Spatiotemporal patterns of agricultural drought and its linkage to large-scaled climate variability: a case study in Gansu, China

At climate transition zones, patterns of drought frequency and severity are shifting because of changing atmosphere-ocean oscillations under global warming. Increasing uncertainty of drought conditions affects regional development, especially related to the agricultural and ecological services. In a transition region between arid and humid climates, the agriculture production of Gansu at northwest China is sensitive and vulnerable to droughts. In this study, the soil water deficit index (SWDI) is used to quantify the spatiotemporal patterns of agricultural droughts in Gansu. Mechanisms between the SWDI and regional quasiperiodic atmosphere-ocean oscillations are explored to quantify how Asian monsoons and the Pacific sea surface temperatures affect deep convections and moisture sources to China. Between 1981 and 2019, weakening Asian monsoons and a cold phase of Central Pacific reduced moisture fluxes to Gansu and caused persistent drought conditions based on the SWDI. Interestingly, there were distinct overall positive vegetation trends in Gansu over the period. Based on vegetation types, the Gansu can be divided into two regions: the irrigated North and the rainfed South. North Gansu is water-limited and irrigation is needed in the region. South Gansu is mainly energy-limited croplands, and the increasing temperature promotes the regional vegetation growth. Moreover, the root zone soil water storage in South Gansu is much higher than that in North part of the province. The vegetations in the south are generally more resilient to agriculture droughts because of the larger soil water storage. Overall, these results have important implications for the local agricultural services and soil conservation programs, particularly for South Gansu having main rainfed croplands in the region. Based on these results, a framework is proposed for planning and managing agriculture droughts at climate transition regions based on atmosphere-ocean oscillations.