Response mechanism of agricultural drought to meteorological drought in the Yellow River Basin
Dehe Xu, Menglong Wang, Ke Zhang, Qi Zhang, Shiwen Wu, Jianli Fu
Abstract
Meteorological drought and agricultural drought interact through the land-air system and hydrological cycle. Understanding how the latter responds to the former is crucial for effective drought mitigation. This study aims to explore the spatially heterogeneous response mechanisms of agricultural drought to meteorological drought in the Yellow River Basin. Based on analyzing the characteristics of meteorological and agricultural drought, Pearson's correlation coefficient was used to evaluate the response time, and then a geographically weighted regression model was used for the first time to study the response mechanism, as it accounts for spatial heterogeneity in environmental influences. The results show: (1) Short-term meteorological droughts are concentrated in the southern basin, with longer durations in the west. In contrast, agricultural droughts occur more frequently in the southwest, but tend to persist longer in the northwest. (2) Response times vary across the basin, ranging from 5 to 12 months in the north, predominantly 12 months in the central region, and 11–12 months in the south. (3) Wind, maximum temperature, and minimum temperature strongly influence response time, with mean GWR coefficients of 0.2121, 0.0357, and −0.0301, respectively. Among them, precipitation, minimum temperature, and hours of sunshine had a negative effect on agricultural drought response time, and wind and maximum temperature had a positive effect. These findings provide a spatially explicit understanding of drought interactions and highlight the importance of targeted water resource and agricultural management strategies. • Meteorological droughts are roughly bounded by the year 2000, with rising severity before and relief after. • Agricultural droughts show a slow increasing trend. • Agricultural response to meteorological drought in the central and southern parts of the basin is mostly 12 months. • For attribution of agricultural response to meteorological drought, GWR has higher accuracy than linear regression. • Wind and temperature had relatively large effects on drought response time.