S. P. Urbanski

Aerosol emissions from prescribed fires in the United States: A synthesis of laboratory and aircraft measurements

Aerosol emissions from prescribed fires can affect air quality on regional scales. Accurate representation of these emissions in models requires information regarding the amount and composition of the emitted species. We measured a suite of submicron particulate matter species in young plumes emitted from prescribed fires (chaparral and montane ecosystems in California; coastal plain ecosystem in South Carolina) and from open burning of over 15 individual plant species in the laboratory. We report emission ratios and emission factors for refractory black carbon (rBC) and submicron nonrefractory aerosol and compare field and laboratory measurements to assess the representativeness of our laboratory-measured emissions. Laboratory measurements of organic aerosol (OA) emission factors for some fires were an order of magnitude higher than those derived from any of our aircraft observations; these are likely due to higher-fuel moisture contents, lower modified combustion efficiencies, and less dilution compared to field studies. Nonrefractory inorganic aerosol emissions depended more strongly on fuel type and fuel composition than on combustion conditions. Laboratory and field measurements for rBC were in good agreement when differences in modified combustion efficiency were considered; however, rBC emission factors measured both from aircraft and in the laboratory during the present study using the Single Particle Soot Photometer were generally higher than values previously reported in the literature, which have been based largely on filter measurements. Although natural variability may account for some of these differences, an increase in the BC emission factors incorporated within emission inventories may be required, pending additional field measurements for a wider variety of fires.

Determination of smoke plume and layer heights using scanning lidar data

The methodology of using mobile scanning lidar data for investigation of smoke plume rise and high-resolution smoke dispersion is considered. The methodology is based on the lidar-signal transformation proposed recently [Appl. Opt. 48, 2559 (2009)]. In this study, similar methodology is used to create the atmospheric heterogeneity height indicator (HHI), which shows all heights at which the smoke plume heterogeneity was detected by a scanning lidar. The methodology is simple and robust. Subtraction of the initial lidar signal offset from the measured lidar signal is not required. HHI examples derived from lidar scans obtained with the U.S. Forest Service, Fire Sciences Laboratory mobile lidar in areas polluted by wildfires are presented, and the basic details of the methodology are discussed.

Impacts of upwind wildfire emissions on CO, CO2, and PM2.5 concentrations in Salt Lake City, Utah

Biomass burning is known to contribute large quantities of CO2, CO, and PM2.5 to the atmosphere. Biomass burning not only affects the area in the vicinity of fire but may also impact the air quality far downwind from the fire. The 2007 and 2012 western U.S. wildfire seasons were characterized by significant wildfire activity across much of the Intermountain West and California. In this study, we determined the locations of wildfire-derived emissions and their aggregate impacts on Salt Lake City, a major urban center downwind of the fires. To determine the influences of biomass burning emissions, we initiated an ensemble of stochastic back trajectories at the Salt Lake City receptor within the Stochastic Time-Inverted Lagrangian Transport (STILT) model, driven by wind fields from the Weather Research and Forecasting (WRF) model. The trajectories were combined with a new, high-resolution biomass burning emissions inventory—the Wildfire Emissions Inventory. Initial results showed that the WRF-STILT model was able to replicate many periods of enhanced wildfire activity observed in the measurements. Most of the contributions for the 2007 and 2012 wildfire seasons originated from fires located in Utah and central Idaho. The model results suggested that during intense episodes of upwind wildfires in 2007 and 2012, fires contributed as much as 250 ppb of CO during a 3 h period and 15 µg/m3 of PM2.5 averaged over 24 h at Salt Lake City. Wildfires had a much smaller impact on CO2 concentrations in Salt Lake City, with contributions rarely exceeding 2 ppm enhancements.

Grassland and forest understorey biomass emissions from prescribed fires in the south-eastern United States – RxCADRE 2012

Smoke measurements were made during grass and forest understorey prescribed fires as part of a comprehensive programme to understand fire and smoke behaviour. Instruments deployed on the ground, airplane and tethered aerostat platforms characterised the smoke plumes through measurements of carbon dioxide (CO2), carbon monoxide (CO), methane (CH4) and particulate matter (PM), and measurements of optical properties. Distinctions were observed in aerial and ground-based measurements, with aerial measurements exhibiting smaller particle size distributions and PM emission factors, likely due to particle settling. Black carbon emission factors were similar for both burns and were highest during the initial flaming phase. On average, the particles from the forest fire were less light absorbing than those from the grass fires due to the longer duration of smouldering combustion in the forest biomass. CO and CH4 emission factors were over twice as high for the forest burn than for the grass burn, corresponding with a lower modified combustion efficiency and greater smouldering combustion. This dataset reveals the evolution of smoke emissions from two different commonly burned fuel types and demonstrates the complexity of emission factors.

Validation of smoke plume rise models using ground-based Lidar

Biomass fires can significantly degrade regional air quality through the emission of primary aerosols and the photochemical production of ozone and secondary aerosols. The injection height of smoke from biomass burning into the atmosphere (‘plume rise height’) is one of the critical factors in determining the impact of fire emissions on air quality. Plume rise models are used to simulate plume rise height and prescribe the vertical distribution of fire emissions for input to smoke dispersion and air quality models. While several plume rise models exist, their uncertainties, biases, and application limits when applied to biomass fires are not well characterized. The poor state of model evaluation is due in large part to a lack of appropriate observational datasets. We have initiated a research project to address this critical observation gap. In August of 2013 we performed a multi-agency field experiment designed to obtain the data necessary to improve the air quality models used by agricultural smoke managers in the northwestern United States. In the experiment, the ground-based mobile lidar, developed at the US Forest Service Missoula Fire Science Laboratory, was used to monitor plume rise heights for nine agricultural fires in the northwestern United States. The lidar measurements were compared with plume rise heights calculated with the Briggs equations, which are used in several smoke management tools. Here we present the preliminary evaluation results and provide recommendations regarding the application of the models to agricultural burning based on lidar measurements made in the vicinity of Walla Walla, Washington, on August 24, 2013.

Modeling crop residue burning experiments to evaluate smoke emissions and plume transport

Crop residue burning is a common land management practice that results in emissions of a variety of pollutants with negative health impacts. Modeling systems are used to estimate air quality impacts of crop residue burning to support retrospective regulatory assessments and also for forecasting purposes. Ground and airborne measurements from a recent field experiment in the Pacific Northwest focused on cropland residue burning was used to evaluate model performance in capturing surface and aloft impacts from the burning events. The Community Multiscale Air Quality (CMAQ) model was used to simulate multiple crop residue burns with 2 km grid spacing using field-specific information and also more general assumptions traditionally used to support National Emission Inventory based assessments. Field study specific information, which includes area burned, fuel consumption, and combustion completeness, resulted in increased biomass consumption by 123 tons (60% increase) on average compared to consumption estimated with default methods in the National Emission Inventory (NEI) process. Buoyancy heat flux, a key parameter for model predicted fire plume rise, estimated from fuel loading obtained from field measurements can be 30% to 200% more than when estimated using default field information. The increased buoyancy heat flux resulted in higher plume rise by 30% to 80%. This evaluation indicates that the regulatory air quality modeling system can replicate intensity and transport (horizontal and vertical) features for crop residue burning in this region when region-specific information is used to inform emissions and plume rise calculations. Further, previous vertical emissions allocation treatment of putting all cropland residue burning in the surface layer does not compare well with measured plume structure and these types of burns should be modeled more similarly to prescribed fires such that plume rise is based on an estimate of buoyancy.

Contiguous United States wildland fire emission estimates during 2003–2015

Wildfires are a major source of air pollutants in the United States. Wildfire smoke can trigger severe pollution episodes with substantial impacts on public health. In addition to acute episodes, wildfires can have a marginal effect on air quality at significant distances from the source, presenting significant challenges to air regulators’ efforts to meet National Ambient Air Quality Standards. Improved emission estimates are needed to quantify the contribution of wildfires to air pollution and thereby inform decision-making activities related to the control and regulation of anthropogenic air pollution sources. To address the need of air regulators and land managers for improved wildfire emission estimates, we developed the Missoula Fire Lab Emission Inventory (MFLEI), a retrospective, daily wildfire emission inventory for the contiguous United States (CONUS). MFLEI was produced using multiple datasets of fire activity and burned area, a newly developed wildland fuels map and an updated emission factor database. Daily burned area is based on a combination of Monitoring Trends in Burn Severity (MTBS) data, Moderate Resolution Imaging Spectroradiometer (MODIS) burned area and active fire detection products, incident fire perimeters, and a spatial wildfire occurrence database. The fuel type classification map is a merger of a national forest type map, produced by the USDA Forest Service (USFS) Forest Inventory and Analysis (FIA) program and the Geospatial Technology and Applications Center (GTAC), with a shrub and grassland vegetation map developed by the USFS Missoula Forestry Sciences Laboratory. Forest fuel loading is from a fuel classification developed from a large set (> 26 000 sites) of FIA surface fuel measurements. Herbaceous fuel loading is estimated using site-specific parameters with the Normalized Difference Vegetation Index from MODIS. Shrub fuel loading is quantified by applying numerous allometric equations linking stand structure and composition to biomass and fuels, with the structure and composition data derived from geospatial data layers of the LANDFIRE project. MFLEI provides estimates of CONUS daily wildfire burned area, fuel consumption, and pollutant emissions at a 250 m× 250 m resolution for 2003–2015. A spatially aggregated emission product (10 km× 10 km, 1 day) with uncertainty estimates is included to provide a representation of emission uncertainties at a spatial scale pertinent to air quality modeling. MFLEI will be updated, with recent years, as the MTBS burned area product becomes available. The data associated with this article can be found at https://doi.org/10.2737/RDS-2017-0039 (Urbanski et al., 2017). Published by Copernicus Publications. 2242 S. P. Urbanski et al.: US wildland fire emission estimates during 2003–2015

Final Report for SERDP Project RC-1649: Advanced Chemical Measurements of Smoke from DoD-prescribed Burns

Abstract : Project RC-1649, Advanced Chemical Measurement of Smoke from DoD-prescribed Burns was undertaken to use advanced instrumental techniques to study in detail the particulate and vapor-phase chemical composition of the smoke that results from prescribed fires used as a land management tool on DoD bases, particularly bases in the southeastern U.S. The statement of need (SON) called for (1) improving characterization of fuel consumption and (2) improving characterization of air emissions under both flaming and smoldering conditions with respect to ... volatile organic compounds, heavy metals, and reactive gases. The measurements and fuels were from several bases throughout the southeast (Camp Lejeune, Ft. Benning, and Ft. Jackson) and were carried out in collaboration and conjunction with projects 1647 (models) and 1648 (particulates, SW bases).