Multi-model high-resolution analysis of Tropical-Like Cyclone Daniel with WRF and ICON: peculiarities and sensitivity to convection schemes.

Mediterranean hurricanes, or "Medicanes," represent one of the most challenging events in contemporary atmospheric physics due to their hybrid nature and the rapid nonlinear processes driving their intensification. Medicane Daniel (September 2023) represents a prime case study, characterized by a long duration and a characteristic transition from a baroclinic disturbance to a tropical-type warm-core vortex. This study delivers a side-by-side, high-resolution (~2 km) assessment of Daniel's predictability and physical representation, using two state of the art weather forecasting models: the Weather Research and Forecasting (WRF) model and the Icosahedral Non-hydrostatic (ICON) model. To isolate the influence of sub-grid-scale atmospheric processes, we conducted seven numerical experiments, specifically investigating the sensitivity of the cyclone's life cycle to different convection parameterization settings. The study compares fully explicit convection with deep cumulus and shallow convection (SH) configurations. Our results reveal that, although both models exhibit remarkable consistency in reproducing the storm's track, large discrepancies emerge regarding simulated intensity, vortex thermodynamic structure, and ground effects. In particular, the results demonstrate that the transition to the tropical phase is extremely sensitive to how moist convection is reproduced at the sub-grid scale. Contrary to the common assumption that fully explicit configurations are inherently superior below 5 km, this study finds that the inclusion of a specific parameterization for shallow convection produces a more robust and physically consistent cyclone. Furthermore, objective spatial verification of accumulated total precipitation using the Fractions Skill Score (FSS) against Integrated Multi-satellite Retrievals for GPM (IMERG) satellite observations highlights that the SH configuration in both WRF and ICON offers better localization of the maximum precipitation extremes responsible for the devastating floods in Greece and Libya. By connecting an academic analysis of atmospheric dynamics to the needs of operational forecasting, this work offers fundamental insights into the peculiarities of these models in simulating extreme events in the Mediterranean basin. It therefore highlights the sensitivity of physical parameterizations for forecasting high-impact mesoscale events, providing important insights for strengthening the resilience of early warning systems in the face of an intensifying Mediterranean climate.