Summertime Urban Heat Island in the Oklahoma City and Implications for Air Quality Assessment

Through many observational and modeling studies, the understanding about the formation of urban heat island (UHI) has improved during the past few decades. UHI intensity is normally defined as the difference between urban and rural near-surface temperature (e.g., at 2 m above ground level). UHI intensity normally increases prominently around sunset and then stays at a roughly constant level throughout the night until early next morning when the atmospheric boundary layer develops rapidly. More detailed variations (e.g., hourly) of nighttime UHI intensity, however, remain to be explained. In addition, three-dimensional dynamic and thermal properties of UHI (which are difficult to observe) and their effects on the dispersion of atmospheric pollutants are not always clear. Large biases of simulated pollutants (e.g., of nitrogen oxides, ozone) in urban regions during nighttime had been reported, partially due to limited capability of numerical models to reproduce UHI features. Nighttime boundary layer structure is critical for UHI and air quality assessments. Investigation of model performance for the nocturnal boundary layer, along with the collection of suitable observations, is highly warranted.

In this study, summertime UHI in Oklahoma City (OKC) is investigated using the comprehensive dataset collected during the Joint Urban 2003 (JU2003) tracer experiment, including surface measurements and vertical profiles obtained by boundary-layer wind profilers and radiosondes. The characteristics of certain chemical species together with rural cooling rate during the early evening transition are found can be used as indications of UHI development. The performance of the Weather Research and Forecasting (WRF) model with different boundary-layer parameterization schemes in terms of simulating nighttime boundary layer structure and reproducing the UHI in OKC is evaluated for July 2003. All the schemes capture the diurnal cycle, day-to-day variations, and horizontal spatial extent of near-surface UHI reasonably well, but predict different vertical extent of nocturnal UHI by a factor of 2 or even larger. The implications for air quality assessments will also be discussed.