Multi-fidelity model assessment of climate change impacts on river water temperatures and thermal extremes and potential effects on cold-water fish in Switzerland

Abstract

River water temperature is a key factor for water quality, aquatic life, and human use. Under climate change, inland water temperatures have increased and are expected to do so further, increasing the pressure on aquatic life and reducing the potential for human use. Here, future river water temperatures are projected for Switzerland based on a multi-fidelity modeling approach. We use 2 different semi-empirical surface water temperature models, 22 coupled and downscaled general circulation to regional climate models, future projections of river discharge from 4 hydrological models, and 3 climate change scenarios (RCP2.6, 4.5, and 8.5). By grouping catchments under representative thermal regimes and by employing hierarchical cluster-based thermal-pattern recognitions, an optimal model and model configuration were selected, thereby improving model performance. Results show that, until the end of the 21st century, average river water temperatures in Switzerland will likely increase by 3.2 ± 0.7 °C (or 0.36 ± 0.1 °C per decade) under RCP8.5, while, under RCP2.6, the temperature increase may remain at 0.9 ± 0.3 °C (0.12 ± 0.1 °C per decade). Under RCP8.5, temperatures of rivers classified as being in the Alpine thermal regime will increase the most, that is, by 3.5 ± 0.5 °C, followed by rivers of the Downstream Lake regime, which will increase by 3.4 ± 0.5 °C. Under RCP2.6, temperatures in the Alpine and Downstream Lake regimes change most, with +1.15 and +0.99 ± 0.5 °C. A general pattern of decreasing river discharge in summer (−10 % to −40 %) and increasing river discharge in winter (+10 % to +30 %), combined with a further increase in average near-surface air temperatures (0.5 °C per decade), bears the potential to result in not only overall warmer rivers but also prolonged periods of extreme summer river water temperatures. This dramatically increases the thermal-stress potential for temperature-sensitive aquatic species such as the brown trout in rivers where such periods occur already but also in rivers where this was previously not a problem. By providing information on future water temperatures, the results of this study can guide the management of climate mitigation efforts.