Assessment Of Diurnal Evolution Of Urban Tree Temperature At Mega-city Scale

Abstract:

Urban trees play a critical role in biosphere-atmospheric interaction and therefore shape urban climates. Trees tolerate urban climates and simultaneously are influenced negatively by human activities in urban settings. Tree canopy temperature was regarded as an indicator of their interactive meteorological effects on urban environments. Understanding knowledge of spatiotemporal variability of canopy temperature in urban settings is thus essential to optimize the environmental benefits from trees in cities. Yet, there is currently a lack of understanding how canopy temperature varies diurnally and it has been rarely explored which factors are attributing to this spatiotemporal variability?

This study takes New York City as an example. We, for the first time, characterize the spatiotemporal patterns of canopy temperature using downscaled ECOSTRESS Land Surface Temperature (LST). Results reveals discernible spatial variability of Canopy LST diurnally, with a remarkable magnitude of 10 K during daytime and 2 K during nighttime. We implemented GAM (Generalized Additive Model) to statistically evaluate effects of explanatory variables on Canopy LST variation. Canopy LST decreases nonlinearly 2 K – 5 K with increasing greenery coverage from 40% - 100% during daytime. This effect is found insignificant during nighttime with a decrease of 0.1 K to 1 K. The similar pattern has been revealed regarding impervious fraction. Increasing impervious coverage from 60% to 100% corresponds to a disproportional increase of Canopy LST 2.3 K – 5 K during daytime and 0.1 K – 1 K at night. Distance to water influences on Canopy LST incredibly at distance adjacent to water bodies less than 1.5 km, which cools canopy temperature by 2 K - 5 K during daytime and 0.1 - 0.5 K at night. Furthermore, the analysis also reveals the strong impact of building height on nocturnal Canopy LST.

Our findings suggest that land cover properties play an important role in shaping Canopy LST during daytime while urban morphological properties explain this variation at night. Our approach is deliverable which can be replicated in other regions.