Shallow Cumulus Cloud Statistics in Real‐Case Large‐Eddy Simulations and Their Dependence on Soil Moisture

Abstract The ability of weather and climate models to reproduce the statistical properties of shallow cumulus clouds (SCCs) is critical for representing boundary‐layer transport, cloud cover, and cloud‐radiative effects. This study investigates statistical properties of SCC ensembles using the Icosahedral Nonhydrostatic model in large‐eddy simulation mode for three cases during the Field Experiment on Submesoscale Spatio‐Temporal Variability in Lindenberg campaign. Feature‐based tracking is applied to identify cloud objects and track their evolution. Each cloud is classified as active or passive according to its in‐cloud buoyancy excess. The cloud‐base mass flux distribution of each class is robustly captured by a Weibull distribution. Active clouds exhibit shape parameters , consistent with convective memory related to their life‐cycle, while passive clouds approach , indicating stochastic, short‐lived nature. The bulk cloud‐base mass flux , representing the convective transport of the SCC population, scales linearly with surface heat input per unit cloud moist‐static‐energy excess, weighted by thermodynamic efficiency of the surface‐driven moist heat cycle. This scaling emerges primarily across the soil‐moisture perturbations and holds from moderately moisture‐limited to energy‐limited regimes, but linearity breaks down under very dry conditions, where a suppressed evaporative fraction and the resulting sub‐cloud moisture deficit prevent further increases in . Soil moisture strongly controls cloud number, updraft strength, cloud size, and the mass‐flux distribution, whereas responds to both soil‐moisture perturbations and day‐to‐day large‐scale conditions, highlighting land‐atmosphere coupling as a key regulator of shallow convective variability over land.