| 20 Aug 2026
Floating photovoltaics and lake thermal responses: monitoring-based assessment of meteorological drivers across contrasting climates
Abstract. Floating photovoltaic (FPV) systems influence lake heat budgets by altering radiative input, wind exposure, and air–water heat exchange. However, harmonised field observations describing these processes across contrasting climatic settings and FPV system designs remain scarce, limiting the calibration and validation of hydrodynamic and ecological impact assessments. Here, we present a high-resolution monitoring dataset from three sites: Lake Toules (deep alpine reservoir, elevated open design, Switzerland), Lake Sekdoorn (humid lowland lake, dense east–west design, Netherlands), and Lake Leimersheim (shallow temperate–continental lake, low-profile design, Germany). Water temperature anomalies between FPV-covered and open-water sites (ΔTw) can be analysed analysed during cold- and warm-period extremes. Water temperature anomalies between FPV-covered and open-water sites (ΔTw) can be analysed during cold- and warm-period extremes. The dataset captures cooler conditions beneath FPV at Toules (–0.12 °C) and Leimersheim (–0.04 °C) but warmer conditions at Sekdoorn (+0.14 °C) during cold periods. Warm-period extremes amplified these contrasts, with alternating cooling and warming at Toules (–0.06 °C mean), negligible differences at Sekdoorn, and strong shading-driven cooling at Leimersheim (–0.55 °C). While mean differences were small, short-term deviations reached –0.75 to +0.5 °C, reflecting variable meteorological forcing. The observations further encompass contrasting meteorological controls and seasonal transitions between lake mixing regimes. Seasonal analysis of the large-scale system at Lake Sekdoorn revealed regime-dependent shifts: air temperature dominated during fully mixed winter conditions, shortwave radiation during stratification onset, and wind speed during stable summer stratification. The ΔTw–air temperature relationship reversed from positive in winter (heat retention) to negative in summer (shading-driven cooling), dampening the seasonal amplitude of surface water temperatures. All monitoring systems were harmonised across sites and synchronised using a common data acquisition framework. The dataset enables applications ranging from hydrodynamic model calibration and FPV impact parameterisation to comparative assessments across climatic regions and system designs. The underlying monitoring dataset is openly available at https://doi.org/10.5281/zenodo.21156967.
Dear authors
I have to concerns about this work:
1) The reported differences between open water and fvp-covered water are usually on the order of 0.1 °C. Yet, the reported accuracy of the thermistors is also
± 0.1 °C. Depending on the sensor type, this accuracy usually gets worse with time (sensor drift can be around 0.1 °C/year). What is the sensor
type that was used in this work and what is the sensor drift? Has there been any cross-validation between sensors before deployment? Without any
cross-validation, I am not convinced that differences on the order of 0.1 °C can be considered significant.
2) What is the horizontal distance between the open water and the fvp-covered thermistor chain? Horizontal transport in lakes is often on the order of
K=1e6 m2/s and the horizontal length scale of mixing is L=sqrt(2*K*T), where T is elapsed time. If T is 1 hour, then the associated length scale is
around 85 m, meaning that the effect of the fvp very quickly extends into the open water. If the horizontal distance between the two thermistor chains
is small, then I think this horizontal mixing effects would need to be discussed.