How urban heterogeneity and turbulence shape street-level heat exposure

Street-level heat exposure, humidity, and cooling energy demand are governed by turbulent exchange within the urban canopy, yet operational models often oversimplify turbulence and miss vertical variability. Using high-resolution large-eddy simulations (LES) and published syntheses, we represent the non-monotonic structure of canopy turbulence via a unified multi-layer closure valid across urban canopy densities and test its city-scale implications in Chicago. Embedded in a 1-D RANS framework, the proposed closure reduces mixing-length and eddy-viscosity errors by 70% and 50% relative to LES. Simulations better match observed temperature and humidity and reveal 1.5–2.5 °C higher thermal exposure and ±$3–6% RH differences in moderately dense local climate zones, consistent with increased cooling-energy demand. Results are robust across coarse and high-resolution morphology, showing that turbulence—not merely morphology detail—controls neighborhood-scale outcomes. By resolving canopy mixing that current schemes oversimplify, this work reduces decision-relevant uncertainty and enables reliable street-level assessments for urban design, operations, and resilience planning.