Simulated and Projected Tropical Cyclone Potential Intensity and Dynamic Genesis Potential Index in HighResMIP

Le résumé fourni par la source

Abstract Potential intensity (PI) and the dynamic genesis potential index (DGPI) provide complementary thermodynamic and dynamical measures of the large‐scale environment relevant to tropical cyclone (TC) intensity and genesis. Here we present a coordinated paired‐resolution, multi‐model assessment of PI and DGPI using seven High Resolution Model Intercomparison Project (HighResMIP) atmosphere‐only model pairs and ERA5. We evaluate the climatology during 1980–2014 and mid‐21st‐century changes, defined as 2030–2049 minus 1995–2014, during each hemisphere's TC season. Both high‐ and low‐resolution multi‐model ensembles reproduce the basin‐scale patterns of PI and DGPI, but systematic biases persist. PI is generally underestimated, whereas DGPI and its variability are overestimated relative to ERA5, particularly in dynamically active basins such as the western North Pacific. Higher resolution yields only modest improvements in basin‐aggregated PI errors, consistent with PI being primarily constrained by the large‐scale thermodynamic structure shared across paired configurations. Projected PI changes are spatially heterogeneous but generally modest by mid‐century and are primarily shaped by CAPE‐related thermodynamic disequilibrium, despite a broadly positive surface‐temperature contribution. DGPI projections are strongly basin dependent, with decreases over the western North Pacific and South Pacific and increases in parts of the eastern North Pacific and North Atlantic. DGPI changes are closely associated with changes in 500 hPa pressure vertical velocity, highlighting the central role of large‐scale dynamics in shaping future genesis favorability. Overall, for these environmental diagnostics, paired HighResMIP simulations suggest that resolution effects are secondary to mean‐state biases and the forced large‐scale thermodynamic and dynamical responses.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.