Effects of microphysics, shallow and deep convection on MJO simulations.

Abstract:

The nonhydrostatic global Model for Prediction Across Scale (MPAS) was used to simulate the November MJO event during the DYNAMO field campaign. Three mesh configurations were used: two with global uniform 60 km and 15 km grid spacing, respectively, and one with a 3 km grid spacing over the tropical Indian Ocean and 50 km for the rest of the world (50/3). The two cumulus parameterization schemes used in this study, Tiedtke and Grell-Freitas (GF), both include separated deep and shallow schemes that can be turned on and off individually. The GF scheme is scale aware. The Single Moment 6-class (WSM6) microphysics scheme was used. MPAS reproduced the MJO event with various degrees of success when the two cumulus schemes and their deep and shallow components were turned on and off. Microphysics and shallow convection each would help to reproduce MJO signals, with greater success at the 50/3 km mesh than 15 km. A deep cumulus scheme, on the other hand, may help or ruin MJO signals produced by microphysics and shallow schemes. The role of shallow convection is to provide moisture in the lower troposphere to reduce dry biases in the model and, in consequence, increase total precipitation. The capability of the model to produce MJO signals does not depend on the mean vertical profiles of diabatic heating profile, but are sensitive to mean vertical profiles of the moisture sink, which vary evidently with the shallow cumulus scheme.