The impact of spatial patterns of urban green infrastructure on urban microclimate

Background/Question/Methods

: The higher temperatures present in cities are associated with adverse impacts for human health and well-being, including increasing rates of illness and mortality. ‘Greening’ urban areas is an adaptation strategy that may mitigate increases in urban temperatures, by increasing the cover and abundance of vegetation. However, space is limited in high-density cities to build new large green spaces. Therefore, there is a need to know how to manage existing green spaces to maximize ecosystem benefits. To do this, we need to better understand the capacity of different types and arrangements of urban green infrastructure (UGI) to mitigate urban heat, at different spatial and temporal scales. We ask: How does composition and configuration of UGI affect mean land surface temperature (LST) in Montreal at multiple spatial scales? We use publicly accessible data to test the relationship between LST and various composition and configuration parameters through a multi-scale approach. We hypothesize both composition and configuration of UGI will affect the magnitude of urban heat mitigation, but that: a) high vegetation will have a greater influence on LST than low vegetation, b) composition will have a greater influence than configuration, c) the impact of UGI configuration will vary with changes in composition.

Results/Conclusions

: Our results show that the percent cover of both high vegetation (e.g. trees > 3 m tall) and low vegetation (< 3 m tall, primarily herbaceous vegetation) are negatively correlated with LST across borough scale and census tract scale, but that the impact of high vegetation is greater than that of low vegetation. For configuration metrics: Mean Patch Size and Largest Patch Index are significantly negatively correlated with LST; Patch density is significantly positively related with LST; the impacts of Edge density and Mean Patch Shape Index are relatively less obvious, especially at borough scale. Further, the impact of several UGI configuration metrics varies with changes in composition. For example, from 10% high vegetation to 70% high vegetation, the slope of the relationship between LST and mean patch size parameter gets lower and lower, which means the action of increasing mean patch size of high vegetation patches might take more effect in place where the percent cover of high vegetation is lower.