Simulation and Sensitivity in a Nested Modeling System for South America. Part I: Reanalyses Boundary Forcing

Abstract

A regional climate model driven by reanalyzed atmospheric forcing is used to investigate 1) the large-scale circulation anomalies that were driven by sea surface temperatures (SSTs), which resulted in extreme rainfall anomalies during January?May 1983 (dry) and 1985 (wet) in tropical South America; 2) the effects of vegetation and soil moisture in the interior Amazon basin on regional circulations, moisture transport, and rainfall; and 3) the sensitivity of regional model results to domain size. Seasonal integrations demonstrated that by prescribing observed SSTs and applying reanalyses-derived forcing along the boundaries of the control domain, the regional climate model (RegCM) was able to simulate the dramatically different large-scale circulations in the two years, as well as the resulting rainfall differences. Thus, the large-scale forcing apparently has a first-order effect on the region. The regional model shows reduced rainfall in the western Amazon compared with observed estimates that are associated with weak low-level moisture transport from the Atlantic. The sensitivity experiments to surface forcing in the Amazon, employing a large (10.8 ? 107 km2) and a small (5.7 ? 107 km2) domain, show that both simulation and sensitivity are a function of domain size in the Tropics. However, the spatial scales and hence the domains required are larger in the Tropics than in the midlatitudes. The perturbations employed in this study influence the large-scale tropical circulation. This feedback is damped by the lateral boundary conditions of the control (smaller) domain.