How Well Do We Simulate Deep Convective Storms in Kilometer-Scale Regional Climate Models?

Deep convective storms play a central role in the global circulation and hydrological cycle and are responsible for a large fraction of extreme precipitation in many regions. Yet they have long been among the most persistent challenges for numerical weather and climate models. The transition from hydrostatic to kilometer-scale (km-scale) convection-permitting regional climate models (RCMs) has fundamentally changed how these storms are represented, enabling the explicit simulation of convective dynamics rather than relying solely on parameterization schemes. However, key physical processes such as turbulent entrainment, microphysical interactions, and surface–atmosphere coupling remain only partially resolved, raising important questions about model fidelity and robustness.

In this contribution, I assess how well kilometer-scale regional climate models simulate deep convective storms across a range of grid spacings and modeling configurations. Using idealized and real-case CP-RCM simulations, I analyze grid-spacing sensitivities in bulk storm properties such as precipitation intensity, vertical mass flux, and storm lifetime, as well as structural convergence in storm morphology and updraft dynamics. The results demonstrate that while bulk precipitation statistics can appear well constrained at kilometer scales, they often arise from compensating errors between microphysics, turbulence, and dynamics, masking deficiencies in the underlying storm processes.

Finally, I show examples in which land–atmosphere coupling and surface heterogeneity exert a first-order control on convective initiation and organization, highlighting the importance of consistent coupling in regional climate simulations. The findings have direct implications for the interpretation of CORDEX and other km-scale RCM ensembles, and point to priority areas for next-generation regional climate model development aimed at more physically grounded projections of extreme rainfall.