ESTIMATING COMMUNITY HEALTH RISKS DURING EXTREME HOT WEATHER EVENTS

Climate change is influencing the severity and frequency of heat waves globally, leading to an expected increase in heat-related morbidity and mortality. In order to address the public health impacts, previous studies have estimated spatial and temporal variability in heat-related mortality and have developed indices to locate vulnerable populations. However, only a few studies have combined vulnerability indices with the spatial distribution of urban heat islands (UHIs) to evaluate the spatial distribution of public health impacts during extreme weather events. These studies have relied on vector-based analysis, introducing data quality concerns as a result of coarse spatial resolution and the modifiable areal unit problem. The objective of this paper is to develop a raster-based spatial model for heat risk mapping, and to predict heat-risk hotspots in Metro Vancouver at high spatial resolution (60m). Population density layers from 8 vulnerable population groups (seniors, infants, people living in old buildings, people living in high-rise building, the unemployed, people with low income, low education and people living alone) were combined in a vulnerability layer. The vulnerability layer was applied to a spatial-weight model with an urban heat island intensity layer, develop from Landsat 5 TM images and field observations, to estimate the heat risk. A spatial hotspot analysis using the Getis-Ord G_i index was used to estimate neighborhood effects at a range of spatial scales. The results indicate that this approach can effectively estimate heat risk hotspots in a large and heterogeneous urban environment.