Enhancing seasonal forecast of heat stress indicators through teleconnection-based subsampling

In recent years, hybrid statistical-dynamical approaches have emerged as a promising avenue to enhance seasonal predictions of the extratropical climate (Slater et al., 2023). These methods combine ensemble-based dynamical prediction systems with statistical post-processing techniques, with the aim of amplifying the predictable components of climate variability and, thus, increasing the signal-to-noise ratio in the model ensemble.

Within this framework, the teleconnection-based subsampling approach has been shown to significantly improve the seasonal prediction of both winter and summer climate across large portion of Eurasian continent, including the occurrence of summer extreme temperature events (Famooss Paolini et al., 2024). This technique relies on selecting a limited subset of ensemble members whose simulations of summer low-frequency atmospheric variability are consistent with its statistical forecasts derived from springtime predictors.

Based on these promising results, we present here an application of the teleconnection-based subsampling approach to assess its potential for enhancing the seasonal prediction of heat stress indicators during the summer. Specifically, the methodology is implemented to mimic real-time operational forecast environment, thus differing from standard retrospective forecast (hindcast) applications. For this purpose, we use the ECMWF seasonal prediction system, which provides data from 1981 to 2024, and the ERA5 reanalysis as surrogate of observations. The model ensemble is subsampled by retaining few ensemble members that adequately capture the teleconnection pattern associated with the summer North Atlantic Oscillation, which represents the leading mode of summer low-frequency atmospheric variability.

The results of this study are particularly relevant, as they contribute to the development and implementation of innovative methodologies for predicting climate conditions that pose risks to human health. This is especially important in the context of climate change, which is projected to increase heat-related mortality by up to 50% in the coming decades unless strong mitigation and adaptation strategies are adopted (Masselot et al., 2025).

Bibliography

Famooss Paolini et al. (2024). Hybrid statistical-dynamical seasonal prediction of summer extreme temperatures in Europe. Quarterly Journal of the Royal Meteorological Society, 151(766). https://doi.org/10.1002/qj.4900

Masselot et al. (2025). Estimating future heat-related and cold-related mortality under climate change, demographic and adaptation scenarios in 854 European cities. Nature Medicine, 1-9.  https://doi.org/10.1038/s41591-024-03452-2

Slater et al. (2023). Hybrid forecasting: blending climate predictions with AI models. Hydrology and earth system sciences, 27(9), 1865-1889. https://doi.org/10.5194/hess-27-1865-2023