Urban morphology impacts on urban microclimate using artificial intelligence – a review

Funding Sponsor

Canada First Research Excellence Fund

Author's Department

Architecture Department

Third Author's Department

Architecture Department

Find in your Library

https://doi.org/10.1016/j.cacint.2025.100221

All Authors

Ahmed Marey Jiwei Zou Sherif Goubran Liangzhu Leon Wang Abhishek Gaur

Document Type

Research Article

Publication Title

City and Environment Interactions

Publication Date

12-1-2025

doi

10.1016/j.cacint.2025.100221

Abstract

Urban morphology, defined by the characteristics and spatial arrangement of urban structures, significantly affects urban microclimate in terms of thermal environments, wind dynamics, energy use, and outdoor air quality. Despite extensive research in this field, these effects are intensified by climate change and rapid urbanization, posing challenges to urban sustainability, such as poor air quality, increased energy demands, and pedestrian discomfort. While artificial intelligence (AI) and machine learning (ML) offer promising solutions for addressing these challenges, the field lacks standardized approaches for implementing these technologies. By leveraging urban morphology indicators such as sky view factor, building density, and green space ratio, AI models can analyze complex interactions across various spatiotemporal scales. However, significant variability in methodologies, indicators, and datasets limits the generalizability and applicability of these techniques. By synthesizing 111 studies over the last decade utilizing urban morphology and AI models to predict urban microclimate, this review aims to bridge these gaps and highlight AI's unique potential to contribute to the field. Analyzed studies reported that key urban morphology indicators, particularly building density and height, explain up to 75% of land surface temperature variance across seasons, while sky view factor accounts for over 67% of heat exposure variations in urban environments, with these findings emerging from multiple independent investigations across diverse urban contexts. Random Forest emerges as the most widely adopted AI technique, demonstrating robust performance across different applications. Emerging trends, such as hybrid approaches combining AI with physics-based models, are highlighted as promising avenues for advancing the field. Our review identifies the need for standardized frameworks and datasets to enhance model applicability. The study presents actionable insights for climate-responsive urban planning and lays the groundwork for interdisciplinary studies, enabling the development of resilient, sustainable urban environments amid the growing challenges of urbanization and climate change.

Link to Full Text
Find this item online

Share

COinS