This study evaluates the applicability of using long-term satellite rainfall estimate (SRE) precipitation products in drought monitoring over mainland China under global warming conditions. Two widely used drought indices, the self-calibrating Palmer Drought Severity Index (scPDSI) and the Standardized Precipitation Evapotranspiration Index (SPEI), were selected as study cases; both indices consider global warming but based on different mechanisms. Two popular long-term SREs were selected to calculate the indices: the Precipitation Estimation from Remotely Sensed Information using the Artificial Neural Networks-Climate Data Record (PERSIANN-CDR), and the Climate Hazards Group Infrared Precipitation with Stations (CHIRPS). A ground-based gridded observation dataset known as the China monthly Precipitation Analysis Product (CPAP) was used as a reference for the evaluation. Research results showed that on a grid cell scale, the SPEI based on both SREs was consistent with observations in eastern China (correlation coefficient over 0.9), while the scPDSI was much less accurate (correlation coefficient of only 0.5) and its accuracy patterns were highly spatially heterogeneous. However, on a regional scale, after spatial errors were offset by spatial averaging, the performance of the SRE-based scPDSI improved, and it showed the same ability as the SPEI in temporally detecting the timing, intensity, and magnitude of drought. The self-calibrating procedure of the scPDSI was determined as the most probable cause of its poorer performance and high heterogeneity, which would increase instability and enlarge the uncertainty of the SREs. It is thus considered that the SPEI should be the first choice for use in monitoring global-warming related drought, primarily because of the high uncertainty and instability of the scPDSI.
Keywords: CHIRPS; Drought; Mainland China; PERSIANN-CDR; Satellite-based precipitation product.
Copyright © 2019 Elsevier Ltd. All rights reserved.