Forested mountain watersheds are important sources of freshwater supply and carbon sequestration, and there has been an interest how climate change and variability will impact forest ecosystem patterns and processes over short to long time scales. Forests are sensitive to hydroclimate variability and they adjust leaf area and duration in response to water and nutrient availability upon topoclimate variation. Therefore, the influence of hydroclimate variability on water yield is already manifested in emergent vegetation dynamics in space and time. Here we characterize catchment-scale vegetation patterns with normalized difference vegetation index (NDVI) from 30-year historic Landsat TM images in six forested headwater catchments in the Coweeta Hydrologic Laboratory, North Carolina, USA. Landsat Thematic Mapper (TM) provides a nearly three-decade multispectral record of landscape vegetation patterns at a fine resolution (30 m) over the period of global warming. We relate vegetation dynamics with seasonal water balances and low flow patterns from observed long-term hydrologic records (35 years). We found that vegetation patterns along hydrologic flow paths have been homogenized over time, which corresponds with long-term leaf litterfall and tree data (20 years) in the study site. This suggests that the hydrologic connectivity between upslope and downslope are decreasing, mostly driven by increased vegetation water use with temperature induced growing season increases. This study suggests that climate change effectively homogenizes landscape vegetation patterns at the catchment scale in the southern Appalachians.