20 and 30 year milestones to determine if additional action is needed for habitat restoration (Portland Water Bureau). These stakeholders have developed decision tree agreements to continue funding, adjust design, or end funding based upon stakeholder monitoring of the above mentioned factors.


The Sandy river system is important because it provides important ecosystem services to benefit human beings. Disease regulation and flood prevention are important for humans because those issues can cause problems and be harmful to us. Nutrient cycling is needed to plants can grow and provide food for us and animals. Fish habitat is needed for ecosystem resilience, as well as food and recreation. And healthy forests around the river provide cultural service that brings happiness to people.

By analyzing these services, we are able to put an economic value of over $6 million per year onto the Sandy river system. The decision to remove the Marmot Dam and provide continuing restoration to the system has come at a cost of over $17 million. But that investment has been a sound management decision due to the repaired services that return that investment every year. By implementing adaptive management practices, the Sandy River system will continue to return on that initial investment.


  1. Acuña, V., Díez, J. R., Flores, L., Meleason, M., Elosegi, A., & Jones, J. (2013). Does it make economic sense to restore rivers for their ecosystem services?. Journal Of Applied Ecology, 50(4), 988-997. doi:10.1111/1365-2664.12107
  2. Amoros, C. & Bornette, G. (2002). Connectivity and bio-complexity in waterbodies of riverine floodplains. Freshwater Biology 47: 761–776.
  3. Armstrong, J., Kemp, P., Kennedy, G., Ladle, M., & Milner, N. (2003). Habitat requirements of Atlantic salmon and brown trout in rivers and streams. Fisheries Research, 62(2), 143-170.
  4.,. (2016). Sandy River Water Trail. Retrieved 29 January 2016, from
  5. Costanza, R., d'Arge, R., de Groot, R., Farber, S., Grasso, M., & Hannon, B. et al. (1997). The value of the world's ecosystem services and natural capital. Nature, 387(6630), 253-260.
  6. Crisp, D., Trevor. (1993). The environmental requirements of salmon and trout in fresh water. Freshwater Forum, 3(3), pp. 176-202.
  7. Downs, P., Cui, Y., Wooster, J., Dusterhoff, S., Booth, D., Dietrich, W., & Sklar, L. (2009). Managing reservoir sediment release in dam removal projects: An approach informed by physical and numerical modelling of non‐cohesive sediment. International Journal of River Basin Management, 7(4), 433-452.
  8. (2016). Impacts of dams on rivers. Accessed January 13, 2016.
  9. Fogarty, C. (2007). PGE Prepares to Remove Two Dams on the Sandy River. Retrieved 21 January 2016, from
  10. Fryburg, M. (2007). News Releases | Portland General Electric Company. Retrieved 7 March 2016, from
  11. Houser, J., & Richardson, W. (2010). Nitrogen and phosphorus in the Upper Mississippi River: transport, processing, and effects on the river ecosystem. Hydrobiologia, 640(1), 71-88. doi:10.1007/s10750-009-0067-4.
  12. Keith, M. K. (2012). "Reservoir Evolution Following the Removal of Marmot Dam on the Sandy River, Oregon". Dissertations and Theses, Portland State University. Paper 532.
  13. Kerr, G., & Swaffield, S. (2012). Identifying Cultural Service Values of a Small River in the Agricultural Landscape of Canterbury, New Zealand, Using Combined Methods. Society & Natural Resources, 25(12), 1330-1339.
  14. Kondolf, G. M. & Wolman, M. G. (1993), The sizes of salmonid spawning gravels, Water Resour. Res.,29(7), 2275–2285, doi:10.1029/93WR00402.
  15. Kumar, P. (2010). Guidance manual for the valuation of regulating services. United Nations Environment Programme.
  16. Larsen, D. P., Kaufmann, P. R., Kincaid, T. M., & Urquhart, N. S. (2004). Detecting persistent change in the habitat of salmon-bearing streams in the Pacific Northwest. Canadian Journal Of Fisheries & Aquatic Sciences, 61(2), 283-291.
  17. Meyer-Zangri Associates, Inc.,. (1982). Net Economic Values for Salmon and Steelhead from the Columbia River System. U.S. Department of Commerce. Retrieved from
  18. Milstein, M. (2008). Oregon's Sandy River successfully reinvents itself after dam removal - Retrieved 20 February 2016, from
  19. O'Connor, W., & Andrew, T. (1998). The effects of siltation on Atlantic salmon, Salmo salar L., embryos in the River Bush. Fisheries Management & Ecology, 5(5), 393.
  20. Portland Water Bureau,. Monitoring, Research, and Adaptive Management Programs. Portland, Oregon: Portland Water Bureau. Retrieved from
  21. Reisinger, A., Tank, J., Rosi-Marshall, E., Hall, R., & Baker, M. (2015). The varying role of water column nutrient uptake along river continua in contrasting landscapes. Biogeochemistry, 125(1), 115-131. doi:10.1007/s10533-015-0118-z
  22.,. (2016). Sandy River, Oregon. Retrieved 29 January 2016, from
  23. Sandy River Basin Working Group,. (2005). Salmon and Steelhead Conservation. Oregon Trout. Retrieved from
  24. Shirvell, C., & Dungey, R. (1983). Microhabitats Chosen by Brown Trout for Feeding and Spawning in Rivers. Transactions Of The American Fisheries Society, 112(3), 355-367.<355:mcbbtf>;2.
  25. The Oregonian (Sunrise ed.) (Portland, Oregon). July 29, 2007. p. E04. Retrieved February, 20, 2016.
  26. United States Department of Agriculture,. Region 6 Working Together. Retrieved 1 March 2016, from
  27. United States Department of Agriculture,. (2007). Bull Run Watershed Management Unit Agreement. Portland, Oregon: USDA Forest Service & Portland Water Bureau.


Joshua on November 03, 2016:

Awesome research! With the table stating Nutrient Cycling Costanza stated that nutrient cycling within lakes/rivers is worth that much. Instead, expand the table and put each individual service in (5445- water reg., 2117 water supply... and so on)

Related Articles