Journal of Climate, 25, 4641-4659.

The tropical subseasonal variability simulated in the NASA GISS general circulation model

Daehyun Kim
Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY.

Adam H. Sobel, Anthony D. Del Genio, Yonghua Chen, Suzana J. Camargo, Mao-Sung Yao, Maxwell Kelley, Larissa Nazarenko


The tropical subseasonal variability simulated by the Goddard Institute for Space Studies (GISS) general circulation model (GCM) Model E2 is examined. Several versions of the Model E2 were developed with changes to the convective parameterization in order to improve the simulation of the Madden-Julian oscillation (MJO). Consistent with previous studies, the simulation of the MJO is found to depend strongly on cumulus parameterization. When the convective scheme is modified to have a greater fractional entrainment rate, Model E2 is able to simulate MJO-like disturbances with proper spatial and temporal scales. The improvement in MJO simulation comes at the cost of increased biases in the mean state, consistent in structure and amplitude with those found in other GCMs when tuned to have a stronger MJO. Increasing the rate of rain re-evaporation has additional positive impacts on the simulated MJO. MJO life-cycle composites of various model fields in a selected version of Model E2 demonstrate realistic features of the simulated MJO. By re-initializing a relatively poor-MJO version with restart files from a relatively better-MJO version, a series of 30-day integrations is constructed to examine the impacts of the parameterization changes on the organization of tropical convection. The poor-MJO version with smaller entrainment rate has a tendency to reduce the contrast between dry and wet regimes, so that tropical convection becomes less organized. Besides the MJO, the number of tropical cyclone-like vortices simulated by the model is also affected by the changes in convection scheme. With a larger entrainment rate, the model simulates a smaller number of such storms globally.