Progress 01/01/05 to 12/31/08
Outputs
OUTPUTS: Activities - Coordinated research activities with the Savannah River Smoke Project at the Department of Energy's Savannah River site to add plume measurements from radar and lidar, through an agreement with the University of Massachusetts Amherst. These efforts netted some of the first detailed views of vertical plume structure along with collocated measurements of surface smoke concentrations. The research team consulted with various field units within the US Forest Service on prescribed fires that produced adverse smoke impacts in urban areas (Asheville NC, Atlanta GA and Fayetteville AR). These consultations led to the development of a set of case studies that led to the recognition of the importance multiple core updrafts in determining plume structure and on subsequent plume transport and dispersion. Presentation of this concept at a series of EPA meetings and workshops revealed that multiple core updrafts was a new way of viewing fires as an emission source that differed markedly from approaches in the past. Data sets produced from the three case studies have helped directly support two Joint Fire Science Program projects dealing with plume rise and smoke dispersion. These JFSP projects are also contributing to a larger JFSP funded effort for the intercomparison of smoke and emissions models. Efforts are underway to incorporate information on multiple core plumes in prescribed fire training material, particularly the Guide to Prescribed Fire in Southern Forests. New modeling tools such as Daysmoke are embracing the multiple core plume concept as a means to improve their simulation of wildland fire smoke dispersion. Preliminary efforts to incorporate this concept into the BlueSky/RAINS smoke modeling framework have led to significant improvements in surface smoke concentration forecasts. PARTICIPANTS: Scott Goodrick, US Forest Service (PI). Project administration and assisted with numerical modeling. Yongqiang Liu, US Forest Service (co-PI). Responsible for FAST analysis and CMAQ simulations. Gary Achtemeier, US Forest Service (co-PI). Daysmoke developer. Ken Forbus and Tim Giddens, US Forest Service (technicians), computer system support. Philip Cunningham, Florida State University (collaborator) fluid dynamics simulations. Stephen Frasier, University of Massachusetts Amherst (collaborator) radar/lidar observations. TARGET AUDIENCES: The primary target for this project was the air quality community to improve their understanding of wildland fire as an emissions source and how assumptions made when modeling this phenomena impact the results. The prescribed fire community developed as a secondary target as new discoveries were made that could impact their smoke management practices. The air quality community was reached through attendance at a number of EPA sponsored conferences and workshops. The prescribed fire community was reached through presentations at state prescribed fire council meetings and participation in different training course on smoke management. PROJECT MODIFICATIONS: An initial delay in receiving funds led to a later need for a no-cost extension
Impacts
The primary goal of this project was to provide the air quality community with the appropriate tools and knowledge to best simulate emissions from wildland fires. The project had three objectives supporting this goal: determine the sensitivity of CMAQ to variations in the vertical distribution and timing of emissions release, determine the best plume rise model for simulating emissions from wildland fires and integrate into CMAQ, and to produce a data set for testing/validation of current/future plume rise models. CMAQ was found to be quite sensitive to the plume rise method with surface concentrations varying by more than an order of magnitude. The Fourier Amplitude Sensitivity Test (FAST) revealed that entrainment of ambient air into the plume was a critical factor controlling plume rise. However, investigation into specific case studies of prescribed fires that produced adverse smoke impacts in urban areas (Asheville NC, Atlanta GA and Fayetteville AR) revealed that many smoke plumes consist of multiple updraft cores and that this number can change during the burn. When the number of updraft cores was added to the FAST analysis it became the leading factor determining variability of plume rise. This result showed that plume descriptions developed for industrial emissions sources are a poor fit for the dynamically complex and evolving wildland fire emissions source. As Daysmoke is currently the only dispersion model capable of describing these multiple core updraft plumes, it has been incorporated into the CMAQ modeling framework and has shown good utility on all case studies presented. Being a hybrid physical/empirical model, Daysmoke plume rise projects were compared against a number of plume rise schemes such as the commonly used Briggs scheme as well as compared to computationally intensive direct numerical simulations (DNS) conducted with a computational fluid dynamics model. The data sets compiled during the course of this project provide a basis for a pair of studies funded by the Joint Fire Science Program examining smoke dispersion. The multiple core updraft plume discovery is leading to a fundamental shift in how smoke plumes are modeled as the concepts developed for industrial sources are found to be too limiting.
Publications
- Achtemeier, G.L., Goodrick, S. and Liu, Y., 2005. A coupled modeling system for connecting prescribed fire activity data through CMAQ for simulating regional scale air quality. EastFire 2005 Proceedings, Fairfax, VA.
- Achtemeier, G., Liu Y.-Q, Goodrick S., 2008, Prescribed fire and air quality in the American South: A review of conflicting interests and a technique for incorporating the land manager into regional air quality modeling, in Remote Sensing and Modeling Applications to Wildland Fires in Geosciences of Springer-Verlag and Tsinghua University Press
- Cunningham, P. and Goodrick, S. 2005. High-resolution numerical simulations of fire plume dynamics. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. Canmore, Alberta Canada
- Liu, Y.-Q., Achtemeier, G. and Goodrick, S., 2005, CMAQ-Daysmoke as a Smoke and Air Quality Management Technique: A Case Study of a Prescribed Burning in Georgia, (abstract) the Sixth Fire and Forest Meteorology Symposium, 25-27 October 2005, Canmore, AB, Canada.
- Liu, Y.-Q., Achtemeier, G., and Goodrick, S., 2005, What parameters are the most important to smoke plume rise (abstract) NOAA/EPA Golden Jubilee Symposium on Air Quality Modeling and Its Applications, 20-21 September, 2005, Durham, NC.
- Liu, Y., Achtemeier, G., Goodrick, S. 2006. Regional air quality effects of Brush Creek burning simulated with CMAQ-Daysmoke. (abstract) Presentation at Third International Fire Ecology and Management Congress, San Diego CA, Nov. 13-17, 2006.
- Liu, Y., Achtemeier, G., Goodrick, S. 2006. Modeling air quality effects of prescribed burn in Georgia with CMAQ-Daysmoke, Proceedings of Workshop on Agricultural Air Quality: State of Science, ed. V.P. Aneja, et al., pp 129-131.
- Liu, Y., Achtemeier, G., Goodrick, S. 2007. Simulation and real-time prediction of air quality effects of prescribed burns in the South with SHRMC-4S. (abstract) USDA Forest Service Southern Research Station All Scientist Meeting, 30 Jan. to 1 Feb. 2007, Emerald Point Hotel at Lake Lanier, GA. Meeting book; p. 118.
- Liu, Y., Achtemeier, G. and Goodrick, S.. 2007. A Sensitivity Study of Air Quality Simulation to Smoke Plume-Core Number. (abstract) EastFire Conference, George Mason University, Fairfax, VA, June 5-8, 2007.
- Liu, Y-Q., Achtemeier, G. and Goodrick, S., 2008, Sensitivity of air quality simulation to smoke plume rise. Journal of Applied Remote Sensing Vol. 2, 021503 (20 May 2008) DOI: 10.1117/1.2938723.
- Liu, Y., Achtemeier, G., Goodrick, S., Jackson, W. 2007. Sensitivity of CMAQ-Daysmoke simulations to the smoke properties and ambient conditions. (abstract) 2nd Fire Behavior and Fuels Conference, Destin, FL. March 26-30, 2007.
- Achtemeier, G.L. 2005. On plume rise: matching Daysmoke with Briggs Equations for industrial stacks. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. 25-27 October 2005, Canmore, AB, Canada.
- Liu, Y.-Q., Goodrick, S., Achtemeier, G., Jackson, W., Qu, J., and Wang, W., 2008, Smoke incursions into urban areas: Simulation of a Georgia prescribed burn, International Journal of Wildland Fire.
- Liu, Y., Goodrick, S., Achtemeier, G. and Naeher, L.. 2007. The air quality effects of the 2007 South Georgia wildfires. (abstract) 7th Symposium on Fire and Forest Meteorology held at Bar Harbor, Maine, 23-25 October 2007. Amerircan Meteorological Society.
|
Progress 01/01/06 to 01/01/07
Outputs
On March 24, 2006 the Cherokee National Forest conducted an 1,800 acre prescribed burn that resulted in extremely high PM 2.5 measurements in Asheville, NC, over 30 miles away. This burn was extensively studied using a variety of smoke dispersion tools (VSmoke, BlueSky, DaySmoke and CMAQ), but only the simplest model, VSmoke, came close to producing the observed results. Visual observations of the smoke plume indicated that the plume was organized into 4 distinct plume cores, likely a result of the aerial ignition pattern. Applying this concept of multiple plume cores to the more complex dispersion tools improved results in all cases. This highlights a major difference between wildland fire plumes and those from industrial stacks, the ability of the spatial pattern of heat release to interact with the atmosphere and alter the physical structure of the plume, effectively subdividing the plume into a collection of smaller units. These smaller plumes are less effective
than a single core plume at transporting material vertically as the buoyancy is quickly lost as entrainment of ambient air into the plume has a larger impact. Describing the number of plume cores appears to be one of the most important factors governing the vertical distribution of wildland fire emissions; however, at this time we lack a systematic method for estimating the number of plume cores as this number is controlled by complex interactions between the fire and atmosphere.
Impacts
Results to date have unearthed a key difference between industrial emissions sources and wildland fires, the complex interplay between the time/space evolution of heat release and resulting plume structure. This interplay results in the smoke column being split among a number of updraft cores. With only one updraft core, a plume can rise to great heights as entrainment of ambient air only slowly reduces plume buoyancy. When the plume is subdivided into multiple cores, the surface area over which entrainment occurs is increased which reduces plume rise dramatically. Describing the relationship between the spatial/temporal patterns of fire-atmosphere interactions and plume structure is key to better describing smoke dispersion.
Publications
- No publications reported this period
|
Progress 01/01/05 to 12/31/05
Outputs
The project has had to shift some priorities during the first year as there were problems with the availability of funds. Transitions in Forest Service Southern Research Station contracts and grants personnel resulted in significant delays in getting access to project funds. Without these funds the opportunity for collecting field data was largely missed; as a result the project focus shifted toward the modeling portion of the study. The modeling portion of the study involved three components. The first of which is a plume model designed specifically for wildland fires, DaySmoke. Results from DaySmoke have been compared to plume trajectories predicted by the commonly used Briggs formulation developed for industrial pollutant sources. The comparison shows that within a given parameter range DaySmoke is capable of reproducing the results of Briggs; however, this parameter range is not typical of conditions for wildland fires. In conjunction with the Savannah River Smoke
Project, further validation of DaySmoke was possible. Measurements of particular matter emissions from several prescribed fires provided a basis for validating the spatial distribution of particular matter predicted by DaySmoke. Additional observations conducted with the University of Massachusetts Doppler radar system provided vertical distributions of particulate matter within the plume provided further validation data. Analysis is underway and initial results are promising. The comparison to plume trajectories provides one means of validating the DaySmoke plume formulation. An additional source of validation is high resolution large eddy models of thermal plumes from fires. Building on previous work by Dr. Phil Cunningham of Florida State University and the PI, a large eddy version of the Weather Research and Forecast (WRF) model was configured for comparison to the to DaySmoke Simulations and Briggs simulations. While computationally expensive, the large eddy model does an
excellent job of simulating the turbulent nature of strongly buoyant plumes in a boundary layer. Comparisons between the large eddy model and DaySmoke provide confidence that the parameterizations within DaySmoke are capturing the important physics of the problem while maintaining a high degree of computational efficiency. The final modeling component, CMAQ sensitivity studies, is also underway. Several case studies utilizing prescribed fire data provided by the Florida Division of Forestry have revealed that DaySmoke produced considerably higher surface concentrations of PM2.5 over the Layer Fraction Method incorporated in the SMOKE component of CMAQ. As initial comparisons of field concentrations have agreed well with DaySmoke, it is likely that the Layer Fraction Method greatly underestimates the contribution of prescribed fires to local air quality issues. Further simulations are planned to expand upon these findings as well as test alternative plume rise strategies.
Impacts
Preliminary results suggest that current air quality modeling efforts may be underestimating the contribution of smoke from wildland fires to surface air quality issues such as regional haze and the revised National Ambient Air Quality Standards.
Publications
- Achtemeier, G.L. 2005. MEASUREMENTS OF GROUND-LEVEL PM2.5 CONCENTRATIONS DOWNWIND FROM SOUTHERN PRESCRIBED BURNS. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. Canmore, Alberta Canada.
- Cunningham, P. and Goodrick, S. 2005. HIGH-RESOLUTION NUMERICAL MODEL SIMULATIONS OF FIRE PLUME DYNAMICS. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. Canmore, Alberta Canada.
- Liu, Y., Achtemeier, G. and Goodrick, S., 2005. Simulation and Experiment of Air Quality Effects of Prescribed Fires in the Southeast. EastFire 2005 Proceedings, Fairfax, VA.
- Achtemeier, G.L. 2005. On plume rise: matching Daysmoke with Briggs Equations for industrial stacks. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. Canmore, Alberta Canada.
- Liu, Y., Achtemeier, G. and Goodrick, S. 2005. CMAQ-DAYSMOKE AS A SMOKE AND AIR QUALITY MANAGEMENT TECHNIQUE: A CASE STUDY OF A PRESCRIBED BURNING IN GEORGIA. American Meteorological Society. Proceedings of the Sixth Symposium on Fire and Forest Meteorology. Canmore, Alberta Canada.
- Achtemeier, G.L., Goodrick, S. and Liu, Y., 2005. A COUPLED MODELING SYSTEM FOR CONNECTING PRESCRIBED FIRE ACTIVITY DATA THROUGH CMAQ FOR SIMULATING REGIONAL SCALE AIR QUALITY. EastFire 2005 Proceedings, Fairfax, VA.
|