Environment-dependent hail hazard maps from high-resolution modelling

Hail is one of the most damaging convective hazards. However, hail hazard maps are commonly derived from long-term climatologies or numerical simulations based on a single, fixed model configuration that do not account for the influence of the large-scale atmospheric environment on hail-producing convection, limiting the physical consistency and reliability of hazard estimates.

In this study, we present hail hazard maps derived from a synoptic-regime-aware modelling framework. To construct these maps, hail days are first classified into distinct synoptic situations using a clustering analysis of large-scale atmospheric fields. For each synoptic regime, a genetic algorithm is used to optimize the physical parameterization configuration of the Weather Research and Forecasting (WRF) model, targeting an improved simulation of hail occurrence evaluated against ground-based hail observations. This approach results in a regime-specific WRF configuration for hazard map generation, rather than a single configuration applied across all atmospheric conditions.

High-resolution, convection-permitting WRF simulations are then performed to generate hail hazard maps. Each simulation is run using the configuration optimized for its corresponding synoptic regime. The regime-specific simulations are subsequently combined to produce hazard maps.

The proposed approach provides a physically informed, flow-dependent strategy for hail hazard mapping, enabling a more realistic representation of extreme convective events and their spatial variability. This methodology could offer a robust framework for regional hail risk assessment.