Assessment of Real Time, Multi-Satellite Precipitation Products under Diverse Climatic and Topographic Conditions

Abstract

A preliminary assessment of real-time multi-satellite precipitation products, Global Precipitation Measurement (GPM) IMERG real time (IMRT), TMPA 3B42 real time (TMPA), PERSIANN (PERS) and CMORPH (CMOR) was carried out in Pakistan on entire study area basis and by dividing the study area in to five regions. It is extended from 61o to 77o East and 23.5 o to 37 o North, with varied climatic and topographic conditions. In south coastline along the Arabian Sea, plateaus, deserts and plain areas in the center and in north snow caped mountains. A set of evaluating (continuous) statistical indices, correlation coefficient (CC), bias (Bias), root mean square error (RMSE) and detective (categorical) statistical indices, probability of detection (POD), false alarm ratio (FAR), critical success index (CSI) were applied to evaluate the satellite products against rain gauge data at daily time scales. However CC values were computed at monthly time scale and comparison between average annual values of satellite data and gauge values was also conducted. At annual basis significant resemblance was observed between spatial distribution of satellite and gauge data. At monthly time scale satellite products showed correlation in the range of 0.7 to 0.9 against reference data at major portion of the study area. But at daily time scale a weak correlation from 0.2 to 0.45, higher values of bias and RMSE in hilly terrains with highest average annual rainfall and the adjacent foot hill areas as compared to the plain areas, lower values of POD, CSI and higher values of FAR were observed. An inter comparison among the used satellite products revealed that overall performance of IMRT was better in the study area.