This dataset provides a spatially explicit classification of potentially groundwater-dependent vegetation (pGDV) in the Sierra Nevada Protected Area (Southern Spain), generated using Sentinel-2 imagery (2019-2023) and ecohydrological attributes derived from NDVI time series. NDVI metrics were calculated from cloud-and snow-filtered Sentinel-2 Level 2A images processed in Google Earth Engine. Monthly NDVI values were used to extract three ecohydrological indicators: dry-season NDVI, dry-wet seasonal NDVI difference, and interannual NDVI variability. Based on quartile classifications of these indicators, 64 ecohydrological vegetation classes were defined. These were further clustered into three levels of potential groundwater dependence using hierarchical clustering techniques, differentiating between alpine and lower-elevation aquifer zones. The dataset includes raster layers (GeoTIFF) of the ecohydrological classes and pGDV types at 10 m spatial resolution, a CSV file with descriptive statistics for each class, and complete metadata. All spatial layers are projected in ETRS89 / UTM Zone 30N (EPSG: 25830) and are ready for visualization and analysis in standard GIS platforms. Partial validation of the classification was performed using spring location data and the distribution of hygrophilous plant species from official conservation databases. This available dataset enables reproducible analysis of vegetation-groundwater relationships in dryland mountain ecosystems. It supports comparative research across regions, facilitates the study of groundwater buffering effects on vegetation function, and offers a transferable framework for ecohydrological classification based on remote sensing. The data can be reused to inform biodiversity conservation, groundwater management, and climate change adaptation strategies in the Mediterranean and other water-limited mountain regions. (c) 2025 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)