A non-stationary hydrological drought assessment method based on four-source driven
To scientifically evaluate non-stationary hydrological drought under the influence of the natural-social system,taking the impact of climate change(precipitation,temperature)and human activities(water intake,reservoirs)into consideration,a non-stationary standardized runoff index(NSRI)was constructed based on the generalized additive model for location scale and shape(GAMLSS)of four-source driven,and a non-stationary hydrological drought assessment method based on four-source driven was proposed.The evaluation results of non-stationary hydrological drought cases in the Yellow River Basin indicate that the runoff sequence at Huayuankou Station in the Yellow River Basin exhibits obvious non-stationary characteristics,with a significant decrease in runoff from 1961 to 2020 and a sudden change in 1990.The non-stationary hydrological drought assessment method based on four-source driven can better fit the non-stationary probability distribution of the runoff data.As for the identification of historical drought events,the drought process identified by NSRI is closer to reality.NSRI achieves better performance than the standardized runoff index(SRI)when identifying drought events,and can better reflect the evolution trend of hydrological drought in changing environments.