smoke management guide for - us forest servicesmoke management guide for prescribed and wildland...

SMOKE MANAGEMENT GUIDE FORPRESCRIBED AND WILDLAND FIRE2001 Edition

EDITORS/COMPILERS: PRODUCED BY:Colin C. Hardy National Wildfire Coordinating GroupRoger D. Ottmar Fire Use Working TeamJanice L. PetersonJohn E. CorePaula Seamon

Additional copies of this publication may be ordered by mail/fax from:National Interagency Fire Center, ATTN: Great Basin Cache Supply Office3833 S. Development AvenueBoise, Idaho 83705. Fax: 208-387-5573.Order NFES 1279

An electronic copy of this document is available at http://www.nwcg.gov.________________________________________________________________________The National Wildfire Coordinating Group (NWCG) has developed this information for the guidance of its member agencies andis not responsible for the interpretation or use of this information by anyone except the member agencies. The use of trade, firm,or corporation names in this publication is for the information and convenience of the reader and does not constitute an endorse-ment by NWCG of any product or service to the exclusion of others that may be suitable.

– i –

2001 Smoke Management Guide Forward

Forward

The National Wildfire Coordinating Group’s (NWCG) Fire Use Working Team1 has assumed overallresponsibility for sponsoring the development and production of this revised Smoke ManagementGuide for Prescribed and Wildland Fire (the “Guide”). The Mission Statement for the Fire Use Work-ing Team includes the need to coordinate and advocate the use of fire to achieve management objec-tives, and to promote a greater understanding of the role of fire and its effects. The Fire Use WorkingTeam recognizes that the ignition of wildland fuels by land managers, or the use of wildland firesignited by natural causes to achieve specific management objectives is receiving continued emphasisfrom fire management specialists, land managers, environmental groups, politicians and the generalpublic. Yet, at the same time that fire use programs are increasing, concerns are being expressedregarding associated “costs” such as smoke management problems. This revised Guide is the Fire UseWorking Team’s contribution to a better national understanding and application of smoke management.

Bill Leenhouts—ChairNWCG Fire Use Working Team

___________________________________

1 The NWCG website [http://www.nwcg.gov] contains documentation and descriptions for all NWCG working teams.

Preface 2001 Smoke Management Guide

– ii –

– iii –

2001 Smoke Management Guide Preface

Preface

The National Wildfire Coordinating Group’s Fire Use Working Team sponsored this 2001 edition of theSmoke Management Guide for Prescribed and Wildland Fire. A six-member steering committee wasresponsible for development of a general outline and for coordination of the Guide’s production. Theeditors/compilers invited the individual contributions, edited submissions, authored many of the sec-tions, obtained comprehensive reviews from the NWCG agencies and other partners, and compiled thefinal material into a cohesive guidebook.

Steering Committee: Bill Leenhouts (chair, NWCG Fire Use Working Team), Colin C. Hardy, RogerD. Ottmar, Janice L. Peterson, John E. Core, Paula Seamon.

Authors:

Gary Achtemeier, Research Meteorologist,USDA Forest Service, Southern ResearchStation. Athens, GA

James D. Brenner, Fire Management Adminis-trator, State of Florida Dept. of Agriculture andConsumer Service, Division of Forestry,Tallahassee, FL

John E. Core, Consultant, Core EnvironmentalConsulting. Portland, OR

Sue A. Ferguson, Research Atmospheric Scien-tist, USDA Forest Service, Pacific NorthwestResearch Station. Seattle, WA

Colin C. Hardy, Research ForesterUSDA Forest Service, Rocky Mountain Re-search Station. Missoula, MT

Sharon M. Hermann, Research Ecologist,Department of Biological Sciences, AuburnUniversity. Auburn, AL

Bill Jackson, Air Resource Specialist, USDAForest Service, Region 8, Asheville, NC

Peter Lahm, Air Resource Program Manager,USDA Forest Service, Arizona National Forests.Phoenix, AZ

Bill Leenhouts, Fire Ecologist, USDI Fish andWildlife Service, National Interagency FireCenter. Boise, ID

Tom Leuschen, Owner—Fire Vision, USDAForest Service, Okanagon National Forest.Okanagon, WA

Robert E. Mutch, Consultant Forester, Sys-tems for Environmental Management. Missoula,MT

Roger D. Ottmar, Research Forester, USDAForest Service, Pacific Northwest ResearchStation. Seattle, WA

Janice L. Peterson, Air Resource Specialist,USDA Forest Service, Mt. Baker-SnoqualmieNational Forest. Mountlake Terrace, WA

Timothy R. Reinhardt, Industrial Hygienist,URS Corp. Bellevue, WA

Paula Seamon, Fire Management Coordinator,The Nature Conservancy, Fire ManagementProgram. Tallahassee, FL

Dale Wade, Research Forester,USDA Forest Service, Southern ResearchStation. Athens, GA

Preface 2001 Smoke Management Guide

– iv –

– v –

2001 Smoke Management Guide Table of Contents

Table of Contents

Forward ..................................................................................................... iBill Leenhouts—Chair, NWCG, Fire Use Working Team

Preface .................................................................................................... iii

1.0 Introduction .............................................................................................3Colin C. HardyBill Leenhouts

2.0 Overview

2.1 The Wildland Fire Imperative .............................................................11Colin C. HardySharon M. HermannRobert E. Mutch

2.2 The Smoke Management Imperative ...................................................21Colin C. HardySharon M. HermannJohn E. Core

3.0 Smoke Impacts3.1 Public Health and Exposure to Smoke ......................................27

John E. CoreJanice L. Peterson

3.2 Visibility .........................................................................................35John E. Core

3.3 Problem and Nuisance Smoke....................................................41Gary L. AchtemeierBill JacksonJames D. Brenner

3.4 Smoke Exposure Among Fireline Personnel ............................51Roger D. OttmarTimothy R. Reinhardt

Table of Contents 2001 Smoke Management Guide

– vi –

4.0 Regulations4.1 Regulations For Smoke Management ......................................... 61

Janice L. Peterson

4.2 State Smoke Management Programs .......................................... 75John E. Core

4.3 Federal Land Management - Special Requirements ................. 81Janice L. Peterson

5.0 Smoke Source Characteristics ............................................................ 89Roger D. Ottmar

6.0 Fire Use Planning ................................................................................ 109Tom LeuschenDale WadePaula Seamon

7.0 Smoke Management Meteorology ..................................................... 121Sue A. Ferguson

8.0 Smoke Management:Techniques to Reduce or Redistribute Emissions ......................... 141

Roger D. OttmarJanice L. PetersonBill LeenhoutsJohn E. Core

9.0 Smoke Dispersion Prediction Systems ............................................ 163Sue A. Ferguson

10.0 Air Quality Monitoring for Smoke ..................................................... 179John E. CoreJanice L. Peterson

11.0 Emission Inventories .......................................................................... 189Janice L. Peterson

12.0 Smoke Management Program Administration and Evaluation ..... 201Peter Lahm

Appendix A – Glossary of Fire and Smoke Management Terminology ..... 209

– 1 –

2001 Smoke Management Guide Introduction

INTRODUCTION

Chapter 1

Chapter 1 – Introduction 2001 Smoke Management Guide

– 2 –

– 3 –


Introduction

Colin C. Hardy

Bill Leenhouts

Why Do We Need A National Smoke Management Guide?

As an ecological process, wildland fire is essen-tial in creating and maintaining functionalecosystems and achieving other land use objec-tives. As a decomposition process, wildland fireproduces combustion byproducts that are harm-ful to human health and welfare. Both the landmanagement benefits from using wildland fireand the public health and welfare effects fromwildland fire smoke are well documented. Thechallenge in using wildland fire is balancing thepublic interest objectives of protecting humanhealth and welfare and sustaining ecologicalintegrity.

Minimizing the adverse effects of smoke onhuman health and welfare while maximizing theeffectiveness of using wildland fire is an inte-grated and collaborative activity. Everyoneinterested in natural resource management isresponsible and has a role. Land managers needto assure that using wildland fire is the mosteffective alternative of achieving the landmanagement objectives. State, regional, tribaland national air resource managers must ensurethat air quality rules and regulations equitablyaccommodate all legal emission sources.

The varied smoke management issues fromacross the nation involve many diverse culturesand interests, include a multitude of strategiesand tactics, and cover a heterogeneous land-scape. No national answer or cookbook ap-

proach will adequately address them. Butpeople with a desire for responsible smokemanagement working in partnership with thelatest science-based smoke management infor-mation can fashion effective regional smokemanagement plans and programs to addresstheir individual and collective objectives. Theintent of the Guide is to provide the latestscience-based smoke management informationfrom across the nation to facilitate these col-laborative efforts.

Awareness of smoke production, transport, andeffects on receptors from prescribed and wild-land fires will enable us to refine existing smokemanagement strategies and to develop bettersmoke management plans and programs in thefuture. This Guide addresses the basic controlstrategies for minimizing the adverse effects ofsmoke on human health and welfare—thusmaximizing the effectiveness of using wildlandfire. These control strategies are:

• Avoidance – using meteorological condi-tions when scheduling burning in order toavoid incursions of wildland fire smokeinto smoke sensitive areas.

• Dilution – controlling the rate of emissionsor scheduling for dispersion to assuretolerable concentrations of smoke indesignated areas.


– 4 –

• Emissions-reduction – using techniques tominimize the smoke output per unit areatreated and decrease the contribution toregional haze as well as intrusions intodesignated areas.

Guide Goals andConsiderations

The Smoke Management Guide steering com-mittee and the NWCG Fire Use Working Teamdeveloped this Guide with the following goals:

• Provide fire use practitioners with afundamental understanding of fire-emis-sions processes and impacts, regulatoryobjectives, and tools for the managementof smoke from wildland fires.

• Provide local, state, tribal, and federal airquality managers with background infor-mation related to the wildland fire andemissions processes and air, land andwildland fire management.

The following considerations provide the con-text within which these goals can be met:

• This document is about smoke manage-ment, not about the decision to use wild-land fire or its alternatives. Its purpose isnot to advocate for or against the use offire to meet land management objectives.

• While the Guide contains relevant back-ground material and resources generallyuseful to development of smoke manage-ment programs, it is not a tutorial on howto develop a state smoke managementprogram.

• Although the Guide is replete with infor-mation and examples for potential applica-tion at the local and regional level, theGuide generally focuses on national smoke

management principles. For maximumbenefit to local or regional applications,appropriate supplements should be devel-oped for the scale or geographical locationof the respective application.

• The Guide is more appropriate for knowl-edgeable air, land, and wildland firemanagers, and is not intended for novicereaders.

Overview andOrganization of the Guide

The Smoke Management Guide for Prescribedand Wildland Fire–2001 Edition follows atextbook model so that it can be used as asupplemental reference in smoke managementtraining sessions and courses such as theNWCG Smoke Management course, RX-410(formerly RX-450). Following an Introduc-tion, a background chapter presents a primer onwildland fire and a discussion of the imperativesfor smoke management. In the Wildland FireImperative, the Guide addresses both theecological and societal aspects of wildland fire(not agricultural, construction debris, or otherbiomass burning), and provides the detailsnecessary for fire use practitioners and airquality managers to understand the fundamen-tals of fire in wildlands. The Smoke Manage-ment Imperative discusses the needs for smokemanagement as well as its benefits and costs.

The background sections are followed by chap-ters presenting details on Wildland Fire SmokeImpacts—public health, visibility, problem andnuisance smoke, and smoke exposure amongfireline personnel—and on Regulations forSmoke Management. The chapter on SmokeSource Characteristics follows a sequencesimilar to the basic pathway that smoke produc-

– 5 –


___________________________________

1 For a comprehensive presentation of fire terminology, the reader should refer to the NWCG Glossary of WildlandFire Terminology (NWCG 1996—PMS #205, Boise, ID).

tion does—from the pre-fire fuel characteristicsand the fire phenomenon as an emissionssource, through the processes of combustion,biomass consumption and emissions production.

The chapter on Fire Use Planning addressesimportant considerations for developing acomprehensive fire use plan (a “burn plan”).The general planning process is reviewed, fromdeveloping a general land use plan, through afire management plan and, ultimately, to a unit-specific burn plan.

The Smoke Management Meteorology chapterpresents a primer on the use of weather observa-tions and forecasts, and then provides informa-tion regarding the transport and dispersion ofsmoke from wildland fires.

Techniques to Reduce or Redistribute Emis-sions are presented in an exhaustive list andsynthesis of emissions reduction and impactreduction practices and techniques. Thesepractices and techniques were initially compiledas the outcomes of three regional workshopsheld specifically for the purpose of synthesizingcurrent and potential smoke management tools.Presented here in a nationally applicable format,they are the fundamental tools available to fireplanners and fire use practitioners for the man-agement and mitigation of smoke from wildlandfires.

The Smoke Dispersion Prediction Systemschapter reviews current prediction tools withinthe context of three “families” of model applica-tions—screening, planning, or regulating.

Air Quality Monitoring for Smoke discussesvarious objectives for monitoring, and empha-sizes the need to carefully match the monitoringobjective with the appropriate equipment. In

addition, the chapter presents information onsome common monitoring equipment, methods,and their associated costs.

Emission Inventories help managers andregulators understand how to better include firein an emissions inventory. This chapter dis-cusses the use of the three basic elementsneeded to perform an emission inventory—areaburned, fuel consumed, and appropriate emis-sion factor(s).

No smoke management effort can succeedwithout continued assessment and feedback.The chapter on Program Administration andAssessment discusses the need to maintain abalance between the level of effort in a programand the level of prescribed or fire use activity aswell as their associated local or regional effects.

Each section in this Guide is now supported byan extensive list of relevant references. Also,authorship for a specific section is given in thetable of contents, where appropriate. In suchcases, the section can be cited with its respectiveauthor(s) as an independent “chapter” in theGuide.

A glossary of frequently used fire and smokemanagement terms1 is provided as an appendixto the Guide.

History of SmokeManagement Guidance

The first guidance document specifically ad-dressing the management of smoke from pre-scribed fires was the Southern SmokeManagement Guidebook, produced in 1976 bythe Southern Forest Fire Laboratory staff


– 6 –

___________________________________

2 The Joint Fire Sciences Program is sponsoring extensive revisions to the Rainbow Series fire effects volumes,including a new volume on fire effects on air.

(1976). It was a comprehensive treatment of thevarious aspects fire behavior, emissions, trans-port and dispersion, and the management ofsmoke in the southern United States.

In 1985, NWCG’s Prescribed Fire and FireEffects Working Team developed the widelyaccepted Prescribed Fire Smoke ManagementGuide that forms the basis for this 2001 revisedGuide (NWCG 1985). The 1985 edition fo-cused on national smoke management principlesand, as a result, was far less comprehensive thanthe Southern guidebook.

One of six state-of-knowledge reports preparedfor the 1978 National Fire Effects Workshop is areview called Effects of Fire on Air (USDAForest Service 1978). The six volumes, calledthe “Rainbow Series” on fire effects, were inresponse to the changes in policies, laws, regula-tions, and initiatives. Objectives specific to thevolume on air were to: “…summarize thecurrent state-of-knowledge of the effects offorest burning on the air resource, and to defineresearch questions of high priority for themanagement of smoke from prescribed and wildfires” (USDA Forest Service 1978, p.5).2

Conflicts between prescribed fire and air qualitybegan to be seriously addressed in the mid-1980s. Prior to this, only a few states haddeveloped or implemented smoke managementprograms, and national-level policies addressingsmoke from wildland burns were only beginningto be drafted. Much has changed since then,with numerous policies and initiatives raisingthe potential for conflicting resource manage-ment objectives—principally air quality andecosystem integrity. The Clean Air Act amend-ments adopted in 1990 specifically addressedregional haze. Smoke Management Plans have

been developed by many states as administrativerules enforceable under state law. These rulesare often incorporated into State and TribalImplementation Plans (SIPs and TIPs) forsubmission to the U.S. Environmental Protec-tion Agency (EPA) and, once promulgated byEPA, are then enforceable under federal law aswell. And now, the role of fire and the need forits accelerated use has become widely recog-nized with respect to maintenance and restora-tion of fire-adapted ecosystems. These issuesall point to the imperative for better knowledgeand more informed collaboration betweenmanagers of both the air and terrestrial re-sources.

The 2001 Edition of the SmokeManagement Guide

Recognizing the increasing likelihood of im-pacting the public, the proliferation of federal,state, and local statutes, rules and ordinancespertaining to smoke, as well as major improve-ments to our knowledge of smoke and its man-agement, the NWCG Fire Use Working Team(formerly named the Prescribed Fire and FireEffects Working Team) sponsored revision ofthe Guide. Conceptually, the Fire Use WorkingTeam identified the need for a revised guide-book that targeted not just prescribed fire practi-tioners, but state and local air quality and publichealth agency personnel as well. A consequenceof this expansion of the target audience was theneed to substantially augment the backgroundinformation with respect to fire in wildlands.

A suite of potential smoke management prac-tices and techniques are not only suggested in

– 7 –


this Guide, but their relative effectiveness andregionally-specific applicability are also pro-vided. This information was acquired throughthree regional workshops held in collaborationwith the U.S. Environmental ProtectionAgency’s Office of Air Quality Planning andStandards.

This revised Guide now emphasizes both emis-sion and impact reduction methods that havebeen found to be practical, useful, and benefi-cial. This new emphasis on reducing emissionsis in response to regional haze and fine particle(PM2.5) control programs that will requireemission reductions from a wide variety ofpollution sources (including prescribed andwildland fire). This is especially important inview of the major increases in the use of fireprojected by federal land managers. Readerswill also find a greatly expanded discussion ofair quality regulatory requirements, reflectingthe growing complexities and demands ontoday’s fire practitioners.

Literature CitationsNWCG. 1985. Prescribed fire smoke management

guide. NWCG publication PMS-420-2. Boise,ID. National Wildfire Coordinating Group.28 p.

Southern Forest Fire Laboratory Staff. 1976. South-ern forestry smoke management guidebook.USDA For. Serv. Gen. Tech. Rep. SE-10.Asheville, N.C. USDA Forest Service, South-east Forest Experiment Station. 140 p.

USDA Forest Service. 1978. Effects of fire on air.USDA For. Serv. Gen. Tech. Rep. WO-9.Washington, D.C. USDA Forest Service. 40 p.

Chapter 2 – Overview 2001 Smoke Management Guide

– 8 –

– 9 –

2001 Smoke Management Guide 2.1 – The Wildland Fire Imperative

OVERVIEW

Chapter 2


– 10 –

– 11 –


The Wildland Fire Imperative

Colin C. Hardy

Sharon M. Hermann

Robert E. Mutch

Perpetuating America’s Natural Heritage: BalancingWildland Management Needs and the Public Interest

Strategies for responsible and effective smokemanagement cannot be developed withoutcareful consideration of the ecological and thesocietal impacts of fire management in thewildlands of modern America. The need toconsider both perspectives is acknowledged bymost land management agencies, as well as bythe U.S. Environmental Protection Agency(EPA) —the primary Federal agency responsiblefor protecting air quality. An awareness of thischallenge is reflected in NWCG’s educationmessage, Managing Wildland Fire: BalancingAmerica’s Natural Heritage and the PublicInterest (NWCG 1998). The preamble to thisdocument not only states that “fire is an impor-tant and inevitable part of America’s wildlands,”but also recognizes that “wildland fires canproduce both benefits and damages—to theenvironment and to people’s interests.”

The EPA’s Interim Air Quality Policy on Wild-land and Prescribed Fires (U.S. EPA 1998)employs similar language to describe relatedpublic policy goals: (1) To allow fire to function,as nearly as possible, in its natural role inmaintaining healthy wildland ecosystems; and,(2) To protect public health and welfare bymitigating the impacts of air pollutant emissions

on air quality and visibility. The documentcomments on the responsibilities of wildlandowners/managers and State/tribal air qualitymanagers to coordinate fire activities, minimizeair pollutant emissions, manage smoke fromprescribed fires as well as wildland fires usedfor resource benefits, and establish emergencyaction programs to mitigate the unavoidableimpacts on the public. In addition, EPA assertsthat “this policy is not intended to limit opportu-nities by private wildland owners/managers touse fire so that burning can be increased onpublicly owned wildlands.”

In this and the following section (2.2–TheSmoke Management Imperative), we outlineboth ecological and societal aspects of wildlandand prescribed fire. We review the historicalrole and extent of fire and the effects of settle-ment and land use changes. The influence offire exclusion policies on historical disturbanceprocesses is considered in light of modernlandscape conditions. This provides the basisfor discussion of significant, recent changes inFederal wildland fire policy and new initiativesfor accelerating use of prescribed and wildlandfire to achieve resource management objectives.Finally, we present examples of the impacts of


– 12 –

wildland smoke on air quality, human health,and safety.

Fire in Wildlands

Recurring fires are often an essential componentof the natural environment—as natural as rain,snow, or wind. Evidence for the recurrence ofpast fires is found in charcoal layers of lakes andbogs, in fire-scars of trees, and in the morpho-logical and life history adaptations of numerousnative plants and animals. Many ecosystems inNorth America and throughout the world arefire-dependent (Heinselman 1978) and periodicburning is essential for healthy ecosystemfunctioning in these wildlands. Fire acts at theindividual, population, and community levelsand can influence:

• Plant succession.

• Fuel accumulation and decay.

• Recruitment pattern and age distribu-tion of individuals.

• Species composition of vegetation.

• Disease and insect pathogens.

• Nutrient cycles and energy flows.

• Biotic productivity, diversity, andstability.

• Habitat structure for wildlife.

For millennia, lightning, volcanoes, and peoplehave ignited fires in wildland ecosystems. Thecurrent emphasis on ecosystem managementcalls for the maintenance of interactions be-tween such disturbance processes and ecosys-tem functions. Therefore, it is incumbent onboth fire and natural resource managers tounderstand the range of historical frequency,severity, and aerial extent of past burns. Thisknowledge provides a frame of reference forapplying appropriate management practices on alandscape scale, including the use and exclusionof fire.

Many studies have described the historicaloccurrence of fires throughout the world. Forexample, Swetnam (1993) used fire scars todescribe a 2000-year period of fire history ingiant sequoia groves in California. He foundthat frequent small fires occurred during a warmperiod from about A.D. 1000 to 1300, and lessfrequent but more widespread fires occurredduring cooler periods from about A.D. 500-1000and after 1300. Swain (1973) determined fromlake sediment analyses in the Boundary WatersCanoe Area in Minnesota that tree species andfire had interacted in complex ways over a10,000-year period. Other studies ranging fromMaine (e.g. Copenheaver and others 2000) toFlorida (e.g. Watts and others 1992) have em-ployed pollen and charcoal deposits to demon-strate shifts in fire frequency correlated with theonset of European settlement.

There is an even larger body of science thatdetails the numerous effects of wildland fires oncomponents of ecosystems. Some of the mostcompelling examples of fire dependency comefrom studies on plant reproduction and estab-lishment. For instance, there are at least tenspecies of pines scattered over the United Statesthat have serotinous cones; that is to say thecones are sealed by resin; the cone scales do notopen and seeds do not disperse until the resin isexposed to high heat (reviewed in Whelan1995). Examples of fire dependency in herba-ceous plants include flowering of wiregrass inSoutheastern longleaf pine forests that is greatlyenhanced by growing season burns (Myers1990) and seed germination of Californiachaparral forbs that is triggered by exposure tosmoke (Keeley and Fotheringham 1997). Ani-mals as diverse as rare Karner blue butterflies inIndiana (Kwilosz and Knutson 1999) to whoop-ing cranes in Texas (Chavez Ramirez and others1996) benefit when fire is re-introduced intotheir habitats. There are numerous other typesof fire dependency in North American ecosys-

– 13 –


tems and many studies on this topic are summa-rized in books and government publications(e.g. Agee 1993, Bond and van Wilgen 1996,Brown and Kapler Smith 2000, Johnson 1992,Kapler Smith 2000, Wade and others 1980,Whelan 1995). In addition, there is a small butgrowing volume of literature that evaluates theinfluence of fire on multiple trophic levels (e.g.Hermann and others 1998).

Knowledge of fire history, fire regimes, and fireeffects allows land stewards to develop informedmanagement strategies. Application of fire maybe one of the tools used to meet resource man-agement objectives. The role of fire as animportant disturbance process has been high-lighted in a classification of continental fireregimes (Kilgore and Heinselman 1990). Theseauthors describe a natural fire regime as the totalpattern of fires over time that is characteristic ofa region or ecosystem. Fire regimes are definedin terms of fire type and severity, typical firesizes and patterns, and fire frequency, or lengthof return intervals in years. Kilgore and

Heinselman (1990) placed natural fire regimesof North America into seven classes, rangingfrom Class 0, in which fires are rare or absent,to Class 6, in which crown fires and severesurface fires occur at return intervals longer than300 years. Intermediate fire regimes, Classes 1-5, are characterized by increasingly longer firereturn intervals and increasingly higher fireintensities. Class 2, for example, describes thesituation for long-needled pines, like longleafpine, ponderosa pine, and Jeffrey pine; in thisclass low severity, surface fires occur ratherfrequently (return intervals of less than 25years). Lodgepole pine, jackpine, and the borealforest of Canada and Alaska generally fall intoClass 4, a class in which high severity crownfires occur every 25 to 100 years; or into Class5, a class in which crown fires occur every 100to 300 years. White bark pine forests at highelevations typically fall into Class 6. For com-parison, three general classes of fire are shownin figure 2.1, including a low-intensity surfacefire, a mixed-severity fire, and a stand-replacingcrown fire.

Figure 2.1. The relative difference in general classes of fire are shown. Thisseries illustrates a low-intensity surface fire (a), a mixed-severity fire (b), and astand-replacing crown fire (c).

(b)

(a)

(c)


– 14 –

A noteworthy aspect of continental fire regimesis that very few North American ecosystems fallinto Class 0. In other words, most ecosystemsin the United States have evolved under theconsistent influence of wildland fire, establish-ing fire as a process that affects numerousecosystem functions described earlier. Thosewho apply prescribed burns or use wildland fireoften attempt to mimic the natural role of fire increating or maintaining ecosystems. Sustainingthe productivity of fire-adapted ecosystemsgenerally requires application of prescribed fireon a sufficiently large scale to ensure thatvarious ecosystem processes remain intact.

Ecological Effects ofAltered Fire Regimes

As humans alter fire frequency and severity,many plant and animal communities experiencea loss of species diversity, site degradation, andincreases in the sizes and severity of wildfires.Ferry and others (1995) concluded that alteredfire regimes was the principal agent of change

affecting vegetative structure, composition, andbiological diversity of five major plant commu-nities totaling over 350 million acres in the U.S.As a way to evaluate the current amount of firein wildland habitat, Leenhouts (1998) comparedestimated land area burned 200-400 years ago(“pre-industrial”) to data from the contemporaryconterminous United States. The result suggeststhat ten times more acreage burned annually inthe pre-industrial era than does in modern times.After accounting for loss of wildland area due toland use changes such as urbanization andagriculture, Leenhouts concluded that theremaining wildland is burned approximatelyfifty percent less compared to fire frequencyunder historical fire regimes (figure 2.2).

Numerous ecosystem indicators serve as alarm-ing examples of the effects of altered fire re-gimes. Land use changes, attempted fireexclusion practices, prolonged drought, andepidemic levels of insects and diseases havecoincided to produce extensive forest mortality,or major changes in forest density and speciescomposition. Gray (1992) called attention to aforest health emergency in parts of the western

Figure 2.2. Estimates of the range of annual area burned in the conterminous United States pre-Europeansettlement (Historic), applying presettlement fire frequencies to present land cover types (Expected), andburning (wildland and agriculture) that has occurred during the recent past (Current). Source: Leenhouts(1998).

– 15 –


United States where trees have been killedacross millions of acres in eastern Oregon andWashington. He indicated that similar problemsextend south into Utah, Nevada, and California,and east into Idaho. Denser stands and heavyfuel accumulations are also setting the stage forhigh severity crown fires in Montana, Colorado,Arizona, New Mexico, and Nebraska, where thehistorical norm in long-needled pine forests was

for more frequent low severity surface fires (fireregime Class 2; Kilgore and Heinselman 1990).The paired photos in figure 2.3 illustrate 85years of change resulting from fire exclusion ona fire-dependent site in western Montana. InNorth Carolina, Gilliam and Platt (1999) quanti-fied the dramatic effects of over 80-years of fireexclusion on tree species composition and standstructure in a longleaf pine forest.

(a)

(b)

Figure 2.3. These two photos, taken of the same homestead near Sula, Montana, show 85 years of change ona fire-dependent site where fire has been excluded. The top photo (a) was taken in 1895. By 1980 (b),encroaching trees and shrubs occupy nearly all of the site. Stand-replacing crown fire visited this site in 2000.


– 16 –

Since the 1960s, records show an alarmingtrend towards more acres consumed by wildfires, despite all of our advances in fire suppres-sion technology (figure 2.4). The larger, moresevere wildfires have accelerated the rate of treemortality, threatening people, property, andnatural resources (Mutch 1994). These wild-fires also have emitted large amounts of particu-late matter into the atmosphere. One studyestimated that more than 53 million pounds ofrespirable particulate matter were producedover a 58-day period by the 1987 Silver Fire insouthwestern Oregon (Hardy and others 1992).

The ecological consequences of past policies offire exclusion have been foreseen for sometime. More than 50 years ago, Weaver (1943)reported that the “complete prevention of forestfires in the ponderosa pine region of California,Oregon, Washington, northern Idaho, andwestern Montana has certain undesirable eco-

logical and silvicultural effects [and that]...conditions are already deplorable and are be-coming increasingly serious over large areas.”Also, Cooper (1961) stated, “…fire has played amajor role in shaping the world’s grassland andforests. Attempts to eliminate it have introducedproblems fully as serious as those created byaccidental conflagrations.” Only more recentlyhave concerns been expressed about potentialloss of biodiversity as a result of fire suppres-sion. This issue may be especially pressing inthe Eastern United States. For example, insouthern longleaf pine ecosystems, at least 66rare plant species are maintained by frequentfire (Walker 1993). The ecological need forhigh fire frequency in large areas of Southeast-ern native ecosystems coupled with the region’slong growing season contribute to the rapidbuildup of fuel and subsequent change in habitatstructure.

Figure 2.4. The average annual burned area for the western States, shown here for the period1916-2000, has generally been increasing since the mid-1960s

– 17 –


Wildland and Prescribed Fire Ter-minology Update

The federal Implementation Procedures Refer-ence Guide for Wildland and Prescribed FireManagement Policy (USDI and USDA ForestService 1998) contains significant changes infire terminology. Several traditional terms haveeither been omitted or have been made obsoleteby the new policy. These include: confine/contain/control; escaped fire situation analysis;management ignited prescribed fire; pre-sup-pression; and prescribed natural fire, or “PNF.”Additionally, there was adoption of several newterms and interpretations that supercedes earlier,traditional terminology:

• Fire Use - the combination of wildlandfire use and prescribed fire application tomeet resource objectives.

• Prescribed Fire - Any fire ignited bymanagement actions to meet specificobjectives. A written, approved prescribedfire plan must exist, and NEPA require-ments must be met, prior to ignition. Thisterm replaces management ignited pre-scribed fire.

• Wildfire - An unwanted wildland fire.This term was only included to give con-tinuing credence to the historic fire pre-vention products. This is NOT a separatetype of fire under the new terminology.

• Wildland Fire - Any non-structure fire,other than prescribed fire, that occurs inthe wildland. This term encompasses firespreviously called both wildfires andprescribed natural fires.

• Wildland Fire Use - the management ofnaturally-ignited wildland fires to accom-plish specific pre-stated resource manage-ment objectives in predefined geographicareas outlined in Fire Management Plans.Wildland fire use is not to be confused

with “fire use,” which is a broader termencompassing more than just wildlandfires.

Taking Action: The Federal Wild-land and Prescribed Fire Policy

The decline in resiliency and ecological “health”of ecosystems has reached alarming proportionsin recent decades, as evidenced by the trendsince the mid-1960’s towards more acres burnedin wildfires (figure 2.4). While national aware-ness of this trend has existed for some time, the1994 fire season created a renewed awarenessand concern among Federal land managementagencies and their constituents regarding theserious impacts of wildfires. The FederalWildland Fire Management Policy and ProgramReview is chartered by the Secretaries of Agri-culture and Interior to “ensure that uniformfederal policies and cohesive interagency andintergovernmental fire management programsexist” (USDI and USDA Forest Service 1995).The review process is directed by an interagencySteering Group whose members represented theDepartments of Agriculture and Interior, theU.S. Fire Administration, the National WeatherService, the Federal Emergency ManagementAgency, and the Environmental ProtectionAgency. In their cover letter accepting the FinalReport of the Review (December 18, 1995), theSecretaries of Agriculture and Interior pro-claimed:

“The philosophy, as well as the specificpolicies and recommendations, of theReport continues to move our approach towildland fire management beyond thetraditional realms of fire suppression byfurther integrating fire into the manage-ment of our lands and resources in anongoing and systematic manner, consistentwith public health and environmental


– 18 –

quality considerations. We strongly sup-port the integration of wildland fire intoour land management planning and imple-mentation activities. Managers must learnto use fire as one of the basic tools foraccomplishing their resource managementobjectives.”

USDI and USDA ForestService 1995—covermemorandum

The Report asserts that “the planning, imple-mentation, and monitoring of wildland firemanagement actions will be done on an inter-agency basis with the involvement of all part-ners.” The term “partners” is all-encompassing,including Federal land management and regula-tory agencies; tribal governments; Departmentof Defense; State, county, and local govern-ments; the private sector; and the public. Part-nerships are essential for establishing collectivepriorities to facilitate use of fire at the landscapelevel. Smoke does not respond to artificialboundaries or delineations. Interaction amongpartners is necessary to meet the dual challengeof using fire for natural resource managementcoupled with the need to minimize negativeeffects related to smoke. Both concerns must bemet to fulfill the public need.

Literature CitationsAgee, J.K. 1993. Fire Ecology of Pacific Northwest

Forests. Island Press, Washington, DC.

Bond, W.J. and B.W. van Wilgen. 1996. Fire andPlants. Chapman Hall, London.

Brown, J.K. and J. Kapler Smith (eds.). 2000.Wildland Fire in Ecosystems: Effects of Fire onFlora. Gen. Tech. Rep. RMRS-GTR-42-vol.2.Ogden, UT.

Chavez Ramirez, F., H.E. Hunt, R.D. Slack and T.V.Stehn. 1996. Ecological correlates of Whoop-ing Crane use of fire-treated upland habitats.Conservation Biology 10:217-223.

Cooper, C.F. 1961. The ecology of fire. Sci. Am.204(4):150-160.

Copenheaver, C.A., A.S. White and W.A. Patterson.2000. Vegetation development in a southernMaine pitch pine-scrub oak barren. JournalTorrey Botanical Soc.127:19-32.

Ferry, G.W. R.G. Clark, R.E. Montomery, R.W.Mutch, W.P. Leenhouts, and G. T. Zimmerman.1995. Altered fire regimes within fire-adaptedecosystems. Pages 222-224. In: Our LivingResources. W.T. LaRoe, G. S. Farris, C.E.Puckett, P.D. Doran, and M.J. Mac eds. U.S.Department of the Interior, National BiologicalService, Washington, D.C. 530p.

Gilliam, F.S. and W.J. Platt. 1999. Effects of long-term fire exclusion on tree species compositionand stand structure in an old-growth Pinuspalustris (Longleaf pine) forest. Plant Ecology140:15-26.

Gray, G.L. 1992. Health emergency imperils westernforests. Resource Hotline. 8(9). Published byAmerican Forests.

Hardy, C. C., D. E. Ward, and W. Einfeld. 1992.PM2.5 emissions from a major wildfire using aGIS: rectification of airborne measurements. In:Proceedings of the 29th Annual Meeting of thePacific Northwest International Section, Air andWaste Management Association, November 11-13, 1992, Bellevue, WA. Pittsburgh, PA: Air andWaste Management Association.

Heinselman, M. L. 1978. Fire in wilderness ecosys-tems. In: Wilderness Management. J. C.Hendee, G. H. Stankey, and R. C. Lucas, eds.USDA Forest Service, Misc. Pub. 1365

Hermann, S.M., T. Van Hook, R.W. Flowers, L.A.Brennan, J.S. Glitzenstein, D.R. Streng, J.L.Walker and R.L. Myers. 1998. Fire andbiodiversity: studies of vegetation andarthropods. Trans. North American Wildlife andNatural Resources Conf. 63:384-401

Johnson, E.A. 1992. Fire and Vegetation Dynamics:Studies from the North American Boreal Forest.Cambridge University Press, Cambridge.

Keeley, J.E., and C.J. Fotheringham. 1997. Tracegas emissions and smoke induced seed germi-nation. Science 276:1248-1250.

– 19 –


Kilgore, B. M., and M. L. Heinselman. 1990. Fire inwilderness ecosystems. In: Wilderness Man-agement, 2nd ed. J. C. Hendee, G. H. Stankey,and R. C. Lucas, eds. North American Press,Golden, CO. Pp. 297-335.

Kwilosz, J.R. and R.L. Knutson. 1999. Prescribedfire management of Karner blue butterflyhabitat at Indiana Dunes National Lakeshore.Natural Areas Journal 19:98-108

Leenhouts, Bill. 1998. Assessment of biomassburning in the conterminous United States.Conservation Ecology [online] 2(1): 1. Avail-able from the Internet. URL:

Mutch, R. W. 1994. Fighting fire with prescribedfire—a return to ecosystem health. J. For.92(11):31-33.

Mutch, R. W. 1997. Need for more prescribed fire:but a double standard slows progress. In Pro-ceedings of the Environmental Regulation andPrescribed Fire Conference, Tampa, Florida.March 1995. Pp. 8-14.

Myers, R.L. 1990. Scrub and high pine. pages 150-193 in (R.L. Myers and J.J. Ewel, eds.) Ecosys-tems of Florida. University of Central FloridaPress, Orlando.

National Wildfire Coordinating Group (NWCG).1998. Managing Wildland Fire: BalancingAmerica’s Natural Heritage and the PublicInterest. National Wildfire Coordinating Group;Fire Use Working Team [online]. Availablefrom the Internet. URL: http://www.fs.fed.us/fire/fire_new/fireuse/wildland_fire_use/role/role_pg8.html

Smith, J. Kapler (ed.). 2000. Wildland fire inecosystems: effects of fire on fauna. Gen. Tech.Rep. RMRS-GTR-42-vol. 1. Ogden, UT.

Swain, A. 1973. A history of fire and vegetation innortheastern Minnesota as recorded in lakesediment. Quat. Res. 3: 383-396.

Swetnam, T. W. 1993. Fire history and climatechange in giant sequoia groves. Science.262:885-889.

USDI and USDA Forest Service. 1995. Federalwildland fire management policy and programreview. Final report. National Interagency FireCenter, Boise, ID. 45 pp.

USDI and USDA Forest Service. 1998. Wildlandand prescribed fire management policy—implementation procedures reference guide.National Interagency Fire Center, Boise, ID. 81pp. and appendices.

U.S. Environmental Protection Agency. 1998.Interim air quality policy on wildland andprescribed fires. Final report. U.S. Environ-mental Protection Agency.

Wade, D.D., J.J. Ewel, and R. Hofsetter. 1980. Firein South Florida Ecosystems. USDA ForestService General Technical Report SE-17.

Walker, J. 1993. Rare vascular plant taxa associatedwith the longleaf pine ecosystems: patterns intaxonomy and ecology. pages 105-125 in (S.M.Hermann, ed.), The Longleaf Pine Ecosystem:ecology, restoration, and management. Pro-ceedings of the Tall Timbers Fire EcologyConference, No. 18.

Watts, W.A., B.C.S. Hansen and E.C. Grimm. 1992.Camel Lake – A 4000-year record of vegeta-tional and forest history from northwest Florida.Ecology 73:1056-1066.

Weaver, H. 1943. Fire as an ecological and silvicul-tural factor in the ponderosa pine region of thePacific Slope. J. For. 41:7-14.

Whelan, R.J. 1995. The Ecology of Fire. Cam-bridge University Press, Cambridge.


– 20 –

– 21 –

2001 Smoke Management Guide 2.2 – Smoke Management Imperative

The Smoke Management Imperative

Colin C. HardySharon M. HermannJohn E. Core

Introduction

In the past, smoke from prescribed burning wasmanaged primarily to avoid nuisance conditionsobjectionable to the public or to avoid traffichazards caused by smoke drift across roadways.While these objectives are still valid, today’ssmoke management programs are also likely tobe driven, in part, by local, regional and federalair quality regulations. These new demands onsmoke management programs have emerged asa result of Federal Clean Air Act requirementsthat include standards for regulation of regionalhaze and the recent revisions to the NationalAmbient Air Quality Standards (NAAQS) onparticulate matter.1

Development of the additional requirementscoincides with renewed efforts to increase use offire to restore forest ecosystem health. Thesetwo requirements are interrelated:

• The purity of the air we breathe is essen-tial to our health and quality of our livesand smoke from wildland and prescribedfire can have adverse effects on publichealth.

• The national forests, national parks andwilderness areas set aside by Congress areamong the nation’s greatest treasures.They inspire us as individuals and as a

nation. Smoke from wildland burning canobscure these natural wonders.

• Although smoke may be an inconvienceunder the best conditions and a publichealth and safety risk under the worstconditions, without periodic fires, thenatural habitat that society holds in suchhigh esteem will decline and ultimatelydissapear. In addition, as ecosystemhealth declines, fuel increases to levelsthat also pose significant risks for wildfireand consequently additional safety risks.

• Wildland and prescribed fire managers areentrusted with balancing these and other,often potentially conflicting responsibili-ties. Fire managers are charged with thetask of increasing the use of fire to ac-complish important land stewardshipobjectives and, at the same time, areentrusted to protect public safety andhealth.

Purpose of a SmokeManagement Program

The purpose of a smoke managementprogram is to:

1 See Chapter 4, Regulations for Smoke Management, for details on specific requirements.


– 22 –

• minimize the amount of smoke enteringpopulated areas, preventing public healthand safety hazards (e.g. visual impairmenton roadways or runways) and problems atsensitive sites (e.g. nursing homes orhospitals),

• avoid significant deterioration of airquality and NAAQS violations, and

• eliminate human-caused visibility impactsin Class I areas.

Smoke management programs create a frame-work of procedures and requirements for man-aging smoke from prescribed fires and aretypically developed by States or tribes withcooperation and participation from stakeholders.Procedures and requirements developed throughpartnerships are more effective at meetingresource management goals, protecting publichealth, and achieving air quality objectives thanprograms that are created in isolation. Sophisti-cated programs for coordination of burning bothwithin a state and across state boundaries arevital to obtain and maintain public support ofburning programs. Fire use professionals areincreasingly encouraged to burn at a landscapelevel. In some cases, when objectives are basedin both ecology and fuel reduction, there is aneed to consider burning during challengingtimes of the year (e.g. during the growingseason rather than the cooler dormant season).Multiple objectectives for fire use are likely toincrease the challenges, consequently increasingthe value of partnerships for smoke manage-ment.

Smoke management is increasingly recognizedas a critical component of a state or tribal airquality program for protecting public health andwelfare while still providing for necessarywildland burning.

Usually, either a state or tribal natural resourcesagency or air quality agency is responsible fordeveloping and administering the smoke man-agement program. Occasionally a smokemanagement program may be administered by alocal agency. California, for example, relies onlocal area smoke management programs. Gen-erally, on a daily basis the administering agencyapproves or denies permits for individual burnsor burns meeting some criteria. Permits may berequired for all fires or only for those thatexceed an established de minimis level (whichcould be based on projections of acres burned,tons consumed, or emissions). Multi-day burnsmay be subject to daily reassessment and re-approval to ensure compliance with smokemanagement program goals.

Advanced smoke management programs evalu-ate individual and multiple burns; coordinate allprescribed fire activities in an area; considercross-boundary (landscape) impacts; and weighdecisions about fires against possible health,visibility, and nuisance effects. With increasinguse of fire for forest health and ecosystemmanagement, interstate and interregional coordi-nation of burning will be necessary to preventepisodes of poor air quality. Development of,and participation in, an effective smoke manage-ment program by state agents and land manag-ers will go a long way towards building andmaintaining public acceptance of prescribedburning.

The Need for SmokeManagement Programs

The call for increasingly effective smoke man-agement programs has occurred because ofpublic and governmental concerns about thepossible risks to public health and safety, as wellas nuisance and regional haze impacts of smoke

– 23 –

2001 Smoke Management Guide 2.2 – Smoke Management Imperative

from wildland and prescribed fires. There arealso concerns about contributions to health-related National Ambient Air Quality Standards.Each of these areas is summarized below.2

Public Health Protection: Fine ParticleNational Ambient Air Quality Standards.–EPA’s most recent review of the National Ambi-ent Air Quality Standards for Particulate Matter(PM

10) concluded that significant changes were

needed to assure the protection of public health.In July of 1997, following an extensive reviewof the global literature, EPA adopted a fineparticle (PM

2.5) standard.3

These small particles are largely responsible forthe health effects of greatest concern and forvisibility reduction in the form of regional haze.More on EPA’s fine particle standard is foundelsewhere in this Guide.

The close link between regional haze and thenew fine particle National Ambient Air QualityStandards means that smoke from prescribedfire is again at the center of attention for airregulators charged with adopting control strate-gies to attain the new standards.

Public Safety and Nuisance Issues.–Perhapsthe most immediate need for an effective smokemanagement program is related to smokedrifting across roadways and restricting motoristvisibility. Each year, people are killed on thenation’s highways because of dust storms,smoke and fog. Wildland and prescribed firemanagers must recognize the legal issues relatedto their professional activities. Special caremust be taken in administering the smokemanagement program to assure that smoke doesnot obscure roadway or airport visibility. Li-ability issues vary by state. Some states such asFlorida have “right-to-burn” laws that provide

some protection for fire use professionals withspecific training and certification.

Probably the most common air quality issuesfacing wildland and prescribed fire managersare those related to public complaints aboutnuisance smoke. Complaints may be about theodor or soiling effects of smoke, poor visibility,and impaired ability to breathe or other health-related effects. Sometimes complaints comefrom the fact that some people don’t like or arefearful of smoke intruding into their lives.Whatever the reason, fire managers have aresponsibility to try to prevent or resolve theissue through smoke management plans thatrecognize the importance of proper selection ofmanagement and burning techniquesand burnscheduling based on meteorological conditions.In additioncommunity public relations andeducation coupled with pre-burn notification cangreatly improve public acceptance of fire man-agement programs.

Visibility Protection.–Haze that obstructs thescenic beauty of the Nation’s wildlands andnational parks does not respect political bound-aries. Any program that is intended to reducevisibility impairment in the nation’s parks andwildlands must be based on multi-state coopera-tive efforts or on national legislation.

In 1999, the U.S. EPA issued regional hazeregulations to manage and mitigate visibilityimpairment from the multitude of regional hazesources.4 Regional haze regulations call forstates to establish goals for improving visibilityin Class I national parks and wildernesses and todevelop long-term strategies for reducing emis-sions of air pollutants that cause visibilityimpairment. Wildland and prescribed fire aresome of the sources of regional haze covered bythe new rules.

2 Details relating to Public Health effects, Problem and Nuisance Smoke, and Regional Haze are given in the sections3.1, 3.3 and 4.1, respectively, of this Guide.

3 One thousand fine particles of this size could fit into the period at the end of this sentence.4 [40 CFR Part 51]


– 24 –

Past Success and Commitmentto Future Efforts

It is clearly noted in the preface to the 2001Smoke Management Guide that conflicts amongnatural resource needs, fire management, and airquality issues are expected to increase. It isequally important to acknowledge the benefitsto air quality resulting from the many successfulsmoke management efforts in the past twodecades.

Since the 1980s, federal, state, tribal, and localland managers have recognized the potential

impacts of smoke emissions from their activi-ties. Additionally, they have sponsored andpursued new efforts to learn the principles ofsmoke management and to develop appropriatesmoke management applications. Many earlysmoke management successes resulted fromproactive, voluntary inclusion of smoke man-agement components in many burn plans asearly as the mid-1980s.

NWCG and its partners are committed to fur-thering their leadership role in the quest for newinformation, technology, and innovative tech-niques. These 2001 revisions to the Guide areevidence of that commitment.

– 25 –

2001 Smoke Management Guide 3.1 – Public Health Effects

SMOKE IMPACTS

Chapter 3

Chapter 3 – Smoke Impacts 2001 Smoke Management Guide

– 26 –

– 27 –


Public Health and Exposure to Smoke

John E. Core

Janice L. Peterson

Introduction

The purity of the air we breathe is an importantpublic health issue. Particles of dust, smoke,and soot in the air from many sources, includingwildland fire, can cause acute health effects.The effects of smoke range from irritation of theeyes and respiratory tract to more serious disor-ders including asthma, bronchitis, reduced lungfunction, and premature death. Airborne par-ticles are respiratory irritants, and high concen-trations can cause persistent cough, phlegm,wheezing, and physical discomfort when breath-ing. Particulate matter can also alter the body’simmune system and affect removal of foreignmaterials from the lung like pollen and bacteria.

This section discusses the effects of air pollu-tion, especially particulate matter, on humanhealth and morbidity. Wildland fire smoke isdiscussed as one type of air pollution that can beharmful to public health1.

Human Health Effects ofParticulate Matter

Many epidemiological studies have shownstatistically significant associations of ambientparticulate matter levels with a variety of humanhealth effects, including increased mortality,hospital admissions, respiratory symptoms and

illness measured in community surveys (Brauer1999, Dockery and others 1993, EPA 1997).Health effects from both short-term (usuallydays) and long-term (usually years) particulatematter exposures have been documented. Theconsistency of the epidemiological data in-creases confidence that the results reported innumerous studies justify the increased publichealth concerns that have prompted EPA toadopt increasingly stringent air quality stan-dards (Federal Register 1997). There remains,however, uncertainty regarding the exactmechanisms that air pollutants trigger to causethe observed health effects (EPA 1996).

Figure 3.1.1 illustrates respiratory pathwaysthat form the human body’s natural defensesagainst polluted air. These pathways can bedivided into two systems - the upper airwaypassage consisting of the nose, nasal passages,mouth and pharynx, and the lower airwaypassages consisting of the trachea, bronchialtree, and alveoli. While coarse particles (largerthan about 5 microns in diameter) are depositedin the upper respiratory system, fine particles(less than 2.5 microns in diameter) can pen-etrate much deeper into the lungs. These fineparticles are deposited in the alveoli where thebody’s defense mechanisms are ineffective inremoving them (Morgan 1989).

___________________________________

1 Information on the effects of smoke on firefighters and prescribed burn crews can be found in Section 3.4.


– 28 –

On a smoggy day in a major metropolitan area,a single breath of air may contain millions offine particles. Some 74 million Americans —28% of the population — are regularly exposedto harmful levels of particulate air pollution(EPA 1997). In recent studies, exposure to fineparticles – either alone or in combination withother air pollutants – has been linked with manyhealth problems, including:

• An estimated 40,000 Americans dieprematurely each year from respiratoryillness and heart attacks that are linkedwith particulate exposure, especiallyelderly people (EPA 1997).

• Children and adults experience aggravatedasthma. Asthma in children increased118% between 1980 and 1993, and it iscurrently the leading cause of child hospi-tal admissions (EPA 1997).

• Children become ill more frequently andexperience increased respiratory problems,including difficult and painful breathing(EPA 1997).

• Hospital admissions, emergency roomvisits and premature deaths increaseamong adults with heart disease, emphy-sema, chronic bronchitis, and other heartand lung diseases (EPA 1997).

The susceptibility of individuals to particulateair pollution (including smoke) is affected bymany factors. Asthmatics, the elderly, thosewith cardiopulmonary disease, as well as thosewith preexisting infectious respiratory diseasesuch as pneumonia may be especially sensitiveto smoke exposure. Children and adolescentsmay also be susceptible to ambient particulatematter effects due to their increased frequencyof breathing, resulting in greater respiratorytract deposition. In children, epidemiologicalstudies reveal associations of particulate expo-sure with increased bronchitis symptoms andsmall decreases in lung function.

Fine particles showed consistent and statisticallysignificant relationships to short-term mortalityin six U.S. cities while coarse particles showedno significant relationship to excess mortality infive of the six cities that were studied (Dockeryand others 1993).

Figure 3.1.1: Particle deposition in the respiratory system.From: Canadian Center for Occupational Health & Safety, available athttp://www.ccohs.ca/oshanswers/ chemicals/how_do.html

– 29 –


Impacts of Wildland FireSmoke on Public Health

There is not much data which specificallyexamines the effects of wildland fire smoke onpublic health, although some studies areplanned or underway. We can, however, inferhealth responses from the documented effects ofparticulate air pollutants. Eighty to ninetypercent of wildfire smoke (by mass) is withinthe fine particle size class (PM2.5), makingpublic exposure to smoke a significant concern.

The Environmental Protection Agency hasdeveloped some general public health warningsfor specific air pollutants including PM2.5(table 3.1.1) (EPA 1999). The concentrations intable 3.1.1 are 24-hour averages, which can beproblematic when dealing with smoke impactsthat may be severe for a short period of time andthen virtually non-existent soon after. Anotherguidance document was developed recently torelate short-term, 1-hour averages to the poten-tial human health effects given in table 3.1.1(Therriault 2001).

Figure 3.1.2 contains these short-term averagesplus approximate corresponding visual range inmiles. Members of the public can use themethods described to estimate visual range anddetermine when air quality may be hazardous totheir health even if they are located in an areathat is not served by an official state air qualitymonitor.

Figure 3.1.3 is an information sheet developedduring a prolonged wildfire smoke episode inMontana during the summer of 2000. Thequestions and answers address many commonconcerns voiced by the public during smokeepisodes.

Other Pollutants of Concernin Smoke

Although the principal air pollutant of concernis particulate matter, there are literally hundredsof compounds emitted by wildland fires that arefound in very low concentrations. Some ofthese compounds that also deserve mentioninclude:

• Carbon monoxide has well known, serioushealth effects including dizziness, nauseaand impaired mental functions but isusually only of concern when people are inclose proximity to a fire (including fire-fighters). Blood levels of carboxyhemo-globin tend to decline rapidly to normallevels after a brief period free from expo-sure (Sharkey 1997).

• Benzo(a)pyrene, anthracene, benzene andnumerous other components found insmoke from wildland fires can cause head-aches, dizziness, nausea, and breathingdifficulties. In addition, they are of con-cern because of long term cancer risksassociated with repeated exposure tosmoke.

• Acrolein and formaldehyde are eye andupper respriatory irritants to which somesegments of the public are especiallysensitive.


– 30 –

Tabl

e 3.

1.1.

EPA

’s p

ollu

tant

sta

ndar

d in

dex

for

PM

2.5

can

be u

sed

for

gene

ral a

sses

smen

t of h

ealth

ris

ks fr

om e

xist

ing

air

qual

ity.

– 31 –


Figure 3.1.2. Visibility range can be used by the public to assess air quality in areas with no state airpollution monitors.

Conclusions

The health effects of wildland smoke are of realconcern to wildland fire managers, public healthofficials, air quality regulators and all segmentsof the public. Fire practitioners have an impor-tant responsibility to understand the potentialhealth impacts of fine particulate matter andminimize the public’s exposure to smoke.

Wildland fire managers should be aware ofsensitive populations and sites that may beaffected by prescribed fires, such as medicalfacilities, schools or nursing homes, and plan

burns to minimize the smoke impacts. This isespecially true when exposure may be pro-longed. Days or weeks of smoke exposure areproblematic because the lung’s ability to sweepthese particles out of the respiratory passagesmay be suppressed over time. Prolonged expo-sure may occur as the result of topographic ormeteorological conditions that trap smoke in anarea. Familiarity with the location and seasonalweather patterns can be invaluable in anticipat-ing and avoiding potential problems while stillin the planning phase.


– 32 –

What’s in smoke from a wildfire?Smoke is made up of small particles, gases and water vapor.Water vapor makes up the majority of smoke. The remainderincludes carbon monoxide, carbon dioxide, nitrogen oxide, irritantvolatile organic compounds, air toxics and very small particles.Is smoke bad for me?Yes. It’s a good idea to avoid breathing smoke if you can help it. Ifyou are healthy, you usually are not at a major risk from smoke.But there are people who are at risk, including people with heart orlung diseases, such as congestive heart disease, chronicobstructive pulmonary disease, emphysema or asthma. Childrenand the elderly also are more susceptible.What can I do to protect myself?

• Many areas report EPA’s Air Quality Index for particulatematter, or PM. PM (tiny particles) is one of the biggestdangers from smoke. As smoke gets worse, that indexchanges — and so do guidelines for protecting yourself. Solisten to your local air quality reports.

• Use common sense. If it looks smoky outside, that’s probablynot a good time to go for a run. And it’s probably a good timefor your children to remain indoors.

• If you’re advised to stay indoors, keep your windows anddoors closed. Run your air conditioner, if you have one. Keepthe fresh air intake closed and the filter clean.

• Help keep particle levels inside lower by avoiding usinganything that burns, such as wood stoves and gas stoves –even candles. And don’t smoke. That puts even more pollutionin your lungs – and those of the people around you.

• If you have asthma, be vigilant about taking your medicines,as prescribed by your doctor. If you’re supposed to measureyour peak flows, make sure you do so. Call your doctor if yoursymptoms worsen.

How can I tell when smoke levels are dangerous? I don’t livenear a monitor.Generally, the worse the visibility, the worse the smoke. InMontana, the Department of Environmental Quality uses visibilityto help you gauge wildfire smoke levels.How do I know if I’m being affected?You may have a scratchy throat, cough, irritated sinuses, head-aches, runny nose and stinging eyes. Children and people withlung diseases may find it difficult to breathe as deeply or vigorouslyas usual, and they may cough or feel short of breath. People withdiseases such as asthma or chronic bronchitis may find theirsymptoms worsening.Should I leave my home because of smoke?The tiny particles in smoke do get inside your home. If smokelevels are high for a prolonged period of time, these particles canbuild up indoors. If you have symptoms indoors (coughing, burningeyes, runny nose, etc.), talk with your doctor or call your countyhealth department. This is particularly important for people withheart or respiratory diseases, the elderly and children.Are the effects of smoke permanent?Healthy adults generally find that their symptoms (runny noses,coughing, etc.) disappear after the smoke is gone.Do air filters help?They do. Indoor air filtration devices with HEPA filters can reducethe levels of particles indoors. Make sure to change your HEPAfilter regularly. Don’t use an air cleaner that works by generatingozone. That puts more pollution in your home.

Do dust masks help?Paper “comfort” or “nuisance” masks are designed to trap largedust particles — not the tiny particles found in smoke.

These masks generally will not protect your lungs from wildfiresmoke.

How long is the smoke going to last?That depends on a number of factors, including the number offires in the area, fire behavior, weather and topography. Smokealso can travel long distances, so fires in other areas can affectsmoke levels in your area.I’m concerned about what the smoke is doing to my animals.What can I do?The same particles that cause problems for people may causesome problems for animals. Don’t force your animals to run orwork in smoky conditions. Contact your veterinarian or countyextension office for more information.How does smoke harm my health?One of the biggest dangers of smoke comes from particulatematter — solid particles and liquid droplets found in air. In smoke,these particles often are very tiny, smaller than 2.5 micrometers indiameter. How small is that? Think of this: the diameter of theaverage human hair is about 30 times bigger.These particles can build up in your respiratory system, causing anumber of health problems, including burning eyes, runny nosesand illnesses such as bronchitis. The particles also can aggravateheart and lung diseases, such as congestive heart failure, chronicobstructive pulmonary disease, emphysema and asthma.What about firefighters?Firefighters do experience short-term effects of smoke, such asstinging, watery eyes, coughing and runny noses. Firefightersmust be in good physical condition, which helps to offset adverseeffects of smoke. In addition to being affected by particles,firefighters can be affected by carbon monoxide from smoke. Arecent Forest Service study showed a very small percentage offirefighters working on wildfires were exposed to levels higherthan occupational safety limits for carbon monoxide and irritants.Why can’t the firefighters do something about the smoke?Firefighters first priorities in fighting a fire are, by necessity,protecting lives, protecting homes and containing the wildfire.Sometimes the conditions that are good for keeping the air clearof smoke can be bad for containing fires. A windy day helpssmoke disperse, but it can help a fire spread.Firefighters do try to manage smoke when possible. As theydevelop their strategies for fighting a fire, firefighters consider firebehavior and weather forecasts, topography and proximity tocommunities – all factors than can affect smoke.Why doesn’t it seem to be as smoky when firefighters areworking on prescribed fires.Land managers are able to plan for prescribed fires. They get tochoose the areas they want to burn, the size of those areas andthe weather and wind conditions that must exist before they beginburning. This allows them to control the fire more easily and limitits size. Those choices don’t exist with wildfires. In addition,wildfires that start in areas that haven’t been managed withprescribed fire often have more fuel, because vegetation in theforest understory has built up, and dead vegetation has not beenremoved.

This document was prepared by the Air Program, U.S. Forest Service – Northern Region, with assistance from the Office of AirQuality Planning & Standards in the US Environmental Protection Agency. For more information, call 406-329-3493. August 2000.

Figure 3.1.3. Public health information developed during the Montana wildfires of 2000.

– 33 –


Literature CitationsBrauer, Michael. 1999. Health impacts of biomass

air pollution. In: Health guidelines for vegeta-tion fire events: background papers. Edited byKee-Tai Goh, Dietrich Schwela, Johann G.Goldammer, and Orman Simpson. WorldHealth Organization. Available at http://www.who.int/peh/air/vegetationfirbackgr.htm.69p.

Dockery DW, Pope CA III, Xu X, Spengler JD,Ware JH, Fay ME, Ferris BG Jr, and SpeizerFE. 1993. An association between air pollutionand mortality in six U.S. cities. New EnglandJournal of Medicine. 329:1753-1759.

Environmental Protection Agency-Region 8. 1997.Public health effects of ozone and fine particlepollution. Environmental Fact Sheet. Availableat http://www.epa.gov/region08/news/news9697/ofpo.html

Environmental Protection Agency. 1996. Air qualitycriteria for particulate matter - Volume 1. USEnvironmental Protection Agency, ResearchTriangle Park, NC. EPA/600/P-95/001aF.

Environmental Protection Agency. 1999. Guidelinefor reporting of daily air quality – air quality

index (AQI). EPA-454/R-99-010. U.S. EPAOffice of Air Quality Planning and Standards,Research Triangle Park, NC 27711. 25p.

Federal Register. 1997. Memorandum from thePresident re: Implementation of Revised AirQuality Standards for Ozone and ParticulateMatter. Federal Register Vol. 62, No. 138. July18, 1997.

Morgan, Michael S. 1989. Health risks from slashfire smoke. In: The Burning Decision – Re-gional Perspectives on Slash. University ofWashington College of Forest Resources.Seattle, WA. pp. 226-236.

Sharkey, Brian, ed. 1997. Health hazards of smoke:recommendations of the April 1997 ConsensusConference. Tech. Rep. 9751-2836-MTDC.Missoula, MT: U.S. Department of Agriculture,Forest Service, Missoula Technology andDevelopment Center. 84 p.

Therriault, Shannon. 2001. Wildfire smoke: a guidefor public health officials. Missoula City-County Health Department. Missoula, MT. 13p.Available at: http://depts.washington.edu/wildfire/resources/pubhealthguide.pdf


– 34 –

– 35 –

2001 Smoke Management Guide 3.2 – Visibility

Visibility

John E. Core

Introduction

Every year there are over 280 million visitors toour nation’s wilderness areas and national parks.Congress has set these special places aside forthe enjoyment of all that seek spectacular andinspiring vistas. Unfortunately, many visitorsare not able to see the beautiful scenery theyexpect. During much of the year, a veil of hazeoften blurs their view. The haze is caused bymany sources of both natural and manmade airpollution sources, including wildland fire.

This section describes measures of scenicvisibility, the properties of the atmosphere andhow these properties are affected by smoke fromwildland fires, natural and current visibilityconditions, as well as sources that contribute tovisibility degradation. This is an importantissue to wildland fire practitioners becausesmoke is of increasing interest to air regulatorsresponsible for solving regional haze problems.

Measures of Visibility Impairment

Visibility is most often thought of in terms ofvisual range or the furthest distance a person cansee a landscape feature. However, visibility ismore than how far one can see; it also encom-passes how well scenic landscape features canbe seen and appreciated. Changes in visualrange are not proportional to human perception.For example, a five-mile change in visual rangecan result in a scene change that is either imper-

ceptible or very obvious depending on thebaseline visibility conditions. Therefore, a moremeaningful visibility index has been adopted.The scale of this index, expressed in deciviews(dv) is linear with respect to perceived visualchanges over its entire range, analogous to thedecibel scale for sound. A one-deciview changerepresents a change in scenic quality that wouldbe noticeable to most people regardless of theinitial visibility conditions. A deciview of zerois equivalent to clear air while deciviews greaterthan zero depict proportionally increased visibil-ity impairment (IMPROVE 1994). The moredeciviews measured, the greater the impairment,which limits the distance you can see. Finally,extinction in inverse megameters (Mm-1) isproportional to the amount of light lost as ittravels through a million meters of atmosphereand is most useful for relating visibility directlyto particulate concentrations. Table 3.2.1 com-pares each of these three forms of measurement(Malm 1999).

Properties of the Atmosphere &Wildland Fire Smoke

An observer sees an image of a distant objectbecause light is reflected from the object alongthe sight path to the observer’s eye. Any of thisimage-forming light that is removed from thesight path by scattering or light absorption


– 36 –

reduces the image-forming information andthereby diminishes the clarity of the landscapefeature. Ambient light is also scattered into thesight path, competing with the image-forminglight to reduce the clarity of the object ofinterest. This “competition” between image-forming light and scattered light is commonlyexperienced while driving in a snowstorm atnight with the car headlights on.

In addition, relative humidity also indirectlyaffects visibility. Although relative humiditydoes not by itself cause visibility to be de-graded, some particles, especially sulfates,accumulate water from the atmosphere andgrow to a size where they are particularlyefficient at scattering light. Poor visibility inthe eastern states during the summer months isa result of the combination of high sulfateconcentrations and high relative humidity.

The sum of scattering and absorption is referredto as atmospheric light extinction. Particles thatare responsible for scattering are categorized asprimary and secondary where primary sourcesinclude smoke from wildland fires and wind-blown dust. Other sources of secondary par-

Table 3.2.1. Comparison of the four expressions of visibility measurement.

ticles include sulfate and nitrate particlesformed in the atmosphere. The closer theparticle size is to the wavelength of light, themore effective the particle is in scattering light.As a result, relatively large particles of wind-blown dust are far less efficient in scatteringlight per unit mass than are the fine particlesfound in smoke from wildland fires. Finally, animportant component of smoke from wildlandfires is elemental carbon (also known as soot),which is highly effective in absorbing lightwithin the sight path. This combination of lightabsorption by elemental carbon and lightscattering caused by the very small particlesthat make up wildland fire smoke explains whyemissions from wildland fire play such animportant role in visibility impairment.

The effect of regional haze on a Glacier Na-tional Park vista is shown in the four panels offigure 3.2.1. The view is of the Garden Wallfrom across Lake McDonald. Particulateconcentrations associated with these photo-graphs correspond to 7.6, 12.0, 21.7 and65.3 µg/m3, respectively (Malm 1999). Notethe loss of color and detail in the mountains asthe particulate concentrations increase andvisibility decreases.

– 37 –


Figure 3.2.1. The effect of regional haze on a Glacier National Park vista.Photo courtesy of the National Park Service, Air Resources Division.

7.6 µg/m3 12.0 µg/m3

21.7 µg/m3 65.3 µg/m3

Natural Visibility Conditions

Some light extinction occurs naturally due toscattering caused by the molecules that make upthe atmosphere. This is called Rayleigh scatter-ing and is the reason why the sky appears blue.But even without the influence of human-causedair pollution, visibility would not always reachthe approximately 240-mile limit defined byRayleigh scattering. Naturally occurring par-ticles, such as windblown dust, smoke fromnatural fires, volcanic activity, and biogenicemissions (e.g. pollen and gaseous hydrocarbon)also contribute to visibility impairment although

the concentrations and sources of some of theseparticles remain a point of investigation.

Average natural visibility in the eastern U.S. isestimated to be about 60-80 miles (8-11 dv),whereas in the western US it is about 110-115miles (4.5-5 dv) (Malm 1999). Lower naturalvisibility in the eastern U.S. is due to higheraverage humidity. Humidity causes fine par-ticles to stick together, grow in size, and becomemore efficient at scattering light. Under naturalconditions, carbon-based particles are respon-sible for most of the non-Rayleigh particle-


– 38 –

Figure 3.2.2. Average annual visual range, in miles, for the years 1996-1998 measured at IMPROVE networkmonitors.

associated visibility reduction, with all otherparticle species contributing significantly less.Scattering from naturally occurring sulfateparticles from volcanic sulfur dioxide emis-sions and oceanic sources of primary sulfateparticles are estimated to account for 9-12% ofthe impairment in the East and 5% in the West(NPS 1997). It is expected that coastlines andhighly vegetated areas may be lower than theseaverages, while some elevated areas (moun-tains) could exceed these background esti-mates.

Current Visibility Conditions

Currently, average visual range in the easternU.S. is about 15-30 miles, or about one-thirdof the estimated natural background for the

East. In the West, visual range currently aver-ages about 60-90 miles, or about one-half of theestimated natural background for the West.Current annual visual range conditions ex-pressed in miles are shown in figure 3.2.2.Notice how much more impaired visibility is inthe East versus the West.

In the East, 60-70% of the visibility impairmentis attributed to sulfates. Sulfate particles formfrom sulfur dioxide gas, most of which is re-leased from coal-burning power plants and otherindustrial sources such as smelters, industrialboilers, and oil refineries. Carbon-based par-ticles contribute about 20% of the impairment inthe East. Sources of organic carbon particlesinclude vehicle exhaust, vehicle refueling,solvent evaporation, food cooking, and fires.

– 39 –


Literature CitationsIMPROVE. 1994. Visibility protection. Interagency

Monitoring of Protected Visual EnvironmentsBrochure. Ft. Collins, Colorado. August 1994.Available at: www2.nature.nps.gov/ard/impr/index.htm

National Park Service. 1997. Visibility protection.National Park Service Air Resources DivisionWeb Page. Available at www2.nature.nps.gov/ard/vis/visprot.html

Malm, William C. 1999. Introduction to visibility.Cooperative Institute for Research in theAtmosphere. NPS Visibility Program, ColoradoState University, Fort Collins, CO: 79 p.

Elemental carbon particles (or light absorbingcarbon) are emitted by virtually all combustionactivities, but are especially prevalent in dieselexhaust and smoke from wood burning.

In the West, sulfates contribute less than 30%(Oregon, Idaho and Nevada) to 40-50% (Ari-zona, New Mexico and Southwest Texas) oflight extinction. Carbon particles in the Westare a greater percentage of the extinction budgetranging from 50% or greater in the Northwest to30-40% in the other western regions. Thehigher percentages of the extinction budgetassociated with carbon particles in the Westappear to be from smoke emitted by wildlandand agricultural fires (NPS 1994).

In summary, the physics of light extinction inthe atmosphere coupled with the chemicalcomposition and physical size distribution ofparticles in wildland fire smoke combine tomake fire (especially in the West) an importantcontributor to visibility impairment. Wildlandfire managers responsible for the protection ofthe scenic vistas of this nation’s wildernessareas and national parks have a difficult chal-lenge in balancing the need to protect visibilitywith the need to use fire for other resourcemanagement goals.


– 40 –

– 41 –

2001 Smoke Management Guide 3.3 – Nuisance Smoke

Problem and Nuisance Smoke

Gary L. Achtemeier

Bill Jackson

James D. Brenner

Introduction

The particulate matter (or particles) producedfrom wildland fires can be a nuisance or safetyhazard to people who come in contact with thesmoke – whether the contact is directly throughpersonal exposure, or indirectly through visibil-ity impairment. Nuisance smoke is defined bythe US Environmental Protection Agency as theamount of smoke in the ambient air that inter-feres with a right or privilege common tomembers of the public, including the use orenjoyment of public or private resources (USEPA 1990).

Although the vast majority of prescribed burnsoccur without negative smoke impact, wildlandfire smoke can be a problem anywhere in thecountry. Complaints about loss of visibility,odors, and soiling from ash fallout are notunique to any region. Reduced visibility fromsmoke has caused fatal collisions on highwaysin several states, from Florida to Oregon. Ac-rolein (and possibly formaldehyde) in smoke islikely to cause eye and nose irritation for dis-tances up to a mile from the fire, exacerbatingpublic nuisance conditions (Sandberg and Dost1990). The abatement of nuisance or problemsmoke is one of the most important objectives ofany wildland fire smoke management plan(Shelby and Speaker 1990).

This section provides information on the issueof visibility reduction from wildland fire smoke,and focuses particularly on smoke as a majorconcern in the Southern states. Meteorology,climate and topography combine with popula-tion density and fire frequency to make nuisancesmoke a chronic issue in the South. Lessonsfrom this regional example can be extrapolatedand applied to other parts of the country. Thissection also briefly summarizes tools currentlyused or under development to aid the landmanager in reducing the problematic effects ofsmoke.

Wildland fire smoke may also be a nuisance tothe public by producing a regional haze, whichis discussed in Section 3.2.

Nuisance Smoke andVisibility Reduction

A prescribed fire is a combustion process thathas no pollution control devices to remove thepollutants. Instead, prescribed fire practitionersoften rely on favorable atmospheric conditionsto successfully disperse the smoke away fromsmoke-sensitive areas, such as communities,


– 42 –

areas of heavy vehicle traffic, and scenic vistas.At times, however, unexpected changes inweather (especially wind), or planning whichdoes not adequately factor in such elements astopography, diurnal weather patterns, or residualcombustion, may result in an intrusion of smokethat causes negative impacts on the public.

Smoke intrusions and nuisance- or safety-related episodes may happen at any time duringthe course of a wildland fire, but they frequentlyoccur in valley bottoms and drainages during thenight. Within approximately one half hour ofsunset, air cools rapidly near the ground, andwind speeds decline as the cooled stable airmass“disconnects” from faster-moving air just aboveit. High concentrations of smoke accumulatenear the ground, particularly smoke from smol-dering fuels that don’t generate much heat.Smoke then tends to be carried through drain-ages with little dispersion or dilution. If thedrainages are wet, smoke can act as a nucleatingagent and can actually assist the formation oflocal fog, a particular problem in the Southeast.Typically, the greatest fog occurs where smokeaccumulates in a low drainage. This can causehazardous conditions where a drainage crosses aroad or bridge, reducing visibility for traffic.

Visibility reduction may also result from thedirect impact of the smoke plume. Fine par-ticles (less than 2.5 microns in diameter) ofsmoke are usually transported to the upperreaches of the atmospheric mixing height, wherethey are dispersed. They may, however, dispersegradually back to ground level in an unstableatmosphere (figure 3.3.1). When this occurs,such intrusions of smoke can cause numerousnuisance impacts as well as specific safetyhazards.

Visibility reduction is used as a metric of smokeintrusions in several State smoke managementprograms. The State of Oregon program opera-tional guidance defines a “moderately” intenseintrusion as a reduction of visibility from 4.6 to

11.4 miles from a background visibility of morethan 50 miles (Oregon Dept. Forestry 1992).The State of Washington smoke intrusionreporting system uses “slightly visible, notice-able impact on visibility, or excessive impact onvisibility” to define light, medium and heavyintrusions (Washington Dept. Natural Resources1993). The New Mexico program requires thatvisibility impacts of smoke be considered indevelopment of the unit’s burn prescription(New Mexico Environmental ImprovementBoard 1995).

Smoke plume-related visibility degradation inurban and rural communities is not subject toregulation under the Clean Air Act. Nuisancesmoke is usually regulated under state and locallaws and is frequently based on either publiccomplaint or compromise of highway safety(Eshee 1995). Public outcry regarding nuisancesmoke often occurs before smoke exposuresreach levels that violate National Ambient AirQuality Standards. The Courts have ruled thatthe taking of private property by interfering withits use and enjoyment caused by smoke withoutjust compensation is in violation of federalconstitutional provisions under the FifthAmendment. The trespass of smoke maydiminish the value of the property, resulting inlosses to the owner (Supreme Court of Iowa1998).

Smoke as a Southern Problem

The Forest Atlas of the United States (figure3.3.2) shows that the thirteen Southern statescontain approximately 40% of U.S. forests –about 200 million acres. While not all of thisforested land is regularly burned, the extensiveforest type generally known as “southern pines”burns with a high fire frequency, about every 2-5years. When shrublands and grasslands areadded to the total, from four to six million acresof southern wildlands are subjected to pre-

– 43 –


Figure 3.3.1. Graphic from the dispersion model VSmoke-GIS, showing the rise anddescent of a smoke plume during a daytime prescribed fire, assuming 25% of thesmoke disperses at ground level.

Figure 3.3.2. National Atlas of Forest Cover Types. Southern forests (outlined in blue extendfrom Virginia to Texas and from the Ohio River southward and account for approximately 40% ofU.S. forest land.


– 44 –

scribed fire each year. This is by far the largestacreage of wildlland subjected to prescribed firein any region of the country.

Figure 3.3.3 shows the 1998 Population DensityClasses for the United States. Of particularimportance regarding problem smoke is theclass “Wildland/Urban Interface,” designated inred. A comparison with figure 3.3.2 shows thatthe wildland/urban interface falls within muchof the range of Southern forests. Southernforests, with highest treatment intervals ofprescribed fire and with the largest acreagessubjected to prescribed fire, are connected withhuman habitation and activity through an enor-mous wildland/urban interface. The potentialexists for significant smoke problems in thisregion.

Smoke and Southern Climate

Several factors regarding climate add to thesmoke problem in the South. The long growing

season allows time for more annual biomassproduction relative to other areas of the countrywith shorter growing seasons. Most of theSouthern forests are located farther south thanforests elsewhere in the country. Consequently,the sun angle is higher in the South and iscapable of supplying warmth well into the latefall and early winter. Further, most southernwildlands are located at low elevations wherethe air is warmer. These factors contribute tothe long growing season, which runs fromMarch/April through October/November.

Abundant rainfall also encourages growth of alarge number of grasses, shrubs, and trees.Most of the South receives 40-60 inches (100-150 cm) of precipitation annually. This copiousrainfall, in combination with the long growingseason, creates conditions for rapid buildup ofboth dead and live fuels. If burns are not con-ducted frequently, the increase in emissionsfrom the accumulated fuels may enhance thelikelihood of negative smoke impacts when firesdo occur.

Figure 3.3.3. Population density classes showing wildland/urban interface in red.Southern forests outlined in blue. [http://www.fs.fed.us/fire/fuelman]

– 45 –


The coincidence of dormant-season burningwith the winter rain season is a third factorcontributing to nuisance smoke. Althoughburning is conducted year round throughout theSouth, a significant amount of burning is doneduring January through March. In a typicalyear, anywhere from 10-20 inches (25-50 cm) ofrain will fall over Southern forests during thisthree-month period. In some areas of the coun-try, the question might be, “Is it wet enough toburn?” In the South, the question is commonly,“Is it dry enough to burn?” Fires burning intomoist fuel burn less efficiently and smolderlonger than fires burning dry fuels. Both factorsincrease smoke production. In addition, lessheat is produced during inefficient combustionand smoldering. Therefore, more smoke staysnear the ground and increases the risk of prob-lem smoke.

Smoke andSouthern Meteorology

All thirteen Southern states have implementedburning regulations designed to limit openburning to those days when burning is consid-ered “safe” and the risks of fire escapes areminimal. Many have implemented smokemanagement regulations. The need to conductburning in a manner to reduce impacts on airquality over sensitive targets has encouraged“best practice” approaches to open burning.

Efforts to avoid smoke incursions over sensitivetargets are often complicated by the highlyvariable meteorology of Southern weathersystems during the extensive burn season. Fourweather features that cause frequent wind shiftsand may be accompanied by rapid changes in airmass stability and mixing height are describedbelow.

1. Synoptic scale high- and low-pressuresystems and accompanying fronts frequentthe South during the winter burn season.In a typical sequence of events, the windsshift to blow from the southeast throughsouthwest in advance of a storm, then shiftrapidly to the northwest with cold frontpassage. Winds blow from the northwestfor a day or so but gradually diminish withthe approach of a high pressure system,becoming light and variable as the systempasses. Then winds shift back to southerlyin advance of the next storm. Low clouds,low mixing heights, and high stability oftenaccompany low-pressure systems. De-pending upon moisture availability, coldfronts may be accompanied by bands oflow clouds and precipitation. Mixingheights are more favorable during high-pressure episodes. Although the movementof synoptic scale weather systems into theSouth can be predicted with lead times ofseveral days, the timing of arrival of frontalwind shifts over specific burn sites is lesscertain.

2. Much of the Piedmont and Coastal Plainare flat and it would be expected that windsthere are steady and predictable. However,the region is frequented by transient eddiesthat can cause unexpected wind shifts andcarry smoke into sensitive areas. Thevertical circulation of air that can forcesmoke plumes to the ground or carrysmoke safely upward are well-understood,but the location, timing and strength of thevertical eddies cannot be predicted. Hori-zontal eddies have not been well docu-mented, and the timing, location andintensity cannot be predicted.

3. The South has the longest coastline of anyfire-prone area in the country. Thus it isaxiomatic that large areas of the South are


– 46 –

subject to wind shifts brought on by seabreezes during the day and by land breezesduring the night. However, the onset,duration, and intensity of these land/water-induced circulations are not consistentfrom one day to the next. The region issubject at different times to warm, humidairmasses drawn northward from the Gulfof Mexico, or cold, dry airmasses drawnsouthward from Canada. Both systemshave an impact on land surface tempera-tures, which results in a significant effecton the duration and extent of land and seabreezes and whether they form at all. Theunpredictability of these wind systems addsto the difficulties faced by Southern landmanagers planning whether smoke from aprescribed burn might impact downwindsensitive targets.

4. The “flying wedge,” a wind system causedby cold air channeled southwestward alongthe eastern slopes of the AppalachianMountains, can cause sudden wind shiftswith large changes in wind direction andlowering of mixing heights. AlthoughVirginia, the Carolinas and Georgia aremost frequently impacted, flying wedgeshave been observed as far south as centralFlorida and as far west as the MississippiRiver. “Flying wedges” occur throughoutthe year but are most intense, and hencebring with them strong shifting winds andlowering of mixing heights, during winterand early spring, the period of maximumwildland burning in the South.

Smoke and Southern Highways

As previously noted, several million acres ofSouthern wildlands are burned each year, thevast majority without incident. However, smoke

and smoke-induced fog obstructions of visibilityon highways sometimes cause accidents withloss of life and personal injuries. Severalattempts to compile records of smoke-impli-cated highway accidents have been made. Forthe 10-year period from 1979-1988, Mobley(1989) reported 28 fatalities, over 60 seriousinjuries, numerous minor injuries and millionsof dollars in lawsuits. During 2000, smoke fromwildfires drifting across Interstate 10 caused atleast 10 fatalities, five in Florida and five inMississippi. In their study of the relationshipbetween fog and highway accidents in Florida,Lavdas and Achtemeier (1995) compared threeyears of accident reports that mentioned fogwith fog reports at nearby National WeatherService stations. Highway accidents were morelikely to be associated with local ground radia-tion fogs than with widespread advection fogs.Accidents tended to happen when fog createdconditions of sudden and unexpected changes invisibility.

There are several reasons why smoke on thehighways is a serious problem in the South,some of them interrelated.

Road density: The density of the road networkin the South is far greater than in other wildlandareas in the country where prescribed fire is inwidespread use. The difference in road densitybetween generally forested areas in the west andin the south exists primarily because of land usehistory. While Western forested lands havealways been in forest, in the Southern area,roads and communities remain essentiallyunchanged from the old agricultural South.

Population in wildland areas: The populationdwelling near or within Southern wildlands isgreater than that in other areas of the countrywhere prescribed fire is in widespread use(figure 3.3.3). Many people live in close prox-imity to Southern forests; many more live in

– 47 –


areas interfacing fire-prone grasslands andshrublands. Southern States are becoming moreurban, and the numbers of tourists driving toresort areas along the Gulf coast, the Atlanticcoast, and the Florida peninsula are increasing.Therefore, the number of accidents related tosmoke and fog can only be expected to increase.

Climate and meteorology: Factors of Southernclimate and meteorology combine to produceairmasses that entrap smoke close to the groundat night. Smoke is most often trapped by eithera surface inversion or inversion aloft. This is acondition in which temperature increases withheight through a layer of the atmosphere. Verti-cal motion is restricted in this very stable airmass. Although most inversions dissipate withdaytime heating, inversions aloft caused bylarge-scale subsidence may persist for severaldays, resulting in a prolonged smoke manage-ment problem

Most smoke-related highway accidents occurjust before sunrise when temperatures arecoldest and smoke entrapment has maximizedunder a surface-level inversion. The high sunangle during the burn season contributes towarm daytime temperatures. Near sunset, underclear skies and near calm winds, temperatures inshallow stream basins can drop up to 20 degreesF. in one hour (Achtemeier 1993). Smoke fromsmoldering heavy fuels can be entrapped nearthe ground and carried by local drainage windsinto these shallow basins where temperaturesare colder and relative humidities are higher.Hygroscopic particles within smoke can assist indevelopment of local dense fog. Weak drainagewinds of approximately 1 mile per hour (0.5m/sec) can carry smoke over 10 miles during thenight—far enough in many areas to carry thesmoke or fog over a roadway.

Problem Smoke: What is beingdone to Minimize the Problem

As population growth in the South continues,there is an increasing likelihood that morepeople will be adversely impacted by smoke.Unless methods are found to mitigate the im-pacts of smoke, increasingly restrictive regula-tions may curtail the use of prescribed fire, orfire as a management tool may be prohibited.Several approaches are underway to reduce theuncertainty in predicting smoke movement.

• Several states have devised smoke man-agement guidelines to regulate the amountof smoke put into the atmosphere fromprescribed burning. The South CarolinaForestry Commission (1998) has estab-lished guidelines to define smoke sensitiveareas, amounts of vegetative debris thatmay be burned, and atmospheric condi-tions suitable for burning this debris.

• The Forestry Weather InterpretationSystem (FWIS) was developed by the U.S.Forest Service in the late 1970’s and early1980’s in cooperation with the southernforestry community (Paul 1981; Paul andClayton 1978). The system has beenenhanced and automated by the GeorgiaForestry Commission (Paul et al. 2000) toserve forestry sources in Georgia andclients in other southern states. The GFCprovides weather information and fore-casts specified for forest districts, andindices used for interpretations for smokemanagement, prescribed fire, fire danger,and fire behavior. Indices include theKeetch-Byram Drought Index, NationalFire Danger Rating System, Ignitioncomponent, Burning Index, and ManningClass Day.


– 48 –

• High resolution weather prediction modelspromise to provide increased accuracy inpredictions of wind speeds and directionsand mixing heights at time and spatialscales useful for land managers. TheFlorida Division of Forestry (FDOF) is aleader in the use of high resolution model-ing for forestry applications in the South(Brackett et al. 1997). Accurate predic-tions of sea/land breezes and associatedchanges in temperature, wind direction,atmospheric stability and mixing heightare critical to the success of the FDOFsystem as much of Florida is locatedwithin 20 miles of a coastline. Highresolution modeling consortia are alsobeing established by the U.S. ForestService to serve clients with interests asdiverse as fire weather, air quality, ocean-ography, ecology, and meteorology.

• Several smoke models are in operation orare being developed to predict smokemovement over Southern landscapes.VSMOKE (Lavdas 1996), a Gaussianplume model that assumes level terrainand unchanging winds, predicts smokemovement and concentration during theday. VSMOKE is now part of the FDOFfire and smoke prediction system. It is ascreening model that aids land managersin assessing where smoke might impactsensitive targets as part of planning forprescribed burns. PB-Piedmont(Achtemeier 2001) is a wind and smokemodel designed to simulate smoke move-ment near the ground under entrapmentconditions at night. The smoke plume issimulated as an ensemble of particles thatare transported by local winds over com-plex terrain characteristic of the shallow(30-50 m) interlocking ridge/valley sys-tems typical of the Piedmont of the South.PB-Piedmont does not predict smoke

concentrations as emissions from smolder-ing combustion are usually not known.Two sister models are planned, one thatwill simulate near ground smoke move-ment near coastal areas influenced by sea/land circulations and the other for theAppalachian mountains.

In summary, the enormous wildland/urbaninterface and dense road network located in aregion where up to six million acres of wild-lands per year are subject to prescribed firecombine to make problem smoke the foremostland management-related air quality problem inthe South. During the daytime, smoke becomesa problem when it drifts into areas of humanhabitation. At night, smoke can become en-trapped near the ground and, in combinationwith fog, create visibility reductions that causeroadway accidents. Public outcry regardingproblem smoke usually occurs before smokeexposures increase to levels that violate airquality standards. With careful planning andknowledge of local conditions, the fire managercan usually avoid problematic smoke intrusionson the public.

Literature CitationsAchtemeier, G.L. 2001. Simulating nocturnal smoke

movement. Fire Management Today, 61: 28-33

Achtemeier, G. L. 1993. Measurements of drainagewinds along a small ridge. In: Proceedings ofthe 86th Annual Meeting and Exhibition, Air &Waste Management Association, 93-FA-155.06,15 pp.

Brackett, D.P., L. G. Arvanitus, J. Brenner, and M.Long, 1997. A high-tech approach to open-burning authorization and wildfire response. J.Forestry, 95: 10-15.

– 49 –


Eshee, W.D. 1995. Legal implications of usingprescribed fire. In: Proceedings: Environmentalregulation & prescribed fire conference. D.C.Bryan, ed. Division of Forestry, Florida Depart-ment of Agriculture and Consumer Services,Tallahassee, Florida. pp. 126-130.

Lavdas, L. G. 1996. Program VSMOKE–usersmanual. USDA Forest Service General Techni-cal Report SRS-6. 147 pp.

Lavdas, L., and G. L. Achtemeier. 1995. A fog andsmoke risk index for estimating roadwayvisibility hazard. National Weather Digest, 20:26-33.

Mobley, H. E. 1989. Summary of smoke-relatedaccidents in the South from prescribed fire(1979-1988). American Pulpwood AssociationTechnical Release 90-R-11.

New Mexico Environmental Improvement BoardRules. 1995. Open Burning 20-NMAC2.60.Santa Fe, New Mexico.

Paul, J. T., 1981: A real-time weather system forforestry (FWIS). Bulletin of the AmericanMeteorological Society, 62: 1466-1472.

Paul, J. T., and J. Clayton, 1978. User manual:Forestry Weather Interpretation System(FWIS). Asheville, NC: U.S. Department ofAgriculture, Forest Service, SoutheasternForest Experiment Station and Atlanta, GA:Southeastern Area State and Private Forestry, incooperation with the U.S. National WeatherService, NOAA, Silver Spring, MD, 83pp.

Paul, J. T., D. Chan, and A. Dozier, 2000. FireWeather and Smoke Management System.Georgia Forestry Commission, 74 pp.

Sandberg, D.V., and F. N. Dost. 1990. Effects ofprescribed fire on air quality and human health.In: Natural and Prescribed Fire in PacificNorthwest Forests. J.D. Walstad; S.R.Radosevich; D.V. Sandberg, eds. Oregon StateUniversity Press, Corvallis, OR. pp. 91-298.

Shelby, B., and R. W. Speaker. 1990. Public atti-tudes and perceptions about prescribed burning.In: J.D. Walstad; S.R. Radosevich; D.V.Sandberg, eds. Natural and Prescribed Fire inPacific Northwest Forests. Oregon StateUniversity Press. Corvallis, OR. pp. 253-260.

South Carolina Forestry Commission, 1998. SmokeManagement Guidelines for Vegetative DebrisBurning Operations in the State of SouthCarolina. SCFC 4th Printing, 21 p.

State of Oregon Department of Forestry. 1992.Smoke Management Program Directives,Appendix 2. Salem, Oregon.

State of Washington Department of Natural Re-sources. 1993. Smoke Management ProgramAppendix 5. Olympia, Washington.

Supreme Court of Iowa, No. 192/96-2276.Bormann, et al vs. Board of Supervisors in andfor Kossuth County Iowa. September 23, 1998.{http://www.webcom.com/bi /iowa.htm}

US Environmental Protection Agency. 1990. AirQuality Criteria for Particulate Matter VolumeII of III. Office of Research and Development.EPA/600/P-95/00ibF. pp. 8-82 – 8-89.


– 50 –

– 51 –

2001 Smoke Management Guide 3.4 – Smoke Exposure

Smoke Exposure AmongFireline Personnel

Roger D. Ottmar

Timothy R. Reinhardt

Wildland firefighting presents many hazards tofireline workers, including inhalation exposureto smoke (Sharkey 1998; Reinhardt and Ottmar1997; Sharkey 1997). Many experiencedfireline personnel consider this to be only aninconvenience, occasionally causing acute casesof eye and respiratory irritation, nausea andheadache. Others express concern about long-term health impacts, especially when large-scale fires occur in terrain and atmosphericconditions that force fireline workers to workfor many days in smoky conditions. At thepresent time, no one can say whether there arelong-term adverse health effects from occupa-tional smoke exposure. This is because therehave been no epidemiological studies to trackthe health of fireline personnel and compare itwith other workers to see if fireline personnelhave more or fewer health problems during andafter their careers. Until such long-term dataare examined to tell us if a problem exists, wecan only assess the occurrence of relativelyshort-term adverse health effects. We canmeasure fireline worker’s exposure to particlesand individual chemicals found in smoke andcompare these exposures to standards estab-lished to protect worker health (Reinhardt andOttmar 2000; Reinhardt and others 2000;Reinhardt and others 1999). We can evaluatethe relative risk of disease among fireline

workers based on the exposure data and thepotency of the health hazards (Booze andReinhardt 1996).

Health Hazards in Smoke

Smoke from wildland fires is composed ofhundreds of chemicals in gaseous, liquid, andsolid forms (Sandberg and Dost 1990; Reinhardtand Ottmar 2000; Reinhardt and others 2000;Sharkey 1998; Sharkey 1997). The chief inhala-tion hazards for fireline personnel and to thegeneral public when they are exposed to smokeappear to be carbon monoxide and respirableirritants which include particulate matter, ac-rolein, and formaldehyde.

Carbon Monoxide — Carbon monoxide (CO)has long been known to interfere with the body’sability to transport oxygen. It does this bybonding with hemoglobin, the molecule in thebloodstream which shuttles oxygen from thelungs throughout the body, to form carboxyhe-moglobin (COHb). When people are exposed toCO, the time until a toxic level of COHb resultscan be predicted as a function of CO concentra-tion, breathing rate, altitude, and other factors(Coburn, Forster and Kane 1965). The harderthe work and the higher the altitude, the more


– 52 –

rapidly COHb forms at a given level of atmo-spheric CO. At the highest CO levels found inheavy smoke, symptoms of excessive COHb canresult in 15 minutes during hard physical labor.

Carbon monoxide causes acute effects rangingfrom diminished work capacity to nausea,headache, and loss of mental acuity. It has awell-established mechanism of action, causingdisplacement of oxygen from hemoglobin in theblood and affecting tissues that do not stand theloss of oxygen very well, such as the brain,heart, and unborn children. Fortunately, most ofthese effects are reversible and CO is rapidlyremoved from the body, with a half-life on theorder of 4 hours. Some studies have linked COexposure to longer-term heart disease, but theevidence is not clearcut.

Respirable Irritants — Experienced firelineworkers can attest to eye, nose and throat irrita-tion at both wildfires and prescribed burns.Burning eyes, runny nose, and scratchy throatare common symptoms in smoky areas atwildland fires, caused by the irritation of mu-cous membranes. These adverse health effectsare symptoms of exposure to aldehydes, includ-ing formaldehyde, acrolein, as well as respirableparticulate matter (PM2.5)—very fine particlesless than a few micrometers (µm) in diameter—composed mostly of condensed organic andinorganic carbon (Dost 1991). Other rapidadverse health effects of aldehydes includetemporary paralysis of the respiratory tract cilia(microscopic hairs which help to remove dustand bacteria from the respiratory tract) anddepression of breathing rates (Kane and Alarie1977), while over the long term, formaldehydeis considered a potential cause of nasal cancer(U.S. Department of Labor, Occupational Safetyand Health Administration 1987).

Adverse health effects of smoke exposure beginwith acute, instantaneous eye and respiratoryirritation and shortness of breath but can de-velop into headaches, dizziness and nausealasting up to several hours. The aldehydes, suchas acrolein and formaldehyde, and PM2.5 causerapid minor to severe eye and upper respiratorytract irritation. Total supsended particulate(TSP) also irritates the eyes, upper respiratorytract and mucous membranes, but the largerparticulates in TSP do not penetrate as deeplyinto the lungs as the finer PM2.5 particles.Longer-term health effects lasting days toperhaps months have recently been identifiedamong fireline workers, including modest lossesof pulmonary function. These include a slightlydiminished capacity to breathe, constriction ofthe repsiratory tract, and hypersensitivity of thesmall airways (Letts and others 1991; Reh andothers 1994).

A discussion of particulate inhalation hazardsfaced by fireline personnel is incomplete with-out mentioning crystalline silica, which can bean additional hazard in the presence of smoke.If crystalline silica is a component of the soil ata site, dust stirred up by walking, digging, mop-up, or vehicles may be a significant irritationhazard, and the threat of silicosis (fibrousscarring of the lungs decreasing oxygenationcapability) is a possibility.

Evaluation Criteria

On what basis do we decide whether smokeexposure is safe or unsafe? Workplace expo-sures to health hazards must be evaluated withcare for several reasons. First, people vary intheir sensitivity to pollutants. Second, personalhabits and physical condition are importantfactors. For example, smokers already com-monly experience 5% COHb because of the CO

– 53 –


from their cigarettes, thus they may be at greaterrisk of adverse health effects from additionalCO exposure at fires. Assumptions are made byregulatory agencies when establishing exposurelimits. These assumptions may not be valid forthe wildland fire workplace. For example, thecurrent CO standard was set to protect a seden-tary worker in an 8-hour per day job over aworking lifetime, not a hard-working firelineworker on a 12-hour/day job for a few summers.

Given these issues, how should we judge thesafety of smoke exposure? At a minimum, afireline worker’s inhalation exposures mustcomply with the occupational exposure limits,called “Permissible Exposure Limits” (PEL’s),by the Occupational Safety and Health Adminis-tration (OSHA) (U.S. Department of Labor,Occupational Safety and Health Administration1994). These limits are set at levels consideredfeasible to attain, and necessary to protect mostworkers from adverse health effects over theirworking lifetime. The more stringent expo-sure limits recommended by the AmericanConference of Governmental Industrial Hygien-ists (ACGIH) are the “Threshold Limit Values”(TLVs) (American Conference of Governmental

Industrial Hygienists 2000). These are alsoestablished to prevent adverse health effects inmost workers, but without adjustment foreconomic feasibility. The ACGIH limits areperiodically updated to incorporate the latestscientific knowledge where as many of thePEL’s have not been revised since the 1960’s.All exposure limits are expressed in terms of atime-weighted average (TWA) exposure, whichis an average exposure over the workshift. Forhealth hazards which quickly cause adverseeffects from acute exposures, the limits aresupplemented by short-term exposure limits(STELs) for 15-minute periods in a workshiftand ceiling exposure limits (C), which are not tobe exceeded at any time. These various expo-sure limits are listed in table 3.4.1, along with athird set of “Recommended Exposure Limits”established by the National Institute for Occupa-tional Safety and Health; these also incorporaterecent scientific evidence. Depending on thepollutant, the units of measure are either milli-grams per cubic meter of air (mg/m3) or partsper million by volume (ppm). Without a moredetailed analysis of a given work/rest regime,adhering to the ACGIH TLV limits shouldprovide reasonable protection for workers.

Table 3.4.1. Occupational exposure limitsa


– 54 –

Smoke Exposure at PrescribedBurns and Wildfires

Several studies (Reinhardt and Ottmar 1997)have evaluated smoke exposure during pre-scribed burns by obtaining personal exposuresamples, which are collected within a foot of aworker’s face (the breathing zone) while theyare on the job (figure 3.4.1). One study inparticular measured smoke exposure amongfireline workers at 39 prescribed burns in thePacific Northwest. The study found that about10% of firefighter exposures to respiratoryirritants and CO exceeded recommended occu-pational exposure limits (Reinhardt and others2000) and could pose a hazard. The actualincidence of illness and mortality among wild-land fireline workers has not been systemati-cally studied, but short-term adverse healthimpacts have been observed among firelinepersonnel at prescribed fires. A study in 1992-93 found small losses in lung function among 76

fireline personnel working at prescribed burns(Betchley and others 1995).

Between 1992 and 1995 a study of smokeexposure and health effects at wildfires in thewestern United States found results similar tothose at prescribed fires. Exposure to carbonmonoxide and respiratory irritants exceededrecommended occupational exposure limits for5 percent of workers (Reinhardt and Ottmar2000).

At wildfires where fireline workers encounterconcentrated smoke, or moderate smoke overlonger times, there is a likelihood that many willdevelop symptoms similar to those seen atprescribed fires. In 1988, engine-basedfirefighters of the California Department ofForestry and Fire Protection underwent lungfunction testing before and after the fire season.Small (0.3 to 2%) losses in lung function wereobserved among the firefighters. These losses

Figure 3.4.1. Bitterroot Hotshot crew member wearing backpack that obtains smokeexposure samples collected within several inches of a worker’s face.

– 55 –


were associated with the amount of recentfirefighting activity in the study period. Thefirefighters also reported increased eye and noseirritation and wheezing during the fire season.

Monitoring Smoke Exposure ofFireline workers

During prescribed fire and wildfire exposurestudies, it was found that exposure to respiratoryirritants could be predicted from measurementsof carbon monoxide (Reinhardt and Ottmar2000). Fire managers and safety officers con-cerned with smoke exposure among fire crewscan use electronic carbon monoxide (CO)monitors to track and prevent overexposure tosmoke (figure 3.4.2). Commonly referred to as

dosimeters, these lightweight instrumentsmeasure the concentration of CO in the airthatfireline personnel breath. Protocols havebeen developed for sampling smoke exposureamong fireline workers with CO dosimeters.These protocols and a basic template have beenoutlined by Reinhardt and others (1999) formanagers and safety officers interested inestablishing their own smoke-exposure monitor-ing program.

Respirator Protection

The Missoula Technology and DevelopmentCenter (MTDC) has the lead role in studyingrespiratory protection for fireline workers(Thompson and Sharkey 1966, Sharkey 1997).

Figure 3.4.2. Carbon monoxide exposure data from a electronic CO data recorder for a fireline workerduring a work-shift on a prescribed fire (Reinhardt and others 2000)


– 56 –

Although respirators reduce work capacity, theymay have merit under certain circumstances tominimize hazardous exposures. Field evalua-tions by MTDC found that disposable respira-tors were acceptable for short-term use but theydeteriorated in the heat during several hours ofuse (Sharkey 1997). Maintenance free half-mask devices were satisfactory, except for theheat stress found with all facemasks. Full-facemasks were preferred for the long-term use onprescribed fires because of the eye protectionthey provided, but workers often complained ofheadaches, a sign of excess CO exposure sincerespirators do not eliminate the intake of CO(Sharkey 1997). Full-face respirators protect theeyes, removing eye irritation as an importantearly warning of exposure to smoke. Any respi-ratory protection program for fireline workerswould require employees to be instructed andtrained in the proper use and limitations of therespirators issued to them.

Management Implications

Evidence to date suggests that fireline workersexceed recommended exposure limits duringprescribed burns and wildfires less than 10percent of the time (Reinhardt and others 2000;Reinhardt and Ottmar 2000). The concept thatfew fireline personnel spend a working lifetimein the fire profession and should be exempt fromoccupational exposure standards which are setto protect workers over their careers is littlecomfort to those who do, and irrelevant forirritants and fast-acting hazards such as CO.Most of the exposure limits that are exceededare established to prevent acute health effects,such as eye and respiratory irritation, headache,nausea and angina. An exposure standardspecifically for fireline workers, and appropriate

respiratory protection, needs to be developed.In addition, a long-term program to managesmoke exposure at wildland fires is needed(Sharkey 1997). The program could include:1) hazard awareness training; 2) implementationof practices to reduce smoke exposure; 3)routine CO monitoring with electronic dosim-eters (Reinhardt and others 1999); 4) improvedrecord keeping on accident reports to includeseparation of smoke related illness amongfireline workers and fire camp personnel; and 4)implementing and training for an OSHA-compliant respirator program to protect firelinepersonnel from respiratory irritants and COwhen they must work in smoky conditions.

Literature Citations

American Conference of Governmental IndustrialHygienists. 2000. 2000 TLVs® and BEIs®:Threshold Limit Values for chemical substancesand physical agents and biological exposureindices. Cincinnati, OH 192 p.

Betchley, C.; Koenig, J.Q.; van Belle G.;Checkoway, H. and Reinhardt, T. 1995. Pulmo-nary function and respiratory symptoms inforest firefighters, Unpublished report, Univer-sity of Washington, Departments of Environ-mental Health and Epidemiology, Seattle, WA

Booze, Tom F.; Reinhardt, Timothy E. 1996.Assessment of the health risks of chronic smokeexposure for wildland firefighters. Unpublishedreport, Pacific Northwest Research Station,Seattle forestry Science Laboratory, 4043Roosevelt Way N.E., Seattle, WA. 29 p.

Coburn, R.F.; Forster, R.E.; Kane, P.B. 1965. Con-siderations of the physiological variables thatdetermine the blood carboxyhemoglobinconcentration in man. Journal of ClinicalInvestigation 44: 1899-1910.

– 57 –


Dost, F.N. 1991. Acute toxicology of components ofvegetation smoke. In: Reviews of Environmen-tal Contamination and Toxicology, Vol 119,Springer-Verlag New York, Inc., New York, NY,1991 pp 2-46.

Kane, L.E.; Alarie, Y. 1977. Sensory irritation toformaldehyde and acrolein during single andrepeated exposures in mice. American IndustrialHygiene Association Journal 38(10): 509-522.

Letts, D.; Fidler, A.T.; Deitchman, S.; Reh, C.M.1991. Health hazard evaluation report 91-152-2140. U.S. Department of the Interior, NationalPark Service, Southern California, U.S. Depart-ment of Health and Human Services, PublicHealth Service, Centers for Disease Control,National Institute for Occupational Safety andHealth, Cincinnati, OH.

Reh, C.M.; Letts, D.; Deitchman, S.D. 1994. Healthhazard evaluation report 90-0365-2415. U.S.Department of the Interior, National ParkService, Yosemite National Park, California,U.S. Department of Health and Human Ser-vices, Public Health Service, Centers forDisease Control, National Institute for Occupa-tional Safety and Health, Cincinnati, OH.

Reinhardt, Timothy E.; Ottmar, Roger D. 1997.Smoke exposure among wildland firefighters: areview and discussion of current literature. Gen.Tech. Rep. GTR-373. Portland, OR: U.S.Department of Agriculture, Forest service,Pacific Northwest research Station. 61 p.

Reinhardt, Timothy E.; Ottmar, Roger D. 2000.Smoke exposure at western wildfires. Res. Pap.PNW-RP-525. Portland, OR: U.S. Departmentof Agriculture, Forest Service, Pacific North-west Research Station. 72 p.

Reinhardt, Timothy E.; Ottmar, Roger D.;Hanneman, Andrew J.S. 2000. Smoke exposureamong firefighters at prescribed burns in thePacific Northwest. Res. Pap. PNW-RP-526.Portland, OR: U.S. Department of Agriculture,Forest Service, Pacific Northwest Research

Station. 45 p.

Reinhardt, Tim E.; Ottmar, Roger D.; Hallett,Michael R. 1999. Guide to monitoring smokeexposure of wildland firefighters. Gen. Tech.Rep. PNW-GTR-448. Portland, OR: U.S.Department of Agriculture, Forest service,Pacific Northwest Research Station. 17 p.

Sandberg, David V.; Dost, Frank N. 1990. Effects ofprescribed fire on air quality and human health.In: Walstad, John D.; Radosevich, Steven R.;Sandberg, David V., eds. Natural and prescribedfire in the Pacific Northwest forests. Corvallis,OR: Oregon State University Press: 191-298.

Sharkey, Brian, ed. 1997. Health hazards of smoke:recommendations of the April 1997 ConsensusConference. Tech. Rep. 9751-2836-MTDC.Missoula, MT: U.S. Department of Agriculture,Forest Service, Missoula Technology andDevelopment Center. 84 p.

Sharkey, Brian, ed. 1998. Health hazards of smoke.Closed caption video. Missoula, MT: U.S.Department of Agriculture, Forest Service,Missoula Technology and Development Center.19 minutes.

Thompson, S.; Sharkey, B. 1966. Physiological costand air flow resistance of respiratory protectivedevices. Ergonomics 9:495.

U.S. Department of Labor, Occupational Safety andHealth Admnistration. 1987. Occupationalexposure to formaldehyde, final rule. FederalRegister 52(233): 46312, Washington D.C.1987.

U.S. Department of Labor, Occupational Safety andHealth Administration. 1994. Code of FederalRegulations, Title 29, Part 1910.1048(d)(2)(i),Washington, DC, 1994.


– 58 –

– 59 –

2001 Smoke Management Guide 4.1 – Smoke Management

REGULATIONS

Chapter 4

Chapter 4 – Regulations 2001 Smoke Management Guide

– 60 –

– 61 –


Regulations For Smoke Management

Janice L. Peterson

Some of the components of smoke from pre-scribed fire are regulated air pollutants. And, aswith any other rule or regulation, fire managersmust understand and follow federal, state, andlocal regulations designed to protect the publicagainst possible negative effects of air pollution.

Air pollution is defined as the presence in theatmosphere of a substance or substances addeddirectly or indirectly by a human act, in suchamounts as to adversely affect humans, animals,vegetation, or materials (Williamson 1973). Airpollutants are classified into two major catego-ries: primary and secondary. Primary pol-lutants are those directly emitted into the air.Under certain conditions, primary pollutants canundergo chemical reactions within the atmo-sphere and produce new substances known assecondary pollutants.

Emissions from prescribed fire are managed andregulated through an often-complex web ofinterrelated laws and regulations. The over-arching law that is the foundation of air qualityregulation across the nation is the Federal CleanAir Act (Public Law 95-95).

Federal Clean Air Act

In 1955, Congress passed the first Federal CleanAir Act with later amendments in 1967, 1970,

1977, and 1990. The Clean Air Act is a legalmandate designed to protect public health andwelfare from air pollution. States developspecific programs for implementing the goals ofthe Clean Air Act through their State Implemen-tation Plans (SIP’s). States may develop pro-grams that are more restrictive than the CleanAir Act requires but never less. Burners mustknow the specifics of state air programs andhow fire emissions are regulated to responsiblyconduct a prescribed fire program.

Roles and Responsibilities

Although the Clean Air Act is a federal law andtherefore applies to the entire country, the statesdo much of the work of implementation. TheAct recognizes that states should have the leadin carrying out provisions of the Clean Air Act,since appropriate and effective design of pollu-tion control programs requires an understandingof local industries, geography, transportation,meteorology, urban and industrial developmentpatterns, and priorities.

The Clean Air Act gives the EnvironmentalProtection Agency (EPA) the task of settinglimits on how much of various pollutants can bein the air where the public has access1 (ambientair). These air pollution limits are the NationalAmbient Air Quality Standards or NAAQS and

___________________________________

1 Note that the Occupational Safety and Health Administration (OSHA), rather than EPA, sets air quality standardsfor worker protection.


– 62 –

are intended to be established regardless ofpossible costs associated with achieving them,though EPA is allowed to consider the costs ofcontrolling air pollution during the implementa-tion phase of the NAAQS in question. In addi-tion, EPA develops policy and technicalguidance describing how various Clean Air Actprograms should function and what they shouldaccomplish. States develop State Implementa-tion Plans (SIPs) that define and describe cus-tomized programs that the state will implementto meet requirements of the Clean Air Act.Tribal lands are legally equivalent to state landsand tribes prepare Tribal Implementation Plans(TIPs) to describe how they will implement theClean Air Act. The individual states and tribescan require more stringent pollution standards,but cannot weaken pollution goals set by EPA.The Environmental Protection Agency mustapprove each SIP/TIP, and if a proposed or activeSIP/TIP is deemed inadequate or unacceptable,

EPA can take over enforcing all or parts of theClean Air Act requirements for that state or tribethrough implementation of a Federal Implemen-tation Plan or FIP (figure 4.1.1).

National Ambient AirQuality Standards

The primary purpose of the Clean Air Act is toprotect humans against negative health orwelfare effects from air pollution. NationalAmbient Air Quality Standards (NAAQS) aredefined in the Clean Air Act as amounts ofpollutant above which detrimental effects topublic health or welfare may result. NAAQSare set at a conservative level with the intent ofprotecting even the most sensitive members ofthe public including children, asthmatics, andpersons with cardiovascular disease. NAAQS

Figure 4.1.1. Role of EPA and the states and tribes in Clean Air Act implementation.

– 63 –


have been established for the following criteriapollutants: particulate matter2 (PM10 andPM2.5), sulfur dioxide (SO2), nitrogen dioxide(NO2), ozone, carbon monoxide and lead (table4.1.1). Primary NAAQS are set at levels toprotect public health; secondary NAAQS are toprotect public welfare. The standards areestablished for different averaging times, forexample, annual, 24-hour, and 3-hour.

Table 4.1.1. National Ambient Air Quality Standards.

The major pollutant of concern in smoke fromwildland fire is fine particulate matter, bothPM10 and PM2.5. Studies indicate that 90percent of smoke particles emitted duringwildland burning are PM10 and about 90percent of PM10 is PM2.5 (Ward and Hardy1991). The most recent human health studieson the effects of particulate matter indicate thatit is fine particles, especially PM2.5, that arelargely responsible for health effects including

___________________________________

2 Particulate matter NAAQS are established for two aerodynamic diameter classes: PM10 is particulate matter 10micrometers or less in diameter, and PM2.5 is particulate matter that is 2.5 micrometers or less in diameter.


– 64 –

mortality, exacerbation of chronic disease, andincreased hospital admissions (Dockery andothers 1993, EPA 1996).

An area that is found to be in violation of aprimary NAAQS is labeled a non-attainmentarea (figure 4.1.2). An area once in non-attain-ment but recently meeting NAAQS, and withappropriate planning documents approved byEPA, is a maintenance area. All other areas areattainment or unclassified (due to lack of moni-toring). State air quality agencies can provide

up-to-date locations of local non-attainmentareas3. States are required through their SIP’s todefine programs for implementation, mainte-nance, and enforcement of the NAAQS withintheir boundaries. A non-attainment designationis a black mark on the states air agency’s abilityto protect citizens from the negative effects ofair pollution so states generally develop aggres-sive programs for bringing non-attainment areasinto compliance with clean air goals. Wildlandfire in and near non-attainment areas will bescrutinized to a greater degree than in attain-

Figure 4.1.2. PM10 nonattainment areas as of August 2001. See the EPA AIRData web page forcurrent nonattainment status for PM10 and al other criteria pollutants(http://www.epa.gov/air/data/mapview.html).

___________________________________

3 PM2.5 is a newly regulated pollutant so attainment/non-attainment status has not yet been determined. Monitoringmust take place for at least 3 years before a designation can be made.

– 65 –


ment areas (and may be subject to GeneralConformity rules, see section 4.3: Federal LandManagement-Special Requirements). Extra pre-planning, documentation, and careful schedulingof wildland fires will likely be required tominimize smoke effects in the non-attainmentarea to the greatest extent possible. In somecases, the use of fire may not be possible ifsignificant impacts to a non-attainment area arelikely.

Natural Events Policy

PM10 NAAQS exceedences caused by naturalevents are not counted toward non-attainmentdesignation if a state can document that theexceedance was truly caused by a natural eventand if the state then prepares a Natural EventsAction Plan (NEAP) to address human healthconcerns during future events4. Natural eventsare defined by this policy as wildfire, volcanicand seismic events, and high wind events.Prescribed fires used to mimic the natural roleof fire in the ecosystem are not considerednatural events under this policy. In response tothis potential conflict of terms, the Interim AirQuality Policy on Wildland and Prescribed Fires(EPA 1998) states that EPA will exercise itsdiscretion not to redisignate an area as non-attainment if the evidence is convincing thatfires managed for resource benefits caused orsignificantly contributed to violations of thedaily or annual PM2.5 or PM10 standards andthe state has a formal smoke managementprogram (see Section 4.2: State Smoke Man-agement Programs for more information).

A NEAP is developed by the state air pollutioncontrol agency in conjunction with the stake-

holders affected by the plan. States shouldinclude input from Federal, state, and privateland managers in areas vulnerable to fire whendeveloping a wildland fire NEAP. Also, agen-cies responsible for suppressing fires, localhealth departments, and citizens in the affectedarea should be involved in developing the plan.The NEAP should include documented agree-ments among stakeholders as to planned actionsand the parties responsible for carrying outthose actions.

A wildfire NEAP should include commitmentsby the state and stakeholders to:

1. Establish public notification and educationprograms.

2. Minimize public exposure to high concen-trations of PM10 due to future naturalevents such as by:

- identifying the people most at risk,

- notifying the at-risk public that an eventis active or imminent,

- recommending actions to be taken by thepublic to minimize their pollutant expo-sure,

- suggesting precautions to take if expo-sure cannot be avoided.

3. Abate or minimize controllable sources ofPM10 including the following:

- prohibition of other burning duringpollution episodes caused by wildfire,

- proactive efforts to minimize fuel load-ings in areas vulnerable to fire,

- planning for prevention of NAAQSexceedances in fire management plans.

___________________________________

4 Nichols, Mary D. 1996. Memorandum dated May 30 to EPA Regional Air Directors. Subject: Areas Affected byPM10 Natural Events. Available from the EPA Technology Transfer Network, Office of Air and Radiation Policy andGuidance at http://www.epa.gov/ttn/oarpg.


– 66 –

4. Identify, study, and implement practicalmitigating measures as necessary.

5. Periodic reevaluation of the NEAP.

Preparation of a NEAP provides the opportunityfor land managers to formally document, incooperation with state air agencies, that it isappropriate to consider prescribed fire a preven-tion, control, and mitigation measure for wild-fire (see item 4 above). Prescribed fire can beused to minimize fuel loadings in areas vulner-able to fire so that future wildfires can be con-tained in a smaller area and will produce lessemissions. This can lead to a greater under-standing by state air agencies of the potential airquality benefits from some types of prescribedfire in certain ecosystems. A recent NEAPprepared for the Chelan county area of Washing-ton State accomplished this goal5. The ChelanCounty NEAP recognizes planned efforts by theWenatchee National Forest to reduce fuelloadings through thinning, pruning of lowerbranches, and careful use of prescribed fire asways to minimize public exposure to particulatematter during wildfire season.

Hazardous Air Pollutants

Hazardous air pollutants or (HAPs) are identi-fied in Title III of the Clean Air Act Amend-ments of 1990 (Public Law 101-549) as 188different pollutants “which present, or maypresent, through inhalation or other routes ofexposure, a threat of adverse human health orenvironmental effects whether through ambientconcentrations, bioaccumulation, deposition, or

other routes.” The listed HAPs are substanceswhich are known or suspected to be carcino-genic, mutagenic, teratogenic, neurotoxic, orwhich cause reproductive dysfunction. Criteriapollutants (the six pollutants that are regulatedthrough established National Ambient AirQuality Standards) are excluded from the list ofHAPs.

De minimis Emission Levels

Air quality regulations allow omission of certainpollution sources in air quality impact analysesif they are considered very minor and are certainto have no detrimental effects. These sourcesare considered to emit pollutant amounts belowde minimis levels. For example, burning a slashpile with less than 100 tons of material is notsubject to permit or regulation in some areas.Emissions below de minimis levels are oftenexcluded from air quality regulations so this isan important concept to define in reference towildland fire. De minimis levels have beendefined for many industrial sources but littleguidance is available for many wildland activi-ties including prescribed fire. Some states havelocally defined de minimis levels for example inUtah, fires less than 20 acres per day in size andemitting less than 0.5 ton of total particulate perday are considered de minimis and can beignited without permit if burners register theproject and comply with clearing index proce-dures. Definition of de minimis levels is a topicthat needs further discussion between wildlandfire managers and regulatory agencies so guid-ance can be developed at the local and/or na-tional level.

___________________________________

5 Washington Department of Ecology. June 1997. Natural event action plan for wildfire particulate matter in ChelanCounty, Washington. 21p. Available from the Washington Department of Ecology, PO Box 47600, Olympia, WA 98504-7600.

– 67 –


Prevention of SignificantDeterioration

Another provision of the Clean Air Act thatsometimes comes up when discussing wildlandburning activities is the Prevention of Signifi-cant Deterioration provisions or PSD. The goalof PSD is to prevent areas that are currentlycleaner than is allowed by the NAAQS frombeing polluted up to the maximum ceilingestablished by the NAAQS. States and tribesuse the permitting requirements of the PSDprogram to manage and limit air pollutionincreases over a baseline concentration. A PSDbaseline is the pollutant concentration at a pointin time when the first PSD permit was issued forthe airshed. New or modified major air pollu-tion sources must apply for a PSD permit priorto construction and test their proposed emissionsagainst allowable PSD increments.

Three air quality classes were established by theClean Air Act, PSD provisions, including ClassI, Class II, and Class III. Class I areas aresubject to the tightest restrictions on how muchadditional pollution, or increment, can be addedto the air. Class I areas include Forest Servicewildernesses and national memorial parks over5000 acres, National parks exceeding 6000acres, and international parks, all of which musthave been in existence as of August 7, 1977,plus later expansions to these areas (figure4.1.3). These original Class I areas are declared“mandatory” and can never be redesignated toanother air quality classification. In addition, afew Indian tribes have redesignated their landsto Class I. Redesignated Class I areas are notmandatory Class I areas so are not automaticallyprotected by all the same rules as defined by theClean Air Act unless a state or tribe chooses,through a SIP or TIP, to do so. Since no areashave ever been designated Class III, all otherlands are Class II, including everything fromnon-Class I wildlands to urban areas.

Historically, EPA has regarded smoke fromwildland fires as temporary and therefore notsubject to issuance of a PSD permit, but whetheror not wildland fire smoke should be consideredwhen calculating PSD increment consumptionor PSD baseline was not defined. EPA recentlyreaffirmed that states could exclude managedfire emissions from increment analyses, pro-vided the exclusion does not result in permanentor long-term air quality deterioration (EPA1998). States are also expected to consider theextent to which a particular type of burningactivity is truly temporary, as opposed to anactivity that can be expected to occur in aparticular area with some regularity over aperiod of time. Oregon is the only state that hasthus far chosen to include prescribed fire emis-sions in PSD increment and baseline calcula-tions.

Visibility

The 1977 amendments to the Clean Air Actestablished a national goal of “the prevention ofany future, and the remedying of any existing,impairment of visibility in mandatory class IFederal areas which impairment results frommanmade air pollution” (Public Law 95-95).States are required to develop implementationplans that make “reasonable progress” towardthe national visibility goal.

Atmospheric visibility is influenced by scatter-ing and absorption of light by particles andgases. Particles and gases in the air can obscurethe clarity, color, texture, and form of what wesee. The fine particles most responsible forvisibility impairment are sulfates, nitrates,organic compounds, elemental carbon (or soot),and soil dust. Sulfates, nitrates, organic carbon,and soil tend to scatter light, whereas elementalcarbon tends to absorb light. Wildland fire


– 68 –

Fig

ure

4.1.

3. M

anda

tory

Cas

s I a

reas

.

– 69 –


smoke is primarily made up of elemental car-bon, organic carbon, and particulate matter.Fine particles (PM2.5) are more efficient perunit mass than coarse particles (PM10 andlarger) at causing visibility impairment. Natu-rally occurring visual range in the East is esti-mated to be between 60 and 80 miles, whilenatural visual range in the West is between 110-115 miles (Trijonis and others 1991). Currently,visual range in the Eastern US is about 15 to 30miles and about 60 to 90 miles in the WesternUS (40 CFR Part 51). The theoretical maximumvisual range with nothing in the air except airmolecules is about 240 miles.

Federal Land Managers (FLMs) have somewhatconflicting roles when it comes to protectingvisibility in the Class I areas they manage. Onthe one hand, FLMs are given the responsibilityby the Clean Air Act for reviewing PSD permitsof major new and modified stationary pollutionsources and commenting to the state on whetherthere is concern for visibility impacts (or otherresource values) in Class I areas downwind ofthe proposed pollution source. In this caseFLMs play a proactive role in air pollutionprevention. On the other hand, however, FLMsalso use wildland fire, which emits visibility-impairing pollutants. In this case the FLM is thepolluter and is often in the difficult position oftrying to explain why wildland burning smokemay be acceptable in wilderness whereas othertypes of air pollution are not. The answer to thisdilemma is that wildernesses are managed topreserve and protect natural conditions andprocesses. So in this context, smoke and visibil-ity impairment from wildland fire that closelymimics what would occur naturally is generallyviewed as acceptable under wilderness manage-ment objectives, whereas visibility impairmentfrom “unnatural” pollutants and “unnatural”pollution sources is not.

The key to successfully promoting this distinc-tion is an honest and scientific definition of howmuch, and what types, of fire are “natural” thatFLMs, air quality regulators, and the public canagree upon. This is a critical area of futurecooperation in smoke management and airquality regulation.

Regional Haze

Regional haze is visibility impairment producedby a multitude of sources and activities that emitfine particles and their precursors, and arelocated across a broad geographic area. Thiscontrasts with visibility impairment that can betraced largely to a single, very large pollutionsource. Until recently, the only regulations forvisibility protection addressed impairment thatis reasonably attributable to a permanent, largeemission source or small group of large sources.Recently, EPA issued regional haze regulationsto manage and mitigate visibility impairmentfrom the multitude of diverse regional hazesources (40 CFR Part 51). The regional hazeregulations call for states to establish goals forimproving visibility in Class I national parksand wildernesses and to develop long-termstrategies for reducing emissions of air pollut-ants that cause visibility impairment. Wildlandfire is one of the sources of regional haze cov-ered by the new rules.

Current data from a national visibility monitor-ing network (Sisler and others 1996) do notshow fire to be the predominant source ofvisibility impairment in any Class I area (40CFR Part 51). Emissions from fire are animportant episodic contributor to atmosphericloading of visibility-impairing aerosols, includ-ing organic carbon, elemental carbon, andparticulate matter. Certainly the contribution tovisibility impairment from fires can be substan-


– 70 –

tial over short periods of time, but fires ingeneral, occur relatively infrequently and thushave a lesser contribution to long-term averages.Fire events contribute less to persistent visibilityimpairment than sources with emissions that aremore continuous.

Reasonable Progress

The visibility regulations require states to make“reasonable progress” toward the Clean Air Actgoal of “prevention of any future, and theremedying of any existing, impairment ofvisibility…”. The regional haze regulations didnot define visibility targets, but instead gave thestates flexibility in determining reasonableprogress goals for Class I areas. States arerequired to conduct analyses to ensure that theyconsider the possibility of setting an ambitiousreasonable progress goal, one that is aimed atreaching natural background conditions in 60years. The rule requires states to establish goalsfor each affected Class I area to 1) improvevisibility on the haziest 20 percent of days and2) ensure no degradation occurs on the clearest20 percent of days over the period of eachimplementation plan.

The states are to analyze and determine the rateof progress needed for the implementationperiod extending to 2018 such that, if main-tained, this rate would attain natural visibilityconditions by the year 2064. To calculate thisrate of progress, the state must comparebaseline visibility conditions to estimate naturalvisibility conditions in Class I areas and deter-mine the uniform rate of visibility improvementthat would need to be maintained during eachimplementation period in order to attain naturalvisibility conditions by 2064. Baseline visibil-ity conditions will be determined from datacollected from a national network of visibilitymonitors representing all Class I areas in the

country for the years 2000 to 2004. The statemust determine whether this rate and associatedemission reduction strategies are reasonablebased on several statutory factors. If the statefinds that this rate is not reasonable, it mustprovide a demonstration supporting an alterna-tive rate.

Regional VisibilityProtection Planning

Regional haze is, by definition, from wide-spread, diverse sources. The regional haze ruleencourages states to work together to improvevisibility. The Environmental ProtectionAgency (EPA) has encouraged the 48 contigu-ous states to engage in regional planning tocoordinate development of strategies for con-trolling pollutant emissions across a multi-stateregion. This means that groups of states will beaddressing groups of “Class I” areas throughestablished organizations. In the West, theWestern Regional Air Partnership, sponsoredthrough the Western Governors’ Association andthe National Tribal Environmental Council iscoordinating regional planning and neededtechnical assessments. In the Eastern U.S., fourformal groups address regional planning issues:CENRAP (Central States Response Air Partner-ship), OTC (Ozone Transport Commission), andVISTAS (Visibility Improvement State andTribal Association of the Southeast) and theMidwest Regional Planning Organization(figure 4.1.4).

Natural Visibility

Air quality regulations often distinguish be-tween human-caused and natural sources of airpollution. Natural sources of air pollutiongenerally are not responsive to control efforts,

– 71 –


Figure 4.1.4. Regional air quality planning groups.

and state air regulatory agencies manage andmonitor them in a manner different from hu-man-caused air pollution. The definition ofnatural sources of air pollution includes volca-noes, dust, and wildfires. The regional hazeregulations propose to measure progress to-wards achieving natural visibility conditions,but how do we define natural visibility impair-ment when considering wildland fires as asource?

In most parts of the country, much less fireoccurs today than historically. Should naturalvisibility consider the contribution to haze fromthese historic, natural fires? And if so, how willwe reconcile a definition of natural visibility

that includes historic levels of smoke with theneed to improve air quality and meet the na-tional visibility goal? Previously, wildfires havebeen considered natural sources while pre-scribed fires have generally been classified ashuman-caused for the purpose of air regulation.That classification is proving to be unsatisfac-tory because aggressive wildfire suppressionand land use changes have made the currentpattern of wildfires anything but natural. Aresome prescribed fires destined to be categorizedas natural emission sources along with theresulting visibility impairment, and how muchprescribed burning should be considerednatural?


– 72 –

How Much Smoke is Natural?

Few wildlands in the United States are withoutsignificant modification by humans, whether byresource utilization, fire suppression, or invasionof exotic species. So in defining natural emis-sions some possible definitions of natural firemay include: 1) historic fire frequency invegetation types present on wildlands today, 2)historic fire frequency only on wildlands wherethe current overriding management goal is tomaintain natural ecosystem processes, 3) hu-man-defined fire needed on wildlands to main-tain natural ecosystem processes, 4) human-defined fire needed to maximize wildfire con-trollability, and 5) prescribed fire needed tominimize the sum of prescribed fire and wildfireemissions.6

Most any approach to estimating natural emis-sions from fire will look to historic fire frequen-cies for preliminary guidance. Historic firefrequency can be defined in numerous ways andcalled by various terms (fire frequency, firereturn interval, natural fire rotation, ecologicalfire rotation). Fire frequency can vary greatlyby vegetative cover type, site-specific meteorol-ogy, stand age, aspect, and elevation. Firefrequency is often defined as a range that re-flects site variation. For example, a given areaof ponderosa pine ecosystem may have a de-fined fire rotation of 7 to 15 years. The driersouthwestern slopes will have an average firerotation of approximately 7 years, whereas thenorthern slopes will have an average fire rota-tion of approximately 15 years. Even within theaverage site fire rotation interval there can besignificant temporal variation depending onweather and ignition potential.

Any change in fire frequency will eventually beexpressed by change in the ecosystem. The

natural fire regime for an ecosystem may not bethe same as the historic fire regime, becauseneither the current fuel condition nor the climateis the same as in the past. Nor will they be thesame in the future.

Wildland fire is highly variable in place andtime. Historic fire regimes are well known anddescribed for most major ecosystem types.These historic frequencies can be used as astarting point for definition of natural emissionsalthough, in many parts of the country,historicfire frequency would likely result in much moreemissions than would be acceptable in today’ssociety (figure 4.1.5). Prescribed burning in thesoutheastern US is, in some cases, near thenatural rotation and the public has been largelytolerant of the smoke. Burning to maintainnatural ecosystem conditions may not need tooccur any more frequently than the middle toupper end of the historical average fire fre-quency. Some areas may be maintained ad-equately even if the infrequent end of thenatural fire frequency range is increased al-though potential long-term effects of this sort ofecological manipulation are uncertain. On theother hand, the environment is not static. Cli-mate change, for example, may change thefrequency of fire necessary to maintain anygiven ecosystem in the future or make retentionof the present ecosystem impossible.

Conclusions

Because smoke from fire can cause negativeeffects to public health and welfare, air qualityprotection regulations must be understood andfollowed by responsible fire managers. Like-wise, air quality regulators need an understand-ing of how and when fire use decisions are

___________________________________

6 Peterson, Janice; Sandberg, David, Leenhouts, Bill. 1998. Estimating natural emissions from wildland andprescribed fire. An unpublished technical support document to the EPA Interim Air Quality Policy on Wildland and Pre-scribed Fires. April 23, 1998. (Available from the author).

– 73 –


Figure 4.1.5. Estimates of the range of annual area burned in the conterminous United States pre-European settlement (Historic), applying presettlement fire frequencies to present land cover types(Expected), and burning (wildland and agriculture) that has occurred during the recent past (Current).Source: Leenhouts (1998).

Table 4.1.2. Recommended cooperation between wildland fire managers and air quality regulators dependingon air quality protection instrument.


– 74 –

made and should become involved in fire andsmoke management planning processes, in-cluding the assessment of when and howalternatives to fire will be used. Many fire andair quality issues need further work including,definition of de minimis emission levels fromfire, prescribed fire as BACM for wildfire,clarification of the difference between visibilityimpairment from fire vs. industrial sources,amounts of smoke from natural ecosystemburning that is acceptable to the public, anddefinition of natural visibility. Cooperationand collaboration between wildland fire man-agers and air quality regulators on these andother issues is of great importance. Table 4.1.2contains recommendations for various types ofcooperation by these two groups depending onthe applicable air quality protection instrument.

Literature Citations40 CFR Part 51. Vol. 64 No. 126. Regional Haze

Regulations – Final Rule. July 1, 1999.

Dockery DW, Pope CA III, Xu X, Spengler JD,Ware JH, Fay ME, Ferris BG Jr, and SpeizerFE. 1993. An association between air pollutionand mortality in six U.S. cities. New EnglandJournal of Medicine. 329:1753-1759.

Environmental Protection Agency. 1996. Reviewof the national ambient air quality standardsfor particulate matter: policy assessment ofscientific and technical information.EPA-452 \ R-96-013. Office of Air QualityPlanning and Standards, U.S. EnvironmentalProtection Agency. Research Triangle Park,NC.

Environmental Protection Agency. 1998. Interim airquality policy on wildland and prescribed fires.Office of Air Quality Planning and Standards,Research Triangle Park, NC 27711. 39p.

Leenhouts, Bill. 1998. Assessment of biomassburning in the conterminous United States.Conservation Ecology [online] 2(1). Availablefrom the Internet. URL: http://ns2.resalliance.org/pub/www/journal/vol2/iss1/art1.

Sisler, James F., William C. Malm, Kristi A.Gebhart. 1996. Spatial and seasonal patternsand long term variability of the composition ofthe haze in the United States: An analysis ofdata from the IMPROVE network. CooperativeInstitute for Research in the Atmosphere,Colorado State University. ISSN: 0737-5352-32.

Trijonis, J., R. Charlson, R. Husar, W. C. Malm, M.Pitchford, W. White. 1991. Visibility: existingand historical conditions - causes and effects.In: Acid Deposition: State of Science andTechnology. Report 24. National Acid Precipi-tation Assessment Program.

Ward, Darold E.; Hardy, Colin C. 1991. Smokeemissions from wildland fires. EnvironmentalInternational, Vol 17:117-134.

Williamson, Samuel J. 1973. Fundamentals of airpollution. Reading, Massachusetts: Addison-Wesley Publishing Co. 472 p.

U.S. Laws, Statutes, etc.; Public Law 95-95. CleanAir Act as Amended August 1977. 42 U.S.C.1857 et seq.

U.S. Laws, Statutes, etc.; Public Law 101-549. CleanAir Act as Amended Nov. 1990. 104 Stat. 2399.

– 75 –

2001 Smoke Management Guide 4.2 – State Programs

State Smoke Management Programs

John E. Core

Introduction

Smoke management programs establish a basicframework of procedures and requirements formanaging smoke from prescribed fires. Thepurposes of a smoke management program areto minimize smoke entering populated areas,prevent public safety hazards (such as smokeimpairment on roadways or runways), avoidsignificant deterioration of air quality andNational Ambient Air Quality Standards(NAAQS) violations, and to avoid visibilityimpacts in Class I areas. Smoke management isincreasingly recognized as a critical componentof a state’s air quality program for protectingpublic health and welfare, while still providingfor necessary wildland burning. Sophisticatedprograms for coordination of burning bothwithin a state and across state boundaries arevital to obtain and continue public support ofburning programs. States typically developthese programs, with cooperation and participa-tion from stakeholders. Smoke managementprograms developed through partnerships aremuch more effective at meeting resource man-agement goals, protecting public health, andmeeting air quality objectives.

Usually, either the state or tribal natural re-sources agency or air quality agency is respon-sible for developing and administering thesmoke management program. Occasionally, aprogram may be administered by a local agency

and apply to a subset of a state. Generally theadministering agency will give daily approval ordisapproval of individual bums. All burningmay be subject to permit, or only burningexceeding an established de minimis level thatcould be based on projections of acres burned,tons consumed, or emissions. Multi-day burnsmay be subject to daily reassessment andreapproval to ensure smoke does not violateprogram goals.

An advanced smoke management program willevaluate individual and multiple bums; coordi-nate all prescribed fire activities in an area;consider cross-boundary impacts; and weighburning decisions against possible health,visibility, and nuisance effects.

With increasing use of fire for forest health andecosystem management, interstate and interre-gional coordination of burning will be necessaryto prevent poor air quality episodes. Every statehas unique needs and issues driving develop-ment of smoke management programs so aspecific program cannot be defined that isapplicable to all. State and land manager devel-opment of, and participation in, an effective,locally specific smoke management programwill go a long way to build and maintain publicacceptance of prescribed burning.


– 76 –

EPA Interim Fire Policy -Recommendations onSmoke Management Programs

In the Interim Air Quality Policy on Wildlandand Prescribed Fires (EPA 1998), EPA urgesState and tribal air quality managers to collabo-rate with wildland owners and managers tomitigate the air quality impacts that could becaused by the increase of fires managed toachieve resource benefits. The EPA especiallyurges development and implementation of atleast basic smoke management programs whenconditions indicate that fires will adverselyimpact the public. In exchange for states andtribes proactively implementing smoke manage-ment programs, EPA intends to exercise itsdiscretion not to redesignate an area asnonattainment if the evidence is convincing thatfires managed for resource benefits caused orsignificantly contributed to violations of thedaily or annual PM2.5 or PM10 standards.Rather, EPA will call on the state or tribe toreview the adequacy of the smoke managementprogram in collaboration with wildland ownersand managers and make appropriate improve-ments to mitigate future air quality impacts. Thestate or tribe must certify in a letter to the EPAAdministrator that at least a basic program hasbeen adopted and implemented in order toreceive special consideration for NAAQS viola-tions under this policy.

To be certifiable by EPA, a smoke managementprogram should include the following basiccomponents, some of which are the responsibil-ity of the administering agency and some ofwhich are provided by the land manager:

1. Process for assessing and authorizing burns.

Reporting of burn plan information to admin-istering agency (not mandatory for states tobe compliant with EPA recommendations fora certified smoke management program, butis highly recommended especially for fires

greater than a predefined de minimis size),including the following information:

• location and description of the area to beburned,

• personnel responsible for managing thefire,

• type of vegetation to be burned,

• area (acres) to be burned,

• amount of fuel to be consumed (tons/acre),

• fire prescription including smoke man-agement components,

• criteria the fire manager will use formaking burn/no burn decisions, and

• safety and contingency plans addressingsmoke intrusions.

2. Plan for long-term minimization of emis-sions and impacts, including promotion ofalternatives to burning and use of emissionreduction techniques.

3. Smoke management goals and procedures tobe described in burn plans (when burn planreporting is required):

• actions to minimize fire emissions,

• smoke dispersion evaluation,

• public notification and exposure reductionprocedures to be implemented during airpollution episodes or smoke emergencies,and

• air quality monitoring.

4. Public education and awareness.

5. Surveillance and enforcement of smokemanagement program compliance.

6. Program evaluation and plan for periodicreview.

– 77 –


7. Optional programs (for example, specialprotection zones or buffers or performancestandards).

Smoke Management Programs

Prescribed burning programs across the nationuse both emission reduction methods and smokemanagement techniques (avoidance and dilu-tion) to minimize the impacts of smoke on airquality as well as concerns about public expo-sure to smoke. The complexity of these pro-grams varies greatly from state to state, rangingfrom the comprehensive and well-funded pro-grams found in Oregon and Washington to thefar simpler program found in Alaska. While thecomprehensive programs gather detailed infor-mation on all burning activity needed for burncoordination, emission inventory calculationpurposes, and to assure compliance with airquality regulations, many prescribed fire practi-tioners work independently with mainly self-imposed constraints. In most cases, smokemanagement programs focus primarily onachieving land management objectives. Otherissues in priority order are: minimizing publicexposure to smoke, achieving and/or maintain-ing healthful air quality, and achieving emissionreductions. Often, emission reductions are onlyan important side benefit of a burning techniqueselected for another management purpose. Fewexisting smoke management programs quantifyemission reductions achieved either intention-ally or unintentionally. Table 4.2.1 summarizesa few of the features of the smoke managementprograms. Significantly, only Oregon andWashington have active, on-going programs tocalculate pollutant emissions and pollutantemission reductions on a daily basis for eachburn. The Utah program has been certifiedunder the EPA Interim Air Quality Policy onWildland and Prescribed Fire; Nevada andFlorida have incorporated the Policy into the

design of their programs. Oregon and Wash-ington have adopted special provisions forprescribed burning for forest health restorationpurposes. The Oregon program includes anemissions cap and offset program for EasternOregon burning. Although most state air agen-cies estimate annual emissions from land man-ager records, only those states that calculateemissions on a daily basis, burn-by-burn, arelisted as having an emissions calculation pro-gram. The adequacy of each program to thespecific state situation is not addressed in table4.2.1. That issue is best addressed by thestakeholders of each program and the citizens ofthe state.

A summary of smoke management programreporting attributes related to emissions trackingis shown in table 4.2.2.

As an example, in the Colorado program, fieldpersonnel collect pre-burn acreage, predominatefuel type and fuel loading information annuallybefore the burning season begins. A generalizedemissions estimate is reported on the SASEMoutput they submit with their permit application(see Chapter 9 for information on SASEM andother models). Post-burn information includingacreage actually burned, fuel types, fuel loading,and fuel consumption is collected in the field atthe end of the season. If the project is classifiedas “High Risk for Smoke Impacts,” the centraloffice Program Coordinator compiles the end-of-year acreage actually burned and fuel actu-ally consumed from all cooperating agencies.The program office then uses this information tocalculate annual emissions. The program officehas no responsibilities related to fuel type data.The Colorado smoke management program isfairly basic compared to some more complexprograms, but is appropriate to the specifics ofthe state burning programs and their potentialimpacts to air quality.


– 78 –

Table 4.2.1. Smoke Management Program features. Smoke Management programs are periodically reviewedand revised; the features listed here reflect program status in 2001.

– 79 –


Tabl

e 4.

2.2.

Sel

ecte

d S

mok

e M

anag

emen

t Pro

g ram

Em

issi

on In

vent

ory

Rep

ortin

g A

ttrib

utes

.


– 80 –

Literature CitationsEnvironmental Protection Agency. 1998. Interim air

quality policy on wildland and prescribed fires.Office of Air Quality Planning and Standards,Research Triangle Park, NC 27711. 39p.

– 81 –

2001 Smoke Management Guide 4.3 – Federal Requirements

Federal Land Management–Special Requirements

Janice L. Peterson

Federal agencies are subject to certain laws andrequirements that are not necessarily applicableto states or private entities in the same manneror at all. Federal agencies are required to dolong-range planning for management of thelands they manage through numerous agency-specific planning mandates. The NationalEnvironmental Policy Act (NEPA) requiresFederal agencies to examine and disclosepotential impacts of their actions on the environ-ment. The General Conformity regulationsrequire federal agencies to examine the effect oftheir actions on the ability of a state to reach airquality goals and modify their actions if airquality targets would be delayed. Federalagencies also manage wilderness areas and theWilderness Act contains language with implica-tions for air quality protection.

Land Management Planning

Each Federal land management agency hassome sort of overarching planning mandate.These broad scale, long-term plans define howFederal lands will be managed for many yearsinto the future. For the USDA Forest Service,the National Forest Management Act (NFMA)(Public Law 94-588) requires National Foreststo prepare plans for land management thataddress a long-term planning perspective andprovide the opportunity for other agencies and

the public to comment on decisions on howthese public lands are managed. Forest Plansare to address protection, management, im-provement, and use of renewable resources onthe National Forests and should “recognize thefundamental need to protect and, where appro-priate, improve the quality of soil, water, and airresources.” Forest Plans must be updated andrevised at least every 15 years and many Na-tional Forests are in the process of, or haverecently completed this task. Other federalagencies have similar land management plan-ning mandates. For the U.S. Department of theInterior, the Bureau of Land Management hasthe Integrated Resource Management Plan; theNational Park Service has the Resource Man-agement Plan; and the Fish and Wildlife Servicehas the Comprehensive Conservation Plan.

In some parts of the country, resource manage-ment agencies have fairly recently recognizedthe importance of fire as an ecological processin the maintenance of sustainable ecosystems.Therefore, existing federal land managementplans do not always adequately address thistopic. Planning revisions provide the opportu-nity to define and resolve issues that involvewildland fire, its relationship to forest health,and its environmental costs and benefits. Revi-sions should address the fact that smoke knowsno boundaries and alternative managementscenarios must be analyzed in this same context.


– 82 –

A Forest Service Example

Forest Plans provide the long-term, big pictureview of goals for management of a NationalForest. Specific projects are planned at a laterdate to fit the goals and framework of the ForestPlan and to meet more short term planninghorizons. For example, the philosophy of howfire will be used to manage various ecosystemson a National Forest and the general effects ofthis fire on air quality will be described in theForest Plan whereas specific prescribed fireprojects and specific air quality effects will bedefined at a later date. The environmentalconsequences of specific projects are analyzedthrough the National Environmental Policy Act(NEPA) planning process.

Recent Forest Service internal guidance1 advisesthat air quality status within 100km of the Forestboundary be assessed for attainment/non-attainment status, Class I or Class II, availabilityof monitoring data, and identification of specialsmoke sensitive areas (such as airports, hospi-tals, etc.). The complexity of the subsequentForest Plan air quality analysis will be deter-mined by what is found in this initial assessmentand can range from preparation of a simpleemissions inventory and development of stan-dards and guidelines for smoke management ifthe complexity is low; up to a detailed emissionsinventory, standards and guidelines for smokemanagement including visibility protection,modeling to estimate mitigation benefits and/orconsequences, worst case emissions analysis,and identification of possible emissions offsetsif complexity is high.

National Environmental Policy Act

The National Environmental Policy Act (NEPA)(Public Law 91-190) directs all federal agenciesto consider every significant aspect of theenvironmental impacts of a proposed action. Italso ensures that an agency will inform thepublic that it has considered environmentalconcerns in its decision-making process. NEPAdoes not require agencies to elevate environ-mental concerns over other appropriate consid-erations; only that agencies fully analyze,understand, and disclose environmental conse-quences before deciding to take an action.NEPA is a procedural mandate to federal agen-cies to ensure a fully informed decision whereshort- and long-term environmental conse-quences are not forgotten.

An analysis of possible air quality impacts maybe needed in a NEPA analysis if the project:

• raised air quality as a significant issue inscoping2,

• includes burning,

• includes significant road construction,road use, or other soil disturbing proce-dures where fugitive dust may be a con-cern,

• includes significant machinery operationin close proximity to publicly accessibleareas,

• may have any impact on air quality in aClass I area,

• may have any impact on sensitive vistas orvisibility in a Class I area,

___________________________________

1 USDA Forest Service. 1999. Draft desk guide for integrating air quality and fire management into land manage-ment planning. USDA Forest Service guidance document. Available at: http://www.fs.fed.us/clean/air/

2 Scoping is the process of determining the issues to be included in NEPA analysis and for identifying any signifi-cant issues that will need to be addressed in depth. Scoping requires the lead agency to invite participation of affectedFederal, State, and local agencies, any affected Indian tribe, the proponent of the action, and other interested persons(including those who might not be in accord with the action on environmental grounds).

– 83 –


• is in close proximity to a non-attainmentarea,

• will make a significant amount of firewoodavailable to the public.

The appropriate level of analysis for eachproject will vary with the size of the project.For example, a small project will likely have abrief analysis and a large project will require adetailed analysis. The complexity and potentialeffects of the project will determine whether anenvironmental impact statement (EIS), anenvironmental assessment (EA), a biologicalevaluation (BE), or a categorical exclusion (CE)is the appropriate NEPA tool. If an air qualityanalysis is deemed unnecessary, the NEPAdocument should state that potential air qualityimpacts were considered but were determined tobe inconsequential. In this case, a justificationfor this determination must be included.

A project NEPA analysis is where specificenvironmental effects from specific projects areanalyzed and assessed. This process provides agood opportunity for fire managers and airquality regulators to come to a common under-standing of how smoke from prescribed fireprojects will be managed and reduced. Section309 of the 1977 Clean Air Act Amendments(Public Law 95-95) gives EPA a role in review-ing NEPA documents and making those reviewspublic. How actively EPA pursues this roletends to vary between EPA regions and with thecomplexity and potential environmental riskfrom the project.

A complete disclosure of air quality impacts in aNEPA document should include the followinginformation:

1. Description of the air quality environmentof the project area

2. Description of alternative fuel treatmentsconsidered and reasons why they were notselected over prescribed fire.

3. Quantification of the fuels to be burned(areas, tons, types).

4. Description of the types of burningplanned (broadcast, piles, understory, etc.).

5. Description of measures taken to reduceemissions and emission impacts.

6. Estimation of the amount and timing ofemissions to be released.

7. Description of the regulatory and permitrequirements for burning (for example,smoke management permits).

8. Modeled estimates of where smoke couldgo under certain common and worst casemeteorological scenarios and focusing onnew or increased impacts on down windcommunities, visibility impacts in Class Iwildernesses, etc. In some areas and forsome fuel types, an appropriate dispersionmodel is not available. In this situation,qualitative analysis will need to suffice.Qualitative analysis can also be used forsimple projects with little risk of airquality impact.

Conformity

“No department, agency, or instrumen-tality of the Federal Government shallengage in, support in any way or providefinancial assistance for, license orpermit, or approve, any activity whichdoes not conform to a State Implementa-tion Plan.”

Clean Air Act Amendments of 1990


– 84 –

The 1990 Clean Air Act Amendments (PublicLaw 101-549) require planned federal actions toconform to state or tribal implementation plans(SIPs/TIPs). EPA’s General Conformity ruleestablished specific criteria and procedures fordetermining the conformity of planned federalprojects and activities. In so doing, EPA choseto apply general conformity directly to non-attainment and maintenance areas only. EPAcontinues to consider application of generalconformity rules to attainment areas but atpresent has not done so, although an activity inan attainment area that causes indirect emissionincreases within a non-attainment area may haveto be analyzed for conformity. Federal agencieshave the responsibility for making conformitydeterminations for their own actions.

General conformity rules prohibit federal agen-cies from taking any action within a non-attain-ment or maintenance area that causes orcontributes to a new violation of air qualitystandards, increases the frequency or severity ofan existing violation, or delays the timelyattainment of a standard as defined in the appli-cable SIP or area plan. If a proposed federalproject (non-temporary) were projected tocontribute pollution to a non-attainment area the

project would likely be canceled or severelymodified. Temporary proposed federal projectsthat could impact a non-attainment area mustalso pass a conformity determination.

Federal activities must not:

1. Cause or contribute to new violations ofany standard.

2. Increase the frequency or severity of anyexisting violations.

3. Interfere with timely attainment or mainte-nance of any standard.

4. Delay emission reduction milestones.

5. Contradict SIP requirements.

A conformity determination is required for eachpollutant where the total of direct and indirectemissions caused by an agency’s actions wouldequal or exceed conformity de minimis levels(table 4.3.1), or are regionally significant.Regionally significant is defined as emissionsrepresenting 10 percent or more of the totalemissions for the area.

Table 4.3.1. Particle and carbon monoxide de minimis levels for general conformity.

– 85 –


The general conformity rule covers direct andindirect emissions of criteria pollutants or theirprecursors that are caused by a Federal action,reasonably foreseeable, and can practicably becontrolled by the Federal agency through itscontinuing program responsibility. In general, aconformity analysis is not required for wildlandfire emissions at the Forest Plan level becausespecifics of prescribed fire timing and locationsare not known, so at this planning level thereasonably foreseeable trigger is not met. Aconformity determination will likely be requiredat a later date when planning specific projectsunder NEPA.

Wilderness Act

The Wilderness Act (Public Law 88-157) (andsubsequent Acts designating individual Wilder-ness Areas) was enacted to preserve and protectwilderness resources in their natural condition.Wildernesses are to be administered for “the useand enjoyment of the American people in suchmanner as will leave them unimpaired for futureuse and enjoyment as wilderness, and so as toprovide for the protection of these areas, thepreservation of their wilderness character, andfor the gathering and dissemination of informa-tion regarding their use and enjoyment aswilderness…” Although air quality is notdirectly mentioned in the Wilderness Act, the

Act requires wilderness managers to minimizethe effects of human use or influence on naturalecological processes and preserve “untram-meled” the earth and its community of life.Federal agencies have interpreted the goals ofthe Wilderness Act to mean that wildernesscharacter and ecosystem health should not beimpacted by unnatural, human-caused airpollution. Most Class I areas are entirely wil-derness although some Class I National Parkscontain areas that are not wilderness.

Literature CitationsU.S. Laws, Statutes, etc.; Public Law 88-157.

Wilderness Act of Sept. 3, 1964. 78 Stat. 890;16 U.S.C. 1131-1136.

U.S. Laws, Statutes, etc.; Public Law 91-190. [S.1075], National Environmental Policy Act of1969. Act of Jan. 1, 1970. In its: United Statesstatutes at large, 1969. 42 U.S.C. sec. 4231, etseq. (1970). Washington, DC: U.S. GovernmentPrinting Office: 852-856. Vol. 83.

U.S. Laws, Statutes, etc.; Public Law 94-588.National Forest Management Act of 1976. Actof Oct. 22, 1976. 16 U.S.C. 1600 (1976).

U.S. Laws, Statutes, etc.; Public Law 95-95. CleanAir Act as Amended August 1977. 42 U.S.C.1857 et seq.

U.S. Laws, Statutes, etc.; Public Law 101-549. CleanAir Act as Amended Nov. 1990. 104 Stat. 2399.


– 86 –

– 87 –

2001 Smoke Management Guide 5.0 – Source Characteristics

SMOKE SOURCE CHARACTERISTICS

Chapter 5

Chapter 5 – Smoke Source Characteristics 2001 Smoke Management Guide

– 88 –

– 89 –


Smoke Source Characteristics

Roger D. Ottmar

Whether you are concerned with particulatematter, carbon monoxide, carbon dioxide, orhydrocarbons, all smoke components fromwildland fires are generated from the incompletecombustion of fuel. The amount of smokeproduced can be derived from knowledge ofarea burned, fuel loading (tons/acre), fuelconsumption (tons/acre), and pollutant-specificemission factors. Multiplying a pollutant-specific emission factor (lbs/ton) by the fuelconsumed, and adding the time variable to theemission production and fuel consumptionequations results in emission and heat releaserates that allow the use of smoke dispersionmodels (figure 5.1). This section discusses thecharacteristics of emissions from wildland fire

and the necessary inputs to obtain sourcestrength and heat release rate for assessingsmoke impacts.

Prefire Fuel Characteristics

Fuel consumption and smoke production areinfluenced by preburn fuel loading categoriessuch as grasses, shrubs, woody fuels, litter,moss, duff, and live vegetation; condition of thefuel (live, dead, sound, rotten); fuel moisture;arrangement; and continuity. These characteris-tics can vary widely across fuelbed types (figure5.2) and within the same fuelbed type (figure

Figure 5.1. Combustion and emission processes.


– 90 –

Figure 5.2. The preburn fuel loading (downed, dead woody, grasses, shrubs, litter, moss, and duff) canvary widely between fuel types as shown in (A) midwest grassland, 2.5 tons/acre; (B) longleaf pine, 4tons/acre; (C) southwest sage shrubland, 6 tons/acre; (D) California chaparral, 40 tons/acre; (E) westernmixed conifer with mortality, 67 tons/acre; and (F) Alaska black spruce, 135 tons/acre.

– 91 –


5.3). For instance, fuel loadings range consider-ably: less than 3 tons/acre for perennial grassesin the Midwest with no rotten material orduff (Ottmar and Vihnanek 1999); 4 tons peracre of mostly grass and a shallow litter and dufflayers for a southern pine stand treated regularlywith fire (Ottmar and Vihnanek 2000b); 6 tons/acre in a Great Basin sage shrubland (Ottmarand others 2000a); 40 tons per acre in a matureCalifornia chaparral shrubland (Ottmar andothers 2000a); 67 tons per acre of 80 percent ofwhich is rotten woody fuels, stump, snags, anddeep duff in a multi-story, ponderosa pine andDouglas-fir forest with high mortality fromdisease and insects (Ottmar and others 1998); to167 tons/acre in a black spruce forest in Alaskawith a deep moss and duff layer (Ottmar andVihnanek 1998). The heaviest fuel loadings

encountered are normally associated withmaterial left following logging, unhealthyforests, mature brush and tall grasses, or deeplayers of duff, moss or organic (muck) soils.The large variation in potential fuel loading cancontribute up to 80 percent of the error associ-ated with estimating emissions (Peterson 1987,Peterson and Sandberg 1988).

Higher fuel loading generally equates to morefuel consumption and emissions if the combus-tion parameters remain constant. For example, afrequently burned southern or western pinestand may have a fuel loading of 12 tons peracre while a recently harvested pine stand withlogging slash left on the ground may have a fuelloading of 50 tons per acre. Prescribed burningunder a moderately dry fuel moisture situation

Figure 5.3. Variability of fuel loading across several fuelbed types. Sources are referenced in the text.


– 92 –

would achieve 50 percent biomass consumptionequating to 3 tons per acre consumed in theunlogged pine stand and 25 tons/acre consumedin the logged stand.

There are several techniques available fordetermining fuel loading (U.S. Department ofInterior 1992). Collecting and weighing the fuelis the most accurate method but is impracticalfor many fuel types except grasses and smallshrubs. Measuring some biomass parameter andestimating the biomass using a pre-derivedequation is less accurate but also less timeconsuming (Brown 1974). Ongoing develop-ment of several techniques including the naturalfuels photo series (Ottmar and others 1998,Ottmar and Vihnanek 2000a) and the FuelCharacteristic Class system (FCC) (Sandbergand others 2001) will provide managers newtools to better estimate fuel loadings and reducethe uncertainty that currently exist with assign-ing fuel characteristics across a landscape. Thephoto series is a sequence of single and stereophotographs with accompanying fuel character-istics. Over 26 volumes are available for log-ging and thinning slash and natural fuels inforested, shrubland, and grassland fuelbed typesthroughout the United States. The Fuel Charac-teristic Class System is a national system beingdesigned for classifying wildland fuelbedsaccording to a set of inherent properties toprovide the best possible fuels estimates andprobable fire parameters based on available site-specific information.

Fuel moisture content is one of the most influen-tial factors in the combustion and consumptionprocesses. Live fuel moisture content can varyby temperature, relative humidity, rainfall, soilmoisture, seasonality and species. Dead fuelmoisture content varies by temperature, relativehumidity, rainfall, species, material size, anddecay class. Fuel moisture content affects theflame temperature that in turn influences the

ease of ignition, the amount and rate of con-sumption, and the combustion efficiency (theratio of energy produced compared to energysupplied). In other words, higher fuel moisturecontent requires more energy to drive off thewater, enabling fuel to reach a point wherepyrolysis can begin. Generally, fuels with lowfuel moisture content burn more efficiently andproduce fewer emissions per unit of fuel con-sumed. On the other hand, even though emis-sions per unit of fuel burned will be greater athigher fuel moistures because of a less efficientcombustion environment, total emissions maybe less if some fraction of the fuels do nottotally burn—typically the large wood fuels andforest floor.

Since combustion generally takes place at thefuel/atmosphere interface, the time necessary toignite and consume an individual fuel particlewith a given fuel moisture content depends uponthe smallest dimension of the particle. Thesurface area to volume ratio of a particle is oftenused to depict a particle’s size—the greater theratio, the smaller the particle. Small twigs andbranches have a much larger surface to volumeratio than large logs and thus a much greaterfuel surface exposed to the atmosphere. Conse-quently, fine fuels will have a greater probabilityof igniting and consuming for a given fuelmoisture.

The arrangement of the particles is also impor-tant. The structuring of fuel particles and airspaces within a fuel bed can either enhance orretard fuel consumption and affect combustionefficiency. The packing ratio (the fraction of thefuel bed volume, occupied by fuel) is the mea-sure of the fuel bed porosity. A loosely packedfuel bed (low packing ratio) will allow plenty ofoxygen to be available for combustion, but mayresult in inefficient heat transfer between burn-ing and adjacent unburned fuel particles. Manyparticles cannot be preheated to ignition tem-

– 93 –


perature and are left unconsumed. On the otherhand, a tightly packed fuel bed (high packingratio) allows efficient heat transfer between theparticles, but may restrict oxygen availabilityand reduce consumption and combustion effi-ciency. An efficiently burning fuel bed willhave particles close enough for adequate heattransfer while at the same time large enoughspaces between particles for oxygen availability.

Fuel discontinuity—both horizontal and verti-cal—isolates portions of the fuel bed from pre-ignition heating and subsequent ignition.Sustained ignition, and combustion will notoccur when the spacing between the fuel par-ticles is too large.

Biochemical differences between species alsoplay a role in combustion. Certain species suchas hoaryleaf ceanothus (Ceanothus crassifolius),palmetto (Serenoa repens) and gallberry (Ilexglabra) contain volatile compounds that makethem more flammable than species such asCarolina azalea (Rhododendron carolinianum)under similar live moisture contents.

Fire Behavior

Fire behavior is the manner in which fire reactsto the fuels available for burning (DeBano andothers 1998) and is dependent upon the type,condition, and arrangement of smaller woodyfuels, local weather conditions, topography andin the case of prescribed fire, lighting patternand rate. Two aspects of fire behavior includefire line intensity (the amount of heat releasedper unit length of fire line) and rate of spread(activity of the fire in extending its horizontaldimensions). These aspects influence combus-tion efficiency of consuming biomass and theresultant pollutants produced from wildlandfires. During fires with rapid rates of spread andhigh intensity but relatively short duration, a

majority of the biomass consumed will besmaller woody fuels and will occur during themore efficient flaming period resulting in lesssmoke. Burning dry grass and shrublands,forestlands with high large woody and duff fuelmoisture contents, clean, dry piles, and rapidlyigniting an area with circular or strip-head fireswill produce these characteristics. In simple,uniform fuelbeds such as pine and leaf litterwith only shallow organic material beneath, abacking fire with lower rates of spreads andintensities may consume fuels very efficientlyproducing less smoke. In more complexfuelbeds, the backing flame may become moreturbulent and this combustion efficiency maylessen. During wildland fires with a range offire intensities and spread rates but long burningdurations, a large portion of the biomass con-sumed will occur during the less efficientsmoldering phase, producing more smokerelative to the fuel consumed. Smoldering firesoften occur during drought periods in areas withhigh loadings of large woody material or deepduff, moss, or organic soils. The EmissionsProduction Model (EPM) (Sandberg andPeterson 1984, Sandberg 2000) and FARSITE(Finney 2000) take into account fire behaviorand lighting pattern to estimate emission pro-duction rates.

Fuel Consumption

Fuel consumption is the amount of biomassconsumed during a fire and is another criticalcomponent required to estimate emissionsproduction from wildland fire. Fuels are con-sumed in a complex combustion process thatadds to a variety of combustion products includ-ing particulate matter, carbon dioxide, carbonmonoxide, water vapor and a variety of varioushydrocarbons. Biomass consumption varieswidely among fires and is dependent on the fueltype (e.g. grass versus woody fuels), arrange-


– 94 –

ment of the fuel (e.g. piled versus non-piledwoody debris), condition of the fuel (e.g. highfuel moisture versus low fuel moisture) and theway the fire is applied in the case of a pre-scribed fire (e.g. a helicopter or fixed wingaircraft ignited high intensity, short durationmass fire versus a slow, low intensity handignition). As with fuel characteristics, extremevariations associated with fuel consumption cancontribute errors of 30 percent or more whenemissions are estimated for wildland fires(Peterson 1987; Peterson and Sandberg 1988).

In the simplest terms, combustion of vegetativematter (cellulose) is a thermal/chemical reactionwhere by plant material is rapidly oxidizedproducing carbon dioxide, water, and heat(figure 5.4). This is the reverse of plant photo-synthesis where energy from the sun combineswith carbon dioxide and water, producingcellulose (figure 5.4).

In the real world, the burning process is muchmore complicated than this. Burning fuels is atwo-stage process of pyrolysis and combustion.Although both stages occur simultaneously,pyrolysis occurs first and is the heat-absorbingreaction that converts fuel elements such ascellulose into char, carbon dioxide, carbonmonoxide, water vapor, and highly combustiblehydrocarbon vapors and gases, and particulatematter. Combustion follows as the escaping

hydrocarbon vapors released from the surface ofthe fuels burn. Because combustion efficiencyis rarely 100 percent during wildland fires,hundreds of chemical compounds are emittedinto the atmosphere, in addition to carbondioxide and water. Pyrolsis and combustionproceed at many different rates since wildlandfuels are often very complex and non-homoge-neous (DeBano and others 1998).

It has been recognized that there are four majorphases of combustion when fuel particles areconsumed (figure 5.5) (Mobley 1976, PrescribedFire Working Team 1985). These phases are:(1) pre-ignition; (2) flaming; (3) smoldering;and (4) glowing (figure 5.4). During the pre-ignition phase, fuels ahead of the fire front areheated by radiation and convection and watervapor is driven to the surface of the fuels andexpelled into the atmosphere. As the fuel’sinternal temperature rises, cellulose and ligninbegin to decompose, releasing combustibleorganic gases and vapors (Ryan and McMahon1976). Since these gases and vapors are ex-tremely hot, they rise and mix with oxygen inthe air and ignite at temperatures between 6170 Fand 6620 F leading to the flaming phase(DeBano and others 1998).

In the flaming phase, the fuel temperature risesrapidly. Pyrolysis accelerates and is accompa-nied by flaming of the combustible gases and

CELLULOSE + O2 CO2 + H20 + ENERGY

Heat/chemical energy

CELLULOSE + O2 CO2 + H20 + ENERGY

Sun/thermal energy

Figure 5.4. The energy flow for combustion is reverse to that for photosynthesis.

– 95 –


vapors. The combustion efficiency during theflaming stage is usually relatively high as longas volatile emissions remain in the vicinity ofthe flames. The predominant products offlaming combustion are carbon dioxide (CO2)and water vapor (H2O). The water vapor is aproduct of the combustion process and alsoderives from moisture being driven from thefuel. Temperatures during the flaming stagerange between 9320 F to 25520 F (Ryan andMcMahon 1976). During the flaming period,the average exterior diameter reduction of roundwood material occurs at a rate of 1 inch per 8minutes (Anderson 1969). For example, a drylimb 3 inches in diameter would take approxi-mately 24 minutes to completely consume ifflaming combustion was sustained during theentire time period.

During the smoldering phase, emissions ofcombustible gases and vapors above the fuel istoo low to support a flaming combustion result-ing in a fire spread decrease and significant

temperature drop. Peak smoldering tempera-tures range from 572oF to 1112oF (Agee 1993).The gases and vapors condense, appearing asvisible smoke as they escape into the atmo-sphere. The smoke consists mostly of dropletsless than a micrometer in size. The amount ofparticulate emissions generated per mass of fuelconsumed during the smoldering phase is morethan double that of the flaming phase.

Smoldering combustion is more prevalent incertain fuel types (e.g. duff, organic soils, androtten logs) due to the lack of oxygen necessaryto support flaming combustion. Smolderingcombustion is often less prevalent in fuels withhigh surface area to volume ratios (e.g. grasses,shrubs, and small diameter woody fuels)(Sandberg and Dost 1990). Since the heatgenerated from a smoldering combustion isseldom sufficient to sustain a convection col-umn, the smoke stays near the ground and oftenconcentrates in nearby valley bottoms, com-pounding the impact of the fire on air quality.

Figure 5.5. The four phases of combustion.

Pre-ignition

Flaming

Smoldering

Glowing


– 96 –

Near the end of the smoldering phase, thepyrolysis process nearly ceases, leaving the fuelthat did not completely consume with a layer ofblack char, high in carbon content.

In the glowing phase, most volatile gases havebeen driven off. Oxygen in the air can nowreach the exposed surface of char left from theflaming and smoldering phase and the remain-ing fuels begin to glow with the characteristicorange color. Peak temperatures of the burningfuel during the glowing phase are similar tothose found in the smoldering phase and rangefrom 572oF to 1117oF (DeBano and others1998). There is little visible smoke. Carbondioxide, carbon monoxide, and methane are theprincipal products of glowing combustion. Thisphase continues until the temperature of the fueldrops or until only noncombustible, mineralgray ash remains.

The combustion phases occur both sequentiallyand simultaneously as a fire front moves acrossthe landscape. The efficiency of combustionthat takes place in each combustion phase is notthe same, resulting in a different set of chemicalcompounds being released at different rates intothe atmosphere. Understanding the combustionprocess of each phase will assist managers inemploying various emission reduction tech-niques. Fuel type, fuel moisture content,arrangement, and the way the fuels are ignitedin the case of prescribed fires, can affect theamount of biomass consumed during variouscombustion stages. Between 20 and 90 percentof the biomass consumed during a wildland fireoccurs during the flaming stage, with the re-mainder occurring during the smoldering andglowing stages (Ottmar and others [in prepara-tion]. The flaming stage has a high combus-tion efficiency; that is it tends to emit the leastemissions relative to the mass of fuel consumed.The smoldering stage has a low combustionefficiency and produces more smoke relative tothe mass of fuel consumed.

Biomass consumption of the woody fuels, piledslash, and duff in forested areas has becomebetter understood in recent years (Sandberg andDost 1990, Sandberg 1980, Brown and others1991, Albini and Reinhardt 1997, Reinhardt andothers 1997, Ottmar and others 1993, Ottmarand others [in preparation]). Large woody fuelconsumption generally depends on moisturecontent of the woody fuel and duff. Approxi-mately 50 percent of the consumption occursduring the flaming period. Duff consumptiondepends on fire duration of woody fuels andduff moisture content. Consumption occursprimarily during the smoldering stage when duffmoisture is low. Consumption of tree crowns inforests and shrub crowns in shrublands arepoorly understood components of biomassconsumption and research is currently underway(Ottmar and Sandberg 2000) to develop ormodify existing consumption equations forthese fuel components.

Since consumption during the flaming phase ismore efficient than during the smoldering phase,separate calculations of flaming consumptionand smoldering consumption are required forimproved assessment of total emissions. Equa-tions for predicting biomass consumption bycombustion phase are widely available in twomajor software packages including Consume 2.1(Ottmar and others [in preparation]) and FirstOrder Fire Effects Model (FOFEM 5.0)(Reinhardt and Keane 2000).

Consume 2.1 is a revision of Consume 1.0(Ottmar and others 1993) and uses a set oftheoretical models based on empirical data topredict the amount of fuel consumption from theburning of logging slash, piled woody debris, ornatural forest, shrub, grass fuels. Input variablesinclude the amount of fuel, woody fuel and duffmoisture content, and meteorological data. Thesoftware product incorporates the original FuelCharacteristic System (Ottmar and others [inpreparation]) for assigning default fuel loadings.

– 97 –


It also incorporates features that allow users toreceive credit for applying fuel consumptionreduction techniques. FOFEM 5.0 (Reinhardtand Keane 2000) is a revision of FOFEM 4.0(Reinhardt and others 1997) and relies onBURNUP. a new model of fuel consumption(Albini and Reinhardt 1997). The softwarecomputes duff and woody fuel consumption formany forest and rangeland systems of theUnited States. Both Consume 2.1 and FOFEM5.0 packages are updated on a regular basis asnew consumption models are being developed.

Smoke Emissions

The chemistry of the fuel as well as the effi-ciency of combustion governs the physical andchemical properties of the resulting smoke fromfire. Although smoke from different sourcesmay look similar to the eye, it is often quitedifferent in terms of its chemical and physicalproperties. Generally, the emissions we cannotsee are gas emissions and the emissions we cansee are particulate emissions.

Carbon dioxide and water—Two products ofcomplete combustion during fires are carbondioxide (CO2) and water (H2O) and generallymake up over 90 percent of the total emissionsfrom wildland fire. Under ideal conditionscomplete combustion of one ton of forest fuelsrequires 3.5 tons of air and yields 1.84 tons ofCO2 and 0.54 tons of water (Prescribed FireEffects Working Team 1985). Under wildlandconditions, however, inefficient combustionproduces different yields. Neither carbondioxide nor water vapor are considered airpollutants in the usual sense, even thoughcarbon dioxide is considered a greenhouse gasand the water vapor will sometimes condenseinto liquid droplets and form a visible whitesmoke near the fire. This fog/smoke mixturecan dramatically reduce visibility and createhazardous driving conditions.

As combustion efficiency decreases, less carbonis converted to CO2 and more carbon is avail-able to form other combustion products such ascarbon monoxide (CO), hydrocarbons (HC),nitrogen oxides(NOx), and sulfur oxides (SOx),all of which are considered pollutants.

Carbon Monoxide—Carbon monoxide (CO) isthe most abundant emission product fromwildland fires. Its negative effect on humanhealth depends on the duration of exposure, COconcentration, and level of physical activityduring the exposure. Generally, dilution occursrapidly enough from the source of the fire thatcarbon monoxide will not be a problem for localcitizens unless a large fire occurs and inversionconditions trap the carbon monoxide near ruralcommunities. Carbon monoxide is always aconcern for wildland firefighters however, bothon the fire line at prescribed fires and wildfires,and at fire camps (Reinhardt and Ottmar 2000,Reinhardt and others 2000).

Hydrocarbons—Hydrocarbons (HC) are anextremely diverse class of compounds contain-ing hydrogen, carbon and sometimes oxygen.Usually, the classes of hydrocarbon compoundsare identified according to the number of carbonatoms per molecule. Emission inventories oftenlump all gaseous hydrocarbons together. Al-though a majority of the HC pollutants mayhave no harmful effects, there are a few that aretoxic. More research is needed to characterizehydrocarbon production from fires.

Nitrogen Oxides—In wildland fires, smallamounts of nitrogen oxides (NOx) are produced,primarily from oxidation of the nitrogen con-tained in the fuel. Thus the highest emissions ofNox occur from fuels burning with a highnitrogen content. Most fuels contain less than 1percent nitrogen. Of that about 20 percent isconverted to NOx when burned.


– 98 –

Hydrocarbons and possibly nitrogen oxidesfrom large wildland fires contribute to increasedozone formation under certain conditions.

Particulate Matter—Particulate matter pro-duced from wildland fires limits visibility,absorbs harmful gases, and aggravates respira-tory conditions in susceptible individuals (figure5.6). Over 90 percent of the mass of particulatematter produced by wildland is less than 10

microns in diameter and over 80-90 percent isless than 2.5 microns in diameter (figure 5.7).These small particles are inhalable and respi-rable. Respirable suspended particulate matteris that proportion of the total particulate matterthat, because of its small size has an especiallylong residence time in the atmosphere andpenetrates deeply into the lungs. Small smokeparticles also scatter visible light and thusreduce visibility.

Figure 5.6. Relative sizes of beach sand, flour, and a PM2.5 particle in smoke.

– 99 –


Emission Factors

An emission factor for a particular pollutant ofinterest is defined as the mass of pollutantproduced per mass of fuel consumed (i.e., lbs/ton in the English system or g/kg as the metricequivalent). Multiplying an emission factor ingrams/kg by a factor of two will convert theemission factor to English units (pounds/ton).

Emission factors vary depending on type ofpollutant, type and arrangement of fuel andcombustion efficiency. The average fire emis-sion factors have a relatively small range andcontributes approximately 16 percent of the totalerror associated with predicting emissionsproduction (Peterson 1987; Peterson andSandberg 1988). In general, fuels consumed byflaming combustion produce less smoke thanfuels consumed by smoldering combustion.Emission factors for several smoke compounds

are presented in table 5.1 for the flaming,smoldering, and fire average for generalized fueltypes and arrangements. Emission factors canbe used by air quality agencies to calculate localand regional emissions inventories or by manag-ers to develop strategies to mitigate downwindsmoke impacts. Additional emission factorshave been determined for other fuel types andwill be available in the future.

Total Emissions, SourceStrength, and Heat Release Rate

Total emissions from a fire or class of fires (thatis, a set of fires similar enough to be character-ized by a single emission factor) can be esti-mated by multiplying that emission factor by thebiomass consumed and an accurate assessmentof the total acreage burned. For instance,assume that 10 tons/acre of fuels will be con-

Figure 5.7. Particulate matter size-class distribution from typical wildland fire smoke.


– 100 –

Table 5.1. Forest and rangeland emission factors 1Ward and others 1989; 2Hardy and others 1996;3Hardy and Einfield 1992).

– 101 –


sumed during a 200 acre landscape prescribedburn in a ponderosa pine stand. Following thefire, ground surveys and aerial reconnaissanceindicate a mosaic fire pattern and only 100 acresof the 200 acres within the fire perimeter actu-ally burned. Since the emission factor forparticulate matter 2.5 microns in diameter orless (PM2.5) for pine fuels is approximately 22lbs/ton, then total emission production wouldbe:

Managers can make better estimates of emis-sions produced from a wildland fire if theamount of fuel consumption in the flaming andsmoldering combustion period is known. Thesame general approach is used although it isslightly more complicated. The fuel consumedduring the flaming period and smoldering periodare multiplied by the appropriate flaming andsmoldering emission factor for a particular fueltype, then summed. Computer software such asConsume 2.1 (Ottmar and others [in prepara-tion]) and FOFEM 5.0 (Reinhardt and Keane2000) use this approach to improve estimates oftotal emissions produced from wildland fire ascompared with the fire average approach. Anemission inventory is the aggregate of totalemissions from all fires in a given period for aspecific geographic area and requires totalemissions.

Modeling emissions from wildland fires requiresnot only total emissions, but also sourcestrength. Source strength is the rate of airpollutant emissions in mass per unit of time orin mass per unit of time per unit of area and isthe product of the rate of biomass consumptionand an emission factor for the pollutant(s) ofinterest. Source strength can be calculated bythe equation:

Emission rates vary by fuel loading, fuel con-sumption, and emission factors. Figure 5.8graphically depicts general trend differences inemission production rate and total emissionsproduction (area under each curve) for variousprescribed fire scenarios. Mechanically treatingfuels before burning, mosaic burning, burningunder high fuel moisture contents, and burningpiles are specific ways emission rates can bereduced to meet smoke management require-ments.

The consumption of biomass produces thermalenergy and this energy creates buoyancy to liftsmoke particles and other pollutants above thefire. Heat release rate is the amount of thermalenergy generated per unit of time or per unit oftime per unit of area. Heat release rate can becalculated by the equation:

Both source strength and heat release rate arerequired by all sophisticated smoke dispersionmodels (Breyfogle and Ferguson 1996). Disper-sion models are used to assess the impact ofsmoke on the health and welfare of the public incities and rural communities and on visibility insensitive areas such as National Parks, Wilder-ness areas, highways, and airports. The Emis-sions Production Model (EPM) (Sandberg andPeterson 1984; Sandberg 2000) is the onlymodel that predicts source strength and heatrelease rate for wildland fires. The EPM soft-ware package imports fuel consumption predic-tions from Consume 2.1 or FOFEM 5.0 and usesignition pattern, ignition periods, and burn areacomponents to calculate source strength, heatrelease rate, and plume buoyancy.


– 102 –

Figure 5.8a. Emission production rate over time forPM2.5 during an underburn with and without fuelsmechanically removed.

Figure 5.8b. Emission production rate over time forPM2.5 during a mosaic burn and a burn where firecovers the entire area within the perimeter.

Figure 5.8c. Emission production rate over time forPM2.5 during an underburn with low and high fuelmoisture content.

Figure 5.8d. Emission production rate over time forPM2.5 during an underburn and a pile burn.

– 103 –


Literature CitationsAgee, James K. 1993. Fire ecology of Pacific

Northwest forests. Island Press, WashingtonD.C. 493 p.

Albini, Frank A.; Reinhardt, Elizabeth D. 1997.Improved calibration of a large fuel burnoutmodel. International Journal of Wildland Fire.7(1): 21-28.

Anderson, Hal E. 1969. Heat transfer and firespread. USDA For. Serv. Res. Pap. INT-69,Intermt. For. And Range Exp. Stn., Ogden, UT,20 p.

Breyfogle, Steve; Ferguson, Sue A. 1996. Userassessment of smoke-disperion models forwildland biomass burning. Gen. Tech. Rep.PNW-GTR-379. Portland, OR: U.S. Depart-ment of Agriculture, Forest Service, PacificNorthwest Research Station. 30 p.

Brown, James K. 1974. Handbook for inventoryingdowned woody material. Gen Tech, Rep. INT-16. Ogden, UT: U.S. Department of Agricul-ture, Forest Service, Intermountain Forest andRange Experiment Station. 24 p.

Brown, Jame K.; Reinhardt, Elizabeth D.; Fischer,Wiliam C. 1991. Predicting duff and woodyfuel consumption in northern Idaho prescribedfires. Forest Science, Vol. 37, (6): 1550-1566.

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

Finney, Mark. 2000. Personal communication.USDA Forest Service Rocky Mountain Re-search Station, Fire Laboratory, Missoula, MT.

Hardy, C.C.; Ward, D.E.; Einfield, W. 1992. PM2.5emissions from a major wildfire using a GIS;rectification of airborne measurements. In:Proceedings of the 29th annual meeting of thePacific Northwest International Section, Air andWaste Management Association; 1992 Novem-ber 11-13; Bellevue, WA. Pittsburgh, PA: Airand Waste Management Association.

Hardy, C.C.; Conard, S.G.; Regelbrugge, J.C.;Teesdale, D.T. 1996. Smoke emissions fromprescribed burning of southern Californiachaparral. Res. Pap. PNW-RP-486. Portland,

OR: U.S. Department of Agriculture, ForestService, Pacific Northwest Research Station.37p.

Mobley, Hugh E. 1976. Smoke management—Whatis it? In southern smoke management guidebook. Gen. Tech. Rep. SE-10. U.S. Departmentof Agriculture, Forest Service, Southeast Rangeand Experiment Station. pp. 1-8.

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

Ottmar, Roger D.; Vihnanek, Robert E.; Wright,Clinton S. 1998. Stereo photo series for quanti-fying natural fuels: Volume I: Mixed-coniferwith mortality, western juniper, sagebrush andgrassland types in the interior Pacific North-west. PMS 830. Boise, ID: National WildfireCoordinating Group, National Interagency FireCenter. 73 p.

Ottmar, Roger D.; Vihnanek, Robert E. 1998. Stereophoto series for quantifying natural fuels:Volume II: Black spruce and white spruce typesin Alaska . PMS 831. Boise, ID: NationalWildfire Coordinating Group, National Inter-agency Fire Center. 65 p.

Ottmar, Roger D.; Vihnanek, Robert E. 1999. Stereophoto series for quantifying natural fuels:Volume V: Midwest red and white pine, north-ern tallgrass prairie, and mixed oak types in thecentral and lake states. PMS 834. Boise, ID:National Wildfire Coordinating Group, NationalInteragency Fire Center. 99 p.

Ottmar, Roger D.; Sandberg, David V. 2000. Modifi-cation and validation of fuel consumptionmodels for shrub and forested lands in theSouthwest, Pacific Northwest, Rockies, Mid-west, Southwest, and Alaska. Abstract. Pre-sented at the Joint Fire Science ProgramPrinciple Investigators Meeting, October 3-5,2000, Reno, Nevada. http://www.nifc.gov/joint_fire_sci/jointfiresci.html

Ottmar, Roger D.; Vihnanek, Robert E. 2000a. Photoseries for major natural fuel types of the United


– 104 –

States—phase II. Abstract. Presented at theJoint Fire Science Program Principle Investiga-tors Meeting; 2000 October 3-5; Reno, Nevada.http://www.nifc.gov/joint_fire_sci/jointfiresci.html.

Ottmar, Roger D.; Vihnanek, Robert E. 2000b.Stereo photo series for quantifying naturalfuels: Volume VI: Longleaf pine, pocosin, andmarshgrass types in the southeast United States.PMS 831. Boise, ID: National Wildfire Coordi-nating Group, National Interagency Fire Center.56 p.

Ottmar, Roger D.; Vihnanek, Robert E.;Regelbrugge, Jon C. 2000a. Stereo photo seriesfor quantifying natural fuels: Volume IV:Pinyon-juniper, sagebrush, and chaparral typesin the Southwestern United States. PMS 833.Boise, ID: National Wildfire CoordinatingGroup, National Interagency Fire Center. 97 p.

Ottmar, Roger D.; Reinhardt, Timothy E.; Anderson,Gary, DeHerrera, Paul J. [In preparation].Consume 2.1 user’s guide. Portland, OR: U.S.Department of Agriculture, Forest Service,Pacific Northwest Research Station.

Peterson, Janice L.; Sandberg, David V. 1988. Anational PM10 emissions inventory approachfor wildfires and prescribed fires. In: Mathai,C.V.; Stonefield, David H., eds. TransactionsPM-10 implementation of standards: an APCA/EPA International Specialty Conference; 1988February 23-24; San Francisco, CA. Pittsburg,PA: Air Pollution Control Association: 353-371.

Peterson, Janice L. 1987. Analysis and reduction ofthe errors of predicting prescribed burn emis-sions. Thesis. Seattle: University of Washing-ton. 70 p.

Prescribed Fire and Fire Effects Working Team.1985. Smoke management guide. PMS 420-2,NFES 1279. Boise ID: National WildfireCoordinating Group, National Interagency FireCenter. 28 p.

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 ofAgriculture, Forest Service, IntermountainResearch Station. 65 p.

Reinhardt, Elizabeth D; Keane, Robert E. 2000. Anational fire effects prediction model—revisionof FOFEM. Abstract. Joint Fire ScienceProgram Principle Investigators Meeting; 2000October 3-5, 2000; Reno, Nevada. http://www.nifc.gov/joint_fire_sci/jointfiresci.html

Reinhardt, Timothy E.; Ottmar, Roger D. 2000.Smoke exposure at western wildfires. Res. Pap.PNW-RP-525. Portland, OR: U.S. Departmentof Agriculture, Forest Service, Pacific North-west Research Station. 72 p.

Reinhardt, Timothy E.; Ottmar, Roger D.;Hanneman, Andrew J.S. 2000. Smoke exposureamong firefighters at prescribed burns in thePacific Northwest. Res. Pap. PNW-RP-526.Portland, OR: U.S. Department of Agriculture,Forest Service, Pacific Northwest ResearchStation. 45 p.

Ryan, P.W.; McMahon, C.K. 1976. 1976. Somechemical characteristics of emissions fromforest fires. In: Proceedings of the 69th AnnualMeeting of the Air Pollution Control Associa-tion, Portland, OR, Air Pollution ControlAssociation, Pittsburgh, PA. Paper No. 76-2.3.

Sandberg, David V. 1980. Duff reduction by pre-scribed unferburning in Douglas-fir. USDA For.Serv. Res. Pap. PNW-272. 18 p.

Sandberg, David V. 2000. Implementation of animproved Emission Production Model. Ab-stract. Joint Fire Science Program PrincipleInvestigators Meeting; 2000 October 3-5; Reno,NV. http://www.nifc.gov/joint_fire_sci/jointfiresci.html.

Sandberg, David V.; Dost, Frank N. 1990. Effects ofprescribed fire on air quality and human health.In: Natural and Prescribed Fire in the PacificNorthwest edited by John D. Walstad, Steven RRadosevich, David V. Sandberg. Orgon StateUniversity Press, Corvallis, Oregon. 191-218.

Sandberg, David V.; Ottmar, Roger D. 2001 Charac-terizing fuels in the 21st century. InternationalJournal of Wildland Fire, 10: 381-387.

Sandberg, D.V.; Peterson, J.L. 1984. A sourcestrength model for prescribed fires in coniferouslogging slash. Presented to the 1984 AnnualMeeting, Air Pollution Control Association,

– 105 –


Pacific Northwest Section. Reprint #84.20.Portland, Oregon: U.S. Department of Agricul-ture, Forest Service, Pacific Northwest Re-search Station. 10 p.

U.S. Department of Interior. 1992. Fire monitoringhandbook. San Francisco: U.S. Department ofInterior, National Park service, Western Region.[Pages unknown].

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 fire atmo-spheric pollution through increased utilizationof hardwoods, piled residues, and long-needledconifers. Final report. U.S. DOE, EPA. Avail-able from: U.S. Department of Agriculture,Forest Service, Pacific Northwest ResearchStation, Seattle WA.


– 106 –

– 107 –

2001 Smoke Management Guide 6.0 – Fire Use Planning

FIRE USE PLANNING

Chapter 6

Chapter 6 – Fire Use Planning 2001 Smoke Management Guide

– 108 –

– 109 –


Fire Use Planning

Tom Leuschen

Dale Wade

Paula Seamon

The success of a fire use program is in large partdependent on a solid foundation set in clear andconcise planning. The planning process resultsin specific goals and measurable objectives forfire application, provides a means of settingpriorities, and establishes a mechanism forevaluating and refining the process to meet thedesired future condition. It is an ongoingprocess, beginning months or even years inadvance of actual fire use, with plans becomingincreasingly specific as the day of the burnapproaches. Although details differ between firepractitioners, the general planning process isessentially the same.

Land and ResourceManagement Planning

Fire use planning should begin as a componentof the overall land and resource managementplanning for a site. Consideration of the inten-tional use of fire to achieve stated resourcemanagement goals should be an integral part ofthis process. In deciding whether or not fire useis the best option to accomplish a given objec-tive, an analysis of potential alternative treat-ments should be completed. This analysisshould describe the risks associated with use ofa given treatment and include expected negativeas well as beneficial outcomes. Care should be

exercised to separate statements that are sup-ported by data (preferably local and ecosystem-specific), from those only purported to be true.

Many private landowners do not have writtenresource management plans, but most have avision of what natural resource attributes theywant to favor and what they want their lands tolook like. We recommend they put this visionon paper to provide guidance to themselves andtheir heirs.

The plans should identify any barriers to imple-menting a treatment judged best from a re-source management standpoint, such asregulations, cost, or insufficient resources. Ifsuch a treatment is not recommended becauseof these barriers, the probable ecological ramifi-cations of this decision should be documented.On sites where fire is selected as the bestalternative to accomplish the desired resourcemanagement objectives, the next step in fire useplanning is to develop a fire management plan.

The Fire Management Plan

The fire management plan addresses fire use atthe level of the administrative unit, such as aforest, nature preserve, park, ranch or planta-tion. It ensures that background information


– 110 –

about the area has been researched, legal con-straints reviewed, and a burn program found tobe both justified and technically feasible. Itproposes how fire will be applied to the land-scape, both spatially and temporally. Whenmanaging for multiple resources (e.g., range,wildlife, and timber) on a tract, guidance shouldbe provided regarding the allocation of benefits;i.e., should benefits to the same resource alwaysbe maximized on given burn units, or should thefocus be rotated among benefits on some, or allburn units over time?

Items commonly addressed in the fire manage-ment plan are:

• Background information on the area, suchas topography, soils, climate and fuels

• Applicable fire laws and regulations,including any legal constraints

• Landowner policy governing fire use onthis tract of land

• Fire history of the area, including thenatural fire regime, and recent fire occur-rence or use

• Justification for fire management

• Fire management goals for the area,including a description of the desiredfuture condition. (Objectives for specificburns are set in the burn unit plan, seebelow.)

• Fire management scheduling, qualitativelydescribing how fire will be applied to thesite over time to achieve stated resourceobjectives. (Quantitative descriptions offireline intensity, fire severity, and seasonof burn are set in the burn unit plan, seebelow.)

• Species of special concern, wildlife habitatissues, invasive species issues

• Definition and descriptions of treatmentunits or burning blocks

• Air quality and smoke managementconsiderations

• Neighbor and community factors

• Maps illustrating fuels distribution, treat-ment units, smoke sensitive areas, etc.

When complete, this document should enablethe resource manager to gain the support (bothinternal and external) and identify the resourcesneeded to effectively and efficiently use fire as amanagement tool.

Community involvement in the fire planningprocess is crucial to public acceptance of fireuse. At what stage to involve the public in theprocess will depend on regional issues, regula-tions, and organizational policy. In general, theearlier the public is involved, the easier it is toreach agreement on any concerns. Whenever itis done, it is important to remember that publicsupport is key to the long-term success of a firemanagement program. Unexpected results,including under-achievement and over-achieve-ment of objectives, are bound to occur. A full,honest discussion of the potential for suchresults, and their ramifications, can defusenegative reaction to the occasional bad outcome,especially if the public was involved early in theplanning process.

Further guidance for developing a fire manage-ment plan is available from a number of federalsources, including Wildland and PrescribedFire Management Policy: ImplementationProcedures Reference Guide (USDI and USDAForest Service 1998), and from The Nature

– 111 –


Conservancy’s Fire Management Manual(www.tncfire.org/manual).

The Burn Plan

Once the fire management plan is completedand approved, the next step is implementation—not an easy task. Resource managers are usu-ally faced with numerous constraints, such asbudget and staff limitations, equipment avail-ability, timing of good burning conditions, and alack of information on potential effects. Asuccessful prescribed fire program requires thecomplete dedication of the fire managementstaff, full cooperation of all personnel andfunctional areas involved, and unwaveringsupport and commitment throughout the chainof command.

Although the overall resource managementgoals for an individual burn unit often remainunchanged for long periods, the specific burnobjectives for a given unit will likely vary overtime, necessitating modifications to the unit planfor each burn. For example, the use of a head-ing fire during the growing season to promotebiodiversity and flowering of ground layerplants may be the current burn objective, while abacking fire during the dormant season mayhave been used to reduce hazardous fuels loadsthe last time the unit was burned.

A written burn plan serves several importantpurposes:

• It makes the planner think about what he/she wants to achieve, and how it will beaccomplished.

• It allows the fire manager to prioritizebetween burn units based on constraintsand objectives.

• It functions as the operational plan thatdetails how a burn will be safely andeffectively conducted.

• It serves as the standard by which toevaluate the burn.

• It provides a record for use when planningfuture burns (which makes it essential todocument any changes when the burn isconducted, directly on the plan).

• It becomes a legal record of the intendedpurpose and execution of the burn project.

There is no standard format for a burn unit plan;numerous examples are available which can beconsulted for guidance. Sources include stateand federal land management agencies, TheNature Conservancy’s internet site(www.tncfire.org), or publications such as AGuide to Prescribed Fire in Southern Forests(Wade and Lunsford, 1989), which is availableonline from the Alabama Private Forest Man-agement Team website (www.pfmt.org/standman/prescrib.htm), and from the FloridaDivision of Forestry (flame.fl-dof.com/Env/Rx/guide/).

Although formats differ, certain componentsshould be included in all burn plans. Theyshould address at least the following 12 topics:

1. Assessment and Description of the BurnUnit. The first step in developing a burnplan is to evaluate and document existingconditions. Factors to include depend on thesite itself, as well as the complexity of theplanned burn. The information recorded herewill serve as the baseline from which successof the burn will be determined, so parametersused in the burn objectives should be as-sessed and described. Include details on theunit size (broken into single-day burn units);date of the last burn; overstory and under-


– 112 –

story vegetation, density and size; fuel type,density and size; soil type and topography;threatened and endangered species present;invasive species present; and current wildlifeuse.

2. Maps. Good maps of the treatment area are akey component of the burn plan. The mapscale should be adequate to show pertinentinformation in meaningful detail. Be carefulnot to include too much information on asingle map, making it difficult to read. Theburn plan should include a series of mapsshowing the following: unit boundaries;adjacent land ownerships, including contactperson and phone numbers; topography andmanmade obstacles such as canals, ditches,and erosion gullies that would impede equip-ment or people; natural and constructed firecontrol lines; areas to be protected or ex-cluded such as sawdust piles, utility polesand sensitive vegetation areas; firing plan;initial placement of equipment and holdingpersonnel, and; escape routes and safetyzones. Every crew member should receive amap with the information essential to person-nel safety and burn operations.

3. Measurable Burning Objectives. Unit-specific treatment objectives identify thedesired changes in affected resources fromthe present to the future condition. Treat-ment objectives are prepared within thecontext and intent of all resource manage-ment objectives. They are the measuresagainst which the success of a burn is deter-mined. Burn objectives make clear to every-one involved what is expected - including theburners, cooperators, managers, and thepublic. The objectives should be detailedstatements that describe what the treatment isintended to accomplish, and as such, must bespecific and quantifiable.

4. Weather and Fuel Prescription. Theprescription defines the range of conditionsunder which a fire is ignited and allowed toburn to obtain given objectives. Fuel mois-ture (by size class) and weather conditions(temperature, humidity, wind, drought,dispersion index) are key factors in achievingobjectives because they in large part deter-mine fire behavior (intensity and severity),which in turn, governs ease of fire controland effects. These same parameters alsoaffect smoke production and transport.Considerable care should therefore be takenin defining the window of conditions underwhich the projected burn may take place.Although there may be an ideal set of condi-tions that will maximize a single objective,the likelihood of this set of conditions occur-ring at the right time is typically extremelylow. Therefore, a range of fuel and weatherconditions are usually specified in the burnprescription that allow the skilled burner tocompensate between various parameters tosafely and efficiently conduct a successfulburn—a burn which meets both the resourceand smoke management objectives.

5. Season and Time of Day. The season ofburn influences many burn parameters.Typically, acceptable burning conditions aremore predictable during certain seasons,making it easier to plan and prepare for burnsdays in advance, but not all burn objectivesmay be achievable under those weather andfuel conditions. Regional effects are impor-tant in decision-making for this factor. Forexample, in the southeast, dormant seasonburns are generally more uniform in effectswhile growing season burns are more likelyto be patchy. Backing fires are much easierto conduct during the dormant season whenground layer herbaceous plants are dead andburn readily, rather than green and succulent

– 113 –


thereby retarding fire spread. In the PacificNorthwest, season of burn can be used toreduce emissions. Broadcast burning ofslash in the wet spring has been shown toproduce 50% fewer emissions when com-pared to burning periods in the dry fall(Sandberg and Dost 1990). Selecting thecorrect season to execute a burn will helpmaximize the probability of achieving theburn objectives.

The timing of ignition determines whetherthe burn can be completed and mopped up asscheduled during the burning period. Timingis also important when considering factorssuch as: when solar radiation will break anighttime inversion or dissipate any dewwhich formed during the night, when atmo-spheric conditions will support adequatetransport and dissipation of smoke, whensurface winds may develop or change speedor direction, or when a sea breeze front mayreach the unit. Experienced burners becomefamiliar with the area, and learn how tofactor these time-sensitive influences intotheir burn plans.

6. Smoke Management. Planning a fire useproject that has the potential to impact areassensitive to smoke requires assessment ofairshed and meteorological conditions thatinfluence both the movement and concentra-tion of smoke. The expected effects of windspeed and direction, air stability, and night-time inversions should be specifically out-lined. Specific regional issues should beaddressed, such as mountainous terrain, fog,or sea breeze effects. This informationnormally will be developed by fire managersusing their personal experience and knowl-edge of fire behavior, smoke transport anddispersion in the area, along with moreformal emissions prediction and dispersionmodeling.

Sensitive areas downwind of the burn unitshould be identified and plotted on a map.Information such as distance and directionfrom the burn unit, the nature of the sensitiv-ity, and when the area is considered sensitiveshould be included. Examples of smokesensitive areas include Class I areas (gener-ally, international parks, and large nationalparks and national wilderness areas), non-attainment areas, communities or individualresidences, airports, highways, and medicalfacilities. Several procedures for predictingthe potential impact of smoke on sensitiveareas are discussed in chapter 9.

Smoke dispersion in areas prone to inver-sions, such as deep, mountainous valleys, isespecially problematic in fire use planning.If the smoke remains trapped by the inver-sion, all of the emissions produced willremain trapped within the airshed.

The following smoke-related questionsshould be addressed in every plan:

• What quantity of emissions will it take tosaturate this airshed?

• Where will the smoke concentrate if itsettles under an inversion?

• Do special arrangements need to be madeto protect populations impacted by theseemissions?

• How many burning projects will it takecumulatively to exceed acceptable levelswithin this airshed?

• How long will the airshed remain stableand harbor the emissions?


– 114 –

In instances where a burn may affect an areaespecially sensitive to smoke, the use of airquality monitors may be advisable to ensurethat an agreed-upon emission level or limit isnot exceeded. Factors to consider in usingmonitors include placement of the device,personnel to operate the instrument, qualitychecks, data analysis, and provisions for real-time feedback if data is to be used in makinga decision to terminate a burn in progress.Monitors are not commonly accessible andare costly to use, so this option is chieflyavailable to federal and state agencies. Airquality monitoring for evaluating a firemanagement program is discussed in Chapter10.

Smoke impacts to fireline personnel shouldalso be considered in a smoke managementplan. The burn planner should considerprojected exposure when determining thesize of the burn crew and the duration of thework shift. More information on smokeexposure to fireline personnel can be foundin Chapter 3.4.

Once an analysis of significant factors iscomplete, the planner should set specific,measurable smoke management objectivesfor the burn. These may include, for ex-ample, minimum visibility standards forroads or viewsheds, and an emissions limit ifair quality monitors are to be used. Objec-tives provide a common understanding for allindividuals involved in or affected by theburn, of what constitutes acceptable smokeimpacts. They also provide a tool for theburn boss when deciding whether to termi-nate a fire because of problematic smokebehavior. If the decision is made to termi-nate a burn because of smoke problems, itshould be remembered that direct suppres-sion often temporarily exacerbates smokeproblems. If ignition has been completed,the best strategy may be too let the fire burnout.

The amount of air quality analysis required atall levels of fire planning will be influencedby air quality laws and smoke managementregulations. Formal state smoke manage-ment programs are becoming increasinglycommon, but are not yet universal. Somestates include only regulatory languageregarding “nuisance smoke.” Complyingwith all applicable laws and regulations is abasic tenet of conscientious land stewardship,but responsible fire use and air qualityplanning include looking beyond the require-ments of the law. Communities likely to beimpacted by a fire-use program should beinvolved in determining what their thresholdof acceptance is for smoke from wildlandfire. Thorough attention to smoke manage-ment planning can prevent future problems.

7. Notification of Local Authorities and thePublic. Early development of a notificationplan will assist in the necessary communica-tion with local authorities and the public. Awide variety of methods have proven suc-cessful, including distribution of pamphletsor flyers, public meetings, newspaper andradio announcements, and Internet postings.The public should be notified well in ad-vance of the proposed burn day, and againwithin a few days of executing the burn.Generally, there is a list of individuals to benotified on the actual burn day. This list isoften unit-specific, and should be includedalong with telephone numbers in the burnplan.

8. Environmental and Legal Constraints. Ifconstraints to the burn plan have not alreadybeen addressed in a fire management plan forthe entire site, they should be addressed herebecause they can limit or determine how aburn is implemented. These may includeenvironmental, economic, operational,administrative, and legal constraints.

– 115 –


9. Operations. The burn plan must describe indetail how fire will be used. This section ofthe plan may take any number of formats, butthe topics to be addressed include:

• Safety. What provisions will be made toensure the safety of the crew?

• Communications. How will the crewcommunicate with each other, and withdispatch or emergency support?

• Equipment and Personnel. What re-sources are needed to effectively accom-plish the burn and how will they bedeployed?

• Fire Lines. What is the width and condi-tion of existing fire lines? How manychains of fireline need to be prepared orcleared? How will this be accomplished?

• Ignition Pattern and Sequence. How willthe burn be ignited? Ignition duration andfiring patterns play an important role inproduction and lofting of emissions.Rapid ignition may reduce consumption,therefore emissions, and be successful inlofting a smoke column high into theatmosphere. Backing fires produce feweremissions than heading fires. Moreinformation on using ignition to manageemissions production can be found inChapter 8, Techniques to Reduce Emis-sions and Impacts.

• Holding. How will the fire be kept withinits predetermined boundaries? How willsnags be dealt with?

• Mop-up. How will the burn be extin-guished? What standard will be used toconsider the burn unit safe to leave?

10. Contingency Planning. Contingency plansoutline procedures for dealing with a burngone awry. They are a normal part of a burnplan and should include provisions to dealnot only with escaped fire, but also withunexpected smoke intrusions during anotherwise controlled burn. Some of theissues to be addressed include safety of thegeneral public and the fire crew, sources ofassistance for fire control and smoke-relatedproblems, deployment of resources, actionsto be taken to rectify the problem, notifica-tion of authorities and the public, and mea-sures to mitigate smoke on roadways. Itshould be recognized that in some caseswhere smoke problems dictate shuttingdown a burn after ignition has been com-pleted, the most prudent action may be toallow the unit to burn out rather than toimmediately extinguish it, which can tempo-rarily exacerbate smoke production.

11. Preburn Checklist. Every burn planshould include a checklist to be reviewedimmediately prior to ignition. The checklistshould include the factors essential to safeexecution of the burn project, and a list ofpoints to review with the crew during thepreburn briefing. The use of the checklistensures that some detail does not slip by theburn manager’s attention in the busy mo-ments preceding a fire.

12. Monitoring and Evaluation. Monitoringand evaluation of the burn are key to learn-ing from the process and making refine-ments for subsequent burns. Whereappropriate and practical, monitoring andpost-fire evaluation protocols describing theeffects on soil, water, air, vegetation, andwildlife should be included in the burn unitplan. Alternatively, the information can beincluded in a post-burn evaluation report orform, which is attached to the burn plan aftercompletion.


– 116 –

• Documenting air quality conditions before,during, and after a fire is useful in identi-fying nuisance smoke thresholds andassuring that air quality standards have notbeen exceeded. Additionally, monitoringand documenting smoke transport, dilu-tion, or concentrations in each airshed canhelp develop local knowledge that is thebasis of predicting smoke impacts. Inaddition to environmental effects, thefollowing topics should be addressed:adequacy of preburn treatments, firebehavior, degree to which objectives wereachieved, discrepancies between plannedfuel and weather components and on sitemeasurements, observations, accidents ornear-accidents, slopovers, and recommendchanges for future burns. A series ofphotographs over time at permanent photopoints is an excellent inexpensive methodto document vegetation changes.

Fire Use Planning for FederalLand Managers

The Wildland and Prescribed Fire ManagementPolicy: Implementation Procedures ReferenceGuide (USDI and USDA Forest Service 1998)represents an effort by Federal wildland firemanagement agencies to establish standardizedprocedures to guide implementation of thepolicy described in the 1995 Federal WildlandFire Management Policy and Program Review.It uses new terminology and definitions toprovide consistency and interpretation tofacilitate policy implementation, and describesrelationships between planning tiers to firemanagement objectives, products, andapplications.

The federal process generally follows the plan-ning process described above. The flow ofinformation begins with the land and resourcemanagement plan, variously called the ForestManagement Plan (FS), Integrated ResourceManagement Plan (BIA), Resource Manage-ment Plan (NPS), Comprehensive ConservationPlan (FWS) and the Forest Management Plan(FS). This plan determines the availability ofland for resource management, predicts levels ofresource use and outputs, and provides for avariety of resource management practices.

The next step is preparation of the Fire Manage-ment Plan (FMP). The FMP is the primary toolfor translating programmatic direction devel-oped in the land management plan into on-the-ground action. The FMP must satisfy NEPArequirements, or follow direction provided by aForest Plan that has been developed through theNEPA process. Comparisons between fire useactivities and no fire use should be described inthe NEPA process. This includes implicationsof wildland fire and prescribed fire use overextended periods of time.

The most detailed step in the process involvesthe tactical implementation of strategic objec-tives for the wildland and prescribed fire man-agement programs. It is at this level wherespecific plans are prepared to guide implementa-tion of fire-related direction on the ground. Thisstep includes Prescribed Fire Plans, WildlandFire Implementation Plans, and the WildlandFire Situation Analysis.

More information on the smoke managementrequirements and federal planning process iscontained in Chapter 4.

– 117 –



Sandberg, D.V.; Dost, F.N. 1990. Effects of pre-scribed fire on air quality and human health. In:Walstad, John D.; Radosevich, S.R.; Sandberg,D.V., eds. Natural and prescribed fire in PacificNorthwest forests. Corvallis, OR: Oregon StateUniversity Press; 191-298.

The Nature Conservancy. 2000. Fire ManagementManual. <http://www.tncfire.org/ manual/>.

USDI and USDA Forest Service. 1998. Wildlandand prescribed fire management policy—implementation procedures reference guide.National Interagency Fire Center, Boise, ID.81 p.

Wade, Dale D. and James D. Lunsford. 1989. Aguide for prescribed fire in southern forests.Tech. Pub. R8-TP11. Atlanta, GA. USDAForest Service, Southern Region. 56 p.


– 118 –

– 119 –

2001 Smoke Management Guide 7.0 – Meteorology

SMOKE MANAGEMENT METEOROLOGY

Chapter 7

Chapter 7 – Smoke Management Meteorology 2001 Smoke Management Guide

– 120 –

– 121 –


Smoke Management Meteorology

Sue A. Ferguson

Once smoke enters the atmosphere, its concen-tration at any one place or time depends onmechanisms of transport and dispersion. Bytransport, we mean whatever carries a plumevertically or horizontally in the atmosphere.Dispersion simply is the scattering of smoke.

Vertical transport is controlled by the buoyancyof the smoke plume and stability of the atmo-sphere. Horizontal transport is controlled bywind. The larger the volume of space thatsmoke is allowed to enter and the farther it canbe transported, the more disperse and lessconcentrated it will become. To begin under-standing stability and wind that control transportand dispersion, we begin with a few elementalconcepts.

Air Pressure

It is helpful to understand air pressure becausestorms and stagnant air conditions are describedin terms of low pressure and high pressure,respectively. Lines of constant pressure are usedto illustrate the state of the atmosphere onweather maps, and pressure influences theexpansion and contraction of smoke parcels asthey travel through the atmosphere. Air pressureis the force per unit area exerted by the weightof the atmosphere above a point on or above theearth’s surface. More simply it can be thoughtof as the weight of an overlying column of air.Air pressure is greatest near the ground, wherethe overlying column of air extends the full

height of the atmosphere. Pressure decreaseswith increasing altitude as the distance to the topof the atmosphere shortens.

In a standard atmosphere, which represents thehorizontal and time-averaged structure of theatmosphere as a function of height only, pres-sure decreases approximately exponentially withheight. With 1,013 millibars (mb) being thestandard atmospheric pressure at sea level, theaverage height of the 850 mb pressure leveltypically occurs at about 5,000 feet (~1,500 m),the 700 mb pressure level typically occurs atabout 10,000 feet (~3,000 m), and the 500 mbheight averages around 20,000 feet (~6,000 m).In the lowest part of the atmosphere (less thanabout 8,000 feet) pressure decreases by approxi-mately 30 mb per 1000 feet. These are usefulvalues to remember when analyzing meteoro-logical data and maps for smoke management.Actual pressure is nearly always within about30% of standard pressure.

Lapse Rates

Lapse rate is the decrease of temperature withheight. Lapse rates help determine whethersmoke will rise from a fire or sink back to thesurface and are used to estimate atmosphericstability. When air is heated it expands, be-comes less dense and more buoyant. Thiscauses it to rise. A parcel of air that is heated atthe ground surface by fire or solar radiationbecomes warmer than its surroundings, causing


– 122 –

it to lift off the surface. As it rises, it encoun-ters lower pressure that causes further expan-sion. The more air expands, the cooler itbecomes. If a parcel of air becomes cooler thanits surroundings, it will sink.

Cooling by expansion without an exchange ofheat at the parcel boundaries is called adiabaticcooling. In dry air, rising air parcels typicallycool at a rate of about 5.5 °F per 1,000 feet (~10 °C/km). This is called the dry adiabatic lapserate (DALR). For example, on a clear day if aheated parcel of air begins at sea level with atemperature of 70 °F (~21 °C), it will cool dry-adiabatically as it rises, reaching a temperatureof 53.5 °F (~12 °C) at 3,000 feet (~915 m).

Rising moist air (relative humidity greater thanabout 70%) is said to undergo a saturation-adiabatic process. The saturated adiabatic lapserate (SALR) or moist adiabatic lapse rate is afunction of temperature and water content. Thisis because as moist air cools its water vaporcondenses, giving off latent heat in the conden-sation process and causing a saturated parcel tocool more slowly than a dry parcel. Near theground in mid-latitudes the SALR can be ap-proximated at a rate of about 3 °F per 1,000 feet(~ 5.5 °C/km). For example, on a humid orrainy day, a heated parcel with a 70 °F (~21 °C)initial temperature at sea level, will reach atemperature of 61 °F (~16 °C) at 3,000 feet(~915 m).

Lapse rates are determined by comparingtemperatures between different elevations. Thetemperature from a ridge-top weather stationcan be subtracted from the temperature at anearby valley-located weather station to calcu-late lapse rate. More commonly, radiosondeobservations (raobs) are used to determine lapserates. These balloon-mounted instruments

measure temperature, wind, pressure, andhumidity at several elevations from the groundsurface to thousands of feet. Raobs are avail-able from weather services or at several sites onthe Internet twice each day: at 0000 UniversalTime Coordinated (UTC)1 and 1200 UTC.

There are several ways of plotting raob data.Typically a pseudo-adiabatic chart is used. Thischart shows measured values of temperature vs.pressure over lines of DALR and SALR. Figure7.1 illustrates how the above examples wouldappear on a standard pseudo-adiabatic chart.More recently, skew-T/log-P diagrams (skew-Tfor short) have become popular. Instead ofplotting temperature and pressure on linear,orthogonal axes, skew-T diagrams plot the logof pressure and skew the temperature axis by45°. The skew-T/log-P view of raob dataallows features of the atmosphere to be moreobvious than when plotted on a standardpseudo-adiabatic chart. Figure 7.2 illustrates theabove examples on a skew-T diagram. On bothstandard pseudo-adiabatic charts and skew-Tdiagrams, elevation in meters or feet (corre-sponding to the pressure of a standard atmo-sphere) may be shown and wind direction andspeed with height is represented parallel to oralong the right-hand vertical axis. Many otherfeatures also may be included.

Atmospheric Stability

Atmospheric stability is the resistance of theatmosphere to vertical motion and provides anindication of the behavior of a smoke plume.Full characterization of a smoke plume requiresa complete estimation of the atmosphere’sturbulent structure that depends on the verticalpatterns of wind, humidity, and temperature,

1 Universal Time Coordinated (UTC) is Standard Time in Greenwich, England. UTC is 9 hours ahead of AlaskaStandard Time (AST), where 0000 UTC = 1500 AST and 1200 UTC = 0300 AST. UTC is 5 hours ahead of EasternStandard Time (EST), where 0000 UTC = 1900 EST and 1200 UTC = 0700 EST.

– 123 –


Figure 7.1. Standard pseudo-adiabatic chart. Short-dashed lines show the saturated adiabatic lapse rate(SALR) and long-dashed lines show the dry adiabatic lapse rate (DALR). Point A marks a parcel of air at thesurface with a temperature of 21 °C (70 °F). If the atmosphere is dry, the parcel will follow a DALR as it risesand reach point B with a temperature of 12 °C (53.5 °F) at 915m (3000 ft). If the atmosphere is saturated, theparcel will follow a SALR as it rises and reach point C with a temperature of 16°C (61 °F) at 915m (3000 ft).

Figure 7.2. Skew-T pseudo-adiabatic chart. Short-dashed lines show the saturated adiabatic lapse rate(SALR) and long-dashed lines show the dry adiabatic lapse rate (DALR). Point A marks a parcel of air at thesurface with a temperature of 21 °C (70 °F). If the atmosphere is dry, the parcel will follow a DALR as it risesand reach point B with a temperature of 12 °C (53.5 °F) at 915m (3000 ft). If the atmosphere is saturated, theparcel will follow a SALR as it rises and reach point C with a temperature of 16°C (61 °F) at 915m (3000 ft).


– 124 –

which are highly variable in space and time.Because this can be a complex calculation, itoften is approximated by estimates of staticstability. The static stability of the atmosphereis determined by comparing the adiabatic lapserate with ambient, environmental lapse rates (aswould be measured from instruments on a risingballoon). By this approximation, an unstableair mass is one in which the temperature of arising parcel of air remains warmer than itssurroundings. In a stable air mass, a risingparcel’s temperature is cooler than ambient anda neutral air mass is one in which the ambienttemperature is equal to the adiabatic lapse rate.

The most common way of estimating staticstability is to note the slope of vertically mea-sured temperature in relation to the slope of thedry (or moist) adiabatic line from a pseudo-adiabatic chart. Figure 7.3 shows raob-mea-sured dry-bulb and dew-point temperatures and

the theoretical trajectory of a parcel being liftedfrom the surface. The parcel trajectory begins atthe current surface temperature then follows aDALR until it becomes saturated. The point ofsaturation is called the lifting condensation level(LCL). Its height in meters can be approxi-mated as 120 x (T

0 – T

d), where T

0 is the tem-

perature at the surface and Td is the mean

dew-point temperature in the surface layers,both in degrees Celsius. From the LCL, theparcel trajectory follows a SALR.

Throughout the depth of the diagram in figure7.3, the slope of the measured temperature isnearly always steeper than the slope of theadiabatic temperature, suggesting that a liftedparcel always will remain cooler than the ambi-ent temperature, which is a sign of stability. Thelarge distance between the measured tempera-ture and the temperature of the theoretical parceltrajectory also gives an indication of strong

Figure 7.3. Skew-T plot of a stable atmosphere. The thick black line on the right is the measured environmental dry-bulbtemperature. The thick black line on the left is the measured environmental dew-point temperature. The red line is atheoretical parcel trajectory. Short-dashed lines are the SALR and long-dashed lines are the DALR.

– 125 –


stability. In a stable atmosphere, smoke ema-nating from relatively cool fires will stay nearthe ground. Hot fires may allow plumes to loftsomewhat through a relatively stable atmo-sphere but fumigation of smoke near the groundremains common. Figure 7.4 shows smokefrom a vigorous wildfire under a stable atmo-sphere. Smoke plumes are trying to develop buta strongly stable layer is trapping most smokejust above the ridge tops.

Parcel trajectories in an unstable atmosphereremain warmer than the measured environmen-tal temperatures (figure 7.5). During unstableconditions, smoke can be carried up and awayfrom ground level. Downwind of the source theinstability causes smoke plumes to develop alooping appearance (figure 7.6). Obviouslythere are many variations between stable andunstable atmospheres that cause various patternsof lofting, fanning, coning, looping, and fumi-gation. Each situation shows characteristic

signatures on a pseudo-adiabatic chart but someexperience may be required to distinguish thesubtle differences.

Because upper-air observations and observationsfrom significantly different elevations are notalways available, Pasquill (1961 and 1974)developed a scheme to estimate stability fromground-based observations. Not only is thisclassification system used to estimate plumecharacteristics; it also is used in many smokedispersion models as a proxy for atmosphericturbulence. Table 7.1 shows the Pasquill classi-fication criteria as modified by Gifford (1962)and Turner (1961, 1964, 1970). In this ex-ample, surface wind is measured at 10 metersabove open terrain. With clear skies, the classof incoming solar radiation is considered strong,moderate, or slight if the solar altitude angle isgreater than 60°, between 35° and 60°, or lessthan 35°, respectively. If more than 50 per centopaque cloud cover is present and the cloud

Figure 7.4. A smoke plume from a vigorous wildfire during stable atmosphericconditions. Photo by Roger Ottmar.


– 126 –

Figure 7.5. Skew-T plot of an unstable atmosphere. The thick black line on the right is the measuredenvironmental dry-bulb temperature. The thick black line on the left is the measured environmental dew-pointtemperature. The red line is a theoretical parcel trajectory. Short-dashed lines are the SALR and long-dashedlines are the DALR.

Table 7.1. Pasquill stability classification criteria, where A = extremely unstable, B = moderately unstable,C = slightly unstable, D = neutral, E = slightly stable, and F= moderately stable. See text for anexplanation of the incoming solar radiation classes.

– 127 –


ceiling height is less than 7,000 feet (~2,100m),the solar class is slight. If ceiling height isbetween 7,000 feet and 16,000 feet (~4,800m),then the solar class is one step below what itwould be in clear sky conditions. At night,classification is based on the amount of sky thatis obscured by clouds. An objective way ofdetermining stability classification is shown inLavdas (1986) and Lavdas (1997).

Mixing Height

Mixing height (also called mixing depth) is theheight above ground level through which rela-tively vigorous vertical mixing occurs. Lowmixing heights mean that the air is generallystagnant with very little vertical motion; pollut-ants usually are trapped near the ground surface.High mixing heights allow vertical mixingwithin a deep layer of the atmosphere and gooddispersion of pollutants. As such, mixingheights sometimes are used to estimate how farsmoke will rise. The actual rise of a smokeplume, however, considers complex interactions

between atmospheric stability, wind shear, heatrelease rate of the fire, initial plume size, densitydifferences between the plume and ambient air,and radiant heat loss. Therefore, an estimate ofmixing height provides only an initial estimateof plume height.

Mixing heights usually are lowest late at nightor early morning and highest during mid to lateafternoon. This daily pattern often causessmoke to be concentrated in basins and valleysduring the morning and dispersed aloft in theafternoon. Average morning mixing heightsrange from 300 m (~980 ft) to over 900 m(~2,900 ft) above ground level (Holzworth1972). The highest morning mixing heightsoccur in coastal areas that are influenced bymoist marine air and cloudiness that inhibitradiation cooling at night. Average afternoonmixing heights are typically higher than morn-ing heights and vary from less than 600 m(~2,000 ft) to over 1400 m (~4,600 ft) aboveground level. The lowest afternoon mixingheights occur during winter and along thecoasts. Mixing heights vary considerablybetween locations and from day to day.

Figure 7.6. A smoke plume during unstable atmospheric conditions.Photo by Roger Ottmar.


– 128 –

Ferguson and others (2001) generated detailedmaps and statistics of mixing heights in theUnited States.

Smoke plumes during the flaming stage of firesoften can penetrate through weak stable layersor the top of mixed layers. Once the plumedynamics are lost, however, the atmosphereretains control of how much mixing occurs.Low-level smoke impacts increase once aconvective column collapses.

The depth of the mixed layer depends on com-plex interactions between the ground surfaceand the atmosphere in a region called the plan-etary boundary layer (PBL). As such, it isdifficult to measure exactly and there are manyways in which it is calculated. At times, it ispossible to estimate the mixing height by notingthe tops of cumulus clouds or the presence of anupper-level inversion, which may appear as adeck of strata-form clouds.

Typically, National Weather Service (NWS)smoke management forecast products willestimate the mixing height by the so-calledparcel method. This method considers turbu-lence related only to buoyancy. When a parcelis lifted adiabatically from the surface, the pointat which it intersects the ambient temperatureprofile, or where it becomes cooler than itssurroundings, is the mixing height. Usually themaximum daily temperature is used as theparcel’s starting temperature and its adiabaticlapse rate is compared with the afternoon (0000UTC) sounding profile. Conversely, the mini-mum daily temperature is used to compare withthe morning (1200 UTC) raob for calculatingmorning mixing heights. If an elevated inver-sion (see next section) occurs before this heightis reached, the height of the inversion basewould determine the mixing height. If a surfaceinversion exists, then its top marks the mixingheight. For example, the mixing height in figure

7.3 is at the top of the surface-based inversion atabout 750 mb (approximately 2,400 meters or7,800 feet above ground level).

Instead of approximating a mixing depth,physical calculations of the PBL are possiblethrough numerical meteorological models.These calculations are more precise than theparcel method because they consider turbulencegenerated by wind shear as well as buoyancy.Each prognostic model, however, may calculatethe PBL slightly differently as some functionsare approximated while others are explicitlyderived to enhance computational efficiency andthe vertical resolution, which varies betweenmodels, affect PBL calculations.

Temperature Inversions

When the ambient temperature increases withheight, an inversion is said to be present. Itusually marks a layer of strong stability. Whena heated air parcel from the surface encountersan inversion, it will stop rising because theambient air is warming faster than the expand-ing parcel is cooling. The parcel being coolerthan its surroundings will sink. Although theheat from some fires is enough to break througha weak inversion, inversions often are referred toas lids because of their effectiveness in stoppingrising air and trapping pollutants beneath it.Smoke trapped under an inversion can substan-tially increase concentrations of particles andgases, aggravating respiratory problems andreducing visibility at airports and along road-ways.

There are three ways that surface-based inver-sions typically form: (1) valley inversions arevery common in basins and valleys during clearnights when radiation heat losses cause air nearthe ground to rapidly cool: the cold surface airflows from the surrounding slopes and collects

– 129 –


in hollows and pockets, allowing warmer air toremain aloft; (2) advective inversions are causedby cold air moving into a region from a nearbylake or ocean, usually during the afternoon whenonshore lake and sea breezes tend to form; and(3) subsidence inversions can occur at any timeof day or night as cold air from high altitudessubsides or sinks under a region of relativelystagnant high pressure. Valley inversions causetremendous problems when managing long-duration fires that continue into the night. Ad-vective inversions can surprise smoke managerswho are unfamiliar with local lake- and sea-breeze effects, creating poor dispersal conditionsin an afternoon when typically good dispersion isexpected. Subsidence inversions are difficult topredict even for a well-trained meteorologist.Figure 7.7 shows smoke caught under a valleyinversion that is being transported by down-valley winds in the early morning.

Surface inversions also occur in the gaps (passesand gorges) of mountain ranges. Approachingstorms usually have an associated center of lowpressure that causes a pressure gradient acrossthe range. If cold air is on the opposite side of

the range, the gradient in pressure causes thecold air to be drawn through the gap, creating aninversion in the gap. Gap inversions are mostcommon in winter but also are frequent duringspring and autumn.

In addition to surface-based inversions, tempera-ture inversions also occur in layers of the atmo-sphere that are above the ground surface, whichsometimes are called thermal belts. Upper-levelinversions usually are associated with incomingwarm fronts that bring moisture and warmth tohigh altitudes well ahead of a storm. Theinversion lowers to the ground as the frontapproaches. Upper-level inversions also may beassociated with subsidence or surface-basedinversions that have been lifted, usually bydaytime heating.

Wind

Not only does smoke mix and disperse verti-cally, the horizontal component of wind readilytransports and disperses pollutants. The stron-

Figure 7.7. A plume of smoke flowing out of a mountain valley with down-slopewinds during the early morning. Photo by Roger Ottmar.


– 130 –

ger the wind, the more scattered particles be-come and the less concentrated they will be.Strong winds at the surface, however, canincrease fire behavior and associated emissionrates. Also, significant surface winds may “lay-down” a plume, keeping smoke close to theground for long distances.

Friction with the ground causes winds to slowdown. Therefore, wind speed usually increaseswith height, causing a smoke column to gradu-ally bend with height as it encounters increas-ingly strong winds. This pattern is complicatedin regions of complex terrain, however, and it iscommon to find stronger surface winds inmountain passes, saddles, and gorges as air issqueezed and funneled through the gap. Forestclearings also allow surface winds to acceleratebecause surface friction is lower in a clearingthan over a forest canopy.

Because smoke from different stages of a firerises to different levels of the atmosphere, it isimportant to know wind speed and direction atseveral different heights. For example, smolder-ing smoke at night responds to surface windswhile daytime smoldering and smoke from theignition and flaming phase of a fire will respondto upper-level winds. Depending on the buoy-ancy of the smoke and stability of the atmo-sphere, winds that influence the upper-levelsmoke trajectories may be from just above aforest canopy to 10,000 feet (about 3,000meters) or more. Because flaming heat cancreate convective columns with strong verticalmotion, most smoke during the flaming portionof a fire will be carried to at least the top of themixing height or an upper-level inversion heightbefore dispersing. In this way, a fire hot enoughto pull itself into a single convection column canreduce concentrations near the ground andknowledge of winds at the top of the mixing

height or inversion level will determine smoketrajectory and dispersion. Smoldering smoke,on the other hand, has very little forced convec-tion so it often fumigates away from a fire as itrises with daytime buoyancy. Knowledge ofwind all the way from the surface to top of themixing height may be needed to determinesmoldering trajectories.

Storm Winds – Storms change the structure ofwinds entirely. Because storms often bring highinstability and good dispersion, it is common toplan fires slightly ahead of an approachingstorm. Knowing storm wind patterns can helpanticipate associated smoke impacts. Figure 7.8shows surface wind directions2 typically associ-ated with a passing cyclonic storm. Because airflows from high pressure to low pressure (likethe rush of air from a punctured tire) and stormsusually have a center of low pressure at thesurface, surface winds ahead of a storm in thenorthern hemisphere will be from the east orsoutheast. As the low center approaches, sur-face winds will become southerly to southwest-erly. After the storm passes, surface winds maybecome more westerly or northwesterly. Thispattern can cause smoke to move toward thewest to northwest then north to northeast aheadof a cyclonic storm, moving toward the east andsoutheast following storm passage.

Each cyclonic storm usually contains at leastone front (a boundary between two different airmasses). A typical storm has a warm frontaligned northwest to southeast ahead of the lowcenter, a cold front trailing northeast to south-west near and closely behind the low, and anoccluded front (formed when a cold frontovertakes a warm front) to the north of the low.Winds change direction most rapidly and be-come gusty when fronts pass by. Warm frontscan bring increasing stability and cause upper-

2 Wind direction is the direction from which the wind is blowing. For example, a west wind is coming from the westand blowing toward the east. If you face east, a west wind will hit your back.

– 131 –


Figure 7.8. Schematic of surface winds associated with a typical cyclonic storm in the Northern Hemisphere.The letter, L, marks position of the surface low pressure center. Thin lines represent isobars (constant pressurecontours that are labeled in millibars) at sea level. The thick line marked with barbs represents a surface coldfront, marked with half-circles is a warm front, and marked with both is an occluded front. East to southeastsurface winds are common ahead of a warm front, south to southwest winds are common ahead of a cold front,and west to northwest winds are common following a cold front.

level inversions, while cold fronts usually areassociated with strong instability. The strongerthe front, the more dramatic the wind shift andthe stronger the gusts. Cold frontal passagetypically improves dispersion of smoke withstronger winds and an unstable air mass that canscour away existing inversions. Smoke trajecto-ries should be expected to change direction withthe passage of a storm front and storms cancause significant changes in fire behavior andresulting emission rates. Storm fronts are notalways typical, however, and the number,strength, and orientation of fronts are quitevariable.

Strong winds above the influence of the earth’ssurface experience forces associated with theearth’s rotation in addition to pressure gradientand other forces. This causes winds in the upper

atmosphere to follow lines of constant pressureinstead of moving across lines of constantpressure as surface or lower-speed winds dowhen air flows from high pressure to low pres-sure. In the upper atmosphere the pressurepattern of a typical storm is shaped like a trough(figure 7.9). As air follows the pressure con-tours around the trough, southwesterly upper-level winds occur ahead of the storm, becomingwesterly as the storm trough passes, and north-westerly following the trough. The upper-leveltrough usually trails the surface low center inmost moving fronts, causing smoke trajectoriesaloft to change directions sometime after trajec-tories at the surface have changed following astorm passage.

Thunderstorms, which are the result of strongconvection, create much different wind patterns


– 132 –

than cyclonic storms. Gusty, shifty winds arecommon at times of strong convection. Strongdown bursts of wind in a direction away fromthe thunder cell may occur several minutesahead the storm, while winds around the cellmay be oriented towards it. Although mixingheights usually are quite high during thunder-storms, allowing for well-lofted plumes, theshifting wind directions and strong gusts cancause variable and unpredictable smoke trajecto-ries and fire behavior in close proximity tothunderstorms.

Diurnal Winds – In the absence of storms,diurnal wind patterns dominate trajectories ofsmoke near the ground. Diurnal patterns are

caused by differences between radiationalcooling at night and solar heating during theday, and by different thermal properties of landand sea surfaces that cause them to heat andcool at different rates. The differential heatingcauses changes in surface pressure patterns thatcontrol air movement. Slope winds and sea andlake breezes, all of which are common in wild-land smoke management situations, typifydiurnal patterns.

Slope winds are caused by the same mecha-nisms that cause valley and basin inversions.When cold air from radiation cooling at nightdrains into a valley or basin, it causes adownslope wind. The cold air, being denser

Figure 7.9. Schematic of upper-level (700 mb) winds associated with a typical stormy trough pattern in theNorthern Hemisphere. Thin lines represent pressure height contours that are labeled in tens of meters. Southto southwest upper-level winds are common ahead of a 700 mb trough, westerly winds are common as thetrough passes, and northwesterly winds are common following an upper-level trough.

– 133 –


than its surroundings, usually hugs the terrain insuch a way that smoke following a drainagewind will follow contours of the terrain. Duringthe day, heated air from the surface rises, caus-ing upslope winds. Because daytime heatingcauses more turbulence than nighttime cooling,the daytime winds do not follow terrain asreadily as nighttime winds, causing thermally-induced upslope winds to be less noticeable thandownslope winds.

Downslope winds at night are notorious forcarrying smoke into towns and across roadways(e.g., Achtemeier et al. 1988), especially whereroads and bridges cross stream channels orwhen towns are located in valleys, basins, ornear outwash plains. Downslope winds aremost likely to occur when skies are clear andambient winds are nearly calm. The speed andduration of a downslope wind is related to thestrength of its associated valley inversion.Downslope winds usually begin around sunsetand persist until shortly before sunrise.

Sea and lake breezes usually occur during theafternoon when land surfaces have had a chanceto heat sufficiently. The heated air rises, as iflifting the overlying column of air. This causesa region of low pressure at the surface. Becauseland heats more rapidly than water, the differen-tial heating causes a pressure gradient to form.Relatively cool air remaining over a lake orocean will flow into the low pressure formedover heated land surfaces. The sea or lakebreeze not only can change smoke trajectoriesbut the incoming cool air can cause surfacebased inversions that will trap smoke at lowlevels near the ground. Also, strong sea breezescan knock plumes down, causing increasingsmoke concentrations near the ground.

Terrain-Influenced Wind – Surface winds arestrongly influenced by small undulations interrain that channel, block, or accelerate air as ittries to move around or over features. For

example, if upper-level winds are orientedperpendicular to a terrain barrier, surface windson the lee side of the barrier often are light andvariable. Upper-level winds oriented in thesame direction as a valley will enhance upvalleyor downvalley winds. Cross-valley winds willbe 90° different than those in the valley itself.

The combination of wind and atmosphericstability determine whether smoke will collecton the windward side of a terrain barrier, moveup, over and away, or traverse the barrier only toaccumulate on the leeward side. Weak windsand a stable atmosphere will enhance blockingand windward accumulations of smoke. Stron-ger winds in a stable atmosphere may allowaccumulations of smoke in leeward valleys andbasins. An unstable atmosphere allows smoketo be lifted over and above the terrain. Theheight, steepness, and orientation of the terrainto the wind direction determine how strong thewind or unstable the atmosphere must be toinfluence smoke trajectories.

Often very small-scale undulations in topogra-phy can affect smoke trajectories, especially atnight when atmospheric stability keeps smokeclose to the ground. Gentle saddles in ridgesmay offer outflow of smoke from a valley.Small streambeds can collect and transportsignificant amounts of smoke even with onlyshallow or weak downslope winds. A simpleband of trees or brush may provide enoughbarrier to block or deflect smoke. As the urban-wildland interface becomes increasingly com-plex, the role of subtle topographic influencesbecomes increasingly important.

Higher in the atmosphere, away from the earth’ssurface, topography plays a decreasing role incontrolling wind speed and direction. Upper-level winds above the influence of underlyingterrain are referred to as “free-air” winds andtend to change slowly from one place to another,except around fronts and thunderstorms.


– 134 –

The Role of Inversions on Wind – Tempera-ture inversions significantly influence winddirection and speed. Under many inversionsthere is little or no transport wind and smoketends to smear out in all directions. Someinversions, such as advected inversions that areassociated with sea breezes and valley inver-sions, may have significant surface wind but itusually is in a different direction to winds aloft.In these cases, surface smoke may be trans-ported rapidly under the inversion in one direc-tion while lofted smoke may be transported inan opposite direction.

Wind Observations – Because surface windsare strongly influenced by small undulations interrain, vegetation cover, and proximity toobstacles and water bodies, it is important toknow where a surface wind observation is takenin relation to the burn site. For example, obser-vations from a bare slope near the ridgeline willgive a poor indication of winds affecting surfacesmoke trajectories if most of the burn area is ona forested slope or in a valley, even if the twosites are very close. Also, if a burn site is in aneast-west oriented valley and the nearest obser-vation is in a north-south oriented valley, ob-served winds can be 90° different from thoseinfluencing the fire and its related smoke.Sometimes, a nearby Remote AutomatedWeather Station (RAWS) will be less represen-tative of burn-site conditions than one that isfarther away if the distant station is in a locationthat better matches terrain effects expected atthe burn site.

There are four principle sources of surface winddata: (1) on-site measurements with a portableRAWS or hand-held anemometer, (2) observa-tions that estimate winds using the Beaufort

wind scale3 or wind sock,4 (3) local measure-ments with a standard RAWS, and (4) measure-ments from NWS observing stations. Becausestations vary in their surroundings, from smallclearings on forested slopes to open fields, anddifferent types of anemometers are used that aremounted at different heights, wind data is verydifficult to compare between one site andanother. Therefore, it is useful to becomefamiliar with measurements and observationsfrom reliable sites and understand local effectsthat make data from that site unique. Also,smoke near the ground can be transported bywinds that are too light to spin the cups orpropeller of an anemometer or turn its tail.Frequently light and variable wind measure-ments actually are responding to very lightwinds that have a preferred direction, ofteninfluenced by surrounding topography or landuse.

Because free-air winds are above the influenceof topography, often it is possible to use anupper-level observation from some point wellaway from the burn site to estimate upper-levelsmoke trajectories. Also, surface RAWS thatare mounted on the tops of ridges or mountainsmay compare well with free-air winds at asimilar elevation. If clouds are in the area,upper-level winds can be estimated by theirmovement relative to the ground. High cloudslook fibrous or bright white. Because the baseof high clouds ranges between 5 km and 13 km(about 16,000 to 45,000 feet) their movementcan indicate wind at those high levels. Mid-level clouds may have shades of gray or bulbousedges with bases ranging from 2 km to 7 km(about 6,6000 to 24,000 feet). Mid-level cloudsoften have a strata-form or layered appearance,which may indicate the presence of an inversion.

3 The Beaufort wind scale estimates wind speed using observations of wind-effects in the landscape. For example,wind speeds of 1.6 to 3.3 m/s (4 to 7 mph) will cause leaves to rustle slightly. If leaves move around vigorously then thewind speed is approximately 3.4 to 5.4 m/s (8 to 12 mph).

4 Wind socks continue to be used at airports and are useful if trying to monitor winds on a nearby ridge that isvisible.

– 135 –


Therefore, movement of these types of cloudsmay closely approximate steering winds for arising smoke plume.

In addition to observations, it is becomingincreasingly common to have available theoutput from wind models. These data do notprovide the detail of a point observation the wayan individual site measurement does, but they doprovide a broad view of wind patterns over thelandscape. Standard analyses from the NWSuse models to interpolate between observations.These products help illustrate upper-level windpatterns and typically are available for 850 mb,700 mb, and 500 mb heights, either from a state,federal, or private meteorological service, or avariety of Internet sites. For surface winds,standard NWS analyses are helpful in regions offlat or gently rolling terrain but mesoscalemeteorological models typically are needed toresolve surface wind fields in regions of com-plex topography. Several regions throughout thecountry are beginning to employ mesoscalemodels (e.g., MM5, RAMS, and MASS)producing wind maps with less than 15 kmhorizontal spacing. Local universities, researchlabs, state offices, and consortia of local, state,and federal agencies have undertaken mesoscalemodeling efforts. Output usually can be foundon a local Internet site through the NWS fore-cast office, a fire weather office, university, stateregulator, EPA office, or regional smoke man-ager. Also, many smoke dispersion modelshave built-in wind models to generate surfacewinds at very fine spatial resolutions (less than 5km grid spacings) from inputs of surface andupper-air observations or data from coarsermeteorological models. Smoke dispersionmodels and their related wind models may beavailable through a regional smoke manager orEPA office (see Chapter 9—Smoke DispersionPrediction Systems).

Atmospheric Moisture

Because water vapor in the atmosphere reducesvisibility, if smoke is added to an already humidenvironment, visibility can be severely de-graded. Also, if the air is saturated with watervapor, particles from smoke may act as conden-sation nuclei causing water droplets to form.This promotes the formation of clouds or fog,which further degrades visibility. Often adeadly combination occurs during the darknessof night as smoldering smoke drains down-valley to encounter high humidities from con-densing cold air under a valley inversion. Theeffect can be fatal, especially along transporta-tion corridors (Achtemeier and others 1998).

Favorable conditions for fog occur when thedew point temperature is within a few degrees ofthe dry bulb temperature, wind is less than a fewmeters per second, and there is a high content ofmoisture in the soil. Fog is most common atnight when temperatures often drop to near thedew point value and winds are most likely to beweak. Common places for fog to form are overlakes and streams and in the vicinity of bogs andmarshes.

There are times when atmospheric moisture canimprove visibility, however. Smoke particlescan adhere to rain droplets, causing them to becarried with the rain as it falls. This “scaveng-ing” effect removes smoke particles out of theatmosphere, reducing smoke concentrations andimproving visibility.

Weather Forecasts

Weather forecasts typically are produced twiceeach day and become available within 3 to 6hours after 0000 UTC and 1200 UTC observa-tions are complete. This is because prognostic


– 136 –

models require input data from the 0000 UTCand 1200 UTC upper-air observations and a fewhours of run-time on a super computer. Prog-nostic models (progs) form the basis of mostforecast products. For example, the first fore-cast of the day should be available by 7 am to 10am local daylight time from Anchorage and by10 am to 1 pm local standard time from Miami.Earlier forecasts or forecasts updated throughoutthe day are possible if the most recently avail-able upper-air observations and prognosticmodel outputs are combined with updatedsurface observations. While public forecastsissued by the NWS and the media are useful,they typically lack the detail needed for smokemanagement. For this reason, spot-weatherforecasts may be requested from state, federal,or private weather services that provide predic-tions of critical variables that influence smoke atspecified times and locations.

Even though there are increasing numbers ofnumerical guidance tools, weather forecastingstill is an art, especially in places with fewobservations or where there are complex localinteractions with terrain, water bodies, andvegetation cover. The primary source of smokeweather forecasts remains the National WeatherService. Their rigorous training, fire weatherprogram, and state-of-the art equipment andanalysis tools help maintain a unique expertise.Most NWS fire weather forecast offices nowissue special dispersion and transport forecasts.In addition to NWS forecasters, many statesmaintain a smoke management program withhighly skilled meteorologists. Also, the numberof inter-agency fire weather offices and privatemeteorological services is growing and canprovide reliable forecast products specificallydesigned for smoke management. Whatever thesource of a forecast, it is helpful to combine theforecast with your own general understanding ofweather conditions by reviewing the manysatellite pictures, current observation summa-

ries, and prognostic model output products nowavailable on the World Wide Web. In this way,apparent trends and local influences can bedetermined and the need for last minute changescan be recognized more quickly. For example,increasing afternoon cloudiness in the forecastmay have indicated an approaching storm thatwas predicted for the following morning. Ifclouds do not increase when predicted, however,it could be suspected that the storm has beendelayed or it was diverted elsewhere. A checkwith the forecaster or updated satellite picturemay confirm the suspicion and the managementplan may be altered.

Because the atmosphere behaves chaotically, theaccuracy of a weather forecast improves as timeto an event shortens. For example, it is possibleto provide an indication of storminess within 30to 90 days. A storm passage, however, may notbe predicted until about 14 days in advance withabout 2 days accuracy. Within 5 days, 1-dayaccuracy on storm passage may be possible.Increasing accuracy should be expected within48 hours and the timing of storm passage within1/2 hour may be possible with 12 hours advancenotice. Spot weather forecasts usually areavailable 24 to 48 hours in advance of a sched-uled burn. This allows a smoke manager toanticipate a potential burning window well inadvance. Specific timing, however, should notbe made before 2 days in advance if the situa-tion is highly dependent on an accurate weatherforecast.

Our increasing knowledge of air-sea interactionsis making it possible to predict some aspects ofweather up to a year in advance as certainregions of the country respond to the El NiñoSouthern Oscillation (ENSO). Precipitation andtemperature during winter and spring are moststrongly related to ENSO. Relating key factorsfor smoke management such as wind and mix-ing height or stability is more difficult, espe-

– 137 –


cially during summer. Nevertheless, an ENSO-based seasonal prediction gives prescribedburners an idea of general weather conditions tobe expected, thereby helping prioritize sched-uled burns and decide if marginal days orweekends early in the burning season should beused or whether a more optimum season willensue.

Climate

Climate simply describes the prevailing weatherof an area. Understanding climate patterns canhelp develop long-range smoke managementplans or adapt short-range plans. For example,afternoon mixing heights in most coastal regionsof the United States are typically lower than theinterior because moist, marine air is relativelystable. This means that there may be fewer dayswith optimum dispersion along the coast thaninterior. It usually is windier along the coast,however, and burns might be scheduled in theearly morning if offshore breezes are desired toreduce smoke impacts on cities and towns.

It is possible to infer climate just by localproximity to oceans, lakes, rivers, and moun-tains. Also, vegetation cover can give an indica-tion of climate. Desert landscapes, with a lot ofbare soil or sand, heat and cool rapidly, causingthem typically to have high daytime mixingheights and very low nighttime mixing heights.Natural landscapes of lush green forests tend toabsorb sunlight while transpiring moisture, bothof which help to modify heating and cooling ofthe ground surface. This can reduce daytimemixing heights and keep nighttime heightsrelatively high, with respect to deserts. Also, thestructural deformation of trees often indicateshigh winds, where the direction of branches orflagging point away from prevailing winddirections.

Quantitative summaries of climate can beobtained from the state climatologist or Re-gional Climate Center (RCC), many of whomalso maintain informative Internet sites and canbe reached through the National Climatic DataCenter (NCDC) <www.ncdc.noaa.gov>. It ismost common to find temperature and precipita-tion in climate summaries. Monthly or annualaverages or extremes are readily available whileclimate summaries of daily data are just begin-ning to emerge. For example, a recently gener-ated climate database by Ferguson and others(2001) provides information on twice-dailyvariations in surface wind, mixing height, andventilation index over a 30-year period.

We know that there are year-to-year variationsin climate (e.g., ENSO) so at least 10 years ofweather data are needed to obtain a preliminaryview of climate in a particular area. There alsoare natural, “decadal” patterns in climate thatlast from 7 to 20 years. Therefore, it is appro-priate to acquire 30 to 50 years of weatherobservation data for any reliable climate sum-mary.

Summary

Managing smoke in ways that prevent seriousimpact to sensitive areas from single burns ormultiple burns occurring simultaneously re-quires knowledge of the weather conditions thatwill affect smoke emissions, trajectories, anddispersion. Not only is it necessary to anticipatethe weather ahead of time through the use ofclimatology and forecasts, but it also is useful tomonitor conditions prior to and during the burnwith regional, local, and on-site observations.On-site observations are helpful because airmovement, and therefore smoke movement, isinfluenced by small variations in terrain andvegetation cover, and proximity to lakes and


– 138 –

oceans, which off-site observations usuallycannot capture. Also, forecasts are not alwaysaccurate and last-minute changes in a burn orsmoke management plan may be needed. Togain more insight into the physical process ofweather in wildland areas and its effect onbiomass fires, refer to the Fire Weather hand-book (Schroeder and Buck 1970).

In using weather observations, forecasts, andclimate summaries effectively for smoke man-agement there are 3 general guidelines; (1)become familiar with local terrain features thatinfluence weather patterns, (2) develop a dia-logue with a reliable local weather forecaster,and (3) ask for and use climate summaries ofwind and mixing height. By combining yourknowledge of local weather effects, trust andcommunication with an experienced forecaster,and understanding of climate patterns, it ispossible to fine-tune or update forecasts to meetyour specific smoke management needs.

Literature CitationsAchtemeier, G.L., W. Jackson, B. Hawkins, D.D.

Wade, and C. McMahon. 1998. The smokedilemma: a head-on collision! Transactions ofthe 63rd North American Wildlife and NaturalResource Conference. 20-24 March 1998.Orlando FL. 415-421.

Ferguson, S.A.; S. McKay; D. Nagel; T. Piepho;M.L. Rorig; and C. Anderson. 2001. Thepotential for smoke to ventilate from wildlandfires in the United States. In Proceedings of the4th Symposium on Fire and Forest Meteorol-ogy, 13-15 November 2001, Reno, Nevada. p.115.

Gifford, F. A. 1962. Use of routine meteorologicalobservations for estimating atmospheric disper-sion. Nuclear Safety. 2(4):47-51.

Holzworth, George C. 1972. Mixing heights, windspeeds and potential for urban air pollutionthroughout the contiguous united states. Envi-ronmental Protection Agency, Office of AirPrograms Publication NO. AP-101. ResearchTriangle Park, NC 27711. 118 p.

Lavdas, L.G. 1986. An Atmospheric DispersionIndex for Prescribed Burning. U.S. Departmentof Agriculture, Forest Service, SoutheasternForest Experiment Station. Research PaperSE-256. 35 pp.

Lavdas, L.G. 1997. Estimating Stability Class inthe Field. U.S. Department of Agriculture,Forest Service, Southeastern Forest ExperimentStation. Research Note SRS-4.

Pasquill, F. 1961. The estimation of the dispersionof windborne material. Meteorological Maga-zine. 90:33-49.

Pasquill, F. 1974. Atmospheric Diffusion: TheDispersion of Windborne Material from Indus-trial and Other Sources. 2nd Edition. JohnWiley and Sons, New York. 429 pp.

Schroeder, M.J. and C.C. Buck. 1970. FireWeather: A guide for application of meteoro-logical information to forest fire control opera-tions. U.S. Department of Agriculture, ForestService, Agriculture Handbook 360. 229 pp.

Turner, D.B. 1961. Relationship between 24-hr.Mean air quality measurement and meteorologi-cal factors in Nashville, TN. Journal of AirPollution Control Association. 11:483-489.

Turner, D.B. 1964. A diffusion model for an urbanarea. Journal of Applied Meteorology. 3:83-91.

Turner, D.B. 1970. Workbook of AtmosphericDispersion Estimates. Office of Air ProgramsPub. No. AP-26, U.S. Environmental ProtectionAgency, Research Triangle Park, NC.

– 139 –

2001 Smoke Management Guide 8.0 – Smoke Management Techniques

SMOKE MANAGEMENT TECHNIQUES

Chapter 8

Chapter 8 – Smoke Management Techniques 2001 Smoke Management Guide

– 140 –

– 141 –


Smoke Management: Techniquesto Reduce or Redistribute Emissions

Roger D. Ottmar

Janice L. Peterson

Bill Leenhouts

John E. Core

Introduction

A land manager’s decision to use a specificburning technique is influenced by many con-siderations, only one of which is a goal toreduce smoke emissions. Other importantconsiderations include ensuring public andfirefighter safety, maintaining control of the fireand keeping it within a given perimeter, comply-ing with numerous environmental regulations,minimizing nuisance and hazard smoke, mini-mizing operational costs, and maximizing thelikelihood of achieving the land managementobjective of the burn. Often these other consid-erations preclude the use of techniques thatreduce emissions. In some cases, however,smoke emission reductions are of great impor-tance and are achieved by compromising othergoals. Emission reduction techniques varywidely in their applicability and effectiveness byvegetation type, burning objective, region of thecountry, and whether fuels are natural or activ-ity-generated.

Emission reduction techniques (or best availablecontrol measures–BACM) are not withoutpotential negatives and must be prescribed andused with careful professional judgment and full

awareness of possible tradeoffs. Fire behavior isdirectly related to both fire effects and fireemissions. Emission reduction techniques alterfire behavior and fire effects and can impair orprevent accomplishment of land managementobjectives. In addition, emission reductiontechniques do not necessarily reduce smokeimpacts and some may, under certain circum-stances, actually increase the likelihood thatsmoke will impact the public. Emission reduc-tion techniques can cause negative effects onother valuable resources such as through soilcompaction, loss of nutrients, impaired waterquality, and increased tree mortality; or theymay be dangerous or expensive to implement.

Land managers are concerned about the repeatedapplication of any resource treatment techniquethat does not replicate the ecological role thatfire plays in the environment. Such applicationsmay result in unintended resource damage,which may only be known far in the future.Some examples of resource damage that couldoccur from the use of emission reduction tech-niques include the loss of nutrients to the soil iftoo much woody debris is removed from the


– 142 –

site, or the effects of soil compaction associatedwith mechanical processing (chipping, shred-ding, or yarding) of fuels. The application ofherbicides and other chemicals and/or theeffects on soils of the intense heat achievedduring mass ignition are also of concern. Theseissues are difficult to quantify but are of univer-sal importance to land managers, who mustweigh the impact of their decisions on long-termecosystem productivity.

Multiple resource values must be weightedalong with air quality benefits before emissionreduction techniques are prescribed. Flexibilityis key to appropriate application of emissionreduction techniques and use of particulartechniques should be decided on a case-by-casebasis. Emission reduction goals may be targetedbut the appropriate mix of emission reductiontechniques to achieve those goals will require acareful analysis of the short and long termecological and social costs and benefits. Airquality managers and land managers shouldwork together to better understand the effective-ness, options, difficulties, applicability, andtradeoffs of emission reduction techniques.

There are two general approaches to managingthe effects of wildland fire smoke on air quality:

1. Use techniques that reduce the emissionsproduced for a given area treated.

2. Redistribute the emissions through meteo-rological scheduling and by sharing theairshed.

Although each method can be discussed inde-pendently, fire practitioners often choose light-ing and fuels manipulation techniques thatcomplement, or are consistent with, meteoro-logical scheduling for maximum smoke disper-sion and favorable plume transport.

Meteorological scheduling is often the mosteffective way to prevent direct smoke impacts tothe public and some emission reduction tech-niques may actually increase the likelihood ofsmoke impacts by decreasing the energy in theplume resulting in more smoke close to theground. A few of the potential negative conse-quences of specific emission reduction tech-niques are mentioned in this chapter althoughthis topic is not addressed comprehensively.

Use of SmokeManagement Techniques

Much of the information presented in thischapter was gathered from fire practitioners atthree national workshops held during the fall of1999. Practitioners were asked to describe how(or if) they apply emission reduction techniquesin the field, how frequently these methods areused, how effective they are, and what con-straints limit their wider use. The informationgained at each of the workshops was thensynthesized into a draft report that was distrib-uted to the participants for further review andcomment. Twenty-nine emission reduction andemission redistribution methods within sevenmajor classifications were identified as currentlyin use to reduce emissions and impacts fromprescribed burning.

The emission reduction methods described inthis document may be used independently or incombination with other methods on any givenburn. In addition, a number of different firingmethods potentially can be applied to any givenparcel of land depending on the objectives andjudgments made by the fire manager. As aresult, no two burns are the same in terms ofpollutant emissions, smoke impacts, fuel con-sumption, or other parameters.

– 143 –


Significant changes in public land managementhave occurred since EPA’s release of the firstdocument describing best available controlmeasures (BACM) for prescribed burning (EPA1992). Some of these changes have dramati-cally impacted when and how emission reduc-tion methods for prescribed fire can be applied.On federally managed lands, the followingconstraints apply to many of the emissionreduction techniques: National EnvironmentalPolicy Act (NEPA), Threatened and EndangeredSpecies (T&E) considerations, water quality andimpacts on riparian areas, administrative con-straints imposed by Congress (eg, roadless andwilderness area designations), impacts onarchaeological resources, smoke managementprogram requirements, and other state environ-mental or forestry regulations.

The following emission reduction and emissionredistribution techniques are a comprehensivecompilation of the current state of the knowl-edge. Any one of these may or may not beapplicable in a given situation depending uponspecifics of the fire use objectives, projectlocations, time and cost constraints, weather andfuel conditions, and public and firefighter safetyconsiderations.

Reducing the Amountof Emissions

Emissions from wildland fire are complex andcontain many pollutants and toxic compounds.Emission factors for over 25 compounds havebeen identified and described in the literature(Ward and Hardy 1991; Ward and others 1993).A simplifying finding from this research is that

all pollutants except nitrous oxide (NOx) are

negatively correlated with combustion effi-ciency, so actions that reduce one pollutantresults in the reduction of all (expect NOx).Nitrous oxide and CO

2 (not considered a pollut-

ant) can increase if the emission reductiontechnique increases combustion efficiency.

Emission reduction techniques may reduceemissions from a given prescribed burn area byas much as about 60 percent to as little asvirtually zero1. Considering all burning nation-ally, if emission reduction techniques wereoptimally used, emissions could probably bereduced by approximately 20-25 percent assum-ing all other factors (vegetation types, acres,etc.) were held constant and land managementgoals were still met1. Individual states or re-gions may be able to achieve greater emissionreductions than this or much less depending onthe state’s or region’s biological decompositioncapability or ability to utilize available biomass.

In the context of air quality regulatory pro-grams, current or future emissions are typicallymeasured against those that occurred during abaseline period (annual, 24-hour, and seasonal)to determine if reductions have or will occur inthe future. Within this framework, land manag-ers need to know their baseline emissions todetermine the degree of emission reduction thata method described here will provide in order toconform to a State Implementation Plan, StateSmoke Management Program, or local nuisancestandards.

Because of all these variables, wildland fireemission models such as the First Order FireEffects Model (FOFEM) (Reinhardt and others1997), Consume 2.1 (Ottmar and others [in

___________________________________

1 Peterson, J. and B. Leenhouts. 1997. What wildland fire conditions minimize emissions and hazardous air pollut-ants and can land management goals still be met? An unpublished technical support document to the EPA Interim AirQuality Policy on Wildland and Prescribed Fires. August 15, 1997. (Available from the authors or online at http://www.epa.gov/ttncaaa1/faca/pbdirs/emissi.pdf


– 144 –

preparation]), and Emissions Production Model(EPM) (Sandberg and Peterson 1984) can beused to estimate particulate, gaseous and haz-ardous pollutant emissions based on the specif-ics of each burn. There are seven generalcategories that encompass all of the techniquesdescribed in this document. Each is describedbelow.

1. Reduce the Area Burned

Perhaps the most obvious method to reducewildland fire emissions is to reduce the areaburned. Area burned can be reduced by notburning at all or by burning a subset of the areawithin a designated perimeter. Caution must beapplied though, and programs to reduce the areaburned must not ultimately result in just a delayin the release of emissions either through pre-scribed burning at a later date or as the result ofa wildland fire. Reducing the area burnedshould be accomplished by methods that trulyresult in reduced emissions over time rather thana deferral of emissions to some future date.

This technique can have detrimental effects onecosystem function in fire-adapted vegetationcommunity types and is least applicable whenfire is needed for ecosystem or habitat manage-ment, or forest health enhancement. In someareas and some vegetation types, when fire isused to eliminate an undesirable species ordispose of biomass waste, alternative methodscan be used to accomplish effects similar towhat burning would accomplish. Examples ofspecific techniques include:

• Burn Concentrations. Sometimes con-centrations of fuels can be burned ratherthan using fire on 100 percent of an arearequiring treatment. The fuel loading ofthe areas burned using this technique tendto be high. The total area burned underthese circumstances can be very difficultto quantify.

• Isolate fuels. Large logs, snags, deeppockets of duff, sawdust piles, squirrelmiddens, or other fuel concentrations thathave the potential to smolder for longperiods of time can be isolated fromburning. This can be accomplished byseveral techniques including: 1) construct-ing a fireline around the fuels of concern;2) not lighting individual or concentratedfuels; 3) using natural barriers or snow; 4)scattering the fuels; and 5) spraying withfoam or other fire retardant material.Eliminating these fuels from burning isoften faster, safer, and less costly thanmop-up, and allows targeted fuels toremain following the prescribed burn.

• Mosaic burning. Landscapes oftencontain a variety of fuel types that are non-continuous and vary in fuel moisturecontent. Prescribed fire prescriptions andlighting patterns can be assigned to usethis fuel and fuel moisture non-homogene-ity to mimic a natural wildfire and createpatches of burned and non-burned areas orburn only selected fuels. Areas or fuelsthat do not burn do not contribute toemissions. For example, an area may becontinuously ignited during a prescribedfire but because the fuels are not continu-ous, patches within the unit perimeter maynot ignite and burn (figure 8.1). Depres-sional wetlands, swamps, and hardwoodstringers can be excluded by burning whensoil moisture is abundant. Furthermore, ifthe burn prescription calls for low humid-ity and high live fuel moisture, continuousburning in the dead fuels may occur whilethe live fuels exceed the moisture ofextinction. In both cases, the unburnedlive fuels may be available for futureburning in a prescribed or wildland fireduring droughts or dormant seasons.

– 145 –


2. Reduce Fuel Load.

Some or all of the fuel can be permanentlyremoved from the site, biologically decom-posed, and/or prevented from being produced.Overall emissions can be reduced when fuel ispermanently excluded from burning.

• Mechanical removal. Mechanicallyremoving fuels from a site reduces emis-sions proportionally to the amount of fuelremoved. This is a broad category and caninclude such techniques as mechanicalremoval of logging debris from clearcuts,onsite chipping of woody material and/orbrush for offsite utilization, and mechani-cal removal of fuels which may or may notbe followed by offsite burning in a morecontrolled environment. Sometimesmechanical treatments (such as whole-treeharvesting or yarding of unmerchantablematerial [YUM]) may result in sufficienttreatment so that burning is not needed.Mechanical treatments are applicable onlands where this activity is allowable (i.e.,non-wilderness, etc.), supported by anaccess road network, and where there is aneconomic market for disposal of theremoved fuel. This technique is mosteffective in forest fuel types and has somelimited applicability in shrub and grassfuel types. A portion of the emissionreduction gains from this technique maybe offset by increased fossil fuel andparticulate emissions from equipment usedfor harvest, transportation, and disposaloperations. Mechanical treatments maycause undue soil disturbance or compac-tion, stimulate alien plant invasion, removenatural nutrient sources, or impair waterquality.

• Mechanical processing. Mechanicalprocessing of dead and live vegetation into

wood chips or shredded biomass is effec-tive in reducing emissions if the material isremoved from the site or biologicallydecomposed (figure 8.2). If the biomass isspread across the ground as additionallitter fuels, emission reductions are notachieved if the litter is consumed either ina prescribed or wildland fire. Use of thistechnique may eliminate the need to burn.

• Firewood sales. Firewood sales mayresult in sufficient removal of woodydebris making onsite burning unnecessary.This technique is particularly effective forpiled material where the public has easyaccess. This technique is generally appli-cable in forest types with large diameter,woody biomass. The emissions fromwildland fuels when burned for residentialheating are not assessed as wildland fireemissions but as residential heating emis-sions. The impact of these emissions onthe human environment is not attributed towildland fire in the national or stateemissions inventories.

• Biomass for electrical generation.Woody biomass can also be removed andused to provide electricity in regions withcogeneration facilities. Combustionefficiency in electricity production isgreater than open burning and emissionsfrom biomass fuel used offset fossil fuelemissions. Although this method ofreducing fuel loading is cost-effectivewhere there is a market for wood chips,there are significant administrative, logisti-cal, and legal barriers that limit its use.

• Biomass utilization. Woody material canbe used for many miscellaneous purposesincluding pulp for paper, methanol produc-tion, wood pellets, garden bedding, andspecialty forest products. Demand forthese products varies widely from place to


– 146 –

Figure 8.1. Mosaic burning creates patches of burned and unburned areasresulting in reduced emissions.

Figure 8.2. Mechanical processing of biomass.

– 147 –


place and year to year. Biomass utilizationis most applicable in forest and shrubtypes that include large diameter woodybiomass and where fuel density andaccessibility makes biomass utilizationeconomically viable.

• Ungulates. Grazing and browsing livegrassy or brushy fuels by sheep, cattle, orgoats can reduce fuels prior to burning orreduce the burn frequency. Goats willsometimes consume even small, deadwoody biomass. However, ungulates areselective, favoring some plants over others.The cumulative effect of this selectivitycan significantly change plant speciescomposition and long-term ecologicalprocesses on an area, eventually convert-ing grass dominated areas to brush. Onmoderate to steep slopes, high populationsof ungulates contribute to increased soilerosion.

3. Reduce Fuel Production.

Management techniques can be used to shiftspecies composition to vegetation types thatproduce less biomass per acre per year, orproduce biomass that is less likely to burn orburns more efficiently with less smoke.

• Chemical treatments. Broad spectrumand selective herbicides can be used toreduce or remove live vegetation, or alterspecies diversity respectively. This oftenreduces or eliminates the need to use fire.Chemical production and application havetheir own emissions, environmental, andpublic relations problems. A NEPA(National Environmental Policy Act)analysis is generally required prior to anychemical use on public lands and statesoften require similar analyses prior tochemical use on state or private lands.

• Site conversion. Natural site productivitycan be decreased by changing the vegeta-tion composition. For example, frequentground fires in southern pine forests willconvert an understory of flammable shrubs(such as palmetto and gallberry) to openwoodlands with less total fuel but alsowith more grass and herbs. Grass andherbs tend to burn cleaner than shrubs.Total fuel loading can also be reducedthrough conversion to species that are lessproductive.

• Land use change. Changing wildlands toanother land use category may result inelimination of the need to burn. Conver-sion of a wildland site to agriculture or anurbanized use significantly alters theecological structure and function andpresents numerous legal and philisophicalissues. This alternative is probably not anoption on Federally managed lands.

4. Reduce Fuel Consumed.

Emission reductions can be achieved whensignificant amounts of fuel are at or above themoisture of extinction, and therefore unavailablefor combustion. Burning when fuels are wetmay leave significant amounts of fuel in thetreated area only to be burned in the future.This may not result in a real reduction in emis-sions then, but rather a delay of emissions to alater date. Real emission reductions areachieved only if the fuels left behind will bio-logically decompose or be otherwise seques-tered at a time of subsequent burning. Eventhough wet fuels burn less efficientlyand produce greater emissions relative to theamount of fuel consumed, emissions from agiven event are significantly reduced because somuch less fuel is consumed.


– 148 –

In the appropriate fuel types, the ability to targetand burn only the fuels necessary to meetmanagement objectives is one of the mosteffective methods of reducing emissions. Whenthe objective of burning is to reduce wildfirehazard, removal of fine and intermediate diam-eter fuels may be sufficient. The opportunity tolimit large fuel and organic layer consumptioncan significantly reduce emissions.

• High moisture in large woody fuels.Burning when large-diameter woody fuels(3+ inches in diameter or greater) are wetcan result in lower fuel consumption andless smoldering. When large fuels are wetthey will not sustain combustion on theirown and are extinguished by their owninternal moisture once the small twigs andbranch-wood in the area finish burning(figure 8.3). The large logs thereforeconsume less in total, they do not smolderas much, and they do not cause as much ofthe organic layer on the forest floor toburn. This can be a very effective tech-nique for reducing total emissions from a

prescribed burn area and can have second-ary benefits by leaving more large-woodydebris in place for nutrient cycling. Thistechnique can be effective in natural andactivity fuels in forest types. When largefuel consumption is needed, burning underhigh moisture conditions is not a viablealternative.

• Moist litter and/or duff. The organiclayer that forms from decayed and par-tially decayed material on the forest flooroften burns during the inefficient smolder-ing phase. Consequently, reducing theconsumption of this material can be veryeffective at reducing emissions. Con-sumption of this litter and/or duff layer canbe greatly reduced if the material is quitemoist. The surface fuels can be burnedand the organic layer left virtually intact.The appropriate conditions for use of thistechnique generally occurs within a fewdays of a soaking rain or shortly aftersnowmelt. This technique is most effec-tive in non-fire adapted forest and brushtypes. This technique may not be appro-priate in areas where removal of theorganic layer is desired. Burning litterand/or duff to expose mineral soil is oftennecessary in fire adapted ecosystems forplant regeneration.

• Burn before precipitation. Scheduling aprescribed fire before a precipitation eventwill often limit the consumption of largewoody material, snags, stumps, and or-ganic ground matter, thus reducing thepotential for a long smoldering period andreducing the fire average emission factor.Sucessful application of this proceduredepends on accurate meteorologicalforecasts for the area.

Figure 8.3. Burning when large fuel moisture is highcan result in less total fuel consumption.

– 149 –


• Burn before large fuels cure. Livingtrees contain very high internal fuel mois-tures, which take a number of months todry after harvest. If an area can be burnedwithin 3-4 drying months of timber har-vest, many of the large fuels will stillcontain a significant amount of live fuelmoisture. This technique is generallyrestricted to activity-generated fuels inforest-types.

5. Schedule Burning Before New Fuels Appear.

Burning can sometimes be scheduled for timesof the year before new fuels appear. This mayinterfere with land management goals if burningis forced into seasons and moisture conditionswhere increased mortality of desirable speciescan result.

• Burn before litter fall. When decidoustrees and shrubs drop their leaves thisground litter contributes extra volume tothe fuel bed. If burning takes place priorto litter fall there is less available fuel andtherefore less fuel consumed and feweremissions.

• Burn before green-up. Burning in covertypes with a grass and/or herbaceousfuelbed component can produce feweremissions if burning takes place beforethese fuels green-up for the year. Less fuelis available therefore fewer emissions areproduced.

6. Increase Combustion Efficiency.

Increasing combustion efficiency, or shifting themajority of consumption away from the smol-dering phase and into the more efficient flamingphase, reduces emissions.

• Burn piles or windrows. Fuels concen-trated into clean and dry piles or windrowsgenerate greater heat and burn moreefficiently (figure 8.4). A greater amountof the consumption occurs in the flamingphase and the emission factor is lower.This technique is primarily effective inforest fuel types but may have someapplicability in brush types also. Concen-trating fuels into piles or windrows gener-ally requires the use of heavy equipment,which can negatively impact soils andwater quality. Piles and windrows alsocause temperature extremes in the soilsdirectly underneath and can result in areasof soil sterilization. If fuels in piles orwindrows are wet or mixed with dirt,extended smoldering of the debris canresult in residual smoke problems.

• Backing fires. Flaming combustion iscleaner than smoldering combustion. Abacking fire takes advantage of this rela-tionship by causing more fuel consump-tion to take place in the flaming phase than

Figure 8.4. Fuels burned in dry, clean piles burnmore efficiently and generate less emissions


– 150 –

would occur if a heading fire were used(figure 8.5). In applicable vegetation typeswhere fuels are continuous and dry, theflaming front backs more slowly throughthe fuelbed and by the time it passes, mostavailable fuel is consumed so the firequickly dies out with very little smolder-ing. In a heading fire, the flaming frontpasses quickly and the ignited fuels con-tinue to smolder until consumed. Theopportunity to use backing fires is notalways an option and often increaseoperational costs.

• Dry conditions. Burning under dryconditions increases combustion efficiencyand less emissions may be produced.However, dryer conditions makes fuel thatwas not available to burn (at or above themoisture of extinction) available to burn.The emissions from additional fuel burnedgenerally more than offsets emissionreduction advantages gained by greatercombustion efficiency. This technique iseffective only if all fuels will consumeunder either wet or dry conditions.

• Rapid mop-up. Rapidly extinguishing afire can reduce fuel consumption andsmoldering emissions somewhat althoughthis technique is not particularly effectiveat reducing total emissions and can be verycostly (figure 8.6). Rapid mop-up prima-rily effects smoldering consumption oflarge-woody fuels, stumps, snags, andduff. Rapid mop-up is more effective asan avoidance technique by reducingresidual emissions that tend to get caughtin drainage flows and end up in smokesensitive areas.

• Aerial ignition / mass ignition. “Mass”ignition can occur through a combinationof dry fine-fuels and very rapid ignition,which can be achieved through a techniquesuch as a helitorch (figure 8.7). Massignition can shorten the duration of thesmoldering phase of a fire and reduce thetotal amount of fuel consumed. Whenproperly applied, mass ignition causesrapid consumption of dry, surface fuelsand creates a very strong plume or convec-

Figure 8.5. Backing fires in uniform,noncomplex fuelbeds consume fuels moreefficiently than during a head fire resulting infewer emissions.

Figure 8.6. Quickly extinguishing a smolderingfire is a costly but effective technique forreducing smoldering emissions and impacts.

– 151 –


tion column which draws much of the heataway from the fuelbed and prevents dryingand preheating of larger, moister fuels.This strong plume may result in improvedsmoke dispersal. The fire dies out shortlyafter the fine fuels fully consume and thereis little smoldering or consumption of thelarger fuels and duff. The conditionsnecessary to create a true mass ignitionsituation include rapid ignition of a large,open area with continuous, dry fuels (Hall1991).

• Air Curtain Incinerators. Burning fuelsin a large metal container or pit with theaid of a powerful fan-like device to forceadditional oxygen into the combustionprocess results in a very hot and efficientfire that produces little smoke (figure 8.8).These devices are commonly used to burnland clearing, highway right-of-ways, ordemolition debris in areas sensitive tosmoke and may be required by air qualityagency regulations in some areas.

Figure 8.7. Mass ignition can shorten theduration of the smoldering phase and reducetotal consumption resulting in fewer emissions

Redistributing theEmissions

Emissions can be spatially and temporallyredistributed by burning during periods of goodatmospheric dispersion (dilution) and whenprevailing winds will transport smoke awayfrom sensitive areas (avoidance) so that airquality standards are not violated. Redistribu-tion of emissions does not necessarily reduceoverall emissions.

1. Burn when dispersion is good.

Smoke concentrations can be reduced by dilut-ing the smoke through a greater volume of air,either by burning during good dispersion condi-tions when the atmosphere is unstable or burn-ing at slower rates. If burning progresses tooslowly, smoke accumulation due to eveningatmospheric stability can occur.

2. Share the airshed.

Establishing a smoke management program thatlinks both local and interstate jurisdictions willcreate opportunities to share the airshed andreduce the likelihood of smoke impacts.

Figure 8.8. Air curtain incinerators result in veryhot and efficient fires that produce little smoke.


– 152 –

3. Avoid sensitive areas.

The most obvious way to avoid smoke impactsis to burn when the wind is blowing away fromall smoke-sensitive areas such as highways,airports, populated areas, and scenic vistas.Wind direction must be considered during allphases of burning. For example, the prevailingwinds during the day time may move the smokeaway from a major highway; however, at night,drainage winds can carry the smoke toward thehighway.

4. Burn smaller units.

Short term emissions and impacts can be re-duced by burning subsets of a large unit overmultiple days. Total emissions are not reducedif the entire area is eventually burned.

5. Burn more frequently.

Burning more frequently does not allow fuels toaccumulate, thus there are less emissions witheach burn. Frequent, low intensity fires canprevent unwanted vegetation from becomingestablished. If longer fire rotations are used, thevegetation has time to grow resulting in theproduction of extra biomass and extra fuelloading at the time of burning. This techniquegenerally has positive effects on land manage-ment goals since it results in fire regimes thatmore closely mimic the frequency of natural firein many ecosystems.

The Use and Effectiveness ofEmission Reduction andRedistribution Techniques

The overall potential for emission reductionsfrom prescribed fire depends on the frequencyof use of emission reduction techniques and the

amount of emission reduction that each methodoffers. This section provides information on theoverall potential for emission reduction andredistribution from prescribed fire based on (a)the frequency of use of each emission reductionand emission redistribution technique by regionof the country, (b) the relative effectiveness ofeach smoke management technique, and (c)constraints on application of the technique(administrative, legal, physical, etc.).

Much of the information in this section wasprovided by participants in regional workshops(as described previously). The informationprovided can, and should, be improved upon bylocal managers who will have better informationabout specific, local burning situations.

The use of each smoke management techniqueis organized by U.S. region as shown in figure8.9. They are the Pacific Northwest includingAlaska (PNW), Interior West (INT), Southwest(SW), Northeast (NE), Midwest (MW), andSoutheast including Hawaii (SE) regions. Eachregion has its own vegetation cover types,climatology, and terrain characteristics, all ofwhich influence the land manager’s decision toburn and the appropriateness of various emis-sion reduction techniques.

Manager use of emission reduction techniques isinfluenced by numerous factors including landmanagement objectives, the type and amount ofvegetation being burned, safety considerations,costs, laws and regulations, geography, etc. Theeffect of some of these many influencing factorscan be assessed through general knowledge ofthe frequency of use of a particular technique ina specific region. Table 8.1 provides generalinformation about frequency of use of eachsmoke management technique by region of thecountry, grouped as shown in figure 8.9.

– 153 –


Figure 8.9. Prescribed burning regions including Pacific Northwest including Alaska (PNW), Intermountain(INT), Midwest (MW), Southwest (SW), Southeast including Hawaii (SE), and Northeast (NE).

Information in table 8.1 summarizes regionalapplicability of each of the twenty-nine smokemanagement methods. Interviews with firepractitioners demonstrate that, on a nationalscale, several smoke management techniquesare rarely used. These include biomass forelectrical generation, biomass utilization, siteconversion, land use change, burning beforelitter fall, burning under dry conditions, aircurtain incineration, and burning smaller units.In most of the regions, firewood sales andchemical treatments are also seldom used. Themethods most commonly applied include aerialignition/mass ignition, burning when dispersionis good, sharing the airshed, and avoidingsensitive areas.

The general effectiveness of the emissionreduction and redistribution techniques isdescribed in table 8.2 based on input frommanagers at the workshops. Local managerswill have better information about specificsituations and can improve upon the informa-tion in the tables. Each technique was assigneda general rank of “High” for those techniquesmost effective at reducing emissions or “Low”for those techniques that are less effective.Some emission reduction techniques also havesecondary benefits of delaying or eliminatingthe need to use prescribed fire. Some smokemanagement techniques, are also effective forreducing local smoke impacts if they promoteplume rise or decrease the amount of residual


– 154 –

Table 8.1. Frequency of smoke management method use by region. Alaska is included in the PacificNorthwest (PNW) region, and Hawaii is included in the southeast region (SE)

– 155 –


Table 8.2. Relative effectiveness of various smoke management techniques.


– 156 –

smoldering combustion where smoke is morelikely to get caught in drainage winds andcarried into populated areas. These factors arealso addressed in table 8.2.

Table 8.3 summarizes significant constraintsidentified by fire managers that limit the widerapplication of techniques to reduce and redis-tribute emissions. This table excludes consider-ation of the objective of the burn, which isgenerally the overriding constraint. Some of thetechniques would probably be used more fre-quently if specific constraints could be over-come.

Smoke management techniques that, in theopinion of workshop participants, show particu-lar promise for wider use in the future are listedbelow:

1. Mosaic Burning: Since this methodreduces the area burned and replicates thenatural role of fire, it is being increasinglyused for forest health restoration burningon a landscape scale.

2. Mechanical Removal: In areas whereslope and access are not a problem andfuels have economic value, the wider useof whole tree yarding, YUM yarding, cut-to-length logging practices and othermethods that remove fuel from the unitprior to burning (if the unit is burned atall) may have potential for wider applica-tion if economic markets for the removedfuels can be found.

3. High Moisture in Large Woody Fuels,and/or Moist Litter and Duff: In situa-tions where the objective is not to maxi-mize the consumption of large woodydebris, litter, and/or duff, this option isfavored by fire practitioners as an effectivemeans of reducing emissions, smolderingcombustion, and smoke impacts.

4. Pile and Windrow Burning: Pile burn-ing, although already widely used in allregions, is gaining popularity among landmanagers because of the flexibility offeredin scheduling burning and the resultantlower impacts on smoke sensitive loca-tions. Lower impacts may not result ifpiles or windrows are wet or mixed withdirt.

5. Aerial/Mass Ignition: Little clear infor-mation currently exists as to the extent towhich aerial ignition achieves true massignition and associated emission reductionbenefits. More effort to achieve true massignition using aerial techniques may yieldsignificant emission reduction benefits.

6. Burn More Frequently: Fire managersgenerally favor more frequent burningpractices to reduce fuel loading on secondand subsequent entry, thereby reducingemissions over long time periods. Thiswill increase daily or seasonal emissions.

Estimated Emission Reductions

While the qualitative assessment of emissionreduction technique effectiveness shown in table8.2 is a useful way to gauge how relativelysuccessful a particular technique may be inreducing emissions, it is also useful to modelpotential quantitative emission reduction. Table8.4 summarizes potential emission reductionsthat may be achieved by employing varioustechniques as estimated by the fuel consumptionand emissions model Consume 2.1 (Ottmar andothers [in preparation]). For example, use ofmosaic burning techniques in natural, mixedconifer fuels in which one-half of a 200-acreproject is burned is projected to reduce PM2.5

emissions from 14.8 to 7.4 tons for a 50%reduction in emissions. A 33% reduction in

– 157 –


Table 8.3. Constraints to the use of emission reduction and redistribution techniques as reported by regionalworkshop participants.


– 158 –

Tabl

e 8.

4. A

ppro

xim

ate

emis

sion

red

uctio

n ef

fect

iven

ess

for

vari

ous

emis

sion

red

uctio

n te

chni

ques

in c

erta

in v

eget

atio

n ty

pes.

(V

alue

s ge

nera

ted

with

Con

sum

e 2.

1 [O

ttmar

and

oth

ers,

in p

repa

ratio

n]).

– 159 –


PM2.5 emissions can be achieved by pile burn-ing mixed conifer fuels under the conditionsnoted in the table. Specific simplifying assump-tions were made in each case to produce theestimates of emission reduction potential seen intable 8.4. Other models using the same fieldassumptions would yield similar trends.

Wildfire Emission Reduction

Little thought has been given to reducing emis-sions from wildfire, but many fire managementactions do affect emission production fromwildfires because they intentionally reducewildfire occurrence, extent, or severity. Forexample, fire prevention efforts, aggressivesuppression actions, and fuel treatments (me-chanical or prescribed fire) all reduce emissionsfrom wildfires. Although fire suppressionefforts may only delay the emissions rather theneliminate them altogether. Allowing fires toburn without suppression early in the fire seasonto prevent more severe fires in drier periodswould reduce fuel consumption and reduceemissions. All fire management plans that allowlimited suppression consider air quality impactsfrom potential wildfires as a decision criterion.So, although only specific emission reductiontechniques for prescribed fires are discussed inthis chapter, we should remember that there isan inextricable link between fuels management,prescribed fire, wildfire severity, and emissionproduction.

Literature CitationsEnvironmental Protection Agency. Sept. 1992.

Prescribed burning background document andtechnical information document for prescribedburning best available control methods. U.S.EPA, Office of Air Quality Planning andStandards. Research Triangle Park, NC. EPA-450/2-92-003.

Hall, Janet. 1991. Comparison of fuel consumptionbetween high and moderate intensity fires inlogging slash. Northwest Science. 64(4): 158-165.

Ottmar, Roger D.; Reinhardt, Timothy E.; Anderson,Gary, DeHerrera, Paul J. [In preparation].Consume 2.1 User’s guide. Gen. Tech. Rep.PNW-GTRxxx. Portland, OR: U.S. Departmentof Agriculture, Forest Service, Pacific North-west Research Station.

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 ofAgriculture, Forest Service, IntermountainResearch Station. 65 p.

Sandberg, D.V.; Peterson, J. 1984. A source strengthmodel for prescribed fires in coniferous loggingslash. Presented at the 1984 Annual Meeting,Air Pollution Control Association, PacificNorthwest Section, 1984 November 12.

Ward, Darold E.; Hardy, Colin C. 1991. Smokeemissions from wildland fires. EnvironmentalInternational, Vol 17:117-134.

Ward, Darold E.; Peterson, Janice; Hao, Wei Min.1993. An Inventory of Particulate Matter andAir Toxic Emissions from Prescribed Fires inthe USA for 1989. In: Proceedings of the 86thAnnual Meeting of the Air and Waste Manage-ment Association; 1993 June 13-18; Denver,CO. 19p.


– 160 –

– 161 –

2001 Smoke Management Guide 9.0 – Dispersion Prediction

DISPERSION PREDICTION SYSTEMS

Chapter 9

Chapter 9 – Dispersion Prediction Systems 2001 Smoke Management Guide

– 162 –

– 163 –


Smoke Dispersion Prediction Systems

Sue A. Ferguson

Smoke dispersion prediction systems are be-coming increasingly valuable tools in smokemanagement. There are a variety of potentialapplications that can help current managementissues. These include screening, where meth-ods and models are used to develop “worst-case” scenarios that help determine ifalternative burn plans are warranted or if morein-depth modeling is required. Such tools alsohelp in planning, where dispersion predictionsaid in visualizing what fuel and weather condi-tions are best suited for burning or when sup-porting data are needed to report potentialenvironmental impacts. Also, prediction sys-tems can be used as communication aids to helpdescribe potential impacts to clients and manag-ers. For regulating, some states use dispersionprediction systems to help determine approvalof burn permits, especially if ignition patternsor fuel complexes are unusual. Other statesrequire dispersion model output in each burnpermit application as supporting proof that aburn activity will not violate clean air thresh-olds.

There are a variety of tools that can be appliedto screening and some planning applications.The easiest of these are simple approximationsof dispersion potential, emission production,and proximity to sensitive receptors. Theapproximations are based on common experi-ence with threshold criteria that consider worst-case conditions or regulatory requirements.More detailed planning and many regulatorysituations require numerical modeling tech-

niques. While numerical models output acalculated physical approximation of dispersionfeatures, they can be adjusted to predict worst-case scenarios by altering such things as emis-sion production or trajectory winds. Often theeasily applied numerical models are used forscreening. Typically, more rigorous applicationsrequire the use of complex models by trainedpersonnel.

Methods of Approximation

A first level of approximation can simply deter-mine whether the atmosphere has the capacity toeffectively disperse smoke by using indexes ofventilation or dispersion. These indexes arebecoming widely used and may be a regularfeature of fire weather or air quality forecasts inyour area. Usually the ventilation index is aproduct of the mixing height times the averagewind within the mixed layer. For example, amixing height of 600 meters (~2,000 feet) abovethe ground surface with average winds of 4 m/s(~7.8 knots or ~8.9 mph) produces a ventilationindex of 2,400 m2/sec (~15,600 knots-feet).With similar wind speeds, the ventilation indexwould increase to 12,000 m2/sec (~78,000knots-feet) if the mixing height rose to 3,000meters (~10,000 feet). Ventilation indexescalculated from model output may use theproduct of the planetary boundary layer (PBL)and lowest level winds (e.g., 10 to 40 metersabove ground level). Others calculate the index


– 164 –

___________________________________

1 Transport winds are those considered most likely to carry smoke away from a fire, usually near mid-level of thehorizontal portion of a spreading plume.

Table 9.1. Common values of the ventilation index (VI) and associated smoke conditions.The Index is calculated by multiplying mixing height (MH) or planetary boundary layer (PBL)times trajectory winds (Traj.), average winds through the depth of the mixed layer (Avg.), orwinds at 40 meters above ground level (40m).

by multiplying the mixing height by a deter-mined transport wind speed,1 which might benear the top of the mixed layer. Because ofdifferent methods of calculating ventilationindex, the scales used for burning recommenda-tions may vary.

It helps to gain experience with a ventilationindex before making management decisionsbased on its value. Defining a uniform methodfor calculating the index and comparing itfrequently with observed smoke dispersalconditions can do this. Ferguson et al. (2001)developed a national historical database ofventilation index based on model generated 10-meter winds and interpolated mixing heightobservations. It is useful in illustrating thespatial and temporal variability of potentialventilation all across the country. In SouthCarolina the index is divided into 5 categoriesthat correspond to specific prescribed burningrecommendations, where no burning is recom-mended if the index is less than 4,500 m2/sec(28,999 knots-feet) and restrictions apply if it isbetween 4,500 and 7,000 m2/sec (29,000-49,999 knots-feet) (South Carolina Forestry

Commission, 1996). In Utah the ventilationindex is referred to as a “clearing index” and isdefined as the mixing depth in feet times theaverage wind in knots divided by 100. In thisway, a clearing index of less than 200 wouldindicate poor dispersion and likely pollution; anindex between 200 and 500 indicates fair disper-sion, while indexes greater than 500 representgood to excellent dispersion. Commonly, theclearing index must be greater than 400 beforeburning is recommended. In the northwesternU.S., where a mesoscale weather model is usedto predict ventilation index, the South Carolinascale has been slightly adjusted to match localburning habits and to accommodate for theslightly different way of computing the index.Table 9.1 gives common values of the ventila-tion index (VI) and associated smoke condi-tions.

Ventilation indexes have no value when there isno mixing height, which is common at night.Also, if the atmosphere is very stable within themixed layer, the ventilation index may be toooptimistic about the ultimate potential of dis-persing a smoke plume. Therefore, to help

– 165 –


Table 9.2. Atmospheric Dispersion Index (ADI) with its current interpretation(Lavdas 1986).

determine the atmosphere’s capacity to dispersesmoke during all atmospheric conditions,Lavdas (1986) developed an AtmosphericDispersion Index (ADI) that combines Pasquill’sstability classes (see table 7.1) and ventilationindexes with a simple dispersion model. Na-tional Weather Service (NWS) fire weatheroffices are beginning to include the ADI as aregular part of their smoke management fore-cast. See table 9.2 for an explanation of the ADIcategories. Commonly the ADI must be greaterthan 30 before burning is recommended.

Another way to approximate smoke impacts isthrough a geometric screening process that isoutlined in “A Guide for Prescribed Fire inSouthern Forests” (Wade 1989) and “SouthernForestry Smoke Management Guidebook”(USDA-Forest Service, Southern Forest Experi-ment Station 1976). The recommended stepsinclude: 1) plotting the direction of the smokeplume, 2) identifying common areas of smokesensitivity (receptors) such as airports, high-ways, hospitals, wildernesses, schools, and

residential areas, 3) identifying critical areasthat already have an air pollution or visibilityproblem (non-attainment areas), 4) estimatingsmoke production, and 5) minimizing risk.

It is suggested that the direction of the smokeplume during the day be estimated by consider-ing the size of the fire and assuming a dispersionof 30° on either side of the centerline trajectoryif wind direction is planned or measured and45° if forecasted winds are used. At night, theguide suggests that smoke follows down-valleywinds and spreads out to cover valley bottoms.Fuel type, condition, and loading are used tohelp estimate the amount of smoke that will beproduced. In minimizing risk, it is suggested toconsider mixing height, transport wind speed,background visibility, dispersion index, andvarious methods of altering ignition and mop-uppatterns.

Because the guidebooks for southern forestryestimate emissions based on fuel types specificto the southeastern U.S., other methods of


– 166 –

estimating emissions are needed to employgeometric screening applications elsewhere.Existing models such as FOFEM (Reinhardt andothers 1997) and CONSUME (Ottmar andothers 1993) are designed for this purpose.

Schaaf and others (1999) describe a similarscreening process for deciding the level ofanalysis for each project. The screening stepsinclude: 1) determining fire size, 2) estimatingfuel load, 3) identifying distance to sensitiveareas, and 4) calculating emission production.Unlike the southern forestry screening method,which estimates downwind impacts from simplegeometry, Schaaf and others (1999) recommendrunning a numerical dispersion model to helpcalculate smoke concentrations if initial screen-ing thresholds are met. Further analysis orefforts to reduce potential impacts are thenrecommended only if predicted concentrationsexceed specified standards.

Before relying on simple screening methods todetermine if additional modeling may be re-quired or if alternatives are necessary, it ishelpful to define appropriate threshold criteriaby consulting regulations, surrounding commu-nity opinions, and management concerns. Forexample, the criteria of sensitive receptor prox-imity may range from fractions of a mile toseveral miles. On the other hand, some placesmay base criteria on total tonnage of emissions,no matter how close or far from a sensitive area.Most often criteria are combinations of proxim-ity to receptors and fire size, which vary fromplace to place.

Numerical Models

Most of the available dispersion predictionsystems are in the form of deterministic numeri-cal models and there are three types designed toestimate the timing and location of pollutant

concentrations; dispersion, box, and three-dimensional grid models. Dispersion modelsare used to estimate smoke and gas concentra-tions along the trajectory of a smoke plume.Box models do not calculate trajectories ofparticles but assume smoke fills a box, such as aconfined basin or valley, and concentrationsvary over time as smoke enters and leaves thebox. Grid models are like expanded box modelsin that every grid cell acts as a confined box.Because trajectories are not explicitly computed,box or grid models may include other enhance-ments, such as complex computations of chemi-cal interactions. Currently, only dispersion andbox models have been adapted for wildlandsmoke management applications. Work isunderway to adapt grid models to smoke prob-lems and this will help in estimates of regionalhaze because grid models can simulate largedomains and usually include critical photo-chemical interactions. The following summaryof numerical models currently used by smokemanagers is updated from an earlier review byBreyfogle and Ferguson (1996).

Dispersion Models – Dispersion models tracktrajectories of individual particles or assume apattern of diffusion to simplify trajectory calcu-lations. Particle models typically are the mostaccurate way to determine smoke trajectories.They are labor intensive, however, and moreoften used when minute changes in concentra-tions are critical, such as when nuclear or toxiccomponents exist, or when flow conditions arewell bounded or of limited extent (e.g., PB-Piedmont by Achtemeier 1994, 1999, 2000).Diffusion models commonly assume that con-centrations crosswind of the plume disperse in abell-shape (Gaussian) distribution pattern. Bothplume (figure 9.1a) and puff (figure 9.1b)patterns are modeled. The plume methodassumes that the smoke travels in a straight lineunder steady-state conditions (the speed anddirection of particles do not change during theperiod of model simulation). SASEM (Sestak

– 167 –


and Riebau 1988), VSMOKE (Lavdas, 1996),and VSMOKE-GIS (Harms and Lavdas 1997)are examples of plume models. Plume modelsmost commonly are applied in regions of flat orgently rolling terrain but can be used whenever aplume is expected to rise above the influence ofunderlying terrain. The puff method simulates acontinuous plume by rapidly generating a seriesof puffs (e.g., NFSpuff: Harrison 1995; Citpuff:in TSARS+ by Hummel and Rafsnider 1995;and CALPUFF: Scire and others 2000a).Therefore, like particle models, puff models canbe used at times when trajectory winds change,such as during changeable weather conditions orin regions where underlying terrain controlssmoke trajectory patterns. Because particletrajectory models and Gaussian diffusion mod-els use coordinate systems that essentiallyfollow particles/parcels as they move(Lagrangian coordinates), sometimes they arereferred to as Lagrangian dispersion models.

Particle and puff models must have high spatialand temporal resolution weather data to modelchanging dispersion patterns. This requires atleast hourly weather information at spatialresolutions that capture important terrain fea-tures (usually less than 1km). For this reason,particle and puff models currently used forsmoke management include a weather modulethat scales observations or input from externalmeteorological information, to appropriatespatial and temporal resolutions. For example,TSARS+ is designed to link with the meteoro-logical model NUATMOS (Ross and others1988) while CALPUFF is linked to CALMET(Scire and others 2000b). NFSpuff (Harrison1995) and PB-Piedmont (Achtemeier 1994,1999, 2000) contain internal algorithms that aresimilar to CALMET and NUATMOS. Mostweather modules that are attached to particleand puff models solve equations that conservemass around terrain obstacles and some haveadditional features that estimate diurnal slopewinds and breezes associated with lakes and

oceans at very fine scales.

Unlike most particle or puff models, plumemodels assume that mixing heights and trajec-tory winds are constant for the duration of theburn. Therefore, they do not require detailedweather inputs and are very useful when meteo-rological information is scarce. Plume models,however, will not identify changing trajectoriesor related concentrations if weather conditionsfluctuate during a burn period. Also, whensmoke extends beyond a distance that is reason-able for steady-state assumptions, which typi-cally is about 50 km (30 miles), plumeapproximations become invalid. When terrainor water bodies interact with the plume, steady-state assumptions become difficult to justify, nomatter how close to the source. Despite thelimitation of plume models in complex terrain,they can be useful if plumes are expected to riseabove the influence of terrain or if plumes areconfined in a straight line that follows a widevalley when dispersion does not extend beyondthe valley walls.

Box and Grid Models – The box method ofestimating smoke concentrations assumesinstantaneous mixing within a confined area,such as a confined basin or valley (figure 9.1c).This type of model usually is restricted toweather conditions that include low windspeeds and a strong temperature inversion thatconfines the mixing height to within valleywalls (e.g., Sestak and others, unpublished;Lavdas 1982). The valley walls, valley bottom,and top of the inversion layer define the boxedges. The end segments of each box typicallycoincide with terrain features of the valley, likea turn or sudden elevation change. Flow isassumed to be down-valley and smoke isassumed to instantaneously fill each box seg-ment. The coordinates used to calculate boxdispersions are fixed in space and time and thuscalled Eulerian coordinates. The box methodprovides a useful alternative to Gaussian diffu-


– 168 –

Figure 9.1. Schematic diagrams of numerical dispersion models; (A) Gaussian plume, (B) Gaussianpuff, and (C) box.

A.

B.

C.

valley wall

valley bottom

inversion height

valley wall

down-valley flow

– 169 –


sion models when understanding patterns ofsmoke concentrations in an isolated valleybecome critical.

Many grid models are called Eulerian gridsbecause of their fixed coordinate system. Thefixed coordinates make it difficult for gridmodels to track the impact of individual plumesbut allows for easier evaluation of cumulativeimpacts from several plumes or chemical inter-actions of particles and gases within plumes.This makes grid models especially useful forevaluating the impact of smoke on regionalhaze. Work is underway to adapt at least twogrid models (REMSAD: Systems ApplicationsInternational 1998; and CMAQ: Byun andChing 1999) for wildland fire applications.REMSAD has very simple chemistry thus isdesirable for use in large domains or over longtime periods. The CMAQ model is more fullyphysical and part of the EPA’s Models 3 project,which is a “one-atmosphere” air quality model-ing framework designed to evaluate all potentialimpacts from all known sources. At this timegrid models require experienced modelers toinitialize and run. Smoke managers, however,may be asked to provide input for grid modelsand could begin seeing results that influenceapplication of regional haze rules.

Uncertainty

All prediction systems include some level ofuncertainty, which may occur from the meteoro-logical inputs, diffusion assumptions, plumedynamics, or emission production. Manydispersion models and methods have beencompared to observations of plumes from pointsources, such as industrial stacks, or tightlycontrolled experiments (e.g., Achtemeier 2000).In these cases, the greatest error usually occursbecause of inaccuracies in the weather inputs;either from a poor forecast or an insufficient

number of data points. If trajectories can bedetermined correctly then dispersion and result-ing down-wind concentrations from pointsources are relatively straightforward calcula-tions. This is because emission rates and subse-quent energy transmitted to the plume fromindustrial stacks, or controlled experiments,usually are constant and can be known exactly.

It is expected that the largest source of uncer-tainty in modeling smoke concentrations fromwildland fires is in estimating the magnitudeand rate of emissions. Highly variable ignitionpatterns and the condition and distribution offuels in wildland fires create complex patternsof source strength. This causes plumes withsimultaneous or alternating buoyant and non-buoyant parts, multiple plumes, and emissionrates that are dependent on fuel availability andmoisture content. Few comparisons of observa-tions from real wildland fires to dispersionmodel output are available. Those that do existare qualitative in nature and from the activephase of broadcast-slash burns (e.g., Hardy andothers 1993), which tend to generate relativelywell-behaved plumes.

To calculate the complex nature of sourcestrength, components of heat and fuel (particleand gas species) must be known. For simulatingwildland fires, additional information is requiredon: 1) the pattern of ignition, 2) fuel moisture bysize of fuel, 3) fuel loading by size, 4) fueldistribution, and 5) local weather that influencescombustion rates. Much of this information isroutinely gathered when developing burn plans.Peterson (1987) noted that 83% of the error incalculating emissions is due to inaccurate fuelload values. Therefore, even the best burn plandata will introduce a large amount of uncertaintyin predicted dispersion patterns.

The shift from burning harvest slash to usingfire in natural fuel complexes for understoryrenovation and stand replacements has intro-


– 170 –

duced another degree of uncertainty by theexistence of decaying fuel and isolated concen-trations of deep duff that have previously beenneglected in pre-burn inventories. This hasprevented emission models from accuratelyestimating the contribution of smolderingcombustion, which is common in the porouselements of rotten wood and deep duff. Untilthis omission is corrected, users must manipu-late source-strength models into expectingsmoldering by inputting very long ignitionperiods and low fuel loads, which simulate theindependent smoldering combustion that occursin porous material.

Currently variable-rate emissions are deter-mined by approximating steady-state conditionsin relatively homogeneous burning segments ofa fire (e.g., Sandberg and Peterson 1984;Ferguson and Hardy 1994; Lavdas 1996; Sestakand Riebau 1988) or by allowing individual fuelelements to control combustion rates (e.g.,Albini and others 1995; Albini and Reinhardt1995; Albini and Reinhardt 1997). The steady-state method has been adapted for many of thecurrently available puff, plume, and box modelsand is most useful when the pattern and durationof ignition are known ahead of time, eitherthrough planning or prediction. The fuel-element approach shows promise for calculatingemissions simultaneously with ignition rates(fire spread) and may become particularly usefulfor coupled fire-atmosphere-smoke models,which currently are being developed.

Principal components (plume rise, trajectory,and diffusion) of all numerical dispersionmodels assume functions that are consistentwith standard, EPA approved, industrial stackemission models. The models themselves,however, may or may not have passed an EPAapproval process. Primary differences in thephysics between the models appear to be thedegree to which they fully derive equations. Allmodels include some empirical coefficients,

approximations, or parameterized equationswhen insufficient input data are expected orwhen faster computations are desired. Thedegree to which this is done varies betweenmodels and between components of each model.Note that it is not clear whether fully physicalcalculations of plume rise and dispersion aremore accurate than approximate calculations inbiomass burning because of the considerableuncertainty in the distribution and magnitude ofavailable fuels in wildland areas.

Output

Useful output products for smoke managers arethose that relate to regulatory standards, showimpact to sensitive receptors, and illustratepatterns of potential impact. Regulatory stan-dards require 24-hour averaged and 24-hourmaximum surface concentrations of respirableparticles at sensitive receptors. In addition,surface concentrations of carbon monoxide(CO), lead, sulfur oxides (SOx), ozone (O

3),

nitrous oxides (NOx), and hydrocarbons (e.g.,methane, ethane, acetylene, propene, butanes,benzene, toluene, isoprene) are needed toconform to health regulations. Quantifying theimpact on regional haze is becoming necessary,which requires an estimate of fine particles,carbon gases, NOx, O

3, relative humidity, and

background concentrations. Safety consider-ations require estimates of visibility, especiallyalong roads (Achtemeier et al. 1998) and atairports. In addition to quantitative output, it ishelpful to map information for demonstratingthe areal extent of potential impact because eventhe smallest amount of smoke can affect humanvalues, especially when people with respiratoryor heart problems are in its path. For example,studies have shown that only 30 to 60 µg/m3 indaily averaged PM10 (particulate matter that isless than 10 micrometers in diameter) can causeincreases in hospital visits for asthma (Schwartz

– 171 –


et al. 1993; Lipsett et al. 1997). These valuesare less than 1/3 of the national ambient airquality standard (U.S. Environmental ProtectionAgency 1997). Sometimes the mere presence ofsmoke, regardless of its concentration, is enoughto force alteration of a burn plan.

The old adage, “you can’t get out what youdon’t put in,” aptly describes the output ofdispersion prediction systems. In a geometricscreening system (Wade 1989), only place ofimpact can be approximated because elementalconstituents of the source emissions are notconsidered. The value in screening processes ofthis type, however, is that they allow an objec-tive, first-guess estimate of smoke impacts soalternative measures can be taken if needed.Also, the process can be done on a map thatillustrates potential receptors and estimatedtrajectory for others to see and discuss. De-pending on the state or tribal implementationplan, a geometric screening may be all that isneeded to conform to regulatory standards.

Numerical models disperse gases and particu-lates that are available from a source-strengthmodel, which uses measured ratios of emissionsto amount of fuel consumed (emission factors).Emission factors vary depending on fuel type,type of fire (e.g., broadcast slash, pile, or undis-turbed) and phase of the fire (e.g., flaming orsmoldering). Currently, emission factorsavailable for wildland fire include total particu-late matter (PM), particulate matter that is lessthan 10 micrometers (µm) in diameter (PM10),particulate matter that is less than 2.5 µm indiameter (PM2.5), carbon monoxide (CO),carbon dioxide (CO

2), methane (CH

4), and non-

methane hydrocarbons (NMHC). Emissionfactor tables (AP-42) are maintained by the U.S.Environmental Protection Agency (1995).

At this time, emissions of lead and SOx frombiomass fires are considered negligible. Emis-sion factors of NOx are uncertain and have not

been quantified to a satisfactory level. It isassumed that ozone is not created at the sourcebut develops downwind of the source as theplume is impacted by solar radiation. Currently,aside from grid models, only one dispersionmodel (CALPUFF: Scire and others 2000a)includes simple photochemical reactions forcalculation of down-wind ozone.

Desired attributes within a dispersion predictionsystem vary in complexity by several orders ofmagnitude. To help potential users determinewhich systems may best apply to their specificneed, three levels of complexity were estimatedfor each desired attribute as shown in table 9.3.The 1st level is the simplest; usually producinggeneralized approximations. At the 3rd level,attributes are determined with the best availablescience and often include a number of perspec-tives or options for output.

Using the estimated levels of complexity fromtable 9.3, it becomes possible to rank dispersionprediction systems for each potential applica-tion. For example, if graphical output is avail-able, the location of impact can be determined.If surface concentrations of particles and gasesare available, then the system can be used todetermine health and visibility impacts. A quickestimate of visibility may require only a 1st levelof complexity, while precise visibility determi-nations may require more complex approaches.A summary of attributes for each dispersionprediction system is provided in table 9.4. Thenumbers in the attribute columns refer to anestimated level of complexity from 1 to 3 assummarized in table 9.3. Ease of use is asubjective determination based on the work ofBreyfogle and Ferguson (1996). It considers thenumber and type of inputs, the availability ofinputs, required user knowledge, and effortneeded to produce useful results. Becausecalculating a ventilation or clearing index issimply a product of two numbers, dispersionindexes typically are computed by others, and


– 172 –

Table 9.3. Desired attributes of dispersion prediction systems are compared to estimated levelsof complexity.

both commonly are available through fireweather or air quality forecasts, they are consid-ered very easy to use.

Several methods/models can show cumulativeimpacts from a number of fires by generalizingthe atmosphere’s capacity to hold the totalemissions (index values) or by displayingmultiple plumes at once (VSmoke-GIS if sepa-rate projects are used as overlays, NFSpuff,TSARS+, and CALPUFF). The ability tonumerically determine the cumulative impact,however, requires concentrations of intersectingplumes to be added together. Currently

CALPUFF (Scire and others 2000a) is capableof additive concentrations.

Only two of the currently available models arespecific to a geographic area. They are NFSpuff(Harrison 1995) and PB-Piedmont (Achtemeier1994, 1999, 2000) that were built for ultimateease by including digital elevation data so theuser would not have to find it or adjust fordifferent formats. Early versions of the NFSpuffmodel contain only elevation data from Wash-ington and Oregon while later versions includeall of the western states. The PB-Peidmontmodel includes data for the piedmont regions of

– 173 –


Table 9.4. Dispersion prediction systems designed for wildland fire applications. Attributes are rankedby their level of complexity, with 1 being simplest and 3 being most complex, where a dash indicatesthat the attribute is unavailable. Ease of use is ranked from 1 being the easiest to 10 being the mostdifficult.

southeastern United States. Other models donot require elevation data (e.g., SASEM andVSmoke) or allow the input of elevation datafrom anywhere as long as it fits the model-specified format (e.g., VSmoke-GIS, TSARS+,and CALPUFF). While there is some concernthat version 1.02 of the Emission ProductionModel (EPM: Sandberg and Peterson 1984) isspecific to vegetation types in Washington andOregon, it has been adapted for use in thesoutheastern U.S. through VSmoke (Lavdas1986) and can be adjusted to function elsewherein the country (e.g., SASEM: Sestak and Riebau1988). Newer versions of EPM (Sandberg2000) and the BurnUp emissions model (Albiniand Reinhardt 1997) are not geocentric but todate neither has been incorporated into anyavailable dispersion prediction system.

Summary

For many projects a simple model often pro-vides as good information as a more complexmodel. Regulations, however, may dictate thelevel of modeling required for each project.Other times, community values will determinethe level of effort needed to demonstrate com-pliance or alternatives. Also, skills available toset up and run models or the availability ofrequired input data may affect whether a predic-tion system is necessary and which one is mostappropriate.

Because regulations vary from state to state andtribe to tribe and because expectations varyfrom project to project there is no simple way todetermine what dispersion prediction system isbest. It is hoped that the information in tables9.3 and 9.4 can be used to help assess the valueof available methods and models. For ex-ample, if a simple indication of visibility im-pacts is required, plume models can be used orvisual indexes can be approximated from


– 174 –

concentrations out of box, plume, or puffmodels. If more detailed visibility impacts arerequired, a sophisticated puff model should beused. Whatever the situation, whether smokedispersion prediction systems are used forscreening, planning, regulating, or simply gameplaying, it is helpful to remember theirstrengths and weaknesses.

Literature CitationsAchtemeier, G.L. 1994. A computer wind model

for predicting smoke movement. South. J.Appl. Forest. 18: 60-64.

Achtemeier, G.L. 1999. Smoke Management:Toward a Data Base to Validate PB-Piedmont -Numerical Simulation of Smoke on the Groundat Night Proceedings of the 1999 TAPPIInternatinal Enviromental Conference, April18-21, 1999, Nashville, Tennessee, OprylandHotel, Vol. 2.

Achtemeier, G.L. 2000. PB-Piedmont: A numericalmodel for predicting the movement of biologi-cal material near the ground at night. In:Proceedings of the 24th conference on agricul-tural and forest meteorology; 14-18 August2000; Davis, CA. Boston, MA: AmericanMeteorology Society: 178-179.

Achtemeier, G.L., W. Jackson, B. Hawkins, D.D.Wade, and C. McMahon. 1998. The smokedilemma: a head-on collision! Transactions ofthe 63rd North American Wildlife and NaturalResource Conference. 20-24 March 1998.Orlando FL. 415-421.

Albini, F.A., J.K. Brown, E.D. Reinhardt, and R.D.Ottmar. 1995. Calibration of a large fuelburnout model. Int. J. Wildland Fire. 5(3):173-192.

Albini, F.A. and E.D. Reinhardt. 1995. Modelingignition and burning rate of large woodynatural fuels. Int. J. Wildland Fire. 5(2): 81-91.

Albini, F.A. and E.D. Reinhardt. 1997. Improvedcalibration of a large fuel burnout model. Int.J. Wildland Fire. 7(2):21-28.

Breyfogle, Steve and S. A. Ferguson. 1996.Userassessment of smoke-dispersion models forwildland biomass burning. U.S. Department ofAgriculture, Forest Service, Pacific NorthwestResearch Station, Portland, OR. PNW-GTR-379. 30 pp.

Byun, D.W. and Ching, J.K.S., editors. 1999.Science Algorithms of the EPA Models-3Community Multiscale Air Quality (CMAQ)Modeling System. U.S. Environmental Protec-tion Agency, Office of Research and Develop-ment. EPA/600/R-99/030. pages unknown.

Ferguson, S.A; Hardy, C.C. 1994. Modeling Smol-dering Emissions from Prescribed BroadcastBurns in the Pacific Northwest. InternationalJournal of Wildland Fires 4(3):135-142.

Ferguson, S.A.; S. McKay; D. Nagel; T. Piepho;M.L. Rorig; and C. Anderson. 2001. Thepotential for smoke to ventilate from wildlandfires in the United States. In Proceedings of the4th Symposium on Fire and Forest Meteorol-ogy, 13-15 November 2001, Reno, Nevada.Paper No. 4.6.

Hardy, C.C.; Ferguson, S.A.; Speers-Hayes, P.;Doughty, C.B.; Teasdale, D.R. 1993. Assess-ment of PUFF: A Dispersion Model for SmokeManagement. Final Report to USDA ForestService Pacific Northwest Region. Fire andEnvironmental Research Applications, USDAForest Service, PNW Research Station, 32 pp.

Harrison, H. 1995. A User’s Guide to PUFFx: ADispersion Model for Smoke Management inComplex Terrain. WYNDSoft Inc., 42 pp.

Hummel, J.; Rafsnider, J. 1995. User’s Guide:TSARS Plus Smoke Production and DispersionModel, Version 1.1. National BiologicalService and the Interior Fire CoordinationCommittee. 107 pp.

Harms, Mary F.; Lavdas, Leonidas G. 1997. Draft:User’s Guide to VSMOKE-GIS for Worksta-tions. Research Paper SRS-6. Asheville, NC:U.S. Department of Agriculture, Forest Service,Southern Research Station. 41 p.

Lavdas, L.G. 1982. A day/night box model forprescribed burning impact in Willamette Valley,Oregon. Journal of Air Pollution ControlAgency, 32:72-76.

– 175 –


Lavdas, L.G. 1986. An Atmospheric DispersionIndex for Prescribed Burning. U.S. Departmentof Agriculture, Forest Service, SoutheasternForest Experiment Station. RP- SE-256. 35pp.

Lavdas, L.G. 1996. VSMOKE User’s Manual. U.S.Department of Agriculture, Forest Service,Southern Research Station. GTR-SRS-6. 147pp.

Lipsett M., Hurley S., Ostro B. 1997. Air Pollutionand Emergency Room Visits for Asthma inSanta Clara County, California. EnvironmentalHealth Perspectives. 105(2):216-222.

Ottmar, R.D., M.F. Burns, J.N. Hall, and A.D.Hanson. 1993. CONSUME User’s Guide.Gen. Tech. Rep. PNW-GTR-304. U.S. Depart-ment of Agriculture, Forest Service, PacificNorthwest Research Station. Portland, OR.117 pp.

Peterson, J.L. 1987. Analysis and reduction of theerrors of predicting prescribed burn emissions.Master of Science thesis. University of Wash-ington, Seattle, Washington. 70 pp.

Ross, D.G.; Smith, I.N.; Mannis, P.C.; Fox, D.G.1988. Diagnostic wind field modeling forcomplex terrain: model development andtesting. Journal of Applied Meteorology,27:785-796.

Reinhardt, E.D., R.E. Keane, and J.K. Brown. 1997.First Order Effects Model: FOFEM 4.0 User’sGuide. Gen. Tech. Rep. INT-GTR-344. U.S.Department of Agriculture, Forest Service,Intermountain Research Station. Ogden, UT.65 pp.

Sandberg, D.V.; Peterson, J. 1984. A sourcestrength model for prescribed fires in coniferouslogging slash. Presented at the 1984 AnnualMeeting, Air Pollution Control Association,Pacific Northwest Section, 1984 November 12-14, Portland, Oregon. 10 pp.

Sandberg, D.V. 2000. Implementation of an im-proved Emission Production Model. Abstract.Joint Fire Science Program PrincipleInvestigator’s Meeting. 3-5 October 2000,Reno, NV.

Schaaf, M.D., J. Nesbitt, and J. Hawkins. 1999. Anintroduction to smoke emissions and dispersionmodeling. National Interagency Fire Center,Great Basin Training Unit, 17-19 March, 1999,Boise Idaho. CH2Mhill, Portland, OR.[unpaginated].

Schwartz J, Slater D, Larson T.V, Pierson W.E,Koenig J.Q. 1993. Particulate air pollution andhospital emergency room visits for asthma inSeattle. Am Rev Respir Dis. 147(4):826-831.

Scire, J.; Strimaitis, D.G.; Yamartino R.J.; XiaomongZhang. 2000a. A User’s Guide for CALPUFFDispersion Model (Version 5). Earth Tech, Inc.,Concord, Massachusetts. 512 pp.

Scire, J.; Robe, F.R.; Fernau, M.E.; Yamartino, R.J.2000b. A User’s Guide for CALMETMeterorological Model. Earth Tech, Inc.,Concord, Massachusetts. 332 pp.

Sestak, M.L.; Marlatt,W.E.; Riebau, A.R. Unpub-lished. Draft VALBOX: Ventilated Valley BoxModel; U.S. Bureau of Land Management, 32pp.

Sestak, M.L.; Riebau, A.R. 1988. SASEM, SimpleApproach Smoke Estimation Model. U.S.Bureau of Land Management, Technical Note382, 31 pp.

South Carolina Forestry Commission. 1996. SmokeManagement Guidelines for Vegetative DebrisBurning Operations: State of South Carolina.SCFC 3rd Printing. 19 pp.

Systems Applications International. 1998. User’sGuide to the Regulatory Modeling System forAerosols and Deposition (REMSAD).SYSAPP98-96/42r2, prepared for the U.S.EPA under contract 68D30032. Available onU.S.EPA website at<www.epa.gov/scram001/>.

USDA Forest Service, Southern Forest ExperimentStation. 1976. Southern forestry smokemanagement guidebook. U.S. Department ofAgriculture, Forest Service, Southern ForestExperiment Station. SE-10. 55 pp.


– 176 –

U.S. Environmental Protection Agency. 1995.Compilation Of Air Pollutant Emission Factors,Volume I: Stationary Point and Area Sources,(AP-42). Fifth edition. Office of Air QualityPlanning and Standards Office of Air andRadiation, U.S. Environmental ProtectionAgency, Research Triangle Park, NC 27711.[unpaginated].

U.S. Environmental Protection Agency. 1997.National Ambient Air Quality Standards forParticulate Matter; Final Rule. 40 CFR Part 50[AD-FRL-5725-2] RIN 2060-AE66. FederalRegister. 62(138): 102 pp.

Wade, D.D. 1989. A Guide for prescribed fire insouthern forests. National Wildfire Coordinat-ing Group, Prescribed Fire and Fire EffectsWorking Team, Boise Interagency Fire Center,ATTN: Supply, 3905 Vista Avenue, Boise, ID83705. NFES #2108. [U.S. Department ofAgriculture, Forest Service, SoutheasternForest Experiment Station. Tech. Pub. R8-TP11.] 56 pp.

– 177 –

2001 Smoke Management Guide 10.0 – Monitoring

AIR QUALITY MONITORING

Chapter 10

Chapter 10 – Air Quality Monitoring 2001 Smoke Management Guide

– 178 –

– 179 –


Air Quality Monitoring for Smoke

John E. Core

Janice L. Peterson

Introduction

There are several reasons why wildland firemanagers may want to conduct an ambient airquality-monitoring program. These include:

• smoke management program evaluationpurposes,

• to fulfill a public information need,

• to verify assumptions used in Environmen-tal Assessments,

• to assess potential human health affects incommunities impacted by smoke,

• and to evaluate wildland burning smokeimpacts on State and Federal air qualitylaws and regulations.

Both visibility data and PM10/PM2.5 concentra-tion data are useful to smoke managementprogram coordinators for assessing air qualityconditions if the information is provided in real-time. Fire managers may also be interested inmonitoring impacts on visibility in Class I areas.Whatever the objective may be, care must betaken to match monitoring objectives to the rightmonitoring method. Monitoring locations,sampling schedules, quality assurance, andmonitoring costs are elements that must also beconsidered.

Particulate Monitoring Techniques

Particulate monitoring instruments generally useone of two particle concentration measurementtechniques: gravimetric or optical. Gravimetricor filter-based instruments collect particulateson ventilated filters. The filters are laterweighed at special laboratory facilities to deter-mine the mass concentration of particulatecollected. Gravimetric monitoring techniqueshave been used for years to quantify massconcentration levels of airborne particulatematter. Filter-based sampling is labor intensive.Filters must be conditioned, weighed beforesampling, installed and removed from theinstrument, and reconditioned and weighedagain at a special facility. Results may not beavailable for days or weeks. Also, airflow ratesand elapsed sampling time must be carefullymonitored and recorded to ensure accurateresults. Filter-based techniques integratesamples over a long period of time, usually 24-hours, to obtain the required minimum mass foranalysis. Gravimetric monitoring is best forprojects where high-accuracy is needed and thetime delay in receiving the data is not a prob-lem. State monitoring networks designed todetect violations of air quality standards relylargely on gravimetric monitors. Specificmonitoring devices must be approved by EPAfor this task and are called Federal ReferenceMonitors (FRM’s).


– 180 –

Optical instruments measure light-scattering(nephelometers) or light-absorbing (aethalo-meters) characteristics of the atmosphere. Thismeasurement can then be converted to obtain anestimate of the concentration of airborne par-ticulates. Optical instruments offer severaladvantages over gravimetric methods, includingreal-time readings, portability, low powerconsumption, and relatively low cost. Opticalinstruments have the disadvantage of beinggenerally less accurate than gravimetric instru-ments at estimating particulate mass concentra-tion. Optical instruments are best for projectswhere real-time or near-real time data is needed,where a high degree of accuracy is not a require-ment, and if instrument portability and rugged-ness is desirable.

Proper conversion of the light scattering mea-surement collected by nephelometers to anestimate of particle concentration requiresdevelopment of customized conversion equa-tions. The light scattering value measureddepends on particle size distribution and opticalproperties of the specific aerosol mix in the areaof interest. The light scattering value measuredvaries as a function of the relative proportions of

fine particles (including smoke) and coarseparticles (such as soil dust). As a result, opticalinstruments should be calibrated against a co-located FRM in the same area, and pollutantmix, in which they will eventually operate. Aformula is then developed to properly convertscattering to a particulate mass per unit volume(µg/m3) estimate.

In a recent monitoring instrument evaluationstudy, sixty-six laboratory measurements weremade with the MIE DataRam, the RadianceResearch nephelometer, and an EPA FRMsampler where the instruments were exposed topine needle smoke (Trent and others 1999).Results from these tests concluded that bothnephelometers overestimated mass concentra-tions of smoke when using the scattering tomass conversion factors provided by the manu-facturer. A follow-up study (Trent and others2000) compared optical instruments fromvarious manufacturers (Radiance, MIE, MetOne, Optec, and Andersen) to FRM instrumentsboth in the field and laboratory and developedpreliminary custom calibration equations (figure10.1). The report provides an estimate of aconversion equation for each instrument tested

Figure 10.1. Three of the nephelometers tested during the Trent and others (2000) study includethe MIE DataRam, the Radiance Research nephelometer, and the Met One GT-640.

– 181 –


but also recommends that optical instruments befield calibrated for a type of fire event, and thatmeteorological conditions and existing levels ofambient particles be included. Specific condi-tions to consider during calibration are age ofthe smoke, type of fire (flaming or smoldering),fuel moisture, relative humidity, and backgroundparticle concentration without smoke from thefire. Figure 10.2 shows the correlation foundbetween PM2.5 measurements made with anEPA FRM gravimetric instrument vs. resultsfrom an MIE DataRam nephelometer (Trent andothers 2000).

Wildland Fire SmokeMonitoring Objectives

Gathering PM10/PM2.5 air quality data down-wind from a prescribed burn or wildfire is animportant fire manager goal in some areas. Thisdata may be used as an input to smoke manage-

ment decision-making, and may or may notinvolve immediate public release of estimatedpollutant levels and health warnings. Thismonitoring can be conducted at a few sensitivelocations within a relatively small area duringspecific events such as a planned large-scaleunderstory burn, or used as a permanent part ofsmoke management effectiveness monitoring.Real-time data access, ease of use, and rugged-ness are all generally required so optical instru-ments are most appropriate (table 10.1).Monitors are often equipped with data loggersand modems to permit downloading of the dataover a telephone line or via radio modem. In thenear future, technology will be available tomake air quality monitoring data from remotesites accessible over the Internet. The USDAForest Service, Missoula Technology andDevelopment Program with Applied DigitalSecurity, Inc have developed a satellite-baseddata retrieval system. Appropriately outfitted

Figure 10.2. Comparison of PM2.5 measurements made with a gravimetric Federal ReferenceMonitor vs. an MIE DataRam nephelometer (Trent and others 2000).


– 182 –

Table 10.1. Equipment appropriate for smoke monitoring differs by program objective.

– 183 –


instruments will send packets of 5-minuteaverage particulate concentrations each hour bysatellite to a stored database to be viewed andretrieved through a Web site.1

A second smoke monitoring objective may be togather data on prescribed fire smoke impacts atsensitive locations over a much longer periodfor purposes of comparison with ambient airquality standards (NAAQS). In these cases,immediate data access is of secondary impor-tance to gathering data that approximates or isequivalent to the high-accuracy official FederalReference Method (FRM) instruments used byair regulatory agencies. A popular option is thesmall, portable, battery powered MiniVol sam-pler although these are not official EPA FRMdesignated monitors. The lag-time limitationmay be overcome by using one of two EPA-approved continuous air monitoring devices(TEOM or Beta Attenuation Monitors [BAM])

__________________________________

1 MTDC Air Program News Issue 1. August 2001. Available at: http://fsweb.mtdc.wo.fs.fed.us/programs/wsa/air_news/issue1.htm

Figure 10.3. A typical IMPROVE monitor installation.

but this equipment is costly and requires a highdegree of technical skill to operate (table 10.1).

Visibility protection is another monitoringobjective for fire managers when wildlandburning smoke may impact nearby Class I areas.For visibility monitoring, information is notonly needed on PM10/PM2.5 concentrations butaerosol chemical composition and particle lightscattering and absorption as well. Since aerosolchemical analysis (speciation) monitoringrequires filter-based methods and extinctionmeasurements require in-situ real-time methods,a combination of techniques are used. Monitor-ing is typically conducted throughout the yearover long time periods to establish trends. In asmuch as data consistency with the nationalvisibility programs is also important, specializedinstruments designed and deployed by theInteragency Monitoring of Protected VisualEnvironments (IMPROVE) Network (Malm


– 184 –

2000) should be used whenever possible(figure 10.3). Monitoring the visual quality ofa vista, called scene monitoring, is often done atthe same time using 35mm cameras. Digitalcamera systems can be used at sites wherereal-time web access to the scene is desirable(table 10.1).

Further monitoring guidance is available on theInternet at the EPA Air Monitoring TechnologyInformation Center (AMTIC) web site (http://www.epa.gov/ttn/amtic) and the EPA VisibilityImprovement site (http://www.epa.gov/oar/vis/index.html).

Monitoring Locations & Siting

Samplers used for smoke impact monitoring arenormally placed at smoke sensitive locationsthat have the greatest likelihood of impact.2

This may be a private residence, within a nearbycommunity, or at a county fair. Care must betaken to ensure that the instrument is located inan open, exposed location removed from localpollution sources such as dirt roads, burn bar-rels, or woodstoves that would influence thedata. The sampler should be located two ormore meters above ground at a secure location.Power availability and access are often control-ling considerations (CH2MHill 1997).

Visibility monitoring sites must be representa-tive of the Class I area of interest and are there-fore best located within the area’s boundary or,in the case of wilderness areas, as close to theboundary as possible. Since visibility data isused to represent conditions over sub-regionalspatial scales, special care is needed in siting to

avoid local source influences. The IMPROVEnetwork has recently been expanded withrepresentative monitors for each of the 156Class I areas in the country. Siting of the instru-ments was accomplished with state and FederalLand Manager input.

Sampling Schedules

The timing, duration, and frequency of samplingdepend on the program objective. Continuous,hourly data is needed to monitor smoke impactsfrom several days prior to burn ignition to a dayor two after the event. In contrast, PM10NAAQS compliance monitoring using filter-based instruments is conducted once every sixdays in attainment areas. In a nonattainmentarea, daily sampling is required for cities withmore than a million people and every three daysotherwise. Filter-based measurements made aspart of the IMPROVE visibility monitoringnetwork are made every third day to reducecosts and operational requirements. Continuousmonitoring instruments always operate 24 hoursper day. Although sampling duration andfrequency decisions are often based largely onoperating costs and technician time require-ments, measurements made as part of the IM-PROVE network or for NAAQS compliancedeterminations must follow the protocols out-lined in EPA regulations found on the AMTICweb site.

2 For NAAQS compliance monitoring, refer to the EPA Monitoring Network Siting Guidance found on the EPAAMTIC web site at. http://www.epa.gov/ttn/amtic.

– 185 –


Quality Assurance

Data integrity is essential in any monitoringprogram. Every monitoring project should havea documented quality assurance plan. In addi-tion to the maintenance and calibration mea-sures outlined by the manufacturer of theinstruments being used, additional qualityassurance measures may also be included in theplan if the monitoring data are of an especiallyimportant nature. These include auditing proce-dures conducted by the state/local air qualityagency to verify proper instrument siting,calibration and data capture as well as traceabil-ity of measurement standards to the NationalBureau of Standards (NBS) (EPA 1984). Meth-ods of calculation and data processing shouldalso be audited. Fire managers may wish toconfer with their state/local air agency to assurethat monitoring results are valid.

Monitoring Costs

Monitoring is expensive. In addition to thecapital cost of the instruments, costs for equip-ment installation, electrical, maintenance,calibration standards, supplies, shipping, dataanalysis, and reporting must also be considered.In the case of filter-based particulate sampling,laboratory costs for filter weighing and chemicalanalysis must also be included. On-goingannual operating costs for technician time toservice the instruments is a major expense thatoften drives the monitoring system design.

Literature CitationsCH2MHill. 1997. When and how to monitor

prescribed fire smoke: a screening procedure.Prepared for the USDA Forest Service, PacificNorthwest Region, Portland, OR. Contract No.53-82FT-03-2. 49p. Available at: http://www.fs.fed.us/r6/aq.

Environmental Protection Agency. 1984. Qualityassurance handbook for air pollution measure-ment systems. EPA-600/9-76-005. US Environ-mental Protection Agency, Office of Researchand Development. Washington DC.

Malm, William C. 2000. Spatial and seasonalpatterns and temporal variability of haze and itsconstituents in the United States – Report III.ISSN: 0737-5352-47. Cooperative Institute forResearch in the Atmosphere. Colorado StateUniversity. Fort Collins, CO 80523.

Trent, Andy; Thistle, Harold, Fisher, Rich; [andothers]. 1999. Laboratory evaluation of twooptical instruments for real-time particulatemonitoring of smoke. Tech. Rep. 9925-2806-MTDC. Missoula, MT: U.S. Department ofAgriculture, Forest Service, Missoula Technol-ogy and Development Center. 14p.

Trent, Andy; Davies, Mary Ann; Fisher, Rich; [andothers]. 2000. Evaluation of optical instrumentsfor real-time, continuous monitoring of smokeparticulates. Tech. Rep. 0025-2860-MTDC.Missoula, MT: U.S. Department of Agriculture,Forest Service, Missoula Technology andDevelopment Center. 38p.


– 186 –

– 187 –

2001 Smoke Management Guide 11.0 – Emission Inventories

EMISSION INVENTORIES

Chapter 11

Chapter 11 – Emission Inventories 2001 Smoke Management Guide

– 188 –

– 189 –


Emission Inventories

Janice L. Peterson

An inventory or estimate of total statewide (orsome other geographically distinct unit) annualemissions of criteria pollutants is a necessarypart of understanding the burden on the airresource in an area and taking appropriatecontrol actions. Emission inventories are a basicrequirement of state air resource managementprograms and are a required element of StateImplementation Plans. Emission inventorieshelp explain the contribution of source catego-ries to pollution events, provide backgroundinformation for air resource management,provide the means to verify progress towardemission reduction goals, and provide a scien-tific basis for state air programs. An accurateemissions inventory provides a measured, ratherthan perceived, estimate of pollutant productionas the basis for regulation, management action,and program compliance. Emission inventoriesshould include all important source categoriesincluding mobile, area, and stationary and arenot complete unless difficult-to-quantify sourceslike agricultural burning, backyard burning,rangeland burning, and wildland and prescribedburning are each addressed.

Wildland and prescribed fires are extremelydiverse and dynamic air pollution sources andtheir emissions can be difficult to quantify.Design and development of an emission inven-tory system is primarily the responsibility ofstate air regulatory agencies. But cooperationand collaboration between air regulatory agen-cies and fire managers is required to design aneffective and appropriate emission inventorysystem. Wildland fire managers should have the

knowledge and data necessary to calculateemissions from their burn programs and beprepared to work with the state in developingemission inventory systems for wildland fire.

At the most basic level, estimation of wildfireemissions requires knowledge of area burned,fuel consumed, and a fuel-appropriate emissionfactor. The estimate of emissions is madethrough simple multiplication of area burned(acres or hectares) times fuel consumed (tons peracre or kilograms per hectare) times an emissionfactor assigned with knowledge of the fuel type(lbs/ton or g/kg) (figure 11.1). Resulting emis-sions are in tons or kilograms.

Greater accuracy, precision, and complexity canbe achieved through increasingly detailed knowl-edge of these basic parameters. For example,area burned is estimated pre-burn in manyexisting reporting systems; if area burned isreassessed post-burn the accuracy of the emis-sion inventory will increase. Accuracy andprecision will also be improved if fuel consumedcan be estimated with knowledge of pre-burnloading and consumption of fuels in each ofmany possible categories based on fuel type,size, and arrangement; and with knowledge offuel moisture conditions, weather parameters,and application of emission reduction tech-niques. A more precise emission factor can beassigned with knowledge of burning conditionsthat can shift fuel consumption from the lessefficient smoldering combustion phase into themore efficient flaming phase (figure 11.2).


– 190 –

Figure 11.1. Basic information components needed to estimate the quantity ofemissions from an individual wildland burn and compile an emissions inventory.

Figure 11.2. Detailed information about fuel loading and consumption by size class plus information to predictconsumption by phase of combustion can increase the accuracy and precision of estimates of emissions fromprescribed wildland fire for an emissions inventory (Modified from Sandberg [1988]). The ranges given in thefigure cover the majority of fuel loading and consumption situations in wildland fuels but do not define theextremes. Numerous exceptions could likely be found in practice.

– 191 –


Sources of Prescribed BurningActivity Level Information

States with incomplete or no centralized burnreporting requirements will need to go to theburners themselves to quantify activity level.Federal agencies generally keep fairly accuraterecords of burning accomplished in a given timeperiod and can also provide estimates of wildfireacres. Federal agencies that may need to becontacted in a given state or area include theBureau of Land Management, Bureau of IndianAffairs or individual Tribes, National ParkService, Forest Service, and Fish and WildlifeService. In some areas other federal agenciesmay need to be contacted. Such as the Depart-ment of Energy, Department of Defense, NaturalResources Conservation Service, AgriculturalResearch Service, U.S. Geological Survey, orthe Department of Reclamation along withmanagers of National Preserves and NationalMonuments.

Specific state agencies with a forestry, wildlife,conservation, or natural resource managementmandate are another source of activity levelinformation. They may use prescribed burningthemselves and may compile burning statisticsfor state lands and sometimes also for privatelands. Private land owners, especially thosemanaging timber-lands should be contacted asshould The Nature Conservancy and theAudubon Society.

In some areas, especially where prescribedwildland burning is infrequent, the only sourcefor activity level information may be a grossestimate for all prescribed fires for an entirestate or area. This can sometimes be obtainedfrom a single federal or state agency, or some-times from an academic institution.

Type of Burn

Prescribed burning can be divided into catego-ries depending on the arrangement of the fuels.Fuel arrangement can help predict total fuelconsumption and the proportion consumed inthe flaming vs. smoldering phases. Broadcastburning refers to fuels burned in place. Thisterm can be used to describe natural woodyfuels scattered under a stand of trees, woodydebris scattered at random after a timber sale,brush burned in place, or grass. Fuels can alsobe concentrated into piles before burning. Inaddition to pile and broadcast burning, othergeneral prescribed-fire-type categories that maybe used include range, windrow, right-of-way,spot, black line, jack-pot, and concentration.Knowledge of the type of burn is valuable forestimating emissions as it can affect the accu-racy and correct interpretation of estimates ofarea burned, fuel consumed, and assignment ofan appropriate emission factor.

Area Burned

Area burned is generally the easiest parameter toobtain from fire managers. One caution is thatarea burned is often estimated prior to pre-scribed burning and not updated with the resultsof the burn, which may be smaller or larger (inthe case of an escaped fire) than originallyestimated. Also, area burned may reflect thearea treated or the area within the wildland fireperimeter, rather than the area actually black-ened by fire. The wildland fire perimeter maybe considerably larger than the area actuallyblackened by fire. For example, a study of theYellowstone fires of 1988 found that about 65%of the wildfire perimeter area within the parkwas actually blackened (Despain and others1989), the remaining 35% was in unburnedislands. In the case of prescribed fire, land


– 192 –

managers may consider a larger area to havebeen treated or to have benefited by the fire thanwas actually blackened by flames. Compilingan accurate emission inventory requires actualacres (or hectares) blackened for an accurateestimate of emissions. Caution should be usedwith estimates of area burned, as this parameteris more prone to systematic overestimation thanany other component of emissions estimation.

Fuel Consumed

Fuel consumed is generally estimated via a two-step process; first fuel loading is estimated, thena percent consumption is applied to calculatefuel consumed. At the most basic level, a singlevalue for both total fuel loading and consump-tion can be used (for example 20 tons of fuel ofwhich 50 percent consumed). In reality, afuelbed is a complex mix of various sizes ofwoody fuels (tree boles, branches, and twigs),needle and/or leaf litter, decayed and partlydecayed organic matter and rotten material(generally called duff or rot), and live fuels likebrush, forbs, and grass. Each of these fuelbedcomponents contributes to the total loading andis consumed to a greater or lesser extent. Forexample 100 percent of woody fuels less than 1inch in diameter may burn whereas just 30percent of those greater than 3 inches in diam-eter burn. In addition, some emission reductiontechniques are specific by fuelbed component.Use of a single estimate of total fuel loading andconsumption will fail to capture this. To gainaccuracy in the emissions inventory and theability to track the use and effectiveness ofemission reduction techniques, further detailconcerning fuel loadings by fuelbed componentwould ideally be tracked.

One simple method for obtaining a gross esti-mate of fuel loading is through the use of stan-

dardized fuel models. The most widely usedexample is the array of National Fire DangerRating System (NFDRS) fuel models (Deemingand others 1977). These 20 models are stan-dardized descriptions of different fuel types thatcan be used with some applicability to virtuallyall wildlands in the US. The NFDRS fuelmodels were designed as predictors of firedanger rather than to characterize the widerange of potential wildland fuel loadings aswould be ideal for compilation of an emissionsinventory. Another commonly used set of fuelmodels is based on predicting fire behavior.Thirteen fire behavior fuel models are describedin Anderson (1982). Since both the NFDRSand fire behavior fuel models were designed forpurposes other than accurate fuel loadingestimation, these models should be used withcaution. In addition, the use of standardizedfuel models to estimate fuel loading means thatefforts to reduce fuel loading for emissionreduction purposes prior to prescribed burningcannot be tracked or reflected in the emissionsinventory.

Other more detailed standardized fuel modelscalled fuel characteristic classes (FCC’s) areunder development (Sandberg and others 2001)that are expected to greatly improve fuel load-ing estimates when they reach widespread use.It is estimated that there will be a core set of 48to 64 FCC’s in common usage with as many as10,000 available in total describing the vastrange of fuel types and conditions that can existin wildlands across the country.

The most accurate method of estimating fuelloading is to have fire managers measure it inthe field. Field estimation also enablesreflection of the effect of emission reductiontechniques on fuel loading. The most accuratemethod of estimating fuel consumption isthrough modeling (field measurement beingunreasonably difficult in virtually all cases). In

– 193 –


the west, two fuel consumption models arecommonly used for this: the First Order FireEffects Model (FOFEM) (Reinhardt and others1997) and Consume (Ottmar and others 1993).These two models can provide very good esti-mates of fuel consumption if some basic knowl-edge of factors influencing fuel loading andmoisture are known.

Estimating fuel loading and consumption forwildfire is much more difficult than for pre-scribed fire. For one thing, large wildfires oftenburn through many different fuel types wherefuel loading can range from just a couple of tonsper acre to over 100 tons per acre. Also, thescience of predicting fuel consumption andemissions from a fire burning in tree crowns isextremely weak. The fuel type available fromwildfire report forms is generally for the pointof ignition rather than a reflection of fuel onthe majority of acres burned.

Emission Factors

Wildland and prescribed-fire emission factorsare contained in the EPA document AP-42 (EPA1995) and in table 5.1 in the Smoke SourceCharacteristics chapter. Accuracy may begained in an emissions inventory throughknowledge of the portion of fuel consumed inthe two primary consumption phases: flamingand smoldering. Flaming consumption emits farless emissions per unit of fuel consumed thansmoldering consumption. Estimation of theflaming vs. smoldering ratio can be obtainedthrough fuel consumption modeling and withknowledge of some influencing factors such asrate of ignition, fuel moisture conditions, anddays since rain.

Federal Agency Reporting

The Forest Service, Bureau of Land Manage-ment, Fish and Wildlife Service, National ParkService, and Bureau of Indian Affairs all havemandatory reporting requirements for wildlandand prescribed fires although at present, they areall somewhat different. These reports containsome of the basic information needed to com-pile an emissions inventory. Within the nextcouple of years, all federal agencies will bemoving toward a consolidated fire reportingdatabase through implementation of the FederalFire Policy.

Record keeping by state and private landownersis much more variable and may or may not beavailable to states wishing to compile an emis-sions inventory.

Forest Service

Forest Service forms FS-5100-29 (wildlandfire) and FS-5100–29T (prescribed fire) requiresome of the basic inputs needed to compile anemissions inventory. The wildland fire reportform requires reporting of acres burned withinthe fire perimeter regardless of landowner plusNational Fire Danger Rating System (NFDRS)fuel model. It is significant to note that theinstructions for estimating acres (USDA ForestService 1999) specify reporting of all acreswithin the fire perimeter, unfortunately thisvalue is not likely to equal acres blackened byfire. The number of acres blackened willalways be less than the number of acres within


– 194 –

the fire perimeter so use of this value withoutsome adjustment will result in a serioussystematic overestimation of acres actuallyburned and therefore of smoke produced. TheNFDRS fuel model reported is the one in whichthe fire was burning at the time and place whereanother required element, the fire intensitylevel, was observed so it may or may not berepresentative of the majority of acres burned.Individual fire reports are collected throughoutthe year and can be analyzed through an elec-tronic system called FIRESTAT (USDA ForestService 1999).

Data collected by the Forest Service aboutprescribed burning that is useful for compilingan emissions inventory includes the prevailingNFDRS fuel model; the total acres plus thepercent of acres burned; the preburn loading ofdead fuels 0-3 inches in diameter; 3+ inches indiameter, and live; and the percent of these fuelsthat consumed. The prescribed fire reportallows more accurate estimation of emissionssince the percent of acres burned is reported andfuel loading and consumption is estimated inthree categories. The Forest Service reportingsystem does not include estimates of duffconsumption which can contribute as much as50 percent of the emissions from a prescribedburn in certain areas under dry conditions,though is generally much less than that.

Fish and Wildlife Service

The Fish and Wildlife Service also has manda-tory fire reporting requirements and uses asystem called the Fire Reporting System (FRS)for data collection. The FRS requires reportingof project area size plus the actual burned areaor acres blackened for both wildland and pre-scribed fire. It also allows multiple entries forNFDRS fuel model and links a specific areaburned to each. Fuel loading is assigned based

on NDFRS defaults in seven categories: deadwoody fuels of diameter 0-1/4”, 1/4-1", 1-3",3+; herbaceous; live woody; and duff. Usersthen specify percent consumption for eachfuelbed category. Custom fuel models may alsobe defined. Data collected as part of the FRSprovides very good information for estimatingemissions from both wildland and prescribedfire on Fish and Wildlife Service burns thoughthis is a very small part of total burning in mostareas of the country with notable exceptions inthe Southeastern states and Alaska.

Bureau of Land Management

The BLM reporting requirements includeestimation of area burned for wildland andprescribed fire, less any unaltered areas as anestimate of acres blackened. The fire behaviorfuel model that best represents the fuels in theburn area is required as is the NFDRS fuelmodel in the vicinity of the fire origin. Themodel representing fuels in the burn area ismore appropriate for emissions estimation. Inaddition, for prescribed fire up to two fire-behavior fuel models can be selected and thepercent of the burned area assigned. Fuelloading (tons per acre) and consumption (per-cent) can be reported in each of six fuel sizeclasses: 0-1", 1.1-3", 3.1-9", greater than 9",shrub and herb, and litter and duff. If actualfield data for fuel loading and consumption isnot available, the most appropriate standard fuelloading and consumption range can be selected.Fuel loads can be assigned as light, average, orheavy for the fire behavior fuel model type andfuel consumption can be assigned as light,average, or heavy making some customizationof the standard fuel models possible. The BLMreporting system also accommodates the uniquerequirements of estimating loading and con-sumption of prescribed burning of debris piles.

– 195 –


National Park Service

The NPS has mandatory fire reporting require-ments but the information collected is of littleuse for emissions estimation, especially forwildland fire. For wildland fires, acres burned isrequired but the instructions don’t specifywhether perimeter acres or acres blackened is tobe reported. The only required description ofvegetation assigns one of three categories:commercial forest land, non-commercial forestland, or non-forest watershed which provideslittle or no information for estimating fuelloading and consumption. There is an optionalfield for input of NFDRS fuel model but howoften this is used is unknown. Prescribed fireand wildland fire for resource benefit requiresinput of both NFDRS fuel model and a firebehavior fuel model.

Bureau of Indian Affairs

Fire reporting requirements for the BIA aresimilar to those for the NPS (see discussionabove). One minor difference exists in thereporting of prescribed and wildland fire forresource benefits, where a fire behavior modelmay be input (but is not required). Further, afire danger rating (NFDR) fuel model cannot beinput.

Choosing the AppropriateAccuracy and Precision in anEmissions Inventory

The appropriate accuracy and precision for astate emissions inventory should be designedthrough analysis of the importance of the source

in the affected area (sub-state, state, or multi-state area). Variables influencing the impor-tance of prescribed burning as a source can beassessed through addressing issues such as:

• whether there are current impacts fromprescribed fire or wildfire smoke,

• the aggressiveness of state goals foremission reduction and air quality im-provement,

• the trend in burning in the local area andthe rate of increase or decrease,

• a professional or financial motivation byburners to track and/or reduce emissions,

• the need to associate wildland fire emis-sions with specific air pollution episodes.

Tables 11.1 and 11.2 summarize informationneeded for a prescribed burning emissionsinventory and for a wildland fire emissionsinventory. Each table lists the categories ofinformation needed to inventory emissions,proposes a minimum requirement for a basicinventory, and lists options for increasing theaccuracy and precision of the inventory whichmay be desirable if wildland fire in the area ofinterest is of concern or controversial.1

Data requirements for producing an emissionsinventory for either prescribed burning orwildland burning are very similar. They bothrequire information about the time period of theburn, the location, the area actually burned, adescription of the fuelbed, how much fuelburned, and site specific information for assign-ing an emission factor. A prescribed burning

___________________________________

1 Sandberg, David, V.; Peterson, Janice. 1997. Emission inventories for SIP development. An unpublished technical supportdocument to the EPA Interim Air Quality Policy on Wildland and Prescribed Fires. August 15, 1997. (Available from the authors oronline at http://www.epa.gov/ttncaaa1/faca/pbdirs/eisfor6.pdf ).


– 196 –

Tabl

e 11

.1. S

umm

ary

of in

form

atio

n ne

eded

to c

ompi

le a

pre

scr ib

ed fi

re e

mis

sion

s in

vent

ory

and

optio

ns fo

r in

crea

sing

acc

urac

y.

– 197 –


Tabl

e 11

.2. S

umm

ary

of in

form

atio

n ne

eded

to c

ompi

le a

wild

land

fire

em

issi

ons

inve

ntor

y an

d op

tions

for

incr

easi

ng a

ccur

acy


– 198 –

emissions inventory includes extra informationabout the type of burn or fuelbed arrangementplus the purpose of the burn. These are optionaldata items that may be useful in some cases. Awildland burning emissions inventory includesinformation about the control strategy used tofight the fire.

Literature CitationsAnderson, Hal E. 1982. Aids to determining fuel

models for estimating fire behavior. USDA For.Serv. Gen. Tech. Rep. INT-122. Intermt. For.And Range Exp. Stn., Ogden, Utah. 22p.

Deeming, John E., Robert E. Burgan, and Jack D.Cohen. 1977. The National Fire-Danger RatingSystem – 1978. USDA Forest Service Gen.Tech. Rep. INT-39. Intermt. For. And RangeExp. Stn., Ogden, Utah. 63 p.

Despain, Don; Rodman, Ann; Schullery, Paul;Shovic, Henry. 1989. Burned area survey ofYellowstone National Park: the fires of 1988.Yellowstone National Park, WY: U.S. Depart-ment of the Interior, National Park Service,Yellowstone National Park. 14 p.

Environmental Protection Agency. 1995. Compila-tion of air pollutant emission factors AP-42.Fifth Edition, Volume 1, Chapter 13.1: Wildfiresand prescribed burning. Supplement B, October1996. U.S. Environmental Protection Agency,Research Triangle Park, NC.

Ottmar, Roger D.; Burns, Mary F.; Hall, Janet N.;Hanson, Aaron D. 1993. CONSUME usersguide. Gen. Tech. Rep. PNW-GTR-304. Port-land, OR: U.S. Department of Agriculture,Forest Service, Pacific Northwest ResearchStation. 17 p.

Reinhardt, Elizabeth D.; Keane, Robert E.; Brown,James K. 1997. First Order Fire Effects Model,FOFEM 4.0 user’s guide. GTR IMT-GTR-344.

Sandberg, David V. 1988. Emission reduction forprescribed forest burning. In: Mathai, C.V.;Stonefield, David H., eds. Transactions—PM10: implementation of standards: AnAPCA/EPA International Specialty Conference;1988 February 22-25; San Francisco: 628-645.

Sandberg, David V.; Ottmar, Roger D., Cushon,Geoffery H. 2001. Characterizing fuels in the21st century. International Journal of WildlandFire 10:1-7.

USDA Forest Service. August 1999. FIRESTAT 5.0User’s Guide. Fire and Aviation Management.Washington DC.

– 199 –

2001 Smoke Management Guide 12.0 – Administration and Assessment

ADMINISTRATION AND ASSESSMENT

Chapter 12

Chapter 12 – Administration and Assessment 2001 Smoke Management Guide

– 200 –

– 201 –


Smoke Management ProgramAdministration and Evaluation

Peter Lahm

Smoke management program administration canrange from activities conducted at the local burnprogram level to a multi-state coordinated effortto manage smoke. The EPA Interim Air QualityPolicy on Wildland and Prescribed Fires (In-terim Policy) (EPA 1998) recommends thatsmoke management programs be administeredby a central authority with clear decision-making capability. As smoke managementprograms range from voluntary efforts to man-datory regulatory driven programs, the adminis-tration will vary accordingly1. On the morelocal level, the programs may be administeredby a group of land managers or private land-holders seeking to coordinate burning efforts toavoid excessive smoke impacts. Mandatoryregulatory driven smoke management programstend to be administered by tribal/state/district airquality regulatory agencies or state forestryentities. The administration of smoke manage-ment programs allows for a number of differentapproaches to meet EPA objectives and tomaintain cooperative and interactive efforts tomanage the dual objectives of good air qualityand land stewardship.

The Interim Policy also recommends periodicevaluation of smoke management programs to

ensure that air quality objectives are being met.From the land management point of view, thesesame reviews are critical to assessing whetherland management objectives are being metunder the smoke management program. EPAalso recommended periodic evaluation of smokemanagement rule or regulation effectiveness aspart of its Interim Policy. For programs that areunder scrutiny by a concerned public or aregrowing rapidly, continuous evaluation shouldalso be considered. All smoke managementefforts—from formal interagency smoke man-agement plans to less structured efforts toaddress smoke from individual fire operations—can benefit from continuous and periodic evalu-ation. If a smoke management program changessize, jurisdiction, or regulatory responsibilities,the level of effort applied to managing smokeshould also change. To keep a program ahead ofgrowing air quality concerns, a continuous effortto evaluate smoke management effectiveness isuseful. This evaluation is also critical for localunit programs that are under formal state ortribal smoke management plans. The evaluationprocess has applicability to all types of fire,including wildland fire under suppression,wildland fire use and prescribed fire.

1 Examples of specific state smoke management programs are provided in chapter 4, section 4.2.


– 202 –

Smoke Management ProgramAdministration

Administration of a smoke management pro-gram is frequently a function of the size of theburn program using a metric such as acresburned or emissions generated, coupled with thecomplexity of the local air quality issues. Fireprograms located in areas that are not rife withClass I areas, PM10 non-attainment areas, orsmoke-sensitive transportation corridors arecommonly under voluntary smoke managementprograms and may be locally administered.These types of programs may be focused onconcerns of local area impacts such as nuisanceor transportation safety and can be well ad-dressed through local level coordination amongburners. State forestry agencies and theirrespective districts are frequently central pointsfor dissemination of information; many ex-amples of this type of program can be found inthe southeastern states.

As air quality complexity rises with potentialsmoke impacts on non-attainment areas or ClassI areas, legal requirements also rise, and fre-quently trigger a more centralized regulatory-based smoke management program. Attendantwith the increased program requirements is thecommensurate increased cost of the program.Direct costs of smoke management programadministration are frequently recovered throughthe charging of fees to burners. Fees are fre-quently based on emissions production ortonnage of material to be consumed and areused to offset an authority’s program adminis-tration costs. The increased indirect cost offrequent reporting requirements and otherpermitting tasks such as modeling of impactsand smoke management plan preparation arefrequently overlooked. The most commoncentralized program approach is administeredby the state or tribal air quality authority and

can be found in such states as Colorado. Statessuch as Florida and Oregon have opted to usetheir forestry agencies to help directly managetheir smoke management programs. Oversightby the respective air quality regulatory authorityis usually a part of such a program. There is anoption for interagency approaches to smokemanagement program administration. Thisapproach blends the lines between air qualityregulatory agencies and land managers. Person-nel from a land management agency may beout-stationed to the respective air quality regula-tory authority to assist in the smoke manage-ment program administration. The states ofUtah and Arizona use this approach respectivelyand have avoided program management fees inthis fashion. This approach can also foster goodinter-agency communication and developmentof joint air quality and land management objec-tives for smoke management programs.

The future of smoke management programadministration will be a reflection of the imple-mentation of the Regional Haze Rule (40 CFRPart 51), which creates a paradigm in which airquality impacts are viewed in a regional senserather than by locality or state. Tribal smokemanagement programs are being rapidly devel-oped and will help support this regional ap-proach. The establishment of multi-state smokemanagement jurisdictions is rapidly becoming areality with a joint effort by Idaho and Montanabeing a recent example. The PM2.5 and ozonestandards will also support this type of approachas the impacts of smoke are viewed as a long-range transport issue. The inclusion of allsources of fire emissions, such as agriculturalburning and wildland burning, into a singularsmoke management program is also a futuredirection in these programs, and can already befound in the Title 17 Rule in California.

– 203 –


Evaluation of SmokeManagement Programs

Size of Program — In lieu of any other param-eter that can describe the activity level of a burnprogram, the number of acres can be used totrigger level of effort for smoke managementand subsequent evaluation of smoke effects. Asmentioned elsewhere, the representation of fireactivity in terms of emissions is more effectivefor air quality purposes. In lieu of emissions,fire size and fuel type can be used for triggeringdifferent smoke management requirements.Small burns located in remote areas with lowemissions may not dictate any evaluation greaterthan tracking the activity level and date of burn.However, more complex situations such as aburn of several days’ duration with heavyemissions located in the wildland/urban inter-face should be tracked more extensively forsmoke management effectiveness. This samecomplex situation may track the effectiveness ofemission reduction practices. It may be benefi-cial if the criteria are established in consultationwith the local or state air regulatory agency. Forfederal agencies, these criteria can also belinked to the management plan’s monitoringprogram. A post burn analysis of the smokemanagement plan and the burn’s smoke effectscan be extremely valuable to all concernedparties.

Intensity and Duration of Smoke Effects —The intensity and duration of smoke impacts arecritical parameters that can represent a variety ofsmoke management effectiveness measures.Duration of smoke impacts upon the public, anon-attainment area, a transportation corridor orClass I area can be tracked and assessed throughdirect air quality monitoring.2 The public can betolerant of one day of heavy levels of smoke,however consecutive day impacts may lead to arash of complaints. The criteria for evaluating a

program may be to assess the number of con-secutive days/hours of impact to a specific area.The intensity level of smoke impact also plays arole, as short bursts of high levels of smokepunctuated by clear air is frequently tolerable byreceptors. An application of this type of criteriaexists in Oregon where number and intensity ofsmoke intrusions is tracked annually. This typeof criteria is applicable to individual incidents aswell.

Methods of tracking the intensity and durationof smoke impact include:

• Number and type of public complaints(citizen, doctor, hospital, etc.);

• Intrusion of smoke into designated smokesensitive areas through specific air qualitymeasurement;

• Violations or percent increase of criteriapollutants attributable to smoke;

• Visibility impacts (local and regional).

As the National Ambient Air Quality Standards(NAAQS) include both short term and annualstandards, the full impact of smoke on theNAAQS may not be readily determined untilwell after the burn season is completed, whichfurther supports the importance of incorporatingevaluation into a smoke management program.Impacts on visibility were previously viewed onan annual basis, however that has changed totracking impacts on Class I areas to determineeffects on the 20% clearest and 20% dirtiestdays. These methods for tracking and evalua-tion should be established prior to the event oras part of the overall smoke management pro-gram as they can take significant planning orcoordination. Pre-planning for the air qualityelement of the Wildland Fire Situation Analysisused by federal agencies for wildland fires


– 204 –

(USDI and USDA Forest Service 1998) can alsobe beneficial as the public, air quality regulatorycommunity, and land management entity has theopportunity to increase acceptance of smokeeffects.

The evaluation criteria should be as quantitativeas possible in light of the complexity of the burnor program and the air quality concerns of thearea. Proximity to non-attainment or Class Iareas should automatically trigger some pro-grammatic evaluation. Visibility should beconsidered in terms of plume blight, regionalhaze and impacts on safety (transportation).Conversely, a small incident with a small quan-tity or short duration of emissions in an areawith few air quality concerns should not warrantextensive programmatic or individual incidentevaluation effort. Again, advance coordinationwith concerned parties can help determine thisvarying level of effort.

If an incident or program results in a smokeintrusion above a pre-defined level such asnumber of complaints or presence of smoke inan avoidance area, the cause should be evaluatedas soon as possible. The breakdown of thesmoke management plan for an incident isequivalent to the breakdown of the fire behaviorprescription for the burn. Smoke managementcontingency programs are another element of asmoke management program included in theInterim Policy (EPA 1998). Factors such asweather/smoke dispersion forecasting or fuelcondition changes can lead to such a smokeintrusion and need to be evaluated quicklyfollowing a failure of the system in order to beaddressed in a proactive fashion. Determinationof what caused the adverse air quality impactallows for growth of the program throughimplementation of changes to avoid futurerecurrence. If a program or incident was con-ducted such that no smoke criteria were ex-ceeded, evaluation of the factors which led to

success are also valuable in building confidenceamong cooperating parties. The development ofan annual report which outlines the air qualityeffects of a burning program or the smokemanagement program demonstrates the commit-ment to addressing both land management andair quality objectives and can show significantand useful trends to concerned parties. Theknowledge that smoke impacts are being ad-dressed effectively in terms of specific criteria isvaluable when working with the concernedpublic and media.

Sources for Evaluation — Evaluation can bethe assessment of air quality monitoring datacollected by the land manager or utilization ofexisting air quality networks as operated by aregulatory agency (state/district/county/EPA/tribe). The meteorological conditions underwhich burns occur is another criteria that can beevaluated to help assess the smoke managementprogram. For complex smoke areas, the use ofdigital camera points could allow distribution ofthe real-time images over the Internet to con-cerned parties, including the public. The con-cerned public can also be directly queried as tothe level of smoke levels and duration of effects.

Annual Evaluation — One of the most effec-tive means of evaluating the smoke managementprogram is to hold periodic meetings amongstthe concerned parties such as the burners,regulators and potentially-concerned public.The frequency of such reviews should dependon the air quality complexity and smoke im-pacts. Many statewide smoke managementprograms meet annually to review the years’activities, successes and problems. Thesemeetings could include review of activity/emissions of burners, record-keeping efforts,effects tracked through the previously men-tioned methods, and discussion of programlogistics and costs. This same review meeting isalso an opportune time to plan for future

– 205 –


changes, discuss emerging issues, and conducttraining if needed. The Interim Policy (EPA1998) also urges such an evaluation processoccur annually. These annual sessions may bean effective way of addressing an Interim Policygoal of assessing the adequacy of the rules andregulations pertaining to smoke management fora respective state, tribe or other managing entity.Reflecting the state of the smoke managementprogram, whether statewide or at the landmanager level, through the issuance of anannual program report on smoke managementcan be another technique for assessing theprogram and informing the public of the invest-ment into smoke management.

Continuous Evaluation — If a specific inci-dent were to have significant adverse effects, itmight trigger immediate review to prevent arepeat occurrence. This immediate incidentassessment can be an effective way of address-ing pressing public concerns that may havearisen due to the impacts. During a wildlandfire use incident, daily conference calls amongstthe land manager and the regulatory agencieswhich discuss acres/fuels/emissions or qualita-tive smoke behavior can be very effective ataddressing smoke concerns. This real-timeevaluation can prevent conflict over smokeimpacts and can ensure accurate information be

provided to the public as well as incorporatedinto the message transmitted to the media by therespective agencies.

Incident debriefings should consider air qualityeffects and how they were addressed. In wild-land fire use, there is a continuous evaluation ofair quality as part of the Wildland Fire SituationAnalysis (USDI and USDA Forest Service1998). Establishment of criteria for evaluationof air quality effects prior to the actual event orimplementation of a program can allow forgreater buy-in by potentially affected partieswhen the fire occurs. Criteria for evaluationshould also include indicators of success.

Literature CitationsEnvironmental Protection Agency. 1998. Interim air

quality policy on wildland and prescribed fires.Office of Air Quality Planning and Standards,Research Triangle Park, NC 27711. 39p.

40 CFR Part 51. Vol. 64 No. 126. Regional HazeRegulations – Final Rule. July 1, 1999.

USDI and USDA Forest Service. 1998. Wildlandand prescribed fire management policy—implementation procedures reference guide.National Interagency Fire Center, Boise, ID. 81pp. and appendices.


– 206 –

2001 Smoke Management Guide Glossary

– 207 –

APPENDIX

Appendix A – Glossary 2001 Smoke Management Guide

– 208 –


– 209 –

Appendix A

Glossary of Fire and SmokeManagement Terminology

The terms listed below were either taken from existing glossaries or developed specifically for thisGuide. Where terms were taken from an existing glossary or document, the source reference is indexedin brackets (e.g. [source number]), with full reference citations provided at the end of the glossary.Note: Although the referenced definitions in this glossary were taken from other sources, the editorshave revised or changed many of them from their original version.

Absorption coefficient A measure of the ability of particles or gases to absorb photons; a num-ber that is proportional to the number of photons removed from the sightpath by absorption per unit length. (See Extinction coefficient). [2]

Activity fuel Debris resulting from such human activities as road construction, log-ging, pruning, thinning, or brush cutting. It includes logs, chunks, bark,branches, litter, stumps, and broken understory trees or brush.

Activity level Fuels resulting from, or altered by, forestry practices such as timberharvest or thinning, as opposed to naturally created fuels. [1]

Adiabatic lapse rate Rate of decrease of temperature with increasing height of a rising airparcel without an exchange of heat at the parcel boundaries. (See Dryadiabatic lapse rate, Saturated adiabatic lapse rate, and Atmosphericstability).

Advection The transfer of atmospheric properties by the horizontal movement of air,usually in reference to the transfer of warmer or cooler air, but may alsorefer to moisture. [1]

Aerial ignition Ignition of fuels by dropping incendiary devices or materials from air-craft. [1]

Aerosol A suspension of microscopic solid or liquid particles in a gaseous me-dium, such as smoke and fog. [2]


– 210 –

Air mass An extensive body of air having similar properties of temperature andmoisture. [1]

Air pollution The general term referring to the undesirable concentration of substances(gases, liquids, or solid particles) to the atmosphere that are foreign to thenatural atmosphere or are present in quantities exceeding natural concen-trations. [1]

Air quality The composition of air with respect to quantities of pollution therein;used most frequently in connection with “standards” of maximum ac-ceptable pollutant concentrations. [1]

Allowable emissions The emissions rate that represents a limit on the emissions that can occurfrom an emissions unit. This limit may be based on a federal, state, orlocal regulatory emission limit determined from state or local regulationsand/or 40 Code of Federal Regulations (CFR) Parts 60, 61, and 63. [3]

Ambient air Any unconfined portion of the atmosphere: open air, surrounding air. [4]

Ambient standards Specific target threshold concentrations and exposure durations of pollut-ants based on criteria gauged to protect human health and the welfare ofthe environment. Ambient standards are not emissions limitations onsources, but usually result in such limits being placed on source operationas part of a control strategy to achieve or maintain an ambient standard.[3]

Anthropogenic Produced by human activities. [2]

Area sources A source category of air pollution that generally extends over a largearea. Prescribed burning, field burning, home heating, and open burningare examples of area sources. [1]

Atmospheric inversion (1) Departure from the usual increase or decrease with altitude of thevalue of an atmospheric property (in fire management usage, nearlyalways refers to an increase in temperature with increasing height). (2)The layer through which this departure occurs (also called inversionlayer). The lowest altitude at which the departure is found is called thebase of the inversion. (See Atmospheric stability; Temperature inversion;Mixing height; Mixing layer; Stable atmosphere; Unstable atmosphere;Subsidence inversion) [1]


– 211 –

Atmospheric pressure The force exerted by the weight of the atmosphere, per unit area. At sealevel the atmospheric pressure fluctuates around 1013 millibars (mb). At5,000 feet (~1,500 m) above sea level the atmospheric pressure fluctuatesaround 850 mb. (See Standard atmosphere).

Atmospheric stability The degree to which vertical motion in the atmosphere is enhanced orsuppressed. (See Atmospheric inversion; Temperature inversion; Mixingheight; Mixing layer; Stable atmosphere; Unstable atmosphere). [1]

Attainment Area An area considered having air quality as good as or better than the Na-tional Ambient Air Quality Standards (NAAQS) as defined in the CleanAir Act. Note that an area may be in attainment for one or more pollut-ants but be a nonattainment area for one or more other pollutants. (SeeNon-attainment area). [3]

Avoidance A smoke emission control strategy that considers meteorological condi-tions when scheduling prescribed fires in order to avoid incursions intosmoke sensitive areas. [1]

Background level In air pollution control, the concentration of air pollutants in a definitearea during a fixed period of time prior to the starting up, or the stoppage,of a source of emission under control. In toxic substances monitoring,the average presence in the environment, originally referring to naturallyoccurring phenomena. [1]

Best Available Control An emission limitation action based on the maximum degree ofMeasures (BACM) emission reduction (considering energy, environmental, and

economic impacts) achievable through application of productionprocesses and available methods, systems, and techniques. [4]

Burn severity A qualitative assessment of the heat pulse directed toward the groundduring a fire. Burn severity relates to soil heating, large fuel and duffconsumption, consumption of the litter and organic layer beneath treesand isolated shrubs, and mortality of buried plant parts. [1]

Carbon dioxide (CO2) A colorless, odorless, nonpoisonous gas, which results from fuel combus-tion and is normally a part of the ambient air. [1]

Carbon monoxide (CO) A colorless, odorless, poisonous gas produced by incomplete fuel com-bustion. Carbon monoxide is a criteria pollutant and is measured in partsper million. (See Criteria pollutants).


– 212 –

Carcinogen Any substance that can cause or contribute to the production of cancer.[1]

Clean Air Act A federal law enacted to ensure that air quality standards are attained andmaintained. Initially passed by Congress in 1963, it has been amendedseveral times. [1]

Combustion efficiency The amount of products of incomplete combustion released relative toamounts produced from theoretically perfect combustion, expressed as adimensionless percentage. Because perfect combustion produces onlyCO2 and water, its combustion efficiency is 1.0. In combustion ofwildland fuels, combustion efficiency can roughly range from as high as0.95 (for flaming combustion) to as low as 0.65 (for smoldering combus-tion).

Condensation nuclei The small nuclei or particles with which gaseous constituents in theatmosphere (e.g., water vapor) collide and adhere. [2]

Consumption The amount of a specified fuel type or strata that is removed through thefire process, often expressed as a percentage of the preburn weight. [1]

Convection column The rising column of gases, smoke, fly ash, particulates, and other debrisproduced by a fire. The column has a strong vertical component indicat-ing that buoyant forces override the ambient surface wind. [1]

Convergence The term for horizontal air currents merging together or approaching asingle point, such as at the center of a low-pressure area producing a netinflow of air. The excess air is removed by rising air currents. Expan-sion of the rising air above a convergence zone results in cooling, whichin turn often gives condensation (clouds) and sometimes precipitation.[1]

Criteria Pollutants Pollutants deemed most harmful to public health and welfare and that canbe monitored effectively. They include carbon monoxide (CO), lead(Pb), nitrogen oxides (NOx ), sulfur dioxide (SO2), Ozone (O3), particu-late matter (PM) of aerodynamic diameter less than or equal to 10 mi-crometers (PM10) and particulate matter of aerodynamic diameter lessthan or equal to 2.5 micrometers (PM2.5). [3]

Deciview A unit of visibility proportional to the logarithm of the atmosphericextinction. (See Extinction coefficient; Visibility; Visual range). [2]


– 213 –

De minimis level A level of emission or impact that is too small to be considered of con-cern. From the Latin phrase “de minimis non curat lex,” meaning the lawis not concerned with trifles.

Dew point Temperature to which a specified parcel of air must cool, at constantpressure and water-vapor content, in order for saturation to occur. Thedew point is always lower than the wet-bulb temperature, which is alwayslower than the dry-bulb temperature, except when the air is saturated andall three values are equal. Fog may form when temperature drops toequal the dew point. (See Dry-bulb temperature; Wet-bulb temperature).[1]

Dormant season burning Prescribed burning conducted during the time of year when vegetation isnot actively growing. In some parts of the country, dormant season burnsare typically less intense than growing season burns.

Drift smoke Smoke that has drifted from its point of origin and is no longer domi-nated by convective motion. May give false impression of a fire in thegeneral area where the smoke has drifted. [1]

Dry adiabatic lapse rate Adiabatic cooling in a dry atmosphere. Usually about -5.5 degrees(DALR) Fahrenheit per 1,000 feet (~-10 degrees centigrade per kilometer).

(See Adiabatic lapse rate; Saturated adiabatic lapse rate).

Dry-bulb temperature Originally, the temperature measured with a mercury thermometer whosebulb is dry. Commonly it is a measure of the atmospheric temperaturewithout the influence of moisture. (See Wet-bulb temperature; Dewpoint).

Duff The partially decomposed organic material above mineral soil that liesbeneath the freshly fallen twigs, needles, and leaves and is often referredto as the F (fermentation) and H (humus) layers. Duff often consumesduring the less efficient smoldering stage and has the potential to producemore than 50 percent of the smoke from a fire.

Ecosystem health A condition where the parts and functions of an ecosystem are sustainedover time and where the system’s capacity for self- repair is maintained,allowing goals for uses, values, and services of the ecosystem to be met.


– 214 –

Ecosystem maintenance A prescribed fire or wildland fire managed for resource benefits that isburn utilized to mimic the natural role of fire in an ecosystem that is currently

in an ecologically functional and fire resilient condition. [5]

Ecosystem Processes The actions or events that link organisms and their environment, such aspredation, mutualism, successional development, nutrient cycling, carbonsequestration, primary productivity, and decay. Natural disturbanceprocesses often occur with some periodicity

Ecosystem Restoration The re-establishment of natural vegetation and ecological processes thatmay be accomplished through the reduction of unwanted and/or unnatu-ral levels of biomass. Prescribed fires, wildland fires managed for re-source benefits and mechanical treatments may be utilized to restore anecosystem to an ecologically functional and fire resilient condition. [5]

Extinction coefficient A measure of the ability of particles or gases to absorb and scatter pho-tons from a beam of light; a number that is proportional to the number ofphotons removed from the sight path per unit length. (See Absorptioncoefficient; Deciview; Visibility; Visual range). [2]

Effective windspeed The mid-flame windspeed adjusted for the effect of slope on fire spread.[1]

Emission factor (EFp) The mass of particulate matter produced per unit mass of fuel consumed(pounds per ton, grams per kilogram). [1]

Emission inventory A listing, by source, of the amount of air pollutants discharged into theatmosphere of a community. [3]

Emission rate The amount of an emission produced per unit of time (lb./min or g/sec).[1]

Emission reduction A strategy for controlling smoke from prescribed fires that minimizes theamount of smoke output per unit area treated. [1]

Emission Standards A general type of standard that limit the mass of a pollutant that may beemitted by a source. The most straightforward emissions standard is asimple limitation on mass of pollutant per unit time (e.g., pounds ofpollutant per hour). [3]


– 215 –

Extinction The attenuation of light due to scattering and absorption as it passesthrough a medium. [2]

Federal Class I area In 1977, Congress identified 156 national parks, wilderness areas, inter-national parks and other areas that were to receive the most stringentprotection from increases in air pollution. It also set a visibility goal forthese areas to protect them from future human-caused haze, and toeliminate existing human-caused haze, and required reasonable progresstoward that goal. [5]

Fine fuel moisture The moisture content of fast-drying fuels that respond to changes inmoisture within 1 hour or less; such as, grass, leaves, ferns, tree moss,pine needles, and small twigs (0-1/4" or 0.0-0.6 cm). (See Fuel moisturecontent; One-hour timelag fuels). [1]

Fire-adapted ecosystem An ecosystem with the ability to survive and regenerate in a fire-proneenvironment.

Fire-dependent An ecosystem that cannot survive without periodic fire.ecosystem

Fire exclusion The policy and practice of eliminating fire from an area to the greatestextent possible, through suppression of wildland fires and a lack of fireuse.

Fire regime Periodicity and pattern of naturally occurring fires in a particular area orvegetative type, described in terms of frequency, biological severity, andarea extent. [1]

Fire regime groups Classes of fire regimes grouped by categories of frequency (expressed asmean fire return interval) and severity. Refers specifically to five groupsused in Federal policy and planning: 0-35 years, low severity; 0-35 years,stand replacement; 35-100 years, mixed severity; 35-100 years, standreplacement; 200+ years, stand replacement. (See Fire return interval;Fire regime).

Fire return interval Mean fire return interval. A mean, area-weighted time (in years) betweensuccessive fires for a respective area (i.e., the interval between two


– 216 –

successive fire occurrences); the size of the area must be specified.

Fire severity (See Burn severity.)

Fire use The combination of wildland fire use and prescribed fire application tomeet resource objectives. [6]

Fireline intensity The rate of heat release per unit time per unit length of fire front. Nu-merically, it is the product of the heat yield, the quantity of fuel con-sumed in the fire front, and the rate of spread. [1]

Flaming combustion Luminous oxidation of gases evolved from the rapid decomposi-phase tion of fuel. This phase follows the pre-ignition phase and precedes the

smoldering combustion phase, which has a much slower combustion rate.Water vapor, soot, and tar comprise the visible smoke. Relatively effi-cient combustion produces minimal soot and tar, resulting in whitesmoke; high moisture content also produces white smoke. (See Soot;Smoldering combustion phase). [1]

Forest floor material Surface organic material, including duff, litter, moss, peat, down-deadwoody pieces.

Forest residue Accumulation in the forest of living or dead (mostly woody) material thatis added to and rearranged by human activities such as harvest, culturaloperations, and land clearing. (See Activity fuel). [1]

Fuel loading The amount of fuel present expressed quantitatively in terms of weight offuel per unit area. This may be available fuel (consumable fuel) or totalfuel and is usually dry weight. [1]

Fuel moisture content The quantity of moisture in fuel expressed as a percentage of the weight;derived by weighing fuel sample both before and after thorough drying at(nominally) 212 degrees F (100 degrees C). (See Fine fuel moisture). [1]

Fuel reduction Manipulation, including combustion, or removal of fuels to reduce thelikelihood of ignition and/or to lessen potential damage and resistance tocontrol. [1]

Fuel size class A category used to describe the diameter of down dead woody fuels.Fuels within the same size class are assumed to have similar wetting anddrying properties, and to preheat and ignite at similar rates during thecombustion process. [1]


– 217 –

Fuel treatment Manipulation or removal of fuels to reduce the likelihood of ignition and/or to lessen potential intensity, rate of spread, severity, damage, andresistance to control. Examples include lopping, chipping, crushing,piling and burning. [1]

Fuel type An identifiable association of fuel elements of distinctive species, form,size, arrangement, or other characteristics that will cause a predictablerate of spread or resistance to control under specified weather conditions.[1]

Glowing combustion Oxidation of solid fuel accompanied by incandescence. Allphase volatiles have already been released and there is no visible smoke. This

phase follows the smoldering combustion phase and continues until thetemperature drops below the combustion threshold value, or until onlynon-combustible ash remains. (See Combustion; Flaming combustionphase; Smoldering combustion phase). [1]

Growing season burning Prescribed burns conducted during the time of year when vegetation isactively growing, or when leaves have matured but not fallen.

Hazard reduction Any treatment of living and dead fuels that reduces the threat of ignitionand spread of fire. [1]

Haze A sufficient concentration of atmospheric aerosols to be visible. Theparticles are so small that they cannot be seen individually, but are stilleffective in visual range restriction. (See Visual range; Extinction; Ab-sorption coefficient; Regional haze). [2]

Heat release rate (1) Total amount of heat produced per unit mass of fuel consumed perunit time. (2) Amount of heat released to the atmosphere from theconvective-lift fire phase of a fire per unit time. [1]

Hydrocarbons Compounds containing only hydrogen and carbon. [2]

IMPROVE Interagency Monitoring of Protected Visual Environments. A cooperativevisibility monitoring effort, using a common set of standards across theUnited States, between the EPA, Federal land management agencies, andstate air agencies. [5]


– 218 –

Integrating nephelometer An instrument that measures the amount of light scattered (scatteringcoefficient) and can be used to measure particulate matter concentrationsfrom fires. [2]

Inversion (See Atmospheric inversion) [2]

Isothermal layer A layer of finite thickness in any medium in which the temperatureremains constant.

Landscape An area composed of interacting and inter-connected ecosystems that arerepeated because of the geology, landform, soils, climate, biota, andhuman influences throughout the area. A landscape is composed ofwatersheds and smaller ecosystems.

Lead (Pb) A criteria pollutant, elemental lead emitted by stationary and mobilesources can cause several types of developmental effects in childrenincluding anemia and neurobehavioral and metabolic disorders. Non-ferrous smelters and battery plants are the most significant contributors toatmospheric lead emissions. (See Criteria pollutants). [3]

Litter The top layer of forest floor, composed of loose debris of dead sticks,branches, twigs, and recently fallen leaves or needles; little altered instructure by decomposition. (See Duff; Forest floor material). [1]

Mass fire A fire resulting from many simultaneous ignitions that generates a highlevel of energy output. [1]

Mean fire interval (See Fire return interval)

Micron Micrometer (mm)—a unit of length equal to one millionth of a meter; theunit of measure for wavelength and also for the mean aerodynamicdiameter of atmospheric aerosols. [2]

Mixing height Measured from the surface upward, the height to which relatively vigor-ous mixing occurs in the atmosphere due to turbulence and diffusion.Also called mixing depth. [1]

Mixing layer That portion of the atmosphere from the surface up to the mixing height.This is the layer of air within which pollutants are mixed by turbulenceand diffusion. Also called mixed layer. (See Ventilation Index). [1]


– 219 –

Mopup Extinguishing or removing burning material near control lines, fellingsnags, and trenching logs to prevent rolling after an area has burned, toreduce the chance of fire spreading beyond the control lines, or to reduceresidual smoke. [1]

Mosaic The central spatial characteristic of a landscape. The intermingling ofplant communities and their successional stages, or of disturbance (espe-cially fire), in such a manner as to give the impression of an interwoven,“patchy” design. [1]

National Ambient Air Maximum recommended concentrations of criteria pollutantsQuality Standards to maintain reasonable standards of air quality. (See criteria(NAAQS) pollutants). [3]

National Wildfire National interagency operational group authorized by the U.S.Coordinating Group Secretaries of Agriculture and Interior and the National Associa-(NWCG) tion of State Foresters, designed to coordinate fire management programs

of participating federal, state, local and private agencies to avoid wastefulduplication and provide a means of constructive cooperation.

Natural background An estimate of the visibility conditions at each Federal Class I areacondition that would exist in the absence of human-caused impairment. [5]

Nitrogen dioxide (NO2) The result of nitric oxide combining with oxygen in the atmosphere. Amajor component of photochemical smog. [1]

Nitrogen Oxide[s] (NOx) A class of compounds that are respiratory irritants and that react x withvolatile organic compounds (VOCs) to form ozone (O3). The primarycombustion product of nitrogen is nitrogen dioxide (NO2). However,several other nitrogen compounds are 2 usually emitted at the same time(nitric oxide [NO], nitrous oxide [NO], etc.), and these may or may notbe distinguishable in available test data. [3]

Non-attainment area An area identified by an air quality regulatory agency through ambientair monitoring (and designated by the Environmental Protection Agency),that presently exceeds federal ambient air standards. (See Attainmentarea). [1]

Nuisance smoke The amount of smoke in the ambient air that interferes with a right orprivilege common to members of the public, including the use or enjoy-ment of public or private resources.


– 220 –

One-hour timelag fuels Fuels consisting of dead herbaceous plants and roundwood less thanabout one-fourth inch (6.4 mm) in diameter. Also included is the upper-most layer of needles or leaves on the forest floor. Fuel elements of thissize usually respond to changes in moisture within one hour or less,hence the term 1-hr timelag. (See Fuel moisture content; Fine fuel mois-ture). [1]

One-hundred-hour Dead fuels consisting of roundwood in the size range of 1 to 3timelag fuels inches (2.5 to 7.6 cm) in diameter and very roughly the layer of litter

extending from approximately three-fourths of an inch (1.9 cm) to 4inches (10 cm) below the surface. Fuel elements of this size usuallyrespond to changes in moisture within about one hundred hours or 3 to 5days, hence the term 100-hr timelag. (See Fuel moisture content). [1]

One-thousand-hour Dead fuels consisting of roundwood 38 inches in diameter and thetimelag fuels layer of the forest floor more than about 4 inches below the surface. Fuel

elements of this size usually respond to changes in moisture within aboutone thousand hours or 4 to 6 weeks, hence the term 1000-hr timelag.(See Fuel moisture content). [1]

Ozone (O3) A criteria pollutant, ozone is a colorless gas, ozone is the major compo-nent of smog. Ozone is not emitted directly into the air but is formedthrough complex chemical reactions between precursor emissions ofvolatile organic compounds (VOCs) and NOx in the presence of sunlight.(See Criteria pollutants). [3]

Particulate matter Any liquid or solid particle. “Total suspended particulates” as used in airquality are those particles suspended in or falling through the atmo-sphere. They generally range in size from 0.1 to 100 microns. [1]

Piling-and-burning Piling slash resulting from logging or fuel management activities andsubsequently burning the individual piles. [1]

PM10 Particulate matter of mass median aerodynamic diameter (MMAD) lessthan or equal to 10 micrometers. A measure of small solid matter sus-pended in the atmosphere that can penetrate deeply into the lung wherethey can cause respiratory problems. Emissions of PM10 are significantfrom fugitive dust, power plants, commercial boilers, metallurgicalindustries, mineral industries, forest and residential fires, and motorvehicles. (See Criteria pollutants). [3]


– 221 –

PM2.5 Particulate matter of mass median aerodynamic diameter (MMAD) lessthan or equal to 2.5 micrometers A measure of fine particles of particu-late matter that come from fuel combustion, agricultural burning,woodstoves, etc. Often called respirable particles, as they are moreefficient at penetrating lungs and causing damage. (See Criteria pollut-ants). [3]

Point sources Large, stationary, identifiable sources of emissions that release pollutantsinto the atmosphere. Sources are often defined by state or local air regu-latory agencies as point sources when they annually emit more than aspecified amount of a given pollutant, and how state and local agenciesdefine point sources can vary. [3]

Precursor emissions Emissions from point or regional sources that transform into pollutantswith varied chemical properties. [2]

Prescribed fire Any fire ignited by management actions to meet specific objectives. Awritten, approved prescribed fire plan must exist, and NEPA require-ments must be met, prior to ignition. This term replaces managementignited prescribed fire. [6]

Prescribed natural fire Obsolete term. (See Wildland fire use) [6]

Prescription A written statement defining the objectives to be attained as well as theconditions of temperature, humidity, wind direction and speed, fuelmoisture, and soil moisture, under which a fire will be allowed to burn.A prescription is generally expressed as acceptable ranges of the pre-scription elements, and the limit of the geographic area to be covered. [1]

Prevention of Significant A program identified by the Clean Air Act to prevent air qualityDeterioration (PSD) and visibility degradation and to remedy existing visibility problems.

Areas of the country are grouped into 3 classes that are allowed certaindegrees of pollution depending on their uses. National Parks and Wilder-ness Areas meeting certain criteria are “Class I” or “clean area” in thatthey have the smallest allowable increment of degradation. [1]

Reasonably Available Control measures developed by EPA that apply to residentialControl Measures wood combustion, fugitive dust, and prescribed and silvicultural(RACM) burning in and around “moderate” PM10 nonattainment areas. RACM is

designed to bring an area back into attainment and uses a smoke manage-ment program that relies on weather forecasts for burn/no-burn days.


– 222 –

(See Best Available Control Measures [BACM]). [1]

Regional Haze Visibility impairment caused by the cumulative air pollutant emissionsfrom numerous sources over a wide geographic area. (See Haze).

Relative humidity (RH) The ratio of the amount of moisture in the air, to the maximum amount ofmoisture that air would contain if it were saturated. [1]

Residual combustion (See Smoldering combustion phase)phase

Residual smoke Smoke produced by smoldering material. The flux of smoke originatingwell after the active flaming combustion period with little or no verticalbuoyancy, and, therefore, most susceptible to subsidence inversions anddown-valley flows. (See Nuisance smoke). [1]

“Right-to-burn” Law A state law that provides liability protection for prescribed burners,providing they meet specified training and planning criteria. The degreeof liability protection varies by state.

Saturated adiabatic Adiabatic cooling in an atmosphere that is saturated with mois-lapse rate (SALR) ture. Usually about -3.0 degrees Fahrenheit per 1,000 feet (~-5.5 degrees

centigrade per kilometer). (See Adiabatic lapse rate; Dry adiabatic lapserate).

Scattering (light) An interaction of a light wave with an object that causes the light to beredirected in its path. In elastic scattering, no energy is lost to the object.[2]

Secondary aerosols Aerosol formed by the interaction of two or more gas molecules and/orprimary aerosols. [2]

Slash (see Activity fuel) [1]

Smoke concentration The amount of combustion products (in micrograms per cubic meter)found in a specified volume of air. [1]

Smoke intrusion Smoke from prescribed fire entering a designated area at unacceptablelevels. [1]


– 223 –

Smoke management The policies and practices implemented by air and natural resourcemanagers directed at minimizing the amount of smoke entering popu-lated areas or impacting sensitive sites, avoiding significant deteriorationof air quality and violations of National Ambient Air Quality Standards,and mitigating human-caused visibility impacts in Class I areas.

Smoke management A standard framework of requirements and procedures for man-program (SMP) aging smoke from prescribed fires, typically developed by States or

Tribes with cooperation from stakeholders.

Smoldering combustion Combined processes of dehydration, pyrolysis, solid oxidation,phase and scattered flaming combustion and glowing combustion, which occur

after the flaming combustion phase of a fire; often characterized by largeamounts of smoke consisting mainly of tars. Emissions are at twice thatof the flaming combustion phase. (See Combustion; Flaming combustionphase, Glowing combustion phase). [1]

Soot Carbon dust formed by incomplete combustion. [4]

Stable atmosphere A condition of the atmosphere in which vertical motion in the atmo-sphere is suppressed. Stability suppresses vertical motion and limitssmoke dispersion. In a stable atmosphere the temperature of a risingparcel of air becomes cooler than its surroundings, causing it to sink backto the surface. Also called stable air. (See Atmospheric stability; Un-stable atmosphere).

Standard atmosphere A horizontal and time-averaged vertical structure of the atmospherewhere standard atmospheric pressure at sea level is 1,013 mb, at 5,000feet (~1,500 m) it is 850 mb, at 10,000 feet (~3,000 m) it is 700 mb, andthe standard atmospheric pressure at 20,000 feet (~6,000 m) is 500 mb.Actual pressure is nearly always within about 30% of standard pressure.(See Atmospheric pressure).

State Implementation Plans devised by states to carry out their responsibilities under thePlan (SIP) Clean Air Act. SIPs must be approved by the U.S. Environmental Protec-

tion Agency and include public review. Same as Tribal ImplementationPlan (TIP). [5]

Subsidence inversion An inversion caused by settling or sinking air from higher elevations.(See Atmospheric inversion; Temperature inversion).


– 224 –

Sulfur dioxide (SO2) A gas (SO2) consisting of one sulfur and two oxygen atoms. Of interestbecause sulfur dioxide converts to an aerosol that is a very efficient atscattering light. Also, it can convert into acid droplets consisting prima-rily of sulfuric acid. (See Criteria pollutants). [2]

Sulfur oxides (SO) A class of colorless, pungent gases that are respiratory irritants andprecursors to acid rain. Sulfur oxides are emitted from various combus-tion or incineration sources, particularly from coal combustion. [3]

Temperature inversion In meteorology, a departure from the normal decrease of temperaturewith increasing altitude such that the temperature is higher at a givenheight in the inversion layer than would be expected from the tempera-ture below the layer. This warmer layer leads to increased stability andlimited vertical mixing of air. [2]

Ten-hour timelag fuels Dead fuels consisting of roundwood 1/4 to l-inch (0.6 to 2.5 cm) indiameter and, very roughly, the layer of litter extending from immedi-ately below the surface to 3/4 inch (1.9 cm) below the surface. Fuelelements of this size usually respond to changes in moisture within aboutten hours or less than a day, hence the term 10-hr timelag. (See Fuelmoisture content). [1]

Total fuel All plant material both living and dead that can burn in a worst-casesituation. [1]

Tribal Implementation Plans devised by tribal governments to carry out their responsi-Plan (TIP) bilities under the Clean Air Act. TIPs must be approved by the U.S.

Environmental Protection Agency and include public review. Same asState Implementation Plan (SIP). [5]

Understory burn A fire that consumes surface fuels but not overstory trees (in the case offorests or woodlands) and shrubs (in the case of shrublands).

Unstable atmosphere A condition of the atmosphere in which vertical motion in the atmo-sphere is favored. Smoke dispersion is enhanced in an unstable atmo-sphere. Thunderstorms and active fire conditions are common inunstable atmospheric conditions. In an unstable atmosphere the tempera-ture of a rising parcel of air remains warmer than its surroundings,allowing it to continue to rise. Also called unstable air. (See Atmo-spheric stability; Stable atmosphere).


– 225 –

Ventilation index An index that describes the potential for smoke or other pollutants toventilate away from its source. Also called clearing index. It is theproduct of mixing height and the mean wind within the mixed layer(trajectory wind).

Visual range Maximum distance at which a given object can just be seen by an ob-server with normal vision. [1]

Volatile Organic Any compound of carbon, excluding carbon monoxide, carbonCompounds (VOC) dioxide, carbonic acid, metallic carbides or carbonates, and ammonium

carbonate that participates in atmospheric photochemical reactions. [3]

Wet-bulb temperature Originally, the temperature measured with a mercury thermometer whosebulb is wrapped in a moist cloth. Commonly it is a measure of theatmospheric temperature after it has cooled by evaporating moisture.(See Dry-bulb temperature; Dew point).

Wildland Fire Any non-structure fire, other than prescribed fire, that occurs in thewildland. This term encompasses fires previously called both wildfiresand prescribed natural fires. [6]

Wildfire An unwanted wildland fire. This term was only included [in the newFederal policy] to give continuing credence to the historic fire preventionproducts. This is NOT a separate type of fire under the new terminology.[6]

Wildland Fire (See Wildland Fire Use) [6]Managed for ResourceObjectives

Wildland Fire Use The management of naturally ignited wildland fires to accomplish spe-cific pre-stated resource management objectives in predefined geographicareas outlined in Fire Management Plans. Wildland fire use is not to beconfused with “fire use,” which is a broader term encompassing morethan just wildland fires. [6]

Wildland Urban The line, area, or zone, where structures and other human devel-Interface (WUI) opment meet or intermingle with undeveloped wildland or vegetative

fuel.


– 226 –


1. National Wildfire Coordinating Group. 1996.Glossary of wildland fire terminology. PMS205. Boise, ID: National Wildfire CoordinatingGroup, National Interagency Fire Center. 162p.

2. Malm, William C. 1999. Introduction tovisibility. Cooperative Institute for Research inthe Atmosphere. NPS Visibility Program,Colorado State University, Fort Collins, CO:79 p.

3. U.S. Environmental Protection Agency. 1999.Handbook for criteria pollutant inventorydevelopment: a beginner’s guide for point andarea sources. EPA-OAQPS-454-99-037.Research Triangle Park, NC. U.S. Environmen-tal Protection Agency, Office of Air QualityPlanning and Standards.

4. Terms of Environment. [online]. 1998. U.S.Environmental Protection Agency, Office ofCommunication, Education, and Public Affairs.Available: URL http://www.epa.gov/OCEPAterms/intro.htm [2001, Dec.].

5. Policy for categorizing fire emissions. [online].2001. Natural Background Task Team, FireEmissions Joint Forum, Western Regional AirPartnership. Available: URL [2001, Nov.].

6. USDI and USDA Forest Service. 1998. Wild-land and prescribed fire management policy-implementation procedures reference guide.National Interagency Fire Center, Boise, ID. 81pp. and appendices.

smoke management guide for - us forest servicesmoke management guide for prescribed and wildland...

Documents