How Much Convective Environment Subgrid Spatial Variability Is Missing Within Atmospheric Reanalysis Data Sets?

Abstract Convective cloud processes are sensitive to environmental conditions that vary on scales smaller than reanalysis data sets (sub‐reanalysis scales). Convective environment variability within areas representative of reanalysis data sets is quantified using large‐domain, high‐resolution (∆ x = 100 m) simulations of convective cloud systems throughout the tropics and subtropics. Even after removing locations of resolved clouds and precipitation, convective environment parameters vary significantly on these scales. For example, for half of the simulated data, 500 hPa relative humidity varies by ∼30% within a typical reanalysis area. Surface winds, convective available potential energy, and middle‐tropospheric moisture are the most variable convective environment parameters for both continental and maritime regimes, while above‐surface temperature and winds are the least variable. While high‐resolution, sub‐reanalysis‐scale extrema are well‐correlated with the reanalysis‐area mean values, some of the most extreme convective environments can occur within regions with moderate reanalysis‐area mean values, particularly for continental regions.