A methodological assessment of extreme heat mortality modeling and heat vulnerability mapping in Atlanta, Detroit, and Phoenix

Urban heat island;Urban climate;Heat vulnerability;Urban planning;Environmental planning;Public health;GIS

Mallen, Evan Sheppard ; Stone, Brian Drummond, William Grundstein, Andrew Russell, Ted O'Neill, Marie City and Regional Planning ; Stone, Brian

University:Georgia Institute of Technology

Department:City and Regional Planning

关键词: Urban heat island;    Urban climate;    Heat vulnerability;    Urban planning;    Environmental planning;    Public health;    GIS;   

Others  :  https://smartech.gatech.edu/bitstream/1853/62338/1/MALLEN-DISSERTATION-2019.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Extreme temperatures pose an increasingly high risk to human health and are projected to worsen in a warming climate with increased intensity, duration and frequency of heat waves, further amplified by the urban heat island, in the coming decades. To mitigate heat exposure and protect sensitive populations, urban planners are increasingly using decision support tools like heat vulnerability indices (HVIs) to identify high priority areas for intervention and investment. However, HVIs often capture only proxy heat exposure indicators at the land surface level, not air temperatures that humans experience, and are highly subjective in their construction methodology. This gap can be filled using regional climate models like the Weather Research & Forecasting (WRF) model to simulate air temperatures comprehensively over a city, coupled with a heat exposure-response function to objectively estimate mortality attributable to heat. But this method is often beyond the capabilities of local planning departments due to limitations in funding or technical expertise to run the model. Careful consideration of decision support tool selection will be an important factor in determining the future resilience of urban populations in a changing climate. Through a comparative analysis, this study investigates the relationship and utility of HVIs and spatial statistical attribution models with a focus on 1) the extent to which HVI methods can replicate spatial prioritization from a WRF-driven mortality model; 2) the relative significance of place-based vulnerabilities used in the HVI; and 3) the potential to reliably replicate a WRF-driven mortality model using publicly available datasets. This information can help urban planners and public health officials improve their emergency response plans and communication strategies for heat mitigation by specifically targeting short and long-term responses where there is greatest need. These techniques equip planners with a useful and accessible tool to protect vulnerable populations effectively and efficiently with minimal public funds and could advance the policies we use to adapt to a changing climate.