Urban areas face significant challenges from extreme heat and urban heat islands (UHIs), which often interact and intensify each other at multi-spatial scales. However, most existing studies examine extreme heat and its interaction with UHIs at the city scale, overlooking the spatial heterogeneity of temperature responses within local areas. Extreme heat does not manifest uniformly across the entire city, and the UHI is a typically localized phenomenon influenced by changes in local climate and urban factors. To address this gap, this study defines local extreme heat (LEH) at the local scale based on 1 km and examines surface urban heat island (SUHI) response to local extreme heat (LEH) in Singapore, a tropical city experiencing more frequent extreme heat events. Using multi-year temperature datasets, we calculated the difference in SUHI intensity (SUHII) between LEH and non-LEH conditions, referred to as SUHII. Our findings revealed that SUHII responses to LEH differed by daytime and nighttime and local areas. Daytime SUHII peaked at 3.2 °C in the northeast, while nighttime SUHII reached 0.6 °C in other regions. To identify the dominant drivers of SUHII responses to LEH, we employed the spatial Random Forest (spatialRF) model. Our results showed that the spatialRF model achieved R-squared values exceeding 63% for predicting daytime SUHII and 45% for nighttime SUHII. LEH, land use, and vegetation dominantly contributed to daytime SUHII, while socioeconomic factors mostly influenced nighttime SUHII. Furthermore, we applied SHAP to interpret the spatialRF model. Hotspots of both daytime and nighttime SUHII were driven by socio-economic factors. Finally, nonlinear associations showed that the cooling effect of vegetation reached saturation, as the SHAP values remained positive, while water bodies, as indicated by a U-shaped SHAP pattern followed by a decline, were more effective in mitigating SUHII increases under LEH conditions.

Greenery plays a vital role in urban environments, providing numerous benefits through diverse pathways. Various metrics and methodologies have been proposed to assess multiple dimensions of greenery exposure. For a comprehensive and precise assessment of greenery exposure for different research purposes, it is crucial to identify the most suitable methods and data sources. However, existing reviews primarily address the health outcomes of urban greenery, rather than the methods of assessing greenery exposure. To address this gap, we conducted a review of 312 research articles, focusing on methodologies and technologies for measuring greenery exposure in urban settings. This review categorizes exposure measurement techniques into three categories: proximity-based, mobility-based, and visibility-based, evaluating their strengths, limitations, and synergies. Proximity-based methods generally assess overall greenery level in residential areas or other locations, but they fall short in capturing the actual interactions between humans and greenery. Mobility-based methods track real-time human location and assess greenery exposure based on travel trajectories, but they neglect the specific nature of human-greenery interactions. In contrast, emerging visibility-based methods offer opportunities to measure potential visual interactions between individuals and greenery. We found emerging metrics tend to integrate 3D data, qualitative aspects, and diverse data sources. We advocate for an integrated approach that encompasses both human mobility and potential interactions with greenery across various areas. We also argue that data granularity is balanced against cost, scalability, and ethical constraints. Our comprehensive review offers a framework and categorization to guide studies in designing exposure measurements aligned with their research objectives.

Understanding people’s preferences is crucial for urban planning, yet current approaches often combine responses from multi-cultural populations, obscuring demographic differences and risking amplifying biases. We conducted a large-scale urban visual perception survey of streetscapes worldwide using street view imagery, examining how demographics—including gender, age, income, education, race and ethnicity, and personality traits—shape perceptions among 1,000 participants with balanced demographics from five countries and 45 nationalities. This dataset, Street Perception Evaluation Considering Socioeconomics, reveals demographic- and personality-based differences across six traditional indicators—safe, lively, wealthy, beautiful, boring, depressing—and four new ones: live nearby, walk, cycle, green. Location-based sentiments further shape these preferences. Machine-learning models trained on existing global datasets tend to overestimate positive indicators and underestimate negative ones compared to human responses, underscoring the need for local context. Our study aspires to rectify the myopic treatment of street perception, which rarely considers demographics or personality traits.