Abstract: Vegetation structure of the riparian buffer zone is of great importance for maintaining various ecological services. Therefore, selecting a typical riparian buffer zone in the transition region from the hills to the plain area of the middle and lower reaches of the Yellow River as the research object, we analyzed the spatialtemporal dynamics of vegetation landscape pattern and the associated drivers, via landscape pattern analysis, buffer analysis, and regression analysis, based on the high resolution remote sensing images in 2003, 2009, and 2015 observed from the Google Earth satellite. The results showed that croplands were the landscape matrix of the typical riparian buffer zone of the middle and lower reaches of Yellow River, and riparian vegetation (forest and grassland) exhibited an increasing trend during 2003-2015. Influenced by the wandering of the Yellow River and agricultural reclamation and abandonment, the mutual transformation between water area and flood pain was frequent, and the transformation frequency from cropland to other landscape types was high. The main transformation directions of forest land were grassland, residential area, and cropland, and the main transformation directions of grassland were cropland and forest land. The spatial distribution of changes in vegetation was different in different periods. Landscape pattern analysis indicated that the landscape connectivity of vegetation in the study area was high, the degree of landscape fragmentation increased gradually, and patch shape tended to be more complicated. However, different vegetation types showed different trends, therein, geometrical shape of forest and grassland patches became more complicated and regular respectively, and spatial connectivity of forest land was higher than that of grassland. Grassland and forest patches showed obvious gradient features along the residential areas, riparian ditches, ponds, and road buffers. Meanwhile, regression analysis indicated that changes in grassland and forest patches were driven less by riparian and road buffers, but significantly by the expansion of residential areas. The response of changes in grassland and forest patches to ditches and ponds were different, therein, changes in grassland patches were significantly driven by ditches, while changes in forest patches were significantly driven by ponds.

Key words: dynamic change, evaluation, Danjiangkou Reservoir, dynamic capacity, spatial information technology