A Regional Modeling System for Weather and Climate Studies in South America

Ana Nunes
Universidade Federal do Rio de Janeiro
Noon April 21 in Room 2155

A version of the NCEP Regional Spectral Model, coupled to the Noah Land-Surface Model, was successfully implemented at the Federal University of Rio de Janeiro in Brazil, as part of an integrated modeling system for the analysis of vulnerability and risk caused by severe weather or extreme climate events. Specifically for climate studies, this regional modeling system employs a new boundary forcing based on scale-selective bias correction and precipitation assimilation. Similar to the spectral nudging technique for dynamical downscaling, the scale-selective bias correction allows high-resolution modes to develop, but maintains the large-scale features from the boundary conditions, whereas the precipitation assimilation procedure acts on the modeled deep-convection and drives the land-surface scheme variables. In this newer version, the scale-selective bias correction was applied only on the rotational part of the wind field, letting the assimilation of satellite-based precipitation estimates to correct moisture convergence. Thus, the two techniques can be used together to improve the regional modeling system solutions, particularly in the tropics and subtropics where moisture convergence can be difficult to depict. Comparisons with the NCEP Climate Forecast System Reanalysis outputs will be shown at resolutions of approximately 38- and 25-km of the regional modeling system.