Detecting multidecadal variation of short-term drought risk by combining frequency analysis and Fourier transformation methods: A case study in the Yangtze River Basin

Abstract

Due to the steady increase in global temperature, both long-term (seasonal or longer) and short-term (monthly to seasonal) drought events have become more frequent around the world (especially for short-term drought events). However, methods for detecting changes in short-term drought risk currently remain inadequate. We proposed a statistical method to quantify the risk of only short-term droughts by combining frequency analysis and Fourier transformation with phase-randomization. The effectiveness of the proposed method was demonstrated using the long-term data (1949–2018) of the largest river basin in China, i.e., Yangtze River Basin (YRB) where drought risk is high and short-term drought risks have rarely been evaluated. The mean short-term drought risk was 71.7 ± 11.9 % (mean ± one standard deviation). Regarding the temporal trend, short-term drought risk for 65.4 % of grids increased over time (0.23 ± 0.15 %·yr −1), of which 90.0 % were statistically significant (p < 0.05). Such spatiotemporal variability is possibly resulted from the joint influence of the Western Pacific Subtropical High, the Atlantic Multidecadal Oscillation and the Pacific Decadal Oscillation. This study proposed a new method to quantify the risk of short-term droughts that enable us to better manage short-term droughts. Our finding can help regional governments give more attention to short-term droughts and promote the rational allocation of regional drought preparedness resources in the YRB.