Spatio-temporal aggregation of convective cell clusters in European MCSs

Mesoscale Convective Systems (MCSs) are organized collections of thunderstorms that typically consist of narrow, intense regions of convective precipitation alongside broader, lighter areas of stratiform precipitation. These systems are the primary contributors to extreme precipitation events across Europe (Da Silva & Haerter, 2023). While both convective and stratiform precipitation rates are expected to increase with temperature according to thermodynamic expectations (the Clausius-Clapeyron relationship), their statistical superposition may intensify at an even faster rate due to increased proportion of convective-type precipitation within MCSs under warmer conditions (Da Silva & Haerter, 2025, accepted).

Both the intensity and proportion of convective-type precipitation in MCSs play a critical role in determining flood risks, but the spatial and temporal organization of convection is equally significant for shaping the characteristics and severity of flooding. Larger, long-lived clusters of convection within MCSs are indeed more likely to trigger severe flooding compared to smaller, isolated clusters.

In this study, we analyze the spatio-temporal characteristics of convective clusters within MCSs. MCSs are identified and tracked using both radar precipitation data (RADOLAN radar; Bartels et al., 2004) and lightning records from the EUropean Cooperation for LIghtning Detection (EUCLID; Schulz et al., 2016) network over Germany. Convective-type precipitation is classified based on its proximity to lightning strikes. To explore links between these clusters and local environmental conditions, we incorporate data from German Weather Service (Deutscher Wetterdienst, DWD) weather stations and the ERA5 reanalysis dataset (Hersbach et al., 2020).

We measure the spatial clustering of convection within MCSs using two novel spatial organization indices that quantify deviations from random distributions. Our preliminary findings suggest that convective clusters within MCSs become wider at higher temperatures, consistent with observations of larger CAPE (Convective Available Potential Energy) environments. Additionally, we observe a geographic trend in the location of convective clusters: they are more frequently concentrated in the southern portions of MCSs. However, under warmer conditions, a larger fraction of MCSs exhibit convective clusters on their northern edges. We hypothesize that this shift is driven by stronger convective instability ahead of the northern flanks of MCSs at higher temperatures. This effect may be linked to increased near-surface baroclinicity and horizontal temperature gradients at warmer temperatures.

The temporal evolution of these convective clusters is further analyzed through the framework of directed percolation, a statistical physics approach that allows us to investigate the growth and connectivity of convective cells over time. Through this lens, we aim to better understand the lifecycle of convective clusters within MCSs, including their formation, propagation, and eventual dissipation. By combining spatial and temporal analyses, this study provides critical insights into how environmental conditions influence the organization of convection within MCSs, thereby advancing our ability to predict and mitigate flood risks in a warming climate.