Assessing rainfall spatial variability in the Brazilian savanna region with TMPA rainfall dataset

Abstract

Rainfall is a major component of the hydrological cycle. The lack of basic information on rainfall spatial variability is a major source of uncertainty in many fields of studies, e.g., meteorology, hydrology, and agriculture. Satellite rainfall measurements are becoming increasingly popular for their extensive database and reliable estimates. The objective of this study was to use the Tropical Rainfall Measuring Mission (TRMM) dataset along with rain gauges to characterize aspects of rainfall spatial variability and discuss the possible impacts from recent trends in the Brazilian savannah biome (Cerrado). Gauge stations were used to assess TRMM bias error and calibrate data for further analyses. Rainfall rates and their spatial variability showed a strong relationship to the transition zones between different biomes. Rainfall showed a decreasing trend for the eastern region of the Cerrado biome, a region characterized by a recent and significant expansion of crop areas. These trends agree with results from different studies which highlight the current drawdown of groundwater levels and reduced discharge, and possible lengthening of the dry season in the long run. As many conflicts have already been documented for this region, these decreasing trends are alarming for urgent and consistent hydroclimatic monitoring, and improved water resources planning and management. Positive trends for rainfall in the central Cerrado are likely a momentary recovery of recent-period drought.