An ongoing PhD thesis addresses the challenge of improving sub-seasonal to seasonal (S2S) forecasts for renewable energy applications, focusing on wind and temperature variables. The current statistical approach, based on the relationship between large-scale atmospheric variables and surface conditions, is refined by exploring nonlinear models like Convolutional Neural Networks. The study also seeks to incorporate various atmospheric variables and extend the application to multi-model ensembles. Consideration of the economic value and effective communication of forecast information to end-users in the energy sector are additional aspects to be explored during the PhD.

Tian, G., Le Coz, C., Alexandre Charantonis, A., Tantet, A., Goutham, N., and Plougonven, R.: Convolutional neural network downscaling to improve sub-seasonal wind-speed predictions in Europe, EMS Annual Meeting 2023, Bratislava, Slovakia, 4–8 Sep 2023, EMS2023-321, https://doi.org/10.5194/ems2023-321, 2023.