Impactos das mudanças climáticas em uma bacia hidrográfica no sul do estado de Minas Gerais

Abstract

Climate change in the 21st century is an undisputed reality. Several effects and impacts have been registered in society in several regions of the planet. In particular, increasingly severe extreme events are being observed, with floods and long periods of drought. This work objective is to evaluate an association of the hydrological model MHD-INPE with the atmospheric model Eta/CPTEC to simulate the impact of climate change in the Jaguara Creek Basin (JCR), located in the south of the state of Minas Gerais. For this purpose, the projections of three global climatic models (GCMs) downscaled by the Eta model (HadGEM2-ES, MIROC5 and CanESM2) were used for the scenarios RCP4.5 and RCP8.5. The MHD-INPE model was calibrated and evaluated using performance statistics. The results of the calibration and validation shows a good performance of the MHD-INPE model in simulating the JCR daily discharge. The Lavras Meteorological Station observations were used to eliminate the bias of atmospheric simulations for the baseline period (1961-2005) because it presents similar characteristics to the studied basin. The bias correction was applied for future scenarios (RCP4.5 and RCP8.5), which were subdivided in time slices (2006-2040, 2041-2070, 2071-2099). The results of the scenarios were used as input to the hydrological model after bias correction. Regarding the corrected climate projections, they resulted in a predominant reduction of the rainfall throughout the 21st century. The results of the projections of temperature showed an increase through the climatic models in both scenarios. In general, the statistical descriptors of the flow duration curve (FDC) resulting from the projections of the Eta model showed a good performance of the MHD-INPE in the simulation of the baseline period. The multi-model’s ensemble mean is consistent about of the reduction of the discharge in the future climate (2006-2040, 2041-2070 and 2071-2099), to RCP4.5 and RCP8.5 in all seasons of the year, reaching a reduction of 35.1% in the summer during the 2071-2099 time slice in RCP8.5 scenario. The MHD-INPE model proved to be able to predict impacts resulting from climate change in small basins such as JCR.