Advanced Journal of Environmental Science and Technology

ISSN 2756-3251

Advanced Journal of Environmental Science and Technology | Vol. 17 (6), pp. 475-486, June, 2026

Research Article

Machine Learning-Guided Optimization of Urban Green Roofs Mitigates Stormwater Runoff and Urban Heat Island Intensity in Tropical Megacities

Amina Yusuf¹*, David Thorne², Kenji Takahashi³

¹Department of Urban and Regional Planning, University of Lagos, Lagos, Nigeria
²Department of Geography and Environment, London School of Economics, London, United Kingdom
³Department of Urban Engineering, University of Tokyo, Tokyo, Japan

Corresponding Author: Amina Yusuf — [email protected]

Abstract

Rapid urbanization and extreme convective rainfall events overwhelm aging drainage infrastructure while intensifying thermal stress in tropical cities. This research combined remote sensing, microclimate sensors, and random forest modeling to optimize substrate depth and native succulent selection across 45 experimental green roofs. Implementing a 15 cm expanded-clay and compost substrate planted with Portulacaria afra reduced peak stormwater runoff volume by 68.4% during high-intensity storms (>50 mm/h). Surface rooftop temperatures dropped by an average of 7.3 °C, lowering indoor cooling energy demand in underlying buildings by 19.5%.

Keywords: urban heat island; green roof infrastructure; stormwater management; machine learning; tropical cities; climate adaptation

Manuscript Timeline: Received 20 May, 2026; Revised 17 June, 2026; Accepted 27 June, 2026; Published 30 June, 2026.

Citation: Yusuf, A., Thorne, D., and Takahashi, K. (2026). Machine Learning-Guided Optimization of Urban Green Roofs Mitigates Stormwater Runoff and Urban Heat Island Intensity in Tropical Megacities. Advanced Journal of Environmental Science and Technology, 17(6), pp. 475-486.