Using the High resolution VOC Atmospheric Chemistry in Canopies (Hi-VACC) Model to Simulate Smoke Dispersion from Forest Fires

We have been working to develop a new post-processing model - High resolution VOC Atmospheric Chemistry in Canopies (Hi-VACC) - which will be able to resolve the dispersion and chemistry of reacting chemical species given their emission rates from the vegetation and soil, driven by high resolution meteorological forcing and wind fields from various high resolution atmospheric regional and large-eddy simulations. Hi-VACC reads in fields of pressure, temperature, humidity, air density, short-wave radiation, wind (3-D u, v and w components) and sub-grid-scale turbulence that were simulated by a high resolution atmospheric model. This meteorological forcing data is provided as snapshots of 3-D fields. Presently, the advection-diffusion portion of the model is fully developed, and we have tested it using a number of RAMS-based Forest Large Eddy Simulation (RAFLES) runs. Here, we use Hi-VACC to perform simulations of smoke dispersion from a theoretical forest fire in a domain in The Pine Barrens in New Jersey. The simulations are driven by RAFLES simulations over the same domain. We perform simulations that include the effects of burning on meteorological conditions – like increased heat flux – as well as simulations that neglected these but had the same smoke source to analyze the resulting differences in smoke dispersion. We prescribed a 90m long line source of smoke in the middle of the domain and fairly weak (~1m/s) mean wind. In both simulations we saw the dispersing smoke plume arise from the source and travel as expected with the wind to - and out of - the edge of the domain (300m away) over the course of the 20 minute simulations. Qualitative differences in plume rise and behavior can be seen in the two simulations, as would be expected, particularly given the higher resultant heat flux from including the effects of burning in one simulation. One of the primary benefits of Hi-VACC is that users of other models can utilize this tool with only minimal work on their part — processing their output fields into the appropriate HI-VACC input format. We have developed our model such that for whatever atmospheric model is being used with it, a MATLAB function must be written to extract the necessary information from the output files of that model and shape it into the proper format. This is the only model-specific work required. As such, this sort of smoke dispersion modeling performed by Hi-VACC – as well as its other capabilities – can be easily performed in other locations and conditions provided simulations from an atmospheric model are available.