Representation of Stratocumulus and Shallow Cumulus Cloud Fraction Near the Southeast Pacific Ocean ITCZ in ERA5 and MERRA‐2 Reanalyses

Abstract Model precipitation biases in the intertropical convergence zone (ITCZ) are often tied to the underestimation of stratocumulus (Sc) and shallow cumulus (Cu) clouds. We design a method to distinguish between Sc and Cu cloud regimes under subsidence on daily timescales based on cloud top pressure and vertical velocity to investigate the spatial distributions and seasonal evolution of their cloud fractions and associated large‐scale atmospheric conditions near the southeast Pacific ITCZ. We compare the Cumulus and Stratocumulus Cloudsat‐CALIPSO Data set (CASCCAD) with the European Center for Medium‐range Weather Forecasts ERA5 reanalysis and the Modern‐Era Retrospective analysis for Research and Applications, Version 2 (MERRA‐2). In both reanalyses, the spatial patterns, vertical structure, and seasonal cycle of Sc cloud fraction are faithful to CASCCAD observations, but the amplitudes are underestimated. Cu cloud fraction vertical structure and spatial patterns are well simulated in ERA5, but both reanalyses vastly underestimate Cu cloud fraction magnitude. MERRA‐2 cannot capture the Cu vertical structure and has no Cu clouds near the southern ITCZ in March. An investigation of Sc and Cu days suggests that both reanalyses have realistic conditions for Sc clouds, however, MERRA‐2 has significantly weaker vertical motions and larger cloud liquid water content during Cu days. Lastly, there is a significant increase in Sc cloud fraction and a decrease in Cu cloud fraction in the southern hemisphere when the northern ITCZs are prevalent. The opposite happens during southern ITCZs. Overall, despite similar mean southern ITCZ precipitation, ERA5 provides more realistic Sc and Cu cloud populations than MERRA‐2.