National Academies Press: OpenBook
« Previous: Wildfires and Human Health - An Overview
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Where Are We Now?

To set the stage for the workshop, the first session focused on the current state of the science and communication around atmospheric chemistry and transport of fire emissions, forecasting, measurement tools, and smoke health effects. Presenters reflected on what is being learned in areas where the science is evolving rapidly and highlighted many resources and tools that have been developed in recent years to advance understanding.

The Changing Fire Regime

Jennifer Balch, University of Colorado Boulder, discussed what she views as the three primary ingredients for the changing fire regime: climate, fuel, and ignition. Hotter, drier conditions caused by human-driven climate change are strongly linked with more burning in the western United States (Abatzoglou and Williams, 2016; Balch et al., 2018; Dennison et al., 2014; McKenzie and Littell, 2016; Stavros et al., 2014; Westerling, 2016; Westerling et al., 2006; Williams et al., 2019). Although this pattern is evident, ecosystems vary in whether and how much the size and severity of fires is changing. For example, the size of fires in conifer forests in the Northern Rockies has increased by about 150% while those in the Southern Rockies/Colorado Plateau have increased by more than 400%. More generally, about one-quarter of ecosystems are experiencing an increase in fire severity, meaning that more plants are dying in those ecosystems during fire events than under previous climate conditions. Fires are also spreading in forests that do not typically burn, and ecosystem regeneration following fire may also change.

Humans are also influencing the fuels available to burn by fragmenting landscapes and introducing invasive species. Invasive grasses can double or triple fire activity across large regions, Balch said. For example, invasive cheatgrass carpets about 40,000 km2 (almost 10 million acres) across the United States and is a “flashy fuel” that allows fire to spread quickly. Cheatgrass has been implicated in some of the largest fire events in the United States.

A third ingredient of the changing fire regime is ignition. Humans were responsible for igniting over 84% of U.S. wildland fires in recent years. Human-ignited fires occur across a larger portion of the year than lightning-ignited fires, which primarily occur during the summer. Consequently, human-ignited fires have lengthened the fire season and led to burning of fuels that are higher in moisture content (Balch et al., 2017) (Figure 1). Human ignitions also co-occur with windier conditions than do lightning ignitions.

Looking to the future, wildfire smoke is expected to be more intense and widespread (McKenzie et al., 2014; Spracklen et al., 2009; Yue et al., 2013), Balch said. Depending on weather conditions, this smoke could reach communities—60 million homes in the United States were within 1 km of a wildland fire between 1992 and 2015, and 13 million people who are also socially vulnerable live in areas of extreme fire risk (Davies et al., 2018; Mietkiewicz et al., 2020), raising equity concerns. Unlike other natural disasters that are not initiated directly by human actions, humans have the agency and power to shift the dialogue from an emergency mindset to one that is more proactive to ameliorate wildland fire disasters, Balch said.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Image
FIGURE 1. Frequency distribution of wildland fires ignited by humans (red) and lightning (blue) on each day of the year across the contiguous United States from 1992 to 2012. SOURCE: Balch et al. (2017), in Balch presentation.

Modeling Smoke Plumes

Brian Potter, U.S. Department of Agriculture Forest Service (USFS), discussed the structure of smoke plumes that come out of wildland fires and considerations when incorporating these dynamics into atmospheric models and estimating smoke transport. Fire plumes vary in their complexity, which can be modeled conceptually. The simplest model is an axis-symmetric plume with no wind and a vertical plume that mixes and spreads out when it reaches a stable layer (Figure 2a). This is most commonly observed in prescribed burns where human ignition of a fire perimeter results in the fire burning more rapidly inward than outward. When wind is added to the model, the fire and plume tilt, taking on more of a boomerang shape (Figure 2b). The ideal model for this situation is one where the plume entrains all the smoke into the updraft that then mixes and reaches a stable level. A further layer of complexity that can match what is observed in the field is to add surface burning with smoldering behind the flaming front (Figure 2c). In this case, some of the smoke is entrained into the major plume while some remains near the ground, creating a much more complex vertical distribution of smoke and variability in the horizontal and downwind transport of the plume.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Image
FIGURE 2. Graphical representation of advancing complexity of fire plumes, including an axis-symmetric plume (a), a boomerang-shaped plume resulting from the addition of wind (b), and additional complexity due to surface burning and smoldering (c). SOURCE: Adapted from Potter presentation, U.S. Forest Service.

There are many other complicating factors to consider when modeling plumes, Potter said. These include spatial and temporal variation in the balance between wind and heat that gives the plume buoyancy; vertical and horizontal wind shear effects on mixing; the influence of canopy structure on turbulence and vertical movement of smoke; burning area geometry and scale; and interactions among fires occurring close to one another, which may draw smoke into different plumes or create a single plume. Another consideration is whether to use a single versus multiple convective cores in the model, which has implications for how to treat vertical distribution and subsequent horizontal transport and mixing of the plume.

Potter noted that merging smoke plume models with atmospheric models requires consideration of assumptions about how the plume mixes vertically as well as with the environment as the smoke rises. It is also important to know whether the mixing assumptions are consistent among the models or whether they are redundant, overmix, or compete with one another. Additionally, the spatial scale of atmospheric models (e.g., 4-km and 12-km grid cells) and the size of the plume matter for interpretation. Most of the plume rise models in use assume that the perturbation due to a plume is small when compared to the other environmental conditions being modeled within a given grid cell. However, if a plume is large and present across a large portion of the grid cell, it may not be possible to determine whether the atmospheric model is accounting for the influence of the plume.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Predicting Smoke

Workshop panelists discussed current predictive, modeling, and forecasting capabilities for smoke, as well as opportunities for advancement in the next decade.

Ravan Ahmadov, NOAA Global Systems Laboratory and the Cooperative Institute for Research in Environmental Sciences at the University of Colorado Boulder (CIRES), explained how well weather forecast models, such as the NOAA High-Resolution Rapid Refresh (HRRR)Smoke model, work for smoke transport. Smoke and air quality models estimate fire emissions using different methods to forecast smoke transport; however, all models also use fire detection data obtained from satellites. The HRRR-Smoke model uses fire radiative power (FRP) observed from satellites as an indicator for heat energy generated by fires. Emissions of both PM2.5 and gas species can be calculated from fire radiation energy (calculated from FRP data in combination with assumptions about fire duration) and emission factors. The HRRR-Smoke model forecasts smoke (primarily PM) as a passive tracer without chemistry at high resolution (3 km x 3 km) for the contiguous United States, producing a new forecast every hour using FRP data from the previous 24 hours and forecasting up to 48 hours into the future. This includes capturing spatial gradients in the smoke distribution influenced by meteorology and complex terrain. The HRRR-Smoke model is also capable of capturing the feedback between smoke and radiation that results in dramatically reduced direct radiation reaching the surface due to the heavy smoke.

However, there are cases where the HRRR-Smoke model is not able to forecast smoke distribution in the atmosphere, Ahmadov explained. This includes situations where pyrocumulonimbus clouds develop, which occurs frequently in the western United States. There is a need to develop next-generation forecast models at fine (subkilometer) scale in order to forecast smoke, especially over complex terrain, and these models should include full O3 and aerosol chemistry, aerosol-meteorology interactions, and assimilation of in situ and satellite atmospheric composition measurements, Ahmadov said.

Integration of Aircraft and Satellite Data into Forecast Models

James Crawford, NASA, used data from the Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) aircraft campaign to demonstrate how in situ measurements from aircraft can be used with satellite data to trace impacts of fire on the atmosphere over time and to help better constrain models. Across numerous western U.S. wildfires, a strong relationship was observed between the relative change in carbon dioxide (CO2) analyzed from an aircraft and the FRP calculated by combining geostationary satellite data, aircraft windows measurements, and data collected downwind of the fire plume. FRP allows researchers to determine whether aircraft measurements have captured the peak of the fire and, with a time series of FRP from geostationary satellites, can be used to apportion emissions in time on a much finer timescale than is typically done in models. Crawford noted that it is still necessary to retrospectively apportion total emissions in terms of area burned and fuel consumed, but the CO2-FRP relationship demonstrates that aircraft campaigns can bring additional information to the interpretation of satellite data. Crawford also explained that VOCs vary across fire plumes, but show a strong and predictable relationship with CO.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Reactive nitrogen species are much more variable in smoke plumes due to factors including fuel type and burning conditions. The availability of reactive nitrogen affects the secondary chemical reactions that occur and what species are present both near and downwind of the fire. For instance, nitrous acid can drive production of radicals close to the fire while peroxyacetyl nitrate (PAN) can halt reactions near the fire and enable reactions to occur much farther downwind. In contrast, when formaldehyde dominates radical production, radical-radical reactions occur and are more permanent, reducing the occurrence of reactions downwind. Additional factors that contribute to complexity in the chemistry and require more study include O3 production, primary and secondary aerosol chemistry, brown carbon absorption, the air toxics generated in the plume, and the balance of day and night chemistry, among other factors, Crawford said. The ultimate challenge is to identify how to simplify these complexities so that processes can be incorporated into global and regional scale models.

Brad Pierce, University of Wisconsin, explained how observations from a variety of satellites can currently be incorporated into air quality forecasting models for wildfires, using the Williams Flats Fire in August 2019 as an example. Pierce first presented the integration of FRP and aerosol optical depth (AOD) data collected at 5-minute intervals by the Geostationary Operational Environmental Satellite with emissions and injection height predictions in the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) for the Williams Flats Fire in August 2019 (Figure 3). Results showed that the WRF-Chem forecast was able to capture the structure of the fire well in terms of AOD. The magnitude of aerosol backscatter and the injection height for this fire were also captured by comparing model data to observations from the High Spectral Resolution Lidar measured from a FIREX-AQ aircraft.

Polar orbiting satellites can also be used, Pierce explained. Coordinated measurements of AOD collected from the NOAA-20 Visible Infrared Imaging Radiometer Suite (VIIRS) instrument and high-resolution trace gas retrievals from the new TROPOspheric Monitoring Instrument (TROPOMI) measuring ultraviolet/near infrared captured CO enhancements in the plume from the Williams Flats Fire. Forecasting this fire plume was also conducted at coarser resolution (0.5 degree) at the global scale using the experimental version of the NOAA Unified Forecasting System Real-time Air Quality Modeling System. Pierce explained that this approach captured the AOD well, but it was only with the assimilation of measurements from TROPOMI that the model captured CO enhancements reasonably well for the Williams Flats Fire at this scale. Because wildfires are a global process and not confined to regional scales, without the data assimilation of global measurements, in this instance from TROPOMI, the smoke plume and resulting impact on air quality over North America would be underestimated.

Looking to the future, Pierce noted that there is a new geostationary constellation with aerosol measurements that includes an Advanced Baseline Imager and Advanced Himawari Imager that are coupled with VIIRS to provide a global perspective for aerosols. The same will be available for trace gases with Tropospheric Emissions: Monitoring of Pollution (TEMPO), Sentinel-4, and Geostationary Environment Monitoring Spectrometer providing hourly capabilities to look at trace gas distributions associated with wildfires that are available now on a daily time step from TROPOMI.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Image
FIGURE 3. Observed and modeled information from the Williams Flats Fire. The left panel illustrates radiative power and aerosol optical depth data obtained by the Geostationary Operational Environmental Satellite, and the right panel shows that information combined with emissions and injection height predictions in the Weather Research and Forecasting model coupled with Chemistry. SOURCE: (left) Aditya Kumar, in Pierce presentation; (right) Kumar et al. (2022).

Forecasting O3 and PM2.5

Kirk Baker, U.S. Environmental Protection Agency (EPA), discussed how accurately models are characterizing O3 and PM2.5 impacts from wildfires at different scales with an aim of differentiating wildfires from other sources for regulatory assessments. O3 produced from wildfire can vary considerably across spatial and temporal scales, with effects local to the fire and across a region. The VOCs and oxidized nitrogen gases (nitric oxide [NO] and NO2, or NOx) produced by wildfires are precursors for O3 formation. Near the fire (within tens of miles), O3 production is typically inhibited by NO emissions, which destroy O3 faster than it can form. However, farther downwind (tens to hundreds of miles), VOCs, NOx, and photochemically produced compounds including oxidized nitrogen like PAN produced from NOx can lead to O3 formation when conditions are favorable, such as warm, sunny days. Predicting O3 production is complicated by the presence of PM in smoke, which can attenuate light and affect O3 formation.

Models can predict high O3 concentrations both close to the fire and much farther downwind. Baker explained that for the 2011 Wallow Fire in eastern Arizona, modeled O3 downwind was a combination of that produced at the fire and transported and O3 formed as a result of NOx (created by the thermal decomposition of PAN) reactions, plus reactions of formaldehyde (a VOC). In areas where there is limited background NOx, the decomposition of

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

PAN to NOx can be a mechanism for O3 formation. When a plume moves through an urban area with high background NOx, such as Denver, Colorado, in Baker’s example, O3 production may be more affected by highly reactive VOCs originating from the fire or in the plume.

Baker compared O3 and PM2.5 model predictions with field data and found that the models perform well in replicating local- to regional-scale smoke plume transport. However, compared to routine monitoring data from rural areas, the models seem to systematically overpredict surface level O3. Data from field studies allow for verification of key aspects of modeling wildfire, O3 production, and PM2.5, Baker explained. To evaluate and develop photochemical models, four aspects of the system need to be constrained: (1) location and fuel consumed and area burned, (2) emissions, (3) plume rise, and (4) plume transport and chemistry (Figure 4). While field studies may not be able to address all these aspects at the same time for the same fire, recent campaigns in particular have provided much more information that can inform which components can be improved on and whether any are compensating for one another, which could lead to misinterpretation of model output.

Some challenges of modeling both specific fires and fires over a broad region remain. It is difficult to get accurate activity data; large fires are well detected by satellites, but smaller prescribed fires are often missed in places like the southeastern United States. Baker suggested

Image
FIGURE 4. Key components of wildland fires for which better understanding is needed to improve air quality modeling. SOURCE: Baker presentation, U.S. Forest Service.
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

this gap could be filled by a better ground-based, bottom-up tracking system for prescribed fire activity, or satellites with more highly resolved temporal and spatial information. Another challenge is synthesizing data on combustion phase, moisture, and fuel type from field and chamber studies, and integrating the data in a manageable way in larger-scale modeling systems. Once there is enough confidence in these modeling systems, they can be used as tools to weigh trade-offs between land management techniques and wildfire impacts on air quality.

Challenges of Modeling Vertical Dynamics of Smoke

Following the presentations, panelists discussed what they think are the biggest uncertainties in forecast modeling. The modeling of vertical distribution and vertical mixing of smoke plumes was raised by all of the panelists as a major challenge, which has direct implications for human health given that pollutants near the Earth’s surface can be inhaled. Smoke plumes are extremely complex, and current models rely on traditional meteorological and weather forecast models for vertical mixing, which do not adequately capture variation in the altitude at which the plume is released, the buoyancy, effects of smoldering versus flaming fires, and other factors, Ahmadov explained. Crawford noted that plume rise is very concentrated in the atmosphere whereas the downward motion of smoke is very diffuse, which is difficult for models to capture. Baker added that it can be difficult to disentangle which processes are influencing vertical dynamics at a given time.

For the role of satellite observations in addressing vertical distribution uncertainties, Pierce explained that satellites currently do not provide vertical information about plumes, but future ultraviolet aerosol retrievals (e.g., TEMPO) will provide some information about the height of the aerosol layer and insights into whether the aerosol layer is in the boundary layer or aloft. Satellites have limited ability to provide information about plume structure, Pierce said, particularly in terms of boundary layer dynamics relevant to surface air quality for health, and that is where ground-based sensor data are needed.

The importance of ground-based sensors to collect information on vertical structure, and their use more generally for air quality modeling and monitoring, was discussed. Baker explained that EPA’s Photochemical Assessment Monitoring Station (PAMS) sites will have ceilometers that measure cloud height and thickness in the future, which will provide information on vertical structure. PAMS sites will also be instrumented with Pandora spectrometers, which measure trace gases including O3, NO2, and formaldehyde throughout the atmospheric column. Other data sources such as the AirNow near-real-time AQI (Box 2) and the PurpleAir Network also provide useful information, which could be used to improve spatial representations of real-time air quality that could inform choices such as when to spend time outside, Baker said. While many challenges remain, panelists highlighted new instruments, capabilities, and data that may help to better constrain models and improve air quality forecasting in the future.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Combustion Chemistry

Carsten Warneke, NOAA and CIRES,1 gave a presentation prepared by Bob Yokelson, University of Montana, on combustion chemistry and how it can be used to improve PM forecasting.

When trying to forecast PM, Yokelson has found that there are key questions related to the timing of impacts that are important to the public. Specifically, “When will it be smoky?” and “When will it go away?” There are many difficulties and major uncertainties associated with smoke production, chemistry, and exposure. Biomass burning that takes place during fires is highly complex and influenced by many factors including fuel characteristics, temperature, humidity, wind, time of day, time of year, and other chemical and physical factors, which interact in nonlinear ways that are difficult to model. Addressing these challenges alongside

___________________

1 Carsten Warneke is currently affiliated with NOAA and at the time of this workshop was also affiliated with CIRES.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

smoke health effects, including the study of metabolomics, is needed to gain a more holistic view of the impacts of smoke, Warneke conveyed.

Warneke next explained Yokelson’s simplified view of biomass combustion processes and the resulting pollutants. When fresh biomass is heated, volatiles evaporate and chemical bonds in the solid fuel break, releasing toxic gases and liquid organic particles (VOCs and PM) in large quantities. These are the processes known as distillation and pyrolysis, and they produce white smoke. As the biomass continues to heat it forms char, which absorbs oxygen and releases heat during the process of gasification. Together, distillation, pyrolysis, and gasification are the processes regulating the composition of smoldering emissions, and the emissions from these processes themselves are often flammable. For instance, VOCs can reach higher temperatures and ignite, resulting in flames. Flames efficiently convert toxic pollutants to less toxic compounds like CO2, black carbon, and NOx and in the process provide a second heat source to drive more pyrolysis, creating a reinforcing feedback loop. Flames also play a large role in the rate of fire spread and provide buoyancy for convection, which can lift flaming and entrained smoldering emissions away from the surface, enhancing the potential for long-range transport. However, these emissions can return to the surface, often with pollutants like O3 that form from further chemical reactions in the smoke plume. Combustion processes, their emissions, and the evolution of the smoke are measured from a variety of platforms, including aircraft, ground sites, and mobile laboratories. Wildfires most commonly produce smoldering emissions from pyrolysis or a combination of pyrolysis, gasification, and flaming. These fires are smokier than most other types of fires, including prescribed fires.

Prescribed fires during the spring and fall seasons can reduce emissions, and are also easier to forecast than wildfires, Warneke explained. Prescribed fires produce about 18 times less PM per unit of area burned when compared to wildfires (Selimovic et al., 2020). Currently, many air quality forecasting models use PM emission factors from prescribed fires, resulting in values that are lower than what is observed for wildfires. Research by Barsanti and colleagues2 and Nergui et al. (2017) included collection of PM data from all monitoring sites in the Northwest for August 2013 and compared those data with a forecast model, which showed that under- and overprediction of PM in the model occurred in different portions of the study region. The area of overprediction suggests there is a PM loss process that is not being captured in the model. Yokelson and colleagues have investigated this loss phenomena, comparing airborne measurements at wildfire sources with downwind surface measurements in Missoula, Montana, and suggest that thermally driven evaporation of PM caused by the change in temperature between aircraft and surface altitude is causing the measured losses (Selimovic et al., 2020).

Chemistry during Smoke Transport

Emily Fischer, Colorado State University, discussed how smoke changes as it is carried away from the source, with respect to pollutants of concern for air quality: PM2.5, O3, and hazardous air pollutants (HAPs).

___________________

2 See http://www.lar.wsu.edu/airpact.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Smoke is the largest contributor to PM2.5 in the Pacific Northwest and directly downwind of this region during extreme fire years and in terms of average annual smoke exposure (McClure and Jaffe, 2018; O’Dell et al., 2020). Organic aerosols (OAs) dominate fine PM mass (Garofalo et al., 2019), and there is a relationship between emissions and subsequent chemical and physical transformations (Hodshire et al., 2019). Large, slowly diluting fire plumes generally exhibit little evaporation, which can allow for accumulation of OA, whereas smaller, quickly diluting fire plumes can lead to faster evaporation, which might decrease OA. There has been considerable uncertainty about the evolution of OA and brown carbon in wildfire plumes because of limited knowledge of precursor emissions and near-source chemical evolution. Recent work has helped to refine understanding of OA sources through the partitioning of OA into the fraction formed from direct oxidation of gas emissions and that resulting from evaporation of primary OA (Palm et al., 2020).

When looking at O3, the presence of wildfire smoke results in an increase in O3 on the order of 5-10 parts per billion, on average, and affects both urban and rural areas (Brey and Fischer, 2016). However, Fischer explained how different timescales and environments make it challenging to predict total O3 production, and in order to accurately predict production there needs to be a focus on chemistry near the fire and urban chemistry downwind, as well as in the space in between. Recent airborne field campaigns including the Western wildfire Experiment for Cloud chemistry, Aerosol absorption and Nitrogen have advanced understanding of daytime chemistry relevant to O3 production (Juncosa Calahorrano et al., 2021; Peng et al., 2020). However, while emissions from large fires can peak in the afternoon to early evening hours (Mu et al., 2011), current understanding of the ability of aircraft observations to constrain nighttime chemistry is limited but will be important for understanding the full life cycle of smoke, Fischer said.

For gas-phase HAPs in smoke, acrolein, formaldehyde, benzene, and hydrogen cyanide are likely the dominant chemicals that pose health risks (O’Dell et al., 2020). For example, heavily fire-prone regions in high-fire years such as 2018 may pose excess cancer risk from smoke mainly due to formaldehyde, the key pollutant for cancer risk. As smoke ages, the risk from HAPs decreases. Fischer noted that an important next phase of understanding HAPs will be to attribute the additional HAP burden to PM2.5 from smoke relative to HAP abundances in urban air.

Generally, models striving to capture chemical transformations as smoke is carried away from the source have both strengths and weaknesses. The connections between aircraft observations and plume-scale models are strong (e.g., Alvarado et al., 2015). At the same time, global-scale models tied to satellite and long-term observations can be useful for identifying problems but do not yet provide solutions for modeling smoke plumes. Fischer stated that significant work is needed to move from plume-scale models to what can easily be implemented in global-scale models in order to adequately capture smoke, and simplifications in order to get at health-centric metrics may be possible.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Health Effects of Wildfire Smoke

Colleen Reid, University of Colorado Boulder, discussed the current state of understanding of various health impacts of wildfire smoke. Generally, there is clear evidence of acute effects of wildfire smoke on mortality, where days with high PM result in increased mortality above what would otherwise be expected. Several recent studies have linked wildfire smoke to diabetic outcomes (Xi et al., 2020; Yao et al., 2020). There is also very consistent evidence that wildfire smoke exacerbates asthma (e.g., Borchers Arriagada et al., 2019), and growing evidence that it exacerbates COPD and respiratory infections. There is a considerable amount of information linking cardiovascular disease to PM effects through a variety of physiological mechanisms (Brook et al., 2010), but there are inconsistent findings in the literature about whether there is an association specifically with wildfire smoke. For instance, Reid et al. (2016) found no association, whereas more recent studies have shown mixed results. Most studies have looked at acute health effects from smoke, and more work is needed to look at both longer-term episodic events and the influence of wildfire smoke on long-term health, Reid said.

There has been growing interest in how PM and air pollution affect birth outcomes, though there are only a few studies looking specifically at the role of wildfire smoke. Reid and colleagues found a small but statistically significant decline in birth weight for babies whose gestation intersected with a wildfire event when compared to those that did not (Holstius et al., 2012). Other research has shown associations with birth weight, preterm birth, and possible impacts on mothers.

Reid also suggested that there is a need for more research on mental health related to smoke plumes. Most mental health studies to date have focused on people who have been evacuated, or lost loved ones or property. However, there is a growing concern about mental health impacts of prolonged periods of poor air quality and smoke when residents are advised to stay indoors for long periods of time.

When it comes to whether the chemical composition of PM2.5 has an effect on health impacts, there is not enough evidence to suggest that PM sources differ in toxicity, according to synthesis reports from the World Health Organization, EPA, and the Health Effects Institute. Consequently, PM2.5 has continued to be regulated by total mass (Adams et al., 2015). While there is not currently evidence that wildfire PM2.5 and PM2.5 originating from other sources affect health differently, recent studies have indicated that it is possible that wildfire PM2.5 is affecting asthma more significantly than other sources (DeFlorio-Barker et al., 2019; Kiser et al., 2020), and, more generally, wildfire smoke is an increasing source of PM2.5, particularly in the western United States (McClure and Jaffe, 2018; O’Dell et al., 2019). More research is also needed into the health implications of O3 produced by wildfires, Reid said.

Reid suggested that more information about other health endpoints of concern is needed, as well as studies that evaluate health impacts of longer-term or episodic, repeated exposures over time. Reid also emphasized the need to investigate why there are differences in findings across health studies (i.e., underlying population health status, components of smoke, severity of outcome, study design). One example of a factor that might lead to differences in findings is how studies quantifying spatiotemporal exposure to wildfire smoke (i.e., from ground monitors, satellite observations, or models) vary considerably.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

Current Communication across Health and Atmospheric Science Fields

Panelists discussed information that is currently available and communicated between health and atmospheric research communities. This included what atmospheric chemistry information is used by the health community, other available information that could be provided, and the usability of information in its current form.

Atmospheric Chemistry Information and Tools for Health Research

Nga Lee “Sally” Ng, Georgia Institute of Technology, shared information and tools that can be contributed by the atmospheric chemistry community to improve understanding of air quality and health impacts of wildfires. As discussed by previous speakers, wildfires produce a number of gases and particles that undergo oxidation and other complex and nonlinear reactions that generate a variety of pollutants. Atmospheric chemistry can provide information on sources and chemical composition, as well as link smoke composition to atmospheric loading, to quantify properties of interest for air quality and health such as particle volatility and toxicity.

The interactions between the atmospheric chemistry and health communities often happen around model output and data from monitoring networks or satellites, with a focus on how PM mass concentration relates to exposure assessments and epidemiological studies, Ng said. However, the atmospheric chemistry community can also help to develop a mechanistic understanding of cause and effect by attributing smoke constituents to their sources and ultimately linking the toxicity of smoke components to their chemical composition. Laboratory experiments, for example, can aid in the development and validation of models and can be used to look at secondary organic aerosol (SOA) formation and the resulting oxidative stress to cells from exposure to those particles. Studies have shown that oxidative stress generated by biomass burning aerosols, such as those produced by wildfires, is higher than aerosols found in urban environments.

Ng also highlighted opportunities for improvement, including quantification of wildfire emissions and understanding of transformations in the atmosphere, particularly the oxidation of VOCs and formation of SOA and O3. This will require more long-term measurements of atmospheric compounds and opportunities to develop low-cost techniques for continuous monitoring and exposure assessment, she said. Enhanced engagement between atmospheric chemistry and health researchers, as well as more platforms for interaction, such as workshops, conferences, and funding opportunities, are needed, Ng added.

Health Community Use of Atmospheric Chemistry Information

Rish Vaidyanathan, U.S. Centers for Disease Control and Prevention (CDC), discussed how the health community uses atmospheric science information, and the format and usability of available data. Because producing atmospheric chemistry information is generally resource intensive, the health community relies on academic partners and federal agencies working in atmospheric chemistry fields to produce the data products that can then be used to quantify exposure and link to health data.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

There are many health applications for highly resolved atmospheric model outputs and data products including public health surveillance, epidemiological assessments, and situational awareness and emergency response. One example Vaidyanathan provided was of a real-time smoke vulnerability assessment where an existing operational forecast of surface smoke from the National Weather Service was linked with various measures of vulnerable populations to estimate where smoke was going and the populations to be impacted. This allowed for identification of areas of interest and opportunities to alert vulnerable groups to the smoke health risks.

Vaidyanathan highlighted challenging areas for the health community as well as current research gaps. He suggested establishing a central repository for historical smoke predictions that follows data collection standards and is available in a format that is accessible to public health agencies and practitioners to use readily. Vaidyanathan also pointed to a need for smoke exposure information to be provided at a county or subcounty level, rather than grid scale, for conducting health risk assessments. Finally, he described a need for atmospheric science information and data products to support comparative assessments of smoke exposure for prescribed burns versus wildfires.

Strengthening Epidemiological Research through Greater Interdisciplinary Collaboration

Ana Rappold, EPA, discussed challenges to using air quality ground monitoring data to define exposure in health effects research, as well as opportunities to build on earlier approaches and data to improve evaluation of the public health burden from wildfire. Looking back to research from a decade ago, studies relied on monitoring data and the time period of exposure, without any additional information on the exposure. Monitors continue to be commonly located close to population centers and far away from fires, and therefore are not representative of populations with the highest smoke exposure. In addition, at high levels of smoke, the data from monitors may be uncertain and monitors can malfunction, creating data gaps. In recent years, epidemiological research has become more interdisciplinary and studies have become more complex, with higher spatial and temporal resolution. Instead of using just sparse monitoring data, studies utilize chemical transport and dispersion models as well as smoke satellite images, which allow for parameterization of population exposures on daily and subdaily timescales and over larger regions and longer time periods.

In order to represent the high level of uncertainty in these models, Rappold suggested making modeling data publicly available, using data fusion models, and making multiple exposure models accessible. Rappold also encouraged the integration of research tools such as sensors, wearable technology, and satellite data in order to improve health outcomes. Improved collaboration among research disciplines to increase the operational capacity of existing applications to better communicate health risks of smoke in real time would also be beneficial, Rappold said.

Improving Communication between Health and Atmospheric Science Experts and with the Public

Following the presentations, speakers discussed the historical focus on the health effects of PM and whether this is limiting, as well as what other constituents should be

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

measured to improve understanding. PM serves as a proxy for wildfire smoke and while smoke contains a complex mixture of pollutants, removal of any single component would not remove all health effects. From a communications standpoint, focusing on PM sends a strong message, even if there are differences in toxicity and particle composition. Speakers noted that studying the chemical components of smoke is important, but greater consistency in what is measured across different fires would also be beneficial for determining whether other constituents should be receiving more attention from health scientists. An additional challenge for understanding health effects is quantifying the contributions of PM and O3 from wildfires compared to their background levels.

Rappold noted that forecast models and real-time information can motivate individuals to make decisions and behavioral changes to protect their health. However, behavioral changes in response to wildfire smoke episodes are not currently enough to combat the observed health effects. There is an opportunity to communicate better with the public, for example, when populations are experiencing smoke from prescribed burns. Ng added that wildfire events are a great opportunity for outreach and engaging community (or “citizen”) scientists by having members of the public measure PM concentrations inside and outside their homes, which could increase public awareness and change behavior.

The panelists agreed that building more bridges between the health and atmospheric science communities is needed. Examples for how this has been done include the NASA Health and Air Quality Applied Sciences Team, the American Geophysical Union GeoHealth Section, and the International Association of Wildland Fire’s International Smoke Symposium. Panelists also pointed out that funding agencies have an opportunity to help to organize and integrate these communities by funding research that marries health effects and atmospheric sciences.

Some Session Themes

Susan Anenberg, George Washington University, summarized the first session of the workshop. Some key themes she identified across the session included the following:

  • It is not what has been learned but rather what is being learned—this is a rapidly evolving area of science with many new resources and tools to expand capabilities.
  • Questions about uncertainties remain and whether uncertainties are narrowing or have grown larger than previously thought based on new information. These are important considerations when trying to bound what is known about fire and smoke and links to human health, Anenberg said.
  • There are many opportunities for synergy between atmospheric scientists and health scientists. There is added complexity in that there are many types of atmospheric scientists (e.g., modelers, remote sensing experts using satellite measurements, those that collect empirical measurements from aircraft campaigns), and health scientists (e.g., clinicians, epidemiologists, toxicologists). It will be helpful to provide more opportunities to bring these communities together.
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×

This page intentionally left blank.

Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 9
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 10
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 11
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 12
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 13
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 14
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 15
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 16
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 17
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 18
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 19
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 20
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 21
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 22
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 23
Suggested Citation:"Where Are We Now?." National Academies of Sciences, Engineering, and Medicine. 2022. Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop. Washington, DC: The National Academies Press. doi: 10.17226/26465.
×
Page 24
Next: Where Do We Want to Be? »
Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop Get This Book
×
 Wildland Fires: Toward Improved Understanding and Forecasting of Air Quality Impacts: Proceedings of a Workshop
Buy Paperback | $30.00 Buy Ebook | $24.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Wildland fires pose a growing threat to air quality and human health. Fire is a natural part of many landscapes, but the extent of area burned and the severity of fires have been increasing, concurrent with human movement into previously uninhabited fire-prone areas and forest management practices that have increased fuel loads. These changes heighten the risk of exposure to fire itself and emissions (smoke), which can travel thousands of miles and affect millions of people, creating local, regional, and national air quality and health concerns.

To address this growing threat, the National Academies brought together atmospheric chemistry and health research communities, natural resource managers, and decision makers to discuss current knowledge and needs surrounding how wildland fire emissions affect air quality and human health. Participants also explored opportunities to better bridge these communities to advance science and improve the production and exchange of information. This publication summarizes the workshop discussions and themes that emerged throughout the meeting.

READ FREE ONLINE

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!