Modelling salmon spawning habitat response to dam removal [Case study : Marmot Dam, Oregon, U.S.A.]

The practice of dam removal in the United States is increasing. Federal standards for safety and environmental protection are stricter now, when compared to the era of dam building in the 1960’s. In consequence some dam owners are faced with an ultimatum to choose between costly improvements or abandoning their license for operation. The Pacific Northwest has many of its rivers designated as critical habitat for salmon, which has some species under threat of extinction. Dam removal on one hand is a restorative measure, but may also impose a significant disturbance to the river system. The fate and magnitude of reservoir sediments is the primary concern. This research employs a two-dimensional morphodynamic model, Delft3D to simulate the transport of graded, non-cohesive reservoir sediments following dam removal. Changes in bed topography, sediment composition, and hydrodynamic conditions are simulated for a 20 km reach in a high-gradient river. This analysis is applied to a case study for the 2007 removal of Marmot Dam, on the Sandy River in Oregon, U.S.A. to assess the short-term impacts to Chinook salmon spawning habitat. Locations of suitable spawning habitat are mapped for pre-removal and post-dam conditions by linking morphological output with physical indicators of salmon survival across multiple life stages. Results indicate that the disturbance to spawning habitat is strongly influenced by morphologic controls of the river system. Therefore, the effect of dam removal on Chinook salmon spawning habitat is variable in time and space, and responds differently to the transport dynamics of fine and coarse sediment. This interdisciplinary approach shows promise as a prediction tool for future dam removal studies, but future research is needed on the study of reservoir erosion