Evaluation of surface soil moisture of global products using measured data in different climates of Iran

Soil moisture is one of the most important variables required from different aspects of research such as agriculture (soil water balance, irrigation schedule, agricultural drought index) and hydrology (infiltration, runoff, recharge). Although the measured soil moisture data is the most accurate data available to researchers, it has disadvantages such as the length of statistical period, missing data in some stations, point source and the inability to extend them to an area such as a basin. Soil moisture data in global products do not have these disadvantages, but their accuracy must be evaluated before using. In the present study, in order to investigate the possibility of using surface soil moisture data from global products in Iran, the accuracy of the global products databases has been evaluated. The main aim of this research is to evaluate the accuracy of surface soil moisture data of global products based on the measured data of soil moisture variable in different climates of Iran. In this regard, the measured data of soil moisture in 43 agricultural meteorological stations across Iran were collected at 7 different depths (5, 10, 20, 30, 50, 70 and 100 cm from the soil surface) with a time step of three hours. After the preprocessing on the collected data, 14 stations from different climates of Iran with the longest and most complete soil moisture data for two depths of 5 and 10 cm with a monthly time step during the period of 2014-2021 were selected in order to compare with the data of global products. In the next step, soil moisture data in different soil layers were extracted from 5 different global products including GLEAM3.6a, ERA5, TERRA, MERRA2 and GLDAS2.1 with a monthly time scale. Surface soil moisture in the measured data, the average soil moisture at two depths of 5 and 10 cm, and in global products, the first layer of the soil surface was considered. The evaluation of the accuracy of surface soil moisture data has been done using three statistical criteria including Pearson's correlation coefficient (R), Mean Bias Error (MBE) and Normalized Root Mean Square Error (NRMSE). Generally, among the studied stations, the highest and lowest values of correlation coefficient are 0.55 and 0.04 respectively for Kahriz and Zahak stations. Between the seasons, the highest correlation has been obtained in the spring, April, equal to 0.51, and the lowest correlation has been obtained in the summer, September, equal to 0.11. The highest correlations are related to the GLEAM product, especially in the months of May, June, November and December. The results of MBE indicate that soil moisture is underestimated in most stations in humid and semi-humid climate stations (Gorgan, Karakhil and Amol stations). The GLEAM products with a range of [-0.03 ~ +0.13] has the least MBE changes among the studied products. The highest accuracy related to the ERA5 product, especially in the humid climate with NRMSE equal to 0.29. In addition, the lowest accuracy among the studied products is related to the GLDAS, especially in dry climate with NRMSE equal to 2.16. The results of this research show that the accuracy of the global products of soil moisture varies according to the temporal and spatial of case study. In order to select the appropriate product with the high accuracy based on the spatiotemporal changes of soil moisture, the present research provides practical results to the researchers.