Evaluation of drought characteristics based on combined global precipitation and runoff products datasets across Iran's sub basins

Recently, with the advancement of technology, extensive facilities have been provided for monitoring climate data with different resolution. Therefore, the aim of this research is to investigate the combination of precipitation databases and simultaneous monitoring of meteorological and hydrological droughts on a monthly time scale using the Combined Drought Index (CDI) at sub basins of Iran. To this end, observational data (100 synoptic stations on a daily scale) and global precipitation and runoff databases including ERA5, MRRRA2, GRUN, GLDAS and TERRA (with different spatial resolution on a monthly scale) were collected and extracted during the period of 1987-2019. Then, based on the aridity index, the climatic classification of the stations and sub basins has been done and the entropy weight method (EW) was employed to combine databases and indices. The accuracy of the datasets was assessed based on Kling Gupta (KGE) and Standardized Mean Square Error (NRMSE) metrics. The results showed that the combination of precipitation databases in the Hyper-arid, Arid, Semi-arid and Humid climates reduced the error index by 32, 10, 24 and 26%, respectively, compared to individual databases. In the 3-month scale, the sub basin of Sefidroud, Lake Namak, and Talesh, and in the 12-month scale, the sub basin of Lake Urmia, Lake Namak, Sefid Rud, Hamon Jazmourian, Hamon Mashkel, South Baluchistan, and parts of Aras River have faced with drought with more severity, duration, and peak values of drought. It should be said that, on average, in both scales, the middle part of the country has less drought severity and duration. The extent of drought in the river basins of Hyper-arid, Arid and Semi-arid climates in the scale of 3 and 12 months on average is 45 and 53%, and 70 and 40% in Humid climates. Based on these findings, it can be concluded that the combining precipitation databases enhances the accuracy compared to using global single databases. Additionally, drought in the humid climate shows higher intensity, duration, and extent in the short-term scale compared to the long-term scale.