Abstract: The Zhangjiakou–Chengde district is an important ecological barrier and water conservation area in the Beijing–Tianjin–Hebei region. Groundwater-dependent ecosystems (GDEs) are therefore crucial for regional ecological stability and water security. Groundwater-dependent vegetation (GDV) constitutes the core component of GDEs. However, studies on the identification of GDE and GDV distributions and on groundwater depth thresholds required to sustain GDV in the Zhangjiakou–Chengde district remain limited. Based on multi-source remote sensing, meteorological, and groundwater monitoring data from 2018 to 2024, the year 2023, a drought year, was selected as the reference year for identification. Available precipitation was calculated using a surface energy balance model and satellite precipitation data. Potential GDEs were delineated by integrating land-use data and groundwater depth data. Within the potential GDE areas, NDVI and groundwater depth were standardized using Z-scores. Areas where the standardized NDVI and groundwater depth showed a significant negative linear correlation were identified as GDV. A quantile-based statistical method was then used to determine groundwater depth thresholds associated with vegetation degradation. GDV in the Zhangjiakou–Chengde district was mainly distributed in northeastern and southwestern Chengde, accounting for 1.5% of the total regional area. In the Bashang area, GDV was dominated by meadow vegetation, whereas in the Baxia area, it was mainly composed of meadows, shrubs, shrublands, and birch forests. Vegetation in the Bashang area, Baxia area, and the entire Zhangjiakou–Chengde district was more likely to decline when groundwater depth exceeded the local mean by 0.78 standard deviations. The corresponding regional risk thresholds of groundwater depth for vegetation degradation were 5.71 m, 8.00 m, and 7.41 m, respectively. These risk thresholds were consistent with the actual thresholds associated with unhealthy vegetation conditions. Among different vegetation types, meadow had the shallowest actual groundwater depth threshold for degradation, at 5.48 m, whereas aspen had the deepest threshold, at 16.60 m. The ecological thresholds of GDV in the Zhangjiakou–Chengde district exhibited pronounced spatial heterogeneity. The actual groundwater depth thresholds corresponding to different health states of typical vegetation types were consistent with their rooting depths. This study provides a simple approach for delineating GDV distribution and confirms the relationship between rooting depth and groundwater table depth for groundwater-dependent vegetation in the Zhangjiakou–Chengde district. The findings offer a quantitative basis for groundwater resource management and ecological restoration in the region.