Introduction
The adverse effects of smoke on air quality and visibility are of great concern for fire and land
managers planning prescribed burns. The Federal Wildland Fire policy and the Clear Air Act
significantly broaden regulatory and management requirements by demanding from fire and land
managers an assessment of the air quality and visibility impacts from wildfires and fire
management programs. Since the fire emissions may lead to a violation of the National ambient
air quality standards, decisions about the prescribed burns must be made not only based on the
fire safety criteria (wind speed, fuel moisture etc.) but also taking into account impacts of the
smoke on air quality and visibility. In this paper, we describe a newly added functionality of the
coupled atmosphere-fire model WRF-SFIRE, allowing for the simulation and forecasting of the
2
smoke emissions, dispersion, and their effect on the air quality both on the local and regional
scale.
Smoke forecasting is a very complex problem that spans across many disciplines. In order to
estimate smoke spread and its effects on air quality, first one has to assess the amount of the fuel
burnt, and convert it into emission fluxes of the particular chemical species (or smoke in
general). In the next step, the vertical distribution of the smoke in the atmosphere (injection
height and vertical profile) has to be estimated, based on the fire intensity and meteorological
conditions (winds, atmospheric stability). Knowing the smoke injection characteristics and local
flow pattern, allows for an estimation of the smoke dispersion and deposition. In the last step, the
chemical processes associated with smoke dispersion in the atmosphere have to be taken into
account in order to assess the smoke impact on the air quality. As described above, the smoke
dispersion is clearly a multidisciplinary problem, with one common denominator though – the
weather, which effects fuel characteristics, fire behavior (intensity and amount of fuel burnt),
smoke injection, smoke dispersion and smoke chemistry. From this standpoint, building a smoke
forecasting system around the weather forecasting system seems to be a logical choice, and in
this paper we present a first attempt toward creating such a model.
There is a wide suite of tools of various complexity levels that may be helpful in assessing the
smoke dispersion. They range from simple Gaussian smoke models like VSMOKE (Lavdas
1996) and SASEM (Sestak and Riebau 1988) providing the area affected by smoke based on the
defined location, fuel type, fire area and wind conditions like, through puff models like
CALPUFF (Scire 2000) to complex multi-model systems like BlueSky (Larkin et al. 2009)
providing estimation of fire emissions, dispersion and air quality effects associated with fires.
For more information on smoke transport models please see the smoke modeling review by
Goodrick et al. (2012). Since the weather plays a key role in fire progression and dispersion of
the fire emissions, some sort of a weather forecast is generally required for any of these tools in
order to make any prediction how the fire smoke will disperse and affect the air quality. In the
simplest case, it may be in a from of a user input who must define the wind speed and direction
for which the smoke dispersion will be assessed. In complex systems like BlueFire, it may come
from a separate numerical weather forecasting model providing the weather input to the fire
emission and dispersion components of the system. Note that in the complex modeling
framework like BlueSky, the weather input generally affects a number of its components. For
instance, the meteorological conditions can be taken into account for the estimation of the burn
rates, but also may be used in the plume rise model computing the vertical smoke distribution as
well the dispersion and chemical models like CMAQ. WRF-SFIRE bases on a similar principle,
but all the components are integrated around the weather forecasting system. The model
comprehensively resolves the fire progression, the heat release associated with the fire, its plume
rise and smoke dispersion and chemistry without any external components, providing the
weather, fire progression, emission and air quality forecasts.