RANS solver for microscale pollution dispersion problems in areas with vegetation: Development and validation

TL;DR

This paper develops and validates a finite volume RANS solver for microscale urban pollution dispersion in vegetated areas, incorporating vegetation effects and dry deposition models to improve urban air quality simulations.

Contribution

It introduces a novel RANS solver that models vegetation as a porous zone within a five-equation system, validated through multiple test cases for urban pollution dispersion.

Findings

Good agreement with experimental and numerical data

Identified limitations due to turbulence model and vegetation properties

Validated across diverse flow scenarios

Abstract

We present a description and validation of a finite volume solver aimed at solving the problems of microscale urban flows where vegetation is present. The solver is based on the five equation system of Reynolds-averaged Navier-Stokes equations for atmospheric boundary layer flows, which are complemented by the k-epsilon turbulence model. The vegetation is modelled as a porous zone, and the effects of the vegetation are included in the momentum and turbulence equations. A detailed dry deposition model is incorporated in the pollutant transport equation, allowing the investigation of the filtering properties of urban vegetation. The solver is validated on four test cases to assess the components of the model: the flow and pollutant dispersion around the 2D hill, the temporal evolution of the rising thermal bubble, the flow through and around the forest canopy, and a hedgerow filtering the particle-laden flow. Generally good agreement with the measured values or previously computed numerical solution is observed, although some deficiencies of the model are identified. These are related to the chosen turbulence model and to the uncertainties of the vegetation properties.