Mapping Real-time Heat Exposure with Social Media Data

Keywords: Urban Heat Island, Social Media, Real-time analysis, CyberGIS
Abstract Type: Paper Abstract

Abstract

As heat-related illness is contributing to a significant portion of mortality, combined with the huge challenge of global warming and urban heat island, it is critical to find the locations and vulnerable groups that are exposed to extreme heat. As individuals, we want to avoid extreme heat areas based on a dynamic heat exposure map. And for the government, quick decisions about the dynamic distribution of medical resources and evacuation can save people’s lives, especially during heatwaves or electricity shortages. To serve this purpose, we develop a framework to analyze and visualize the heat exposure dynamically based on real-time collected location-based social media data (LBSMD). Large volumes and low-cost LBSMD, together with the content analysis algorithm from Natural Language Processing are used effectively to generate real-time heat exposure maps in both local city scale in 1-km spatial resolution/census tract resolution and national scale with counties as spatial units. In this study, we conduct a case study to visualize and analyze the heat exposure map in Chicago and the United States with real-time collected geo-referenced data from Twitter. Enabled with high-performance computing resources, heat exposure from social media data can be used as a tool to assist the temperature and heat vulnerability index to effectively pinpoint vulnerable locations and vulnerable groups as well as usual changes in heat exposure and in real-time.