To improve the accuracy of karst agricultural drought monitoring and further reveal its driving mechanism,a downscaling study of GLDAS soil moisture was conducted based on MODIS-NDVI/LST data using a geographically weighted regression model;agricultural drought was identified based on SSI,and the spatial and temporal evolution and joint probability characteristics of agricultural drought at different time scales were analyzed;finally,the driving mechanism of karst agricultural drought was revealed by geographic detector.The results showed that:(1) The drought intensity at different time scales in Guizhou Province showed an overall weakening trend over the 20-years period,and the drought area also showed a decreasing trend;the drought intensity and drought frequency showed a spatial distribution pattern of high in the west and low in the east.(2) The joint characteristic values of drought showed a pattern of autumn > winter > growing season > summer > year-round > spring,indicating that agricultural drought of higher intensity and larger area was prone to occur in autumn and winter in Guizhou Province.(3) Karst development intensity,rainfall and elevation were the main driving factors of agricultural drought in karst agriculture,and there was a strong coupling relationship with their spatial distribution;the interaction of different factors showed both two-way enhancement and non-linear enhancement of SSI,and there were few combinations without significant differences among factors,indicating that agricultural drought was the result of synergy among factors.The results can provide a reference basis for the monitoring of agricultural drought in karst agriculture and the formulation of drought prevention and anti-drought measures.