united states department wildland fire in forest service … · 2008. 5. 21. · united states...

United StatesDepartmentof Agriculture

Forest Service

Rocky MountainResearch Station

General TechnicalReport RMRS-GTR-42-volume 5

December 2002

Wildland Fire inEcosystems

Effects of Fire on Air

Abstract _____________________________________Sandberg, David V.; Ottmar, Roger D.; Peterson, Janice L.; Core, John. 2002. Wildland fire on

ecosystems: effects of fire on air. Gen. Tech. Rep. RMRS-GTR-42-vol. 5. Ogden, UT: U.S.Department of Agriculture, Forest Service, Rocky Mountain Research Station. 79 p.

This state-of-knowledge review about the effects of fire on air quality can assist land, fire, and airresource managers with fire and smoke planning, and their efforts to explain to others the sciencebehind fire-related program policies and practices to improve air quality. Chapter topics include airquality regulations and fire; characterization of emissions from fire; the transport, dispersion, andmodeling of fire emissions; atmospheric and plume chemistry; air quality impacts of fire; socialconsequences of air quality impacts; and recommendations for future research.

Keywords: smoke, air quality, fire effects, smoke management, prescribed fire, wildland fire, wildfire,biomass emissions, smoke dispersion

The volumes in “The Rainbow Series” will be published through 2003. The larger bold check-mark boxes indicate the volumescurrently published. To order, check any box or boxes below, fill in the address form, and send to the mailing address listed below.Or send your order and your address in mailing label form to one of the other listed media.

RMRS-GTR-42-vol. 1. Wildland fire in ecosystems: effects of fire on fauna.

RMRS-GTR-42-vol. 2. Wildland fire in ecosystems: effects of fire on flora.

RMRS-GTR-42-vol. 3. Wildland fire in ecosystems: effects of fire on cultural resources and archeology.

RMRS-GTR-42-vol. 4. Wildland fire in ecosystems: effects of fire on soil and water.

RMRS-GTR-42-vol. 5. Wildland fire in ecosystems: effects of fire on air.

Send to: ________________________________________________________________________________Name

________________________________________________________________________________Address

Fort Collins Service Center

Telephone (970) 498-1392

FAX (970) 498-1396

E-mail rschneider/[email protected]

Web site http://www.fs.fed.us/rm

Mailing Address Publications DistributionRocky Mountain Research Station240 W. Prospect RoadFort Collins, CO 80526-2098

Cover photo—Photo by Roger Ottmar. Smoke blots outthe sun during the 1994 Anne Wildfire in western Montana.

Wildland Fire in EcosystemsEffects of Fire on Air

AuthorsDavid V. Sandberg, Research Physical Scientist, Corvallis ForestrySciences Laboratory, Pacific Northwest Research Station, U.S. Depart-ment of Agriculture, Corvallis, OR 97331

Roger D. Ottmar, Research Forester, Seattle Forestry Sciences Labo-ratory, Pacific Northwest Research Station, U.S. Department of Agricul-ture, Seattle, WA 98103

Janice L. Peterson, Air Resource Specialist, Mt. Baker-SnoqualmieNational Forest, U.S. Department of Agriculture, Mountlake Terrace,WA 98053

John Core, Consultant, Core Environmental Consulting, Portland, OR97229

Preface _____________________________________

In 1978, a national workshop on fire effects in Denver, Colorado, provided the impetusfor the “Effects of Wildland Fire on Ecosystems” series. Recognizing that knowledge offire was needed for land management planning, state-of-the-knowledge reviews wereproduced that became known as the “Rainbow Series.” The series consisted of sixpublications, each with a different colored cover, describing the effects of fire on soil,water, air, flora, fauna, and fuels.

The Rainbow Series proved popular in providing fire effects information for professionals,students, and others. Printed supplies eventually ran out, but knowledge of fire effectscontinued to grow. To meet the continuing demand for summaries of fire effects knowledge,the interagency National Wildfire Coordinating Group asked Forest Service research leadersto update and revise the series. To fulfill this request, a meeting for organizing the revision washeld January 4-6, 1993, in Scottsdale, Arizona. The series name was then changed to “TheRainbow Series.” The five-volume series covers air, soil and water, fauna, flora and fuels, andcultural resources.

The Rainbow Series emphasizes principles and processes rather than serving as asummary of all that is known. The five volumes, taken together, provide a wealth of informationand examples to advance understanding of basic concepts regarding fire effects in the UnitedStates and Canada. As conceptual background, they provide technical support to fire andresource managers for carrying out interdisciplinary planning, which is essential to managingwildlands in an ecosystem context. Planners and managers will find the series helpful in manyaspects of ecosystem-based management, but they will also need to seek out and synthesizemore detailed information to resolve specific management questions.

–– The AuthorsDecember 2002

Acknowledgments____________________________

The Rainbow Series was compiled under the sponsorship of the Joint Fire Science Program,a cooperative fire science effort of the U.S. Department of Agriculture, Forest Service, and theU.S. Department of the Interior, Bureau of Indian Affairs, Bureau of Land Management, Fishand Wildlife Service, National Park Service, and U.S. Geological Survey.Several scientists provided significant input without requesting authorship in this volume. Weacknowledge valuable contributions by Sue A. Ferguson, Timothy E. Reinhardt, RobertYokelson, Dale Wade, and Gary Achtemeier. We also thank the following individuals for theirsuggestions, information, and assistance that led to substantial technical and editorialimprovements in the manuscripts: Scott Goodrick, Allen R. Riebau, Sue A. Ferguson, and PattiHirami. Finally, we appreciate Marcia Patton-Mallory and Louise Kingsbury for persistence andsupport.

ii

Contents________________________________________________Page Page

iii

Summary ........................................................................ ivChapter 1: Introduction .................................................... 1Objective ............................................................................. 2Related Publications ........................................................... 2Scope .................................................................................. 2Framework .......................................................................... 2Prior Work ........................................................................... 3

Smoke Management Guide For Prescribed and Wildland Fire: 2001 Edition ............................... 3Wildland Fire and Air Quality: National Strategic Plan.......................................................................... 4Introduction to Visibility ............................................... 4The Federal Advisory Committee Act White Papers ..................................................................... 4Environmental Regulation and Prescribed Fire Conference .............................................................. 5Southern Forestry Smoke Management Guidebook ............................................................... 6

Changes in Fire Policy ........................................................ 6Joint Fire Science Program ........................................ 6Cohesive Strategy ...................................................... 7National Fire Plan ....................................................... 7

Chapter 2: Air Quality Regulations and Fire .................. 9Roles and Responsibilities Under the Clean Air Act ......... 10National Ambient Air Quality Standards ............................ 11Prevention of Significant Deterioration .............................. 11Visibility ............................................................................. 12

Regional Haze .......................................................... 13Reasonable Progress ............................................... 15

Hazardous Air Pollutants .................................................. 15EPA Interim Air Quality Policy on Wildland and

Prescribed Fires..................................................... 16Natural Events Policy ........................................................ 16Collaboration Among Stakeholders .................................. 16Best Available Control Measures ...................................... 16

Reducing Emissions ................................................. 17Redistributing Emissions .......................................... 17

Ozone and Fire ................................................................. 17Chapter 3: Overview of Air Pollution from Fire ............ 19Magnitude of Fire Contributions ........................................ 19

Smoke from Wildland Fires ...................................... 20Smoke from Prescribed Fires ................................... 24Impacts on National Ambient Air Quality Standards .............................................................. 24Significance of Visibility Degradation ........................ 24Greenhouse Gas Emissions from Fires .................... 24

Smoke Management Programs ........................................ 25Chapter 4: Characterization of Emissions from

Fires ....................................................................... 27Area Burned ...................................................................... 27Preburn Fuel Characteristics ............................................ 28Fire Behavior ..................................................................... 29Combustion Stages ........................................................... 30Fuel Consumption ............................................................. 31Emission Factors ............................................................... 32Source Strength ................................................................ 32Chapter 5: Transport, Dispersion, and Modeling

of Fire Emissions ................................................... 35

Basic Elements of Trajectory and Dispersion ................... 35Heat Release ............................................................ 36Plume Rise and Buoyancy ....................................... 36Advection and Diffusion ............................................ 37Scavenging ............................................................... 38Chemical Transformations ........................................ 38

Transport and Dispersion Models ..................................... 38Plume Models ........................................................... 38Puff Models ............................................................... 39Particle Models ......................................................... 39Grid Models .............................................................. 39

Model Application .............................................................. 40Chapter 6: Atmospheric and Plume Chemistry ........... 41Ozone Formation in Plumes ............................................. 41Factors Affecting Plume Chemistry ................................... 42Emission Factors for Reactive Species ............................ 43Particle Formation in Plumes ............................................ 43Chapter 7: Estimating the Air Quality Impacts of

Fire ................................................................. 45Emission Inventories ......................................................... 45

State Emission Inventories ....................................... 46Regional Emission Inventories ................................. 46National Emission Inventories .................................. 47Improving Emission Inventories ................................ 47

Air Quality Monitoring ........................................................ 48Current Monitoring Techniques ................................ 48

Source Apportionment ...................................................... 49Source Apportionment Methods ............................... 50Receptor-Oriented Approaches ................................ 50Factor Analysis and Multiple Linear Regression ...... 52Summary .................................................................. 52

Mechanistic Models .......................................................... 53Chapter 8: Consequences of Fire on Air Quality ......... 55Health Effects .................................................................... 55

National Review of Health Effects ............................ 55Occupational Exposure to Wildland Fire Smoke ...... 56Research Issues ....................................................... 57

Welfare Effects .................................................................. 58Soiling of Materials ................................................... 58Public Nuisance and Visibility Loss .......................... 58

Economic and Social Consequences ............................... 59Soiling-Related Economic Losses ............................ 59Visibility-Related Costs ............................................. 59

Highway Safety ................................................................. 60Magnitude of the Problem ........................................ 60Measures to Improve Highway Safety ...................... 60

Climate Change ................................................................ 61Chapter 9: Recommendations for Future Research

and Development .................................................... 63Established Research Framework .................................... 63Emerging Research Needs ............................................... 65

Emissions Source Strength and Emissions Inventory ................................................................ 65Ambient Air Quality Impacts ..................................... 66Effects on Receptors ................................................ 66

Conclusion ........................................................................ 67References ........................................................................ 69

SummaryWildland fire is an integral part of ecosystem manage-

ment and is essential in maintaining functional ecosys-tems, but air pollutants emitted from those fires can beharmful to human health and welfare. Because of thepublic and governmental concerns about the possiblerisk of wildland fire smoke to public health and safety, aswell as nuisance, visibility, ozone generation, and re-gional haze impacts, increasingly effective smoke man-agement programs and air quality policies are beingimplemented with support from research and land man-agement agency programs.

This state-of-knowledge review of what is known aboutthe effects of fire on air quality has been prepared toassist those in the fire and air quality managementcommunities for future discussion of management, policy,and science options for managing fire and air quality. Theintroduction sets up a framework in which to discuss theinteraction between pollutants emitted from fire, and airquality at the national, State, and local levels applied toair resource management, fire management, and geo-graphical scale components. It also provides an over-view of science reviews conducted since 1979 anddiscusses recent changes in fire policy, strategies, andfunding. The Clean Air Act and its amendments arediscussed in chapter 2, in the context of how and why fireimpacts each issue, what information is needed, and whoneeds it to fulfill legal requirements under the act. Na-tional ambient air quality standards, regional haze andvisibility, hazardous air pollutants, and best availablecontrol methods are some of the topics covered. Chapter3 covers the magnitude of the impacts of prescribed andwildland fire on air quality, and contains an overview ofsmoke management plans intended to manage thoseimpacts.

Chapters 4 through 7 present scientific and technicaldiscussions. Chapter 4 discusses the characterizationand production rate of emissions from fire in terms offuels, fire behavior, stages of combustion, fuel consump-tion, and emission factors of various pollutants. The basicelements and modeling of transport and dispersion arecovered in chapter 5, including, plume, puff, particle, andgrid models. Chapter 6 considers plume and atmo-spheric chemistry, the chemical reactions that occur inplumes, with a focus on ozone formation and particleformation. Use of emission inventories, air quality moni-toring, and source apportionment methods, and mecha-nistic models to estimate the impacts of fire on air qualityare covered in chapter 7. Chapter 8 reviews the health,welfare, economic, and safety consequences of theseimpacts. The final chapter recommends priorities forfuture research to better understand and quantify fire andits effect on air quality.

iv

Metric EquivalentsWhen you know: Divide by: To find:

Feet (ft) 3.28 Meters

Pounds (lb) 2.21 Kilograms

Acres 2.47 Hectares

Pounds per acre 0.89 Kilograms per hectare

Fahrenheit (°F) 1.8 and subtract 32 Celsius

USDA Forest Service Gen. Tech. Rep. RMRS-GTR-42-vol. 5. 2002 1

Chapter 1:Introduction

A state-of-knowledge review, Effects of Fire on Air,was written in 1979 to inform environmental agen-cies, fire managers, and land management planners,and to guide research strategies in the interveningyears (Sandberg and others 1979). That review is stilltechnically sound for the most part, but substantialnew knowledge is now available. In this volume, weupdate that review of knowledge important for man-aging the effects of fire on air and for adjusting thecourse of new research. In addition, we expand thescope of our review to place the information in thecontext of new policies regarding fire managementand air quality management

Acquisition of scientific knowledge regarding airpollution from fires is motivated by active policy devel-opment both to restore the role of fire in ecosystemsand to improve air quality. Land managers requirequantitative analysis and goal-seeking solutions tominimize the negative consequences of fire manage-ment. Managing fire and air quality to the standardsset by Congress requires an increasingly detailed baseof scientific knowledge and information systems.

The Federal Wildland Fire Policy (U.S. Department ofthe Interior and U.S. Department of Agriculture 1995)and the Clean Air Act as Amended 1990 (PL 101-549)resulted in the need to significantly raise the level ofknowledge about fire’s effects on air in order to meetregulatory and management requirements. For example,

new information is needed to assess, monitor, predict,and manage:

• Emissions and air quality impacts from wild-fires

• Acute health effects of human exposure tosmoke

• Natural and anthropogenic sources of visibil-ity reduction

• Cumulative air quality impacts from expandedfuel management programs

• Tradeoffs between air quality impacts fromwildland fire and prescribed fire

Likewise, management of fire and air quality is alsoundergoing substantial policy development that hasled to the need for new and different information tosatisfy regulatory and management requirements. Asboth legal and management issues mature, there isless a sense that environmental regulation is a limita-tion on fire management, and more of a sense thatecosystem management goals, fire safety, and airquality are goals to be met collectively. For example,new air quality rules recognize the importance of therole of fire in sustaining ecosystems and the inherenttradeoffs between prescribed fire and wildland fireoccurrence. At the same time, land management plansand real-time fire management decisions increasinglyfactor in the expected consequences to air quality.

2 USDA Forest Service Gen. Tech. Rep. RMRS-GTR-42-vol. 5. 2002

Since 1995, researchers and land managers haveconcentrated a great deal of energy to extend what isknown about fire and its effect on air quality; to expandinformation systems that make knowledge readilyavailable to policy, management, and public clients; tomerge what is known about sustainable ecosystemsand disturbance ecology with what is known about thechemistry, physics, biology, and social impacts of airpollution; and to redefine the research agenda.

Objective ______________________This review summarizes the current state of knowl-

edge of the effects of fire on air, and defines researchquestions of high priority for the management ofsmoke from fires. We also intend this as a referencedocument for future discussion of management, policy,and science options for managing fires and air quality.This review is limited to readily available publishedand unpublished knowledge and to original contribu-tions by the authors. No new analysis of data or policy,nor assessment of impacts and options, is includedherein.

Related Publications _____________This document does not stand alone. There are

several excellent sources for information on the effectsof fire on air. We advise the reader to include at leastthe following publications, each of which will be ab-stracted in this document, in your reference library:

• Smoke Management Guide for Prescribed andWildland Fire: 2001 Edition (Hardy and oth-ers 2001)

• National Strategic Plan: Modeling and DataSystems for Wildland Fire and Air Quality(Sandberg and others 1999)

• Introduction to Visibility (Malm 2000)• Fire Effects on Air (Sandberg and others 1979)• Southern Forestry Smoke Management Guide-

book (Southern Forest Fire Laboratory Per-sonnel 1976)

• Development of Emissions Inventory Methodsfor Wildland Fire (Battye and Battye 2002)

Why, then, is another state-of-knowledge reviewnecessary on the subject of fire effects on air? First,because policy and regulatory development in airquality management and in fire management is ad-vancing rapidly, and there is a continuing need toreassess current knowledge about what is required tomeet new expectations. Second, this document ad-dresses the advancement of science at a much higherlevel than the above-mentioned references. Third,because the Joint Fire Science Program has sponsoreda series of reviews, nicknamed the Rainbow Series (see

“Preface”), to compile a broad reference of fire effectsto serve practitioners and policymakers charged withusing and managing fire, and this is the third volumein that series. Finally, we hope you will find thisvolume a useful attempt to abstract and fill in the gapsleft by the previous publications.

Scope _________________________This review includes all health and welfare effects of

air pollution from fires, but does not include the effectsof air resource management on ecosystem health orany other value. Unless otherwise specifically stated,the term “fires” in this manuscript includes all pre-scribed and wildland fires on wildlands. Prescribedfires are ignited intentionally to achieve ecosystemmanagement or fire protection objectives, whereaswildland fires result from unplanned ignitions onwildlands. Wildlands include all the nonagriculturaland nonresidential rural lands of the United States,including the wildland-urban interface, regardless ofownership, sovereignty, or management objective.Management response to wildland fires differs greatlyaccording to economic efficiency, the values at risk(including air quality), and the expected ecologicalconsequences. Wildfires are at one end of the spectrumof wildland fires in that they are unwanted and un-planned, and are managed to minimize cost plus loss.At the other end of the spectrum are wildland firesthat benefit ecosystem values, and are managed tomaximize their benefit. Ideally, each wildland fire isevaluated with respect to expected costs, losses, risks,and benefits in order to provide an appropriate andpreplanned response. Because fires are a significantemitter of air pollutants, many other fire managementactivities such as fire prevention or fuel treatmentmay have an indirect effect on air quality.

Framework _____________________The issues, responsibilities, and tools that address

fire and air quality are varied and complex, sometimesresulting in confusion about the physical scale andtemporal stage of three characteristics: the applica-tion to fire management, the application to air re-source management, and the physical process of airpollution. National Strategic Plan: Modeling and DataSystems for Wildland Fire and Air Quality (Sandbergand others 1999) provides a conceptual framework forvisualizing fire’s effects on air by representing thescope of the problem as a three dimensional array ofair resource management, fire management, and scalecomponents (fig. 1-1). The air resource component isordered in time from emissions source strength, toambient air quality, and to effects. The fire manage-ment component includes planning, operations, and


monitoring. The scale component includes the event,landscape, state or tribal, and regional scales.

We have organized this volume around the air re-source component and expanded it to include a regu-latory perspective (fig. 1-2). Fire in the context of theregulatory environment is the subject of chapters 2and 3. Biomass consumption and emissions are the

subject of chapter 4; transport and dispersion of pollut-ants in the atmosphere the subjects of chapters 5 and6; air quality impacts the subject of chapter 7; and theeffect on human values from exposure to air pollutantsthe subject of chapter 8. We conclude with a review ofrecommendations for future research in chapter 9.

Prior Work _____________________Since the publication of Effects of Fire on Air

(Sandberg and others 1979), significant changes havecome to pass in both the technical and policy issuesthat surround the fire and air quality dilemma. Theconferences, stakeholder group discussions, and tech-nical publications discussed here have helped to shapethe current fire management programs and will influ-ence future programs.

Smoke Management Guide For Prescribedand Wildland Fire: 2001 Edition

Smoke Management Guide for Prescribed and Wild-land Fire: 2001 Edition (Hardy and others 2001) hasbeen developed by the Fire Use Working Team of theNational Wildfire Coordinating Group (NWCG) andinvolves most of the same authors as this currentpublication. The guide provides fire managementand smoke management practitioners with a funda-mental understanding of fire emissions processesand impacts, regulatory objectives, and tools for themanagement of smoke from fires. It is a comprehen-

Figure 1-1—Three primary components of the issues, respon-sibilities, and tools related to wildland fire and air quality: airresource management, fire management, and scale (Sandbergand others 1999).

Figure 1-2—The relations of air regulations and physical processes to the three categories within the airresource component. OSHA/NIOSH = Occupational Safety and Health Administration/National Institute forOccupational Safety and Health (Sandberg and others 1999).


sive treatment of the state of knowledge regardingfire and air quality and provides guidance to practi-tioners. We will not attempt to duplicate its level ofdetail in this volume. Rather, we add some technicalbackground and analysis of research needs relativeto new requirements for management.

First published in 1985, the guide is intended toprovide national guidance for the planning and man-aging of smoke from prescribed fires to achieve airquality requirements through better smoke manage-ment practices (NWCG 1985). This guide has beenwidely distributed within the fire community and airquality regulatory agencies, and to private and Triballand managers, providing a single comprehensivesource of information on fire and air quality issues.

Much has changed since 1985 in prescribed burningpractices, smoke management programs, and air qual-ity regulatory requirements. These changes are re-flected in the 2001 edition of the guide, which includesexpanded sections on fire and emissions processes,smoke impacts on health, welfare, safety, and nui-sance; regulations for smoke management; and thefundamentals of responsible smoke management(Hardy and others 2001). These fundamentals includefire planning, use of smoke management meteorology,techniques to reduce emissions, smoke dispersion pre-diction systems, air quality monitoring methods, andprogram assessment.

The most significant change in the guide is theexpanded and updated section on techniques to reduceemissions and impacts. While the 1985 guide focusedprimarily on minimizing smoke impacts by meteoro-logical scheduling and dispersion, the 2001 guideprovides detailed information on emissions reductiontechniques, used in different regions of the country,that have been useful, practicable, and effective in thefield. This emphasis on actual reduction of emissionsrather than dispersion was provided in response to airquality regulations that now target regional emissionsreductions.

Readers will also find that the 2001 guide has a greatdeal more information on the latest developments innational air quality regulations that affect fire pro-grams including the regional haze and visibility pro-tection programs, Clean Air Act’s conformity require-ments, EPA’s Interim Air Quality Policy on Wildlandand Prescribed Fires (EPA 1998), and NEPA planningguidance. The guide was drafted by 16 authors andfive editor/compilers working under the sponsorshipof the NWCG Fire Use Working Team with supportfrom the EPA.

Wildland Fire and Air Quality: NationalStrategic Plan

Another recent publication also provides a system-atic review of the state of knowledge and information

systems. This strategic plan was also sponsored bythe NWCG and the Environmental Protection Agency(EPA).

In 1997, the NWCG Fire Use Working Team sanc-tioned a small group of fire research scientists and airquality managers to develop a National StrategicPlan: Modeling and Data Systems for Wildland Fireand Air Quality (Sandberg and others 1999) to fosterdevelopment and implementation of models and datasystems that could be used to manage air qualityimpacts of fires. The resulting report provides a con-ceptual design and strategic direction toward meetingthe increasing need for information required to man-age emissions from fire (Sandberg and others 1999). InNovember 1997, after 2 years of drafting and exten-sive review of a draft plan, 86 experts attended anational workshop, and using the discussion frame-work presented in this chapter, they defined the cur-rent state of knowledge, desired future condition, andrecommendations for research and development foreach cell in the discussion framework.

The strategic plan targets a more technical, scien-tific, and policy-oriented audience than the smokemanagement guide, and recommends a research anddevelopment strategy to reach a desired future statefor smoke management information systems. It alsoprovides a comprehensive treatment of policy andtechnical issues that we will not duplicate in thisvolume.

Introduction to Visibility

Air pollution impacts on visibility are discussed indetail in Introduction to Visibility (Malm 2000). Thediscussion is not specific to the impacts of fire but isrelevant because of the regulatory attention given tofire in the EPA Regional Haze Rule (40 CFR Part 511999) and because Federal land managers have theresponsibility of managing fires and the impacts offires and all other pollution sources on visibility inmany National Parks and wilderness areas. We makeno attempt in this volume to duplicate this discussionof the atmospheric physics, meteorology, historic vis-ibility trends, monitoring and apportionment method-ologies, or human perceptions that are so admirablycovered in Introduction to Visibility.

The Federal Advisory Committee ActWhite Papers

During the 1997 to 1998 development of proposednational ambient air quality standards (NAAQS) forPM2.5 (particulate matter with an aerodynamic diam-eter less than or equal to 2.5 microns) and regional hazeregulations, EPA used provisions of the Federal Advi-sory Committee Act (FACA) to convene a large group ofstakeholders who were interested in providing input to


the regulatory process. A FACA committee for thedevelopment of ozone, particulate matter, and regionalhaze implementation programs was formed to addressboth policy and technical issues. The committee’s Sci-ence and Technology Wildland Fire Issues Group, oneof several working groups and subcommittees, re-searched and drafted five reports that are briefly sum-marized below (EPA 2000c).

Air Monitoring for Wildland Fire Operations pro-vides recommendations for conducting air-monitoringprograms designed to support fire activities that alsomonitor for compliance with NAAQS. It also describeshow monitoring can support burning programs andhow land managers can collaborate with air agencies,and it provides guidance for selecting monitoringequipment.

Elements of a Smoke Management Program dis-cusses recommendations for a basic level smoke man-agement program. The document summarized infor-mation from an EPA-sponsored workshop held torespond to specific questions posed by EPA. The docu-ment describes the six basic components of a smokemanagement program:

• Authorization to burn• Minimizing emissions• Burn plan components• Public education• Surveillance and enforcement• Program evaluation

It also provides examples of monitoring methods,public awareness programs, and program enforcement.

Emission Inventories for State Implementation Plan[SIP] Development describes several levels of inven-tory complexity: a default level based on currentlyavailable information; a basic level program that isconsidered the minimal program needed to supportSIP development; and a detailed inventory level whena greater level of analysis or accountability in inven-tory precision is needed. Elements of each level ofinventory are described, data sources are identifiedand data management issues are discussed.

What Wildland Fire Conditions Minimize Emis-sions and Hazardous Air Pollutants and Can LandManagement Goals Still be Met? This paper is a dis-cussion of fire conditions and techniques that mini-mize pollutant emissions. Both wildland emissionsand prescribed fire emissions are discussed. The dis-cussion of emissions reduction techniques for pre-scribed burning is also found in Smoke ManagementGuide for Prescribed and Wildland Fire: 2001 edition(Hardy and others 2001).

Estimating Natural Emissions from Wildland andPrescribed Fire addresses how best to define “naturalemissions” from fire. This is critical to implementing

regional haze goals of reducing visibility degradationcaused by human-made sources of air pollution. Thepaper discusses a matrix of choices: (1) emissions fromfire necessary to restore and sustain desired ecosys-tem characteristics, (2) fire needed to manage fuels toa condition where they can be dealt with most effec-tively from a wildfire control standpoint, (3) no netincrease in fire emissions, and (4) no change fromcurrent emissions.

Stakeholders reviewed, discussed, and drafted, ad-ditional work on these five reports. The reports andother technical references were considered by EPAduring the formulation of the regional haze regula-tions and revisions to the particulate matter NAAQS.

Environmental Regulation and PrescribedFire Conference

In March 1995 a conference on new developments inenvironmental regulations related to prescribed firewas held in Tampa, FL (Conference Proceedings: Envi-ronmental Regulation & Prescribed Fire: Legal andSocial Challenges, Bryan 1997). This 3-day meetingincluded sessions on challenges and strategies regard-ing the use of fire, air quality regulation, and liability,as well as social and economic issues. Sponsored bynumerous State and Federal environmental and for-estry agencies, the conference provided a forum fordiscussion of the Clean Air Act, Endangered SpeciesAct, and other Federal statutes that guide national,State, and local regulations pertaining to prescribedfire.

Significantly, a joint declaration drafted by the con-ference steering committee and presented to confer-ence attendees was later signed by representatives ofthe EPA, State of Florida, National Biological Survey,The Wilderness Society, Forest Service, and MariposaCounty, Florida. In summary, the declaration upheldthe following principles:

• Practitioner liability is a major obstacle to theincreased use of fire. Legislation should beconsidered on the Federal level to protectproperly certified fire practitioners except incases where negligence is proven.

• Partnerships among all of the stakeholdersare vital to the future use of fire. Efforts toenhance such partnerships must be encour-aged especially in the exchange of informa-tion, development of best management prac-tices, public education campaigns, and fundinginitiatives.

• Agencies should work together to evaluatetradeoffs between public health risks fromfire and ecological damage caused by fireexclusion.


• Public education regarding the use of pre-scribed fire, ecosystem health, and risks ofwildfire versus those from prescribed burningis encouraged.

• The role of fire in ecosystem managementneeds to be understood by all stakeholders.The ramifications of not using prescribed fireare serious and must also be appreciated aslimits on fire use may conflict with otherpublic mandates.

• Actions pertaining to the use of fire must bebased on sound science. There are severalcrucial knowledge gaps that must be filled.Consequences to public safety caused by de-laying the increases of prescribed fire aregreat.

• Public and private property owners need toretain the right to use prescribed fire to pro-tect and enhance the productivity of theirlands while also protecting nearby propertyowners from adverse impacts of burning.

• Administrators responsible for allocating fundsshould do so on the basis of regional prioritieswith greater emphasis on prevention than inthe past.

• An increased emphasis on training for pre-scribed fire practitioners is needed to enhancepublic acceptance.

Southern Forestry Smoke ManagementGuidebook

The Southern Forestry Smoke Management Guide-book (Southern Forest Fire Laboratory Personnel 1976)was one of the first smoke management guidebooksdeveloped in the United States for use by land, fire,and air resources managers. The guide provides animproved understanding of: (1) smoke managementand air quality regulations; (2) contents of smoke andvariables affecting production; (3) smoke transportand dispersion; (4) potential effects on human health,human welfare, and visibility; and (5) what can bedone to mitigate its impacts. A system for predictingand modifying smoke concentrations from prescribedfires was introduced for Southern fuels.

Changes in Fire Policy ___________The Federal Wildland Fire Policy (USDI and USDA

1995; USDI and others 2001) requires that “… fire, asa critical natural process, must be reintroduced intothe ecosystem to restore and maintain sustainableecosystems. This will be accomplished across agencyboundaries and will be based on the best availablescience.” The policy requires “the use of fire to sustainecosystem health based on sound scientific principles

and balanced with other social goals including publichealth and safety, air quality, and other specific envi-ronmental concerns.” Early in the planning process,action is required to “involve public health and envi-ronmental regulators in developing the most workableapplication of policies and regulations.” Agencies arecalled on to “create a system for coordination andcooperation among land managers and regulatorsthat explores options within existing laws to allow forthe use of fire to achieve goals of ecosystem healthwhile protecting individual components of the envi-ronment, human health, and safety.” The policy alsorequires that air quality values be considered duringpreparedness and fire protection. When setting pro-tection priorities, land managers must “define valuesto be protected working in cooperation with state,local, and tribal governments, permittees, and publicusers. Criteria will include environmental, commod-ity, social, economic, political, public health, and othervalues.”

Several strategies and funding programs were de-veloped to improve the ability of managers to fullyimplement this policy.

Joint Fire Science Program

The Joint Fire Science Program (JFSP) was createdby Congress in the 1998 Appropriations to Interiorand Related Agencies bill to augment the delivery ofscience and information systems necessary to managethe increased use of fire and other fuel treatments. Thelegislation provides a mandate to protect air quality inconjunction with economic efficiency and ecologicalconsequences. The program (National InteragencyFire Center 2002 unpaginated) recognizes that:

Land managers are rapidly expanding the use of firefor managing ecosystems while air resource managersare accelerating efforts to reduce the local and regionalimpacts of smoke. Smoke management (meeting airquality standards) is a legal requirement of the CleanAir Act, as well as a health and safety issue for thegeneral populace and fireline personnel. The JFSP willattempt to define these social relationships and de-velop analytical tools and communication practices tohelp mangers include social considerations in decisionmaking.

One of the goals of the JFSP is “to evaluate varioustreatment techniques for cost effectiveness, ecologicalconsequences, and air quality impacts.” The programplan states:

Methods have not been developed to assess the oppor-tunities, costs, and effectiveness of employing smokereduction techniques throughout the country. Currentmodels to assess regional scale cumulative effects onair quality and water quality will need to be expanded.The program will develop a nationally consistent sys-tem of models for fuel consumption, emissions produc-tion, and smoke dispersal that can assess cumulative


effect. This research would also contribute to under-standing the potential national and global impacts ofchanges in biomass use, prescribed fire, and wildlandfire on wood supply, atmospheric chemistry, and car-bon sequestration.

Cohesive Strategy

Protecting People and Sustaining Resources in Fire-Adapted Ecosystems: A Cohesive Strategy (Lavertyand Williams 2000) is the Federal framework estab-lished to restore and maintain ecosystem health toreduce the threat and consequences of wildfires. It ispresumed that fire suppression over the past 100years has excluded fire from many ecosystems, fuelingconditions for unnaturally intense fires that, amongother effects, threaten air quality. Citing serious airquality impacts from long duration wildfire episodesin recent years, the report expresses concern that:

The extent to which management for ecosystem resil-ience can improve air quality over the long term is notcompletely known. Present regulatory policies measureprescribed fire emissions, but not wildland fire emis-sions. The emissions policy tends to constrain treat-ments and – in short interval fire systems — may act toinadvertently compound wildland fire risks. (p 34)

The cohesive strategy directs land managementagencies to collaborate with the EPA in addressinglong-term impacts, tradeoffs, and issues regarding airquality and other impacts. The report acknowledgesthat programmatic analysis of air quality impacts willbe a necessary step in implementing the plannedincreases in prescribed burning necessary to restorethe health of fire-prone ecosystems. The strategy esti-mates that the USDA Forest Service Regions wouldincrease fuel treatments by five-fold in the West andtwo-fold in the East and South to achieve restorationgoals within 10 years; or employ a slightly smallerincrease to obtain results in 20 years. Most, but not all,of the treatments would involve burning.

The relative risk to air quality was projected todecrease by about 25 percent as a result of improvingthe resilience of ecosystems, according to currentmodels.

The cohesive strategy is responsive to regulatoryresponsibilities. The planned increase in burning isconstrained in part by the consideration to regulatoryobligations, with an acknowledgment that a morerigorous assessment of impacts could substantiallychange the planned extent and schedule of treat-ments. Concerns for public health issues and firefightersafety in relation to smoke are also expressed. Thestrategy acknowledges that air quality issues must beanalyzed more thoroughly at smaller scales as it isstepped down to landscape and project level planning.

National Fire Plan

The National Fire Plan was established in A Reportto the President In Response to the Wildfires of 2000(USDA and USDI 2000), and implemented using Col-laborative Approach for Reducing Wildfire Risks toCommunities and the Environment: 10-Year Compre-hensive Strategy (Western Governors’ Association 2001).Stakeholder groups under the sponsorship of the USDAForest Service, USDI, and the Western Governors’Association prepared the implementation strategy.This strategy recognizes that key decisions in settingpriorities for restoration, fire, and fuel managementshould be made at local levels. As such, the planrequires an ongoing process whereby the local, Tribal,State and Federal land management, scientific, andregulatory agencies exchange the required technicalinformation, including the assessment of air qualitytradeoffs, to inform this decisionmaking process. Thestrategy has a goal of maintaining and enhancingcommunity health and economic and social well-be-ing; and requires that public health risks from smokeare reduced, airshed visibility is improved, and smokemanagement plans are developed in conjunction withprescribed fire planning and implementation.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


This chapter introduces the regulatory environmentfor smoke from prescribed and wildland fire, providingupdated discussion of the laws, regulations, stan-dards, and regulatory strategies that have changedsince about 1980. We explain roles and responsibili-ties of the regulatory agencies and land managers, andwe frame the technical discussion in the context of whoneeds what information to fulfill legal requirements.

Air pollution is the presence in the atmosphere ofone or more contaminants of a nature, concentration,and duration to be hazardous to human health orwelfare (Sandberg and others 1999). Welfare includespotential to harm animal or ecosystem health, eco-nomic activity, or the comfortable enjoyment of lifeand property. Air pollution is created from both hu-man (that is, anthropogenic) and natural sources.Anthropogenic air pollution is the presence in theatmosphere of a substance or substances added di-rectly or indirectly by a human act, in such amounts asto adversely affect humans, animals, vegetation, ormaterials (Williamson 1973). Air pollutants are clas-sified into two major categories: primary and second-ary. Air pollutant emissions, or simply “emissions,”are the production and release of air contaminantsemitted from fires that have a potential to cause airpollution. This definition includes particulates, hydro-carbons, carbon monoxide (CO), metals, and all other

Chapter 2: Air QualityRegulations and Fire

trace gases that may be hazardous or that are chemi-cal precursors to secondary air pollution. Primarypollutants are those directly emitted into the air.Under certain conditions, primary pollutants undergochemical reactions within the atmosphere and pro-duce new substances known as secondary pollutants.Hazardous air pollutants are a special class of airpollutants identified in the Clean Air Act Amend-ments of 1990 as constituting a hazard to humanhealth.

Air quality is a measure of the presence of airpollution. Ambient air quality is defined by the CleanAir Act of 1963 as the air quality anywhere people haveaccess, outside of industrial site boundaries. Ambientair quality standards are standards of air qualitydesigned to protect human health or welfare. Airresource management includes any activity to antici-pate, regulate, or monitor air pollution, air pollutantemissions, ambient air quality, or the effects of airpollution resulting from fires or fire management.

In the past, emissions from prescribed fire wereconsidered human-caused, and wildland fires wereconsidered natural sources of emissions. But recentpolicy debate has focused on what should be consid-ered natural; that is, to be reasonably unaffected byhuman influence. This debate resulted from theparadox that not all wildland fires are vigorously


suppressed and that some prescribed burning isdone to maintain healthy natural ecosystems wherefire has previously been excluded.

Air resource management includes any activity toanticipate, regulate, or monitor air pollution, air pol-lutant emissions, ambient air quality, or the effects ofair pollution resulting from fires or fire management.

Emissions and impacts on air quality from fires aremanaged and regulated through a complex web ofinterrelated laws and regulations. The primary legalbasis for air quality regulation across the nation is theFederal Clean Air Act (CAA), which is actually a seriesof acts, amendments, and regulations that include:

• Federal Air Pollution Control Act of 1955 (PL84-159). Provides for research and technicalassistance and authorizes the Secretary ofHealth, Education, and Welfare to work to-ward a better understanding of the causes andeffects of air pollution.

• Federal Clean Air Act of 1963 (PL 88-206).Empowers the Secretary of Health, Educa-tion, and Welfare to define air quality criteriabased on scientific studies. Provides grants tostate and local air pollution control agencies.

• Federal Air Quality Act of 1967 (PL 90-148).Establishes a framework for defining “air qual-ity control regions” based on meteorologicaland topographical factors of air pollution.

• Federal Clean Air Act Amendments of 1970(PL 91-604). Principal source of statutory au-thority for controlling air pollution. Estab-lishes basic U.S. program for controlling airpollution.

• Environmental Protection Agency (EPA) pro-mulgates national ambient air quality stan-dards (NAAQS) for particulates, photochemi-cal oxidants (including ozone), hydrocarbons,carbon monoxide, nitrogen dioxide, and sulfurdioxide (1971).

• Clean Air Act Amendments of 1977 (PL 95-95). Sets the goal for visibility protectionand improvement in Class I areas and as-signs Federal land managers the affirma-tive responsibility to protect air quality re-lated values.

• Clean Air Act Amendments of 1990 (PL 101-549). Establishes authority for regulating re-gional haze and acknowledges the complexityof the relation between prescribed and wild-land fires.

• Regional Haze Regulations, Final Rule (40CFR Part 51) (1999). EPA promulgates theRegional Haze Rule supported in part by the1998 Interim Air Quality Policy on Wildlandand Prescribed Fires.

Roles and Responsibilities Underthe Clean Air Act ________________

States have the lead in carrying out provisions of theClean Air Act because appropriate and effective de-sign of pollution control programs requires an under-standing of local industries, geography, transporta-tion, meteorology, urban and industrial developmentpatterns, and priorities. The EPA has the task ofsetting air quality standards (national ambient airquality standards, or NAAQS). In addition, EPA de-velops policy and technical guidance describing howvarious Clean Air Act programs should function andwhat they should accomplish. States develop Stateimplementation plans (SIPs) that define and describecustomized programs they will implement to meetrequirements of the Clean Air Act. Tribal lands arelegally equivalent to State lands, and Tribes prepareTribal implementation plans (TIPs) to describe howthey will implement the Clean Air Act. IndividualStates and Tribes can require more stringent airquality standards but cannot weaken clean air goalsset by EPA.

Federal land managers have the complex role ofmanaging a fire as a source of air pollutants, whilefulfilling monitoring and regulatory responsibilitiestied to visibility and regional haze. Federal landmanagers are given the responsibility by the CleanAir Act for reviewing prevention of significant dete-rioration (PSD) permits (discussed later in this chap-ter) of major new and modified stationary pollutionsources and commenting to the State on whetherthere is concern for visibility impacts (or other re-source values) in Class I areas downwind of theproposed pollution source. Some States require mod-eling of source impacts on Class I areas, and Federalland managers customarily comment on the modelresults.

The 1990 Clean Air Act Amendments requireplanned Federal actions to conform to SIPs. This“general conformity rule” prohibits Federal agenciesfrom taking any action within a nonattainment ormaintenance area that (1) causes or contributes to anew violation of air quality standards, (2) increasesthe frequency or severity of an existing violation, or(3) delays the timely attainment of a standard asdefined in the applicable SIP or area plan. The gen-eral conformity rule covers direct and indirect emis-sions of criteria pollutants, or their precursors, whichare caused by a Federal action, are reasonablyforeseeable, and can practicably be controlled bythe Federal agency through its continuing programresponsibility.


National Ambient Air QualityStandards______________________

The purpose of the Clean Air Act is to protecthumans against negative health or welfare effectsfrom air pollution. National ambient air quality stan-dards (NAAQS) are defined in the Clean Air Act asamounts of pollutant above which detrimental effectsto public health or welfare may result. NAAQS havebeen established for the following criteria pollutants:particulate matter (PM10 and PM2.5; NAAQS forparticulate matter are established for two aerody-namic diameter classes: PM10 is particulate matterless than 10 microns in diameter, and PM2.5 is lessthan 2.5 microns in diameter; total suspended particu-late matter is called PM or sometimes TSP), sulfurdioxide (SO2), nitrogen dioxide (NO2), ozone, carbonmonoxide (CO) and lead (Pb) (table 2-1). PrimaryNAAQS are set at levels to protect human health;secondary NAAQS are to protect human welfare ef-fects including visibility as well as plant and materialsdamage.

An area that is found to be in violation of a primaryNAAQS is labeled a nonattainment area (fig. 2-1); anarea once in nonattainment but recently meetingNAAQS, and with appropriate planning documentsapproved by EPA, is a maintenance area; all otherareas are attainment or unclassified (due to lack ofmonitoring). State air quality agencies can provideup-to-date locations of local nonattainment areas(PM2.5 is a newly regulated pollutant, so attainment/nonattainment status had not been determined at thetime of publication of this document; monitoring must

take place for at least 3 years before designation can bemade, which means PM2.5 status will likely not beknown until at least 2003). States are required throughtheir SIPs to define programs for implementation,maintenance, and enforcement of the NAAQS withintheir boundaries. Wildland fire in and nearnonattainment areas will be scrutinized to a greaterdegree than in attainment areas and may be subject togeneral conformity rules. Extra planning, documenta-tion, and careful scheduling of prescribed fires willlikely be required to minimize smoke effects in thenonattainment area to the greatest extent possible. Insome cases, the use of fire may not be possible ifsignificant impacts to a nonattainment area are likely.

The major pollutant of concern in smoke from fire isfine particulate matter, both PM10 and PM2.5. Stud-ies indicate that 90 percent of all smoke particlesemitted during wildland burning are PM10, and 90percent of PM10 is PM2.5 (Ward and Hardy 1991). Themost recent human health studies on the effects ofparticulate matter indicate that fine particles, espe-cially PM2.5, are largely responsible for health effectsincluding mortality, exacerbation of chronic disease,and increased hospital admissions (Dockery and oth-ers 1993; Schwartz and others 1996).

Prevention of SignificantDeterioration ___________________

Another provision of the Clean Air Act with someapplicability to wildland burning activities is the pre-vention of significant deterioration (PSD) provisions.

Table 2-1—National ambient air quality standards (NAAQS) (U.S. Environmental ProtectionAgency 2000b). Primary NAAQS are set at levels to protect human health;secondary NAAQS are to protect human welfare.

Pollutant Averaging time Primary Secondary

PM10 Annual arithmetic mean 50 µg/m3 a 50 µg/m3

24-hour average 150 µg/m3 150 µg/m3

PM2.5 Annual arithmetic mean 15 µg/m3 15 µg/m3

24-hour average 65 µg/m3 65 µg/m3

Sulfur dioxide (SO2) Annual average 0.03 ppmb —24-hour average 0.14 ppm —3-hour average — 0.50 ppm

Carbon monoxide (CO) 8-hour average 9 ppm —1-hour average 35 ppm —

Ozone (O3) 8-hour average 0.12 ppm 0.12 ppm1-hour average 0.08 ppm 0.08 ppm

Nitrogen dioxide (NO2) Annual average 0.053 ppm 0.053 ppmLead (Pb) Quarterly average 1.5 µg/m3 1.5 µg/m3

aµg/m3 = micrograms per cubic meter.bppm = parts per million.


The goal of PSD is to prevent areas that are currentlycleaner than is allowed by the NAAQS from beingpolluted up to the maximum ceiling established by theNAAQS. Three air quality classes were established bythe Clean Air Act PSD provisions including Class I(which allows very little additional pollution), Class II(which allows some incremental increase in pollution),and Class III (which allows pollution to increase up tothe NAAQS). Class I areas include wildernesses andnational memorial parks over 5,000 acres, NationalParks exceeding 6,000 acres, and all internationalparks that were in existence on August 7, 1977, as wellas later expansions to these areas (fig. 2-2).

Historically, EPA has regarded smoke from wild-land fires as temporary and therefore not subject toissuance of a PSD permit; whether or not wildlandfire smoke should be considered when calculatingPSD increment consumption or PSD baseline was not

defined. EPA recently reaffirmed that States couldexclude prescribed fire emissions from incrementanalyses provided the exclusion does not result inpermanent or long-term air quality deterioration(EPA 1998). States are also expected to consider theextent to which a particular type of burning activityis truly temporary, as opposed to an activity thatcould be expected to occur in a particular area withsome regularity over a long period. Oregon is the onlyState that has chosen to include prescribed fire emis-sions in PSD increment and baseline calculations.

Visibility _______________________The 1977 amendments to the Clean Air Act include

a national goal of “the prevention of any future, andthe remedying of any existing, impairment of visibility

Figure 2-1—PM10 nonattainment areas as of May 2002. Current nonattainment status for PM10 and all other criteriapollutants are available from the Environmental Protection Agency (EPA) aerometric information retrieval system(AIRS) Web page at http://www.epa.gov/air/data/index.html (EPA 2002).


in mandatory Class I Federal areas which impairmentresults from manmade air pollution” (42 U.S.C §7491). States are required to develop implementationplans that make “reasonable progress” toward thenational visibility goal.

Atmospheric visibility is affected by scattering andabsorption of light by particles and gases. Particlesand gases in the air can obscure the clarity, color,texture, and form of what we see. Fine particles mostresponsible for visibility impairment are sulfates, ni-trates, organic compounds, elemental carbon (or soot),and soil dust. Sulfates, nitrates, organic carbon, andsoil tend to scatter light, whereas elemental carbontends to absorb light. Fine particles (PM2.5) are moreefficient per unit mass than coarse particles (PM10and larger) at causing visibility impairment. Natu-rally occurring visual range in the Eastern UnitedStates is estimated to be between 60 and 80 miles,while natural visual range in the Western UnitedStates is between 110 and 115 miles (these estimatesdo not consider the effect of natural fire on visibility)

(Trijonis and others 1991). Currently, visual range inthe Eastern United States is about 15 to 30 miles andabout 60 to 90 miles in the Western United States. (40CFR Part 51). The theoretical maximum visual rangeabout 240 miles.

Regional Haze

Regional haze is visibility impairment produced bya multitude of sources and activities that emit fineparticles and their precursors and are located across abroad geographic area. This contrasts with visibilityimpairment that can be traced largely to a single, largepollution source. Until recently, the only regulationsfor visibility protection addressed impairment that isreasonably attributable to a permanent, large emis-sions source or small group of large sources. In 1999,EPA issued regional haze regulations to manage andmitigate visibility impairment from the multitude ofdiverse regional haze sources (40 CFR Part 51). Theregional haze regulations call for States to establish

Figure 2-2—Mandatory class 1 areas (Hardy and others 2001).


goals for improving visibility in Class I National Parksand wildernesses, and to develop long-term strategiesfor reducing emissions of air pollutants that causevisibility impairment.

Regional Haze Planning Process—Because re-gional haze is a multi-State issue, regional haze regu-lations encourage States, land managers, and otherstakeholders to work together to develop control pro-grams through regional planning organizations thatcan coordinate development of strategies across amulti-State region. In the Western United States, theWestern Regional Air Partnership (WRAP), sponsoredthrough the Western Governors’ Association and theNational Tribal Environmental Council, is coordinat-ing regional planning and technical assessments. TheWRAP was the first of five regional planning organiza-tions to be established and has been active in manytechnical and policy developments. Other regionalplanning organizations have begun assessments offire and air quality in their regions. In the Eastern

United States, four formal groups are addressingplanning issues: CENRAP (Central States RegionalAir Partnership), OTC (Ozone Transport Commis-sion), VISTAS (Visibility Improvement State and TribalAssociation of the Southeast); and the Midwest Re-gional Planning Organization (fig. 2-3).

As inter-State smoke transport becomes a largerissue, agencies are expanding coordination of theirburns. Multi-State, interagency partnerships are de-veloping to help coordinate burning and mitigate cu-mulative impacts of smoke. For example, the Mon-tana/Idaho airshed group includes private, State,Tribal, and Federal partners in supporting an inte-grated smoke management program that includesemissions monitoring and smoke forecasting (Levinson2001).

Regional Haze and Fire Emissions—The adop-tion of regional haze regulations marks a turningpoint in how fire emissions are treated under thenation’s Federal and State air quality regulations,

Figure 2-3—Regional air quality planning groups (Hardy and others 2001).


although the regulations leave several definitionsopen to subsequent policy interpretation:

• The role of fire in forest ecosystems is formallyrecognized for the first time.

• Emissions from “natural” sources are distin-guished from “anthropogenic” sources andtreated differently under the rule.

• The rule is the first to require development ofemissions inventories for fire, including wild-land fires.

• Emissions from fire are now subject to re-gional air quality planning processes as wellas requirements to achieve “reasonableprogress” in emissions reductions

The policy discussion to determine what types of fireemissions are considered natural is still in progress,but the WRAP has recommended a national policythat would (1) define “natural background” as fireemissions that would occur in the future without firemanagement; that is, without reference to historic fireoccurrence or historic vegetation types; and (2) includeprescribed burning as natural sources of visibilityimpacts when fire is used to maintain healthy andsustainable ecosystems.

Current data from a national visibility-monitoringnetwork (Sisler and others 1996) do not show fire to bethe predominant long-term source of visibility impair-ment in any Class I area (40 CFR Part 51), althoughemissions from fire are an important episodic con-tributor to visibility-impairing aerosols. Certainly thecontribution to visibility impairment from fires can besignificant over short periods, but fires in generaloccur relatively infrequently and thus have a lessercontribution to long-term averages. Specific goals forvisibility improvement focus efforts on improving airquality on the most impaired days, so fires may proveto be an important target for control efforts in someareas

Fire Consortia for Advanced Modeling of Me-teorology and Smoke (FCAMMS)—Multiagencyconsortia are building in the Pacific Northwest, RockyMountain region, and Northeastern and Southeast-ern United States as part of the U.S. Department ofAgriculture, Forest Service, Fire Consortia for Ad-vanced Modeling of Meteorology and Smoke. The Pa-cific Northwest consortium is developing a real-timesmoke prediction and emission tracking system thataddresses needs of several smoke management plansfrom collaborating States, Tribes, and local air agen-cies (Ferguson and others 2001). California andNevada are working together through the Californiaand Nevada Smoke and Air Committee (CANSAC)with similar objectives of tracking and predictingcumulative smoke impacts (Chris Fontana, personalcommunication).

Each group or regional consortium must respond tolocal, State, and Tribal smoke management programs.In addition, each region of the country has its ownparticular atmospheric processes that impact fire be-havior and smoke dispersion in different ways. Forexample, while in the Southeast, timing of frontalpassages and onshore flow regimes become critical, inthe Western United States, complex flow throughmountainous terrain is an important consideration inmanaging smoke. These regionally specific demandsare forcing research to focus on subtle aspects of smokeemissions and dispersion instead of traditional devel-opment of worst-case air pollution scenarios.

Reasonable Progress

Visibility rules require States to make “reasonableprogress” toward the Clean Air Act goal of “preventionof any future, and the remedying of any existing,impairment of visibility.” The regional haze regula-tions did not define visibility targets but instead gaveStates flexibility in determining reasonable progressgoals for Class I areas. States are required to conductanalyses to ensure that they consider the possibility ofsetting an ambitious reasonable progress goal, onethat is aimed at reaching natural background condi-tions in 60 years. The rule requires States to establishgoals for each affected Class I area to (1) improvevisibility on the haziest 20 percent of days, and (2)ensure no degradation occurs on the clearest 20 per-cent of days over the period of each implementationplan.

States are to analyze and determine the rate ofprogress needed for the implementation period ex-tending to 2018 such that, if maintained, this ratewould attain natural visibility conditions by the year2064. To calculate this rate of progress, each Statemust compare baseline visibility conditions to esti-mate natural visibility conditions in Class I areas andto determine the uniform rate of visibility improve-ment that would need to be maintained during eachimplementation period to attain natural visibility con-ditions by 2064. Baseline visibility conditions will bedetermined from data collected from a national net-work of visibility monitors representing all Class Iareas in the country for the years 2000 to 2004. EachState must determine whether this rate and associ-ated emissions reduction strategies are reasonablebased on several statutory factors. If the State findsthat this rate is not reasonable, it must provide ademonstration supporting an alternative rate.

Hazardous Air Pollutants _________Hazardous air pollutants (HAPs) are identified in

Title III of the Clean Air Act Amendments of 1990


(PL 101-549) as 188 different pollutants “which present,or may present, through inhalation or other routes ofexposure, a threat of adverse human health or envi-ronmental effects whether through ambient concen-trations, bioaccumulation, deposition, or other routes.”The list of HAPs identified in the Clean Air Act aresubstances that are known or suspected to be carcino-genic, mutagenic, teratogenic, neurotoxic, or whichcause reproductive dysfunction.

EPA Interim Air Quality Policy onWildland and Prescribed Fires_____

In 1998, the EPA issued a national policy to addresshow best to achieve national clean air goals whileimproving the quality of wildland ecosystems throughthe increased use of fire. The Interim Air QualityPolicy on Wildland and Prescribed Fires (U.S. Envi-ronmental Protection Agency 1998) was developedthrough a partnership effort involving EPA, the U.S.Departments of Agriculture, Defense, and the Inte-rior, State foresters, State and Tribal air regulators,and others. The group that developed the policy reliedon the assumption that properly managed prescribedfires can improve the health of wildland ecosystemsand reduce the health and safety risks associated withwildfire, while meeting clean air and public healthgoals through careful planning and cooperationamong land managers, air quality regulators, and localcommunities.

Natural Events Policy ____________PM10 NAAQS exceedances caused by natural events

are not counted toward nonattainment designation ifa State can document that the exceedance was trulycaused by a natural event and prepares a naturalevents action plan (NEAP) to address human healthconcerns during future events (Nichols 1996). Naturalevents are defined by this policy as wildfire, volcanic,seismic, and high wind events.

A wildfire NEAP should include commitments bythe State and stakeholders to:

1. Establish public notification and education pro-grams.

2. Minimize public exposure to high concentrationsof PM10 due to future natural events such as by:a. Identifying the people most at risk.b. Notifying the at-risk public that an event is

active or imminent.c. Recommending actions to be taken by the

public to minimize their pollutant exposure.d. Suggesting precautions to take if exposure

cannot be avoided.

3. Abate or minimize controllable sources of PM10including the following:a. Prohibition of other burning during pollution

episodes caused by wildfire.b. Proactive efforts to minimize fuel loadings in

areas vulnerable to fire.c. Planning for prevention of NAAQS exceedances

in fire management plans.

4. Identify, study, and implement practical mitigat-ing measures as necessary.

5. Periodic reevaluation of the NEAP.

Collaboration AmongStakeholders ___________________

Because smoke from fire can negatively affect publichealth and welfare, air quality protection regulationsmust be understood and followed by responsible firemanagers. Likewise, air quality regulators need anunderstanding of how and when fire use decisions aremade and should become involved in fire and smokemanagement planning processes, including the as-sessment of when and how alternatives to fire will beused. Cooperation and collaboration between firemanagers and air quality regulators is of great impor-tance. Table 2-2 contains recommendations for vari-ous types of cooperation by these two groups depend-ing on the applicable air quality protection instrument.

Best Available ControlMeasures ______________________

The application of best available control measures(BACM) for prescribed fire is a required element ofState implementation plans for PM10 nonattainmentareas that are significantly impacted by prescribedfire smoke (EPA 1992a). The application of BACM isalso a requirement of EPA’s Air Quality Policy onWildland and Prescribed Fires (EPA 1998) (see “PriorWork” section in chapter 1). EPA’s BACM guidanceincludes basic smoke management program elementsand emissions reduction techniques that can be usedby land managers to minimize air quality impactsfrom fire. These program elements and emissionsreduction techniques are fully documented in theSmoke Management Guide for Prescribed and Wild-land Fire: 2001 Edition (Hardy and others 2001).

Briefly, the BACM guidance notes that there aretwo basic approaches to minimizing the impact ofprescribed fire on air quality: reducing the amount ofpollutants emitted, or reducing the impact of thepollutants emitted on sensitive locations or regionalhaze through smoke dilution or transport (redistribut-ing emissions). Although each method can be discussed


independently, fire practitioners often choose fire andfuels manipulation techniques that complement or areat least consistent with meteorological scheduling formaximum smoke dispersion and favorable plume trans-port. The following emissions reduction and redistrib-uting emissions techniques are a compilation of ourknowledge base, and depending on specific fire useobjectives, the project locations, time, and cost con-straints may or may not be applicable.

Reducing Emissions

At least 24 methods within six major classificationshave been used to reduce emissions from prescribedburning (Hardy and others 2001). These techniquesinclude methods designed to minimize emissions byreducing the area burned; reducing the fuel load byreducing the fuel production, or fuel consumption, orboth; scheduling burns before new fuels appear; andincreasing combustion efficiency. Each of these meth-ods has specific practices associated with it.

Redistributing Emissions

These measures are commonly practiced in smokemanagement programs and include burning whendispersion is good, cooperating with other burners in

Table 2-2—Recommended cooperation between wildland fire managers and air quality regulators, depending on air qualityprotection instrument (Hardy and others 2001).

Air quality protection instrument Wildland fire managers Air quality regulators

National ambient air quality standards (NAAQS) Awarea Lead b

Attainment status Aware LeadState implementation plan (SIP) planning and development Involvedc LeadConformity Involved LeadSmoke management programs Partnerd LeadVisibility protection Involved LeadRegional planning groups Partner LeadNatural emissions Partner LeadNatural events action plan Partner LeadLand use planning Lead InvolvedProject NEPA documents Lead InvolvedOther fire planning efforts Lead Involved

aAware: Responsibility to have a complete working knowledge of the air quality protection instrument but likely little or no involvement in itsdevelopment or daily implementation.

bLead: Responsibility to initiate, bring together participants, complete, and implement the particular air quality protection instrument.cInvolved: Responsibility to participate in certain components of development and implementation of the air quality protection instrument although

not at full partner status.dPartner: Responsibility to fully participate with lead organization toward development and implementation of the air quality protection instrument

in a nearly equal relationship.

a single airshed to schedule burns, avoiding sensitiveareas, burning smaller units, and burning morefrequently.

Ozone and Fire _________________Ozone is a criteria air pollutant, but there is little

monitoring or research data that directly link fireemissions with ground-level ozone concentrations.Regulating efforts to reduce ozone have thereforefocused on more obvious industrial and urban sourcesof the pollutants that form ozone (NOX and VOCs).Fires are known to emit VOCs and a minor amount ofNOX, but much is uncertain about the magnitude ofozone formation in the plume, the degree of mixingwith urban sources of ozone precursors, and transportof ozone to ground level. EPA plans to begin includingfire emissions in future regional ozone strategy model-ing. Field observations of ozone formation in smokeplumes from fires date back nearly 25 years whenmeasurements from aircraft detected ozone at theedge of forest fire smoke plumes aloft. A recent study(Wotawa and Trainer 2000) did link high ground-levelozone concentrations to forest fire plumes that hadbeen transported great distances. Chapter 6 exploresthese issues more fully.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


Chapter 3: Overview of AirPollution from Fire

This chapter provides a brief overview of and anappreciation for the national, regional, and local im-portance of smoke to ambient air quality. We discussthe significance of fire emissions and air quality im-pacts on a national and regional scale. Chapter 7 ofthis document adds additional depth to this discus-sion.

Magnitude of FireContributions___________________

Air quality impacts associated with wildland firesare distinguished from those resulting from prescribedburning because emissions from these two sourceshave in the past been treated differently under theClean Air Act and by State and local air qualityregulations. In addition, it is important to have ahistorical perspective of these issues given the in-creased use of fire in the recent past.

A comparison by Leenhouts (1998) of estimatedlevels of biomass burning suggests that 10 times morearea burned annually in the pre industrial era than inthe contemporary era. After accounting for land usechanges such as urbanization and agriculture,Leenhouts concluded that about 50 percent of histori-cal levels would burn today if historical fire regimeswere restored to all wildlands to maintain ecosystem

health (figs. 3-1 and 3-2). This suggests a four- to six-fold increase from the current magnitude of wildlandfire emissions.

This section discusses: (1) smoke from wildlandfires; (2) smoke from prescribed fires; (3) impacts onnational ambient air quality standards (NAAQS); (4)and magnitude with respect to regional and subre-gional scale visibility degradation. The second sectiondiscusses smoke management programs.

Smoke from Wildland Fires

Although wildland fires occur throughout the na-tion, the largest fires and greatest number of firesoccur in Alaska, the Southeastern States, and theWest. Figure 3-3 shows the location of major firesduring the 2000 fire season when 90,674 fires burned7,259,159 acres (2,938,931 ha) at a fire suppressioncost of $1.6 billion. The 10-year average acreage burnedbetween 1990 and 1999 was 3.78 million acres (1.53million ha), testifying to the severity of the 2000wildfire season. Figure 3-4 shows those States thathad more than 100,000 acres (40,486 ha) burned peryear, on average, over the 1987 through 1997 period,illustrating that Alaska wildfires burn far moreacres than fires in any other State. Area burned inCalifornia, the States in the Intermountain West,Florida, and the Southwest follow (Peterson 2000).


Figure 3-2—Estimated annual preindustrial, expected con-temporary, and contemporary biomass consumed (Tg x102) forthe conterminous United States (from Leenhouts 1998).

Wildfires occur throughout the year. The 2000 wild-fire season began with a Florida fire on January 1,continued with two 40,000-acre fires in New Mexico,an early May, 47,000-acre fire near Los Alamos andpeaked on August 29, 2000, when fires that eventuallyburned 1,642,579 acres were burning in 16 States(NIFC 2001a). Generally, the occurrence of wildfiresmoves northward from the Southeastern and South-western States as summer approaches, fuels dry andfire danger increases.

Figure 3-1—Estimated annual preindustrial, expected con-temporary, and contemporary area (Mha) for the conterminousUnited States (from Leenhouts 1998).

Wildfires, both in number and total acreage burned,vary widely from year to year and from region toregion. Figures 3-5 and 3-6 show no consistent relationbetween the number of fires and acres burned. It isknown, however, that smoke from these fires impactsair quality on both an episodic and long-term averagebasis over wide regions.

Wildfires occur as episodic events. For example, in1999, smoke from fires reduced visibility to less than100 feet (30 m) in Florida, prompting officials to advisepeople with respiratory problems to stay indoors (NewYork Daily News 1999). In the West, fires in six States(California, Nevada, Oregon, Montana, Washington,and Idaho) put thick smoke in many communities. InReno and cities in California’s Central Valley, smokefrom nearby wildfires prompted authorities to warnresidents with asthma to avoid unnecessary activity(USA Today 1999). Wildfire smoke is also transportedacross international boundaries. Fires in Canada werefound to cause high concentrations of carbon monoxideand ozone over a period of 2 weeks in the SoutheasternUnited States and across the Eastern seaboard duringthe summer of 1995 (Wotawa and Trainer 2000).

Smoke impacts during these episodic events canthreaten public health, cause smoke damage to build-ings and materials, and disrupt community activities.Although particulate concentrations in ambient airrarely reach health-threatening levels within majorcities, several communities in the United States haveexperienced particulate matter concentrations fromwildfire smoke that exceeded the Environmental Pro-tection Agency (EPA) significant harm emergencyaction level of 600 µg/m3 defined as an “imminent andsubstantial endangerment of public health” (EPA1992b).

For example, the Yellowstone National Park wildfiresof 1988 impacted communities in three States. Concen-trations of suspended particulate matter — both totalsuspended particulate (TSP) and PM10 — measured incommunities near the fires exceeded NAAQS, triggeringpublic health alerts and advisories (Core 1996). Anestimated 200,000 people were exposed to high concen-trations of smoke. In 1987, the Klamath fires of northernCalifornia burned for more than 60 days, resulting inwidespread smoke intrusions into numerous communi-ties in northern California and southern Oregon. Morerecently, wildfire impacts during the 2000 season werealso severe in several communities. Twenty-four aver-age PM10 concentration measured in Salmon, ID, reached225 µg/m3 on August 15, 2002, and 281 µg/m3 on August18, 2000, during wildfire smoke intrusions (Idaho De-partment of Environmental Quality n.d.).

Wildfire smoke can also be the dominant cause ofvisibility reduction during episodic events in theRocky Mountain States, on the Pacific Coast, and inthe Southeast (National Research Council [NRC]


Figure 3-3—Location of major wildfires in 2000 available at http://www.nifc.gov/fireinfo/2000/Top10fires.html.

Figure 3-4—States with more than 100,000 acres per year burned by wildfires.

Wildfire Acreage for States Exceeding 100,000 AcresAverage Acreage 1987–1997

Oklahoma

Arizona

Utah

Nevada

Montana

Oregon

Kansas

New Mexico

Florida

Wyoming

Idaho

California

Alaska

Sta

te

Acres900,000800,000700,000600,000500,000400,000300,000200,000100,0000


Figure 3-5—Number of wildfires per year 1990 through 1999 (National Interagency Fire Center 2002).

Figure 3-6—Number of acres burned by wildfires per year 1990 through 1999.


1993). Figures 3-7 and 3-8 are examples of the denseplumes of smoke that can be transported over hun-dreds of kilometers across State and internationalboundaries, degrading air quality, scenic values, and

highway safety. Between 1979 and 1988, 28 fatali-ties and more than 60 serious injuries were attrib-uted to smoke that drifted across roadways in theSouthern United States (Mobley 1989).

Figure 3-7—Big Bar Fire, Shasta-Trinity National Forest, California, August 1999(National Interagency Fire Center 2000).

Figure 3-8—Wildfire smoke transported across State lines, August 14,2000 (NASA).


Smoke from Prescribed Fires

On a national annual basis, PM10 emissions fromprescribed burns in 1989 were estimated to be over600,000 tons, half of which (380,000 tons) occurred inthe Southeastern States. Of the remaining 42 States,seven (Arizona, California, Idaho, Montana, Oregon,Texas, and Washington) were estimated to have an-nual emissions over 10,000 tons of PM10 from pre-scribed forest and rangeland burning (EPA 1992a;Peterson and Ward 1990). More recent estimates ofprescribed fire PM2.5 emissions in the West (EPAregions 8, 9 and 10) totaled 193,293 tons (Dickson andothers 1994). These national, annual estimates areless significant in terms of air quality impact thanthose prepared at the State level. For example, the211,000 tons of prescribed fire PM10 emissions inGeorgia in 1989 is about 30 percent of the total esti-mated particulate inventory for all sources (EPA1992a). On a seasonal basis, emissions from pre-scribed burning are likely to be an even more signifi-cant percentage of total emissions in some States.

Acreage treated by prescribed burning on Federallands increased from 918,300 acres in 1995 to 2,240,105acres in 1999, demonstrating renewed interest in theuse of fire as an important tool in the management ofwildlands (NIFC 2001b).

Impacts on National Ambient Air QualityStandards

Characterization of the true extent of effects ofprescribed and wildland fires on ambient air quality isincomplete due to the deficiency of air quality monitor-ing sites in rural areas. Also, particulate standards arebased on 24-hour and annual averages, whereas smokeplumes may significantly degrade air quality in acommunity for just a few hours before moving ordispersing. These short-term, acute impacts likelycause discomfort at the least, and possibly even affecthealth, but may not result in a violation of the NAAQS.

Numerous exceedances of 24-hour PM10 and PM2.5standards have been attributed to wildfires but, asmentioned previously, violations of NAAQS caused bywildfire do not result in nonattainment if a State candocument that the cause of the violation was trulywildfire and then prepares a natural events actionplan for future events.

At present, prescribed fires are not considered to bea significant cause of nonattainment, but with in-creased burning to reduce fuels, this situation maychange as land managers move forward with imple-menting a several-fold increase in the use of fire tosustain ecosystems (USDI and USDA 1995; USDA1997). In general, little information is available on anational level to identify the contribution of prescribedburning to PM10 or PM2.5 within nonattainment

areas (EPA 1992a). It appears, however, that there isno clear relation between total acres burned (or par-ticulate emissions) and the nonattainment status ofnearby airsheds, possibly because of successful smokemanagement programs.

In areas where air quality standards are being ormay be violated, however, land managers are beingdirected to reduce air quality impacts through smokemanagement programs. This is because any sourcethat contributes even a few micrograms per cubicmeter of particulate matter toward violation of theNAAQS may be required to reduce emissions to assurethat air quality standards are attained.

Significance of Visibility Degradation

As noted above, wildland fires can significantlydegrade visibility during episodic events. With thenew emphasis on the reduction of regional haze in theClass I National Parks and wilderness areas of thenation, smoke from fire is of special concern, especiallyin the West. In their report to the EPA, the GrandCanyon Visibility Transport Commission (GCVTC)noted that emissions from fire, both wildland fire andprescribed fire, are likely to have the single greatestimpact on visibility at Class I areas through 2040.During periods of intense fire activity, smoke fromwildland fires is likely to make the worst 20 percent ofdays at the Grand Canyon even worse rather thanimpair visibility on clear days (GCVTC 1996b). TheCommission recommended several actions to reduceimpacts on regional haze including enhanced smokemanagement programs and establishment of annualemissions goals for all fire programs.

Greenhouse Gas Emissions from Fires

Globally, fires are a significant contributor of carbondioxide and other greenhouse gases in the atmosphere.Fires account for approximately one-fifth of the totalglobal emissions of carbon dioxide (Levine and Cofer2000; Schimel 1995). Andreae and Merlet (2001) cal-culate that 5,130 Tg per year of biomass is consumedin fires, emitting 8,200 Tg per year of carbon dioxide,413 Tg per year of carbon monoxide, and 19.4 Tg peryear of methane. The accuracy of these global esti-mates is thought to be within plus or minus 50 percent,with the bulk of the error resulting from inaccuraciesin the estimates of the area burned and the mass offuel consumed.

Fires in temperate ecosystems are minor contribu-tors compared to the world’s savannas, boreal forests,and tropical forests. More than 60 percent of the totalslisted in the previous paragraph are released fromsavannas and grasslands, and another 25 percentfrom tropical forests. Burning in tropical Africa isdominated by savanna fires; in tropical Asia, by forest


fires; and in tropical South America, about equallyrepresented by savannas and tropical forests (Hao andLiu 1994). Lavoué and others (2000) detail contribu-tions from temperate and boreal fires, demonstratingthat about 90 percent of the global boreal fire area isin Russia and Canada. Alaska accounts for only about4.5 percent of the global boreal forest, but it accountsfor at least 10 percent of the emissions from thatsource, because of the heavier fuel loads in Alaska.Alaska accounts for an average of 41 percent of totalU.S. fire emissions, with a huge year-to-year variabil-ity. In 1990, 89 percent of U.S. fire emissions werefrom Alaska fires.

Smoke ManagementPrograms ______________________

Smoke management programs establish a basicframework of procedures and requirements whenmanagers are considering resource benefits. Theseprograms are typically developed by States and Tribeswith cooperation and participation by wildland own-ers and managers. The purposes of smoke manage-ment programs are to mitigate the nuisance (such asimpacts on air quality below the level of ambientstandards) and public safety hazards (such as visibil-ity on roads and airports) posed by smoke intrusionsinto populated areas; to prevent significant deteriora-tion of air quality and NAAQS violations; and toaddress visibility impacts in Class I areas.

The Interim Air Quality Policy on Wildland andPrescribed Fires (EPA 1998) provides clear guidelinesfor establishing the need for and content of smoke man-agement programs and assigns accountability to Stateand Tribal air quality managers for developing andadopting regulations for a program. Measured PM10NAAQS exceedances attributable to fires, includingsome prescribed fires and wildland fires managed forresource benefits, can be excluded from air quality datasets used to determine attainment status for a State.Special consideration will be given if the State or Tribalair quality manager certifies in a letter to the adminis-trator of EPA that at least a basic smoke managementprogram has been adopted and implemented.

States with smoke management programs that haveauthorized a central agency or office to make burn/no-burn decisions include Arizona, Colorado, Oregon,Idaho/Montana, Washington, California, Nevada, NewMexico, Florida, South Carolina, Utah, North Caro-lina, and Wyoming (Battye and others 1999). In manyother States, the decision to burn rests in the hands ofthe persons conducting the burn, local fire depart-ments, or local authorities. These States include Alaska,Alabama, Arkansas, Georgia, Louisiana, Mississippi,Tennessee, Texas, and Virginia. In yet other States

(New York, Illinois, Massachusetts, and others), burnpermits are required and may be subject to State airagency oversight if burning is conducted nearnonattainment areas or areas sensitive to smoke (Core1998; Hardy and others 2001). In addition, manyprivate landowners, nonprofit conservation organiza-tions and government agencies voluntarily practiceresponsible smoke management to maintain goodwillin their communities.

Smoke management programs have been estab-lished and are operated on an on-going basis becauseof local, regional, and national concerns about theimpact of prescribed burning on air quality. The num-ber, complexity, and cost of operating these programsunderscore the potential significance of prescribedfire’s impact on air quality on a national scale.

Smoke management programs across the nationhave changed significantly since the mid-1980s. In thePacific Northwest, there have been reductions in pre-scribed fire smoke management programs because ofthe decline in large-scale clearcut burning of forestharvesting residues. Current smoke management pro-grams across the West have to place a much greaterfocus than in the past on understory burning to restoredeclining forest health, on burns to reduce fire haz-ards, or on burns to meet wildlife habitat objectives.All across the nation, an increasing number of peopleliving within the wildland-urban interface have placednew emphasis on the need to minimize smoke impactson residents living near fires. Increasing air qualityregulatory pressures, fire manager liability issues,and the increased likelihood of fire escapement inoverstocked forestlands have all placed ever-greaterdemands on fire practitioners.

As these demands have increased, so have thenumber and complexity of smoke management pro-grams nationwide (Hardy and others 2001). Althoughthe complexity of these programs varies widely fromState to State, the key to a successful program alwayslies in its ability to balance the use of prescribed firewith air quality, environmental, legal, and socialrequirements. Increasingly, this has meant adoptionof formalized burn authorization procedures issuedby program managers who are responsible for over-seeing burning on both public and private lands on adaily basis. Coordinated burn operations are basedon meteorological forecasts, the location of smoke-sensitive receptors, fuel conditions, and a myriad ofother considerations. Increasingly, public notifica-tion of planned burning activity and monitoring ofsmoke transport, as well as fire practitioner trainingand program enforcement, are becoming more com-mon (Battye and others 1999).

As inter-State smoke transport becomes a largerissue, agencies are expanding coordination. For ex-ample, land management agencies in California’s


San Joaquin Valley are using a new centralized,electronic database, Prescribed Fire Incident Report-ing System (PFIRS), to schedule fires and to shareinformation on expected emissions and smoke trans-port with California and Nevada air and land man-agement agencies (Little n.d.). This trend is likely tocontinue as States begin to work on regional hazecontrol programs.

The Western Regional Air Partnership (WRAP) FireEmissions Joint Forum (FEJF) has issued a draftpolicy to set the criteria for enhanced smoke manage-ment plans for visibility protection in the West (FireEmissions Joint Forum 2002). The policy documentconcludes that the regional haze rule can be satisfiedonly by the States and Tribes establishing an emissiontracking system for all prescribed fires and wildlandfires; by managing smoke from all fires; and by imple-menting smoke management systems that includenine elements:

1. Actions to minimize emissions from fire2. Evaluation of smoke dispersion3. Alternatives to fire4. Public notification of burning5. Air quality monitoring6. Surveillance and enforcement7. Program evaluation8. Burn authorization9. Regional coordination

The enhanced smoke management plan (ESMP)policy would enable Western States and Tribes tominimize increases in emissions and show reasonableprogress toward the natural visibility goal. The FireEmissions Joint Forum is developing additional policyand technical tools that will support ESMP policy andits implementation, such as recommendations for cre-ation of an annual emissions goal, availability andfeasibility of alternatives to burning, recommenda-tions for managing fire emissions sources, guidancefor feasibility determinations, and a method for track-ing fire emissions.


All fires emit air pollutants in addition to nonpollut-ing combustion products; but fires vary widely in whatpollutants are emitted in what proportion. Character-izing and managing air pollution from fires first re-quires knowledge of the amount and timing of whatpollutants are emitted. Fires are a complex combus-tion source that involve several stages of combustion,several categories of fuels, and fire behavior thatchanges over time and with fuel and weather condi-tions; so the amount, rate, and nature of pollutantsalso vary widely. Characterizing emissions from firesrequires explicit knowledge of fuelbed character andcondition, combustion environment, and fire behav-ior.

This chapter reviews the state of knowledge andpredictive models necessary to characterize air pollut-ant emissions from prescribed and wildland fires.

All components of smoke from fires, with the excep-tion of carbon dioxide and water, are generated fromthe inefficient combustion of biomass fuels. The amountof smoke produced is derived by determining the fuelconsumed (tons per acre) in each combustion stage andknowing the size of the area burned, fuel characteris-tics, fire behavior, and combustion conditions (fuelmoisture, weather parameters, and so forth). The fuel

Chapter 4:Characterization ofEmissions from Fires

consumption is then multiplied by an emissions factorfor each pollutant, which is an expression of theefficiency of combustion. An emission factor is theratio of the mass of pollutant per unit mass of fuelconsumed, and is a statistical average of measure-ments made in the plumes of fires containing differingfuel types and combustion stages. Errors and uncer-tainties arise in the estimates made during each stepin the process of estimating emissions.

Area Burned____________________At first glance, amount of area burned seems rela-

tively easy to calculate. However, individual esti-mates of fire size tend to be systematically exagger-ated, and fires are frequently double-counted ininventories. For example, geographic features, non-uniform fuelbeds, or a change in the weather will oftencause a fire to create a mosaic of burned, partiallyburned, and unburned areas, although the entirelandscape within the fire perimeter is often reportedas burned. In addition, large-scale (such as continen-tal) inventories of area burned are often derived fromremote sensing data that have resolutions from 250 mto 1 km (SAI 2002), limiting their precision. Remote


sensing accuracy is currently inadequate in land-scapes that change slope and fuel characteristics overa few tens of meters.

Preburn Fuel Characteristics ______Large variations in fuel characteristics can contrib-

ute up to 80 percent of the error associated withpredicting emissions (Peterson 1987; Peterson andSandberg 1988). Fuel characteristics can vary widelyacross the landscape (figs. 4-1 and 4-2). For instance,fuel loads can range from less than 3 tons per acre forperennial grasses with no rotten woody material orduff, 6 tons per acre in a sagebrush shrubland, 60 tonsper acre in a ponderosa pine and Douglas-fir forestwith rotten woody material, stumps, snags, and deepduff, to 160 tons per acre in a black spruce forest withdeep moss and duff layer. The greatest errors occurwhen the fuel load is inferred from vegetation type as

is usual when deriving biomass emissions from re-motely sensed data (Crutzen and Andrae 1990; Levine1994). Preburn fuel characteristics, such as relativeabundance for particular fuelbed components (grasses,shrubs, woody fuels, litter, duff, and live vegetation)and the condition of the fuel (live, dead, sound, rotten)are needed to calculate fuel consumption, and theresulting smoke.

The ongoing development of several techniques,including the natural fuels photo series (Ottmar andVihnanek 2000a) and the fuel characteristic classifi-cation (FCC) system (Sandberg and others 2001), willprovide managers new tools to better estimate fuelloadings and reduce the uncertainty that currentlyexists when assigning fuel characteristics across alandscape. The photo series is a sequence of single andstereo photographs with accompanying fuel charac-teristics. The FCC is a national system designed forclassifying wildland fuelbeds according to a set of

Figure 4-1—Fuelbed types and fuel loads (a) grassland (3 tons per acre), (b) sagebrush (6 tons per acre), (c) ponderosapine with mortality in mixed fir (60 tons per acre), and (d) black spruce with deep duff and moss (160 tons per acre).(Photos by Roger Ottmar)

d

a

c

b


inherent physical properties, thereby providing thebest possible fuels estimates and probable fire param-eters based on available site-specific and remotelysensed information.

Fire Behavior ___________________Fire behavior is the manner in which fire reacts to

the fuels available for burning (DeBano and others1998) and is dependent upon the type, condition, andarrangement of fuels, local weather conditions, topog-raphy, and in the case of prescribed fire, ignitionpattern and rate (fig. 4-3). Important aspects of firebehavior include:

• Fire intensity (rate of energy release per unitarea or unit length of fire perimeter, generallyduring the flaming combustion period).

• Rate of spread (rate of advancement of flamingfront, length per unit time), crowning poten-tial (involvement of tree and shrub foliage andspread within the canopy), smoldering poten-tial (smoldering combustion of fuels that havebeen preheated or dried during the flamingstage).

• Residual smoldering potential (propagation ofa smoldering combustion front within porousfuels such as rotten logs or duff, independentof preheating or drying).

• Residence time in the flaming, smoldering,and residual stages of combustion.

These aspects influence combustion efficiency of con-suming biomass, as well as the resulting pollutantchemistry and emission factor (fig. 4-4).

The Emissions Production Model (EPM) (Sandberg2000; Sandberg and Peterson 1984) and FARSITE(Finney 1998) take into account fire behavior andignition pattern to estimate emission production rates.Fire behavior during the flaming stage of combustionin surface woody fuels and some shrub vegetation iseffectively predicted within models such as BEHAVE(Andrews and Bevins 1999) and its spatial applica-tion, FARSITE (Finney 1998). However, EPM andother applications do not consider fire intensity orother fire behavior attributes when estimating emis-sions from flames, and that may result in a reasonableapproximation for criteria pollutants but also be alimitation to the estimate of hazardous air pollutantsor trace gases. BURNUP (Albini and Reinhardt 1997),FARSITE (Finney 1998), and EPM v2.0 (Sandberg2002) attempt to model the extent and duration offlaming and smoldering combustion in downed woodyfuels and duff. Current capability to model residualcombustion, combustion in rotten logs and duff, andfire behavior in the foliage canopies of trees and someshrubs remains inadequate to predict emission rateswith any reasonable degree of accuracy.

The Los Alamos and Lawrence Livermore nationallaboratories offer an approach to predicting fire be-havior, plume trajectory, and dispersion, by combin-ing a fire physics model, FIRETEC, with a dynamicatmosphere model, HIGRAD, to produce a highlydetailed numerical simulation of fire spread and atmo-spheric turbulence (Bradley and others 2000). Theapproach builds on prior experience in predicting thedispersion of hazardous air pollutants from fires suchas burning oil fields or “nuclear winter” scenarios.This modeling approach is limited to the propagatingfront but is unique in its coupling of atmospheric andfire physics.

Figure 4-2—Various fuelbeds across a single landscape.(Photo by Roger Ottmar)

Figure 4-3—Fire behavior in the leaf layer of a longleaf pineforest. (Photo by Roger Ottmar)


Figure 4-4—Fuel consumption in (a) large rotten log duringa fall prescribed burn, (b) pile burning during a prescribedburn, (c) litter and duff during a prescribed burn, (d) grassduring a wildfire, and (e) sagebrush during a prescribed fire.(Photos by Roger Ottmar)

Combustion Stages _____________At least three important stages of combustion exist

when fuel particles are consumed (Mobley 1976; NWCG1985): flaming, smoldering, and residual (also knownas “glowing,” “residual smoldering,” or “residual com-bustion”) (fig. 4-5). The efficiency of combustion isdistinct for each stage, resulting in a different set ofchemical compounds and thermal energy being re-leased at different rates into the atmosphere. In theflaming phase, combustion efficiency is relatively highand usually tends to emit the least amount of pollutantemissions compared with the mass of fuel consumed.

The predominant products of flaming combustionare CO2 and water vapor. During the smolderingphase, combustion efficiency is lower, resulting inmore particulate emissions generated than during theflaming stage.

Smoldering combustion is more prevalent in certainfuel types such as duff, organic soils, and rotten logs,and often less prevalent in fuels with high surface tovolume ratios such as grasses, shrubs, and smalldiameter woody fuels (Sandberg and Dost 1990).

ba

dc

e


Figure 4-5—Flaming, smoldering, and residual combustionstages during a fire. (Photo by Roger Ottmar)

The residual stage differs from the smoldering stagein that the smoldering stage is a secondary processthat occurs in fuels preheated or dried by flamingcombustion, while residual is an independent processof propagation in a fuelbed unaffected by the flamingstage. This phase is characterized by little smoke andis composed mostly of CO2 and carbon monoxide. Allcombustion stages occur sequentially at a point, butsimultaneously on a landscape.

Fuel Consumption_______________Fuel consumption is the amount of biomass con-

sumed during a fire and is another critical componentrequired to estimate emissions production from fire.Biomass consumption varies widely among individualfires depending on the fuelbed type, arrangement, andcondition, weather parameters, and the way the fire isapplied in the case of prescribed fire. As with fuelcharacteristics, extreme variations can be associatedwith fuel consumption resulting in an error contribu-tion of 30 percent or more when emissions are esti-mated (fig. 4-6) (Peterson 1987; Peterson and Sandberg1988).

Biomass consumption of woody fuels, piled slash,and duff in forested areas has become better under-stood in recent years (Albini and Reinhardt 1997;Brown and others 1991; Ottmar and others 1993;Ottmar and others [N.d.]); Reinhardt and others 1997;Sandberg 1980; Sandberg and Dost 1990). Consump-tion of forested crowns and shrublands are the leastunderstood components of biomass consumption, andresearch is currently under way (Ottmar and Sandberg2000) to develop or modify existing consumption equa-tions for these fuel components. Equations for predict-ing biomass consumption in the flaming and smolder-ing combustion stages are widely available in twomajor software packages, Consume 2.1 (Ottmar andothers [N.d.]) and the First Order Fire Effects Model(FOFEM 5.0) (Reinhardt and Keane 2000).

Figure 4-6—The largest errors are associated with fuel loading and fuel consumptionestimates when determining emission production and impacts from wildland fire (Petersonand Sandberg 1988).


Emission Factors _______________Emissions from fires or from points over fires have

been observed extensively by researchers since about1970. The result is a complete set of emission factors(pounds of pollutant per ton of fuel consumed) forcriteria pollutants and many hazardous air pollutantsfor most important fuel types. These are available inseveral publications (for example, Battye and Battye2002, EPA 1972, Hardy and others 2001, Ward andothers 1989) and are not reproduced here.

Less complete compilations of emission factors arefor particulate matter components such as size classdistribution, elemental and organic carbon fractions,and particulate hazardous air pollutants; and formethane, ammonia, aldehydes, compounds of nitro-gen, volatile organic hydrocarbons, and volatile haz-ardous air pollutants (for example, Battye and Battye2002, Goode and others 1999, Goode and others 2000,Lobert and others 1991, McKenzie and others 1994,and Yokelson and others 1996).

Source Strength ________________Source strength is the rate of air pollutant emissions

in mass per unit of time, or in mass per unit of time perunit of area. Source strength is the product of the rateof biomass consumption (that is, fuel consumption)and an emission factor for the pollutant(s) of interestand is representative of the physical and chemical fuelcharacteristics (fig. 4-7). Source strength or emissionrate is required as an input to dispersion models(Breyfogle and Ferguson 1996), or to break downemission inventories into time periods shorter thanthe duration of a fire event. Source strength is alsorequired in photochemical models such as the commu-nity multiscale air quality model (CMAQ) (Byun andChing 1999) to account for timing of chemical reac-tions with diurnal patterns and interaction with othersources.

Total emissions from a fire or class of fires are thesource strength integrated over the time of burning.Total emissions from a single class of fires (that is, aset of fires similar enough to be characterized by asingle emission factor) can be estimated by multiply-ing that emission factor by the level of activity, whichis the total biomass consumed by the class of fires. Anemission inventory is the aggregate of total emissionsfrom all fires or classes of fire in a given period for aspecific geographic area.

Managing the source strength (or level of activity) offires is the most direct way to control air pollution fromwildland and prescribed fires. Prediction of sourcestrength is sometimes used to manage the rate ofemissions from fires, and it also is needed as an inputto dispersion models. Standards or regulations are

commonly set to limit the total emissions of pollutants,emission of specific hazardous air pollutants, or thelevel of activity, so that estimates of biomass consump-tion can be essential for environmental assessment,permitting of prescribed fires, or measuring compli-ance. Emission inventories are a critical part of impactanalyses and strategy development so the level ofactivity must be estimated whenever there is a regu-latory application.

The Emissions Production Model (Sandberg 2000;Sandberg and Peterson 1984) is currently the mostwidely used model for predicting source strength forprescribed fires. EPM v.1 predicts flaming and re-sidual emissions rates for each criteria pollutantbased on a simple formula that assumes a constantrate of ignition of a prescribed fire in uniform fuels.The software package pulls fuel consumption predic-tions from Consume 2.1 or FOFEM 5.0 and usesignition pattern, ignition periods, and burn areacomponents to calculate source strength for the flam-ing and residual combustion phases. EPM v.1 doesnot consider smoldering emissions (for example, long-duration, self-propagating glowing combustion),multiple fires or multiple burn periods, wildland fireor piled burning emissions, or diurnal and spatialchanges in the fire environment. EPM v.2, now underdevelopment (Sandberg 2000), corrects all of theseshortcomings in a dynamic simulation model. EPM v.2will satisfy the requirement to provide hourly esti-mates of emission rates for most fires and fuelbedsneeded for input into Models-3/CMAQ (see the “GridModels” section in chapter 5) and into currentlyenvisioned smoke management screening systems.

Figure 4-7—A high-intensity Alaska wildfire with heavy fuelloads, causes a high rate of emissions. (Photo by RogerOttmar)


FARSITE (Finney 2000) has been modified to pre-dict emissions source strength as well as fire behaviorin a detailed spatial simulation. FARSITE incorpo-rates BURNUP (Albini and Reinhardt 1997), whichestimates consumption and rates of individual fuelelements.

Accurate characterization of emissions from fires iscritical to predicting the impact emissions will have

upon communities and across broader landscapes andairsheds. Managers will increasingly be required toprovide this type of information prior to prescribedburns, as well as during the course of wildland fires,and the information provided here summarizes thestrengths and weaknesses of the various means ofprediction.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


Chapter 5: Transport,Dispersion, and Modeling ofFire Emissions

To anticipate the impacts of smoke, the timing andlocation of smoke concentrations become important.Data on the site-specific surface concentrations ofrespirable particles and gases often are needed forestimating impacts on public health and welfare, re-quiring atmospheric dispersion and transport modelsthat can approximate the atmospheric physics andchemical reactions that occur during transport nearthe ground. Data on the cumulative concentrations ofelements that scatter and absorb light also are neededto estimate impacts on visibility and haze, requiringmodels that can approximate aqueous reactions aswell as physical and chemical reactions at all levels ofthe atmosphere.

Although progress is being made, none of the cur-rently available models fully meet the needs of fireplanners and air resource managers. Much of thedeficiency in current modeling approaches is causedby inherent uncertainties associated with turbulentmotions between the fire, smoke, and the atmospherethat are compounded by the highly variable distribu-tion of fuel elements, composition, and condition.

Another source of deficiency is that most availablemodels were originally designed for well-behaved sourcessuch as industrial stacks or automobile emissions,

while emissions from fire can be extremely variable inboth time and space. Also, outputs from currentlyavailable models do not always match the temporal orspatial scale needed for land management application.

To help readers understand the strengths and weak-nesses of available models, we describe basic elementsof the trajectory and dispersion of smoke. This chapterconcludes with a summary of currently available mod-els and a brief guide to applications.

Basic Elements of Trajectory andDispersion _____________________

Ambient air quality can be measured at a point or asdistribution of air quality over any space and time ofinterest. Ambient air quality is affected by the pollut-ants emitted to the atmosphere from fires, the back-ground air quality that has already been degraded byother sources, the transport of the polluted parcels ofthe atmosphere, dispersion due to atmospheric move-ment and turbulence, secondary reactions, and re-moval processes. Plume rise is an important componentof transport, because it determines where in the verti-cal structure of the atmosphere dispersion will begin.


Overall, dispersion has proven extremely difficultto model accurately, especially in complex terrain.For example, detailed, gridded, three-dimensionalmeteorological data are required to model transportand dispersion, but expert judgment is often re-quired to supplement or substitute for such modeledpredictions.

Despite the difficulties of modeling, since about1990 modeling systems used to assess the air qualityimpact of fires have grown increasingly important toboth the fire planning and air quality communities.There is a broad range of acceptable tools from rela-tively simple methods used by local fire managers forestimating likely impacts on air quality standards (forexample, SASEM: Riebau and others 1988; andVSMOKE: Lavdas 1996), to complex terrain and re-gional-scale models that incorporate atmosphericchemistry to assess impacts on regional haze (forexample, Calpuff: Scire and others 2000a, and Models-3: Byun and Ching 1999).

The tremendous growth in model application placesincreasingly greater demands on the user, requiringaccess to detailed fuel characteristics, fuel consump-tion, ignition pattern, fire behavior, and meteorologi-cal inputs. Also needed is the ability to interpret thecomplex smoke dispersion model outputs.

In this section we describe such processes of heatrelease, plume rise and buoyancy, advection and diffu-sion, scavenging, and chemical transformations.

Heat Release

The consumption of biomass produces thermal en-ergy, and this energy creates buoyancy to lift smokeparticles and other pollutants above the fire. Heatrelease rate is the amount of thermal energy gener-ated per unit of time. Total heat release from a fire orclass of fires is a function of the heat content of thebiomass, fuel consumed, ignition method and pattern,and area burned.

The early work of Anderson (1969) and Rothermel(1972) created fundamental equations for combustionenergy in a variety of fuelbeds. Sandberg and Peterson(1984) adapted the combustion equations to model thetemporal change in energy during flaming and smol-dering combustion (Emission Production Model,EPMv.1.02). Currently, EPM provides heat releaserates for most biomass smoke dispersion models(Harms and others 1997; Harrison 1995; Lavdas 1996;Sestak and Riebau 1988; Scire and others 2000a) andhas been used to estimate the change in global biomassemissions patterns due to changes in land use(Ferguson and others 2000). The model, however,requires a constant rate of ignition with constant slopeand wind. Such homogeneous conditions may be ap-proximated during prescribed fires that are ignitedwith a deliberate pattern of drip torches or airborne

incendiaries, or during portions of wildfires that expe-rience relatively constant spread rates, both overfuelbed strata that retain a relatively consistent spa-tial and compositional pattern. To use EPM effectivelyfor modeling source strength, the fire area and ignitionduration are broken into space and time segments thatmeet the steady-state criteria.

Albini and others (1995), Albini and Reinhardt (1995),and Albini and Reinhardt (1997) do not explicitlyderive temporal changes in combustion energy in theirmodel, BurnUp, but they do assign source heat insteps of flaming and smoldering that are estimatedfrom total fuel consumption. They have linked theirmodel with the fire spread model, FARSITE (Finney1998), which allows ignition rates and subsequentheat-release rates to vary over the landscape. Thecoupled system is computationally expensive and notyet associated with a plume rise component but mayoffer a reasonable approximation of the temporal andspatial varying emission rates of fires.

Plume Rise and Buoyancy

Heat, particle, and gas emissions from fires vary intime and space, causing unique patterns of convectionand resulting plume rise. This plume rise is a functionof free convection in the atmosphere, which is causedby density differences within the fluid. As a fire heatsand expands air near the ground, large density differ-ences between the heated volume and the surroundingair mass are created, causing the heated parcel to rise.The potential height of the resulting plume dependson the heat energy of the source and rise velocity,which is affected by the exchange and conservation ofmass, radiant heat loss, the buoyancy force, and tur-bulent mixing with the ambient air.

Hot, flaming fires can develop central convectivecolumns with counter-rotating vortices that involvemassive entrainment of the surrounding air mass(Clark and others 1996; Haines and Smith 1987;Haines and Updike 1971). This stage of fire can pro-duce fast-rising plumes and turbulent downdrafts,carrying sparks that ignite new fires. Cumulonimbusclouds often develop with accompanying lightning andrain. Dynamic plume rise brings gas and particleshigh into the atmosphere where strong winds candisperse the smoke hundreds to thousands of kilome-ters. As high intensity fires cool, however, the centralcolumn often collapses, creating numerous small con-vective cells that are less dynamic but equally active incarrying smoke into the atmosphere. Smoldering firesoften create plumes that are neutrally buoyant, limit-ing widespread dispersion but allowing surface windsto dominate smoke trajectories. This can lead to accu-mulations of smoke in valleys and basins at night.

Because plume rise can eventually result in wide-spread dispersion, plume rise calculations are essential


for determining the height above ground from whichplume dispersion is initiated. Uncertainties in suchcalculations can result in inaccurate predictions ofplume transport and downwind smoke impacts. Giventhe pressing need to predict the impact of plumes fromfires, the need for improved plume rise calculations isapparent.

The basic mechanisms and algorithms used to de-scribe plume rise and buoyancy were developed in themid-1960s by Briggs (1969) for industrial, ductedemissions. These methods are still used today to esti-mate the plume rise and buoyancy of fires in spite ofthe significant differences in characteristics betweenducted emissions and prescribed and wildland fires:

• Heat released from ducted sources is preciselyknown and usually emitted at relatively con-stant rates during a single phase of combus-tion. Heat released from fires is a function offuel loading, fuel conditions, and ignitionmethod through several phases of combustion(pre-ignition, flaming, smoldering, and re-sidual), which create highly variable magni-tudes and rates of heat release.

• Nearly all of the energy generated at thesource of a ducted plume is transmitted toconvection energy. In open burning, however,significant amounts of energy are lost by con-duction and radiation, reducing the amount ofavailable energy for convection.

• Plumes from ducted sources create single con-vective columns, but low intensity understoryburning that occurs over broad areas does notdevelop a cohesive plume.

To improve plume rise predictions, emission produc-tion models need to do a better job of characterizing thespatial and temporal pattern of heat release fromfires, and plume rise models need to be improved toaccount for the energy lost from the convective systemthrough radiation and turbulent mixing. While mod-els such as EPM and Burnup described in the previoussection simulate variable rates of heat release fromfires, both models use general estimates of spatialdistributions of fuel, including structure, composition,and moisture content. Also, significant elements offires that influence convective energy — such as thedistribution of naturally piled fuel (“jackpots”), amountand density of rotten fuel and duff, and release ofwater vapor — are not adequately captured.

Rough approximations on the proportion of energyavailable for convection were made more than 40 yearsago (Brown and Davis 1959). Despite efforts to improveplume rise calculations by removing the density dif-ference assumption (Scire and others 2000a), theystill are in use today.

Low intensity fires that typically do not have acohesive convective column must be treated, from amodeling perspective, as an area source in Euleriangrid models. In Lagrangian dispersion models, there iscurrently no valid means of calculating plume risefrom unconsolidated convection. Eulerian coordinates(used by box and grid models) are coordinate systemsthat are fixed in space and time, and there is noattempt to identify individual particles or parcels fromone time to the next. Lagrangian models (bell-shape orGaussian distribution pattern, often applied to plumeand puff models) are used to show concentrationscrosswind of the plume.

Another complication for modeling is that onceplumes from fires enter the atmosphere, their fluctu-ating convection dynamics make them more suscep-tible to erratic behavior than well-mannered indus-trial stacks. For example, different parts of a plumecan be carried to different heights in the atmosphereat the same time. This causes unusual splitting pat-terns if there is a notable wind shear between loftedelevations, causing different portions of the plume tobe transported in different directions. Therefore, pre-dictions of the plume’s impact on visibility and airquality under these conditions become highly uncer-tain (Walcek 2002). Even when the behavior of plumesfrom fires resembles that of stack plumes, the varyingand widely distributed locations of wildland sourcesprevent consistent study. For example, down-wash ofplumes has been observed from ducted (stack) emis-sions after an inversion breaks up — conditions thatare common at the end of an onshore breeze if theplume is above the inversion at its source (de Nevers2000; Venkatram 1988) or if horizontal stratificationin the lower atmosphere is disrupted by mountains (deNevers 2000).

These characteristics of plumes from fire are strik-ingly different than those of ducted industrial emis-sions yet little research has been done on this topic inthe past several decades.

Advection and Diffusion

In most existing models, the horizontal advectionof smoke and its diffusion (lateral and vertical spread)are assumed to be controlled mainly by wind, and theformation and dissipation of atmospheric eddies.These elements are greatly simplified by assumingconstant wind (at least for an hourly time step) insome cases (such as VSMOKE and SASEM), and aGaussian dispersion is nearly always imposed. Per-haps the most critical issues are the constantly chang-ing nature of the plume due to scavenging, chemicaltransformation, and changing convection dynamicsthat affect plume transport.

Many photochemical and dispersion models de-pend on gridded meteorological inputs. Unfortunately,


numerical formulations of dynamic meteorologicalmodels (for example, MM5: Grell and others 1995;RAMS: Pielke and others 1992) do not adequatelyconserve several important scalar quantities (Byun1999a,1999b). Therefore, modelers often introducemass-conserving interpolations. For example, Mod-els-3/CMAQ (Byun and Ching 1999) uses the MCIPscheme (Byun and others 1999), Calpuff (Scire andothers 2000a) employs CALMET (Scire and others2000b), and TSARS+ (Hummel and Rafsnider 1995)is linked with NUATMOS (Ross and others 1988).Driving a photochemical or dispersion model withoutthese mass-conserving schemes will produce inaccu-rate results, especially near the ground surface.

Scavenging

Smoke particles by nature of their small size provideefficient cloud condensation nuclei. This allows clouddroplets to condense around fine particles, called nucle-ation scavenging. Scavenging within a cloud also canoccur as particles impinge on cloud droplets throughBrownian diffusion, inertial impaction, or collision byelectrical, thermal, or pressure-gradient forces(Jennings 1998). Cloud droplets eventually coalesceinto sizes large enough to precipitate out, thus re-moving smoke aerosols from the atmosphere. Whileinterstitial cloud scavenging, especially nucleationscavenging, is thought to dominate the pollutionremoval process, particles also may be removed byimpacting raindrops below a cloud. Jennings (1998)reviews several theories on pollution scavenging butcontends that there is little experimental evidence tosupport such theories.

The size and chemical structure of particles deter-mine their efficiency in nucleation or other scavengingmechanisms. While the chemical composition of smokeis reasonably well known (see chapter 6), distributionsof particle size from fire are not. The few airbornemeasurements (Hobbs and others 1996; Martins andothers 1996; Radke and others 1990) do not distin-guish fire characteristics or combustion dynamics,which play important roles in the range of particlesizes emitted from a fire. Therefore, the efficiency ofscavenging biomass smoke particles out of the atmo-sphere by cloud droplets, rain, or other mechanism hasnot been quantified.

Chemical Transformations

Chemical transformations provide another mecha-nism for changing particle and gas concentrations withina plume. Chemical transformation in the plume can beimportant in regional-scale modeling programs wheresulfate chemistry and ozone formation are of interest(see chapter 6). Oxidation within the smoke plumecauses a loss of electrons during chemical transforma-

tion processes, which increases polarity of a moleculeand improves its water solubility (Schroeder and Lane1988). This improves scavenging mechanisms by cloudand rain droplets. Chemical transformation rates de-pend on complex interactions between catalysts andenvironmental conditions such as turbulent mixing rates.

Transport and DispersionModels ________________________

Trajectories show the path of air parcels along astreamline in the atmosphere. Their simplicity allowstrajectory methods to be used as a diagnostic tool foridentifying the origin of air parcels from a potentialreceptor. This commonly is called a backward trajec-tory or back trajectory analysis. Because these modelsintegrate over time the position of a parcel of air thatis transported by wind, their accuracy is limited by thegrid resolution of the model. Also, the flow path of asingle parcel may have little relation to an actualplume dispersion pattern.

Current models to predict trajectory or air qualityimpacts from fires are inadequate in coverage and areincomplete in scope (Sandberg and others 1999). Butbecause of new interest in modeling emissions on aregional scale, land managers need transport anddispersion models that include all fire and fuel typesas well as multiple sources. Such models need to belinked to other systems that track fire activity andbehavior as well as provide for variable scaling to fitthe area of interest. At the operational level, modelsthat support real-time decisionmaking during fireoperations in both wildland fire situation analysis andgo/no-go decision making are also needed (Breyfogleand Ferguson 1996). Transport and dispersion modelsfall into four major categories. These categories in-clude plume, puff, particle, and grid.

Plume Models

One of the simplest ways of estimating smoke con-centrations is to assume that plumes diffuse in aGaussian pattern along the centerline of a steady windtrajectory. Plume models usually assume steady-stateconditions during the life of the plume, which meansrelatively constant emission rates, wind speed, andwind direction. For this reason, they can be used onlyto estimate concentrations relatively near the sourceor for a short duration. Their steady-state approxima-tion also restricts plume models to conditions that donot include the influence of topography or significantchanges in land use, such as flow from a forest tograssland or across a land-water boundary.

Gaussian plume models have a great benefit in placesand circumstances that restrict the amount of availableinput data. They can be run fast and have simple butrealistic output that can be easily interpreted. Many


regulatory guidelines from the EPA are based onGaussian plume models.

Plume models typically are in Lagrangian coordi-nates that follow particles or parcels as they move,assigning the positions in space of a particle or parcelat some arbitrarily selected moment. (Lagrangiancoordinates are used by plume, puff, and particlemodels.) Examples adapted for wildland biomass smokeinclude VSMOKE (Harms and others 1997; Lavdas1996) and SASEM (Riebau and others 1988; Sestakand Riebau 1988). Both models follow regulatory guide-lines in their development and offer a simple screeningtool for examining potential concentrations at recep-tor locations from straight-line trajectories relativelynear the source. However, SASEM directly comparesdownwind concentrations with ambient standards andcalculates visibility impairment in a simple manner.It is also used as a State regulatory model in Wyoming,Colorado, New Mexico, and Arizona, and has beenrecommended for use by the EPA.

Plume rise models developed for other applicationsmight be useful if adapted to fire environments. Forexample, ALOFT-FT (A Large Outdoor Fire PlumeTrajectory Model - Flat Terrain), developed for oil-spillfires (Walton and others 1996), is a computer-basedmodel to predict the downwind distribution of smokeparticulate and combustion products from large out-door fires. It solves the fundamental fluid dynamicequations for the smoke plume and its surroundingswith flat terrain. The program contains a graphicaluser interface for input and output, and a database offuel and smoke emission parameters that can bemodified by the user. The output can be displayed asdownwind, crosswind, and vertical smoke concentra-tion contours.

Puff Models

Instead of describing smoke concentrations as asteadily growing plume, puff models characterizethe source as individual puffs being released overtime. Each puff expands in space in response to theturbulent atmosphere, which usually is approxi-mated as a Gaussian dispersion pattern. Puffs movethrough the atmosphere according to the trajectoryof their center position. Because puffs grow andmove independently of each other, tortuous plumepatterns in response to changing winds, varyingtopography, or alternating source strengths can besimulated with some accuracy.

Some models allow puffs to expand, split, compact,and coalesce (Hysplit: Draxler and Hess 1998; Calpuff:Scire and others 2000a) while others retain coherentpuffs with constantly expanding volumes (NFSpuff:Harrison 1995). In either case, the variability of puffgeneration, movement, and dispersion does not re-strict the time or distance with which a plume can be

modeled. Most puff models are computed in Lagrangiancoordinates that allow accurate location of specificconcentrations at any time.

Particle Models

In a particle model, the source is simulated by therelease of many particles over the duration of the burn.The trajectory of each particle is determined as well asa random component that mimics the effect of atmo-spheric turbulence. This allows a cluster of particles toexpand in space according to the patterns of atmo-spheric turbulence rather than following a parameter-ized spatial distribution pattern, such as commonGaussian approximations. Therefore, particle modelstend to be the most accurate way of simulating concen-trations at any point in time. Because of their numeri-cal complexity, however, particle models usually arerestricted to modeling individual point sources withsimple chemistry or sources that have critical compo-nents such as toxins that must be tracked precisely.Particle models use Lagrangian coordinates for accu-rate depiction of place of each time of particle move-ment (for example, Hysplit: Draxler and Hess 1998;PB-Piedmont: Achtemeier 1994, 2000).

Grid Models

Grid models use Eulerian coordinates, disperse pol-lutants uniformly within a cell, and transport them toadjacent cells. The simplicity of advection and diffu-sion in a grid model allows these models to moreaccurately simulate other characteristics of the pollu-tion, such as complex chemical or thermal interac-tions, and to be used over large domains with multiplesources. This is why grid models commonly are usedfor estimating regional haze and ozone and are oftencalled Eulerian photochemical models. Much of thefuture work on fire impact assessment and planning atregional to national scales will be done by using gridmodels.

Because of their nature, grid models are not used todefine accurate timing or locations of pollutant con-centrations from individual plumes, only concentra-tions that fill each cell. This means that sourcessmall relative to the grid size, which create individualplumes, will introduce unrealistic concentrations inplaces that are outside of the actual plume. Ways ofapproximating plume position and its related chemi-cal stage include nesting grids to finer and finerspatial resolutions around sources of interest (Changand others 1993; Odman and Russell 1991), estab-lishing nonuniform grids (Mathur and others 1992),and creating “plume-in-grid” approximations (Byunand Ching 1999; Kumar and Russell 1996; Morrisand others 1992; Myer and others 1996; Seigneur andothers 1983).


Many regional haze assessments use the RegulatoryModeling System for Aerosols and Acid Deposition(REMSAD) (Systems Applications International 2002).This model was adapted from the urban airshed model–variable grid (UAMV) by removing its plume-in-gridfeature and parameterizing explicit chemistry to im-prove computational efficiency. REMSAD incorporatesboth atmospheric chemistry and deposition processesto simulate sulfate, nitrate, and organic carbon par-ticle formation and scavenging. As such, it is quiteuseful for simulations over large regions.

The Models-3/ CMAQ modeling system is designedto integrate the best available modules for simulatingthe evolution and dispersion of multiple pollutants ata variety of scales (Byun and Ching 1999). It includeschemical transformations of ozone and ozone precur-sors, transport and concentrations of fine particlesand toxics, acid deposition, and visibility degradation.

At the other end of the grid modeling spectra aresimple box models that describe pollution characteris-tics of a small area of interest. Box models instanta-neously mix pollutants within a confined area, such asa valley. This type of model usually is restricted toweather conditions that include low wind speeds anda strong temperature inversion that confines the mix-ing height to within valley walls (Lavdas 1982; Sestakand others 1988). The valley walls, valley bottom, andtop of the inversion layer define the box edges. The endsegments of each box typically coincide with terrainfeatures of the valley, such as a turn or sudden eleva-tion change. Flow is assumed to be down-valley, andsmoke is assumed to instantaneously fill each boxsegment. Few box models include the complex chemi-cal or particle interactions that are inherent in largergrid models.

Model Application _______________Modeling of the transport and dispersion of indus-

trial stack plumes has occurred for decades, prompt-ing a variety of techniques. But application to fires ismuch more limited (Breyfogle and Ferguson 1996).Part of the reason for this is that source strength fromundulating and meandering fires is so difficult tosimulate accurately. Therefore, applications have beenappropriate mainly for relatively homogeneousfuelbeds and steady state burn conditions. This hasrestricted most transport and dispersion modeling tofires on a local scale and to those started in harvestresidue from land clearing operations where fuels arescattered uniformly over the landscape or collectedinto piles (Hardy and others 1993; Hummel andRafsnider 1995; Lavdas 1996; Sestak and Riebau1988). Global-scale modeling also has taken placewhere fuelbed and ignition patterns are assumed to be

approximately steady state in relation to the grid size(Kasischke and Stocks 2000; Levin 1996).

Gaussian plume models (Harms and others 1997;Lavdas 1996; Sestak and Riebau 1988; Southern For-est Fire Laboratory Personnel 1976) are useful forplaces with relatively flat terrain, for circumstanceswhen input data are scarce, and for evaluating surfaceconcentrations relatively near the source. These mod-els typically require only an estimate of atmosphericstability, trajectory wind speed and direction, andemission rates. Fires are modeled independently.Therefore, accumulations of smoke from multiple firesare ignored. Some Western States require SASEMmodeling of prescribed burns before they can be per-mitted (Battye and Battye 2002).

Puff models (Draxler and Hess 1998; Harrison 1995;Hummel and Rafsnider 1995; Scire and others 2000a)are needed when simulating long-range transport, ortransport that occurs during changeable environmen-tal conditions such as influences from complex terrainor variable weather. NFSpuff has an easy user inter-face, but because of its internal terrain data files it isrestricted to applications in the Western States, ex-cluding Alaska (Harrison 1995). Hysplit (Draxler andHess 1998) currently is programmed to accept only 16individual sources and assumes a constant rate ofemissions with no plume rise. Hysplit (Draxler andHess 1998) and Calpuff (Scire and others 2000a) bothinclude simple chemistry. NFSpuff is the most com-monly used puff model for prescribed fire planning(Dull and others 1998). All three models are linked tothe MM5 meteorological model (Grell and others 1995).NFSpuff can function with a simple trajectory wind,and Hysplit and Calpuff can accept other griddedweather input data.

Particle models are used in coupled fire-atmospheremodeling (Reisner and others 2000) and for trackingcritical signature elements (Achtemeier 1994, 2000;Draxler and Hess 1998). The sophistication of thesetypes of models and their computational requirements,however, has thus far limited their application toresearch development or individual case studies.

Eulerian photochemical grid models are highly use-ful in estimating smoke concentrations from manysources over large domains. In addition, their abilityto model secondary chemical reactions and transfor-mations is needed for determining ozone concentra-tions and regional haze conditions. Regional planningorganizations such as the Western Regional Air Part-nership (WRAP), are evaluating the photochemicalmodels Models-3/CMAQ (Byun and Ching 1999) andREMSAD (Systems Applications International 2002)for use in guiding State implementation plans (SIPs)and Tribal implementation plans (TIPs).

Additional work is needed to fill critical gaps in themodeling systems identified above. As the need for


Traditionally, ozone and secondary aerosol precur-sors have been discussed within the context of urbansmog caused by auto exhaust and reactive organiccompounds emitted from industrial facilities. But thesame pollutant and tropospheric chemical reactionsoccur in both urban settings and in rural areas wherewildfire smoke may be an important if not dominantsource of ozone precursor emissions. In these situa-tions, emissions from fire may play an important rolein ozone formation as well as nitrate and, indirectly,sulfate aerosol formation, which results in visibilityimpairment and increased PM2.5 concentrations.

At present, there is an urgent need to understandthe impact of fire emissions on emerging visibility andambient air standards as they relate to fire planningat the strategic, programmatic, and operational scales(Fox and Riebau 2000; Sandberg and others 1999).Chemical processes that occur in plumes from fires,directly or indirectly, touch on a number of theseissues and are critical to the development of a regionalmodel that will be used to assess the impact of fire onair quality.

Because of the Environmental Protection Agency’s(EPA) pressing regulatory need to assess inter-Stateozone transport and sources of precursor emissions, anew regional-scale mechanistic model called Models-3/CMAQ (Byun and Ching 1999) is being used by theOzone Transport Commission (OTC) region of North-

Chapter 6: Atmospheric andPlume Chemistry

eastern and Mid-Western States, and the WesternRegional Air Partnership (WRAP). Future applica-tions will likely involve regional haze modeling inother areas of the country. Oxides of nitrogen (NOX)and volatile organic compounds (VOCs) emissionsfrom fire in the OTC have not previously been consid-ered significant, but the new model photochemistrymodule requires that precursor emissions be includedfor all sources. As Models-3/CMAQ develops, NOX andVOC emissions from fire will be included in ozone andsecondary modeling.

Ozone Formation in Plumes_______Field observations of ozone formation in smoke

plumes from fires date back nearly 25 years whenaircraft measurements detected elevated ozone at theedge of forest fire smoke plumes far downwind (Stithand others 1981). More recent observations (Wotawaand Trainer 2000) suggest that high concentrations ofozone are found in forest fire plumes that are trans-ported great distances and across international bound-aries. Measurements made during EPA’s 1995 South-ern Oxidant Study indicate that Canadian forest fireschanged the photochemical properties of air massesover Tennessee on days with strong fire influence.Regional background ozone levels were elevated by 10to 20 ppb on fire impact days as compared with


nonimpact days during the study. Aircraft measure-ments found that, although forest fire plumes werealways well defined with respect to carbon monoxide,they gradually lost their definition with respect toozone after being mixed into the boundary layer. Theamount of ozone transported to the surface measure-ment sites was found to depend upon where and whenthe plumes reached the ground. Elevated plumes werealways marked by enhanced ozone concentrations, attimes reaching values of 80 to 100 parts per billion(ppb) above tropospheric background.

Stith and others (1981) mapped ozone mixing ratiosin an isolated, fresh, biomass-burning plume. At thesource, or near the bottom, of the horizontally driftingplume they measured low or negative changes in ozonevalues, which they attributed to titration by NO andlow ultraviolet (UV) intensity. Near the top of theplume, 10 km downwind, and in smoke less than 1hour old, they measured change in ozone values ashigh as 44 parts per billion by volume (ppbv). Greaterchanges in ozone were positively correlated with highUV. Thus the initial destruction of ozone by reactivespecies in the plume followed by its gradual formationwas documented.

A new and potentially useful tool for assessingimpacts of long-range plume transport is based on theconcept of using ∆O3/∆CO (excess O3 over excess CO)as a “photochemical clock” to denote the degree ofphotochemical processing in a polluted air mass byusing carbon monoxide as a stable plume signature.As the plume disperses, its volume expands and abso-lute values of ozone can drop even though net produc-tion of ozone is still occurring. The ∆O3/∆CO normal-izes for plume expansion and is a useful measure of netozone production. In the course of atmospheric chem-istry research, numerous observations of ∆O3/∆COratios have been made in biomass burning haze layers.Unfortunately, the observations represent haze ofvarious ages and uncertain origin. In haze layers 1 to2 days old, changes in the ∆O3/∆CO ratios of 0.04 to0.18 were measured over Alaska (Wofsy and others1992) and ratios of 0.1 to 0.2 were measured overEastern Canada (Mauzerall and others 1996). Highratios, up to 0.88, were measured at the top of hazelayers that had aged about 10 days in the tropics(Andreae and others 1994).

In 1997, airborne Fourier transform infrared spectro-scopy (FTIR) measurements in large isolated biomassburning plumes in Alaska revealed new details ofdownwind chemistry. Downwind smoke samples thathad aged in the upper part of one plume for 2.2 ± 1hours had ∆O3/∆CO ratios of 7.9 ± 2.4 percent, result-ing from initial, absolute ozone formation rates of about50 ppb/hr. Downwind samples obtained well insideanother plume, and of similar age, did not have detect-able ∆O3, but did have ∆NH3/∆CO ratios about one-third of the initial value. ∆HCOOH/∆CO (formic acid)

and ∆CH3COOH/∆CO (acetic acid) usually increasedabout a factor of 2 over the same time scale in samplesfrom both plumes. NOX was below the detection limitin all the downwind samples. These data provided thefirst precise in-plume measurements of the rate ofO3/CO increase and suggested that this rate dependedon relative position in the plume. The apparentlyrapid disappearance of NOX is consistent with thesimilar early observation, and the drop in NH3 wasconsistent with a reaction with HNO3 to form ammo-nium nitrate, which is a NOX sink. Secondary sourcesof formic acid relevant to polluted air have beendescribed (Finlayson-Pitts and Pitts 1986). Jacob andothers (1992, 1996) discussed several gas-phase sourcesof acetic acid that could occur in biomass burningplumes. These experiments provide the first experi-mental indication of the approximate time scale ofsecondary organic acid production in actual plumes.

A large number of photochemical modeling studiesof biomass burning plumes have been published(Chatfield and Delaney 1990; Chatfield and others1996; Crutzen and Carmichael 1993; Fishman andothers 1991; Jacob and others 1992, 1996; Koppmannand others 1997; Lee and others 1998; Lelieveld andothers 1997; Mauzerall and others 1998; Olson andothers 1997; Richardson and others 1991; Thompsonand others 1996). Nearly all these studies concludethat the net production of ozone occurs either in theoriginal plume, or as a result of the plume mixing withthe regional atmosphere. Several studies have showna strong dependence of the final modeled results on thedetails of the post-emission-processing scenario suchas the timing between production of the emissions andtheir convection to the free troposphere (Chatfield andDelaney 1990; Jacob and others 1996; Lelieveld andothers 1997; Pickering and others 1992; Thompsonand others 1996).

Factors Affecting PlumeChemistry______________________

The specific chemical composition of the plume de-pends on many factors: the details of post-emissionatmospheric reactions including dilution rates, pho-tolysis rates, position within the plume, altitude, andsmoke temperature, which varies by time of day andcombustion stage. Equally important is the chemistryof the downwind air that mixes with the plume, whichcould be clean air or contain aged plumes from urbanareas or other fires. In addition, the physical aspects ofthe plume mixing are important. For example, at therelatively low temperatures typical of higher altitudesin the troposphere, peroxyacetyl nitrate (PAN) is astable molecule, which can be transported. At loweraltitudes, PAN can thermally decompose and rereleaseNOX. Nitric acid (HNO3) can also be an important,


transportable reservoir species for NOX at high alti-tudes but for a different reason. HNO3 has a narrowerabsorption cross-section at lower temperatures andtherefore is less susceptible to photolysis. The rate ofbimolecular reactions among smoke components usu-ally decreases with temperature (thus typically withaltitude or at night). Reaction rates depend even morestrongly on the dilution rate, at least initially. Dilutionby a factor of 2 will decrease a bimolecular reactionrate by a factor of 4.

Emission Factors for ReactiveSpecies________________________

Emission factors for hydrogen oxide (HOX, a collec-tive term for OH and HO2) precursors, NH3, and NOXhave been estimated with the Missoula, MT, open-pathspectroscopic system (Yokelson and others 1997). Theseexperiments reveal that smoke contains high levels ofoxygenated organic compounds, methanol (CH3OH),acetic acid (CH3COOH), and formaldehyde (HCHO).These compounds typically oxidize or photolyze withinhours in a smoke plume to release HOX that is impor-tant in sulfate aerosol formation processes. Underclear-sky conditions typical for noon on July 1 at 40°Nlatitude, the formaldehyde photolysis lifetime is about3.8 hours (Yokelson and others 1997). Since theHCHO/CO source ratio for fires is typically near 2percent, this process clearly injects large quantities ofHO2 into fresh plumes (Yokelson and others 1997).HOX emissions from fire may become a critical input toregional haze models that simulate secondary sulfateformation processes.

The H2O2 is soluble in cloud droplets where it wouldplay a major role influencing reaction rates duringaqueous-phase sulfate formation chemistry (NRC1993).

Particle Formation in Plumes______A number of processes are important in plume par-

ticle formation and growth. Many of these processesinvolve interaction with the trace gases in a plumeoriginating from nucleation in which two gases reactto form a solid nucleus for subsequent particle growth.An example of nucleation is the reaction of ammoniaand nitric acid. In addition, condensation can createnew particles when gases cool or through particlegrowth when a trace gas collides with and condenseson an existing particle. The second condensation pro-cess is quite common because biomass burning aerosolis hydrated. Soluble nucleilike ammonium nitrate

promotes this process. There is a little evidence thatorganic gases also condense on particles. Nucleationand condensation are both examples of trace-gas-to-particle conversion, which will increase the mass ofparticles in a plume, decrease the concentration ofcertain trace gases in the plume, and, in the case ofcondensation, contribute to an increase in averageparticle diameter. Andreae and others (1988) mea-sured particle-NH4

+/CO2 ratios of 0.7 to 1.5 percent inslightly aged biomass burning plumes. Measurementsof NH3/CO in fresh smoke are typically near 2 percent.Thus, there is probably rapid conversion of gas-phaseNH3 to particle NH4

+ either through nucleation ordissolution in the surface water of other hydratedparticles.

Coagulation is when two particles collide and com-bine. This increases the average particle diameter,reduces particle number, and does not effect totalparticle mass. Coagulation probably contributes to theincrease in average particle diameter that occurs down-wind from fires (Reid and others 1998).

At any given point in its evolution a particle mayimpact the trace gas chemistry in a smoke plume. Forinstance, it is known that NO2 reacts on the surface ofsoot particles to yield gas phase HONO. This and otherheterogeneous reactions such as ozone destructionmay occur on smoke aerosol. Some recent researchsuggests that oxygenated organic compounds emittedfrom fires could also be important in heterogeneousprocesses. Hobbs and Radke (1969), Desalmand andothers (1985), Andreae and others (1988), and Rogerand others (1991) found that a high percentage (25 to100 percent) of fire aerosol particles from fires could beactive as condensation nuclei (CCN). Radke and oth-ers (1990) observed that cumulus clouds greater than2 km in depth scavenged 40 to 80 percent of smokeparticles. The high concentrations of CCN in smokeplumes can contribute to the formation of clouds withsmaller than “normal” cloud droplet size distributions.This type of cloud is more reflective to incoming solarradiation and less likely to form precipitation. Somework suggests that absorbing aerosol can reduce cloudformation. Finally, clouds can evaporate and leavebehind chemically altered particles.

All of these mechanisms alter both the chemicalnature and number of particles contained within smokeplumes from fires. In addition, reactive species emit-ted from fires (see previous section) may alter theconversion rate of gaseous precursors of secondarysulfate and nitrate particles, affecting regional hazemodeling results.

Although the regulatory implications of reactive spe-cies emissions from fire are yet to be determined, muchmore attention to these issues will occur once fire isincluding in regional haze and ozone modeling efforts.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


Chapter 7: Estimating the AirQuality Impacts of Fire

State-of-the-science methods used to determine theimpact of fire on air quality and visibility include: (1)emission inventories; (2) air quality monitoring in-struments to measure smoke concentrations in real-time; and (3) filter-based monitoring techniques andreceptor-oriented methods that quantify wildfire smokecontribution to air pollution based on the chemicalcharacteristics of smoke particles or the spatial and/ortemporal variability. Fire also contributes to ground-level ozone. These topics have become increasinglyimportant to both air quality regulators and landmanagers as efforts to identify, or apportion, thecontributions that fire makes to particulate air pollu-tion, regional haze, and ground-level ozone come un-der increased scrutiny.

Because the health effects of air pollution are sodifficult to measure in the broad population, there hasbeen little effort to regulate or manage those effectsdirectly. Many smoke management decisions are madeon the basis of nuisance complaints as an indicator,rather than on quantitative measurements of impactsto health and welfare. Close to the source, efforts arebeing made to keep the exposure of firefighters tohazardous air pollutants within the standards set bythe Occupational Safety and Health Administration.Hazard assessment describes the nature, concentra-tion, and duration of pollutants. Exposure assessmentquantifies the population exposed and the degree of

exposure. Risk assessment describes the probableresult for a population from all exposures. Integratedhealth risk assessments and economic assessmentsare still rare.

Modeling and data systems are needed to predict,measure, and monitor the ultimate effects of air pollu-tion from fires on human or ecosystem health, on theeconomy, and on the comfortable enjoyment of life andproperty. Risk assessment methods are needed tocompare these effects with those from other sources.

Emission Inventories ____________An emission inventory is an estimate of the mass of

emissions by class of activity within a specified geo-graphic area in a specified amount of time. Usually, aninventory is compiled by multiplying the appropriateemission factor (see chapter 4) by the estimated levelof activity (in other words, tons of fuel consumed).

Development of emission inventory methods forfires was recently reviewed in detail by Battye andBattye (2002). The report considers prior attempts atemission inventory, describes approaches to estimat-ing emissions from fires, and reviews the scientificinformation available as components of an inventory.The report also reviews emission reduction strategiesand smoke management techniques.


An emission inventory provides an understandingof the relative burden on the air resource from par-ticular air pollution source categories. Emission in-ventories help explain the contribution of sourcecategories to pollution events, provide backgroundinformation for air resource management, providethe means to verify progress toward emission reduc-tion goals, and provide a scientific basis for State airprogram development. An accurate emission inven-tory provides a measured, rather than perceived,estimate of pollutant production as the basis forregulation, management action, and program com-pliance. Emission inventories should include all im-portant source categories including mobile, area, andstationary, and the inventories are not completeunless difficult-to-quantify sources such as agricul-tural burning, backyard burning, rangeland burn-ing, and wildland and prescribed burning are ad-dressed. Emission inventories are a basic requirementof State air resource management programs and area required element of State implementation plans(SIPs). Emission inventories are also compiled annu-ally at the national level and for specific geographicregions (sub-State, multi-State, or multi-jurisdic-tion) to address a particular regional air qualityissue.

The science necessary to accurately estimate emis-sions from prescribed burning is quite good for mostfuel types in the United States if good quality informa-tion about several critical variables is known. Areaburned, fuel type, fuel loading, fuel arrangement, fuelconsumption, and emission factors are all needed toaccurately estimate emissions. Some of these requireonsite reporting for reasonable accuracy includingarea burned, fuel type, and fuel arrangement. Otherfactors can be defaulted or estimated with reasonableaccuracy if some other information is known. Fuelloading can be defaulted with knowledge of the fueltype and arrangement. Fuel consumption can be cal-culated with knowledge of the fuel type, fuel loading,and fuel moisture. Emission factor assignment is madewith knowledge of the fuel type.

The science of predicting emissions from wildlandfire is much weaker than for prescribed fire. In addi-tion, it is generally far more difficult to obtain decentquality information about individual wildland fires.

In most cases, the information gap that makes fireemissions prediction a difficult endeavor is good qual-ity, consistent, and regular reporting of the specificonsite variables needed for emissions estimation. Datacollection systems that are supported and utilized byfire managers need to be developed for every Statewhere a reasonable estimate of prescribed fire emis-sions is desired. Data collection for wildfire emissionsestimation will be more difficult because some of theneeded information is not currently available in a way

that is compatible with emissions estimation require-ments. For example, a single wildfire often burnsthrough many different fuel types, but current report-ing requirements request the fuel type at the point ofignition. This fuel type may or may not be representa-tive of the majority of acres burned in the wildfire.Also, acres burned in wildland fires may be the areawithin the fire perimeter rather than the actual acresblackened by fire as is needed for emissions estima-tion. Similarly, the area reported as burned in pre-scribed fires is often the area authorized for burningwhether or not the entire burn was completed.

State Emission Inventories

High quality Statewide inventories of daily emis-sions from prescribed fire have been developed byOregon and Washington since the 1980s (Hardy andothers 2001). Eleven other States (Alabama, Alaska,Arizona, California, Colorado, Florida, Idaho, Mon-tana, Nevada, South Carolina, and Utah) estimateannual prescribed fire emissions from records of acre-age burned by fuel type and fuel loading at the end ofthe burning season. Many other States (such as Michi-gan, New Mexico, and Tennessee) currently have noannual reporting program.

No State has a reporting system for wildland firesthat is based on actual, reported data from individualwildland fires events. Any estimate a State may haveof wildland fires emissions is based on gross assump-tions about fuel loading and consumption, and on anarea-burned figure that may systematically overesti-mate the true value.

Regional Emission Inventories

Several recent regional inventories compiled insupport of regional haze program development haveshown new approaches to fire emission inventorydevelopment.

The Fire Emissions Project (FEP) calculated anemissions inventory for 10 Western States for a cur-rent year (1995) using actual reported data, plus twofuture years (2015 and 2040) using manager projec-tions of fire use. Fourteen vegetative cover types werechosen to characterize the range of species types withinthe 10-State domain. Within each vegetative covertype, up to three fuel loading categories (high, me-dium, and low) could be specified by field fire manag-ers. Fuel consumption calculations relied on expertestimates of fuel moisture believed to be most fre-quently associated with a particular type of burning.Emission factors were assigned based on the vegeta-tive cover type. The FEP inventory was used duringthe Grand Canyon Visibility Transport Commission(GCVTC) effort to apportion sources of visibility im-pairment in the Western States.


The GCVTC also sponsored the development of awildland fire emissions inventory for the period 1986through 1992. The GCVTC wildfire inventory includedonly wildland fires greater than 100 acres in size(capturing approximately 98 percent of the acreageburned). The variability of wildland fire emissions,which ranged from 50,000 tons per year of PM2.5 tomore than 550,000 tons per year over the 7 yearsstudied, indicates the difficulty in selecting a single 1year period that is representative of “typical” fireemissions (GCVTC 1996a).

In 1998, analysts at the Forest Service’s MissoulaFire Sciences Laboratory, Rocky Mountain ResearchStation, used the FEP management strategies withnew, additional data to estimate emissions from wild-land fires in the Western States (Hardy and others1998). This inventory of potential emissions used asuite of new or improved spatial data layers, includingvegetation/cover type, ownership, fuel and fire charac-teristics, modeled emissions and heat release rates,and fuels treatment probability distributions. Theseinventories are included in the Environmental Protec-tion Agency’s (EPA) National Emission Inventory(NEI).

Wildland fire frequency and occurrence are highlyvariable in time and space (fig. 7-1). The impact ofwildland fire smoke on Class I area visibility is alsoexpected to be highly variable from year to year withepisodic air quality and visibility impact events thatare difficult to predict. Seasonal impacts may be manytimes higher than annual averages.

National Emission Inventories

National emission inventories for prescribed firehave been compiled and reported by several investiga-tors (Chi and others 1979; Peterson and Ward 1992;Ward and others 1976; Yamate and others 1975). Ofthese, only the Peterson and Ward inventory of par-ticulate matter and air toxic emissions from pre-scribed fires during 1989 is still useful today, despitethe inconsistencies in the information available tocompile the emission estimates. The poor data collec-tion and inconsistent or nonexistent reporting sys-tems in use at the time of the 1989 inventory continuetoday.

Improving Emission Inventories

Significant barriers to compiling better regionalinventories include:

• Varying degrees of availability and number ofrecords describing burning activity over mul-tiple States, multiple agencies, ownerships,and Tribes.

• Lack of a national wildland fuel classificationsystem with spatial attributes.

• Limited and inappropriate modeling of fuelconsumption and emission characterizationfor prescribed burning in natural fuels.

Sandberg and others (1999) describe remedies toovercome some of the limitations of data collection andavailability. These remedies are intended to guide

Figure 7-1—Number of acres burned by wildfire between 1960and 2000 (National Interagency Fire Center 2002).


future inventory development efforts. Significantly,these remedies include adoption of standardized burnreporting protocols to be used by all agencies, Tribes,and ownerships to report daily emissions for eachburn, location of the burn, and many other param-eters.

The Fuel Analysis, Smoke Tracking, and ReportAccess Computer System (FASTRACS) is a sophisti-cated system developed by the Forest Service andBureau of Land Management in the Pacific North-west. FASTRACS tracks all the information neededfor accurate estimation of emissions from Federal useof prescribed fire in Washington and Oregon includingthe ability to track use of emission reduction tech-niques. As long as field fire managers are doing areasonable job of reporting the information requiredby FASTRACS, this system provides excellent emis-sions calculation capabilities and the best data report-ing standards in the country. Currently other land-owners, such as State and private, are not usingFASTRACS in Washington and Oregon although thereis an effort under way to bring them into the system.FASTRACS is also being looked at by other regionsand may be adopted or emulated across the country.For more information about FASTRACS, see http://www.fs.fed.us/r6/fastracs/index.htm.

Another data reporting system is under develop-ment in California. The Prescribed Fire InformationRetrieval System (Cal/PFIRS) is a centralized elec-tronic database that allows all users immediate accessto detailed information on burns on a day-to-day basis.Cal/PFIRS does not include the kind of detailed re-porting of information that could be used to assessthe use of emission reduction techniques but doesprovide a reasonable estimate of the amount of burn-ing taking place. For more information on Cal/PFIRS,see http://www.arb.ca.gov/smp/progdev/techtool/pfirs.htm.

Research since about 1970 has significantly im-proved the completeness and accuracy of emissioninventory techniques. However, the science is beingpressed forward because of new demands for regionalscale emission transport information needed to assessthe impact of wildland smoke on PM2.5 air qualitystandards and regional haze. Because of new air regu-latory demands, emission inventories, when used inconcert with regional models, have become an impor-tant means of apportioning fire smoke impacts on airresources.

Air Quality Monitoring ___________Unlike emission inventories, air quality monitors

determine actual pollutant loading in the atmosphereand are therefore the most direct measure of airquality on which air regulatory programs are based.Samples of particulate matter in the atmosphere (PM10

or PM2.5, or both) are also used for source apportion-ment purposes to identify the origin of the aerosols.Monitoring of smoke from fires, however, presentsseveral unusual technical challenges that affect re-sults. These challenges center on the fact that smokefrom fires has several unique characteristics.

Current Monitoring Techniques

The three principal methods of measuring air pollu-tion are samplers, optical instruments, and electro-chemical devices. Samplers are most common for long-term monitoring. Data from optical meters andelectrochemical devices can be stored in a computer ordatalogger on site or transmitted from remote loca-tions to provide real-time information.

Samplers—Samplers collect aerosols on a filter orchemical solution. A simple gravimetric measure ofmass concentration may be obtained, or different typesof filters or solutions can be used, to help definechemical species and particle sizes. For chemical spe-ciation, filters must be sent to a laboratory for analy-sis. For this reason, sampling information usually isdelayed by days to weeks after the sampling period.Active samplers are the most accurate as they use apump to pass a known volume of air through thecollector. Passive samplers are the least expensive,allowing air to reach the collector by some physicalprocess such as diffusion. Tapered Element Oscilla-tion Microscales (TEOMs) are a special class of sam-plers that provide a gravimetric measure of massconcentration at the studied site without having totransport filters to a laboratory.

All sampling devices lose some degree of semivolatilefine particulates (Eatough and Pang 1999). Positiveand negative organic carbon artifacts are just two ofseveral factors that contribute to variability betweendifferent colocated instruments. To minimize thisvariability, consistent sampling methods are usedthroughout a sampling network to help recognizesuch artifacts.

The analytical technique used to quantify carbonconcentrations from filters also can cause discrepan-cies between measurements (Chow 2000). For ex-ample, the NIOSH 5040 method (Cassinelli andO’Conner 1994) is a thermal-optical transmittancemethod of speciating total, organic, elemental, andcarbonate (inorganic) carbon being adopted by theEPA’s PM2.5 program. This method is a departurefrom the thermal-optical reflectance method that hasbeen used in the IMPROVE program. Recent compari-sons between ambient samples have identified differ-ences as great as 17.5 ± 15 percent (EPA 2000a), whichcan be significant when monitoring for National Am-bient Air Quality Standards (NAAQS) violations.


Because filters can become overfull, they must bechanged regularly and are not suitable for sites closeto fires where particulate concentrations are heavy.

Optical Instruments—Optical instruments use alight source to measure the atmosphere’s ability toscatter and absorb light. Common devices are photom-eters, which measure the intensity of light, and trans-missometers, which are photometers used to measurethe intensity of distant light. Photometers and trans-missometers have a direct relation to visual range.Nephelometers measure the scattering function ofparticles suspended in air. They can be used to deter-mine the visual range, as well as the size of thesuspended particles, by changing the wavelength ofthe light source. Wavelengths of 400 to 550 nm arecommon for monitoring smoke from biomass fires,while wavelengths of 880 nm are more common forroad dust measurements. Because the instrumentshave increasing application for both long-term andreal-time monitoring of smoke, Trent and others (2000)evaluated the accuracy of several different opticalinstruments by comparing their output to gravimetricsamples.

Investigators have found some problems in fieldreliability and temperature drift among photometersand nephelometers (Trent and others 1999, 2000).While Davies (2002) recommends a general coefficientfor relating scattering coefficient to drift smoke from aDataRAM nephelometer, a precise relation between anephelometer’s measured scattering coefficient andparticle concentration depends on the wavelength ofthe instrument and the particle distribution of themedium, which varies by combustion stage and fueltype.

Electrochemical Devices—Electrochemical de-vices have been used in industrial applications formany years. Their small size and ability to measurecriteria pollutants, such as carbon monoxide, makethem suitable for personal monitoring or monitoringin extremely remote locations. Thus, they are gainingvalue for monitoring wildland smoke impacts. Forexample, Reinhardt and Ottmar (2000) recommendthe use of an electrochemical dosimeter for monitoringexposure levels experienced by wildland fire fighters(Reinhardt and Ottmar 2000).

States, Tribes, and local air agencies use a variety ofinstruments to monitor long-term and real-time smokeimpacts for both NAAQS and visibility to suit theirlocal interests and regulatory needs. The InteragencyMonitoring of Protected Visual Environments (IM-PROVE) program is one of few nationally coordinatedmonitoring projects.

IMPROVE was established in 1985 in response tothe 1977 amendment of the Clean Air Act requiringmonitoring of visibility-related parameters in Class I

areas throughout the country (fig. 7-2). The IMPROVEnetwork uses a combination of speciation filters onactive samplers to measure physical properties ofatmospheric particles (PM2.5 and PM10) that arerelated to visibility. Many sites also include transmis-someters and nephelometers optical devices. Also,cameras are used document the appearance of scenicvistas. Because the samplers collect for 24 hours every3 days, their information is used for determining long-term trends in visibility. The optical and cameradevices can monitor more frequently and can helpdefine short-term or near real-time changes in visibil-ity impact

Source Apportionment ___________Most air monitoring programs are designed to mea-

sure particulate mass loading to provide data forPM10 and PM2.5 NAAQS and visibility. Becausethese sizes of particles can come from many sources,they are not useful for apportioning to one source oranother. While the IMPROVE program provides spe-ciated aerosol data that are helpful in source attribu-tion analysis, the averaging periods of samples andsparse location of sites make IMPROVE measure-ments difficult to use for source attribution withoutsupplemental measurements or modeling tools.

Wotawa and Trainer (2000) found that 74 percent ofthe variance in the average afternoon carbon monox-ide levels could not be attributed to anthropogenicsources during the 1995 Southern Oxidant Study(Chameides and Cowling 1995). Analysis of weatherpatterns indicated that transport of wildland firesmoke from Canada could explain the elevated carbonmonoxide levels. Also, they discovered a statisticallysignificant relationship between the elevated carbonmonoxide and ground-level ozone concentrations.

Characterization of organic carbon compounds foundwithin the organic carbon fraction of fine particulatematter coupled with inclusion of gaseous volatile or-ganic compounds (VOCs) holds substantial promise inadvancing the science of source apportionment (Watson1997). The key to the use of chemical mass balancemethods is the acquisition of accurate data describingthe chemical composition of both particulate matterand VOCs in the ambient air and in emissions fromspecific sources. Several organic compounds unique towood smoke have been identified including retene,levoglucosan, thermally altered resin, and polycyclicaromatic hydrocarbons (PAH) compounds. These com-pounds are present in appreciable amounts and can beused as signatures for source apportionment if specialprecautions are taken during sampling to minimizelosses (Standley and Simoneit 1987). Inclusion ofthese aerosol and VOC components in the speciation


analysis appears worthwhile but would increase moni-toring and sample analysis costs.

Source Apportionment Methods

Apportionment of particulate matter mass to therespective contributing sources is done through bothmechanistic models (dispersion models) and receptor-oriented techniques that are based on the characteris-tics of the particles collected at the receptor. The bestapproach is through the use of both techniques, ap-plied independently, to develop a “weight of evidence”assessment of source contributions of smoke from fire.A third approach is through the use of visual andphotographic systems that can document visibilityconditions over time or track a plume from its sourceto the point of impact within a Class I area.

Figure 7-2—IMPROVE monitoring network in 1999 (http://vista.cira.colostate.edu/improve/Overview/IMPROVEProgram.htm).

Receptor-Oriented Approaches

Receptor-oriented approaches range from simplesignature applications to complex data analysis tech-niques that are based on the spatial, temporal, andchemical constituents (“fingerprint”) of various sources.

Simple signature applications for smoke from fireare based on chemically distinct emissions from fire.For example, methyl chloride (CH3Cl) is a gas emittedduring wood combustion that has been used in thismanner to identify impacts of both residentialwoodstove smoke and smoke from prescribed fires(Khalil and others 1983).

Speciated Rollback Model —The speciated roll-back model (NRC 1993) is a simple hybrid model thatuses aerosol data collected at the receptor with emis-sion inventories to estimate source impacts. It is a


The limitations of the speciated rollback model areseveral:

• Deviations from the assumption of spatiallyhomogeneous emissions are likely to occurwhen air quality is most critical at a singlereceptor where a single emission source canhave an inordinate impact.

• Secondary particle formation is assumed to belinear to changes in precursor emissions.

• Meteorological conditions do not change fromyear to year.

• Emission inventory errors have a direct, pro-portional effect on the model estimates.

The model can be applied to any temporal concentra-tion such as annual average, worst 20th percentile, orworst daily average scenarios in any region that meetsthe constraint on the spatial distribution of emissionchanges. It is straightforward, necessary input dataare available, and the model assumptions are easilyunderstood. It makes use of chemical speciation datacollected from the IMPROVE network but cannotapportion contributions made from source classes notincluded in the inventory.

Chemical Mass Balance Model—The chemicalmass balance model, CMB7 (Watson 1997; Watsonand others 1990), infers source contributions based onspeciated aerosol samples collected at a monitoringsite. Chemical elements and compounds in ambientaerosol are “matched” to speciated source emissionprofiles “fingerprints” by using least-squares, linearregression techniques to apportion the aerosol mass.CMB7 has been widely used within the regulatorycommunity to identify and quantify the sources ofparticles emitted directly to the atmosphere. The modelis based on the relationship between characteristics ofthe airborne particle (ci), the summation of the productof the ambient mass concentration contributed by allsources (Sj), and the fraction of the characteristiccomponent in the source’s fingerprint (fij).

ci = ∑jSjfij (2)

Given detailed information about the chemical spe-ciation of the ambient aerosol and similar informationabout all of the emission sources impacting the recep-tor, the CMB7 model can apportion the aerosol massamong the sources if certain assumptions are met.

To minimize error, there must be more aerosolcomponents than sources to be included in the least-squares linear regression fit. If there are more compo-nents measured than sources, then the comparison ofmodel-estimated concentrations of these additionalcomponents provides a valuable internal check onmodel consistency.

spatially averaged model that disaggregates majorparticle components into chemically distinct groupsthat are contributed by different types of sources. Alinear rollback model is based on the assumption thatambient concentrations (C) above background (Cb) aredirectly proportional to total emissions in the region ofinterest (E):

C – Cb = kE (1)

The proportionality constant, k, is determined overa historical time period when both concentrations Cand Cb as well as regional emissions E are known.Once k is determined, new concentration estimatescan be derived for other emission levels of interestassuming that meteorological conditions are constantover the same averaging time. Because the anthropo-genic components in the particle mass consist almostentirely of sulfates, nitrates, organic carbon, elemen-tal carbon, and crustal material, a maximum contribu-tion from fire can be made based on the assumptionthat all of the organic carbon or elemental carbon isfrom primary fire emissions. Various complexities canbe added to this model; components can be disaggre-gated by particle-size fraction (coarse versus fine par-ticles) as well as by chemical composition. Additionaldistinctions can be made between primary and sec-ondary particles, and nonlinear transformation pro-cesses can be approximated to account for atmosphericreactions.

Simple proportional speciated rollback models re-quire data on the chemical composition of airborneparticles, knowledge or assumptions regarding sec-ondary particle components, an emission inventoryfor the important source categories for each particlecomponent and each gaseous precursor, and knowl-edge or assumptions regarding background concen-trations for each component of the aerosol and eachgaseous precursor.

The speciated rollback model was applied by theNRC Committee on Haze in National Parks and Wil-derness Areas to apportion regional haze in the threelarge regions of the country (East, Southwest, andPacific Northwest) by including extinction coefficientsto the estimated mass concentrations (NRC 1993). Thepercentage of anthropogenic light extinction appor-tioned to forest management burning was estimatedat 11 percent in the Northwestern United States on anannual basis assuming that about one-third of themeasured organic carbon is of natural origin. The 1985National Acid Precipitation Assessment Program(NAPAP) inventory was used in this analysis, whichalso assumed that the elemental carbon and organiccarbon fractions of the PM2.5 emissions for forestmanagement burning were 6 percent and 60 percent,respectively.


The chemical components in the source “finger-print” must be conserved and not altered during atmo-spheric transport — a rather large limitation.

Model resolution is typically limited to five or sixsource types, and separation of two sources with simi-lar emission profiles (for example, prescribed burningand residential woodstove smoke) is difficult if bothsources are active at the same time.

Systematic error analysis procedures have beendeveloped for the CMB7 model, and the results havebeen published in model validation studies (NRC1993). However, the model cannot apportion second-ary aerosols (sulfate and nitrate); it is limited in itsability to apportion all of the mass to specific sources.

The ability of the model to apportion smoke from firedepends on several factors:

• The presence or absence of smoke from otherforms of vegetative burning (woodstoves, agri-cultural burning, open burning, and others).

• The magnitude of the smoke impact at thereceptor (for example, well-dispersed smokethat contributes small amounts of aerosol massis more difficult to distinguish).

• The uncertainty in both the ambient aerosoland the source “fingerprint” components thatthe model most heavily weights in the regres-sion analysis, typically organic carbon, potas-sium, and elemental carbon. The greater theuncertainty of these measurements, the less“fitting pressure” they have in influencing theregression solution.

• Inclusion of multiple aerosol components thatare as nearly unique to smoke from fires (en-demic signatures) as possible. These includeorganic compounds such as retene andlevoglucosan, as well as gaseous signaturesuch as carbon monoxide and methyl chloride.The more the source profile distinguishes pre-scribed or wildland fire smoke from othersources, the more accurate the source appor-tionment is likely to be.

Factor Analysis and Multiple LinearRegression

When many ambient samples are available, linearregression and factor analysis techniques can be ap-plied to the dataset to obtain empirical insights intothe origin of the particles. Factor analysis is based onthe assumption that chemical components of the aero-sol that covary are emitted from a common source.Cluster patterns can then be matched to the sourceprofiles of known sources to identify the degree ofcovariance associated with a specific source category.

Source profiles can be recovered from the ambientdata by using two special forms of factor analysis(VARIMAX rotation) or, when the profiles are approxi-mately known, target transformation factor analysis(Hopke 1985). Factor analysis can therefore serve torefine the source profile information used in chemicalmass balance analysis. In the context of wildfire smokeapportionment, investigators have historically lookedfor a high degree of covariance between organic car-bon, elemental carbon, and potassium (total, watersoluble, and/or nonsoil potassium) as the cluster com-ponents that signal particles emitted from vegetativeburning of all kinds. Unfortunately, these componentsof the aerosol are not necessarily unique to smoke fromvegetative burning.

Linear regression analysis is a well-established sta-tistical procedure for estimating unknown coefficientsin linear relationships where a large dataset of obser-vations of both the dependent and independent vari-ables are present. In the terminology of regressionanalysis, c in equation (2) is the variable, Sj is theindependent variable, and fj the regression coeffi-cients. In practice, the independent variable is takento be proportional to source strength rather than thesource strength themselves. Multiple linear regres-sion has been widely used to apportion total particlemass, the most common approach being use of signa-ture concentrations taken directly as the independentvariable, fj. Significantly, gaseous pollutant data canbe included in the regression to increase the model’sability to resolve sources. Although carbon monoxidewould greatly enhance the success of the model as it isemitted by wildfires in large quantities and is stable inthe atmosphere, carbon monoxide is not routinelyincluded in nonurban monitoring programs.

Regression analysis has been used successfully toapportion the total carbon portion of the aerosol massbetween wood smoke, vehicle exhaust, and othersources by using nonsoil potassium . The regression-derived estimates were then validated by 14C isotopeanalysis, which is a direct indicator of “contemporary”versus fossil fuel carbon sources. The 14C measure-ments nicely confirmed the source apportionment re-sults by regression analysis (r=0.88) (NRC 1993).

Summary

Receptor-oriented methods of particle mass sourceapportionment have proven successful in a large num-ber of urban studies worldwide. A number of thesestudies have attempted to apportion wildfire smoke onthe basis of a set of aerosol and source emission traceelements and compounds. The experimental design ofthese studies has limited the ability of receptor modelsto resolve wildfire smoke from other sources. With


improvements in speciation of the organic carboncomponent of the aerosol, and inclusion of carbonmonoxide, methyl chloride, and other endemic signa-tures, the ability of these techniques to resolve sourcesand minimize uncertainties will increase. Sensitivitystudies are needed to determine which additionalcomponents beyond the standard array of trace ele-ments, ions, and carbon fractions would be most ben-eficial to include in future monitoring programs.

Mechanistic Models _____________As noted in chapter 6, multiple dispersion models

have been used to estimate air quality impacts ofsingle or multiple fires at local and regional scales.Eulerian regional-scale models have been principallyused for source apportionment application both toestimate contributions to particulate air quality andregional haze. The suitability of such models for ap-portionment applications largely depends on the com-pleteness and accuracy of the emission inventory in-puts used by the model. Unfortunately, few fieldvalidations are available.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


Chapter 8: Consequences ofFire on Air Quality

The potential impacts of fire-induced degradation ofair quality on public health and welfare range fromoccupational exposure of smoke on firefighters tobroader economic and social impacts and highwaysafety.

Health Effects __________________

National Review of Health Effects

In 1996, the Environmental Protection Agency (EPA)conducted an extensive review of the science relatinghuman health effects to particulate matter (PM), theprincipal pollutant of concern from fires (EPA 1996).The review found that (1) epidemiological studiessuggest a variety of health effects at concentrationsfound in several U.S. cities and (2) ambient particles ofgreatest concern to health were those smaller than 10micrometers in diameter. Results of efforts to trace thephysiological and pathological responses of the body toPM are unclear, and demonstration of possible mecha-nisms linking ambient PM to mortality and morbidityare derived from hypotheses in animal and humanstudies. It is known, however, that PM produces physi-ological and pathological effects by a variety of mecha-nisms, including:

• Increased airflow obstruction by PM-inducednarrowing of airways.

• Impaired clearance of lung pathways causedby hypersecretion of mucus caused by PMexposure.

• Lung responses to PM exposure includinghypoxia, broncho-constriction, apnea, impaireddiffusion and production of inflammatorymediators.

• Changes in the epithelial lining of the alveolarcapillary membrane that increase the diffu-sion distances across the respiratory mem-brane, thereby reducing the effectiveness ofblood gas exchange.

• Inflammatory responses that cause increasedsusceptibility to asthma, chronic obstructivepulmonary disease (COPD) and infections.

Recent information also suggests that several sub-groups within the population are more sensitive to PMthan others. Children are more likely to have de-creased pulmonary function, while increased mortal-ity has been reported in the elderly and in individualswith cardiopulmonary disease. Asthmatics are espe-cially susceptible to PM exposure. In addition, coarse(2.5 to 10µm) particles from road dust or windblownsoil were found to have less toxicity than fine particles(less than 2.5µm) that include acid aerosols, diesel


emissions, smoke from fires, and potentially carcino-genic PAH compounds.

Occupational Exposure to Wildland FireSmoke

Wildland firefighters and fire managers have longbeen aware that smoke exposure occurs during theirwork (Reinhardt and Ottmar 1997; Sharkey 1997).Although the long-term health effects from occupa-tional smoke exposure remain unknown, the evidenceto date suggests that brief, intense smoke exposurescan easily exceed short-term exposure limits in peakexposure situations such as direct attack and holdingfirelines downwind of an active wildfire or prescribedburn. Shift-average exposure only occasionally exceedsrecommended instantaneous exposure limits set by theAmerican Conference of Governmental Industrial Hy-gienists (ACGIH), and rarely do they exceed Occupa-tional Safety and Health Administration (OSHA) timeweighted average (TWA) limits (fig. 8-1) (Reinhardtand Ottmar 2000; Reinhardt and others 2000). Overex-posure increases to 10 percent of the time if the expo-sure limits are adjusted for unique aspects of the firemanagement workplace; these aspects include hardbreathing, extended hours, and high elevations, allfactors which intensify the effects of many of the healthhazards of smoke (Betchley and others 1995; Maternaand others 1992; Reinhardt and Ottmar 2000; Reinhardtand others 2000). It could be argued that few firefightersspend a working lifetime in the fire profession, and thusthey should be exempt from occupational standards

that are set to protect workers over their careers. Butthis argument is irrelevant for irritants and fast-actinghealth effects such as eye and respiratory irritation,headache, nausea, and angina. An exposure standardspecifically for wildland firefighters and appropriaterespiratory protection may need to be developed(Reinhardt and Ottmar 2000).

In spite of the studies that have been done, majordata gaps remain:

• In the area of health hazards, not enoughevidence is available to defend the commonlycited “inert” classification of total and respi-rable particulate in dust and smoke; there islittle knowledge of the occurrence of crystal-line silica in dust at fires; and there is incom-plete characterization of aldehydes and otherrespiratory irritants present in smoke(Reinhardt and Ottmar 1997, 2000).

• The differences in smoke exposure betweenlarge and small wildland fires have not beencharacterized in spite of the fact that one ortwo crews extinguish the vast majority ofwildfires (Reinhardt and Ottmar 2000).

• The long-term health experience of wildlandfirefighters is unknown, although anecdotalreports and the biological plausibility of cu-mulative health effects indicate a potentiallygreater incidence of disease and death than inthe general population of workers (Booze andReinhardt, in press; Sharkey 1997).

Although data gaps remain, enough information hasbeen gathered to chart a course to alleviate many ofthe overexposures. Respiratory protection is availablefor irritants such as aldehydes and particulate matterbut not for carbon monoxide. Respirators can be heavy,hot, and impede the speed of work, but some newmodels are light, simple and could be worn only whenneeded (Beason and others 1996; Rothwell and Sharkey1995). The entire costly process of medical evalua-tions, fitness testing, maintenance, and training mustbe employed if respirators are to be used. But there areimmediate benefits to reducing respiratory irritantexposure. Small electrochemical dosimeters can pro-vide instant warnings about carbon monoxide levels ina smoky situation, and fire crew members equippedwith respirators and carbon monoxide monitors haveall the protection necessary to stay and accomplishobjectives safely and withdraw when the carbon mon-oxide levels become the limiting factors (Reinhardtand others 1999). In the future, a respirator for useduring wildland fires may be developed that offerswarning and protection against carbon monoxide aswell. Although some work has been done in this area,we need more significant development. Smoke expo-sure is a hazard only a small portion of the time and is

Figure 8-1—Firefighters being monitored for smoke exposure.Monitoring equipment seen includes a red backpack that col-lects gas samples from the breathing zone of the firefightersand a white-colored particulate matter filter sampler attached tothe chest. (Photo by Roger Ottmar)


manageable because the situation where it occurs canbe predicted. A long-term program to manage smokeexposure at wildland fires could include (1) hazardawareness training, (2) implementation of practices toreduce smoke exposure such as rotating crews andproviding clean air sites, (3) routine carbon monoxidemonitoring with electronic dosimeters, (4) improvedrecordkeeping on accident reports to include separa-tion of smoke related illness among fireline workersand fire camp personnel, and (5) improved nutritionaland health habits. Fire management practices such ascrew rotation, awareness training, and carbon monox-ide monitoring can mitigate the hazard and allowfirefighters to focus on the job of fire management,lessening the distraction, discomfort, and health im-pacts of smoke exposure (Reinhardt and Ottmar 2000).

Research Issues

A number of wildland fire health effect researchissues flow from the EPA staff report (Clean AirScientific Advisory Committee1995) and occupationalhealth exposure studies.

Research into the health effects of particulate mat-ter is largely based on epidemiological studies con-ducted over long periods in urban centers with highhospital admittance or large air quality databases, orboth. Consequently, inadequate information is avail-able that relates short-term, acute smoke exposure(such as would be experienced by a visitor to aNational Park or to a community near a wildfire) tohuman health effects. As a result, little or no specificguidance is available to wildland fire managers, airquality regulators, or public health officials who needto responsibly judge the public health risks of expo-sure to extremely high smoke concentrations. Thisgap in knowledge was clearly evident during the 1988Yellowstone fires and later wildfire events whenquick decisions had to be made on how best to protectpublic health in communities near major wildfires(WESTAR 1995). The best available guidelines arethose published by EPA (1999) for assessing the riskto health from air pollution (table 8-1). These guide-lines may or may not reflect the specific hazards ofpollutants from fires, which will have a differentchemical composition.

Table 8-1—Pollutant-specific breakpoints for the air quality index (AQI) and accompanying health effects statements (adapted fromEPA 1999).

Category PM2.5 (24-hour) PM10 (24-hour)Concentration Health effects Concentration Health effectsbreakpoints statements breakpoints statement

µg/m3 µg/m3

Good 0.0-15.4 None 0-54 None

Moderate 15.5-40.4 None 55-154 None

Unhealthy 40.5-65.4 Increasing likelihood of respiratory 155-254 Increasing likelihood of respiratoryfor sensitive symptoms in sensitive individuals, symptoms and aggravation of lunggroups aggravation of heart or lung disease disease, such as asthma.

and premature mortality of personswith cardiopulmonary disease andthe elderly.

Unhealthy 65.5-150.4 Increased aggravation of heart or lung 255-354 Increased respiratory symptoms anddisease and premature mortality in aggravation of lung disease, suchpersons with cardiopulmonary disease as asthma; possible respiratoryand the elderly; increased respiratory effects in general population.effects in the general population.

Very unhealthy 150.5-250.4 Significant aggravation of heart or lung 355-424 Significant increase in respiratorydisease and premature mortality in symptoms and aggravation of lungpersons with cardiopulmonary disease disease, such as asthma; increasingand the elderly; significant increase in likelihood of respiratory effects inrespiratory effects in general population. general population.

Hazardous 250.5-500.4 Serious aggravation of heart or lung 425-604 Serious risk of respiratory symptomsdisease and premature mortality in and aggravation of lung disease,persons with cardiopulmonary disease such as asthma; respiratory effectsand the elderly; serious risk of likely in the general population.respiratory effects in general population.


The long-term health effects of smoke exposure towildland firefighters are unknown in spite of anec-dotal evidence that indicates the possibility of agreater incidence of cardiopulmonary disease anddeath than in the general population. Although car-bon monoxide monitoring and respiratory protectioncan mitigate the hazard, personal protection equip-ment is still needed that allows firefighters to workeffectively without discomfort or distraction(Reinhardt 2000).

Welfare Effects _________________Air quality-related effects of smoke include the soil-

ing of materials, public nuisance, and visibility loss.Because these and other consequences of smoke havecome increasingly into conflict with the public’s inter-est in clean air, an understanding of these effects isimportant to fire managers.

Soiling of Materials

The deposition of smoke particles on the surface ofbuildings, automobiles, clothing, and other objectsreduces aesthetic appeal and damages a variety ofobjects and building structures (Baedecker and others1991). Studies of the effect of aerodynamic particlesize on soiling have concluded that coarse particles(2.5 to 10µm) initially contribute more to soiling ofboth horizontal and vertical surfaces than do fineparticles (less than 2.5µm), but that coarse particlesare more easily removed by rainfall (Haynie andLemmons 1990). Smoke from fires is largely within thefine mode, although ash fallout in the near vicinity ofa fire is often also a concern. Smoke may also discolorartificial surfaces such as building bricks or stucco,requiring cleaning or repainting. Increasing the fre-quency of cleaning, washing, or repainting soiled sur-faces becomes an economic burden and can reduce thelife usefulness of the soiled material (Maler and Wyzga1976).

Soiling from smoke also changes the reflectance ofopaque materials and reduces light transmissionthrough windows and other transparent materials(Beloin and Haynie 1975).

When fine smoke particles (less than 2.5µm) infil-trate indoor environments, soiling of fabrics, paintedinterior walls, and works of art may occur. Curtainsmay require more frequent washing because of soilingor may deteriorate along folds in the fabric after beingweakened by particle exposure (Yocom and Upham1977). As in the case of corrosion damage from acidi-fied particles, these same particles accelerate damageto painted surfaces (Cowling and Roberts 1954). Stud-ies of the soiling of works of art at a museum insouthern California concluded that a significant frac-tion of the dark-colored fine mode elemental carbon

and soil dust originated from outdoor sources (Ligockiand others 1993). Smoke from fires is one source ofelemental carbon.

Public Nuisance and Visibility Loss

Nuisance smoke is the amount of smoke in theambient air that interferes with a right or privilegecommon to members of the public, including the use orenjoyment of public or private resources (EPA 1990).The abatement of nuisance smoke is one of the mostimportant objectives of successful smoke manage-ment (Shelby and Speaker 1990). Public complaintsabout nuisance smoke are linked to loss of visibility,odors, and ash fallout that soils buildings, cars, laun-dry, and other objects. Acrolein (and possibly formal-dehyde) in smoke at distances of 1 mile from thefireline are likely to cause eye and nose irritation,exacerbating public nuisance conditions (Sandbergand Dost 1990).

Perhaps the most significant nuisance effect of smokefrom fire is local visibility reduction in areas impactedby the plume. While visibility loss within Class I areasis subject to regulation under the Clean Air Act, smokeplume-related visibility degradation in urban andrural communities is not. Nuisance is usually regu-lated under State and local laws and is frequentlybased on public complaint or, when highway safety iscompromised, the risk of litigation (Eshee 1995). Thecourts have also ruled that the taking of privateproperty by interfering with its use and enjoymentcaused by smoke (and without just compensation) is inviolation of Federal Constitutional provisions underthe Fifth Amendment. The trespass of smoke maydiminish the value of the property, resulting in lossesto the owner (Iowa Supreme Court 1998).

Because the public links visibility loss with concernsabout the health implications of breathing smoke,smoke management programs have been under in-creasing pressure to minimize emissions and reducesmoke impacts to the greatest degree possible (Core1989). Visibility reduction is used as a measure ofsmoke intrusions in several smoke management plans.The State of Oregon program operational guidancedefines a “moderately” intense intrusion as a reduc-tion of from 4.6 to 11.4 miles from a backgroundvisibility of more than 50 miles (Oregon Department ofForestry 1992). The State of Washington smoke intru-sion reporting system uses a “slightly visible,” “notice-able impact on visibility” or “excessive impact onvisibility“ to define light, medium, and heavy intru-sions (Washington Department of Natural Resources1993). The State of New Mexico program requires thatvisibility impacts of smoke be considered in develop-ment of the unit’s burn prescription (New MexicoEnvironmental Improvement Board 1995).


Economic and SocialConsequences__________________

The economic consequences of smoke are principallyin the areas of soiling-related losses and costs relatedto reduced visibility.

Soiling-Related Economic Losses

Economic costs associated with materials damageand soiling caused by airborne particles include reduc-tion in the useful life of the damaged materials and thedecreased utility of the object. Losses caused by theneed for more frequent maintenance and cleaning arealso significant. Amenity losses occur when the in-creased cleaning or repair of materials results ininconvenience or delays, many of which are difficult toquantify (Maler and Wyzga 1976).

Within the United States, however, the soiling ofbuildings constitutes the largest category of surfaceareas at risk to pollution damage (Lipfert and Daum1992). Soiling on painted surfaces on residential build-ings, resulting in a need to repaint exterior walls, hascaused damage approaching $1 billion per year (Haynieand others 1990).

Willingness-to-pay estimates developed using thecontingent valuation method found that householdswere willing to pay $2.70 per µg/m3 charge in particlepollution to avoid soiling effects (McClelland and oth-ers 1991). No estimates are available for costs specifi-cally associated with smoke from fires.

Visibility-Related Costs

The importance of clean, clear air within the wild-lands and National Parks of this nation is hard tooveremphasize. People go to these special places toenjoy scenery, the color of the landscapes, and clarityof the vistas. At Grand Canyon, 82 percent of 638respondents rated “clean, clear air” as very importantor extremely important to their recreational experi-ence (Ross 1988). Three National Park Service (NPS)studies determined that air quality conditions affectthe amount of time and money visitors are willing tospend at NPS units (Brookshire and others 1976;MacFarland and others 1983; Schulze and others1983). These studies found estimated onsite use val-ues for the prevention or elimination of plumes thatranged from about $3 to $6 (1989 dollars) per day pervisitor party at the park. Based on these results, theimplied preservation value for preventing a visibleplume most days (the exact frequency was not speci-fied) at the Grand Canyon was estimated at about $5.7billion each year when applied to the total U.S. popu-lation (EPA 1996). Other investigators have suggestedthat these estimates are overstated by a factor of 2 or3 (Chestnut and Rowe 1990).

In the studies noted above, park visitors generallyresponded that they would be willing to spend moretime and money if visibility conditions were betterand, conversely, less if visibility conditions were worse(Ross 1988). The average amount of time visitors werewilling to spend traveling to a vista for every unitchange in visibility (.01 km–1 extinction coefficient)was between 15 minutes and 4 hours. These resultsprovide evidence that changes in visual air quality canbe expected to affect visitor enjoyment and satisfac-tion with park visits.

Even given the limitations and uncertainties ofcontingent valuation surveys, economic values re-lated to visibility degradation are clearly likely to besubstantial.

Public Perception of Haze—Perceived visual airquality (PVAQ) has been used as a measure of thepublic’s acceptance of haze conditions (Middleton andothers 1983). Subjects were asked to judge the visualair quality in several photos depicting vistas underdifferent haze conditions using a scale of 1 to 10, 1being the worst and 10 being the best. These 1 to 10scales reflect people’s perceptions and judgments con-cerning visibility conditions. By matching particulateair quality conditions that occurred at the time of thephotographs, researchers have been able to develop arelationship between PVAQ and particulate matterconcentrations (Middleton and others 1985). Evensmall increases in particulate concentrations in theatmosphere result in dramatic decreases in PVAQ.Because of the light scattering efficiency of smoke, thisrelationship is especially applicable to fire emissions.

Cultural Consequences of Visibility Loss—“Na-tional parks and wilderness areas are among ournation’s greatest treasures. Ranging from invitingcoastal beaches and beautiful shorelines to colorfuldeserts and dramatic canyons to towering mountainsand spectacular glaciers, these regions inspire us asindividuals and as a nation” (NRC 1993). With thesewords, the National Research Council (NRC) notedthe importance of preserving the scenic vistas of thenation. Congress, in recognition of the scenic values ofthe nation, adopted the Clean Air Act Amendments of1977, which established a national visibility protec-tion program. The GCVTC was later established in the1990 amendments to the act to address visibilityimpairment issues relevant to the region surroundingGrand Canyon National Park. Following 4 years ofstudy, the GCVTC concluded that smoke from wild-land fires is likely to have the single greatest impact onvisibility in Class I areas of the Colorado Plateauthrough the year 2040 (GCVTC 1996c). While difficultto quantify, there is consensus that visibility lossassociated with smoke from wildland fire and othersources has important cultural consequences on thenation.


Highway Safety _________________Smoke can cause highway safety problems when it

impedes a driver’s ability to see the roadway (fig. 8-2)and can result in loss of life and in property damage atsmoke levels that are far below NAAQS. This sectionfocuses on highway safety issues in the SoutheasternUnited States because this is where the foremostforestry-related air quality problem has been in thepast. We also describe tools being developed to aid theland manager in avoiding highway safety problems.

Although smoke at times can become a problemanywhere in the country, it is in the Southern States,from Virginia to Texas and from the Ohio River south-ward, where highway safety is most at risk fromprescribed fire smoke, principally because of theamount of burning done in the South and the proxim-ity of wildlands to population centers. Roughly 4million acres of Southern forests are treated withprescribed fire each year (after Wade and Lunsford1988). This area is by far the largest acreage subjectedto prescribed fire in the country. Prescribed fire treat-ment intervals, especially in Southern pine (in an areaextending roughly from Virginia to Texas), is every 3to 5 years. These forests are intermixed with homes,small towns, and scattered villages within an enor-mous wildland/urban interface. During the daytime,smoke becomes a problem when it drifts into theseareas of human habitation. At night, smoke can be-come entrapped near the ground and, in combinationwith fog, creates visibility reductions that cause road-way accidents. The potential exists for frequent andsevere smoke intrusions onto the public roads andhighways from both prescribed and wildland fires.

Magnitude of the Problem

Smoke and smoke/fog obstructions of visibility onSoutheastern United States highways cause numer-ous accidents with loss of life and personal injuriesevery year. Several attempts to compile records ofsmoke-implicated highway accidents have been made.For the 10 years from 1979 through 1988, Mobley(1989) reported 28 fatalities, over 60 serious injuries,numerous minor injuries, and millions of dollars inlawsuits. During 2000, smoke from wildfires driftingacross Interstate 10 caused at least 10 fatalities, fivein Florida and five in Mississippi.

As the population growth in the South continues,more people will likely be adversely impacted bysmoke on the highways. Unless methods are found toadequately protect public safety on the highways,there exists the prospect that increasingly restrictiveregulations will curtail the use of prescribed fire orthat fire as a management tool may be altogetherprohibited.

Figure 8-2—Smoke can cause highway safety problemswhen it impedes a driver’s ability to safely see the roadway.(Photo by Jim Brenner)

Measures to Improve Highway Safety

Several approaches are being taken to reduce theuncertainty of predicting smoke movement overroadways:

High-resolution weather prediction models promiseto provide increased accuracy in predictions of windspeeds and directions and mixing heights at time andspatial scales useful for land managers. The FloridaDivision of Forestry (FDOF) is a leader in the use ofhigh resolution modeling for forestry applications inthe South (Brenner and others 2001). Because muchof Florida is located within 20 miles of a coastline,accurate predictions of sea/land breezes and associ-ated changes in temperature, wind direction, atmo-spheric stability, and mixing height are critical to thesuccess of the FDOF. High-resolution modeling con-sortia are also being established by the USDA ForestService to serve clients with interests as diverse asfire weather, air quality, ecology, and meteorology.These centers involve scientists in development ofnew products and in technology transfer to bring theproducts to consortia members.

Several smoke models are in operation or are beingdeveloped to predict smoke movement over Southernlandscapes. VSMOKE (Lavdas 1996), a Gaussianplume model that assumes level terrain and unchang-ing winds, predicts smoke movement and concentra-tion during daytime. VSMOKE has been made part ofthe FDOF fire and smoke prediction system. It is ascreening model that aids land managers in assessing


where smoke might impact sensitive targets as part ofplanning for prescribed burns. PB-Piedmont(Achtemeier 2001) is a wind and smoke model de-signed to simulate smoke movement near the groundunder entrapment conditions at night. The smokeplume is simulated as an ensemble of particles thatare transported by local winds over complex terraincharacteristic of the shallow (30 to 50 m) interlockingridge/valley systems typical of the Piedmont of theSouth. Two sister models are planned — one that willsimulate near-ground smoke movement near coastalareas influenced by sea/land circulations, and theother for the Appalachian Mountains.

Climate Change _________________Globally, fires are a significant contributor of carbon

dioxide and other greenhouse gases in the atmosphere.Fires are also an important mechanism in the redistri-bution of ecosystems in response to climate stress,which in turn affects the atmosphere-biosphere car-bon balance. Currently, there is no policy mandate,nor widely accepted methodology for managing fires,for the conservation of terrestrial carbon pools ormitigation of greenhouse gas emissions. However, wemay expect carbon accounting and perhaps conserva-tion to become a part of fire and air resource manage-ment if and when global agreements are made toaddress biomass burning and resultant greenhousegas emissions.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


Chapter 9:Recommendations for FutureResearch and Development

Managing smoke and air quality impacts from firesrequires an increasing base of knowledge obtainedthrough research and the development of informationsystems. Fire and air resource managers have had theresponsibility since the 1960s to mitigate direct intru-sions of smoke into areas where it presents a health orsafety hazard, or where it is simply objectionable to anaffected population. In more recent years, that respon-sibility has broadened because of an increase in theuse of fire, more people in the wildland/urban inter-face, tightening of regulatory standards, and decreas-ing public tolerance for air pollution. More Statesrequire smoke management plans, and the plans areincreasingly complex due to increased coverage andgreater requirements for notification, modeling, moni-toring, and recordkeeping.

Established ResearchFramework _____________________

There is ample strategic analysis and workshopoutput to guide research. The most comprehensiveand up-to-date recommendations for research anddevelopment are found in National Strategic Plan:Modeling and Data Systems for Wildland Fire and Air

Quality (Sandberg and others 1999). Workshop ses-sions, internal discussion, and review comments werecompiled into more than 200 proposals from which 46priority projects were selected that support the ninesummary recommendations outlined here.

Recommendation 1: Fuels and fire character-istics—An ability to estimate emissions from all typesof fires over the wide variation in fuels in the contigu-ous United States and Alaska is needed. Expandedmodels and fuel characteristics data are needed to fillthis gap.

Recommendation 2: Emissions modeling sys-tems—Current models to estimate emissions are in-adequate in coverage and incomplete in scope. Emis-sions production models need to be expanded to includeall fire and fuel types as well as linked to fire behaviorand air quality models in a geographically resolveddata system.

Recommendation 3: Transport, dispersion, andsecondary pollutant formation—Air quality andland management planners lack spatially explicit plan-ning and real-time systems for assessing air qualityimpacts. A geographic information system (GIS) based


system linked to emissions production, meteorologi-cal, and dispersion models is needed.

Recommendation 4: Air quality impact assess-ment—Better wildland and prescribed fire informa-tion is needed to compile emissions inventories, forregional haze analysis and for determination of com-pliance with National Ambient Air Quality Standards(NAAQS).

Recommendation 5: Emissions tradeoffs anddetermination of “natural” visibility backgroundassessments—No policy-driven or scientific defini-tion of “natural” background visibility exists for re-gional haze assessments. The tradeoffs between wild-fire and prescribed fire emissions are also not known.To address these issues, the policy community needsto decide what types of fires contribute to naturalimpairment after which a scientific assessment couldbe done and tradeoffs evaluated.

Recommendation 6: Impact and risk assess-ment of emissions from fire—A comprehensiveassessment of smoke exposure of prescribed and wild-land firefighters and the public at current levels of fireactivity should be done to provide a baseline for futurerisk assessments. Exposures should be periodicallyreassessed to evaluate increased risks from futureincreases in fire emissions.

Recommendation 7: Monitoring guidelines andprotocols—Guidelines are needed on how best tomonitor source strength, air quality, visibility, andnuisance impacts from fires to support consistent andquantitative evaluation of air impacts.

Recommendation 8: National fire and air qual-ity information database—A readily accessiblesource of information on past, current, and predictedfuture fire activity levels, emissions production, andair quality impacts from fires does not exist. Such adatabase is needed to analyze past experiences andreplicate successes.

Recommendation 9: Public information andprotection—A centralized system is needed to pro-vide information to the public on air quality impactsfrom fires. Also needed are general criteria for howland managers, air regulators, and public health offi-cials should respond to adverse smoke impacts andemergency notifications of the public to health haz-ards associated with smoke from fire.

The authors of this plan hoped that these recom-mendations would be used in future joint agencyefforts to advance the fire sciences, minimize duplica-tion of effort, and share information among agenciesand the public.

The technically advanced smoke estimation tools, orTASET, project (Fox and Riebau 2000) was funded by

the Joint Fire Sciences Program (JFSP) to develop astructured analysis of smoke management and recom-mend specific developments for advancing the state ofscience. The report confirmed and refined the recom-mendations of Sandberg and others (1999) above, anddeveloped 10 recommendations for research activities:

• Fire community participation in regional airquality modeling consortia.

• Conduct a national smoke and visibility con-ference and reference guide.

• Develop a national smoke emissions data struc-ture or database system.

• Apply remote sensing for fuels and fire areaemissions inventories.

• Develop a fire gaming system to quantify emis-sions and impacts from alternative fire man-agement practices.

• Improve the CalMet/CalPuff smoke manage-ment model.

• Upgrade a nationalized screening model/simple approach smoke estimation model(SASEM).

• Provide onsite fire emissions verification.• Utilize back-trajectory modeling and filter

analysis for fire smoke contributions fornonattainment areas.

• Develop a method to identify the specificsources of organic carbon fine particulatematerial.

Research priorities established in the Effects of FireAir (Sandberg and others 1979) are unfortunately stillvalid today, although some progress has been made inevery category. We list these here, slightly rewordedfrom the original for brevity and to conform to modernnomenclature:

1. Provide quantitative smoke management systems.a. Develop information systems necessary to

support smoke management decisions.b. Provide a smoke management reporting sys-

tem for emission rates based on the predictionof fuel consumption, fire behavior, heat re-lease rates, and source control measures.

c. Provide the data network and modeling schemeto calculate the change in pollution concentra-tions and character between the source andpotential receptors.

d. Adapt plume rise models necessary to predictthe vertical distribution of emissions fromfires.

2. Characterize the chemistry and physics ofemissions.a. Relate emissions and heat release rates to

fuelbed characteristics and fire behavior.


b. Advance our knowledge of hazardous and re-active compounds in smoke.

c. Develop field methods to monitor emissionrates and smoke chemistry from operationalfires.

d. Investigate the potential for secondary reac-tions of emissions downstream from theirsource.

3. Model atmospheric transport, diffusion, trans-formation, and removal mechanisms.a. Continue development of winds and disper-

sion models for boundary layer flow and me-soscale transport of smoke over mountainousterrain.

b. Investigate the mechanisms of removal; forexample, canopy interactions, fallout, and lo-cal deposition.

c. Interact with the wider scientific communityto establish the effect of reactive pollutants onthe biosphere.

d. Evaluate the potential contribution of wild-land fires to climate change.

4. Identify receptor responses to wildland smoke.a. Identify and quantify the visibility needs of

wildland users, and recommend standards forparticulate and sulfate pollution from allsources affecting Class I visibility areas.

b. Evaluate the potential impact of wildlandsmoke on human health.

c. Investigate the role of wildland ecosystems as asink and receptor for atmospheric contaminants.

5. Investigate tradeoffs made in the substitution ofalternatives to fire use.a. Develop simulation models to evaluate inter-

actions of land use policy with air resourcemanagement. Incorporate air resource man-agement and fuels management needs intothe land use planning process.

b. Evaluate the effect on wildland fire occurrenceand air pollution from changes in the amountof prescribed fire activity.

c. Describe the resource and economic tradeoff ofwildland fire occurrence resulting from achange in prescribed fire activity.

d. Investigate the effect of changes in fire use onnutrient cycling, successional response, andecosystem stability.

Emerging Research Needs________Several new responsibilities create the need for

additional information systems that require new re-search and development, including:

• Planning rules that require the considerationof cumulative pollution and visibility impactsof fuel management programs.

• Wildland fire situation analysis requirementsthat smoke impacts from wildland be antici-pated and communicated to the public.

• Increased requirements for emission reduction.• Policies that require hourly and daily tracking

of emissions and the management of smokefrom all fires.

• Increased management of wildland fires forresource benefits.

• Increased use of long-duration landscape-scalefires.

• Regulatory concern over secondary pollutants,especially ozone formation and the reentrain-ment of mercury.

• Questions about the role of fire and globalbiomass emission on atmospheric carbon andglobal warming.

• Increased attention to firefighter health ef-fects from exposure to smoke.

Each of these factors requires information systemsfor planning, operations, and monitoring the effects offire on air. Using the framework illustrated in figures1-1 and 1-2 (in chapter 1) and the background ofprevious chapters, some emerging research needs areoutlined below.

Emissions Source Strength andEmissions Inventory

Level of burn activity: Accurately predict, de-termine, and record the area burned and time ofburning for all types of prescribed and wildlandfire—Area burned is still the parameter that impartsthe greatest error into predictions of source strengthand emission inventory. Needed are: a balanced pro-gram of new planning models that project area burnedand fire residence times; remote-sensing technologiesthat track fire sizes at hourly intervals; ground basedsampling, reporting, and communication systems; andanalysis tools. Planning models include those thatproject fire use and predict wildland fire activity from1 to 50 years in the future must be included, as well asaccurate predictions made a day in advance.

Biomass: Accurately predict, determine, andrecord the mass, combustion stage, and resi-dence time of fuels burned in all types of fires—Inadequate representation of fuelbed characteristicsand the ability to infer fuelbed characteristics andflammability conditions from remote sensing or eco-system physiognomy is the second greatest remainingsource of error. Models of the combustion process,


while improving, are still inadequate to predict orcharacterize emission rates and durations. New clas-sification systems, inference models, inventory andsensing processes, and process models are needed.

Heat release and emissions: Predict and mea-sure physical and chemical characteristics ofemissions from all types of fires—Among the great-est advances since about 1980 has been the nearlycomplete characterization of primary and criteria pol-lutants from a wide range of fire environments. Newmodels also greatly improve the prediction and char-acterization of emissions source strength. Emissionfactors for criteria pollutants are adequate. There issubstantial remaining uncertainly in the measure-ment and prediction of precursors to ozone and othersecondary chemical formations, secondary entrain-ment of mercury, production and stimulation of nitro-gen compounds, air toxics, and greenhouse gases.Continuing research on these trace constituents areneeded. In addition, we lack models that characterizethe complex spatial and temporal distribution of heatrelease from fires.

Emissions inventory methods: Integrate mea-surements and reporting from remote sensing,airborne platforms, simulation models, and sur-face observations into a fine-scale spatial andtemporal emission inventory—Emission invento-ries are a fundamental tool that air resource managersuse to calculate the relative importance of air pollutionsources and to design control strategies. Hourly, point-specific emission estimates as well as daily, monthly,and yearly summaries are necessary to compare firewith other sources or as inputs to dispersion models.Fire managers currently lack a system of observationsand reporting mechanisms required for planning, track-ing, and monitoring emissions.

Ambient Air Quality Impacts

Background air quality: Improve the accessi-bility of girded detail about background airquality and meteorological conditions—Fireemissions are inserted into an already complex atmo-sphere, and current ability to predict pollutant inter-actions, transformations, and combined effects arelimited by the availability of hourly fine-scale atmo-spheric profiling.

Plume rise and transport: Improve the pre-diction, detection, and tracking of plumes fromall types and stages of fires—Fire plumes arecomplex; often splitting into lofted and unlofted por-tions; plumes that split in two directions at different

altitudes, and plumes that change rapidly over time.Plumes are transported long distances, often overcomplex terrain, and the accuracy and availability ofmodels to predict transport are inadequate. Methodsto track plume trajectories and measure pollutantconcentrations in near real time using remote sensingare emerging but not yet available.

Dispersion, dilution, and pollutant transfor-mation: Improve the ability on all scales to pre-dict, model, and detect changes in the proper-ties and concentration of pollutants over timeand space—Data and models are needed to initiateand predict local, regional, national, and global airquality impacts from individual fires to the cumula-tive effects of tens of thousands of fires.

Atmospheric carbon balance and climaticchange: Develop consistent technologies to as-sess the contribution of fires to greenhouse gasesin the atmosphere and the effect of fire andecosystem management practices—For a sourceof greenhouse gas emissions as large as wildland andprescribed fires, there is a regrettable lack of consen-sus on the magnitude or even the methods for assess-ment and accountability. This emerging issue re-quires much of the same research on sourcecharacteristics and air quality as do the health, safety,and visibility issues, but also requires integrationwith the global science and policy communities.

Effects on Receptors

Visibility and other welfare effects: Predict,measure, and interpret the impact of naturaland anthropogenic fire sources on visibility,economic, and other welfare effects—The impactof smoke exposure from fires on human health stan-dards is minor relative to the nuisance it creates andthe impacts on visibility. New science is required tomonitor and predict effects on visibility, and to appor-tion visibility impacts to specific sources and classes ofsources.

Health and safety risk assessment: Developknowledge and systems to assess the risk ofindividual and collective fires to personal andcommunity health and safety—This broad topichas received limited attention in recent years, mostlyin the prediction of visibility impacts on highwaysafety and in the assessment of individual firefighterexposure to hazardous air pollutants. But all aspectsof risk management, including hazard identification,exposure assessment, dose-response, risk assessment,and mitigation measures are lacking.


Conclusion_____________________Knowledge and information requirements for man-

aging fire effects on air quality continue to increase.Policy advancements require the understanding, mod-eling, prediction, monitoring, and tracking of fires andtheir effect on air at greater detail and in greatervolume than ever before. Research and developmenthas progressed logically over the past 25 years due tostrategic planning and prioritization that has included

the needs of the managers of ecosystems and of airquality. Analytical and information transfer capacityhas increased dramatically in the past decade, soinformation is more readily accessible to those whoneed it. Thanks largely to the National Fire Plan, theJoint Fire Science Program, the Western Regional AirPartnership, and EPA’s implementation of the Re-gional Haze Rule, there is currently more active re-search and development the effects of fire on air thanever before.


Notes________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________

________________________________________________________________


40 CFR Part 51. Vol. 64 No. 126. Regional Haze Regulations—FinalRule. July 1, 1999.

Achtemeier, G.L. 1994. A computer wind model for predictingsmoke movement. Southern Journal of Applied Forestry. 18:60–64.

Achtemeier, Gary L. 2000. PB-Piedmont: A numerical model forpredicting the movement of biological material near the groundat night. In: Proceedings of the 24th conference on agriculturaland forest meteorology. Boston, MA: American Meteorology Soci-ety: 178–179.

Achtemeier, Gary L. 2001. Simulating nocturnal smoke movement.Fire Management Today. 61: 28–33.

Albini, F.A.; Brown, J.K.; Reinhardt, E.D.; Ottmar, R.D. 1995.Calibration of a large fuel burnout model. International Journalof Wildland Fire. 5(3): 173–192.

Albini, F.A.; Reinhardt, E.D. 1995. Modeling ignition and burningrate of large woody natural fuels. International Journal of Wild-land Fire. 5(2): 81–91.

Albini, F.A.; Reinhardt, E.D. 1997. Improved calibration of a largefuel burnout model. International Journal of Wildland Fire. 7(1):21–28.

Anderson, Hal E. 1969. Heat transfer and fire spread. Res. Pap.INT-69. Ogden, UT: U.S. Department of Agriculture, ForestService, Intermountain Forest and Range Experiment Station.20 p.

Andreae, M.O.; Browell, E.V.; Garstang, M.; Gregory, G.L.; Harriss,R.C.; Hill, G.F.; Jacob, D.J.; Pereira, M.C.; Sachse, G.W.; Setzer,A.W.; Silva Dias, P.L.; Talbot, R.W.; Torres, A.L., Wofsy, S.C.1988. Biomass-burning emissions and associated haze layersover Amazonia. Journal of Geophysical Research. 93: 1509–1527.

Andreae, M.O.; Anderson, B.E.; Blake, D.R.; Bradshaw, J.D.; Collins,J.E.; Gregory, G.L.; Sachse, G.W.; Shipman, M.C. 1994. Influenceof plumes from biomass burning on atmospheric chemistry overthe equatorial and tropical South Atlantic during CITE 3. Jour-nal of Geophysical Research. 99(D6): 12,793–12,808.

Andreae, M.O.; Merlet, P. 2001. Emission of trace gases and aero-sols from biomass burning. Global Biogeochemical Cycles. 15(4):955–966.

Andrews, Patricia L.; Bevins, Collin D. 1999. BEHAVE fire model-ing system—redesign and expansion. Fire Management Notes.59(2): 16–19.

Baedecker, P.A.; Edney, E.O.; Morgan, P.J.; Simpson, T.C.;Williams, R.S. 1991. Effects of acidic deposition on materials.In: Irving, P.M., ed. Acidic deposition: state of science andtechnology, volume III: terrestrial, materials, health andvisibility effects. Washington, DC: The U.S. National AcidicPrecipitation Assessment Program (NAPAP). AtmosphericEnvironment. 26: 147–158.

Battye, R.; Bauer, B.; MacDonald, G. 1999. Features of prescribedfire and smoke management rules for western and southernStates. Contract 68-D-98-026. Chapel Hill, NC: EC/R Incorpo-rated. Prepared for U.S. Environmental Protection Agency.

Battye, William; Battye, Rebecca. 2002. Development of emissionsinventory methods for wildland fire. Final report. Contract 68-D-98-046. Research Triangle Park, NC: U.S. Environmental Protec-tion Agency.

Beason, Donald G.; Johnson, James S.; Foote, Kenneth L.; Weaver,William A. 1996. Summary report. California Department ofForestry and Fire Protection evaluation of full-face air-purifyingrespirators for wildland fire fighting use. California Departmentof Forestry and Fire Protection Contract WN-02-19-05-0.Livermore, CA: Lawrence Livermore National Laboratory.February.

Beloin, N.J.; Haynie, F.H. 1975. Soiling of building materials.Journal of the Air Pollution Control Association. 25: 399–403.

Betchley, C.; Koenig, J.Q.; van Belle, G. [and others]. 1995. Pulmo-nary function and respiratory symptoms in forest firefighters.Unpublished report. On file with: University of Washington,Departments of Environmental Health and Epidemiology,Seattle, WA.

ReferencesBooze, Thomas F.; Reinhardt, Timothy E. [In press]. A screening-

level assessment of the health risks of chronic smoke exposure forwildland firefighters. American Industrial Hygiene AssociationJournal.

Bradley, Michael M.; Schomer, Christina L.; Sumikawa, Denise A.;Walker, Hoyt; Younker, Leland W.; Bossert, James E.; Hanson,Howard P.; Linn, Rodman R.; Reisner, Jon M. 2000. The nationalwildfire prediction program: a key piece of the wildfire solution.In: Neuenschwander, Leon, F.; Ryan, Kevin C., tech. eds. Proceed-ings from the joint fire science conference and workshop: crossingthe millennium: integrating spatial technologies and ecologicalprinciples for a new age in fire management. University of Idaho:64–76.

Brenner, J.; Suddaby, R.M.; Carr, R.J.; Lee, B.S.; Brackett, D.P.;Arvanitus, L.G. 2001. GIS-based fire management in Florida.Journal of Forestry. 95(6): 140–147.

Breyfogle, Steve; Ferguson, Sue A. 1996. User assessment of smoke-disperion models for wildland biomass burning. Gen. Tech. Rep.PNW-GTR-379. Portland, OR: U.S. Department of Agriculture,Forest Service, Pacific Northwest Research Station. 30 p.

Briggs, G.A. 1969. Plume rise. Prepared for: Nuclear Safety Infor-mation Center, Oak Ridge National Laboratory. Oak Ridge, TN:U.S. Atomic Energy Commission, Division of Technical Informa-tion. Springfield, VA: Clearinghouse for Federal Scientific andTechnical Information, National Bureau of Standards, U.S. Dept.of Commerce.

Brookshire, D.S.; Ives, B.C.; Schulze, W.D. 1976. The valuation ofaesthetic preferences. Journal of Environmental Economics andManagement. 3: 325–346.

Brown, Arthur A.; Davis, Kenneth P. 1959. Forest fire: control anduse. New York: McGraw-Hill Book Company. 686 p.

Brown, J.K.; Reinhardt, E.D.; Fischer, W.C. 1991. Predicting duffand woody fuel consumption in northern Idaho prescribed fires.Forest Science. 37(6): 1550–1566.

Bryan, Dana C., ed. 1997. Conference proceedings: Environmentalregulation & prescribed fire: legal and social challenges. Talla-hassee, FL: Florida State University, Center for ProfessionalDevelopment. 246 p.

Byun, D.W. 1999a. Dynamically consistent formulations in meteo-rological and air quality models for multiscale atmospheric stud-ies. Part I: Governing equations in a generalized coordinatesystem. Journal of Atmospheric Sciences. 56: 3789–3807.

Byun, D.W. 1999b. Dynamically consistent formulations in meteo-rological and air quality models for multiscale atmospheric stud-ies. Part II: Mass conservation issues. Journal of AtmosphericSciences. 56: 3808–3820.

Byun, D.W.; Ching, J.K.S., eds. 1999. Science algorithms of the EPAModels-3 community multiscale air quality (CMAQ) modelingsystem. EPA/600/R-99/030. U.S. Environmental ProtectionAgency, Office of Research and Development.

Byun, D.W.; Pleim, J.E.; Tang, R.T.; Bourgeois, A. 1999. Chapter 12:meteorology-chemistry interface processor (MCIP) for MODELS-3 community multiscale air quality (CMAQ) modeling system. In:Byun, D.W.; Ching, J.K.S., eds. Science algorithms of the EPAModels-3 community multiscale air quality (CMAQ) modelingsystem. EPA/600/R-99/030. U.S. Environmental ProtectionAgency, Office of Research and Development.

Cassinelli, M.E.; O’Connor, P.F. eds. 1994. NIOSH manual ofanalytical methods, 4th ed. Washington, DC: U.S. GovernmentPrinting Office.

Chameides, W.L.; Cowling, E.B. 1995. The state of the SouthernOxidants Study (SOS): policy-relevant findings in ozone pollutionresearch, 1988–1994. Raleigh, NC: North Carolina State Univer-sity, College of Forestry. 94 p.

Chang, J.S.; Chang, K.H.; Jin, S. 1993. Two-way and one-waynested SARMAP air quality model. In: International conferenceon regional photochemical measurement and modeling studies.Pittsburgh, PA: Air & Waste Management Association.


Chatfield, R.B.; Delaney, A.C. 1990. Convection links biomass burn-ing to increased tropical ozone: however, models will tend to overpredict O3. Journal of Geophysical Research. 95: 18,473–18,488.

Chatfield, R.B.; Vastano, J.A.; Singh, H.B.; Sachase, G.W. 1996. Ageneral model of how fire emissions and chemistry produceAfrican/oceanic plumes (O3, CO, PAN, smoke) in TRACE A.Journal of Geophysical Research. 101(D19): 24,279–24,306.

Chestnut, L.G.; Rowe, R.D. 1990. Preservation values for visibilityin the National Parks. Washington, DC: U.S. EnvironmentalProtection Agency.

Chi, C.T.; Horn, D.A.; Reznik, R.B.; [and others]. 1979. Sourceassessment: prescribed burning, state of the art. Final report.EPA-600/2-79-019h. Research Triangle Park, NC: U.S. Environ-mental Protection Agency, Office of Research and Development.106 p.

Chow, J.C. 2000. Comparison of IMPROVE and NIOSH carbonmeasurements. Presented at the PM2000: Particulate matterand health conference. Pittsburg, PA: Air & Waste ManagementAssociation.

Clark, T.L.; Jenkins, M.A.; Coen, J.; Packham, D. 1996. A coupledatmosphere-fire model: convective feedback on fire-line dynamics.Journal of Applied Meteorology. 35(6): 875–901.

Clean Air Scientific Advisory Committee. 1995. Clean Air ScientificAdvisory Committee (CASAC) Comments on the April 1995 draftair quality criteria for particulate matter. EPA-SAB-CASAC-LTR-95-005. U.S. Environmental Protection Agency, ScienceAdvisory Committee, Clean Air Scientific Advisory Committee,Particulate Matter Criteria Document Review Panel. 8 p.

Core, J.E. 1989. Air quality and forestry burning: public policyissues. In: Hanley, D.P.; Kammenga, J.J.; Oliver, C.D., eds. Theburning decision: regional perspectives on slash. Seattle, WA:University of Washington, College of Forest Resources: 237–245.

Core, J.E.1996. Wildfire smoke emergency action plan implementa-tion guideline: draft of 11/15/95. In: WESTAR Council wildfireand prescribed fire workshop. Portland, OR: WESTAR Council.

Core, J.E. 1998. Survey of smoke management programs in thewestern United States. Prepared for the USDI Bureau of LandManagement, National Applied Resources Sciences Center. May.

Cowling, J.E.; Roberts, M.E. 1954. Paints, varnishes, enamels, andlacquers. In: Greathouse, G.A.; Wessel, C.J., eds. Deterioration ofmaterials: causes and preventive techniques. New York: ReinholdPublishing Corporation: 596–645.

Crutzen, P.J.; Andreae, M.O. 1990. Biomass burning in the tropic:impacts on atmospheric chemistry and biogeochemical cycles.Science. 250(4988): 1669–1678.

Crutzen, P.J.; Carmichael, G.J. 1993. Modeling the influence of fireson atmospheric chemistry. In: Crutzen, P.J.; Goldammer, J.G.,eds. Fire in the environment: the ecological, atmospheric, andclimatic importance of vegetation fires. New York: Wiley: 89–105.

Davies, Mary Ann. 2002. DataRAM4 particulate monitor: ForestService user’s guide. Tech. Rep. 0225-2810-MTDC. Missoula, MT:U.S. Department of Agriculture, Forest Service, Technology andDevelopment Program. 26 p.

DeBano, Leonard F.; Neary, Daniel G.; Ffolliott, Peter F. 1998.Fire’s effects on ecosystems. New York: John Wiley and Sons, Inc.333 p.

de Nevers, N. 2000. Air pollution control engineering. Boston, MA:McGraw Hill. 586 p.

Desalmand, F.; Serpolay, R.; Podzimek, J. 1985. Some specificfeatures of the aerosol particle concentrations during the dryseason and during a bush fire event in West Africa. AtmosphericEnvironment. 19(9): 1535–1543.

Dickson, R.J.; Oliver, W.R.; Dickson, E.L. 1994. Development of anemissions inventory for assessing visual air quality in the west-ern United States. Prepared for the Western Governors’ Associa-tion and the Electric Power Research Institute. Radian Corpora-tion 674-050-04-01. July.

Dockery, D.W.; Pope, C.A., III; Xu, X.; Spengler, J.D.; Ware, J.H.;Fay, M.E.; Ferris, B.G., Jr.; Speizer, F.E. 1993. An associationbetween air pollution and mortality in six U.S. cities. NewEngland Journal of Medicine. 329: 1753–1759.

Draxler, R.R.; Hess, G.D. 1998. An overview of the HYSPLIT-4modelling system for trajectories, dispersion and deposition.Australian Meteorological Magazine. 47(4): 295–308.

Dull, K.; Acheson, A.; Thomas, D.; Chapell, L.; Volkland, S. 1998.Forecasting smoke dispersion and concentration in Idaho andMontana. Unpublished report. On file at: U.S. Department ofAgriculture, Forest Service, Pacific Northwest Region, Portland,OR. 13 p.

Eatough, D.J.; Pang, Y. 1999. Determination of PM2.5 sulfate andnitrate with a PC-BOSS designed for routine sampling for semi-volatile particulate matter. Journal of the Air & Waste Manage-ment Association. 49: 69–75.

Eshee, W.D. 1995. Legal implications of using prescribed fire. In:Bryan, D.C., ed. Proceedings: environmental regulation andprescribed fire conference: legal and social challenges. Tallahas-see, FL: Division of Forestry, Florida Department of Agricultureand Consumer Services: 126–130.

Ferguson, S.A.; Peterson, J.; Acheson, A. 2001. Automated, real-time predictions of cumulative smoke impacts from prescribedforest and agricultural fires. In: Fourth symposium on fire andforest meteorology. Boston, MA: American Meteorological Soci-ety: 168–175.

Ferguson, Sue A.; Sandberg, David V.; Ottmar, Roger. 2000.Modelling the effect of landuse changes on global biomassemissions. In: Innes, John L.; Beniston, Martin; Verstraete,Michel M., eds. Biomass burning and its relationships with theclimate system. Dordrecht, The Netherlands: Kluwer AcademicPublishers: 33–50.

Finlayson-Pitts, Barbara J.; Pitts, James N., Jr. 1986. Atmosphericchemistry: fundamentals and experimental techniques. NewYork: John Wiley and Sons, Inc. 1098 p.

Finney, M. 1998. FARSITE: Fire area simulator—model develop-ment and evaluation. Res. Pap. RMRS-RP-4. Ogden, UT: U.S.Department of Agriculture, Forest Service, Rocky MountainResearch Station. 47 p.

Finney, Mark. 2000. [Personal communication]. Research forester.Rocky Mountain Research Station, Fire Sciences Laboratory,Missoula, MT.

Fire Emissions Joint Forum. 2002. WRAP policy: enhanced smokemanagement programs for visibility. http://www.wrapair.org/forums/FEJF/esmptt/policy/WRAP_ESMP_Policy_090202.pdf.[September 2002].

Fishman, J.; Fakhruzzaman, K.; Cros, B.; Ngana, D. 1991. Identi-fication of widespread pollution in the Southern Hemispherededuced from satellite analyses. Science. 252: 1693–1696.

Fox, Douglas G.; Riebau, Allen R. 2000. Technically advancedsmoke estimation tools (TASET). Final report. Fort Collins, CO:Colorado State University, Cooperative Institute for Research inthe Atmosphere. 99 p.

Goode, J.G.; Yokelson, R.J.; Susott, R.A.; Ward, D.E. 1999. Trace gasemissions from laboratory biomass fires measured by open-pathFourier transform infrared spectroscopy: fires in grass and sur-face fuels. Journal of Geophysical Research. 104: 21,237–21,245.

Goode, J.G.; Yokelson, R.J.; Susott, R.A.; Babbitt, R.E.; Ward, D.E.;Davies, M.A.; Hao, W.M. 2000. Measurements of excess O3, CO2,CO, CH4, C2H4, C2H2, HCN, NO, NH3, HCOOH, CH3COOH,HCHO, and CH3OH in 1997 Alaskan biomass burning plumes byairborne Fourier transform infrared spectroscopy (AFTIR). Jour-nal of Geophysical Research. 105: 22,147.

Grand Canyon Visibility Transport Commission. 1996a. Alterna-tive assessment committee report. Denver, CO: Western Gover-nors’ Association.

Grand Canyon Visibility Transport Commission. 1996b. Recom-mendations for improving western vistas. Denver, CO: WesternGovernors’ Association.

Grand Canyon Visibility Transport Commission. 1996c. Report ofthe Grand Canyon Visibility Transport Commission to the UnitedStates Environmental Protection Agency (1996). Denver, CO:Western Governors’ Association. 85 p.

Grell, G.A.; Dudhia, J.; Stauffer, D.R. 1995. A description of thefifth-generation Penn State/NCAR mesoscale model (MM5). NCARTechnical Note, NCAR/TN-398+ STR. Boulder, CO: NationalCenter for Atmospheric Research.

Haines, D.A.; Smith, M.C.1987. Three types of horizontal vorticesobserved in wildland mass and crown fires. Journal of Climateand Applied Meteorology. 26: 1624–1637.

Haines, Donald A.; Updike, Gerald H. 1971. Fire whirlwind forma-tion over flat terrain. Res. Pap. NC-71. St. Paul, MN: U.S.


Department of Agriculture, Forest Service, North Central ForestExperiment Station. 12 p.

Hao, Wei Min; Lui, Mei-Huey. 1994. Spatial and temporal distribu-tion of tropical biomass burning. Global Biogeochemical Cycles.8(4): 495–503.

Hardy, C.C.; Conard, S.G.; Regelbrugge, J.C.; Teesdale, D.T. 1996.Smoke emissions from prescribed burning of southern Californiachaparral. Res. Pap. PNW-RP-486. Portland, OR: U.S. Depart-ment of Agriculture, Forest Service, Pacific Northwest ResearchStation. 37 p.

Hardy, C.C.; Ferguson, S.A.; Speers-Hayes, P.; Doughty, C.B.;Teasdale, D.R. 1993. Assessment of PUFF: a dispersion modelfor smoke management. Final report. Submitted to: U.S. De-partment of Agriculture, Forest Service, Pacific NorthwestRegion. 32 p.

Hardy, C.C.; Menakis, J.P.; Long, D.G.; Garner, J.L. 1998. FMI/WESTAR emissions inventory and spatial data for the WesternUnited States. Final report. EPA agreement numberDW12957250-01-0. U.S. Department of Agriculture, Forest Ser-vice, Rocky Mountain Research Station, Fire Effects ResearchWork Unit, Missoula, MT.

Hardy, C.C.; Ward, D.E.; Enfield, W. 1992. PM2.5 emissions from amajor wildfire using a GIS: rectification of airborne measure-ments. In: Proceedings, 29th annual meeting of the Pacific North-west International Section, Air & Waste Management Associa-tion. Pittsburgh, PA: Air & Waste Management Association.

Hardy, Colin C.; Ottmar, Roger D.; Peterson, J.L.; Core, John E.;Seamon, Paula, comps., eds. 2001. Smoke management guide forprescribed and wildland fire: 2001 edition. PMS 420-2. Boise, ID:National Wildfire Coordinating Group. 226 p.

Harms, Mary F.; Lavdas, Leonidas G. 1997. Draft user’s guide toVSMOKE-GIS for workstations. Research Paper SRS-6. Asheville,NC: U.S. Department of Agriculture, Forest Service, SouthernResearch Station. 41 p.

Harrison, H. 1995. A user’s guide to NFSPUFF: a dispersion modelfor smoke management in complex terrain. WYNDSoft Inc. 42 p.

Haynie, F.H.; Lemmons, T.J. 1990. Particulate matters oiling ofexterior households paints. Journal of the Air Pollution ControlAssociation. 34: 941–944.

Haynie, F.H.; Spence J.W.; Lipfert, F.W. Cramer, S.D.; McDonald,L.G. 1990. Evaluation of an atmospheric damage function forgalvanized steel. In: Baboian, R.; Dean, S.W., eds. Corrosiontesting and evaluation: silver anniversary volume. ASTM SpecialTech. Publ. 1000. Philadelphia, PA: American Society for Testingand Materials: 225–240.

Hobbs, P.V.; Radke, L.F. 1969. Cloud condensation nuclei from asimulated forest fire. Science. 163: 279–280.

Hobbs, P.V.; Reid, J.S.; Herring, J.A.; Nance, J.D.; Weiss, R.E.; Ross,J.L.; Hegg, D.A.; Ottmar, R.D.; Liousse, C. 1996. Particle andtrace-gas measurements in the smoke from prescribed burns offorest products in the Pacific Northwest. In: Levine, J.S., ed.Biomass burning and global change: volume I remote sensing,modeling and inventory development, and biomass burning inAfrica. Cambridge, MA: The MIT Press: 697–715.

Hopke, P.K. 1985. Receptor modeling in environmental chemistry.New York: Wiley.

Hummel, J.; Rafsnider, J. 1995. User’s Guide, TSARS plus smokeproduction and dispersion model. Unpublished. National Biologi-cal Service and the Interior Fire Coordination Committee. 107 p.

Idaho Department of Environmental Quality. [n.d.] Supportingtechnical document for PM10 excursions in Salmon, Idaho, dur-ing the summer of 2000. Boise, ID: Idaho Department of Environ-mental Quality. http://www2.state.id.us/deq/air/smoke/NEAP/NEAP_SupportingDoc.pdf. (August 8, 2002).

Iowa Supreme Court. 1998. Bormann and others versus Board ofSupervisors in and for Kossuth County, Iowa. September 23,1998. No. 192/96-2276. http://www.judicial.state.ia.us/supreme/opinions/19980923/96-2276.asp. (August 8, 2002).

Jacob, D.J.; Heikes, B.G.; Fan, S.-M.; Logan, J.A.; Mauzerall, D.L.;Bradshaw, J.D.; Singh, H.B.; Gregory, G.L.; Talbot, R.W.; Blake,D.R.; Sachse, G.W. 1996. Origin of ozone and NOX in the tropicaltroposphere: a photochemical analysis of aircraft observationsover the South Atlantic Basin. Journal of Geophysical Research.101(24): 24,235–24,250.

Jacob, D.J.; Wofsy, S.C.; Bakwin, P.S.; Fan, S.-M.; Harriss, R.C.;Talbot, R.W.; Bradshaw, J.D.; Sandholm, S.T.; Singh, H.B.;Browell, E.V.; Gregory, G.L.; Sachse, G.W.; Shipham, M.C.;Blake, D.R.; Fitzjarrald, D.R. 1992. Summertime photochemistryof the troposphere at high northern latitudes. Journal of Geo-physical Research. 97: 16,421–16,431.

Jennings, S.G. 1998. Wet processes affecting atmospheric aerosols.In: Harrison, R.M.; Grieken, R.V., eds. Atmospheric particles:475–508.

Kasischke, E.S.; Stocks, B.J., eds. 2000. Fire, climate change, andcarbon cycling in the boreal forest. Ecological Studies, Vol. 138.New York: Springer. 461 p.

Khalil, M.A.K.; Edgerton, S.A.; Rasmussen, R.A. 1983. A gaseoustracer model for air pollution from residential wood burning.Environmental Science and Technology. 22: 53–61.

Koppmann, R.; Khedim, A.; Rudolph, J.; Poppe, D.; Andreae, M.O.;Helas, G.; Welling, M.; Zenker, T. 1997. Emissions of organictrace gases from savanna fires in southern Africa during the 1992Southern Africa Fire Atmosphere Research Initiative and theirimpact on the formation of tropospheric ozone. Journal of Geo-physical Research. 102: 18,879–18,888.

Kumar, N.; Russell, A.G. 1996. Development of a computationallyefficient, reactive sub-grid scale plume model and the impact inthe northeastern United States using increasing levels of chemi-cal detail. Journal of Geophysical Research. 101: 16,737–16,744.

Lavdas, L.G. 1982. A day/night box model for prescribed burningimpact in Willamette Valley, Oregon. Journal of Air PollutionControl Agency. 32: 72–76.

Lavdas, L.G. 1996. Program VSMOKE—user manual. Gen. Tech.Rep. SRS-6. Asheville, NC: U.S. Department of Agriculture,Forest Service, Southern Research Station. 147 p.

Laverty, Lyle; Williams, Jerry. 2000. Protecting people and sus-taining resources in fire-adapted ecosystems: a cohesive strategy.Washington, DC: U.S. Department of Agriculture, Forest Ser-vice. 85 p.

Lavoué, David; Liousse, Catherine; Cachier, Hélène. 2000. Model-ing of carbonaceous particles emitted by boreal and temperatewildfires at northern latitudes. Journal of Geophysical Research.105(D22): 26,871–26,890.

Lee, M.; Heikes, B.G.; Jacob, D.J. 1998. Enhancements of hydro-peroxides and formaldehyde in biomass burning impacted air andtheir effect on atmospheric oxidant cycles. Journal of GeophysicalResearch. 13,201–13,212.

Leenhouts, Bill. 1998. Assessment of biomass burning in theconterminous United States. Conservation Ecology [online].2(1). http://www.consecol.org/vol2/iss1/art1 (August 8, 2002).

Lelieveld, J.; Crutzen, P.J.; Jacob, D.J.; Thompson, A.M. 1997.Modeling of biomass burning influences on tropospheric ozone.In: Wilgen, B.W., ed. Fire in southern African savannas: ecologi-cal and atmospheric perspectives. Johannesburg, South Africa:Witsatersrand University Press: 217–238.

Levin, J.S., ed. 1996. Global biomass burning: Atmospheric, cli-matic, and biospheric implications. Cambridge, MA: MIT Press.569 p.

Levine, Joel S.1994. Biomass burning and the production of green-house gases. In: Zepp, Richard G. Climate-biosphere interac-tions: biogenic emissions and environmental effects of climatechange. New York: John Wiley and Sons, Inc.: 139–160.

Levine, Joel S.; Cofer, Wesley R., III. 2000. Boreal forest fireemissions and the chemistry of the atmosphere. In: Kasischke,Eric S.; Stocks, Brian J., eds. Fire, climate change, and carboncycling in the boreal forest. Ecological Studies Vol. 138. NewYork: Springer-Verlag: 31-48.

Levinson, D. 2001. The Montana/Idaho airshed group operatingguide. Missoula, MT: Montana/Idaho Airshed Group, SmokeMonitoring Unit. 48 p.

Ligocki, M.P.; Salmon, L.G.; Fall, T.; Jones, M.C.; Nazaroff, W.W.;Cass, G.R. 1993. Characteristics of airborne particles insidesouthern California museums. Atmospheric Environment. PartA. 27: 697–711.

Lipfert, F.W.; Daum, M.L. 1992. The distribution of common con-struction materials at risk to acid deposition in the United States.Atmospheric Environment Part B. 26: 217–226.


Little, Beth. [n.d.]. CalPFIRS [California prescribed fire incidentreporting system] program manual. Redding, CA: U.S. Depart-ment of Agriculture, Forest Service, Northern California ServiceCenter. 43 p.

Lobert, J.M.; Scharffe, D.H.; Hao, W.M.; Kuhlbusch, T.A.; Seuwen,R.; Warneck, P.; Crutzen, P.J. 1991. Experimental evaluation ofbiomass burning emissions: Nitrogen carbon containing com-pounds. In: Levin, J.S., ed. Global biomass burning: Atmospheric,climatic, and biospheric implications. Cambridge, MA: MIT Press.569 p.

Maler, K.G.; Wyzga, R.E. 1976. Economic measurements of environ-mental damage: a technical handbook. Paris: Organization forEconomic Cooperation and Development.

MacFarland, K.K.; Malm, W.; Molenar, J. 1983. An examination ofmethodologies for assessing the value of visibility. In: Rowe, R.D.;Chestnut, L.G., eds. Managing air quality and scenic resources atNational Parks and wilderness areas. Boulder, CO: WestviewPress: 151–172.

Malm, W.C. 2000. Introduction to visibility. #CA-2350-97. Ft. Collins,CO: Colorado State University, Cooperative Institute for Re-search in the Atmosphere: T097-04, T098-06.

Materna, B.L.; Jones, J.R.; Sutton, P.M. [and others]. 1992. Occupa-tional exposures in California wildland fire fighting. AmericanIndustrial Hygiene Association Journal. 53(1): 69–76.

Martins, J.V.; Artaxo, P.; Hobbs, P.V.; Liousse, C.; Cachier, H.;Kaufman, Y.; Plana-Fattori, A. 1996. Particle size distributions,elemental compositions, carbon measurements, and optical prop-erties of smoke from biomass burning in the Pacific Northwest ofthe United States. In: Levin, J.S., ed. Biomass burning and globalchange, volume 2: Biomass burning in South America, southeastAsia, and temperate and boreal ecosystems, and the oil fires ofKuwait. Cambridge, MA: The MIT Press: 716–732.

Mathur, R.; Peters, L.K.; Saylor, R.D. 1992. Sub-grid presentationof emission source clusters in regional air quality monitoring.Atmospheric Environment. 26A: 3219–3238.

Mauzerall, D.L.; Jacob, D.J.; Fan, S.M.; Bradshaw, J.D.; Gregory,G.L.; Sachse, G.W.; Blake, D.R. 1996. Origin of troposphericozone at remote high northern latitudes in summer. Journal ofGeophysical Research. 101(D2): 4175–4188.

Mauzerall, D.L.; Logan, J.A.; Jacob, D.J.; Anderson, B.E.; Blake,D.R.; Bradshaw, J.D.; Heikes, B.; Sachse, G.W.; Singh, H.; Talbot,R. 1998. Photochemistry in biomass burning plumes and implica-tions for tropospheric ozone over the tropical South Atlantic.Journal of Geophysical Research. 103: 8401–8423.

McClelland, G.; Schulze, W.; Waldman, D.; Irwin, J.; Schenk, D.;Stewart, T.; Deck, L.; Thayer, M. 1991. Valuing eastern visibility;a field test of the contingent valuation method. Cooperativeagreement CR-815183-01-3. Washington, DC: Draft report to theU.S. Environmental Protection Agency.

McKenzie, L.; Hao, W.M.; Richards, G.; Ward, D. 1994 Quantifica-tion of major components emitted from smoldering combustion ofwood. Atmospheric Environment. 28(20): 3285–3292.

Middleton, P.; Stewart, T.R.; Dennis, R.L. 1983. Modeling humanjudgments of urban visual air quality. Atmospheric Environ-ment. 17: 1015–1022.

Middleton, P.; Stewart, T.R.; Leary, J. 1985. On the use of humanjudgment and physical/chemical measurements in visual airquality management. Atmospheric Environment. 12: 1195–1208.

Mobley, Hugh E. 1976. Smoke management—What is it? In: South-ern Forest Fire Laboratory Personnel. Southern smoke manage-ment guidebook. Gen. Tech. Rep. SE-10. Asheville, NC: U.S.Department of Agriculture, Forest Service, Southeastern ForestExperiment Station: 1–8.

Mobley, Hugh E. 1989. Summary of smoke-related accidents inthe South from prescribed fire (1979–1988). Technical Release90-R-11. American Pulpwood Association.

Morris, R.E.; Yocke, M.A.; Myers, T.C.; Mirabella, V. 1992. Over-view of the variable-grid urban airshed model (UAM-V). 85thAnnual Meeting of the A&WMA. Pittsburgh, PA: Air & WasteManagement Association.

Myer, T.C.; Guthrie, P.D.; Wu, S.Y. 1996. The implementation of aplume-in-grid module in the SARMAP air quality model (SAQM).SYSAPP-96-06, Systems Applications International, Inc. Sacra-mento, CA: California Air Resources Board, Technical SupportDivision.

National Interagency Fire Center. 2001a. National fire news: fireseason 2000. http://www.nifc.gov/fireinfo/2000/highlights.html.(August 8, 2002).

National Interagency Fire Center. 2001b. Prescribed fire statistics.http://www.nifc.gov/stats/prescribedfirestats.html. (August 8,2002).

National Interagency Fire Center. 2002. Wildland fires statistics.http://www.nifc.gov/stats/wildlandfirestats.html. (October 5, 2002).

National Research Council. 1993. Protecting visibility in nationalparks and wilderness areas: committee on haze in National Parksand wilderness areas, National Research Council. Washington,DC: National Academy Press. 446 p.

National Wildfire Coordinating Group. 1985. Smoke managementguide. PNW 420-2. NFES 1279. Boise, ID: National InteragencyFire Center, National Interagency Coordinating Group, Pre-scribed Fire and Fire Effects Working Team. 28 p.

New Mexico Environmental Improvement Board. 1995. Open burn-ing: 20-NMAC2.60. http://www.nmenv.state.nm.us/NMED_regs/aqb/20nmac2_60.html. (August 8, 2002).

New York Daily News. 1999. Florida fire stirs health warnings.April 20; Sports Final Edition; News Section. Page 2.

Nichols, Mary D. 1996. Memorandum dated May 30 to EPA Re-gional Air Directors. Subject: Areas Affected by PM-10 NaturalEvents. http://www.epa.gov/ttn/oarpg/t1/memoranda/nepol.pdf.

Odman, M.T.; Russell, A.G. 1991. Multiscale modeling of pollutanttransport and chemistry. Journal of Geophysical Research. 96(D4):7363–7370.

Olson, J.; Prather, M.; Berntsen, T.; Carmichael, G.; Chatfield, R.;Connell, P.; Derwent, R.; Horowitz, L.; Jin, S.; Kanakidou, M.;Kasibhatla, P.; Kotamarthi, R.; Kuhn, M.; Law, K.; Penner, J.;Perliski, L.; Sillman, S.; Stordal, F.; Thompson, A.; Wild, O. 1997.Results from the Intergovernmental Panel on Climate Changephotochemical model intercomparison (PhotoComp). Journal ofGeophysical Research. 102 (D5): 5979–5991.

Oregon Department of Forestry. 1992. Smoke management pro-gram directives, appendix 2. Salem, OR: Oregon Department ofForestry.

Ottmar, R.D.; Reinhardt, T.E.; Anderson, G.; DeHerrera, P.J. [Inpreparation]. Consume 2.1 user’s guide. Manuscript. On file with:Roger D. Ottmar, Pacific Northwest Research Station, Seattle,WA.

Ottmar, Roger D.; Burns, Mary F.; Hall, Janet N.; Hanson, Aaron D.1993. CONSUME users guide. Gen. Tech. Rep. PNW-GTR-304.Portland, OR: U.S. Department of Agriculture, Forest Service,Pacific Northwest Research Station. 117 p.

Ottmar, Roger D.; Sandberg, David V. 2000. Modification andvalidation of fuel consumption models for shrub and forestedlands in the Southwest, Pacific Northwest, Rockies, Midwest,Southwest, and Alaska. Abstract. Joint fire science programprinciple investigators meeting; 2000 October 3–5; Reno, NV.http://www.nifc.gov/joint_fire_sci/jointfiresci.html

Ottmar, Roger D.; Vihnanek, Robert E. 2000a. Photo series formajor natural fuel types of the United States—phase II. Abstract.Joint fire science program principle investigators meeting; 2000October 3–5; Reno, NV. http://www.nifc.gov/joint_fire_sci/jointfiresci.html

Peterson, J. 2000. Personal communication. December.Peterson, Janice L. 1987. Analysis and reduction of the errors of

predicting prescribed burn emissions. Seattle: University ofWashington. 70 p. Thesis.

Peterson, J.; Ward, D. 1990. An inventory of particulate matter andair toxic emissions from prescribed fire in the United States for1989. Seattle, WA: U.S. Department of Agriculture, Forest Ser-vice, Pacific Northwest Research Station.

Peterson, J.L.; Ward, D. 1992. An inventory of particulate matterand air toxic emissions from prescribed fires in the United Statesfor 1989. Final report. U.S. EPA Office of Air Quality Programsand Standards.

Peterson, Janice L.; Sandberg, David V. 1988. A national PM10emissions inventory approach for wildfires and prescribed fires.In: Mathai, C.V.; Stonefield, David H., eds. Transactions PM-10implementation of standards: an APCA/EPA international spe-cialty conference. Pittsburgh, PA: Air Pollution Control Associa-tion: 353–371.


Pickering, K.E.; Thompson, A.M.; Scala, J.R.; Tao, W.K.; Simpson,J. 1992. Ozone production potential following convective redistri-bution of biomass burning emissions. Journal of AtmosphericChemistry 14(1–4): 297–313.

Pielke, R.A.; Cotton, W.R.; Walko, R.L.; Tremback, C.J.; Lyons,W.A.; Grasso, L.D.; Nicholls, M.E.; Moran, M.D.; Wesley, D.A.;Lee, T.J.; Copeland, J.H. 1992. A comprehensive meteorologicalmodeling system—RAMS. Meteorology and Atmospheric Phys-ics. 49(1–4): 69–91.

Public Law 84-159. Air Pollution Control Act of 1955. Act of July 14,1955. 42 U.S.C. 7401, et seq. 69 Stat. 322.

Public Law 88-206. Clean Air Act of 1963. Act of December 17, 1963,77 Stat. 392.

Public Law 90-148. Air Quality Act of 1967. Act of November 1, 1967.42 U.S.C. 7401. 81 Stat. 485, 501.

Public Law 91-604. Clean Air Act Amendments of 1970. Act ofDecember 31, 1970. 42 USC 1857h-7 et seq.

Public Law 95-95. Clean Air Act as Amended August 1977. 42 U.S.C.s/s 1857 et seq.

Public Law 101-549. Clean Air Act as Amended. November 15, 1990.104 Stat. 2399.

Radke, L.F.; Lyons, J.H.; Hobbs, P.V.; Hegg, D.A.; Sandberg, D.V.;Ward, D.E. 1990. Airborne monitoring and smoke characteriza-tion of prescribed fires on forest lands in western Washington andOregon. Gen. Tech. Rep. PNW-GTR-251. Portland, OR: U.S.Department of Agriculture, Forest Service, Pacific NorthwestResearch Station. 81 p.

Reid, J.S.; Hobbs, P.V.; Ferek, R.J.; Blake, D.R.; Martins, J.V.;Dunlap, M.R.; Liousse, C. 1998. Physical, chemical and opticalproperties of regional hazes dominated by smoke in Brazil.Journal of Geophysical Research (SCAR-B Special Issue). 103:32,059–32,080.

Reinhardt, Elizabeth D.; Keane, Robert E. 2000. A national fireeffects prediction model—revision of FOFEM. Abstract. Joint firescience program principle investigators meeting; 2000 October3–5; Reno, NV. http://www.nifc.gov/joint_fire_sci/jointfiresci.html

Reinhardt, Elizabeth D.; Keane, Robert E.; Brown, James K. 1997.First Order Fire Effects Model: FOFEM 4.0, users guide. Gen.Tech. Rep. INT-GTR-344. Ogden, UT: U.S. Department of Agri-culture, Forest Service, Intermountain Research Station. 65 p.

Reinhardt, T.E. 2000. Effects of smoke on wildland firefighters.Seattle, WA: URS/Radian International. April 2000.

Reinhardt, T.E.; Ottmar, R.D. 2000. Smoke exposure at westernwildfires. Res. Pap. PNW-RP-525. Portland, OR: U.S. Depart-ment of Agriculture, Forest Service, Pacific Northwest ResearchStation. 72 p.

Reinhardt, Tim E.; Ottmar, Roger D.; Hallett, Michael J. 1999.Guide to monitoring smoke exposure of wildland firefighters.Gen. Tech. Rep. PNW-GTR-448. U.S. Department of Agriculture,Forest Service, Pacific Northwest Research Station. 15 p.

Reinhardt, Timothy E.; Ottmar, Roger D. 1997. Smoke exposureamong wildland firefighters: a review and discussion of currentliterature. Gen. Tech. Rep. PNW-GTR-373. U.S. Department ofAgriculture, Forest Service, Pacific Northwest Research Station.61 p.

Reinhardt, Timothy E.; Ottmar, Roger D.; Hanneman, Andrew J.S.2000. Smoke exposure among firefighters at prescribed burns inthe Pacific Northwest. Res. Pap. PNW-RP-526. U.S. Department ofAgriculture, Forest Service, Pacific Northwest Research Station. 45 p.

Reisner, J.; Wynne, S.; Margolin, L.; Linn, R. 2000. Coupled atmos-pheric-fire modeling employing the method of averages. MonthlyWeather Review. 128(10): 3683–3691.

Richardson, J.L.; Fishman, J.; Gregory, G.L. 1991. Ozone budgetover the Amazon: regional effects from biomass burning emis-sions. Journal of Geophysical Research. 96(D7): 13,073–13,087.

Riebau, A.R.; Fox, D.G.; Sestak, M.L.; Daily, B.; Archer, S.F. 1988.Simple approach smoke estimation model. Atmospheric Environ-ment. 22(4): 783–788.

Roger, C.F.; Hudson, J.G.; Zielinska, B.; Tanner, R.L.; Hallett, J.;Watson, J.G. 1991. Cloud condensation from biomass burning.In: Levin, J.S., ed. Global burning: atmospheric, climatic, andbiospheric implications. Cambridge, MA: MIT Press.

Ross, D. 1988. Effects of visual air quality on visitor experience. In:Air quality in the national parks: A summary of findings from theNational Park Service Air Quality Research and MonitoringProgram. Natural Resources Report 88-1. Chapter 3.

Ross, D.G.; Smith, I.N.; Mannis, P.C.; Fox, D.G. 1988. Diagnosticwind field modeling for complex terrain: model development andtesting. Journal of Applied Meteorology. 27: 785–796.

Rothermel, R.C. 1972. A mathematical model for predicting firespread in wildland fuels. Res. Pap. INT-115. Ogden, UT: U.S.Department of Agriculture, Forest Service, Intermountain For-est and Range Experiment Station. 40 p.

Rothwell, T.; Sharkey, B. 1995. The effect of an air-purifyingrespirator on performance of upper body work. In: Health haz-ards of smoke: fall 1995. U.S. Department of Agriculture, ForestService, Missoula Technology and Development Center: 56–65.

Sandberg, D.V. 1980. Duff reduction by prescribed underburning inDouglas-fir. Res. Pap. PNW-272. Portland, OR: U.S. Departmentof Agriculture, Forest Service, Pacific Northwest Forest andRange Experiment Station. 18 p.

Sandberg, David V. 2000. Implementation of an improved EmissionProduction Model. Abstract. Joint fire science program principleinvestigators meeting; 2000 October 3–5; Reno, NV. http://www.nifc.gov/joint_fire_sci/jointfiresci.html.

Sandberg, David V. 2002. Personal communications. SupervisoryResearch Biologist. U.S. Department of Agriculture, Forest Ser-vice, Pacific Northwest Research Station. October.

Sandberg, D.V.; Dost, F.N. 1990. Effects of prescribed fire on airquality and human health. In: Wasltad, J.W.; Radosevich, S.R.;Sandberg, D.V., eds. Natural and prescribed fire in Pacific North-west forests. Corvallis: Oregon State University Press: 191–218.

Sandberg, David V.; Hardy, Colin C.; Ottmar, Roger D.; Snell, J.A.Kendall; Acheson, Ann; Peterson, Janice L.; Seamon, Paula;Lahm, Peter; Wade, Dale. 1999. National strategic plan: Model-ing and data systems for wildland fire and air quality. U.S.Department of Agriculture, Forest Service, Pacific NorthwestResearch Station. 60 p.

Sandberg, David V.; Ottmar, R.D.; Cushon, G.H. 2001. Character-izing fuels in the 21st century. International Journal of WildlandFire. 10: 1–7.

Sandberg, D.V.; Peterson, J.L. 1984. A source strength model forprescribed fires in coniferous logging slash. Annual Meeting, AirPollution Control Association, Pacific Northwest Section. Re-print #84.20. Portland, OR: U.S. Department of Agriculture,Forest Service, Pacific Northwest Research Station. 10 p.

Sandberg, D.V.; Pierovich, J.M.; Fox, D.G.; Ross, E.W. 1979. Effectsof fire on air: a state-of-knowledge review. Gen. Tech. Rep. WO-9.U.S. Department of Agriculture, Forest Service. 40 p.

Schimel, D.S. 1995. Terrestrial ecosystems and the global carboncycle. Global Change Biology. 1: 77–71.

Schroeder, W.H.; Lane, D.A. 1988. The fate of toxic airborne pollut-ants. Environmental Science and Technology. 22(3): 240–246.

Schulze, William D.; Brookshire, David S.; Walther, Eric G.;MacFarland, Karen Kelley; Thayer, Mark A.; Whitworth, Regan L.;Ben-David, Shaul; Malm, William; Molenar, John. 1983. The eco-nomic benefits of preserving visibility in the national parklandsof the Southwest. Natural Resources Journal. 23: 149–173.

Schwartz, Joel; Dockery, Douglas W.; Neas, Lucas M. 1996. Is dailymortality associated specifically with fine particles? Journal ofthe Air & Waste Management Association. 46: 927–939.

Scire, J.; Robe, F.R.; Fernau, M.E.; Yamartino, R.J. 2000a. A user’sguide for CALMET meteorological model. Concord, MA: EarthTech, Inc. 332 p.

Scire, J.; Strimaitis, D.G.; Yamartino R.J.; Xiaomong, Zhang.2000b. A user’s guide for CALPUFF dispersion model (Version 5).Concord, MA: Earth Tech, Inc. 512 p.

Seigneur, C.; Tesche, T.W.; Roth, P.M.; Liu, M.K. 1983. On thetreatment of point source emissions in urban air quality. Atmos-pheric Environment. 17(9): 1655–1676.

Sestak, M.L.; Riebau, A.R. 1988. SASEM, Simple approach smokeestimation model. U.S. Bureau of Land Management, TechnicalNote 382. 31 p.

Sestak, M.L.; Marlatt, W.E.; Riebau, A.R. 1988. Unpublished draft.VALBOX: ventilated valley box model. U.S. Bureau of LandManagement. 32 p.

Sharkey, Brian, ed. 1997. Health hazards of smoke: recommenda-tions of the April 1997 Consensus Conference. Tech. Rep. 9751-2836-MTDC. Missoula, MT: U.S. Department of Agriculture,Forest Service, Missoula Technology and Development Center.84 p.


Shelby, B.; Speaker, R.W. 1990. Public attitudes and perceptionsabout prescribed burning. In: Wasltad, J.W.; Radosevich, S.R.;Sandberg, D.V., eds. Natural and prescribed fire in Pacific North-west forests. Corvallis: Oregon State University Press: 253–260.

Sisler, James F.; Malm, William C.; Gebhart, Kristi A. 1996. Spatialand seasonal patterns and long term variability of the composi-tion of the haze in the United States: an analysis of data from theIMPROVE network. Cooperative Institute for Research in theAtmosphere, Colorado State University. ISSN: 0737-5352-32.[Pages unknown].

Southern Forest Fire Laboratory Personnel. 1976. Southern for-estry smoke management guidebook. Gen. Tech. Rep. SE-10.Asheville, NC: U.S. Department of Agriculture, Forest Service,Southeastern Forest Experiment Station. 140 p.

Standley, L. J.; Simoneit, B. R. T. 1987. Characterization of extract-able plant wax, resin and thermally matured components insmoke particles from prescribed burns. Environmental Scienceand Technology. 21: 163–169.

Stith, J.L.; Radke, F.L.; Hobbs, P.V. 1981. Particle emissions andthe production of ozone and nitrogen oxides from the burning offorest slash. Atmospheric Environment. 7: 73–82.

Systems Applications International. 2002. User’s guide to the regu-latory modeling system for aerosols and deposition (REMSAD).SYSAPP98-96/42r2. Contract 68D30032. U.S. EnvironmentalProtection Agency. http://remsad.saintl.com/ or http://www.epa.gov/scram001/

Thompson, A.M.; Pickering, K.E.; McNamera, D.P.; Schoeberl, M.R.;Hudson, R.D.; Kim, J.H.; Browell, E.V.; Kirchoff, V.W.J.H.;Ngana, D. 1996. Where did tropospheric ozone over southernAfrica and the tropical Atlantic come from in October 1992?Insights from TOMS, GTE TRACE A, and SAFARI 1992. Journalof Geophysical Research. 101(D19):24,251–24,278.

Trent, Andy; Davies, Mary Ann; Fisher, Rich; Thistle, Harold;Babbitt, Ronald. 2000. Evaluation of optical instruments for real-time continuous monitoring of smoke particles. Tech. Rep. 9925-2806-MTDC. Missoula, MT: U.S. Department of Agriculture,Forest Service, Technology and Development Program. 38 p.

Trent, Andy; Thistle, Harold; Fisher, Rich; Babbitt, Ronald;Holland-Sears, Andria. 1999. Laboratory evaluation of twooptical instruments for real-time particulate monitoring of smoke.Tech. Rep. 9925-2806-MTDC. Missoula, MT: U.S. Department ofAgriculture, Forest Service, Technology and Development Pro-gram. 38 p.

Trijonis, J.; Charlson, R.; Husar, R.; Malm, W.C.; Pitchford, M.;White, W. 1991. Visibility: existing and historical conditions—causes and effects. In: Acid deposition: state of science andtechnology: Report 24. National Acid Precipitation AssessmentProgram. Washington, DC: Government Printing Office. [Pagesunknown].

U.S. Code Title 42, Chapter 85. Air Pollution Prevention andControl, As Amended.

USA Today. 1999. Fires blaze in six Western states. http://www.usatoday.com/weather/news/1999/w827fire.htm. August 27,1999.

U.S. Department of Agriculture. 1997. Course to the future: the RPAprogram. http://www.fs.fed.us/land/RPA/chp3sec1.htm. April1997.

U.S. Department of Agriculture and U.S. Department of the Inte-rior. 2000. A report to the President in response to the wildfiresof 2000. http://www.fireplan.gov/president.cfm. September 2002.

U.S. Department of the Interior; U.S. Department of Agriculture.1995. Federal wildland fire management policy and programreview. Final report. Boise, ID: Bureau of Land Management.45 p.

U.S. Department of the Interior; U.S. Department of Agriculture;Department of Energy; [and others]. 2001. Review and update ofthe 1995 federal wildland fire management policy. Boise, ID:Bureau of Land Management. 78 p.

U.S. Environmental Protection Agency. 1972. Compilation of airpollutant emission factors. OAP Publ. AP-42. Research TrianglePark, NC: U.S. Environmental Protection Agency.

U.S. Environmental Protection Agency. 1990. Air quality criteriafor particulate matter: Volume II of III. EPA/600/P-95/00ibF.Washington, DC: U.S. Environmental Protection Agency, Officeof Research and Development, National Center for Environmen-tal Assessment: 8-82–8-89.

U.S. Environmental Protection Agency. 1992a. Prescribed burningbackground document and technical information document forprescribed burning best available control measures. EPA-450/2-92-003. Office of Air Quality Planning and Standards. September.

U.S. Environmental Protection Agency. 1992b. Prevention of airpollution emergency episodes. 40 CFR 51 Appendix L.

U.S. Environmental Protection Agency. 1996. Review of the na-tional ambient air quality standard for particulate matter: Policyassessment of scientific and technical information. EPA-452\R-96-013. Washington, DC: U.S Environmental Protection Agency,Office of Air Quality Planning and Standards. July.

U.S. Environmental Protection Agency. 1998. Interim air qualitypolicy on wildland and prescribed fires.

U.S. Environmental Protection Agency, Office of Air Quality Plan-ning and Standards. April 23. 29 p.

U.S. Environmental Protection Agency. 1999. Guideline for report-ing of daily air quality – air quality index (AQI). EPA-454/R-99-010. Research Triangle Park, NC: Office of Air Quality Planningand Standards. 25 p.

U.S. Environmental Protection Agency. 2000a. Compilation of airpollutant emission factors AP-42, fifth edition, volume I: station-ary point and area sources. Research Triangle Park, NC: U.S.Environmental Protection Agency. January 1995–September2000. AP-42 reference.

U.S. Environmental Protection Agency. 2000b. National ambientair quality standards (NAAQS). http://www.epa.gov/airs/criteria.html. 5 December 2000.

U.S. Environmental Protection Agency. 2000c. Wildland fire issuesgroup. http://www.epa.gov/ttncaaa1/faca/fa08.html. December 5,2000.

U.S. Environmental Protection Agency. 2002. Models-3 Air QualityModeling System. http://www.epa.gov/asmdnerl/models3/July2002.

Venkatram, A. 1988. Topics in applied dispersion modeling. In:Venkatram, A.; Wyngaard, eds. Lectures on air pollution model-ing. Boston, MA: American Meteorological Society: 267–324.

Wade, Dale D.; Lunsford, James D. 1988. A guide to prescribed firein southern forests. Tech. Publ. R8-TP-11. NFES 2108. U.S.Department of Agriculture, Forest Service, Southern Region.Boise, ID: Publication Management System.

Walcek, C.J. 2002. Effects of wind shear on pollution dispersion.Atmospheric Environment. 36: 511–517.

Walton, W.D.; McGrattan, K.B.; Mullin, J.V. 1996. ALOFT-PC: Asmoke plume trajectory model for personal computers. In: Pro-ceedings 19th Arctic and Marine Oilspill Programme (AMOP)Technical Seminar: Volume 2. Ottawa, Ontario, Canada: Envi-ronment Canada: 987–997.

Ward, D.E.; Hardy, C.C.; Sandberg, D.V.; Reinhardt, T.E. 1989.Part III-emissions characterization. In; Sandberg, D.V.; Ward,D.E.; Ottmar, R.D., comp. eds. Mitigation of prescribed fireatmospheric pollution through increased utilization of hard-woods, piled residues, and long-needled conifers. Final report.U.S. DOE, EPA. Seattle, WA: U.S. Department of Agriculture,Forest Service, Pacific Northwest Research Station.

Ward, D.E.; McMahon, C.K.; Johansen, R.W. 1976. An update onparticulate emissions from forest fires. In: Transactions of the69th annual meeting of the Air Pollution Control Association.

Ward, Darold E.; Hardy, Colin C. 1991. Smoke emissions fromwildland fires. Environmental International. 17: 117–134.

Washington Department of Natural Resources. 1993. Smoke man-agement program, Appendix 5. Olympia, WA.

Watson, J.W. 1997. Review and comment on observational models—the use of chemical mass balance methods. In: Proceedings of theEPA source attribution workshop. Research Triangle Park, NC:U.S. Environmental Protection Agency.

Watson, John G.; Robinson, Norman F.; Chow, Judith C.; [andothers]. 1990. CMB7 user’s manual. Receptor model technicalseries, volume III (1989 revision). EPA-450/4-90-004. ResearchTriangle Park, NC: U.S. Environmental Protection Agency, Of-fice of Air and Radiation, Office of Air Quality Planning andStandards.

WESTAR. 1995. Wildfire emergency action plans draft report.Portland, OR: Western States Air Resources Council.


Western Governors’ Association. 2001. A collaborative approach forreducing wildland fire risks to communities and the environ-ment: 10-year comprehensive strategy. http://www.westgov.org/wga/initiatives/fire/final_fire_rpt.pdf. August 2001.

Williamson, Samuel J. 1973. Fundamentals of air pollutions. Read-ing, MA: Addison-Wesley Publishing Company. 472 p.

Wofsy, S.C.; Sachse, G.W.; Sachse, G.L.; Blake, D.L.; Bradshaw,J.D.; Singh, H.B.; Barrick, J.A.; Harriss, R.C.; Talbot, R.W.;Shipman, M.A.; Browell, E.V.; Jacob, D.J.; Logan, J.A. 1992.Atmospheric chemistry in the Arctic and Subarctic: influence ofnatural fires, industrial emissions, and stratospheric inputs.Journal of Geophysical Research. 97: 16,731–16,746.

Wotawa, G.; Trainer, M. 2000. The influence of Canadian forestfires on pollutant concentrations in the United States. Science.288: 324–328.

Yamate, G.; Stockham, J.; Vatavuk, W.; Mann, C. 1975. An inven-tory of emissions from forest wildfires, forest managed burns andagricultural burns. In: Transactions of the 68th annual meeting ofthe Air Pollution Control Association. June 15–20, 1975.

Yocom, J.E.; Upham, J.B. 1977. Effects of economic materials andstructures. In: Stern, Arthur C., ed. Air pollution: The effects ofair pollution. 3rd ed. New York, NY: Academic Press, Inc.: 93–94.

Yokelson, R.J.; Griffith, D.W.T.; Ward, D.E. 1996. Open-path Fou-rier transform infrared studies of large scale laboratory biomassfires. Journal of Geophysical Research. 101(D15): 20,167–21,080.

Yokelson, R.J.; Ward, D.E.; Susott, R.A.; Reardon, J.; Griffith,D.W.T. 1997. Emissions from smoldering combustion of biomassmeasured by open-path Fourier transform infrared spectroscopy.Journal of Geophysical Research. 102(D15): 18,865–18,877.


A

Africa. See World: Africa: tropicalair pollutants 2, 3, 9, 10, 14, 27

carbon dioxide (CO2) 24, 27, 61carbon monoxide

(CO) 9, 10, 11, 20, 31, 42, 49, 52, 53, 56, 57, 58criteria 10, 11, 29, 32, 36, 49, 66hazardous air pollutants

(HAP) 5, 9, 16, 29, 32, 45, 67hydrocarbons 9, 10, 32particulate matter (PM) 45PM10 11, 13, 16, 20, 24, 25, 48, 49PM2.5 4, 11, 13, 24, 41, 47, 48, 49, 51primary 9secondary 9

air pollution 1, 2, 4, 5, 9, 10, 11, 13, 15, 27, 45, 46, 48,57, 63, 65, 66

Alabama. See United States of America: AlabamaAlaska. See United States of America: AlaskaALOFT-FT. See models: ALOFT-FTambient air 2, 9, 10, 19, 20, 24, 35, 36, 41, 49, 58, 66Appalachian Mountains. See United States of America:

Appalachian MountainsArizona. See United States of America: ArizonaArkansas. See United States of America: ArkansasAsia. See World: Asia: tropical

B

BEHAVE. See models: BEHAVEBURNUP. See models: BURNUP

C

Cal/PFIRS. See models: Cal/PFIRSCalifornia. See United States of America: CaliforniaCalpuff. See models: CalpuffCanada. See World: Canadacarbon dioxide. See air pollutantscarbon monixide. See air pollutantsCentral States Regional Air Partnership. See regional

planning organizationsClean Air Act. See laws and regulationsCMAQ. See modelsColorado. See United States of America: Coloradocombustion stages 27, 30, 31, 42, 49, 65

flaming 29, 30, 31, 32, 36, 37residual 29, 30, 31, 32, 37smoldering 29, 30, 31, 32, 36, 37

Consume. See modelscontrol measures 16, 64

E

emission factor 27, 29, 32, 43, 45, 46, 66emission inventory 2, 45, 46, 47, 48, 64, 65

emission production model. See also modelsemission(s)

anthropogenic 1, 9, 15, 49, 51, 66chemistry 36estimating 27, 29greenhouse gas 61natural 1, 5, 9, 15, 64, 66prescribed fire 5, 7, 13, 16, 64rates 29, 32, 38, 40. See also source strengthredistribution 17reduction 4, 6, 16, 17, 45, 46, 48, 65wildland fire 5, 7

EPA 4, 5. See U.S. Environmental Protection AgencyEPM. See models: Emission production model (EPM)

F

FARSITE. See modelsFASTRACS. See modelsFCAMMS. See Fire Consortia for Advanced Modeling of

MeteorologyFCC. See fuel: fuel characteristic classification system

(FCC)Federal Wildland Fire Policy 1, 6FEJF. See Fire Emissions Joint Forum (FEJF)fire behavior 15, 27, 29, 36, 63, 64, 65Fire Consortia for Advanced Modeling of Meteorology

15. See regional planning organizationsFire Emissions Joint Forum (FEJF) 26FIRETEC. See modelsFlorida. See United States of America: FloridaFOFEM. See modelsfuel

arrangement 46characteristics 27, 28, 31, 32, 36, 63, 65consumption 7, 17, 27, 28, 30, 31, 32, 36, 46,

47, 64fuel characteristic classification system (FCC) 28loading 16, 28, 31, 37, 46moisture 27, 46

G

GCVTC. See Grand Canyon Visibility Transport Commission(GCVTC)

general conformity. See laws and regulationsGeorgia. See United States of America: Georgiaglobal impacts

biomass burning and emissions 19, 36, 42, 43, 61global change 36greenhouse gases 24, 66

Grand Canyon. See United States of America: Grand CanyonGrand Canyon Visibility Transport Commission

(GCVTC) 24, 46, 47, 59

Index


H

HAP. See air pollutants: hazardous air pollutants (HAP)health 2, 4, 6, 11, 16, 20, 24, 35, 45, 55, 56, 57, 63, 66

community 7effects 1, 11, 45, 55, 56, 57, 65firefighter 65human 6, 9, 11, 16public 6, 7, 20risks 5, 45, 57, 64

HIGRAD. See modelshydrocarbons. See air pollutantsHysplit. See models

I

Idaho. See United States of America: IdahoIllinois. See United States of America: IllinoisIMPROVE 48, 49, 51Iowa. See United States of America: Iowa

J

Joint Fire Science Program (JFSP) 2, 6, 64, 67

L

laws and regulationsClean Air Act 1, 4, 6, 9, 10, 11, 12, 13, 15, 16, 49, 58, 59

roles and responsibilities 10general conformity 10, 11National Environmental Policy Act (NEPA) 4

liability. See safetyLouisiana. See United States of America: Louisiana

M

Massachussetts. See United States of America: Massachusettsmeteorology 4, 10, 60, 64Michigan. See United States of America: MichiganMidwest Regional Planning Organization. See regional

planning organizationsMississippi. See United States of America: Mississippimodel types

chemical mass balance (CMB) 51dispersion 32, 35, 36, 37, 38, 40, 50, 64, 65, 66grid 32, 37, 39, 40particle 39plume 37, 38, 39, 40, 60puff 37, 39, 40scavenging 38transport 35, 37

modelsALOFT-FT 39BEHAVE 29BURNUP 29, 33, 36, 37Cal/PFIRS 48CALMET 38, 64Calpuff 36, 39, 40, 64CMAQ 32Consume 31, 32Emission production model (EPM) 29, 32, 36, 37FARSITE 29, 33, 36FASTRACS 48

FIRETEC 29FOFEM 31, 32HIGRAD 29Hysplit 39, 40MM5 38Models-3/CMAQ 32, 36, 38, 40, 41NFSpuff 39, 40NUATMOS 38PB-Piedmont 39, 61RAMs 38REMSAD 40SASEM 36, 37, 39, 40, 64TSARS+ 38VSMOKE 36, 37, 39, 60

Models-3/CMAQ. See modelsmonitoring 3, 4, 5, 11, 15, 17, 25, 26, 45, 48, 49, 50,

51, 52, 53, 57, 63, 64, 66Montana. See United States of America

N

NAAQS 4, 5. See national ambient air quality standards(NAAQS)

national ambient air quality standards(NAAQS) 4, 5, 10, 11, 16, 17, 19, 24, 25, 48, 49, 60, 64

National Fire Plan 7, 67National Interagency Fire Center (NIFC) 6, 24National Wildfire Coordinating Group (NWCG) 3, 4NEAP. See smoke management planning: natural events

action plan (NEAP)NEPA. See laws and regulations: National Environmental

Policy Act (NEPA)Nevada. See United States of America: NevadaNew Mexico. See United States of America: New MexicoNew York. See United States of America: New YorkNFSpuff. See modelsNIFC. See National Interagency Fire Center (NIFC)nonattainment 10, 11, 16, 24, 25, 64North Carolina. See United States of America: North CarolinaNWCG. See National Wildfire Coordinating Group (NWCG)

O

Oregon. See United States of America: OregonOTC. See regional planning organizationsozone 5, 10, 11, 17, 39, 40, 41, 42, 43, 45, 49, 65, 66Ozone Transport Commission. See regional planning

organizations

P

Pacific Northwest. See United States of America: PacificNorthwest

particulate matter 4, 5, 11, 20, 24, 32, 48, 49, 50, 55,56, 59, 64

particulates 9, 10PB-Piedmont. See modelsPFIRS. See smoke management programs: Prescribed Fire

Incident Reporting Systemphoto series 28Piedmont. See United States of America: Piedmontplume 17, 23, 24, 27, 29, 36, 37, 38, 39, 40, 41, 42, 43,

50, 58, 59, 61


chemistry 38, 39, 42plume rise 35, 36, 37, 39, 40, 66

prevention of significant deterioration (PSD) 10, 12, 13

R

RAMs. See modelsregional haze 4, 5, 10, 13, 14, 15, 17, 24, 26, 36, 39,

40, 41, 43, 45, 46, 48, 51, 53, 64, 67regional planning organizations 14, 40

Central States Regional Air Partnership 14Midwest Regional Planning Organizationrship 14Ozone Transport Commission 41Visibility Improvement State and Tribal Associatio 14

REMSAD. See modelsresearch questions 1, 2, 4, 7, 10, 17, 31, 37, 40, 42,

43, 47, 48, 57, 59, 63, 64, 65, 66, 67Rocky Mountain. See United States of America: Rocky

MountainRussia. See World: Russia

S

safety 1, 4, 6, 16, 63, 66firefighter 7highway 23, 58, 60, 66liability 5, 25public 6, 25, 60roadway 23, 60

SASEM. See modelsscavenging 36, 37, 38, 40SIP. See smoke management programs: State

implementation plan (SIP)smoke

dispersion 4, 15, 17, 26, 36prescribed fire 25, 52, 60reduction. See emission(s): reductiontransport 6, 14, 25, 26wildland fire 49, 52, 56

smoke management guidancesmoke management guide for prescribed and wildland

fire 2, 3, 4, 5, 17southern smoke management guidebook 2, 6

smoke management guide for prescribed and wildlandfire. see smoke management guidance

smoke management planning 16enhanced smoke management plan 26natural events action plan (NEAP) 16, 24

smoke managementprograms 4, 5, 15, 16, 17, 19, 24, 25, 58

Prescribed Fire Incident Reporting System 26State implementation plan

(SIP) 5, 10, 11, 16, 17, 40, 46Tribal implementation plan (TIP) 10, 40

source strength 2, 32, 33, 36, 39, 40, 52, 64, 65, 66Southern Smoke Management Guidebook. See smoke

management guidanceSouth America. See World: South AmericaSouth Carolina. See United States of America: South

Carolina

T

Tennessee. See United States of America: TennesseeTexas. See United States of America: TexasTIP. See smoke management programs: Tribal implementa-

tion plan (TIP)

U

U.S. Environmental Protection Agency 4, 5, 7, 10, 12,14, 16, 17, 19, 20, 24, 25, 39, 41, 47, 48, 55, 57, 58, 59

United States of AmericaAlabama 25Alaska 19, 25, 40, 42, 63Appalachian Mountains 61Arizona 24, 25, 39, 46Arkansas 25California 15, 19, 20, 24, 25, 26, 46, 48, 58

Central Valley 20San Joaquin Valley 26southern 58

Colorado 25, 39, 46Colorado Plateau 59Eastern Seaboard 20Eastern U.S. 51Florida 19, 20, 25, 46, 60Georgia 24, 25Grand Canyon 59Idaho 20, 24, 46

Salmon 20Illinois 25Intermountain West 19Iowa 58Louisiana 25Massachusetts 25Michigan 46Midwestern States 41Mississippi 25, 60Montana 20, 24, 46Montana/Idaho 15, 25Nevada 20, 25, 26, 46

Reno 20New Mexico 20, 25, 39, 46, 58New York 25North Carolina 25Northeastern 15Northwestern U.S. 51Ohio 60Ohio River 60Oregon 12, 20, 24, 25, 46, 48, 58

southern 20Pacific Coast 20Pacific Northwest 15, 48, 51Piedmont 61Rocky Mountain 15Rocky Mountain States 20South Carolina 25, 46Southeastern 15, 19, 20, 23, 24, 60Southern U.S. 60Southwest 20, 51Tennessee 25, 41, 46


Texas 24, 25, 60Utah 25, 46Virginia 25, 60Washington 20, 24, 25, 46, 48, 58Western Regional Air Partnership

(WRAP) 14, 26, 40, 41, 67Western U.S. 15, 19Wyoming 25, 39Yellowstone National Park 20

Utah. See United States of America: Utah

V

Virginia. See United States of America: Virginiavisibility 2, 4, 6, 7, 10, 13, 14, 15, 19, 20, 24, 25, 26, 35,

39, 41, 45, 49, 50, 58, 59, 60, 64, 65, 66Class I

areas 10, 12, 14, 15, 24, 25, 47, 49, 50, 58, 59, 65visibility impairment and reduction 1, 20, 46, 58, 60

VISTAS. See regional planning organizationsVSMOKE. See models

W

Washington. See United States of America: Washingtonwelfare 2, 4, 6, 9, 11, 16, 35, 45, 55, 58, 66Western Regional Air Partnership (WRAP) 14, 67World

Africatropical 24

Asiatropical 24

Canada 20, 25, 42, 49Russia 25South America

tropical 25WRAP. See Western Regional Air Partnership (WRAP)Wyoming. See United States of America: Wyoming

Y

Yellowstone National Park. See United States of America:Yellowstone National Park

Federal Recycling Program Printed on Recycled Paper

The U.S. Department of Agriculture (USDA) prohibits discrimination in all itsprograms and activities on the basis of race, color, national origin, sex, religion,age, disability, political beliefs, sexual orientation, or marital or family status. (Notall prohibited bases apply to all programs.) Persons with disabilities who requirealternative means for communication of program information (Braille, large print,audiotape, etc.) should contact USDA’s TARGET Center at (202) 720-2600 (voiceand TDD).

To file a complaint of discrimination, write USDA, Director, Office of Civil Rights,Room 326-W, Whitten Building, 1400 Independence Avenue, SW, Washington,DC 20250-9410 or call (202) 720-5964 (voice or TDD). USDA is an equalopportunity provider and employer.

The Rocky Mountain Research Station develops scientific informationand technology to improve management, protection, and use of theforests and rangelands. Research is designed to meet the needs ofNational Forest managers, Federal and State agencies, public andprivate organizations, academic institutions, industry, and individuals.

Studies accelerate solutions to problems involving ecosystems,range, forests, water, recreation, fire, resource inventory, land recla-mation, community sustainability, forest engineering technology,multiple use economics, wildlife and fish habitat, and forest insectsand diseases. Studies are conducted cooperatively, and applicationsmay be found worldwide.

Research Locations

Flagstaff, Arizona Reno, NevadaFort Collins, Colorado* Albuquerque, New MexicoBoise, Idaho Rapid City, South DakotaMoscow, Idaho Logan, UtahBozeman, Montana Ogden, UtahMissoula, Montana Provo, UtahLincoln, Nebraska Laramie, Wyoming

*Station Headquarters, Natural Resources Research Center,2150 Centre Avenue, Building A, Fort Collins, CO 80526

RMRSROCKY MOUNTAIN RESEARCH STATION

united states department wildland fire in forest service … · 2008. 5. 21. · united states...

Documents