COS 97-6 - Alteration of flow variability by dams in the U.S. and implications for freshwater biodiversity

Wednesday, August 9, 2017: 3:20 PM
B114, Oregon Convention Center
Albert Ruhí, National Socio-Environmental Synthesis Center (SESYNC), Annapolis, MD and Julian D. Olden, School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA

Hydropower dams are a major driver of hydrological alteration in the U.S. Dams and reservoirs increase water residence time, muting high flows and dampening the longer scales of flow variability (seasonality). Additionally, the periodic fluctuation in streamflow due to the production of electricity by hydropower plants, or hydropeaking, represents an important alteration to the natural flow regime. Here we asked to what extent dams have affected the different scales of flow variability in the U.S., by comparing pre- to post-dam periods in river reaches affected by dams, and dammed to nearby undammed river sites. We used time-series analyses – namely the wavelet transform – to localize changes in flow variability in long-term USGS gages U.S.-wide. By comparing changes in the frequency domain between pre and post-dam periods, and between dammed and undammed gages, we identified the unique effects of dams and dam operations on downstream flow regimes. Finally, we quantified how the dampening of ‘natural’ signals (i.e. yearly, seasonal scales) and the introduction of ‘artificial’ signals (i.e. daily, weekly scales) in dammed rivers may be decoupling downstream hydrographs from the life histories of riverine and riparian communities.


Most pre-dam periods presented a marked yearly and/or seasonal periodicity in flow, ‘natural’ signals that were largely dampened during post-dam periods due to reservoir storage. Additionally, ‘artificial’ daily and weekly signals arising from dam operation emerged during post-dam periods. These effects were not present in nearby undammed river sites, suggesting that the effects of dams on flow periodicity may be overriding other influences (e.g., climate). We also found that whereas ‘natural’ signals were generally not restored downstream of dams, ‘artificial’ signals ameliorated progressively, easing the filter that high-frequency cycles (daily and weekly signals) impose on long-lived organisms. Our work advances the notion that dams alter high-level properties of flow regimes like flow periodicity and predictability, threatening downstream freshwater biodiversity. Because hydropower dams are increasingly planned and under construction in emerging economies, and hydropower installed capacity will almost double over the next 10-20 years, research on mitigating impacts of high head storage hydropower plants is extremely timely.