Hybrid statistical-dynamical seasonal prediction of summer extreme temperatures over Europe
- University of Bologna, Department of Physics and Astronomy, Bologna, Italy
Several studies show that the occurrence of summer extreme temperatures over Europe is increased since the middle of the twentieth century and is expected to further increase in the future due to global warming. Thus, due to the impacts that extreme temperatures have on socio-economic and environmental systems, predicting heat extremes and their statistics several months ahead is crucial.
State-of-the-art dynamical prediction systems show low skills in predicting European heat extremes on seasonal timescale. This is deemed to be due to the combination of the internal chaotic nature of the atmosphere and the underestimation of predictable component of the climate variability in the model ensemble. Recent studies have shown that our skills in predicting extratropical climate can be largely improved by refining the ensemble-based dynamical prediction systems with statistical post-processing techniques. Such techniques are based on sub-sampling the model ensemble by selecting only members that verify specific conditions.
The present study assesses the prediction skill of summer extreme temperatures over Europe on seasonal timescale in the Copernicus Climate Change Service (C3S) multi-systems seasonal forecasts for the period 1993—2016. Then, a hybrid statistical-dynamical prediction system is presented, where the model ensemble is sub-sampled by retaining only a subset of members which predict the summer extreme temperature statistics over Europe in agreement with a teleconnection-based statistical prediction (following Dobrynin et al. 2022). This approach relies on predictors of European heat extremes during spring season (e.g. sea surface temperature in the North Atlantic and tropical Pacific, soil moisture and sea-ice concentration) and thus allows us to retain only those members with a reasonable representation of summer heat extreme teleconnections.
Results on the skill of the hybrid statistical-dynamical prediction system and its potential applications for the health-sector are discussed.
How to cite: Famooss Paolini, L., Ruggieri, P., Pascale, S., Brattich, E., and Di Sabatino, S.: Hybrid statistical-dynamical seasonal prediction of summer extreme temperatures over Europe, EMS Annual Meeting 2023, Bratislava, Slovakia, 4–8 Sep 2023, EMS2023-440, https://doi.org/10.5194/ems2023-440, 2023.
