Improving medicanes representation with a high-resolution regional coupled atmosphere-ocean configuration of the ICON Earth System Modelling framework

A first regional configuration of the ICON-Ocean model in Limited Area Mode (ICON-O-LAM) is now available within the ICON Earth System Model framework, enabling fully coupled regional ocean–atmosphere simulations with ICON-NWP over the Mediterranean Sea. This configuration can be used to investigate Mediterranean high-impact weather and provides a flexible framework that can be systematically applied to multiple medicane events occurring in the basin.Medicanes are characterized by intense small-scale dynamics and a strong dependence on air-sea interactions, requiring a modelling framework capable of resolving key feedback processes such as sea surface temperature cooling, surface heat fluxes, and wind-driven ocean responses. Uncoupled atmospheric simulations with prescribed or static SSTs, typical of many operational setups, are unable to represent these interactions and may therefore misrepresent storm intensity, structure, and evolution.As a first application, the coupled ICON-O-LAM/ICON-NWP system has been applied at 2.5 km resolution to simulate Medicane IANOS (September 2020), one of the strongest Mediterranean tropical-like cyclones on record, and is used here as a benchmark case to assess the performance of the coupled regional ICON system. The coupled simulations show clear improvements over uncoupled experiments, reproducing IANOS intensity more realistically, capturing SST cooling effects, and providing a better representation of precipitation patterns.Beyond this initial application, the modelling framework is designed to be extended to other recent Mediterranean medicanes, such as Zorbas (September 2018), Apollo (October 2021), and Daniel (September 2023), enabling systematic, high-resolution analyses across multiple events. This configuration offers new opportunities to investigate medicane intensification, air–sea coupling mechanisms, and event-to-event variability, providing a valuable platform for both research applications and future operational forecasting of Mediterranean tropical-like cyclones.