Press release from the Swiss Federal Institute for Forest, Snow and Landscape Research (WSL) in Birmensdorf, dated 18 June 2026:
Sub-seasonal forecasting of thermal stress for Swiss river fishes during heatwaves
WSL researchers are proposing an early-warning system that provides up to three weeks’ notice when river fish in Switzerland are at risk from heat. As heatwaves are becoming increasingly frequent due to climate change, this will allow time to take action and protect fish populations from future heatwaves.
* A forecasting tool provides advance warning when river fish in Switzerland are at risk from heat stress.
* The system assesses future risk based on forecasts of water temperature and the heat sensitivity of local fish populations.
* The forecasts are freely available online to support the proactive protection of fish populations.
Preparing for extreme weather
When will rivers in Switzerland become too warm for river fish? To predict this, ecologists and climatologists from the Swiss Federal Institute for Forest, Snow and Landscape Research (WSL), together with experts from across Switzerland, have developed a forecasting tool and published their findings in the *Journal of Ecological Modelling*. The forecasts are updated twice a week and are publicly available on drought.ch.
As heatwaves are becoming more frequent and intense due to climate change, fish are particularly at risk. Fish are ectothermic, or ‘cold-blooded’, and their body temperature depends on their environment. This means they cannot cool themselves down. ‘When the water temperature rises, the chemical reactions in the fish’s cells change – we refer to this as heat stress,’ explains Camille Albouy from the joint research group ‘Ecosystems and Landscape Development’ at WSL and ETH Zurich.
From historical data to future risks
The tool predicts, up to three weeks in advance, the extent to which fish populations across Switzerland are at risk from heat. It is based on three factors: water temperature forecasts, fish physiology and species distribution. The first of these was developed by the WSL’s Hydrological Forecasting Group. ‘Our machine learning model, which we have trained using data from the last ten years, predicts river temperatures based on current temperature measurements and weather forecasts,’ explains Ryan Padrón Flasher, one of the group’s members.
Not all fish are equally sensitive to heat. The authors have compiled information from previously published data on the temperature thresholds of 59 fish species found in Switzerland. This refers to the temperature at which their ability to swim – and thus their ability to escape danger – is impaired. The authors found that non-native fish species, on average, have a temperature tolerance that is 1.4 °C higher than that of native species. The marbled carp (Hypophthalmichthys nobilis) exhibited the highest tolerance at 32.3 °C, whilst the burbot (Lota lota) showed the lowest tolerance at 24.1 °C.
As fish species are unevenly distributed across Switzerland, taking their distribution into account makes the forecasts location-specific. “It’s a constant trade-off between the complexity and the quality of the model,” explains Albouy. “As our aim was to develop a risk map for the whole of Switzerland based on the best climate forecasts, we had to keep it simple and focus on what is important for management.”
Looking back
To test the forecasts, the scientists applied their tool to data from the summer of 2018, when around three tonnes of river fish died in Switzerland. “We wanted to see how well we could predict the observed fish mortality in retrospect,” says Albouy. However, this proved more difficult than expected. “Data on fish mortality is limited,” he explains, “so we turned to practitioners to see whether their observations matched our predictions.”
Of the three reported cases of fish mortality in 2018, the tool correctly predicted two. At the remaining sites, the predictions were 70 per cent accurate and successfully identified areas of low and medium risk. However, at five sites, the tool overestimated the risk level, probably because it did not take into account more detailed information such as local population size or the presence of cooler refuges in rivers. “It’s a promising start,” says Albouy, “now we need to see whether this will develop into a long-term solution for Switzerland’s rivers.”
Scientific contacts:
Dr. Konrad Bogner
Wissenschaftlicher Mitarbeiter
Gebirgshydrologie und Massenbewegung
Hydrologische Vorhersagen
konrad.bogner@wsl.ch
+41 44 739 2495
Publication:
Bonaglia, A., Shen, C., Padrón, R. S., Bogner, K., Fopp, F., Rubin, A., … Pellissier, L. (2025). Sub-seasonal forecasting of thermal stress for Swiss river fishes during heatwaves. Ecological Modelling, 507, 111171 (14 pp.). https://doi.org/10.1016/j.ecolmodel.2025.111171
Background photo: https://www.swr.de/swraktuell/baden-wuerttemberg/friedrichshafen/wieder-kaeltezonen-im-hochrhein-100.html