Statistical assessment of IMERG-FRV6, MSWEP, TRMM-3B43V7, and PERSIANN-CDR satellite precipitation in monthly, seasonal, and annual time-scale over Iran

Abstract

Abstract Knowing the spatial and temporal distribution of precipitation around the world is beneficial for improving climate knowledge and enhancing weather and climate forecasting models. While determining the distribution of precipitation is complicated, many satellite precipitation products (SSPs) have been developed in the past few decades to estimate precipitation with sufficient coverage and accuracy. In this research, the performance of four SSPs namely IMERG-FRV6, MSWEP, TRMM-3B43V7, and PERSIANN-CDR were evaluated at monthly, seasonal, and annual scales in Iran. For this purpose, the measured rainfall data in 81 synoptic stations located in the entire country of Iran in the period of 2008 to 2019 were used. For a more accurate evaluation of the selected SSPs, several statistical indices including Correlation Coefficient, Kling-Gupta Efficiency, Root Mean Square Error (RMSE), and Bias were calculated and analyzed at the location of all synoptic stations. The results indicate that, in general, MSWEP has a significant advantage over other products in all time scales. Of course, the performance of all four products in areas with high monthly rainfall is associated with more errors. Our assessments show that the highest amount of monthly RMSE is observed in PERSIANN-CDR. TRMM-3B43V7 performance is more satisfactory in drier regions, i.e., areas with low to moderate precipitation. It is worthy to mention that MSWEP has the closest average precipitation to the observational data in spring, summer, and winter. Furthermore, IMERG-FRV6 overestimates precipitation in all seasons.