Characterizing spatio-temporal patterns of multi-scalar drought risk in mainland China

Abstract Drought is complex geographical elements, which endangers China's food security, with coexistence of spatial and temporal heterogeneity and similarity. Thus, the recent abnormal characteristics of Chinese droughts warrant further study. This paper proposes a three-step framework to identify the spatiotemporal evolution characteristics of drought events in China Mainland, using the standardized precipitation evapotranspiration index (SPEI) at different time scales (1, 6 and 12 months). And then the drought characteristics are separated using the run theory. Finally, the spatiotemporal evolution patterns of meteorological, agricultural and hydrological drought risk in distance are investigated based on the geographic statistics analysis method (i.e., global and local Moran's index). Results show that the drought trend is gradually increasing in some basins, especially the Yellow River Basin, the Hai River Basin and the Southwest River Basins. The multi-variate attributes of drought characteristics in spatiotemporal evolution patterns at different time scales have the characteristics of regularity and mutation, for example, the drought frequency in northwestern China shows a low values in a low value neighbourhood, but the drought duration and intensity shows a high values in a high value neighbourhood. This indicates that the hydrological drought disaster in the region is extremely prominent. In addition, the risk of drought in inland River basins is the highest, followed by the Hai River basin and the Yellow River basin; the Pearl River basin has the lowest. Our results offer insights into the development of drought in China and provide support for disaster risk assessment and response planning.