How to bring UHI to the urban planning table? A data-driven modeling approach. (August 2021)

• Record Type: Journal Article
• Authors: Pena Acosta, Monica; Vahdatikhaki, Faridaddin; Santos, João; Hammad, Amin; Dorée, Andries G.
• Is Part Of: Sustainable cities and society. Volume 71(2021)
• Publication Date: 2021-08
• Subjects: Urban heat island -- Urban decision making -- Data-driven modeling -- Decision trees; Sustainable urban development -- Periodicals; Sustainable buildings -- Periodicals; Urban ecology (Sociology) -- Periodicals; 307.76
• DOI: 10.1016/j.scs.2021.102948
• Languages: English
• ISSNs: 2210-6707

Abstract

Highlights: A data-driven modeling approach is used to develop a decision support tool for consideration of UHI in urban design. Urban morphology and socio-economic factors are considered. Two different machine learning methods are used and compared. An accuracy of 93 % is achieved in a case study about the city of Montreal. An easy-to-use assessment matrix is developed that categorizes streets into five UHI classes. Abstract: While temperature rises in urbanized area there is a growing concern among key decision-makers and urban planners to actively incorporate Urban Heat Island (UHI)-related considerations in their development/design. However, given that the existing models (mainly physics-based) are too complex to use, there is a need for an easy-to-use decision support tool that provides an explicit understanding of the contributions of different urban planning decision-making parameters on UHI. To this end, this research uses publicly available data to develop a data-driven methodology that mines explicit rules about the correlation between socio-economic and urban morphology features and UHI at a street-level. By implementing a tree-regression approach, five distinct categories of potential UHI were identified. These categories represent five levels of UHI, from low to high, where explicit thresholds are identified for each feature. The optimal model based on accuracy and interpretability is a decision tree (DT), with an accuracy of 93 %. With the results of the case study, it is demonstrated that (1) the proposed methodology leads to an easy-to-use tool that can be implemented by urban planners to investigate the impact of their design choices at the street-level, and (2) the results obtained are consistent with the current body of knowledge, which in turn alleviates the drawbacks of traditional methods.