Urban Heat Island (UHI) phenomena have emerged as a major environmental issue in fast-developing tropical megacities, where increasing land surface temperature poses threats to public health, energy systems, and ecosystem equilibrium. The current work proffers an interdisciplinary research approach involving remote sensing, Geographic Information Systems (GIS), and Artificial Intelligence (AI) climate modelling for the study, forecasting, and mitigation of UHI in Lagos, Nigeria, which is among Africa's fastest-developing urban agglomerations. With the help of multi-temporal satellite data of Landsat 8/9, MODIS, and Sentinel-2, along with geospatial indices (NDVI, NDBI, Albedo), land surface temperature (LST) variation and urbanization growth were measured between 2013 and 2024. There was tremendous growth in UHI intensity, with the LST gradient of the urban-rural zone increasing from 3.2°C in 2013 to 5.4°C in 2024, especially in highly developed areas like Apapa, Mushin, Ikeja, and Lagos Mainland. Advanced machine learning algorithms Artificial Neural Network (ANN), Random Forest (RF), and Gradient Boosted Trees (GBT) were created for predicting LST based on urban morphological and environmental variables. The GBT model performed the best with R² of 0.95 and RMSE of 1.28°C and well captured the non-linear dependencies between land use, vegetation, and surface temperature. Scenario-based simulation revealed that enhanced urban greens by 20%, cool roof technology, and water-sensitive urban design would decrease LST up to 2.4°C, thereby decreasing UHI hotspots by 38%. In addition, a decision-support system with WebGIS was developed to visualize spatial UHI patterns and simulate mitigation effect, providing urban planners with an interactive platform for climate-responsive decision-making. The research concludes that there exists an intersection of geospatial and AI technologies that offers a scalable, data-enabled method for managing the tropical cities' heat risks. It suggests institutionalizing UHI mitigation into urban planning, green infrastructure prioritization, and predictive analytics utilization for adaptive urban planning. The framework thus established can act as a model for climate-resilient urbanization for Sub-Saharan Africa and other exposed tropical regions.
Afolabi et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: