Analysis of estimated precipitation data by remote sensing in the Sergipe state

Abstract

Currently, remote sensing data, such as that of the Tropical Rainfall Measuring Mission (TRMM), has been used to monitor the distribution of rain over time and space. The objective of this work was to evaluate the quality of the rainfall data estimated by the product 3B43-TRMM in the state of Sergipe, on the monthly and annual scales, between 1998 and 2013. The point values estimated by the TRMM were compared with the precipitation data obtained in 13 pluviometric stations of the National Water Agency (ANA). The statistical indications considered were the coefficient of determination (R2), mean absolute error (EMA), root of the mean square error (REQM) and Willmott's agreement index (d). The R2 values were 0.49 and 0.16 on the monthly and annual scales, respectively. For the monthly time scale, the best estimates of the TRMM product were found in the semi-arid region of the state of Sergipe, with values of R2, EMA, REQM and d equal to 0.54, 27.18 mm and 38.71 mm and 0.83, respectively.