This file was created by scanning the printed publication.

Errors identified by the software have been corrected;

however, some errors may remain.

United States
Department of
Agriculture

Forest Serv~ce
Charts for
Intermountain
Forest and Range
Experiment Station
Interpreting Wildland
Ogden, UT 84401

General Technical Fire Behavior

Report INT-131
September 1982 Characteristics
Patricia L. Andrews
and Richard C. Rothermel
 THE AUTHORS RESEARCH SUMMARY
PATRICIA L. ANDREWS is a mathematician stat~onedat The fire characteristics chart is proposed as a
the Northern Forest Fire Laboratory in Missoula, Mont. graphical method of presenting two primary character-
She rece~vedher B.A. in mathematics and chemistry from istics of fire behavior: spread rate and intensity. Its
Eastern Montana College, Bill~ngs,in 1970, and her M.A. primary use is communicating and interpreting either site-
in mathematics and computer science in 1973 from the specific predictions of fire behavior or National Fire-
University of Montana, Missoula. She has been employed Danger Rating System (NF3RS) indexes and components.
at the Northern Forest Fire Laboratory since 1973. Rate of spread, heat per unit area, flame length, and
fireline intensity are plotted on a fire behavior chart.
RICHARD C. ROTHERMEL is a research engineer Spread component, energy release component, and
stationed at the Northern Forest Fire Laboratory in burning index are plotted on an NFDRS chart.
Missoula, Mont. Rothermel received his B.S. degree in Specific examples illustrate use of a fire characteristics
aeronautical engineering at the University of Washington, chart in conjunction with fire prescriptions, fire behavior
Seattle, in 1953. He served in the U.S. Air Force as a forecasts, fire .management plans, and briefings.
special weapons aircraft development officer from The equations used in creating the charts are given;
1953-55. Upon his discharge he was employed at Douglas and a method of obtaining heat per unit area from fire
Aircraft Company as a designer and troubleshooter in the behavior nomograms is illustrated.
Armarnent Group. From 1957 to 1961 Rothermel was em-
ployed by General Electric Co. in the aircraft nuclear pro-
pulsion department at the National Reactor Testing
Station in Idaho. In 1961, Rothermel joined the Northern
Forest Fire Laboratory where he has been engaged in
research on the mechanisms of fire spread. He received
his master's degree in mechanical engineering at the
University of Colorado, Fort Collins, in 1971. He was
project leader of the Fire Fundamentals research work
unit from 1966 until 1979, and is currently project leader
of the Fire Behavior research work unit at the fire
laboratory.

The use of trade, firm, or corporation names in this publication is for the
information and convenience of the reader. Such use does not constitute an
official endorsement or approval by the U.S. Department of Agriculture of any
product or service to the exclusion of others which may be suitable.

Approved for publication by Intermountain Station

September 1981
United States
Department of
Agricullure
Forest Serv~ce
Charts for
Intermountain
Forest and Range
Experiment Station
Interpreting Wildland
Ogden, UT 84401
General Techn~cal
Fire Behavior
Report INT-131
September 1982 Characteristics
Patricia L. Andrews
and Richard C. Rothermel

FIRE CHARACTERISTICS CHARTS

A fire characteristics chart is a graph that illustraies
the two primary characteristics of fire behavior-spread Fire Characteristics Chart
rate and intensity. Overall fire severity, a s well a s the
character of the fire, can be inferred from the location
of a point representing the fire on the chart. The chart
is mainly useful a s a communication aid. The fire char-
acteristics chart can be used for site-specific predictions
of fire behavior (fig. 1) or for National Fire-Danger Rat-
ing System (NFDRS) indexes and components (fig 2).
There is a pressing need for analysis of fire behavior
and a clear understanding of the analysis a t all levels of
management. Fire policy on National Forests has
shifted emphasis from fire control to fire management.
In addition to traditional fire control and use activitles,
a successful fire manager must also evaluate altcrnative
fire management strategies in relation to land and
resource management objectives. A vital part of this
process is communication with other resource special-
ists who may not be familiar with the National Fire-
Danger Rating System or methods for predicting slte-
specific fire behavior. Quantitative descriptors of fire
behavior are becoming more widely used due to the
prevalence of automated systems (Rothermel 1980).
The hand-held TI-59 calculator with a Fire DangeriFire
Behavior Custom Read Only Memory (CROM)is an
examplc of techllology that is reaching every level of
lire manager- from dispatchers to regional planners
(Burgan 1980). Fire characteristics charts allow graphic
presentation of quantitative fire behavior information in 0
a form that is readily understood. HEAT PER UNIT AREA, BTUl FT2
Figure I.-Fire behavior fire characteristics
chart.
 combustion zone (Byram 1959).This value has been
directly related to flame length, a n observable char-
acteristic of fire behavior. Fireline intensity is indicative
Fire Characteristics Chart of the heat that would be experienced by a person
working near the fire.
Flame length and fireline intensity can be interpreted
in terms of suppression capabilities a s shown in table 1.
The curved lines on the fire behavior chart define the
areas of interpretations shown in table 1. The inter-
pretations range from fires being easily controlled by
hand crews, to fire o ich equipment can be effec-
tive, lo fires on whlcl ,trol effort a t the head will be
ineffective.
As a n illustration of how the fire behavior chart
works, the fire behavior prediction values listed in table
2 are plotted in figure 3. These predictions are for fires
in three fuel types burning under the same wind, slope,
and fuel moisture conditions. The fuel types were
chosen from the 13 stylized fire behavior fuel models
(Anderson 1982). Fuel model 1 represents continuous
stands of arid western grass: fuel model 10, litter and
understory of a timber stand with heavy accumi~lations
of deadfall; and fuel model 8,short-needle litter.
The differences among the characteristics of the fires
in these three fuel types are readily apparent from their

r-
FIRE BEHAVIOR
Fire Characteristics Chart

ENERGY RELEASE COMPONENT, ERC

F~gure2.-Natronal F~re-DangerRating
System (NFDRS)frre characterrstrcs chart.
b
FIRE BEHAVIOR CHARTS .t\f'
The values that are plotted on the fire behavior charts
are based on a mathematical model for predicting fire
spread in wildland fuels (Rothermel 1972). Calculations
are made using nomograms (Albini 1976a), computer
(Albini 1976b), or calculator (Burgan 1979). The model
estimates actual fire behavior from specific descriptions
of fuel type, fuel moisture, topography, and wind. The
four descriptors of fire behavior that are plotted as a
single point on the chart are:
1. Rate of spread (chainslhour),the forward rate of
spread at the head of a surface fire.
2. Heat per unit area ( ~ t u l f t a~ )measure
, of the
amount of heat that is released by a square foot of fuel
while the flaming zone of the fire is in that area.
(Procedures for obtaining this value from nomograms
and the equation for modifying existing computer pro-
grams are given in appendix A.)
3. Flame length (feet), the length of the flame at the
head of the fire measured from the middle of the com-
bustion zone to the average position of the flame tip.
Flame length is determined by the rate of spread and HEAT PER U N I T AREA. BTUI Fr2
the heat per unit area of the fire.
4. Fireline intensity (Btulftls], the amount of heat Figure 3.-Fire behavior predictions from
released per second by a foot-wide slice of the flaming table 2 plotted on a fire behavior chart.
Table 1.-Fire suppression interpretations of flame length and fireline intensity

Flame length Fireline internsity Interpretation

Feet B tulf tls
<4 < 100 Fire can generally be attacked at the head or flanks by persons using handtools.
Handline should hold the lire.
4-8 100-500 Fires are too intense for direct attack on the head by persons using handtools.
Handline cannot be relied on to hold fire.
Equipment such as plows, dozers, pumpers, and retardant aircraft can be effective.
500 - 1,000 Fires may present serious control problems-torching out, crowning, and spotting.
Control efforts at the fire head will probably be ineffective.
'11 > 1,000 Crowning, spotting, and major fire run's are probable.
Control efforts at head of fire are ineffective.

Table 2.-Fire behavior predictions for specific fuel, moisture, wind, and slope conditions

Fire

Descriptors
Fuel and environmenthl conditions:
Fuel model
Dead fuel moisture, percent
Live fuel moisture, percent
Midflame windspeed, milh
Slope, percent

Fire behavior predictions:

Rate of spread, chainslh
Heat per unit area, Btulft2
Fireline intensity, Btulftls
Flame length, ft

placement on the chart. Notice that flame length and Although a point on the chart represents the charac-
fireline intensity are essentially the same for fires A and teristics of a fire, a circle around the point would more
B. Fire A is very fasi spreading and has a low heat per appropriately indicate the probable range of fire
unit area. On the other hand, fire B is slow spreading, behavior. The numbers used to characterize fire
but has a high heat per unit area. Both fires A and B behavior are a best estimate based on a mathematical
are predicted to be too intense for direct attack a t the model, and are subject to the assumptions and limita-
head by persons using handtools (table l ) ,but this tions of that model as described by Rothermel (1972)
degree of severity is caused by fires of very different and Albini (1976a). In addition, fire is inherently vari-
character. able and cannot be uniquely described over a n area.
Fire C occurs under the same wind, slope, and fuel The circle around a point becomes larger with more
moisture conditions a s fires A and B. But the handline nonuniform fuels, more variable wind, and increasing
should easily hold fire C, whereas plows, dozers. fire severity. There are no simple techniques for esti-
pumpers, or retardant would probably be required on mating the range of fire behavior characteristics caused
fires A and B. Notice that the character of these three by nonuniformities a t this time. I

fires can be determined by a glance at the chart. The Because several fire behavior characteristics are
same information is on table 2, but is not as easily plotted a s a single point and because it is easier to
interpreted. interpret illustrations than arrays of numbers, the fire
In general, fires with a high rate of spread and low behavior chart lends itself to many applications. The
heat per unit area are plotted near the upper end of the chart can be used to illustrate the fire management
vertical axis, whereas fires with low rates of spread and activities and associated fire characteristics listed below:
high heat per unit area fall to the right, near the hori- Project fires
zontal axis. Fires with both high spread rate and high Expected fire behavior given a s written narrative
values of heat per unit area will lie in the center of the in the fire behavior forecast.
graph far from the origin. The overall severity of the Expected change in fire behavior that may ac-
fire is shown by increasing flame length and fireline company a forecasted weather change.
intensity curves. @ Level of fire behavior considered in a n escaped
 fire analysis. burning plan, and report. The fire prescription portion
Expected change in fire behavior with a change includes a section on fire behavior and associated en-
in fuel type. vironmental conditions desired to meet burn objectives.
Prescribedfire Increasingly specific fire management objectives have
Fire characteristics required to achieve specific created a need for more quantitative descriptions of fire
burn objectives. behavior. Rate of spread described a s fast or slow, or
Fire behavior expected under a range of weather fire intensity a s hot or cool, is often not adequate. The
conditions. four values plotted on the fire cliaracteristics chart
Behavior of a n escaped fire. describe aspects of fire behavior that are important in
Conditions that will require using ignition both fire control and fire effects considerations. Flame
patterns to draw the fire and produce the desired length and fireline intensity are directly related Lo the
intensity. effectiveness of control forces. Many prescribed burns
Behavior of strip fires-bracketed by plotting the are conducted under conditions that produce flame
behavior of a backing fire and a free-burning lengths less than 3 feet. Rothermel and Deeming (1980)
head fire. have suggested that fireline intensity be correlated to
Long-range planning fire effects in the flames or in the convection column,
Variation of fire behavior between planning units and heat per unit area be correlated to fire effects near
under the same weather conditions. the base of the fire in the duff and litter. Fire behavior
Effect of a change in fuel type on fire behavior. can be quantified on a fire characteristic chart without
dwelling on tables and numbers. An example of a sec-
Application of Fire Behavior Charts tion of a fire use plan and report and the associated fire
EXAMPLE: FIRE PRESCRIPTIONS behavior chart are shown in figures 4 and 5.
Successful prescribed burning requires planning.
Fischer (1978) propbses a four-point fire use plan a n d
report: treatment area and objectives, fire prescription,

300 -
FIRE BEHAVIOR
Part 2 - Fire Prescription Fire Characteristics Chart
2.1 Treatment Specifications

2.11 Desired Accomplishment 250 -.

- Kiii shrubs and trees C 5 " D.B.H. without kiillng the
overstory of western larch.
4 s Fee - bumin4 he~dGre
- Expose mineral soils over 60% of the area B : Backin Cire
200 - 9
- Reduce duff by 86%.
c: behavior
Desired &.;-re
2.12 Desired Fire Behavior

- The fiame iength must be iess than 4 feet to ensure that

less than 60% of the crowns will be scorched, therefore
aliowlng the larch t o survive.

- Due t o dlscontinulty of fuels, a head flre wlth a flame

length of at least 2 feet IS required to achleve a unlform
burn pattern

- The fiame iength should be iess than 4 feet to reduce the

risk of torching and spot fires. The burn area is within
a mile of some 2.year old logging siash. if a spot f r e
occurs in the siash, it wiii require pumpers and possibiy
retaidant for control.

- Under the burning conditions that are required t o meet

desired duff reduction objectives, a free-burning, forward.
spreadng fire would make torching. crowning, and spottirig
probable. Strip headfires will be used to keep the fire
behavior w t h i n the required range.

2.13 Required Environmental Conditions

Relative Humidity - 20.30%

Temperature - 70-80 F
Windspeed (midflame) - N-NW - 3-7 0 500 1000 1500 2000
1.H Fuel Moisture - 10.14%
10.H Fuel Moisture - 12.14% HEAT PER U N I T AREA, BTUI d 2
Live Fuel Moisture - 75.125%
Duff Moisture - 50%
Figure 5.-Use of a fire characteristics chart
Figure 4.-The fire prescription part of a fire t o illustrate the desired fire behavior portion
use plan and report. of the prescription given in figure 4.
EXAMPLE: FIRE BEHAVIOR FORECAST nomograms or the TI-59 calculator. The numerical out-
A fire behavior officer (FBO) is normally part of a n put from the model must be interpreted by the FBO
overhead team on a project fire. His duties include before it is communicated. In addition to the narrative
identifying critical fire behavior conditions and project- fire behavior forecast and a map indicating expected
ing fire size. He prepares a fire behavior forecast which fire growth, a fire characteristics chart can be used in
includes a written narrative describing potential fire the oral briefing to illustrate potential fire behavior for
behavior. This is done for briefing the fire overhead different sectors of the fire. A fire behavior forecast and
team and is revised whenever conditions warrant a new the associated fire characteristics chart are shown in
appraisal of fire behavior. One of the tools the FBO uses figures 6 and 7.
in his job is the fire behavior model in the form of

FlRE BEHAVIOR FORECAST NO.

NAME OF FIRE: PREDICTION FOR: Day SHIFT
FOREST: SHIFT DATE:
TIME AND DATE
FORECAST ISSUED: 0900 SIGNED:
FlRE BEHAVIOR OFFICER

WEATHER SUMMARY RED FLAG ALERT SEE ATTACHED SPOT FORECAST

Note Terrain channeling of the general wind up the river will produce up-canyon winds of 5 to 10 milh
along the lower slopes and 10 to 15 milh along upper slopes by early afternoon.

FlRE BEHAVIOR
GENERAL: Fire will be relatively inactive until the inversion breaks, about 1200. Fire activity will
increase sharply at that time with fire spread mainly up-canyon to the southwest. Some torching
will occur where fuel concentrations exist with short range spotting possible.

SPECIFIC:
SECTOR A. Fire spread will be relatively low on this sector (3 chains per hour), but rolling fire-
brands will be a problem, especially in the small draws to the south. Roll into unburned fuels will result in
upslope runs with some torching and short range spotting.

SECTOR B. This will be a hot sector on the fire today. Direct attack with hand crews will be
marginal until 1100 and impossible after the inversion breaks. Roll and spotting will cause short runs to the
ridge, especially in the small draws to the south.

SECTOR C. The fire will back slowly down canyon against the wind. Rolling material may cause
some problems, but this will be the coolest sector on the fire. Conditions will be good for direct attack to
succeed.

AIR OPERATIONS: Strong inversion will limit air operations until about noon. Any thunderstornls that
occur later in the day will produce turbulent flying conditions.

SAFETY: Crews should be alert to the danger of roll igniting fuels below them on steep slopes.
If thunderstorms enter the fire area, be alert for the possibility of erratic fire behavior from down-
draft winds. A weather watch has been established to give warning of approaching thunderstorms.
Figure 6.-F~re behavior forecast.
 3. Burning index (B1)-related to the magnitude of
the fire containment problem. The burning index is
derived from the spread component and the energy
release component.
The three values are related to the corresponding
values on the fire behavior-fire characterisiics chart.
Procedures used to calculate spread component, energy
release component, and burning index were in fact
derived from the equations for rate of spread, heat per
unit area, and flame length. NFDRS indexes and com-
ponents are designed to give broad area rating of fire
potential and are not meant to predict actual site-
specific fire behavior. The indexes are based on fuel
models that describe large areas and on weather taken
a t a specific location once per day.
Because NFDRS components and indexes are relative
indicators of fire danger, a value is meaningful only
when it is compared with other values. Seasonal plots
illustrated in figure 8 help one make such comparisons.
An index value can be readily related to previous
values in the season or, if plots are overlaid, to those
from other seasons. As can be seen in figure 8, ERC
had climbed to the highest point thus far that season
by August 3, and did not change on August 4. How-
ever, SC for August 4 was four times a s high a s that on
August 3, and BI nearly doubled from August 3 to
August 4. Notice also that on October 3, BI was
approximately the same a s it had been on August 3,
but SC was m u c h higher and ERC much lower.
The fire characteristics chart offers another means of
HEAT PER UNIT AREA, BTUI FT2 interpreting SC, ERC, and BI in terms of potential fire
behavior. The values for August 3, August 4, and
Figure 7.-Use of a fire characteristics chart
October 3 are plotted on the NFDRS chart in figure 9.
to illustrate the fire behavior glven in the f ~ r e The curved lines on the fire characteristics chart cor-
behavior forecast in figure 6. respond to the horizonal lines on the seasonal plot of
burning index. Notice how the character of fire danger
on all 3 days is readily apparent.
NATIONAL FIRE-DANGER RATING The seasonal plots and the fire characteristics chart
SYSTEM CHARTS illustrate different aspects of SC, ERC, and BI. A
The National Fire-Danger Rating System [NFDRS) seasonal plot reflects the change of conditions in a
(Deeming and others 1977) is used throughout the component or index over time, whereas the NFDRS
country for fire management planning. Most users chart is best suited for illustrating the relationship
obtain daily ratings using a centralized computer among all three values a t a particular time.
(Helfman and others 1980): some use nomograms NFDRS calculations are normally done on a day-to-
(Burgan and others 1977), or the TI-59 calculator with day basis using standard weather observations a s
a fire dangerifire behavior CROM (Burgan 1979). input. The TI-59 calculator with a fire dangerlfire
Management guidelines are often based on indexes b;havior CROM allows another option: fuel moisture
obtained by processing historical weather records can be entered directly in lieu of weather input. This
through standard computer programs (Main and others allows the user to look at situations that are not based
in press: Bradshaw a n d Fischer 1981a, 1981b). on the seasonal trend. For example, "under the same
T h e NFDRS is comprised of many components a n d conditions, how does the fire danger change with a
indexes related to fire occurrence, fire behavior, and fire change of fuel models?" or "what would the fire danger
suppression. The following items can be plotted on a be if the moisture content of the large fuels were 3
fire characteristics chart similar to the one used for site- percent less?" The NFDRS chart is a n ideal way of
specific fire behavior estimation: illustrating the change in fire danger due to a specific
1. Spread component (SC)-related to rate of fire change in conditions.
spread.
2. Energy release component (ERC)-related to
energy or heat that will be released in a passing fire
front.
ERC

MAY JUNE JULY AUG. SEPT. OCT.

Figure 8.-Seasonal plots of spread component (SC),

energy release component (ERC), and burnrng index (BI).
 The Troy Ranger District in northwestern Montana
NATIONAL FIRE DANGER
prepared a fire management plan that covered the
RATING SYSTEM
entire district. The plan was approved in early 1979,
Fire Characteristics Chart and several unplanned ignitions were used a s pre-
scribed fires that summer. A fire management prescrip-
tion covers each of five fire management categories.
Figure 10 is a flowchart for managing fires on big-game
spring and winter ranges on operational fire manage-
ment areas. Among the conditions that must be satis-
fied for a fire to be allowed to burn in these areas are
limits on burning index and energy release component.
The shaded area of the NFDRS chart: in figure 11 illus-
trates these limits. If the plotted points from the last 4
days fall to the left of the vertical line where ERC = 30,
and if the current day's point and the point forecasted
for the next day fall in the smaller area limited by BI =
28. then the portion of the fire prescription based on
the National Fire-Danger Rating System is satisfied.
As the fire season progresses, current information on
weather, fire danger, fire activity, and prescription cri-
teria must be readily available, as illustrated in figure
12. The NFDRS chart is a visual aid that clearly illus-
trates the limits determined from the National Fire-
Danger Rating System.

ENERGY RELEASE COMPONENT, ERC

-,
Fire Start

+I
Lifei property endangered ?
4
-
Smoke management conditions favorable ?
Yes
Suppress

Suppress
I Yes
F~gure9 -SC, ERC, and BI for August 3,
August 4, and October 3 from the seasonal
plots In f~gure8 plotted on an NFDRS chart.
Suppress
A review of each index and the relationship between
them will clarify the use of the NFDRS chart. Spread
component is related to the rate of fire spread and is I + Yes
Fire weather,acceptable ? Nob
Suppress
strongly affected by windspeed; SC can therefore have 1 Yes
wide daily fluctuations. On the other hand. windspeed Fire behavior acceptable ? N ob Suppress
I Yes
has no effect on energy release component. ERC is
related to the energy released in the flaming zone and
+
Forecasted f i r e weather and behavior acceptablep?-.
NO
Suppress
I Yes
mainly reflects changes in fuel moisture. Because ERC
+
is not affected by wind, it shows a more definite
seasonal trend than either SC or BI. Burning index is
1 Equipment and manpower a,vailabe if needed ?p-
NO
Suppress

derived from the spread and energy release components

and has the same underlying trend a s ERC with the
No , Suppress

daily fluctuations of SC imposed on it. Depending on

the type of management decisions to be made, SC,
No , Suppress

ERC, BI, or a combination of these values can be used. / Yes

Boundary threatened ? Yes+
Suppress

Application of NFDRS Charts

EXAMPLE: FIRE MANAGEMENT PLANS
1
Allow t o b u r n ,
Continue t o evaluate
(limited
o r total)

USDA Forest Service fire management policy cur-

rently recognizes two categories of fires on its lands: +I
Management objective satisfied ?
wildfires and prescribed fires. Every wildfire requires an I
appropriate suppression response. Every prescribed fire Contain and a$\hl to b u r n out
is to be conducted in compliance with an approved
plan. Unplanned ignitions can be allowed to burn a s Figure 10.-Example flow chart for
prescribed fires a s long a s they meet criteria established managing fires on big-game spring and
in approved fire prescriptions. winter ranges on operational fire
management areas.
 EXAMPLE: BRIEFINGS
During the fire season, briefings are often held to
describe the general fire situation. The audience can
include members of the news media, staff personnel, or
others who are not familiar with the National Fire-
Danger Rating System. Statements such a s "the fire
danger is high" or "the burning index is 85" can be
misinterpreted. The NFDRS chart can serve a s a yisual
aid and focal point for discussion. The chart can be
explained easily: a s a point falls farther to the right on
the chart, there is an indication that fires will be hotter.
A point falling farther u p on the chart indicates the
potential for faster spreading fires. Overall severity
increases as a point falls farther from the origin in
either direction.
Consider the following briefing of fire danger of a
USDA Forest Service region, represented by the fire
characteristics chart in figure 13.
The fire danger of most of the west side of the
region is low a s indicated by point A, although
there are a couple of districts that may cause prob-
lems [point B). Point C refers to the fire danger on
the east side of the region. If we have another week
of dry weather, the situation on the east side could
become critical (point D).

RATING SYSTEM
0 20 40 60 80 100 120 140
Fire Characteristics Chart
ENERGY RELEASE COMPONENT, ERC
F~gure11 -NFDRS chart used to illustrate
the llm~tson BI and ERC glven In the flow
chart In flgure 10 For the port~onof thls flre
prescr~pt~on that I S based on NFDRS
Indexes to be satsfled, plotted polnts from
the last 4 days must fall to the left of the
vertlcal llne where ERC = 30 and the
current day's polnt and the polnt forecasted
for the next day must fall In the smaller area
llmlted by the BI = 28 Ilne.

Figure 12.-The fire characteristics chart ENERGY RELEASE COMPONENT, ERC

can be part of a display that includes
Figure 13.-NFDRS chart used in a briefing
current information on weather, fire danger,
of current and potential fire danger.
fire activity, and prescription criteria that a
fire manager uses to make decisions.
FORMS OF THE CHART 11 may be necessary to change the scales of a chart
primarily used for a specific fuel type. For example, a
The fire behavior chart and the NFDRS chart illus-
chart primarily used to display fire behavior predictions
trated in figures 1 and 2 have scales that do not cover
in logging slasli should have a n expanded heat-per-unit-
the entire range of possible values. The scales were
area axis and a truncated rate-of-spread axis. Such a
chosen to give resolution to the lower values and allow
chart is shown in appendix B, along with other full-size
the majority of values to be plotted on the chart. Never-
charts suitable for reproduction. Appendix A offers the
theless, some points will fall beyond the limits of the
equations used in creating the fire characteristics
graph. A form of the fire behavior chart that overcomes
charts.
this limitation uses logarithmic scales on the axes. In
An NFDRS chart can illustrate NFDR manning
this way the entire range of values can be covered. The
classes for a specific area. USDA Forest Service fire
curved flame length lines become straight lines on the
specialists generally determine manning classes from
log scale version of the chart.
the 90th and 97th percentile burning index values for a
Figure 14 shows six points plotted on both linear a n d
specific station and fuel model. The percentile values
logarithmic scale fire behavior charts. This illustrates
are determined from historical fire weather obser-
that a relatively small change in the behavior of a low-
vations and provide criteria for ranking the relative
intensity fire is amplified, and a large change in the
severity of the burning conditions on a given day.
behavior of a high-intensity fire is compressed on the
Appendix A includes a n example showing how to
logarithmic chart. Because a primary purpose of the fire
design a n NFDRS chart where BI lines designate
characteristics chart is to visually illustrate changes in
manning class levels.
fire behavior, care should be taken in interpretation of
relative location or points plotted on the logarithmic
chart.

FIRE BEHAVIOR
Fire Ctiaracteristics C h a r t
(Lcqarithmic Scale)

HEAT PER UNIT AREA, BTUl FTZ 148

Figure 14.-The same six points (A, B,C. X, Y.Z) plotted for comparison on both a linear
scale (A) and a logarithmic scale (B)fire behavior chart.
SUMMARY Burgan, Robert E. Fire dangerlfire behavior computa-
Fire characteristics charts can be used to plot either tions with the Texas Instruments TI-59 calculator:
site-specific fire behavior predictions or National Fire- user's manual. Gen. Tech. Rep. INT-61. Ogden, UT:
Danger Rating System indexes. Because several aspects U.S. Department of Agriculture, Forest Service, In-
of fire behavior are plotted as a single point, the chart termountain Forest and Range Experiment Station;
aids a user in assessing overall fire cliaracteristics. The 1979. 25 p.
chart lends itself to a wide range of potential uses, the Byram, G. M. Combustion of forest fuels. In: Forest'
most significant being communication of quantitative fire: control and use. Edited by K. P. Davis.
values in a form that is easily understood. McGraw-Hill, New York, 1959. pp. 61-89.
Deeming, John E.; Burgan, Robert E.; Cohen, Jack D.
The national fire-danger rating system-1978. Gen.
Tech. Rep. INT-39. Ogden, UT: U.S. Department of
PUBLICATIONS CITED Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station; 1977. 63 p.
Albini, Frank A. Estimating wildfire behavior and
effects. Gen. Tech. Rep. INT-30. Ogden, UT: U.S. Fischer, William C. Planning and evaluating prescrib-
Department of Agriculture, Forest Service, Inter- ed fires-a standard procedure. Gen. Tech. Rep.
mountain Forest and Range Experiment Station; INT-43. Ogden, UT: U.S. Department of Agri-
1976a. 92 p. culture, Forest Service, Intermountain Forest and
Albini, Frank A. Computer-based models of wildland Range Experiment Station; 1978. 19 p.
fire behavior: a user's manual. Ogden, UT: U.S. Helfman, Robert S.; Straub, Robert J.: Deeming,
Department of Agriculture, Forest Service, Inter- John E. User's guide to AFFIRMS: time share
mountain Fprest and Range Experiment Station; computerized processing of fire danger rating. Gen.
1976b. 68 p. Tech. Rep. INT-82. Ogden, UT: U.S. Department of
Anderson, Hal E. Heat transfer and fire spread. Res. Agriculture, Forest Service, Intermountain Forest and
Pap. INT-69. Ogden, UT: U.S. Department of Agricul- Range Experiment Station; 1980. 150 p.
ture, Forest Service, Intermountain Forest and Range Main, William A.; Straub, Robert J.; Paananen,
Experiment Station; 1969. 20 p. Donna M. FIREFAMILY: fire planning with historic
Anderson, Hal E. Aids to determining fuel models for weather data. Gen. Tech. Rep. St. Paul, MN: U.S.
estimating fire behavior. Gen. Tech. Rep. INT-122. Department of Agriculture, Forest Service, North
Ogden, UT: U.S. Department of Agriculture, Forest Central Forest and Range Experiment Station; in
Service, Intermountain Forest and Range Experiment press.
Station; 1982. 22 p.
Rothermel, Richard C. How to predict the spread and
Bradshaw, Larry S.; Fischer, William C. A computer intensity of forest and range fires. Gen. Tech. Rep.
system for scheduling fire use. Part I: the system. INT- . Ogden, UT: U.S. Department of Agri-
Gen. Tech. Rep. INT-91. Ogden, UT: U.S. Department culture, Intermountain Forest and Range Experi-
of Agriculture, Forest Service, Intermountain Forest ment Station. In preparation.
and Range Experiment Station; 1981a. 6 3 p. Rothermel, Richard C. A mathematical model for pre-
Bradshaw, Larry S.; Fischer, William C. A computer dicting fire spread in wildland fuels. Res. Pap.
system for scheduling fire use. Part II,: computer INT-115, Ogden, UT: U.S. Department of Agriculture,
terminal operator's manual. Gen. Tech. Rep. Forest Service, Intermountain Forest and Range
INT-100. Ogden, UT: U.S. Department of Agriculture, Experiment Station; 1972. 40 p.
Forest Service, Intermountain Forest and Range Rothermel, Richard C. Fire behavior systems for fire
Experiment Station; 1981b. 3 3 p. management. In: Sixth conference on fire and forest
Burgan, Robert E.; Cohen, Jack D.; Deeming, J o h n E. meteorology: proceedings; 1980 April 22-24; Seattle,
Mannually calculating fire-danger ratings- 1978 WA. Washington, DC: Society of American Foresters;
National Fire-Danger Rating System. Gen. Tech. Rep 1980: 58-64.
INT-40. Ogden, UT: U.S. Department of Agriculture, Rothermel, Richard C.; Deeming, John E. Measuring
Forest Service, Intermountain Forest and Range and interpreting fire behavior for fire effects. Gen.
Experiment Station; 1977. 49 p. Tech. Rep. INT-93. Ogden, UT: U.S. Department of
Burgan, Robert E. A handheld calculator-fire danger Agriculture, Forest Service, Intermountain Forest and
and fire behavior. In: Sixth conference on fire and Range Experiment Station; 1980. 3 p.
forest meteorology: proceedings; 1980 April 22-24;
Seattle, WA. Washington, DC: Society of American
Foresters; 1980: 65-69.
APPENDIX A To modify an existing computer program, use a
reformulation of equation (2)
Equations for Compiling the Fire
Characteristics Charts
The equation used to plot the curves on the NFDRS
chart is
where
0.091 x B I ~ . " HA = heat per unit area, ~ t u l f t ~
SC =
ERC IB = fireline intensity, Btu/ft/s
R = rate of spread, chainslh.
where
SC = spread component
ERC = energy release component The X axis of the upper right-hand graph of the
BI = burning index. nomograms as originally published by Albini (1976a) is
reaction intensity (Btulft21min).The label was
The equation used to plot the curves on the fire eliminated on subsequent revisions for fire behavior
behavior chart officer (FBO) training. The nomograms to be published
by Rothermel (in preparation) will have heat per unit
area on that axis,

where The relationship between reaction intensity and heat

R = rate of spread, chainslh per unit area is
IB = fireline intensity, Btulftls
HA = heat pkr unit area, ~ t u l f t ~ .

The relationship between fireline intensity and flame where

length is given by the following equation: HA = heat per unit area, Btulft2
IR = reaction intensity, Btu/ft2/min
F, = 0.45 1~0.4" (3) = characteristic surface-area-to-volume ratio of
where the fuel array, ft2/ft3
FL = flame length, ft 384
m
= residence time, min (Anderson 1969).
0
IB = fireline intensity, Btulftls.
The labels for heat per unit area can be added to the
Determination of Heat Per Unit Area horizontal axis on the upper right-hand quadrant of the
Heat per unit area is a direct output of the TI-59 fire nomograms, using the scales given in figure 15. The
behavior program. However, it cannot be read directly scale is the same for both the low and high windspeed
from the nomograms, nor is it a n output value from options on all versions of the nomograms. Use of the
most available computer programs. Although a point nomograms does not change, a s illustrated in the
can be plotted on the fire behavior chart given rate of example in figure 16. Heat per unit area is read where
spread and fireline intensity or flame length, it is easier the first constructed vertical line intersects the newly
if a heat per unit area value is available. labeled axis.
 FUEL FUEL
MODEL HEAT PER UNIT AREA, BTU IFT2 MODEL HEAT PER UNIT AREA, BTU I F T ~
1 r 1 I I I 1 1
8 1 I 1 I
I I

0 50 150 250

Frgure 15.-Heat-per-unrt-area scales can be transferred to the high or low w~ndspeed

optrons of the frre behav~ornomograms (FBO or Albrnr 1976a). The scale for the
approprrate fuel model IS transferred to the horrzontal axis of the upper r~ght-handgraph
 Input: Output:
Fuel Model 8 Rate of spread = 1.75 chainslh
Dead fuel moisture = 8% Flame length = I ft.
Effective windspeed = 5 milh Fireline intensity = 5 Btulltls
Heat per unit area = 175 ~ f u l f t ~

8. C L O S E D T I M B E R L I T T E R - L O W WINDSPEEDS
2 FLAME LENGTH. FT

Figure 16.-Example of the use of a nomogram after the heat-per-unit-area label

has been transferred from figure 15.
Creation of a Custom NDFRS Chart 3. Draw the axes for the chart based on the
An example of a custom NFDRS chart is shown in maximum values for SC and ERC.
figure 17. This chart is for the Ninemile District (Lolo In this case, the axes are extended beyond the
National Forest), fuel model G. Manning classes are 10-year maximums to SC = 60 and ERC = 70.
based on burning index. 4. Locate the curves for each of the BI values
The steps to creation of such a chart are a s follows: determined in step 2 above using equation (1).
1. Run the FIRDAT program a s described in the Find SC for several values of ERC. The upper e ~ d -
User's Guide to FIREFAMILY (Main and others in points of the BI curves can most easily be located
press) lo determine the 90th and 97th percentile by using a n alternate farm of equation (1):
values of BI and the maximum SC and ERC 0.091 x ~ 1 2 . l ~
values. ERC =
SC
For this example, based on the 1970-79 weather
data, the 90th percentile of BI is 56.38 and the To locate the BI = 5 6 curve in figure 17, find ERC
97th percentile is 69.74. The maximum SC for this when SC = 6 0 (the maximum SC for this chart):
period was 49 and the maximum ERC was 60.
0.091 x 562.17
2. Calculate the manning class boundary values a s ERC = = 9.4
described in the User's Guide to AFFIRMS 60
(Helfman and others 1980). Find the other endpoint of the curve by calculating
For six manning classes in this example, the SC when ERC = 70 (the maximum ERC for this
results are: chart):
Displayed Upper value
manning for class
class (burning index)
1 l(M190)/4 = 14 Sirnilarly when ERC = 40 and 20, the calculated
2 (MI90)12 = 28 values for SC are 14 and 28, respectively. A
3- (MI90)(3/4) = 42 smooth curve is drawn through the following four
3+ (MI90) = 56 points:
4 MI97 = 70 ERC SC
5 More than MI97 9 60
20 28
40 14
ni his notation corresponds to that used in the AFFIRMS manual 70 8
MI90 = 90th percentile m a n n ~ n gindex and MI97 = 97th percentile
manning index. The curves for the other manning class divisions
(BI = 14, 28, 42, and 70) are located in a similar
manner.
 I \ \ L
NATIONAL FIRE DANGER
RATING SYSTEM
F i r e Characteristics C h a r t
NINEMILE DISTRICT

FUEL MODEL G

\ M a n n i n g Class
5

\ 4 \ Burnina

ENERGY RELEASE COMPONENT, ERC

F~gure17 -Example of a custom NFDRS chart, where the BI l ~ n e scorrespond to
manning class cutoff values. Mann~ngclasses are based on 1970-79 weather for Ninem~le
Ranger Statlon (241507)
APPENDIX B
Fire Characteristics Charts
Suitable for Copying
0 500 1000 1500 2000

HEAT PER UNIT AREA, BTUl FT*

GTR.INT-131 Charts for Interpreting Wildland Fire Behavior Characteristics

18
0 500 1000 1500 2000 2500 3000 3500 4000

HEAT PER UNIT AREA, BTUl F T ~

 FIRE BEHAVIOR
Fire Characteristics C h a r t
(Loqarithmic Scale)

HEAT PER UNIT AREA, BTUI FT2

GTR-INT-131 Charts for Interpreting Wildland Fire Behavior Characteristics

20
ENERGY RELEASE COMPONENT, ERC
GTR-INT-131 Charts for Interpreting Wildland Fire Behavior Charactersitics

21 %' U S . GOVERNMENT PRINTING OFFICE: 1982-576.026119 REGION No. 8

 Andrews, Patricia L.; Rothermel, Richard C. Charts for interpreting w~ldlandfire
behavior characteristics. Gen. Tech. Rep. INT-131. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1981. 21 p.

The fire characteristics chart is proposed a s a graphical method of

presenting two primary characteristics of fire behavior-spread rate and intensity.
Its primary use is c o m m u n i c a t ~ n gand interpreting either site-specific predictions
of fire behavior or National Fire-Danger Rating System (NFDRS) indexes and
components. Rate of spread, heat per unit area, flame length, and fireline
intensity, are plotted on a fire behavior chart. Spread component, energy release
component, and burning index are plotted on a n NFDRS chart. Specific examples
illustrate potential application.

1 KEYWORDS: fire behavior, fire spread, fire intensity, fire-danger rating, National
Fire-Danger Rating System
!
The Intermountain Station, headquarted in Ogden,
Utah, is one of eight regional experiment stations charged
with prov'i,ding scientific knowledge to help resource
managers meet human needs and protect forest and range
ecosystems.
The Intermountain Station includes the States of
Montana, Idaho, Utah, Nevada, and western Wyoming.
About 231 million acres, or 85 percent, of the land area in
the Station territory are classified as forest and rangeland.
These lands include grasslands, deserts, shrublands, alpine
areas, and well-stocked forests. They supply fiber for
forest industries; minerals for energy and industrial
development; and water for domestic and industrial con-
sumption. They also provide recreation opportunities for
millions of visitors each year.
Fi$lcj,programs and research work units o f the Station
are maintained-in

Boise, Idaho
Bbzeman, Montana (in cooperation with
Montana State University)
Logan, Utah (in cooperation with Utah State
University)
Missoula, Montana (in cooperation with the
University of Montana)
Moscow, Idaho (in cooperation with the
University of Idaho)
Provo, Utah (in cooperation with Brigham
Young University)
Reno, Nevada (in cooperation with the Univer-
sity of Nevada)