Floating solar photovoltaic (FPV) deployments are increasing globally as the switch to renewable energy intensifies, representing a considerable water surface transformation. FPV installations can potentially impact aquatic ecosystem function, either positively or negatively. However, these impacts are poorly resolved given the challenges of collecting empirical data for field or modelling experiments. In particular, there is limited evidence on the response of phytoplankton to changes in water body thermal dynamics and light climate with FPV. Given the importance of understanding phytoplankton biomass and species composition for managing ecosystem services, we use an uncertainty estimation approach to simulate the effect of FPV coverage and array siting location on a UK reservoir. FPV coverage was modified in 10% increments from a baseline with 0% coverage to 100% coverage for three different FPV array siting locations based on reservoir circulation patterns. Results showed that FPV coverage significantly impacted thermal properties, resulting in highly variable impacts on phytoplankton biomass and species composition. The impacts on phytoplankton were often dependent on array siting location as well as surface coverage. Changes to phytoplankton species composition were offset by the decrease in phytoplankton biomass associated with increasing FPV coverage. We identified that similar phytoplankton biomass reductions could be achieved with less FPV coverage by deploying the FPV array on the water body's faster-flowing area than the central or slower flowing areas. The difference in response dependent on siting location could be used to tailor phytoplankton management in water bodies. Simulation of water body-FPV interactions efficiently using an uncertainty approach is an essential tool to rapidly develop understanding and ultimately inform FPV developers and water body managers looking to minimise negative impacts and maximise co-benefits.

Phytoplankton blooms are a global water quality issue, and successful management depends on understanding their responses to multiple and interacting drivers, including nutrient loading and climate change. Here, we examine a long-term dataset from Lake 227, a site subject to a fertilization experiment (1969–present) with changing nitrogen:phosphorus (N:P) ratios. We applied a process-oriented model, MyLake, and updated the model structure with nutrient uptake kinetics that incorporated shifting N:P and competition among phytoplankton functional groups. We also tested different temperature and P-loading scenarios to examine the interacting effects of climate change and nutrient loading on phytoplankton blooms. The model successfully reproduced lake physics over 48 yr and the timing, overall magnitude, and shifting community structure (diazotrophs vs. non-diazotrophs) of phytoplankton blooms. Intra- and interannual variability was captured more accurately for the P-only fertilization period than for the high N:P and low N:P fertilization periods, highlighting the difficulty of modeling complex blooms even in well-studied systems. A model scenario was also run which removed climate-driven temperature trends, allowing us to disentangle concurrent drivers of blooms. Results showed that increases in water temperature in the spring led to earlier and larger phytoplankton blooms under climate change than under the effects of nutrient fertilization alone. These findings suggest that successful lake management efforts should incorporate the effects of climate change in addition to nutrient reductions, including intensifying and/or expanding monitoring periods and incorporating climate change into uncertainty estimates around future conditions.

Phytoplankton blooms are a global water quality issue, and successful management depends on understanding their responses to multiple and interacting drivers, including nutrient loading and climate change. Here, we examine a long-term dataset from Lake 227, a site subject to a fertilization experiment (1969–present) with changing nitrogen:phosphorus (N:P) ratios. We applied a process-oriented model, MyLake, and updated the model structure with nutrient uptake kinetics that incorporated shifting N:P and competition among phytoplankton functional groups. We also tested different temperature and P-loading scenarios to examine the interacting effects of climate change and nutrient loading on phytoplankton blooms. The model successfully reproduced lake physics over 48 yr and the timing, overall magnitude, and shifting community structure (diazotrophs vs. non-diazotrophs) of phytoplankton blooms. Intra- and interannual variability was captured more accurately for the P-only fertilization period than for the high N:P and low N:P fertilization periods, highlighting the difficulty of modeling complex blooms even in well-studied systems. A model scenario was also run which removed climate-driven temperature trends, allowing us to disentangle concurrent drivers of blooms. Results showed that increases in water temperature in the spring led to earlier and larger phytoplankton blooms under climate change than under the effects of nutrient fertilization alone. These findings suggest that successful lake management efforts should incorporate the effects of climate change in addition to nutrient reductions, including intensifying and/or expanding monitoring periods and incorporating climate change into uncertainty estimates around future conditions.