Abstract
<jats:p>Weather forecasting has traditionally relied on Numerical Weather Prediction (NWP) models, which simulate weatherby solving the governing fluid equations. Recently, the emergence of Deep Learning Weather Prediction (DLWP)models has opened a new era in weather forecasting, offering a data-driven alternative to classical NWP approaches.Regional DLWP models such as the stretched-grid model Bris developed by Met Norway, have demonstrated perfor-mance on par with, or even slightly better than regional NWP models across a range of standard forecast metrics.By overcoming the coarse horizontal resolution that constrained earlier global data-driven models, the operationaluse of regional DLWP systems now appears increasingly promising. Nevertheless, the performance of such modelsduring extreme events is generally inferior to that of regional NWP models, and comprehensive evaluations of theirability to generate physically realistic forecasts are still lacking.Here, we present a study comparing the physical consistency of the deterministic version of Bris with the controlrun of the operational MetCoOp Ensemble Prediction System (MEPS) in forecasting the severe extratropical cyclonePoly, which hit the Netherlands on 5 July 2023. We examine whether Bris accurately represents deviations fromkey atmospheric balances and whether it reproduces expected dynamics of the storm. We show that, despite itsrelatively good performance in terms of RMSE, Bris struggles to capture important mesoscale features of the eventand that it significantly disrupts several atmospheric balances. This unrealistic disruption is mainly linked to thefine-scale noise evidenced in its output fields, which leads to incorrect and unrealistic spatial gradients. The analysisof the amplitude spectra also reveals that, in the stretched-grid DDM, fine-scale noise coexists with a seeminglycompeting smoothing of the same meteorological variables at larger scales. This tendency for large-scale smoothingis commonly observed in DLWP models trained with MSE loss and we show that it has notable consequences whenforecasting extreme windstorms such as Poly. These results raise critical questions for improving AI-based modelsto better represent extreme events and how to ensure physical consistency in their predictions.</jats:p>