Evaluation of drought monitoring in China using IMERG satellite precipitation products

In the Global Precipitation Measurement (GPM) era, the Integrated Multi-satellitE Retrievals for GPM (IMERG) with high spatiotemporal resolution (spatial resolution of 0.1° and the maximum temporal resolution of 0.5 h) has become one of the most mainstream satellite precipitation products.This study aims to primarily evaluate the ability of the latest retrospective IMERG Final Run precipitation product for drought monitoring in China. The China gauge-based monthly precipitation analysis product (CPAP) gridded data was selected as the reference. Two dimensionless and commonly used meteorological drought indices, including the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI), and a comprehensive drought index, the Composite Index of meteorological drought (CI), were calculated using IMERG and CPAP data at conventional scale, respectively. The performance of IMERG product was verified for drought monitoring, where three drought indices were compared over China from June 2000 to December 2017. The results showed that: 1) The IMERG product can well capture the spatial pattern of mean monthly precipitation in China (spatial correlation coefficient of 0.96 and relative bias of 0.02), indicating high consistency with the CPAP data, especially in humid and semi humid areas. The regions with a correlation coefficient over 0.9 were occupied 73.7% of the continental area. However, there was relatively high spatial heterogeneity in the Xinjiang and Qinghai Tibet Plateau regions. 2) The SPI and SPEI were calculated using IMERG product, where the regional mean correlation coefficient was higher than 0.8 in most sub-regions, indicating high reliability and good consistency with the CPAP data at multiple temporal scales (1, 3, 6, and 12 months). Specifically, the IMERG-based SPEI and SPI presented the best performance on the 12-month scale. Moreover, the CI was calculated using IMERG product, showing good usability with the regional mean correlation coefficient 0.756. Generally, the IMERG-based SPEI had better applicability than the SPI and the CI in most regions. 3) In southwestern China, the IMERG can accurately reflect the spatial and temporal variations of drought (drought intensity and drought area), particularly for the IMERG-based SPEI. Temporally, the correlation coefficient between the IMERG-based and CPAP-based SPEI was 0.968 in this region. Spatially, the IMERG-based SPEI precisely captured the spatial characteristics of typical drought disasters on the specific month (March 2010) as well. In short, the retrospective IMERG precipitation product performed a great potential for drought monitoring over China, where the accuracy was influenced by the selection of different drought indices. If the precipitation factor was considered only, the IMERG-based SPI could be applied to identify and evaluate meteorological drought at large scale. However, under the background of global warming, a suggestion could be that the IMERG-based SPEI had a higher potential than the SPI and the CI (considering precipitation and temperature factors) to be utilized in drought monitoring, particularly in semi-arid and arid areas. The IMERG-based CI was suitable for coastal areas, such as Northern China, the middle and lower reaches of the Yangtze River, and Southeastern China. © 2021, Editorial Department of the Transactions of the Chinese Society of Agricultural Engineering. All right reserved.