Spatial and temporal patterns of agricultural drought in China during 1960-2020 characterized by use of the crop water deficit Abnormal Index

Agricultural drought intensifies crop water stress and limits photosynthetic productivity, thereby threatening food security and socio-economic development. However, current drought indices from model simulations or remote sensing have limitations with regard to fully considering the complex interactions between various climate factors and crop water demands. Furthermore, estimates from remote sensing-based drought indices are subject to significant uncertainties due to the coarse spatiotemporal resolution of the data sources, inherent systematic errors in methodologies, and the limited number of ground observations. This study aimed to bridge these gaps by employing the nationwide Agricultural Drought Monitoring Network covering over 2000 in-situ observations and the Crop Water Deficit Abnormal Index (CWDIa) to quantify the spatiotemporal characteristics of agricultural drought from 1960 to 2020. The duration and severity of drought in northern China were significantly greater than in the southern regions, especially in the North China Plain, and the northern Tibetan Plateau. The occurrence of severe drought in southwestern China and in the southern Tibetan Plateau during the spring and winter seasons was driven by persistent water deficits. Over the past 60 years, China has experienced an overall alleviation in the severity and duration of agricultural drought. However, the drought situation has continued to worsen in certain regions, such as in southwestern China and the southern North China Plain. Multiscale decomposition of the ensemble empirical mode has further strengthened this finding (i.e., the drying trend in southwestern China since 1990). As seen in authoritative drought statistics, the integration of CWDIa and dense observations in the drought monitoring framework have demonstrated robust efficacy in capturing over 80 % of the actual drought-covered and drought-affected area. These findings highlight the importance of incorporating climate and crop water demand in agricultural drought monitoring to provide reliable and longterm spatiotemporal estimates of drought characteristics based on a nationwide observation network.