This article demonstrates a way to model the urban microclimate using a combination of statistical analysis and a geographic information system (GIS). Field measurements of ambient temperature and relative humidity were collected in various urban settings at ground level over two surface materials and at two different times of the day. A panel autoregressive (PAR) model was created to conduct a regression analysis capable of accounting for the spatial autocorrelation. With the PAR analyses, selected urban street characteristics showed varying effects over the different surface materials and at different times of the day. The overall influences were consistent with previous findings. PAR models were then applied in GIS to create thermal prediction maps for the study area to determine the hot and cool spots. Since GIS is the most commonly used technology among a wide range of researchers and professionals, this article provides a means to improve the synthesis, integration, and sharing of information to understand the relationship between the heat-vulnerable population and heat stress within the urban environment. This methodology will be useful in the effort to reduce the heat-related morbidity and mortality, which are expected to increase with projected global warming.
展开▼
