Simulations of the South American Monsoon System: A Multi-Model Study

The present study focuses on a major component of the hydrological cycle: the South American Monsoon System (SAMS). The variability of monsoons affects populations and economies, and their better simulation and prediction at regional scales can help mitigate climate change impacts. We aim to provide a comparative assessment of simulations by global high-resolution models. The topics addressed are climatology, hydrology, intraseasonal and interannual variability, and effects of remote climate variability. The global models used are the coupled atmosphere-ocean general circulation models (GCMs) participating in the Coupled Model Intercomparison Project Phase 5 (CMIP5) and several versions of the NCAR Community Atmospheric Model version 4 (CAM4) with prescribed sea surface temperature. The CAM4 simulations are made with dynamical cores that differ in numerics as well as in spatial resolution. Simulations performed with CAM4 coupled with the Community Land Model (CLM) are also used in this study. For model evaluation we use the NCEP Reanalysis, the CPC Merged Analysis of Precipitation data (CMAP) and data from TRMM (the Tropical Rainfall Measuring Mission). Our main focus is on the period from 1979-2010 (present climate), but we also consider the period from 2050-2100 (future climate).

Our scientific goals are three-fold. The first goal is to analyze the simulations of SAMS made with the global models, assessing how well the models capture fundamental mechanisms of the monsoon system at different spatial and temporal resolution. The second goal is to evaluate whether or not increasing model resolution to more explicitly represent finer scale processes is a key to improving simulations of the hydrologic cycle. The last and overarching goal is to understand the variability within the system and how it relates to the myriad of processes affecting precipitation, which provides the dominant forcing for the surface water budgets.