Assessing Cloud and Precipitation Properties on Temporal Scales Using LASSO Simulations Over the ARM Eastern North Atlantic (ENA) Site

Abstract

The representation of low clouds and precipitation remains a major source of uncertainty in Earth System Models (ESMs), particularly due to challenges in representing their sub‐grid variability and scale‐dependent sampling. This study evaluates the performance of preliminary simulations from the Large‐Eddy Simulation (LES) Atmospheric Radiation Measurement Symbiotic Simulation and Observation (LASSO) project over the Eastern North Atlantic (ENA), focusing on liquid water path (LWP), ice water path (IWP), cloud fraction (CF), and surface precipitation rate across closed‐cell, open‐cell, and transitional cloud regimes. Using LES (100 m horizontal grid spacing) driven by ERA5 and MERRA‐2 reanalyses, we assess the representativeness of ground‐based point observations by analyzing their correspondence to model‐resolved temporal means. Results suggest that observational sampling of at least 6 hr is required to achieve consistency with domain‐scale averages, in particular for observations that exhibit pronounced sub‐grid heterogeneity, such as precipitation. Differences between ERA5‐ and MERRA‐2‐driven simulations are detectable but secondary relative to regime and temporal‐scale effects, and diminish with extended averaging. These findings highlight the importance of regime‐aware model evaluation strategies and potentially demonstrate how LES can inform observation‐model comparison practices and the development of cloud and precipitation parameterizations in ESMs.