Development of a fine scale smoke dispersion modeling system: Part II—Case study of a prescribed burn in the New Jersey Pine Barrens

Smoke dispersion from wildland fires is a critical health and safety issue, impacting air quality and visibility across a broad range of space and time scales. Predicting the dispersion of smoke from low-intensity fires is particularly challenging due to the fact that it is highly sensitive to critical factors such as near-surface meteorological conditions, local topography, vegetation, and atmospheric turbulence within and above vegetation layers. Existing integrated smoke dispersion modeling systems (e.g., BlueSky), which are designed for prediction of smoke from multiple sources on a regional scale, do not have the necessary resolution to accurately capture smoke from low-intensity fires that tends to meander around the source and may stay underneath forest canopies for a relatively long period of time. Recently, a project has been launched by the Joint Fire Science Program (JFSP) to develop and validate modeling tools for predicting smoke dispersion from low-intensity fires.

As part of achieving this goal, the Advanced Regional Prediction System (ARPS) atmospheric model has been modified to allow simulation of flow through a multi-layer canopy. The effects of vegetation elements (e.g., branches, leaves) on drag, turbulence production/dissipation, and the surface energy budget are accounted for through modifications to the ARPS model equations. Three-dimensional vegetation density data obtained from LIDAR measurements are used to initialize the canopy model. To account for the first order effects of a wildland fire, upward sensible heat fluxes are imposed within a fixed area of the model domain, with fire intensity derived from observed data. As a final step in the development process, ARPS has been coupled to the Pacific Northwest National Laboratory (PNNL) Integrated Lagrangian Transport (PILT) model.

This paper presents results from a recent modeling case study of a March 2011 prescribed burn in the New Jersey Pine Barrens. To accurately represent regional and local forcing within the region of the burn, a series of one-way nested simulations are executed, spanning from 9-km to 100-m horizontal grid spacing. Momentum, scalar, and turbulence fields are compared between the innermost domain simulation and data obtained from a series of flux towers located inside and outside of the burn unit. This work is one part of a coordinated effort to evaluate the performance of atmospheric dispersion modeling systems; papers detailing data analysis efforts as well as smoke dispersion modeling will be presented elsewhere at the conference.