The Effects of Urban Geometry on Point-Source Scalar Plume Statistics: A Large Eddy Simulation Study

Particle transport by turbulent flows is a study of interest in a variety of fields, from agricultural modeling and planning, to snow and sediment deposition, and cloud microphysics. One application directly pertinent to human health is the transport and deposition of pollutants, such as PM2.5, in urban environments. Numerous observational studies have attempted to quantify the characteristics of pollutant dispersion in urban canopies, but in recent years, Large Eddy Simulations (LES) have been shown to properly represent turbulent flows at sufficiently high Reynolds numbers. The computational efficiency of LES means it can resolve entire central business districts (CBDs) or neighborhoods, making it very attractive for urban modeling.

This study aims to utilize LES to quantify scalar plumes in an idealized urban canopy at atmospheric Reynolds numbers. A suite of LES was run over simplified urban geometry (cuboids), spanning a wide range of plan area and frontal area fractions. The immersed boundary direct forcing approach is used to account for the presence of buildings on momentum conservation, and a Cartesian cut cell finite volume method, which conserves mass accurately, is used for scalar transport. Simulations will be used to characterize how point-source scalar plumes are influenced by urban geometry; we will demonstrate how plume lateral and vertical moments (centerline, spread, skewness, kurtosis) depend on urban geometry, and how they deviate from the Gaussian plume model in the near-source region.