Thresholds of key disaster-inducing factors and drought simulation in the Xilinguole Grassland

Abstract The frequency of extreme events such as droughts is increasing due to climate change. Grassland ecosystems are extremely sensitive to drought. This study aims to analyse how droughts respond to meteorological factors, establish a method to quantitatively evaluate the threshold values of key meteorological factors that can induce drought in different time periods, and explore a drought prediction model applicable to the Xilinguole Grassland. The Xilinguole Grassland has experienced an obvious aridification trend, with notable fluctuations in short dry and wet periods and the probability of drought (generally ranked in the order of moderate drought > light drought > severe drought > extreme drought). The Xilinguole Grassland is expected to face more intense drought conditions in the next 5 years. Among the meteorological elements, temperature, precipitation, water vapor pressure and solar radiation are the key factors influencing drought. Both short-term and long-term changes in precipitation and temperature have a strong influence on drought. Short-term changes in water vapor pressure have a significant impact on drought. Drought is sensitive to extreme climate events and is most sensitive to the influence of meteorological factors in May–September, particularly in July. In July, drought can be induced by any of the following: monthly precipitation below 50 mm, an average temperature above 23 °C, sunshine percentage above 69%, a monthly sunshine duration above 319 h, or an average water vapor pressure below 12 hPa. In addition, Markov chain modelling is a feasible approach for predicting drought in the Xilinguole Grassland. This study provides a scientific basis for drought prevention and control.