Evaluation of Grid Precipitation data Products for Hydrological Application over Upper Northern Part of Thailand

Abstract

The purpose of this study was to evaluate and compare the datasets between 7 global gridded precipitation products including rainfall product GPCC - First Guess Monthly, PERSIANN - CCS, PERSIANN - CDR, GSMaP - MVK, GPM - 3IMERGM v06, TRMM - 3B43 and TRMM - 3B42 with the 90 rain gauged stations of the Royal Irrigation Department and the Meteorological Department in the upper northern region of Thailand in 2014 – 2020, as a point comparison. The statistical values such as correlation coefficient (R), mean error (ME), relative bias (RB), and square root mean squared error (RMSE) were analyzed. According to the investigation and analysis, there were differences in the amount of rainfall of each product, In statistical analysis, the R value is between 0.68 - 0.86, indicated that the gridded precipitation data of all products has a direct correlation with the station data, ME and RB values were between -12.31 - 17.19 mm and -10.2 - 24.46 %, while the RMSE values were between 54.73 - 69.53 mm. When analyzing the correlation between the statistical values and the digital elevation values (ASTER DEM) of the rain gauged stations, it was found that the grid products were estimated to increase the inaccurate rainfall with the elevation value.