Determine whether explicit convection in ICON improves representation of individual extreme Sahel rainfall events

Determine whether enabling explicit convection in the ICOsahedral Nonhydrostatic (ICON) numerical weather prediction model improves the representation of individual extreme precipitation events in the Sahel compared to parameterized convection, i.e., whether the improvements seen for diurnal rainfall translate to event-scale extremes.

Background

The paper compares two ICON configurations at 6.5 km grid spacing: explicit convection (EXPLC) and parameterized convection (PARAM) using the Tiedtke-Bechtold scheme. Prior work (Pante and Knippertz, 2019) showed that resolving convection generally improves the representation of Sahelian rainfall systems, particularly the diurnal cycle.

However, the authors note that it is not established whether such improvements extend to the representation of individual extreme rainfall events. They analyze two contrasting Malian cases (San 2012 and Kenieba 2019) to explore this question, finding case-dependent performance, which underscores the unresolved nature of the general relationship between explicit convection in ICON and event-scale extremes.