Role of Land Surface Vegetation on Rainfall Bias in the CFSv2 Model During Early and Late Phases of the Indian Summer Monsoon

Although the Climate Forecast System version-2 (CFSv2) model simulates an overall dry bias in boreal summer mean rainfall over Indian land, the deficiency is particularly prominent over northwestern India. The dry bias not only limits the inter-annual prediction skill of the Indian Summer Monsoon Rainfall (ISMR) but also may reduce predictability at sub-seasonal time scales because of poor representation of latent heating due to weak moist convection and the resulting misrepresentation of circulation. In this paper we show that land surface vegetation plays a crucial role in determining the dry bias. We replaced the land surface models existing vegetation type over India with that derived from recent satellite-based observations. The modifications helped improve the seasonal mean rainfall over northwestern India by 10-20%. The improvements are especially noticeable in the early and later phases of the season. We further perform a few other idealistic experiments with extreme vegetation cover to show the role of spatial asymmetry in vegetation cover on the moisture flux and rainfall over land. Our study indicates necessity of greater attention to land surface representations and initialization for improved predictions of the seasonal mean rainfall.