Effects of floating photovoltaic on aquatic environment and turbulence and the response of phytoplankton

• A 3D numerical model is applied to explore FPV impacts on the water environment and trubulence in (semi-) enclosed waters. • FPV coverage shifts turbulence effects on phytoplankton from inhibition at low-medium to promotion at high coverage. • FPV coverage shifts phytoplankton growth controls from nutrient limitation (≤60%) to TKE, WT, and DO (70-100%). • Regression models predict Chl-a well in non-covered waters but show higher uncertainty in FPV-covered areas. The large-scale exploitation of mineral resources has generated extensive subsidence waters. To relieve the pressure on energy demand and reduce greenhouse gas emissions, Floating Photovoltaic (FPV) has been widely deployed over these waters, which are mainly (semi-)closed waters with limited water exchange rates, and the primary driving force of hydrodynamic processes is the wind field. However, in these waters, the variations in water environment and turbulence caused by FPV sheltering the water’s surface from the wind and limiting the solar radiation reaching the water column are unresolved, despite water environment and turbulence being key drivers of the ecosystem response to FPV deployment. In this study, the influence of FPV coverage on water quality indicators and turbulent kinetic energy (TKE) was investigated through numerical simulation. Furthermore, multiple regression analysis was applied to establish quantitative relationships between chlorophyll-a (Chl-a) concentration, water quality indicators, and turbulence intensity. Simulation results indicated that increasing FPV coverage progressively lowered water temperature (WT) and dissolved oxygen (DO), with significantly steeper declines observed within the FPV array compared to adjacent open water. In contrast, major solute concentrations remained largely unaffected. Turbulence exerts a coverage-dependent influence on phytoplankton biomass: strong suppression without FPV coverage, moderate suppression at low-medium FPV coverage (<60%), re-strengthened inhibition at medium–high FPV coverage (60–80%), and a shift toward promotion under extremely high FPV coverage (90–100%). Consequently, the dominant control on phytoplankton biomass switches from nutrient limitation (≤60%) to a physical regime governed by TKE, WT and DO (70–100%). The regression model shows acceptable predictive performance in the NP region, whereas larger deviations are observed in the FPV-covered P region, reflecting the complex environmental interactions under FPV shading. This integrated modelling framework effectively isolates coverage-specific ecological impacts and provides a practical decision-support framework for FPV siting, subject to local calibration and system characteristics, thereby supporting sustainable water-body management in mining subsidence landscapes.