Simulating Personal Heat Exposure in Cities with the ICARUS Model

Many of the mitigation and adaptation strategies proposed to improve urban thermal environments are designed to reduce the frequency, intensity, and/or duration of personal exposure to dangerously hot conditions. Despite this implicit focus, knowledge of spatial and temporal patterns in personal heat exposure in urban areas is quite limited. More information is needed to precisely pinpoint the places, times, situations, activities, and people associated with higher heat exposure such that intervention measures that aim to provide relief from the heat can be accurately targeted and efficiently delivered.

We are developing a model called ICARUS that is capable of simulating at a high spatial and temporal resolution the personal heat exposure of simulated agents in urban environments. This presentation will introduce major components of model development, including the representation of individual-level trips and activities as informed by a regional travel demand model and the quantification of urban microclimates across urban regions using deep learning. We will also present findings from technical advisory committee meetings in which professional stakeholders in southwestern cities interacted with prototypes of the ICARUS model and provided feedback on model design, visualization and application. Finally, we will share results that emerged from and contributed to model development. Among the most notable findings is that we estimate that approximately 30% of the more than one million walking trips that occur in Maricopa County on a summer day likely fall out of compliance with accepted standards for occupational heat safety, suggesting significant opportunities for infrastructure investment.