Precipitation Bias over the Western Indian Ocean in an Atmospheric General Circulation Model: Implications for Studying Changes in South Asian Summer Monsoon

Most of current atmospheric general circulation models (AGCMs) show a remarkable positive precipitation anomaly over the western south-equatorial Indian Ocean (IO). The anomaly develops in spring and gradually extends eastward while moving northward, and then reduces by the end of the monsoon season.

Here, we employ the latest GFDL AGCM AM3, which is to be used for the IPCC fifth assessment (AR5), to characterize this anomalous precipitation pattern, and its influence on the development of the South Asian summer monsoon to the north. It is found that the precipitation bias is related to anomalous near-surface meridional convergence over the western IO (with a noticeable lower-tropospheric secondary circulation and overestimated low-level clouds), which is suggested to be in large part modulated by the local meridional sea surface temperature (SST) gradient. The enhanced equatorial precipitation is responsible for an anomalous Hadley-type meridional circulation over the Indian Ocean. As a result, the simulated monsoon evolution over India is greatly affected, with a significant delay in some areas.

A simplified AGCM, consisting of a spectral dry dynamical core nudged toward the AM3.0 zonal-mean climatological temperature and winds and forced by three-dimensional diabatic heating, is run as a diagnostic tool to investigate the impact of anomalous heat sources over the IO (mimicking the AM3.0 precipitation anomaly) on the circulation over the Indian Subcontinent. It is found that anomalous precipitation over the western IO leads to local ascent and upper-level divergence, and induces a poleward anomalous convergence and associated large-scale subsidence over northwestern India, with eastward Rossby wave propagation.

Sensitivity experiments with AM3.0 are also run to test the impact of the strength of a prescribed meridional SST gradient over the western IO, keeping the mean SST constant. As expected, the pattern of the vertical circulation (e.g., shallow versus deep convection) is very sensitive to the magnitude of the gradient, with important feedbacks on clouds, diabatic heating, and radiation.

These findings reveal that meridional SST gradients in the western south-equatorial IO play a crucial role in current AGCMs simulations of the South Asian monsoon, largely affecting their overall performance. Important implications follow: regional (e.g., aerosols) or global (e.g., greenhouse gases) climate forcing factors may alter the SST meridional gradient in the IO, with the potential to significantly affect the spatial and temporal evolution of the South Asian monsoon.