A machine-learning approach to thunderstorm forecasting through post-processing of simulation data

Thunderstorms pose a major hazard to society and the economy, which calls for reliable thunderstorm forecasts. In this work, we introduce SALAMA, a feedforward neural network model for identifying thunderstorm occurrence in numerical weather prediction (NWP) data. The model is trained on convection-resolving ensemble forecasts over central Europe and lightning observations. Given only a set of pixel-wise input parameters that are extracted from NWP data and related to thunderstorm development, SALAMA infers the probability of thunderstorm occurrence in a reliably calibrated manner. For lead times up to 11 h, we find a forecast skill superior to classification based only on NWP reflectivity. Varying the spatiotemporal criteria by which we associate lightning observations with NWP data, we show that the time-scale for skillful thunderstorm predictions increases linearly with the spatial scale of the forecast. We present SALAMA, a neural network model for predicting the probability of thunderstorm occurrence by post-processing convection-resolving ensemble forecasts. Shown is a case visualization for June 23, 2023, at 2000 UTC, with pixels in which thunderstorm probability exceeds a decision threshold (red), and pixels in which lightning occurs (black contours). We carefully evaluate model reliability and investigate how model skill depends on the spatial scale of the forecast. image