Performance assessment of Large Eddy Simulation (LES) for modeling dispersion in an urban street canyon with tree planting

• Published in: Atmospheric environment (1994) Vol. 75; pp. 66 - 76
• Main Authors: Moonen, P., Gromke, C., Dorer, V.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01-08-2013; Elsevier
• Subjects: Analysis methods; Applied sciences; Atmospheric pollution; Computational fluid dynamics (CFD); Dispersion; Exact sciences and technology; Large eddy simulation (LES); Model performance; Pollution; Statistical analysis; Validation metrics; Wind tunnel measurements; Wind tunnel measurements; Model performance; Statistical analysis; Validation metrics; Computational fluid dynamics (CFD); Large eddy simulation (LES); Dispersion; Performance evaluation; Computational fluid dynamics; Urban area; Vegetation effect; Street canyon; Large eddy simulation; Wind tunnel; Tree crown; Atmospheric pollution forecasting; Air pollution; Numerical simulation; Atmospheric dispersion
• ISSN: 1352-2310; 1873-2844
• DOI: 10.1016/j.atmosenv.2013.04.016

The potential of a Large Eddy Simulation (LES) model to reliably predict near-field pollutant dispersion is assessed. To that extent, detailed time-resolved numerical simulations of coupled flow and dispersion are conducted for a street canyon with tree planting. Different crown porosities are considered. The model performance is assessed in several steps, ranging from a qualitative comparison to measured concentrations, over statistical data analysis by means of scatter plots and box plots, up to the calculation of objective validation metrics. The extensive validation effort highlights and quantifies notable features and shortcomings of the model, which would otherwise remain unnoticed. The model performance is found to be spatially non-uniform. Closer agreement with measurement data is achieved near the canyon ends than for the central part of the canyon, and typical model acceptance criteria are satisfied more easily for the leeward than for the windward canyon wall. This demonstrates the need for rigorous model evaluation. Only quality-assured models can be used with confidence to support assessment, planning and implementation of pollutant mitigation strategies.
•Detailed time-resolved numerical simulations (LES) of coupled flow and dispersion.•Novel inlet profile parameterization and line source modeling technique.•Rigorous model performance assessment following COST Action 732 recommendations.•We evidence that model performance is not spatially uniform.•Performance metrics are shown to be sensitive to the magnitude of the quantity being compared.