Evaluating the Accuracy of Precipitation Products Over Utah, United States, Using the Google Earth Engine Platform

Satellite-based precipitation missions can be used to estimate precipitation distribution, especially in areas where there are no rain gauging stations. Nevertheless, these products are still less used because of the lack of accuracy evaluation. This study evaluates the monthly rainfall values of five satellite precipitation products, including ERA5, GPM, CHIRPS, TRMM 3B43, and PERSIANN-CDR, at eight rain gauge networks over the Utah, United States using Google Earth Engine platform (GEE). For this purpose, different validating indices such as R2, RMSE, and MAE were used to evaluate the accuracy of mentioned products from 2009 to 2019. The results showed that CHIRPS outperformed other rainfall products in this region with an R2 value of 0.63. ERA5 ranked second with an R2 of 0.6, and GPM, TRMM, and PERSIANN-CDR were in the subsequent ranks with R2 values of 0.53, 0.52, and 0.32, respectively. The results also indicated that spatial resolution is directly related to the accuracy of the results. CHIRPS rainfall product had the highest spatial resolution (0.05°) among all studied products, which led to the most reliable results. On the other hand, the lowest spatial resolutions belonged to TRMM and PERSIANN-CDR (0.25°), which resulted in the weakest results. The results also revealed that the ERA5 precipitation product was more influenced by elevation, longitude, and rainfall factors than other products.