Final Report: Investigation of the Effects of Changing Climate on Fires and the Consequences for U.S. Air Quality, Using a Hierarchy of Chemistry and Climate Models

EPA Grant Number: R832275
Title: Investigation of the Effects of Changing Climate on Fires and the Consequences for U.S. Air Quality, Using a Hierarchy of Chemistry and Climate Models
Investigators: Logan, Jennifer A. , Byun, Daewon , Diner, David , Jacob, Daniel J. , Li, Qinbin , Mazzoni, Dominic M. , Mickley, Loretta J.
Institution: Harvard University , Jet Propulsion Laboratory - Pasadena , University of Houston
EPA Project Officer: Chung, Serena
Project Period: April 1, 2005 through March 31, 2008 (Extended to March 31, 2010)
Project Amount: $750,000
RFA: Fire, Climate, and Air Quality (2004) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Global Climate Change , Climate Change , Air

Objective:

This project is an assessment of the impacts of climate change on forest fires and ozone and particulate matter air quality in the United States from the present day till 2050. The project explores the relationships between climate and frequency and intensity of forest fires in North America. Future climate predicted using a general circulation model (GCM) and relationships between fire and observed meteorology which we have developed are used to predict future fires in the United States. The height of forest fire plumes over North America are investigated using the MISR satellite data. Using global and regional scale chemistry-aerosol transport models, we investigate the role of future fires on air quality.

Summary/Accomplishments (Outputs/Outcomes):

  1. Fire prediction model for the western United States.

    The first part of this project focused on developing a fire prediction scheme to predict the effect of future climate change on fire activity in the western United States. Stepwise linear regression was used to determine the best relationships between observed area burned and variables chosen from temperature, relative humidity, wind speed, 24-hour rainfall, and from three fuel moisture codes and four fire weather indices from the FWI model, following the approach of Flannigan et al. [2005]. We used observed area burned from the 1ºx1º database of Westerling et al. [2003] for 1980-2004. Area burned was binned according to the ecological stratification of Bailey et al. [1994], which defines 18 ecosystem classes in the western United States. These ecosystems were further aggregated to produce 6 ecoregions with similar vegetation and climate, as we found that we could better fit area burned for the larger ecoregions.

    Our results showed that May-October mean temperature and fuel moisture explain 24-57% of the observed variance in annual area burned in the Western United States. Explained variance is generally greater in forest dominated ecosystems (48-52%) than in shrub and grass dominated ecosystems (24-49%) [Spracklen et al., 2009]. The lower explained variance in the latter ecosystems is likely due to the importance of the previous year's weather for fire activity in these areas [Westerling et al., 2003; Westerling and Bryant, 2008; Littell et al., 2009], which we did not take into account.

    We also used stepwise linear regression to develop predictive relationships between annual area burned in Alaska and Canada and monthly/seasonal surface meteorology, 500 hPa geopotential heights (GPH), and moisture indices from the FWI model for different ecozones. The regressions capture much of the variability in annual area burned over Alaska (53-57%), and Canada (15-62%). Key predictors vary by ecosystem, but temperature, 500 mb geopotential height anomaly, and seasonal severity ratings are often key terms [Hudman et al., in preparation, 2010].

  2. Simulation of future area burned.

    We used the GISS GCM III to generate meteorological boundary conditions from the present day to 2050. Greenhouse gas concentrations in the model were updated yearly following the A1B scenario, which describes a homogeneous future world with rapid economic growth, introduction of new technologies, and balanced energy generation from fossil and alternative fuels. We used the GISS predicted meteorology as input to the FWI model to calculate daily moisture and temperature parameters for May to October of 2000-2050. We then calculated future area burned by applying our regression analysis to the model meteorology and drought and fire indices for these years. Results showed that the A1B climate leads to a 54% increase in area burned in the Western United States by 2050, relative to the present day. Changes in area burned are ecosystem dependent, with the forests of the Pacific Northwest and Rocky Mountains experiencing the greatest increases of 78% and 175%, respectively [Spracklen et al., 2009]. In Alaska, the area burned increased by 34%, but over Canada, it increased by only 8%, as there were regions with both increases and decreases (with a range of -34 to +118%), due to increases in GCM precipitation [Hudman et al., in preparation, 2010].

  3. Effect of future fires on present-day and future air quality.

    We used GEOS-Chem to examine the impact of wildfires on organic carbon (OC) aerosol concentrations in the western United States. We derived OC emissions from wildfires using data for area burned for 1980–2004 and ecosystem specific fuel loadings. Our results showed conclusively that wildfires drive the interannual variability of observed OC concentrations in the West [Spracklen et al., 2007]. We estimated that the observed increase in wildfire activity after the mid 1980s has increased mean summertime OC concentrations by 30% relative to 1970–1984 for this region [Spracklen et al., 2007].

    To calculate the effect of future wildfires on air quality, we used the GEOS-Chem global 3-D model, forced by the GISS calculated meteorology for the present day and future. To compute wildfire emissions, we located each predicted fire randomly within a given ecosystem and distributed the area burned according to observed wildfire behavior. We then used the resulting area burned maps together with ecosystem specific fuel loadings derived from the Fuel Characteristic Classification System (FCCS) of the USDA Forest Service [McKenzie et al., 2007] and emission factors from Andreae and Merlet [2001]. Our results from GEOS-Chem showed for the first time that changing wildfires could have a large impact on OC emissions and thus on U.S. air quality (Figure 1). We found that the increased area burned in the future climate leads to a near doubling of wildfire carbonaceous aerosol emissions by mid-century. Western United States by 40% and elemental carbon (EC) concentrations by 20% from 2000 to 2050. Most of this increase (75% for OC, 95% for EC) is caused by larger wildfire emissions in a warming climate, with the rest caused by direct effect of changing climate on air quality.

    We have also considered the effect of changing wildfires on ozone air quality. Our results show that future wildfires result in an increase of 1-3 ppb in mean summertime afternoon ozone in the Western United States, but episodic enhancements are larger. Long-range transport from Canadian and Alaskan wildfire are predicted to increase ozone by 1-2 ppb over populated southern Quebec cities, and by ~1 ppb in the Midwest of United States, exacerbating predicted increases from climate change due to stagnation [Hudman et al., in preparation, 2010].

  4. Analysis of forest fire plume heights from MISR

    We collaborated with the David Diner of the Jet Propulsion Laboratory, P.I. of the Multi-angle Imaging SpectroRadiometer (MISR) instrument, to develop a data-base of fire plume heights for five years over North America. JPL first tested an automated data mining algorithm to search through the MISR data for plumes that were connected to MODIS fire pixels. This approach identified 77 plumes over Alaska/Yukon in 2004, and showed that the median plume height was 2.2 km [Mazzoni et al., 2007]. The automated approach missed many plumes, however, and the MISRtool was then developed. This tool also uses MODIS fire pixels to find candidate plumes, but relies on a person to digitize each plume shape; the subsequent extraction of MISR data is automated. With this tool, 664 plumes were identified in Alaska/Yukon in 2004, and ~10% were above the boundary layer at the local time of the MISR overpass, ~11:00-12:00 [Kahn et al., 2008]. The MISR group produced height data for almost 3000 plumes in 2002-2007 (except for 2003), and the plume database is available on the web [http//www-misr2.jpl. nasa.gov/EPA-Plumes/]. We used the MODIS Land Cover Map to relate the fire plumes to underlying vegetation, and the GEOS assimilated meteorological products to relate the plume heights to the boundary layer height. We found 4-12% of plumes are injected above the boundary layer (BL), and that most plumes above the BL (> 83%) are trapped within stable layers. A full analysis of the plume data-base, funded by the National Science Foundation as well as by EPA, is given in Val Martin et al. [2010].

  5. CMAQ regional model simulation.

    For a more accurate picture of the impact of future wildfires on air quality, we collaborated with Daewon Byun, of University of Houston to perform simulations with the regional chemical model CMAQ. Harvard supplied the University of Houston with calculated area burned statistics and GEOS-Chem chemical output to use as boundary conditions for CMAQ. Meteorological fields for CMAQ were calculated by MM5, using boundary conditions from the GISS model. University of Houston also developed a downscaling approach for future area burned provided from the coarser resolution GISS model. The downscaling used two probability distributions: one based on the Haines index, which combines atmospheric stability and the moisture content of the lower atmosphere to give the likelihood of a fire occurrence, and one based on several years of data on fire occurrence as a function of different land cover types. We used the CMAQ model for 2000 and 2050 with and without fire emissions to predict ozone and aerosols.

    We used the CMAQ model for 2000 and 2050 with and without fire emissions to predict ozone and aerosols. Results for afternoon ozone in summer are shown in Figure 2. With no fires, ozone is higher in 2050 than in 2000 (Figure 2e), particularly in the Eastern United States, and Mexico, primarily because of the higher temperatures. The effect of fires in 2050 (compared to the 2000 case with fires, Figure 2f) is to increase ozone in the west by 1-7 ppb. Figures 2g and 2h show the spatial extent of the effect of fires alone in 2000 and 2050 for their respective climates. As expected, CMAQ model simulations also show that OC is significantly enhanced in the Western United States by the fires in 2050. This work is being prepared for publication.

Conclusions:

Our main findings are as follows:

  • Wildfires drive the interannual variability of observed OC concentrations in the West. The observed increase in wildfire activity after the mid 1980s has increased mean summertime OC concentrations by 30% relative to 1970–1984 for this region.
  • The A1B climate leads to a 54% increase in area burned in the Western United States by 2050, relative to the present-day. Changes in area burned are ecosystem dependent, with the forests of the Pacific Northwest and Rocky Mountains experiencing the greatest increases of 78% and 175%, respectively. In Alaska, the area burned increased by 34%, but over Canada, it increased by only 8%, as there were regions with both increases and decreases (with a range of -34 to +118%), due to increases in GCM precipitation.
  • Climate change will increase summertime OC aerosol concentrations over the western United States by 40% and elemental carbon (EC) concentrations by 20% from 2000 to 2050. Most of this increase (75% for OC, 95% for EC) is caused by larger wildfire emissions in a warming climate, with the rest caused by direct effect of changing climate on air quality.
  • Results from the regional chemical model show that by 2050 wildfires in the West could increase ozone by 1-7 ppb, relative to the present-day.

In summary, our research has made important contributions toward the EPA’s overarching goals of safeguarding human health and the environment. In particular, our work has enhanced the ability of policymakers to gauge the coming “climate penalty” on ongoing efforts to reduce air pollution across the United States. (Here climate penalty is defined as the additional emission controls necessary to meet a given air quality target [EPA, 2007].) We quantified the impact of climate change on area burned for the western United States, in contrast to previous studies that either estimated changes in a fire index or focused on a small geographic area. We showed that wildfires in the western United States have a significant effect on air quality in the United States in the present-day, and that this effect will likely increase by 2050. Our work represents the first-ever assessment of the effects of fires in a future climate on air quality in the United States. In addition, our analysis of the height of plumes from forest fires has led to more realistic simulations of the effects of fires on present day and future air quality.

References:

Andreae, M. and P. Merlet, Emission of trace gases and aerosols from biomass burning, Global Biogeochemical Cycles, 15, 955-966, 2001.
 
Bailey, R., P. Avers, T. King, and W. McNab, Ecoregions and subregions of the United States (map), Tech. rep., Washington, DC: USDA Forest Service, 1994.
 
EPA, Assessment of the Impacts of Global Change on Regional U.S. Air Quality: A Preliminary Synthesis of Climate Change Impacts on Ground-Level Ozone, EPA, Washington, DC., 2007.
 
Flannigan, M. D., K.A. Logan, B. D. Amiro, W. R. Skinner, and B. J. Stocks, Future area burned in Canada, Clim. Change,72, 1-16, 2005.
 
Kahn, R. A., Y. Chen, D. L. Nelson, F.-Y. Leung, Q. Li, D. J. Diner, and J. A. Logan, Wildfire smoke injection heights – two perspectives from space, Geophys. Res. Lett., 35, L04809, doi:10.1029/2007GL032165, 2008.
 
Littell, J.S., D. McKenzie, D.L. Peterson, and A.L.Westerling, Climate and wildfire area burned in western U.S. ecoprovinces, 1916-2003. Ecol. App., 19, 1003-1021, 2009.
 
Mazzoni, D., J. A. Logan, D. Diner, R. Kahn, L. Tong, and Q. Li, A data-mining approach to associating MISR smoke plume heights with MODIS fire measurements, Remote Sens. Environ., 107, 138-148, 2007.
 
McKenzie, D., C. Raymond, L.-K. Kellogg, R. Norheim, A. Andreu, A. Bayard, K. Kopper, and E. Elman, Mapping fuels at multiple scales: landscape application of the Fuel Characteristic Classification System, Can. J. For. Res., 37, 2421-2437, 2007.
 
Spracklen, D. V., J. A. Logan, L. J. Mickley, R. J. Park, R. Yevich, A. L. Westerling, and D. Jaffe, Wildfires drive interannual variability of organic carbon aerosol in the western U.S. in summer, Geophys. Res. Lett., L16816, doi:10.0129/GL030037, 2007.
 
Spracklen, D. V., L. J. Mickley, J. A. Logan. R. C. Hudman, R. Yevich, M. D. Flannigan, and A. L. Westerling, Impacts of climate change from 2000 to 2050 on wildfire activity and carbonaceous aerosol concentrations in the western United States, J. Geophys. Res., 114, D20301, doi:10.1029/2008JD010966, 2009.
 
Val Martin, M., J. A. Logan, R. Kahn, F.-Y. Leung, D. Nelson, and D. Diner, Smoke injection heights from fires in North America: Analysis of five years of satellite observations, Atmos. Chem. Phys., 10, 1491-1510, 2010.
 
Westerling, A. L., and T. S. Swetman, Interannual to decadal drought and wildfire in the western United States, EOS Trans., 84, (49), 545-560, 2003.
 
Westerling, A.L, and B.P. Bryant, Climate change and wildfire in California, Climatic Change, 87 (Suppl 1):S231–S249, 2008.


Journal Articles on this Report : 5 Displayed | Download in RIS Format

Other project views: All 14 publications 5 publications in selected types All 5 journal articles
Type Citation Project Document Sources
Journal Article Kahn RA, Chen Y, Nelson DL, Leung F-Y, Li Q, Diner DJ, Logan JA. Wildfire smoke injection heights: two perspectives from space. Geophysical Research Letters 2008;35:L04809, doi:10.1029/2007GL032165. R832275 (2007)
R832275 (2008)
R832275 (Final)
  • Full-text: PDF Full-text
  • Abstract: AGU Abstract
    Exit
  • Journal Article Mazzoni D, Logan JA, Diner D, Kahn R, Tong L, Li Q. A data-mining approach to associating MISR smoke plume heights with MODIS fire measurements. Remote Sensing of Environment 2007;107(1-2):138-148. R832275 (2006)
    R832275 (2007)
    R832275 (2008)
    R832275 (Final)
  • Full-text: ScienceDirect-HTML
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Other: ScienceDirect-PDF
    Exit
  • Journal Article Spracklen DV, Mickley LJ, Logan JA, Hudman RC, Yevich R, Flannigan MD, Westerling AL. Impacts of climate change from 2000 to 2050 on wildfire activity and carbonaceous aerosol concentrations in the western United States. Journal of Geophysical Research 2009;114:D20301, doi:10.1029/2008JD010966. R832275 (Final)
  • Full-text: Citeseer
    Exit
  • Abstract: AGU-Abstract
    Exit
  • Journal Article Spracklen D, Logan JA, Mickley LJ, Park RJ, Yevich R, Westerling AL, Jaffe DA. Wildfires drive interannual variability of organic carbon aerosol in the western U.S. in summer. Geophysical Research Letters 2007;34:L16816, doi:10.0129/2007GL030037. R832275 (2006)
    R832275 (2007)
    R832275 (2008)
    R832275 (Final)
  • Full-text: University of California, Merced-Full Text PDF
    Exit
  • Abstract: American Geophysical Union-Abstract
    Exit
  • Journal Article Val Martin M, Logan JA, Kahn RA, Leung F-Y, Nelson DL, Diner DJ. Smoke injection heights from fires in North America:analysis of 5 years of satellite observations. Atmospheric Chemistry and Physics 2010;10(4):1491-1510. R832275 (Final)
  • Full-text: Atmospheric Chemistry and Physics-Full Text PDF
    Exit
  • Abstract: Atmospheric Chemistry and Physics-Abstract
    Exit
  • Supplemental Keywords:

    forest fires, wildfires, biomass burning, air quality, tropospheric ozone, tropospheric aerosol, PM, visibility, climate models, air pollution, climate change, health effects, pollution prevention, public policy,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Environmental Chemistry, climate change, Air Pollution Effects, Aquatic Ecosystem, Monitoring/Modeling, Environmental Monitoring, Ecological Risk Assessment, Atmosphere, anthropogenic stress, environmental measurement, meteorology, climatic influence, global ciruclation model, ozone depletion, tidal marsh, socioeconomics, ecosystem indicators, climate models, aquatic ecosystems, environmental stress, coastal ecosystems, global climate models, climate model, ecosystem stress, forest resources, ecological models, sea level rise, air quality, atmospheric chemistry, climate variability

    Relevant Websites:

    http://www.people.fas.harvard.edu/~logan/research2.htmlexit EPA
    www.as.harvard.edu/chemistry/trop/curresh.html#wildfiresexit EPA
    http://www.people.fas.harvard.edu/~mickley/wildfires.htmlexit EPA
    www.imaqs.uh.eduexit EPA

    Progress and Final Reports:

    Original Abstract
  • 2005 Progress Report
  • 2006 Progress Report
  • 2007 Progress Report
  • 2008 Progress Report