A comparative study of two land surface schemes in regional climate integrations over South America

[1] A case study comparison using two different land surface models in the Regional Spectral Model (RSM) is presented here. This comparison is motivated from recent studies that show a significant impact of land surface processes on the predictability of precipitation at seasonal to interannual scales. We find in this comparative study that coupled land-atmosphere interactions using the simplified simple biosphere (SSiB) scheme and a two-layer soil (control) model with uniform vegetation fraction yield mixed results. It does not clearly indicate the superiority of one scheme over the other for this particular case. The warmer mean surface temperatures simulated by RSM-SSiB improve the seasonal (January-February-March [JFM]) simulation over Amazon River Basin, along Brazilian Highlands, over Bolivian Plateau, and over central America relative to the control model. However, the RSM-SSiB runs exhibit a strong warm bias in the surface temperature over the subtropics of South America. The mean JFM onshore easterly flow from the tropical Atlantic Ocean and the low-level jet are relatively stronger in the SSiB model. The mean JFM precipitation from the RSM-SSiB model shows an improvement over Guianan Highlands, Venezeulan Llanos, and Amazon River Basin. However, the mean JFM precipitation of the control model over the Caribbean Sea, Central America, and equatorial eastern Pacific Ocean is better than the SSiB model. INDEX TERMS: 183