NDWI–NDBI Overlay–Based Analysis of Coastal Fence Expansion and Hydrological Changes in Victoria Island, Lagos, Nigeria (2010–2015), with Machine Le

Coastal ecosystems are becoming increasingly susceptible to anthropogenic impact and the process of infrastructure construction, particularly within fast-urbanizing zones. This study examines the spatiotemporal processes of coastal fence growth and surface water loss during 2010–2015 in Victoria Island, Lagos. Landsat imagery was employed in extracting the Normalized Difference Water Index (NDWI) and the Normalized Difference Built-up Index (NDBI), the former being the prime indicator of hydrological alteration, while the latter being the key indicator of urban growth. NDWI loss areas coincide with NDBI gain areas to identify areas where natural water bodies are substituted by artificial structures in the guise of coastal fences. Supervised machine learning–based classification was employed for improved change detection and spatial precision. This study introduces a new application of NDWI–NDBI overlay and supervised machine learning to coastal fencing identification in Lagos, Nigeria, and the approach provides a reproducible model for monitoring structural change in coastal communities that have limited data. The result clearly demonstrates that the reduction in the extent of surface water has a strong correlation with the expansion of built-up land, particularly along the coasts, where coastal fences have been extended seaward. This study demonstrates the application of fused spectral indices and machine learning to monitor coastal change and offers a reproducible technique for sustainable coastal zone management.