In recent years, the water level of the Dongting Lake (DTL) has been continuously low, and the wetland area and landscape pattern have changed significantly. Considering the obvious spatial heterogeneity of water regime changes in different waters of the DTL, this paper takes two core areas of the South Dongting Lake Nature Reserve (SDLNR) as study areas and analyzes the spatial distribution characteristics of the wetland blue–green landscape patterns by using remote sensing image data and hydrological and meteorological data. The multi-scale correlation between runoff, precipitation, temperature, and evapotranspiration in the SDLNR was studied via cross-wavelet transform analysis. The results show the following: (1) The change in the blue–green spatial patterns in different regions in different periods is inconsistent, and this inconsistency is related to the topography, climate, and human activities in each region; (2) there are seasonal fluctuations in precipitation, air temperature, and evapotranspiration in the SDLNR. Among them, the annual mean temperature shows a rising trend and passes the significance test with 95% confidence, while the annual mean precipitation and annual mean evapotranspiration show no significant change trend; and (3) our Pearson correlation analysis and cross-wavelet change results show that precipitation and temperature are strongly correlated with runoff, with a resonance period of 8–16 months, while the correlation between evapotranspiration and runoff is not significant. We recommend that policymakers establish an effective early warning system and make plans to store water through micro-terrain transformation in possible climate change treatments and strategies.