Drought was one of the main natural disasters which influencing agricultural production in China. As the largest maize production region in China, the frequent droughts caused by climate change have significantly affected the production of spring maize in the Northeast China. In this study, the risk of spring maize drought and its spatial pattern under future climate scenarios in the Northeast China was assessed, so as to provide a scientific basis for preventing spring maize drought and ensuring high-stable spring maize yield.

The spring maize potential planting area was focused on. Based on the daily weather data of three climate scenarios (shared socioeconomic pathways, SSPs), i.e., SSP1-2.6, SSP3-7.0, and SSP5-8.5 output from MPI-ESM1.2-HR model under the Inter Statistical Impact Model Intercomparision Project (ISIMIP) from 1981 to 2060 and the spring maize phenology data from 53 agro-meteorological observation stations, the Crop Water Deficit Index (CWDI) as the agricultural drought index was selected to analyze the spatio-temporal characteristics of different levels of drought in different growth periods of spring maize in Northeast China. The optimal probability theory distribution function was selected to estimate the probability of drought index series. The drought risk index was constructed by using different levels of drought risk for spring maize at each point estimated based on information diffusion theory. Then, spring maize drought risk in the Northeast China was assessed under different climate scenarios and future changes in the spatial pattern of risk areas by class.

(1) The drought index of spring maize in the whole growth period in research region from 1981 to 2014 was generally characterized by a high drought in the southwest and low in the northeast, showing that the four eastern leagues of Inner Mongolia (57.3%)>Heilongjiang Province (40.6%)>Liaoning Province (39.5%)>Jilin Province (38.9%). (2) The drought intensity in the middle of spring maize growth was overall higher than that in the early and late growth periods in the study area. In the 2030s and 2050s, the probability of drought risk in early growth was light drought>moderate drought ≈ severe drought>extreme drought. The probability of drought risk in middle growth was extreme drought>severe drought>light drought ≈ moderate drought. And the probability of drought risk in late growth was light drought>medium drought>heavy drought>exceptional drought. (3) From 1981 to 2060, under the SSP1-2.6 low emission scenario, the probability of occurrence of higher-grade drought risk for spring maize in Northeast China decreased, and the extremely high and higher drought risk zones was obviously shrink to the southwest, with the area share decreasing by 5.4% and 9.6% in the 2030s, and by 0.8% and 2.5% in the 2050s, respectively; while under the SSP3-7.0 and SSP5-8.5 two high emission scenarios, the probability of occurrence of higher-grade drought risk increased, and the higher drought risk area expanded to the northeast, with the area share increasing by 8.5% and 9.7% in the 2050s, respectively.

According to the spatial and temporal distribution pattern of future drought risk based on the drought risk index, the drought risk of spring maize in Northeast China decreased from southwest to northeast, and the higher-grade drought risk area expanded to northeast under the scenarios of SSP3-7.0 and SSP5-8.5 in the future, so it was necessary to pay attention to the critical growth period of maize to propose targeted defense measures.