MJO events during DYNAMO Field Campaign: Cloud-Resolving Model and Radar Observation Comparisons

The Goddard Cumulus Ensemble (GCE) model is used to simulate the two MJO events in October and November 2011, during the DYNAMO field campaign. The emphasis is on comparing two sets of large-scale forcing: Forcing 1 derived exclusively from in-situ observations by Colorado State University group; Forcing 2, derived by the Lawrence Livermore National Lab group, utilized mainly ECMWF re-analyses constrained by radar rainfall observations. Simulations using both forcing are able to produce rich distributions of cloud population transitioning from shallow dominant, to congestus dominant, to deep convection with stratiform anvil, as the MJO events progressing. The global models generally have difficulties representing these details. It is hoped that the cloud-resolving model simulations will be able to shed some light on the mechanisms of the transitioning and ultimately on how to parameterize them in a GCM.

In order to validate cloud-resolving model simulations, TRMM data during the DYNAMO period, as well as S-POL and KAZR radar observations, are compiled and compared with model results. Each dataset have their strengths. TRMM satellite has large coverage over the whole sounding domain, but provides data only twice a day. S-POL offers continuous volume scanning, but has limited spatial coverage. KAZR has continuous coverage, too, at a single point. However, it offers data for shallow convection and non-or lightly- precipitating clouds which are missed by the S-POL. To ensure direct comparisons, the Goddard Satellite Data Simulator Unit (SDSU) was used to convert the model output to radar reflectivity, with the same sampling strategy as each instrument.