The study was conducted with an aim to provide practical solution for the groundwater management in three villages namely Singhola, Ghoga and Dhirpur of the North-West Delhi, India. LANDSAT remote-sensing datasets for the last four decades (1977–2018) were assessed to determine changes in vegetation cover at the selected sites. The Google Earth Engine was used to determine how values of the Normalized Difference Vegetation Index (NDVI) were found to have varied spatially and over time for the selected sites. Strong correlations were found between the NDVI values of surface features including waterbodies, forest land, agricultural land and urban areas in Singhola, Ghoga and Dhirpur, respectively. The relative infiltration capacity of soils was highest (92.9%) and lowest (57%) for Singhola and Dhirpur, respectively, due to spatial differences in soil texture. In each village, locations exhibiting a higher soil infiltration capacity could be used for implementing managed aquifer recharge schemes using rainwater harvested from rooftops in the villages. This assessment indicated that the village of Ghoga has the highest potential (3,76,98,013.08 m3) for aquifer recharge through rooftop rainwater harvesting as compared to the other two villages.