Imroz et al. (2025) Integrated assessment of urban flooding and heat island interactions: A systematic review of geospatial technologies, machine learning approaches, and microclimate dynamics

Identification

• Journal: Journal of Environmental Management
• Year: 2025
• Date: 2025-11-19
• Authors: Mohammad Imroz, M. P. Akhtar, Meena Sharma, Fahad Alshehri
• DOI: 10.1016/j.jenvman.2025.127984

Research Groups

• Department of Civil Engineering, School of Engineering (SCE), Manipal University Jaipur, Jaipur, India
• Department of Civil Engineering, College of Engineering, University of Bisha, Bisha, Saudi Arabia

Short Summary

This systematic review synthesizes geospatial technologies and machine learning approaches used to assess the integrated dynamics of urban flooding and Urban Heat Island (UHI) interactions, identifying critical methodological gaps and proposing a unified data-driven framework for enhanced urban climate resilience.

Objective

• To systematically review the evolution and significance of Land Use and Land Cover (LULC), microclimate variables, and empirical/GIS-based techniques, including machine learning, in assessing the integrated dynamics of urban flooding and Urban Heat Island (UHI) interactions.
• To identify methodological and geographical gaps in current approaches and propose a unified data-driven framework for simulating compound hazards and enhancing urban climate resilience.

Study Configuration

• Spatial Scale: Urban to regional, encompassing microclimate variations.
• Temporal Scale: Review of peer-reviewed studies published between 2010 and 2025.

Methodology and Data

• Models used:
  • Hybrid learning models (e.g., CNN-LSTM) for Urban Heat Island analysis.
  • Ensemble techniques (e.g., Random Forest, Gradient Boosting) for flood danger analysis.
  • Deep learning models for complex, nonlinear relationships.
  • Transformers and Graph Neural Networks (GNNs) for complex, large-scale scenarios.
• Data sources:
  • Scientific literature from Scopus, Web of Science, ScienceDirect, and Google Scholar.
  • Studies categorized by hazard type, analytical technique, and validation metrics (e.g., RMSE, R², MAE).

Main Results

• Approximately 30–32% of reviewed studies employed hybrid learning models (e.g., CNN-LSTM) for Urban Heat Island (UHI) analysis.
• 60–65% of studies utilized ensemble techniques (e.g., Random Forest, Gradient Boosting) for flood danger analysis.
• Deep learning is effective for modeling complex, nonlinear relationships, while ensemble models are valued for interpretability and multi-source information integration.
• CNN-LSTM models performed best for moderate-resolution spatiotemporal data, whereas transformers and Graph Neural Networks (GNNs) excelled for complex, large-scale scenarios.
• Key gaps identified include the lack of scalable models leveraging real-time Internet of Things (IoT) data, insufficient accounting for urban morphological and climatic variability, and challenges in model interpretability.
• A unified data-driven framework is proposed to simulate compound hazards and enhance urban climate resilience.

Contributions

• Provides a comprehensive systematic review of geospatial technologies, machine learning approaches, and microclimate dynamics in the integrated assessment of urban flooding and Urban Heat Island interactions.
• Identifies critical methodological and geographical gaps in current analytical frameworks, particularly regarding the reciprocal spatio-temporal interplay among these phenomena.
• Synthesizes cutting-edge findings on the physical and data-driven interlinkages between urban flooding and UHIs, offering insights for more resilient and sustainable city planning.
• Proposes a unified data-driven framework to simulate compound hazards and enhance urban climate resilience, addressing current limitations in separate hazard assessments.

Funding

Not specified in the provided text.

Citation

@article{Imroz2025Integrated,
  author = {Imroz, Mohammad and Akhtar, M. P. and Sharma, Meena and Alshehri, Fahad},
  title = {Integrated assessment of urban flooding and heat island interactions: A systematic review of geospatial technologies, machine learning approaches, and microclimate dynamics},
  journal = {Journal of Environmental Management},
  year = {2025},
  doi = {10.1016/j.jenvman.2025.127984},
  url = {https://doi.org/10.1016/j.jenvman.2025.127984}
}