Search for a command to run...
Hydropeaking, resulting from flexible hydropower operations, causes rapid hydraulic fluctuations and disrupts downstream ecosystems, posing significant ecological challenges. This study develops an enhanced ecohydraulic modeling framework to assess the responses of fish habitats to different hydropeaking scenarios for the European grayling ( Thymallus thymallus ) and brown trout ( Salmo trutta ). A two-dimensional hydrodynamic model was coupled with a novel habitat module that introduces a machine learning-based habitat suitability computation method. This method effectively captures the nonlinear and coupled effects between flow velocity and water depth, outperforms traditional habitat suitability approaches (e.g., geometric mean and fuzzy logic), and yields stable, ecologically consistent habitat suitability results. To complement local habitat assessment, a habitat connectivity index (HCI) was also developed to quantify longitudinal connectivity and migration pathways across life stages under dynamic flow conditions. Results demonstrated that hydropeaking frequency was the primary driver of habitat variability, while spill gate closing time played a secondary role. Lower frequencies increased temporal variability but expanded opportunities for accessing high-quality patches; higher frequencies stabilized conditions but reduced ecological flexibility. Longer closing times enhanced short-term stability, but their effect was relatively minor compared with that of frequencies. The HCI approach enabled the prediction of migration pathways and the identification of critical corridors under extreme flow regulation. The proposed framework offers a valuable tool for evaluating habitat quality and connectivity dynamics under hydropower regulation, providing practical insights for adaptive ecological flow management in regulated rivers. • An improved ecohydraulic model is developed to assess the impacts of hydropeaking events on fish habitats. • Species- and life-stage-specific modeling reveals differential habitat requirements and sensitivities. • Hydropeaking frequency is identified as the primary driver affecting habitat stability of both suitability and connectivity compared to spill gate closing time. • An explicit spatial method is proposed to predict potential migration pathways under extreme flow regulation.