An effective approach to evaluate GCM simulated diurnal variation of clouds

Cloud radiative effects strongly depend on diurnal variations of insolation and cloud radiative properties. In general circulation models (GCMs), even when the daily-mean cloud properties agree with observations, errors in cloud diurnal cycle can still significantly impact the shortwave radiation and induce model biases. However, this aspect is overlooked in GCM evaluation and intercomparison programs (e.g., Coupled Model Intercomparison Project Phase 5 (CMIP5)), which mainly consider the daily-mean cloud fraction. This study presents a simple approach of using a diagnostic parameter, the effective-daytime cloud fraction which accounts for the concurrent variation of clouds and insolation, to reveal GCM biases in cloud diurnal variations. The usefulness of the approach is illustrated by the significant biases of cloud diurnal cycle in the Modern-Era Retrospective analysis for Research and Applications (MERRA) reanalysis when compared with that in the International Satellite Cloud Climatology Project (ISCCP) data. It is thus suggested that the parameter be included as one of the GCM diagnostics for evaluating cloud diurnal cycle in model intercomparisons.