Climate change, topographic features, soil components, and hydrological processes in desert regions effectively control vegetation distribution. Identifying key drivers of vegetation dynamics under climate-hydrological changes is crucial for policymaking. This study focused on oasis vegetation in the Taklamakan Desert, China. The study employs Geodetector to analyze the drivers of oasis vegetation change using structural equation modeling. Multi-year Landsat imagery was used; the relative distribution frequency of surface water was constructed using the water body index and the threshold segmentation method, and meteorological and topographical factors were combined to quantify the effects of potential factors and their interactions on vegetation changes. Surface water frequency and soil properties influenced the spatial distribution of the normalized difference vegetation index (NDVI) for Populus euphratica, whereas, for Tamarix chinensis, meteorological factors and soil environmental factors in the middle and deep layers were the main influences. Interactions had a greater impact on NDVI than individual factors; the effects of various drivers on P. euphratica and Tamarix chinensis NDVI were characterized by nonlinear and mutual enhancement effects. Structural equation modeling indicated that hydrological (0.35) and topographical (-0.28) factors affected P. euphratica NDVI; meteorological (0.27) and topographical (-0.18) factors affected the T. chinensis NDVI. These findings elucidate the impact of environmental factors on desert vegetation patterns, offering a scientific basis for the rational development of water conveyance programs, as well as the restoration and development of desert riverbanks.