UC Davis

Hydropeaking, a hydroelectricity generation strategy involving rapid changes to flow releases from dams in response to fluctuations in hourly-adjusted electricity markets, has been widely applied due to its economic efficiency. However, these operational practices produce sub-daily flow fluctuations that pose substantial hazards to riverine ecosystems and human activities. To ascertain the downstream impacts of hydropeaking, features of hydropeaking have been analyzed with respect to ecologically relevant hydrologic variables. However, since studies aiming to characterize hydropeaking regime often require manual feature extraction, they are limited to small temporal and spatial scales. Additionally, riverine ecologists have commonly treated hydropeaking as a broadly similar flow-alteration pattern regardless of the complexities of the electricity market and differences in the natural settings where it is applied. Therefore, this study sought to determine whether significantly different hydropeaking patterns exist on a regional scale, as revealed by the variation in hydropeaking over a long temporal scale (>five years). To fulfill this goal, a new algorithm, the Hydropeaking Event Detection Algorithm (HEDA), was developed in R to automate the characterization of hydropeaking flow regimes. Clustering analyses were conducted to explore the similarities and differences of hydropeaking regimes among 33 sites in numerous hydrologic regions of California. Four distinct classes of hydropeaking flow regimes were identified and distinguished by the duration and frequency of hydropeaking. Meanwhile, rate of change, amplitude and timing of hdyropeaking played less important roles in the classification.