An Application of Machine-Learning Model for Analyzing the Impact of Land-Use Change on Surface Water Resources in Gauteng Province, South Africa

The change in land-use diversity is attributed to the anthropogenic factors sustaining life. surface water bodies and other crucial natural resources study area are being depleted at an alarming rate. This explored implications of changing on by using a random forest (RF) classifier machine-learning algorithm remote-sensing models Gauteng Province, South Africa. Landsat datasets from 1993 2022 were used processed Google Earth Engine (GEE) platform, RF classifier. results indicate nine classes having increased decreased tendencies, with high F-score values ranging 72.3% 100%. In GP, spatial coverage BL has shrunk 100.4 km2 every year over past three decades. Similarly, BuA exhibits annual decreasing rate 42.4 due effect dense vegetation within same land use type. Meanwhile, bodies, marine quarries, arable lands, grasslands, shrublands, forests, wetlands expanded annually 1.3, 2.3, 2.9, 5.6, 11.2, 29.6, 89.5 km2, respectively. content level been poor throughout years. MNDWI NDWI have stronger Pearson correlation radius 5 km (r = 0.60, p 0.000, n 87,260) than 10 15 km. research essential improve current planning management techniques reduce environmental impacts change.