Cities are often 2-5°C warmer than their rural surroundings, intensifying health risks, straining infrastructure, and amplifying inequities. This additional warming is caused by the urban heat island effect, in which the built environment traps heat during the day, and retains that heat through the night. Despite this risk, we lack reliable, high-resolution observational data to adequately characterize the urban heat island effect and its impact on health and energy systems. When these data do exist, they are often noisy, biased, and unevenly distributed. In this talk, I will present statistical and machine learning approaches that transform these fragmented data into reliable, high-resolution maps of neighborhood-level heat exposure. Using techniques from causal inference, this work provides actionable insights for climate adaptation, so we can better evaluate cooling strategies such as reflective surfaces and urban greenery. I will conclude by outlining future research directions.

*Refreshments: 10:30 AM - 11:00 AM (Atrium)