Tropical and Subtropical Cloud Regimes in MERRA Reanalyses Using an ISCCP Simulator

Considerable debate remains whether reanalyses can be used to identify long-term dynamical and physical climate trends given the nature of discontinuous data assimilation and the uncertainty associated with analysis fields prior to the epoch of global satellite coverage. Significant variability may also exist among datasets, particularly for those regions constrained by fewer observations (e.g., the tropical Hadley circulation). Whereas radiative feedbacks from simulated clouds often explain much of the sensitivity in global climate models (GCMs), no studies exist that document the variability of simulated cloud types and frequency using reanalyses.

This study presents a climatology of simulated low-latitude cloud regimes derived from MERRA and MERRA-2 data using an International Satellite Cloud Climatology Project (ISCCP) simulator. Physically meaningful cloud types are predicted from 25 years of reanalysis data using a clustering algorithm on joint histograms of simulated cloud-top pressure (pc) and optical thickness (τ). The ability of reanalysis to predict the full range of observed tropical and subtropical clouds or “weather states” is evaluated in addition to documenting the effects of including MERRA precipitation fluxes on the simulator results. Trends in the simulated cloud properties and cloud fraction as related to potential long-term changes in the tropical circulation are also discussed.

The simulated MERRA clouds qualitatively matched the observed distributions of pc and τ, though the tallest (pc < 310 hPa) and thickest (τ > 23) clouds were often missing from the reanalysis. The presence of tall, thin clouds is also unaccounted for in the simulated regimes, though this result was partially improved when using a random vertical cloud overlap parameterization. Convectively suppressed regimes are simulated well in MERRA data, albeit the reanalysis significantly underpredicts the observed cloud fractions for all the regimes. Finally, composite profiles of MERRA vertical velocity, temperature, and moisture for each ISCCP regime largely matched expectations and observations from previous studies, suggesting that the dynamic and thermodynamic properties of the cloud regimes are well captured by the reanalysis even if the simulated properties do no fully align with ISCCP observations.