Landsat remote sensing image data during 1979-2019 were selected to extract natural and constructed wetlands in the Dawen River Basin of the lower Yellow River. The spatial-temporal evolution and differentiation characteristics of wetlands in the Dawen River Basin and their driving factors were analyzed with the standard deviation ellipse method and spatial autocorrelation analysis. The results show that the wetland area in the Dawen River Basin had an increasing trend from 1979 to 2019, and the standard ellipse centroid migrated to the northwest. The watershed wetland had significant spatial autocorrelation aggregation characteristics, and the global mean Moran’s I was 0.67. According to the local Moran’s I, natural wetlands such as lakes and rivers showed “high-high” spatial distribution characteristics and were more likely to aggregate in space, while constructed reservoir and pond wetlands showed “high-low” and “low-high” spatial distribution characteristics and had spatial dispersion distributions, with large spatial heterogeneity. The wetland distribution in the Dawen River Basin was more significantly correlated with the aggregation of the spatial locations when there were no strong natural or human interferences. Extreme disasters and human activities, such as the drought event in the Dawen River Basin in 1989 and the start of the South-to-North Water Diversion Project in Dongping Lake in 2019, led to great fluctuations in wetland area, standard elliptical centroid, spatial aggregation, and heterogeneity in the river basin. The protection and restoration of ecological functions of Dongping Lake wetland play an important role in ensuring the water quality of the South-to-North Water Diversion Project, and it is the key to ensuring ecological functions of wetlands in the Dawen River Basin. Keywords: remote sensing; wetland; spatial autocorrelation analysis; standard ellipse method; Dawen River Basin