Acessibilidade / Reportar erro

Projections of precipitation and temperature changes over northeast brazil using the dynamical downscaling technique

ABSTRACT

This paper presents analysis of projected changes on precipitation and temperature perfomed by dynamical downscaling simulations in the CORDEX Project context. The simulation domain encompasses the Tropical Americas but the focus of this study is restricted to the Northeast Brazil subregions: North (NNEB) and South (SNEB). The regional model run in the simulations was RAMS6.0, forced by HadGEM2-ES global model data. The present period and three future time slices under the RCP8.5 climate scenario were analyzed. Current precipitation and temperature were compared against CRU observed data, to evaluate the performance and the biases from the regional and the global model. In comparison to the HadGEM2-ES model, RAMS6.0 is superior in some aspects as the phase representation of the annual precipitation and temperature cycle, reduction of the dry bias during the dry season over the NNEB and spatial distribution of the two variables, especially temperature. However, RAMS6.0 exacerbates some errors from the global model, like the wet bias in the rainy season. In the future climate, changes in the climatological value and annual mean fields of both variables were analyzed. Some patterns were projected by both models, like decreasing precipitation during austral winter and increasing temperature throughout the century.

Keywords:
Climate Change; CORDEX; Annual Cycle.

Sociedade Brasileira de Meteorologia Rua. Do México - Centro - Rio de Janeiro - RJ - Brasil, +55(83)981340757 - São Paulo - SP - Brazil
E-mail: sbmet@sbmet.org.br