Hydropower peaking and stalled salmon migration are linked by altered reservoir hydraulics: A multidisciplinary synthesis and hypothesis

Abstract

AbstractAdding variable renewable energy (solar, wind) in electricity portfolios will increase need for fast grid responses through hydropower peaking. Over 60 years of daily hydropeaking by four dams on the lower Snake River, United States of America provide an example of long‐term environmental impacts. Downstream‐migrating Chinook salmon (Oncorhynchus tshawytscha) subyearlings that normally transit the dammed river in summer slowed their migration into autumn with about one‐third of those delayed overwintering in the reservoirs. Specific cause for stalled migration is controversial, with options including evolution of a new migratory strain and action of environment on individual migrants. Analysis of archived dam data shows evidence of reservoir seiches (standing waves of waterbody oscillations) caused by within‐day hydropower peaking during the October–February period of stalled migration. Analyses of limnological literature identified typical water movements in seiches and analyses of biological literature identified typical effects on fish kinematics (shape, motion) of changing flows in a fish's immediate proximity. Process‐focused inference predicts anomalous fish‐migration behavior in seiche hydraulics, which matches fish‐tracking data obtained by others in Lower Monumental Reservoir. Fish tracks include upstream swimming (‘downstream’ in reverse seiche flows) and periods of disoriented movements typical of responses to repeatedly changing flows. This multi‐disciplinary, process‐focused synthesis yields a testable hypothesis that seiche‐flow hydraulics in the lower Snake River reservoirs from hydropower peaking contribute to known out‐migration delay and overwintering of late‐migrating fall Chinook salmon subyearlings. As hydropower peaking causes seiches in downstream reservoirs elsewhere in North America and Europe, migratory species elsewhere may be susceptible to similar migration delays with long‐term population effects.