Abstract: Abstract: Since more frequent and severe droughts will occur in China under the background of global climate change in the future, understanding temporal and spatial characteristics of drought and its potential impacts on agricultural production are especially significant for ensuring agricultural sustainability and national food security. In this study, we selected 241 prefecture-level units as the study area in 5 main maize planting regions of China: the northern spring-maize region, the Huang-Huai-Hai summer-maize region, the southwest mountain-hilly maize region, the south hilly maize region and the northwest irrigated maize region. Based on the AquaCrop crop growth model, we simulated the historical crop water stress and corresponding maize yield during the historical period (1980-1999) using ERA-Interim Daily dataset from European Centre for Medium-Range Weather Forecasts and CN05.1 dataset from Laboratory for Climate Studies, China Meteorological Administration, and simulated the future crop water stress and corresponding maize yield for two future periods: Mid-21st Century (2030-2049) and End-21st Century (2080-2099) under 3 representative concentration paths (rcp2.6, rcp4.5 and rcp8.5) using meteorological data from 5 global coupled models in Phase 5 of Coupled Model Inter-comparison Project (CMIP5). Based on the AquaCrop model's irrigation management function, we set up 2 irrigation scenarios in each simulation: rain-fed and complete irrigation. We defined drought hazard index as the daily average value of crop water stress indicator under the rain-fed scenario to indicate drought intensity and defined the drought loss as the difference between the crop yield under the rain-fed scenario and the complete irrigation scenario to indicate the loss intensity. Then, the change of historical and future drought intensity was compared and analyzed, and the yield loss rate under different climate scenarios in the future was discussed. After that, a regression model was created to evaluate the effect of drought intensity on yield loss on that basis. The results showed that: 1) In terms of spatial distribution, both drought intensity and maize yield loss in China were decreasing from northwest to southeast in the future under all representative concentration paths. The northwest irrigated maize region had the largest drought intensity and crop yield loss, followed by the northern spring-maize region, the Huang-Huai-Hai summer-maize region, the south hilly maize region and the southwest mountain-hilly maize region. 2) Except for a slight decrease in the southwest mountain-hilly maize region, China's drought intensity in the future would increase in general compared with the historical period, especially under rcp4.5 and rcp8.5 scenarios. Among the 4 maize regions, the increased rate of drought intensity was the largest in the south hilly maize region, followed by the Huang-Huai-Hai summer-maize region, the northern spring-maize region, and the northwest irrigated maize region. 3) The effect of drought intensity on yield loss in all rcp scenarios could be described well by a logistic curve, the R2 of the regression result was 0.96. The logistic curve showed that the yield loss rate of maize began to increase rapidly when the drought hazard index reaches around 0.2 and it approached the maximum value when the drought hazard index was near 0.6. 4）Drought would be most serious in Northwest China in the future, and its corresponding maize yield was estimated to more than 90% on average without irrigation. The drought hazard index of the northern spring-maize region and Huang-Huai-Hai summer-maize region were mostly in the middle of the "S" shape curve, where the slope of the curve was largest. Thus, the benefits of irrigation in these two regions would be the highest with the most sensitive response of yield loss to drought intensity. Besides, considering the maize production was also the highest in these two regions, adequate water should be provided for agricultural irrigation. The results of this study would provide an important reference for the development of regional agricultural drought adaptation strategy and have certain guiding significance for planning and reducing the potential negative impact of drought on agricultural production.