Deep learning-based spatial optimization of green and cool roof implementation for urban heat mitigation

Area of Science:

• Urban planning
• Climate change adaptation
• Building science

Background:

• Urban heat extremes are intensifying due to climate change.
• Effective mitigation strategies are crucial for urban resilience.

Purpose of the Study:

• To develop a methodological framework for optimizing the implementation of urban green and cool roofs.
• To maximize cost-effectiveness in reducing urban heat stress.

Main Methods:

• A surrogate model based on the Multi-ResNet deep learning algorithm was developed.
• The model was trained on data from the Weather Research and Forecasting model coupled with an urban canopy model (WRF-UCM).
• The framework was applied to the Greater Seoul region under the SSP585 climate scenario for 2090-2099.

Main Results:

• The Pareto optimal scenario involves implementing cool roofs over 89.2% of urban areas at current green roof costs.
• This scenario reduces the effective heat stress index by 8.8% and decreases costs by 19.6%.
• An optimal cost range of $117.4–146.1/m over 40 years was identified for green roof cost-effectiveness.

Conclusions:

• Deep learning techniques can provide efficient quantitative assessments for urban planning.
• The proposed framework significantly reduces computational demands compared to traditional models.
• This approach supports climate-resilient urban building planning by optimizing green and cool roof strategies.