Multi-scale modeling of roadway air quality impacts: Development and evaluation of a Plume-in-Grid model

Eulerian three-dimensional (3D) grid-based models are widely used in air quality modeling. In such models, emissions are instantaneously diluted within the grid cells and, therefore, the near-source impacts of large point and line sources cannot be properly resolved. Plume-in-Grid models (PinG) use a subgrid-scale treatment to better represent local source contributions in an Eulerian grid-based simulation. PinG models already exist for point sources. However, modeling emissions from roadway traffic with point sources implies a very large computational burden. We present here a new PinG model that uses a Gaussian line source model, better suited than point sources to model roadway traffic emissions, embedded within an Eulerian model. The model is evaluated with a large dataset of nitrogen dioxide (NO2) concentrations over a 800 km road network. The PinG model leads to greater NO2 concentrations and shows better performance than the Eulerian model.