Mesoscale Convective Systems over Europe: A Comparison of CMIP5 and CMIP6-driven EURO-CORDEX Ensembles

The accurate simulation of Mesoscale Convective Systems (MCSs) is a critical benchmark for the skill of regional climate models (RCMs), as these systems are principal drivers of high-impact weather and hazardous flash floods across Europe.However, evaluating MCSs in climate ensembles has been historically hindered by a lack of tracking-compatible variables and the difficulty of distinguishing organized, self-sustaining convection from Europes frequent frontal precipitation.To address this, we introduce the EMMA-Tracker (Evolution-based MCS Model Assessment), a novel algorithm designed to identify and track MCSs using only standard model output variables..This design choice ensures physical consistency when comparing observations with RCM ensembles. The tracker utilizes a series of physics-based post-processing filters, to isolate genuine MCSs based on their full spatiotemporal lifecycle.We first present a new 27-year reference warm season climatology (1998–2024) generated by applying the EMMA-Tracker to IMERG precipitation and ERA5-derived instability.This dataset reveals that MCS contribution to heavy hourly precipitation (P99.9) exceeds 60% across most of continental Europe and 80% over parts of the Mediterranean.Building on this benchmark, we investigate the representation of MCSs within the different generations of the EURO-CORDEX ensembles.This process-oriented evaluation provides a pathway to understand how mesoscale organization evolves in RCMs and offers insights into the uncertainties of future projections of high-impact convective weather in a warming climate.