Wind tunnel measurements of pollutant plume dispersion over hypothetical urban areas

Abstract

Non-computational-fluid-dynamics (Non-CFD) solutions, such as Gaussian plume models, are commonly employed to predict ground-level pollutant concentrations because of their cost-effectiveness. Whilst, they should be applied with caution for pollutant plume dispersion over complicated urban morphology in view of their implicit limitation of empirically determined dispersion coefficients σz. Skin-friction coefficient cf, which is a measure of aerodynamic resistance induced by rough surfaces, is proposed to parameterize the dispersion coefficient over urban areas in isothermal conditions. Analytical derivation shows that σz is proportional to the newly proposed friction length scale Lf (= x1/2 δ1/2 cf 1/4 where x and δ are the distance after pollutant source and the turbulent boundary layer thickness, respectively). Its functional form is verified by wind tunnel experiments for flows and tracer plume dispersion over hypothetical urban areas in the form of idealized street canyons of different building-height-to-street-width (aspect) ratios (ARs = 1/2, 1/4, 1/8 and 1/12). A ground-level, pollutant line source in crossflows is modeled by atomizing water vapor using ultrasonic. Ranges of turbulent boundary layer thickness (240 mm ≤ δ ≤ 285 mm) and skin-friction coefficient (8 × 10−3 ≤ cf ≤ 13 × 10−3) are tested. The tracer concentrations over rough surfaces exhibit the Gaussian distribution. A close correlation between σz and Lf is revealed (coefficient of determination R2 = 0.93), demonstrating the influence of drag on the transport processes. The analytical solution and wind tunnel results collectively suggest an improved parameterization of pollutant plume dispersion coefficient over rough surfaces, refining the practice of the air quality forecast in urban areas. © 2018 Elsevier Ltd