An improved understanding of thermal environments for sustainable urban development

Keywords: Land surface temperature, air temperature, global, urban thermal environment, sustainability
Abstract Type: Paper Abstract

Abstract

Land surface temperature (LST) and near-surface air temperature (Ta) that influence human and nature systems from local to global scales are two of the most important variables for understanding thermal environments. Although satellite data can provide detailed spatiotemporal information of LST, the applications of these data are hampered because of missing values caused by factors such as cloud contamination. Moreover, Ta is generally only available at weather stations with limited spatial coverage. In this study, we first developed a spatiotemporal gap-filling framework and created 1-km seamless global datasets of daily maximum and minimum LST using this framework and MODIS LST product. We then introduced a class of Spatially Varying Coefficient Models with Sign Preservation (SVCM-SP) method and using this method, built 1-km seamless global datasets of daily maximum and minimum Ta with the seamless LST and elevation data. Our algorithms not only are computing efficient, but also perform well regarding accuracy as shown from the cross-validation. The resulting global products of the seamless daily (mid-daytime and mid-nighttime) LST and daily (maximum and minimum) Ta at 1 km spatial resolution are of great use in global studies such as climatology, terrestrial ecology, and epidemiology, especially in urban systems.