Green and cool roofs to mitigate urban heat island effects in the Chicago metropolitan area: evaluation with a regional climate model

The effects of urban heat islands (UHIs) have a substantial bearing on the sustainability of cities and environs. This paper examines the efficacy of green and cool roofs as potential UHI mitigation strategies to make cities more resilient against UHI. We have employed the urbanized version of the Weather Research and Forecasting (uWRF) model at high (1 km) resolution with physically-based rooftop parameterization schemes (conventional, green and cool), a first-time application to the Chicago metropolitan area. We simulated a hot summer period (16-18 August 2013) and assessed (i) UHI reductions for different urban landuse with green/cool roofs, (ii) the interaction of lake breeze and UHI, and (iii) diurnal boundary layer dynamics. The performance of uWRF was evaluated using sensible heat flux and air temperature measurements from an urban mini-field campaign. The simulated roof surface energy balance captured the energy distribution with respective rooftop algorithms. Results showed that daytime roof temperature reduced and varied linearly with increasing green roof fractions, from less than 1 °C for the case of 25% green roof to ∼3 °C during peak daytime for 100% green roof. Diurnal transitions from land to lake breeze and vice versa had a substantial impact on the daytime cycle of roof surface UHI, which had a 3-4 hour lag in comparison to 2 m UHI. Green and cool roofs reduced horizontal and vertical wind speeds and affected lower atmosphere dynamics, including reduced vertical mixing, lower boundary layer depth, and weaker convective rolls. The lowered wind speeds and vertical mixing during daytime led to stagnation of air near the surface, potentially causing air quality issues. The selection of green and cool roofs for UHI mitigation should therefore carefully consider the competing feedbacks. The new results for regional land-lake circulations and boundary layer dynamics from this study may be extended to other urbanized areas, particularly to coastal areas. © 2016 IOP Publishing Ltd.