This file was created by scanning the printed publication.
Errors identified by the software have been corrected;
however, some errors may remain.
A Publication of the
National Wildfire
Coordinating Group
Sponsored by
United States
Department of Agriculture
United States
Department of the Interior
BEHAVE: Fire
Behavior Prediction
and Fuel Modeling
National Association of
State Foresters
FUEL Subsystem
Robert E. Burgan
Richard C. Rothermel
PMS 439-1
NFES 0275
MAY 1984
PREFACE
O v e r t h e p a s t decade t h e science o f f i r e
m o d e l i n g h a s made g r e a t a d v a n c e s .
T h e 13
o r i g i n a l f i r e b e h a v i o r f u e l models h a v e b e e n
used successfully to represent a wide a r r a y
o f fuel t y p e s i n t h e U n i t e d States. Nevertheless, f i r e managers, who are u s i n g f i r e
p r e d i c t i o n s in a n i n c r e a s i n g n u m b e r o f
applications, have f o u n d t h a t existing fuel
models d o n o t a d e q u a t e l y m a t c h some f u e l
situations.
T h e y therefore have developed
a need f o r techniques t h a t will enable them
t o m o d i f y e x i s t i n g f u e l models o r t o d e v i s e
e n t i r e l y n e w ones.
The purpose o f this
publication i s to p r o v i d e them w i t h t h i s
capability.
T h e FUELS s u b s y s t e m o f B E H A V E
contains programs t h a t w i l l enable f i r e
m a n a g e r s t o assemble f u e l models a n d t e s t
t h e i r performance before releasing them f o r
o p e r a t i o n a l use.
Fuel modeling is n o t y e t a
r i g o r o u s process; consequently science a n d
good judgment a r e b o t h needed.
Nevertheless, p i l o t tests have shown t h a t t h e
methods a r e r e a d y f o r application i n t h e
field b y well- trained personnel.
T h e programs contain new a n d simplif i e d p r o c e d u r e s f o r e x a m i n i n g f u e l s in t h e
f i e l d a n d d e v e l o p i n g f u e l models.
I t is n o t
a l w a y s n e c e s s a r y t o c o n s t r u c t new models,
h o w e v e r ; m o d i f i c a t i o n s t o e x i s t i n g models
may b e s u f f i c i e n t in some cases, w h i l e in
o t h e r s more r i g o r o u s f i e l d i n v e n t o r y proced u r e s may b e d e s i r a b l e .
There are four
w a y s t o o b t a i n a f u e l model f o r o p e r a t i o n a l
use i n BEHAVE:
1 . Choose o n e o f t h e 13 s t a n d a r d
models.
2. M o d i f y o n e o f t h e 13 s t a n d a r d
models.
3.
Use measured data t a k e n b y invent o r y techniques.
4.
Use t h e n e w f u e l m o d e l i n g p r o c e d u r e s d e s c r i b e d in t h i s m a n u a l .
T h e fastest solution is choosing one o f
t h e s t a n d a r d 13 models [ A n d e r s o n 1982).
I f t h a t does n o t s a t i s f y t h e u s e r , t h e most
r e p r e s e n t a t i v e model o f t h e 13 c a n b e
modified.
F o r example, one c a n c h a n g e
l o a d i n g a n d d e p t h , a d d g r e e n f u e l , make it
a d y n a m i c model, a n d so o n . I f m o d i f i cation i s n o t satisfactory, t h e n e x t fastest
expedient would b e t o use o u r new procedures.
Although any method o f measuring
a n d modeling fuels yields o n l y approximate
answers, o u r new procedures a r e simple,
i n e x p e n s i v e , a n d r a p i d t o use.
B u t if t h e
user prefers t o inventory, o r t o use previously inventoried data, t h e programs will
accommodate t h e f u e l loads b y size class
a n d w i l l a s s i s t t h e u s e r in p r o v i d i n g i n f o r mation n e e d e d t o assemble a complete f u e l
model.
Several features b u i l t i n t o t h e modeling
program c o n t r i b u t e t o reasonable fuel
models a n d f i r e p r e d i c t i o n s :
1.
T h e system will b u i l d either static o r
d y n a m i c models.
T h i s overcomes t h e p r o b lem t h a t t h e p r e s e n t 13 models a r e p r i m a r i l y
d e s i g n e d f o r t h e time o f y e a r w h e n f u e l s
are cured.
2. T h e p r o c e d u r e s a r e d e s i g n e d t o
combine t h e data f r o m m i x t u r e s o f l i t t e r ,
grass, s h r u b s , a n d slash t o p r o d u c e a
composite model.
I n this process, depths
a n d loads o f each t y p e a r e a d j u s t e d b y a r e a
covered.
S u c h a model s h o u l d b e c a r e f u l l y
examined, tested, a n d i t s f i r e p r e d i c t i o n s
compared w i t h f i e l d data a n d s t a n d a r d
models- - a t a s k s i m p l i f i e d b y t h e F U E L
programs.
3.
I f t h e f u e l s o c c u r in i n d i v i d u a l
p a t c h e s , models may b e b u i l t t o d e s c r i b e
t h e dominant fuel cover a n d t h e fuel t h a t
i n t e r r u p t s t h e dominant fuel. BURN will
u s e b o t h in t h e t w o - f u e l - m o d e l c o n c e p t
d e s c r i b e d b y R o t h e r m e l ( 1 983).
4.
T h e slash p r o c e d u r e s u t i l i z e several
t e c h n i q u e s f o r e s t i m a t i n g load.
These are
patterned after t h e research o f B r o w n
(1974) a n d i n c l u d e t h e n u m b e r o f i n t e r c e p t s
a s w e l l as load a n d d e p t h r e l a t i o n s h i p s .
T h e y also can u t i l i z e fuel photo series such
as t h o s e d e v e l o p e d b y F i s c h e r (1981a,
1981 b, 1981c), K o s k i a n d F i s c h e r (1 979),
a n d M a x w e l l a n d Ward (1978a, 1978b, 1979,
1980).
T h e site- specific fuel modeling techn i q u e s d e s c r i b e d in t h i s manual a r e a p p r o priate f o r constructing f i r e behavior fuel
models o n l y .
They are n o t intended f o r
c o n s t r u c t i n g National Fire- Danger Rating
f u e l models.
Basic differences between t h e
mathematical e q u a t i o n s u s e d in t h e f i r e
danger a n d f i r e behavior computer programs
preclude t h i s possibility.
These differences
o c c u r p r i m a r i l y in t h e p r o c e d u r e s f o r
weighting t h e influence o f v a r i o u s f u e l size
classes, t h u s p r o d u c i n g o u t p u t s meant t o
have different interpretations.
A s a result,
t o r e a s o n a b l y r e p r e s e n t t h e same " a c t u a l "
f u e l s s i t u a t i o n , a f i r e d a n g e r f u e l model
must b e assigned d i f f e r e n t values than a
f i r e b e h a v i o r f u e l model.
T h u s , f u e l models
are applicable only w i t h t h e f i r e processor
used t o c o n s t r u c t them, a n d t h e f i r e danger
processor is n o t p a r t o f t h e BEHAVE
system.
-,
d
I
T H E AUTHORS
RESEARCH SUMMARY
ROBERT E. BURGAN received h i s
bachelor's degree i n f o r e s t e n g i n e e r i n g i n
1963 a n d h i s master's degree i n f o r e s t f i r e
c o n t r o l i n 1966 from t h e U n i v e r s i t y o f
Montana. From 1963 t o 1969, he s e r v e d on
t h e t i m b e r management s t a f f o f t h e Union
a n d Bear- Sleds D i s t r i c t s , Wallowa-Whitman
National Forest. From 1969 t o 1975, h e was
a research f o r e s t e r o n t h e s t a f f o f t h e
I n s t i t u t e o f Pacific Islands F o r e s t r y ,
Honolulu, Hawaii. Since 1975, he has been
a t t h e N o r t h e r n Forest F i r e L a b o r a t o r y ,
Missoula, Mont., f i r s t as a member o f t h e
National Fire- Danger Rating r e s e a r c h w o r k
u n i t , a n d c u r r e n t l y as a research f o r e s t e r
i n t h e f i r e b e h a v i o r research w o r k u n i t .
T h e BEHAVE system i s a set o f i n t e r a c t i v e computer programs t h a t ( 1 ) p e r m i t
c o n s t r u c t i o n o f site- specific f i r e b e h a v i o r
f u e l models, a n d ( 2 ) contain state- of- thea r t wildland f i r e behavior prediction procedures t h a t w i l l be p e r i o d i c a l l y u p d a t e d .
T h i s manual documents t h e f u e l modeling
p o r t i o n o f BEHAVE. New a n d simplified
p r o c e d u r e s f o r collecting f u e l s data a r e
described.
ln s t r u c t i o n s a r e p r o v i d e d f o r
t h e use o f two programs: ( 1 ) NEWMDL,
w h i c h i s used t o c o n s t r u c t a " f i r s t d r a f t ' '
f u e l model f r o m r a w f i e l d data, a n d ( 2 )
TSTMDL, w h i c h i s u s e d t o t e s t new f u e l
models a n d a d j u s t them u n t i l t h e y p r o d u c e
reasonable f i r e b e h a v i o r p r e d i c t i o n s . A n
e x t e n s i v e section describes concepts a n d
technical aspects o f f u e l modeling.
RICHARD C. ROTHERMEL received h i s
bachelor o f science d e g r e e i n aeronautical
e n g i n e e r i n g a t t h e U n i v e r s i t y o f Washington
i n 1953 a n d h i s master's degree i n mechanical e n g i n e e r i n g a t Colorado State U n i v e r s i t y , F o r t Collins, i n 1971. He s e r v e d i n
t h e U.S. A i r Force as a special weapons
a i r c r a f t development o f f i c e r f r o m 1953 t o
1955. Upon h i s d i s c h a r g e he was employed
a t Douglas A i r c r a f t Co. as a d e s i g n e r a n d
t r o u b l e shooter i n t h e armament g r o u p .
From 1957 t o 1961 Rothermel was employed
b y t h e General E l e c t r i c Co. i n t h e a i r c r a f t
n u c l e a r p r o p u l s i o n department a t t h e
National Reactor T e s t i n g Station i n Idaho.
I n 1961 h e joined t h e N o r t h e r n Forest F i r e
L a b o r a t o r y where he has been engaged i n
r e s e a r c h on t h e mechanisms o f f i r e spread.
He was p r o j e c t leader o f t h e f i r e f u n d a mentals research w o r k u n i t f r o m 1966 u n t i l
1979 a n d i s c u r r e n t l y p r o j e c t leader o f t h e
f i r e behavior research w o r k u n i t a t t h e f i r e
laboratory.
CONTENTS
Page
Introduction
.........................
Fuel Model File..
T h e Common L i n k
f o r t h e BEHAVE System
.............
Program MEWMDL ...................
General Concept
.................
S t r u c t u r e a n d Operation ...........
Field Procedures .....................
...
Specific Field Procedures .............
Reconnaissance ...................
Data Forms .......................
I n d i v i d u a l Data Forms .............
Grass Fuel Data E n t r y Form ...
S h r u b Fuel Data E n t r y Form ...
L i t t e r Fuel Data E n t r y Form ...
Slash Fuel Data E n t r y Form ...
General Data Collection Concept
Multiple 1- Hour
Data E n t r y F o r m
.............
Common Data Items .............
Specific Data ...................
Grass Component ...........
S h r u b Component ...........
L i t t e r Component ...........
Slash Component ...........
M u l t i p l e 1- Hour Fuels .......
Program T S T M D L ...................
General Concept .................
Program S t r u c t u r e ...............
Program O p e r a t i o n ...............
Fuel Modeling Concepts .............
l n t r o d u c t i o n .......................
T h e F i r e S p r e a d Model ...........
D e f i n i t i o n o f T e r m s in t h e
Spread Equation
..................
.........
Reaction ln t e n s i t y ( I
Propagating F l u x Ratio ( 5 ) .....
Wind C o e f f i c i e n t (Owl ..........
Slope C o e f f i c i e n t ( OS) ..........
B u l k D e n s i t y ( p b ) ..............
E f f e c t i v e H e a t i n g Number ( 1...
Heat o f P r e i g n i t i o n (Q. 1 .......
'FI
Weighting o f Fuel Size Classes .....
E
Response of Fuel Models
t o Fuel M o i s t u r e
.................
General T e c h n i q u e s f o r A d j u s t i n g
Fuel Models
Changing
Changing
Changing
......................
Fuel Load ............
Fuel B e d D e p t h ......
S / V Ratios ...........
C h a n g i n g Dead Fuel M o i s t u r e
o f Extinction
Changing
..................
Heat C o n t e n t .........
R e c o r d i n g a n d U s i n g Site- Specific
Fuel Models w i t h t h e
TI- 59 Calculator
.....................
M o d i f y i n g t h e K e y b o a r d O v e r l a y ...
Recording a Fuel Model ............
U s i n g a Fuel Model ................
References ...........................
A p p e n d i x A : Grass a n d S h r u b
Fuel T y p e s
..........................
A p p e n d i x B: Example NEWMDL
Session
.............................
A p p e n d i x C : Example T S T M D L
Session
.............................
A p p e n d i x D : Fuel Model
File S t r u c t u r e
......................
A p p e n d i x E: Weighting Procedures
Used in Program NEWMDL
...........
United States
Department of
Agriculture
Forest Service
u
BEHAVE: Fire Behavior
Prediction and Fuel Modeling
System-
Intermountain
Forest and Range
Experiment Station
Ogden, UT 84401
FUEL Subsystem
General Technical
Report INT-167
Robert E. Burgan
Richard C. Rothermel
May 1984
INTRODUCTION
The site-specific fuel modeling programs described in this manual
a r e p a r t of the BEHAVE System--a series of interactive fire behavior
computer programs for estimating wildland fire potential under various
fuels, weather, and topographicsituations. t he field procedures and
t h e two interactive computer programs described here--NEWMDL and
TSTMDL--provide fire managers t h e capability to construct sitespecific fuel models and to t e s t their fire behavior characteristics
under a variety of simulated environmental conditions. The BURN
subsystem of BEHAVE described by Andrews ( n . d . ) i s designed to
use t h e fuel models developed in FUEL along with state-of-the-art fire
prediction techniques for predicting fire behavior for operations,
planning, or training. The general s t r u c t u r e of the BEHAVE system
and the relation of these programs to each other a r e illustrated in
figure 1.
BEHAVE SYSTEM
BURN
FUEL
-
FUEL MODELING
SUBSYSTEM
(COYYUNICATION
LINK)
FIRE PREDICTION
SUBSYSTEM
----DEVELOPMENT
STATE- OF- THE- ART
F l l E PREDICTION
STORE
FUEL YODELS
TSTMDL
-----
TECHNIQUES
INCLUDINO USE OF
SITE - SPECIFIC
FUEL YODELS
TEST INITIAL
Figure 1 . --General s t r u c t u r e of the B E H A V E system. The B E H A V E
system utilizes a " fuel model file" to q i v e t h e f i r e prediction subsystem access to site-specific fuel models constructed in the fuel
modeling subsystem.
Until now, the library of fire behavior fuel models available to
match fuels situations encountered in the field has been limited to the
13 stylized fuel models developed at t h e Northern Forest Fire Laboratory (Andel-:;on 1982) or specialized models developed for certain
p a r t s of the country such a s t h e southern California b r u s h models
(Rothermel and Philpot 1973; Cohen, review d r a f t ) or the southern
rough models (Hough and Albini 1978). These fuel models have
served well in a variety of applications, b u t methods a r e needed to
accommodate a wide a r r a y of fire management activities.
Careful consideration should be given to the methods of obtaining
a fuel model. The matters of cost, time, and values at r i s k should
b e considered. The following guidelines a r e suggested to aid in t h e
choice:
Use t h e s t a n d a r d 13 models without modification :
a . To illustrate fire behavior of different fuels in general without
reference to any particular site.
b . For estimating f i r e behavior when t h e r e a r e no other fuel
models for t h e area and no time to develop them.
c. When some of t h e s t a n d a r d models have been found to work
well for fuels in an area.
d . For instruction and training about fuels o r f i r e behavior.
Use one of t h e s t a n d a r d 13 models with modifications:
a . When experience indicates b e t t e r representation of fire
behavior r e q u i r e s a change, such a s
...
changing a g r a s s model from static to dynamic,
adding live fuel to a model such a s slash,
adjusting load a n d / o r depth to b e t t e r r e p r e s e n t local fuels,
i . e . , 3-ft b r u s h at 10 tons p e r acre ( T / A ) r a t h e r than 6-ft
at 25 ( T I A ) ,
increasing the heat content of v e r y flammable b r u s h .
Use inventory techniques a s developed b y Brown (1974) and Brown
and o t h e r s (1982):
a . For fuel appraisal, o r whenever i t is important to compare t h e
relative differences in flammability between fuels complexes.
b . For developing fuel models where fuels a r e relatively uniform
and values at risk warrant highly accurate fuel models for fire
prediction.
Use the new procedures in NEWMDL:
a. When an estimate of fire behavior is needed b u t t h e time and
expense of inventory is not cost effective.
b . For developing a fuel model to produce fire behavior predictions t h a t a r e consistent with observed behavior in fuels difficult to
model b y other means.
c . For constructing f i r e behavior fuel models to mimic t h e
behavior of t h e National Fire-Danger Rating System (NFDRS) models
used in an area.
If one of t h e standard 13 models i s to be u s e d , i t may b e called
directly in both BURN and TSTMDL.
If one of t h e s t a n d a r d 13 models i s to b e modified, follow t h e
TSTMDL instructions.
If t h e new fuel modeling procedures a r e to b e u s e d , follow t h e
NEWMDL instructions.
If fuel load inventory data is to b e u s e d , i t i s entered in NEWhlDL
when you a r e asked for loading by size class.
Successful fuel modeling r e q u i r e s a working knowledge of both t h e
mathematical fire spread model (Rothermel 1972) and t h e fire behavior
characteristics of any given vegetation t y p e , u n d e r a variety of
environmental conditions. Therefore, fuels and fire behavior
specialists a r e the intended operational u s e r s of t h e BEHAVE system.
Nevertheless, the BEHAVE system may also s e r v e a s an effective
educational tool for those interested in learning more about how fuels
and environmental parameters influence fire behavior prediction.
The new procedures introduced in NEWMDL use a few key observations about one o r more of four major fuel components: g r a s s ,
l i t t e r , s h r u b s , o r slash. NEWMDL prompts the u s e r for values of
t h e fuel descriptors in a sequence t h a t gradually assembles the
fuel model. Once assembled, the model can be tested in a variety of
ways, including comparisons with any of the original 13 fire behavior
models.
The philosophy used in developing the new fuel modeling subsystem has been to assemble a fuel model with minimal field sampling.
To accomplish t h i s , the programs have the flexibility to allow e n t r y of
information from :
*
*
*
previously inventoried fuels data
relationships compiled from past research
new data obtained using field procedures described in
this manual.
The new field procedures are simplified through the use of a photo
series to help determine general vegetation type and density; ocular
assessments of the percentage of area covered by g r a s s , l i t t e r ,
s h r u b s , or slash; and simple measurements of their approximate
d e p t h s , o r if available from inventory data, loads. Then loadldepth
relationships defined in NEWMDL a r e used to determine depths from
loads or loads from depths. Load assessment will be most accurate if
measured. Depth is more difficult to estimate (Brown 1982). For
instructional purposes, where the model will not b e keyed to a site,
this consideration i s not important. Sample loadldepth relationships
a r e illustrated in figure 2 . The NEWMDL program contains a more
complete representation of the data in this figure. The relationships
in figure 2 show the distinction between fuel t y p e s , b u t t h e r e i s , of
course, considerable variation in the loadldepth relationship for any
one fuel type. Consequently, the f i r s t approximation may not produce reasonable fire behavior and the values may require adjustment.
LOAD (TONS1 ACRE)
F i g u r e 2.- - An example of loadldepth relationships established for
general fuel types and used i n the NEWMDL program.
The interactive computer programs contribute to fuel modeling in
several ways: breaking total load into loads by size (timelag) class,
estimating heat content, surface-area-to-volume ratios, moisture of
extinction, and testing and adjusting a fuel model until it provides
fire behavior estimates that closely match known fire behavior for the
fuel complex it represents.
ith her d y namic-or static fuel models can be constructed. Dynamic
models transfer fuel between the live herbaceous and the 1-hour
timelag categories as appropriate for seasonal changes in t h e moisture
content of herbaceous fuels. This process uses the herbaceous fuel
load transfer algorithm developed for t h e 1978 National Fire-Danger
Rating System (Burgan 1979). Static fuel models have fixed loads in
all fuel categories. The 13 fire behavior fuel models are an example
of static models that were designed for use during t h e more critical
portion of a fire season. The fuel loads in all live and dead classes
remain constant regardless of fuel moisture in this type of fuel model.
Both NEWMDL and TSTMDL meet t h e constraints imposed b y
80-column-by-24-row video display terminals and 80-column printing
terminals. Although graphics a r e employed, specialized graphics
terminals are not required. This generality was achieved a t the
expense of graphics resolution.
To increase " u s e r friendliness, " the fuel modeling programs are
tutorial and have both t'wordy" and "terse" response modes. The
"wordy" mode provides full prompting, which i s helpful for first time
o r occasional u s e r s , while the "terse" mode produces minimal prompting desired by experienced u s e r s . In addition, program control i s
through keywords that a r e descriptive of the task to be performed.
The details of these features a r e provided in t h e sections on operating NEWMDL and TSTMDL.
o n c e an acceptable fuel model has been developed, i t can either be
used with the BURN subsystem of BEHAVE, or be recorded on a magnetic card and used with t h e fire behavior program developed for the
TI-59 calculator (Burgan 1979). Instructions for using t h e TI-59 to
predict fire behavior are given by Rothermel (1983). Instructions for
testing and verifying fire behavior predictions with any fuel model
a r e given by Rothermel and Rinehart (1983).
FUEL MODEL FILE--THE COMMON LINK FOR THE BEHAVE SYSTEM
Fuel model files provide a communications link between t h e
NEWMDL, TSTMDL, and BURN programs of the BEHAVE system
(fig. 1 ) .
Both NEWMDL and TSTMDL enable you to build and save fuel
models in a disk file for easy access. You may manage the contents
of t h e file by listing, adding, replacing, o r deleting fuel models.
The first record in each fuel model file is a "header" containing
(1) a password and ( 2 ) a short description of t h e file.
The password i s user-defined and must be matched before fuel
models are added to, deleted from, or replaced in a file. This protects users from unauthorized or accidental alteration of their file.
Nevertheless, there is no restriction on creating new fuel models for
your own file, or listing the names and numbers of models currently
in any file.
The file description provides very general information about the
models in the file. They might b e described as being for a particular
Forest, Ranger District, o r project.
Use of keyword "FILE" may be made from any of the t h r e e
programs. TSTMDL will allow you to:
1.
2.
3.
4.
5.
6.
Get a previously built site-specific fuel model.
List t h e names and numbers of fuel models in t h e file.
Change a fuel file header.
Add the fuel model just built to the fuel model file.
Replace a fuel model in the file.
Delete a model from the fuel model file.
NEWMDL can perform all of these functions except get a previously
built fuel model.
The BURN program i s intended to be used with previously cons t r u c t e d fuel models, in an operational mode. It will access models in
the file, b u t cannot alter the file.
The s t r u c t u r e of the fuel model file i s described in appendix D .
PROGRAM NEWMDL
General Concept
Construction of a new site-specific fuel model should begin by
using program NEWMDL. NEWMDL defines initial values for fuel model
parameters under user control. NEWMDL i s especially helpful if
extensive fuel inventory information is not available and permits construction of a "compositett fuel model containing any combination of
litter, g r a s s , s h r u b , o r slash.
Although most fuel models can b e constructed with t h e standard
t h r e e dead and two live fuel classes, special cases may arise where it
i s necessary to e n t e r data for two different sizes of 1-h fuels. An
example i s ponderosa pine (Pinus ponderosa) s l a s h , which may have
fine needles, b u t r a t h e r coarse twigs.
When such a model i s being built, the program assumes measured
data i s available for direct input. Upon completion of data e n t r y ,
NEWMDL will "condense" the four-dead, two-live class model to a
standard three- dead, two-live class model for use in the BEHAVE
system or t h e TI-59. The "condensed" model should produce fire
behavior very similar to a four-dead fuel class model.
Litter, g r a s s , and s h r u b fuel information can b e entered as
follows :
1.
woody
2.
3.
Direct input of dead fuel loads by timelag class, live loads as
o r herbaceous, and fuel depth for each vegetation type.
Total load by vegetation type--depth calculated
Total depth by vegetation type--load calculated.
Option 1 i s used when fuel inventory data a r e available for both load
by size class and depth by fuel component--grass, l i t t e r , or s h r u b .
The program then calculates a mean depth for t h e composite fuel
complex in addition to suggesting reasonable values for heat content,
surface-to-volume ratios, and moisture of extinction. Options 2 or 3
a r e used when only loads or only depths a r e known. In fuels with
poorly defined d e p t h s , such as forest l i t t e r , option 3 should be used
cautiously and the calculated loads checked for reasonableness.
Slash fuels may also b e entered directly by load within each timelag class and depth (option 1 ) if complete inventory data a r e available. Otherwise relationships developed for intermountain conifers
(Brown 1978; Albini and Brown 1978) a r e used to estimate the slash
fuels. These relationships permit e n t r y of:
1.
2.
3.
4.
Total slash load.
Total 10-hour timelag load only.
Ten-hour timelag load by species.
Number of 10-hour intercepts p e r foot, by species.
The program then assists the u s e r in partitioning the total load into
size classes and in reducing slash depth and twig and foliage retention, as a function of harvest method and slash age.
One hundred percent ground coverage i s assumed for total l i t t e r ,
g r a s s , s h r u b , or slash loads initially entered into the program. Such
coverage by a single fuel component is possible, but not necessarily
the case. When less than 100 percent ground coverage is specified
for any fuel component, the load and the depth of t h a t component will
be reduced accordingly. Both load and depth must b e reduced so the
bulk density (amount of fuel [pounds] p e r cubic foot) of fuel bed will
remain the same. In addition, the same ground area may be covered
by more than one component (example: g r a s s , l i t t e r , and s l a s h ) .
Subsequent program operations sum t h e loads for each component,
and partition them among the size classes.
The final output of NEWMDL i s a display of the completed fuel
model (fig 3 ) . The model should b e exercised in the TSTMDL program to examine i t s fire behavior characteristics and to possibly
adjust some parameters.
A detailed explanation of the weighting procedures used to produce' the completed fuel model from the users 1 input is provided in
appendix E.
CUliRENT VALUES OF FUEL MODEL PARAMIZTERS
DYNAMIC 1 4 . SAMPLE MODEL
BY: BURGAN
LOAD ( T / A O )
1 HR
1 0 HR
1 0 0 HR
I...I V E HE::K E{
V
J
Y
S/V
RATIOS
-
-----------------
4.07
1 00
0.09
I H R
1800.
L I V E HERE{
1'300.
LIVE: WOODY
1700.
9 / V :z (S(IFT/CUFT)
.
0 . (33
0Tt.IlZR
DEPTH ( F E E T )
I-IE:AT CON'TI:INT (E{'I'U/LB)
EX'TMOISI'\JRE ! X )
0.94
8000.
17.
1.13
Figure 3.--NEWMDL o u t p u t . The final output of the NEWMDL
program is a display of the completed fuel model. At this point
the model can be saved in a fuel model file.
S t r u c t u r e and
Operation
The specific procedure for accessing your computer and the
NEWMDL program must be obtained from your computer specialist.
Once s t a r t e d , you will find the interactive, tutorial nature of NEWMDL
eliminates the need for a detailed explanation of program operation.
Nevertheless, a general overview of program s t r u c t u r e and operation
is h e l ~ f u l .
You will first be asked to enter your name (maximum of 20 l e t t e r s )
and indicate whether you want to use the "TERSE" mode (minimal
prompting for experienced u s e r s ) or the "WORDY" mode (full prompting for new u s e r s ) . After accepting or declining a list of keywords
used for program control, you will b e asked whether you want to
build a model with one or two sizes of fine (1-h) fuel. Normally one
size of fine fuel should be selected. A number and name must then
be entered for the proposed fuel model. Acceptable numbers are 14
through 99. Numbers 1 through 13 a r e reserved for the 13 fire
behavior fuel models (Anderson 1982)
Program control is throu g h the use. of keywords. This provides a
great deal of operational flexibility. Any keyword above the dashed
fine in the following tabulation can be entered whenever the message
"CONTROL SECTION. KEYWORD?" is printed. There is no specific
order in which litter, g r a s s , s h r u b , or slash fuel loads must b e
determined. In addition, you can ask for a keyword list, set terse
or wordy mode, display current values of the four fuel components,
r e s t a r t the program, access the fuel model file, or quit the session,
whenever you a r e prompted for a keyword. But notice the restrictions associated with the keywords below the dashed lines.
.
Keyword
KEY
TERSE
WORDY
LITTER
GRASS
SHRUB
SLASH
COMP
FILE
RENUMBER
QUIT
RESTART
Function
Prints this keyword list
Set t e r s e mode for minimal prompting
Set wordy mode for full prompting
Determine load and depth of litter fuels
Determine load and depth of g r a s s fuels
Determine load and depth of s h r u b fuels
Determine load and depth of slash fuels
Display values currently assigned to
each of the above four fuel components
Access fuel model file
Renumber the fuel model
Quit session
S t a r t program at beginning again
SURF
Determine surface-to-volume ratios
( a t least one of the keywords LITTER,
GRASS, SHRUB, or SLASH must be used
first to assign some fuel loads)
HEAT
Determine heat content (keyword SURF
must be used before this keyword)
MODEL
Display tabulation of completed fuel
model (keywords SURF and HEAT must be
used before this keyword)
Figure 4 reemphasizes the limitations associated with the keywords
below the dashed line and also illustrates the general flow of t h e
program. Loads and depth must be defined for a t least one of t h e
four fuel components before surface-to-volume (SIV) ratios can be
assigned. The S/V ratios must be assigned before heat contents of
t h e fuel components a r e entered, because SIV ratios a r e used to
calculate a single, weighted heat content for the completed fuel model.
Surface-to-volume ratios and heat contents must be reentered i f a
keyword for a fuel model component--LITTER, GRASS, SHRUBS, o r
SLASH--is u s e d , because you may have modified one o r more fuel
components.
The program will not accept the keyword "MODEL" until all usercontrolled fuel model parameters have been defined, or adjusted if the
fuel model h a s been changed. The fuel model should b e added to t h e
file only after you judge that reasonable values have been assigned to
all the fuel model parameters under your control. This i s best done
by looking at t h e listing obtained from keyword "MODEL". Use of
keyword "FILE" will provide an opportunity to save t h e fuel model on
disk. After saving a fuel model, you may either "QUIT" to exit from
NEWMDL, or "RESTART" to begin constructing another fuel model.
The procedure for accessing any fuel model to test and adjust i t s fire
behavior characteristics i s given in the section for o p eratin g program
TSTMDL.
You should not b e able to "crash" the NEWMDL program, so feel
free to experiment with i t . Appropriate messages a r e presented and
correct actions suggested whenever improper procedures a r e
attem~ted.
We strongly recommend that you become familiar with t h e operation
and capabilities of NEWMDL before collecting any fuels data in t h e
field. -while learning how to construct fuel- models, the accuracy of
your answers to the questions posed by NEWMDL i s much less
important than gaining insight into t h e relationships between the
program and the field procedures.
PROGRAM NEWMDL
----SET MODE
- -
-
-
DETERMINE LOADS & DEPTH FOR APPLICABLE
FUEL COMPONENTS
SURF
---SIV RATIOS MUST BE ASSIGNED
(OR REASSIGNED IF ANY COMPONENT LOAD HAS BEEN
CHANGED) BEFORE HEAT CONTENT CAN BE ENTERED
I---- I
HEAT
HEAT CONTENT MUST BE ASSIGNED
(OR REASSIGNED IF ANY COMPONENT LOAD HAS BEEN
CHANGED) BEFORE THE COMPLETED FUEL MODEL
CAN BE LISTED
COMPLETED FUEL MODEL MAY BE LISTED
BY USING KEYWORD " MODEL "
SAVE FllEL MODEL
I N A FILE
.
Figure 4 . --General flow of program NEWMDL
The general procedure in using the NEWMDL
program is to establish fuel load, assign
surface-area-to-volume ratios and heat
con tents, lis t the model for reference, and
save it in a fuel file.
General Data
Collection
Concept
FIELD PROCEDURES
When building a fuel model t h e task is more one of describing
vegetation as a fuel complex rather than precisely measuring i t s
biomass, although t h e two a r e related.
When considering how a particular vegetation type might b u r n ,
remember the following limitations of the fire behavior model that will
use the fuels data.
1. The fire i s assumed to b e a line fire burning steadily in
surface fuels.
2. The fire model i s intended to predict fire behavior produced
by fine fuels at the perimeter of the fire, usually t h e fire front.
3 . The fire model works best in uniform, continuous fuels such
as g r a s s , long-needle pine litter, uniform brushfields, and continuous
logging slash.
These limitations have important implications regarding how to view
vegetation a s a forest or range fuel. For example, because a surface
fire i s assumed, it i s wrong to include vegetation that i s in a separ a t e and distinctly higher canopy level than that near the ground.
Consider only vegetation that can influence fires before erratic
behavior such as crowning or spotting begins.
The fire model predicts behavior on the fire perimeter, normally
at the fire front. Inventory only t h e fine fuel that propagates the
fire, that i s , dead fuels less than 3 inches in diameter and live fuels
of less than 114-inch diameter. This is often much less than t h e
total fuel load per acre. Ignore fuels that b u r n long after t h e fire
front has passed. These include deep duff, stumps, large logs, and
SO on.
The assumption of uniform and continuous fuel means t h a t t h e fire
model will calculate fire behavior as though the fuel components in the
model Gere mixed and distributed uniformly throughout the specified
depth.
These are reasonable assumptions when nearly all t h e fuel is represented by just one component, such as a field of g r a s s or a relatively continuous litter layer. The assumptions still hold even when
t h e fuel complex is composed of more than one component--grass,
litter, s h r u b , o r slash--if t h e components are fairly well mixed.
When t h e data for a mixed fuel complex are entered in NEWMDL i t will
produce a representative fuel model for the mixture.
But if t h e fuel components occur in separated patches, and the
fire will b u r n from one to another and back again, consider building
separate fuel models. Then t h e two-fuel-model concept available in
BURN can be used to predict r a t e of spread for this situation.
The fact that the assumptions and limitations do not always match
reality accounts in part for differences between predicted and
observed fire behavior. Nevertheless, a properly developed and
tested fuel model can be used with the fire model to produce s u r prisingly accurate fire potential estimates.
Perhaps the greatest difficulty in constructing a site-specific fuel
model is clearly defining t h e fuel complex it represents. The infinite
variability produced by changes in fuel composition, quantity, depth,
continuity, and so on, make it imperative that even site-specific fuel
models must represent a r a t h e r broad range of conditions. T h u s ,
although the first step in constructing a site-specific fuel model may
be to obtain field data, a t least the following points should b e carefully considered in the planning phase:
1. To what general vegetation type will the model apply?
Fire should be a recurring problem in this vegetation type, and the
vegetation must b e readily identifiable and sufficiently abundant to
justify the need for a separate fuel model.
2. Should t h e model b e dynamic or static?
Dynamic models are needed only if the model i s to be used throughout
the growing and curing season.
3. Should the two-fuel-model concept be considered?
4. What a r e the intended uses of the model?
This can dictate how accurate the data must be.
What is the range of fuel conditions to which t h e fuel model
5.
will apply? Can it be used in similar fuels in other areas? How will
it be described so others will know i t s intended application?
These and other questions arising in your fire management operations will be difficult to answer, b u t considering such questions in
advance is helpful both in the initial collection of field data and in
later attempts to apply the model to new situations.
NEWMDL is designed to accept fuel data from a variety of sources.
This i s not necessarily simpler than a single process, but it does
allow the user to utilize data on hand or design field collection procedures to match the needs of the intended application.
If you have discarded the idea of choosing one of the standard 13
models or modifying one of them, you must now select one of the
following sources of data:
utilize inventory data already collected
collect new inventory data
use photo series
use new procedures offered here
use knowledge about fuels gained from experience
combination of t h e above.
The inventory procedures by Brown (1974) a r e designed to
measure fuel load and depth b y size class for naturally fallen debris
and logging slash. In a later handbook, Brown and others (1982)
give more complete procedures for inventorying surface fuels in the
interior West. The restriction of their methods to the interior West is
necessitated by relating s h r u b and conifer reproduction measurements
to previously measured characteristics of specific species. Their procedures provide estimates of fuel load by size class for duff, litter,
grasses and h e r b s , s h r u b s , fallen debris, and conifer reproduction.
Both living and dead loads a r e included, b u t depth of s h r u b s and
duff (not used h e r e ) i s the only depth tallied.
An ever-expanding photo series i s being developed for describing
and classifying fuels. Each photographic scene of a fuel complex
includes a description of the fuel, a fire potential r a t i n g , and data
about fuel load b y size class.
Fuel inventory procedures and photo series provide data primarily
about fuel load. In some cases depth i s included, b u t not always.
Brown and others (1982) discuss the difficulty of measuring depth.
To construct a fuel model, however, a depth must be p r o v i d e d a ~ o n ~
with load. The bulk density determined by these two factors i s a
primary variable needed to drive the fire model (Rothermel 1972).
The new procedures presented here overcome this problem by allowing
the u s e r to determine a depth t h a t can be used with inventoried
loads. The new procedures may also be used to infer fuel loads from
estimated fuel depths if inventory data a r e not available or if the
assessment does not warrant the time for inventory.
Figure 2 illustrates the heart of the new procedures, which rely
upon the fact t h a t if the bulk density of a fuel component can be
estimated, then i t s load can be calculated using a measurement of the
depth, or the depth can be determined from a load measurement.
(Bulk density is the fuel load [ I b / f t 2 ] divided b y the depth [ f e e t ] . )
Note t h a t in figure 2 , the bulk densities a r e t h e inverse of the slopes
of the lines. There i s , of course, scatter about these lines for different fuels. The specific field procedures in the next section allow
you to choose a bulk density most appropriate for g r a s s or s h r u b
data. Figure 2 illustrates relationships used within the program in
greater detail, and is used to define the loadldepth relationships
needed.
L
Data forms described in t h e n e x t section have been designed to
r e c o r d t h e data needed to develop a fuel model. They will accommodate d a t a obtained b y any of t h e methods described above. As you
work with t h e forms and p r o c e d u r e s , you will find t h a t only p a r t of
the data can be obtained from t h e field; o t h e r data r e g a r d i n g particle
size a n d heat content must be provided a f t e r prompting by t h e computer.
Some fuel factors essential t o t h e fire model a r e held constant
because they e i t h e r have a small effect over t h e i r naturally occurring
r a n g e or would be v e r y difficult for the u s e r to determine.
These a r e :
Fuel factor
Particle density
Total mineral content*
Effective mineral content*
10-h surface-to-volume ratio
100-h surface-to-volume ratio
Assumed value
32 l b / f t Z
0.0555
0.010
109
30
* F r a c t i o n of d r y w e i g h t .
SPECIFIC FIELD PROCEDURES
Reconnaissance
The f i r s t s t e p i s to conduct a field reconnaissance t o obtain a
general impression of t h e fuels to b e modeled. A fire t h a t covers a
significant a r e a will often b e influenced b y considerable fuel variability. T r y to develop an impression of t h e "typical" situation b y
looking a t t h e vegetation in broad terms. During y o u r reconnaissance, consider t h e following questions about t h e fuel:
1 . Which fuel components--litter, g r a s s , s h r u b s , and slash--are
p r e s e n t in significant q u a n t i t y ?
2 . How continuous a r e t h e various fuel components?
3. What fuel stratum i s most likely to c a r r y fire?
4 . Are t h e r e l a r g e variations i n t h e amount of one o r more fuel
components?
5. What proportion of t h e fuel i s in t h e 1-h, 10-h, 100-h, live
herbaceous, a n d live woody categories?
6. How many g r a s s a n d s h r u b t y p e s must b e dealt with?
7 . Which bulk density photos b e s t r e p r e s e n t t h e bulk
densities of t h e important g r a s s e s a n d s h r u b s in t h e a r e a ?
8 . What i s a representative d e p t h of t h e g r a s s e s , s h r u b s , l i t t e r ,
o r slash in t h e area?
9 . A r e t h e fuels sufficiently intermixed t h a t they can b e r e p r e s e n t e d b y a single model, o r do t h e y occur in independent "patches"
t h a t may r e q u i r e use of t h e two-fuel-model concept?
Field measurements a r e time consuming a n d expensive; therefore
t h e new procedures described h e r e have been made a s simple a s
possible. T h e equipment needed i s limited to data forms, a tape
measure, a g r a s s clipper, a n d a photo s e r i e s , if applicable.
Data Forms
A s e p a r a t e data form i s provided for each fuel component- - grass,
s h r u b s , l i t t e r , a n d s l a s h . T h e s e four forms a r e for e n t e r i n g d a t a on
a single size of 1-h fuels- - that i s , t h e familiar three- dead- class,
two-live-class fuel model. Each form i s divided into two sections: one
for summarizing existing inventory data t h a t include both fuel load
and d e p t h ("previously inventoried fuel d a t a " ) , a n d t h e other for
recording new observations o r inventory data t h a t do not contain both
load a n d d e p t h ("new fuel d a t a " ) . If you have complete information
for either portion, you will be able t o answer all t h e questions
NEWMDL will a s k . Depth may not b e available from y o u r existing
fuels data. I n t h a t case you can u s e t h e new fuel d a t a portion of t h e
form b y supplying t h e additional r e q u i r e d information. Note t h a t this
gives you t h e option of e n t e r i n g e i t h e r load o r d e p t h . E n t e r load
a n d l e t NEWMDL calculate a d e p t h for you, b u t b e s u r e to check i t
f o r reasonableness. You will also h a v e t o e n t e r p e r c e n t a g e s of loads
i n t h e various size classes r a t h e r t h a n t h e actual values.
A fifth form i s f o r e n t e r i n g data on two sizes of 1-h fuels. Such
d a t a have t o come from e i t h e r detailed field measurements o r from
supplemental computed programs t h a t analyze o r p r e d i c t d e b r i s .
Individual Data Forms
GRASS FUEL DATA ENTRY FORM
I.
Previously Inventoried Fuel Data
A.
Model type (1
1.
2.
11.
-
2)
Dynamic
Static
B.
Total g r a s s load (0-30 t o n s l a c r e )
C.
Depth (0-10 f t )
D.
For dynamic models e n t e r maximum
percentage t h a t can be l i v e (0-100%)
E.
For. s t a t i c models e n t e r c u r r e n t
percentage l i v e (0-100%)
F.
Percentage of a r e a covered by g r a s s
(0-100%)
New Fuel Data
A.
Model type (1
1.
2.
B.
2.
3.
4.
2)
-
4)
Dynamic
Static
Grass type (1
1.
-
Fine--e.g.,
cheatgrass
Medium--e.g., rough fescue
Coarse--e.g.,
fountaingrass
Very coarse--e.g.,
sawgrass
C.
Bulk density c l a s s (1 - 6)
( r e f e r t o photos i n u s e r ' s manual)
D.
T o t a l g r a s s load (0-30 t o n s l a c r e )
E.
Grass depth (0-10 f t )
F.
For dynamic models e n t e r maximum
percentage t h a t can be l i v e (0-100%)
G.
For s t a t i c models e n t e r c u r r e n t
percentage l i v e (0-100%)
H.
Percentage of a r e a covered by g r a s s
(0-100%)
'I
SHRUB FUEL DATA ENTRY FORM
I. Previously Inventoried Fuel Data
A. Loads (tonslacre)
1. 1-HR (0-30)
2. 10-HR (0-30)
3. 100-HR (0-30)
4. Leaves and live twigs (0-30)
11.
B.
Depth (0-10 it)
C.
Percentage of area covered by shrubs
(0-100%)
D.
Oils and waxes (circle one)
Yes
No
New Fuel Data
A.
Shrub type (1-5)
1.
2.
3.
4.
5.
Fine stems, thin leaves--e.g., huckleberry
Medium stems, thin leaves--e.g., ninebark
Medium stems, thick leaves--e.g., ceanothus
Densely packed fine stems and leaves-e.g., chamise
Thick stems and leaves--e.g., manzanita
B.
Bulk density class (1-6)
(refer to photos in user's manual)
C.
Total shrub load (0-80 tonslacre)
D.
Shrub depth (0-10 it)
E.
Percentage of total shrub load in each size
class. Enter as whole percentile (must
total 100%)
1.
1-HR (0-114 inch)
2.
10-HR (114-1 inch)
3.
100-HR (1-3 inches)
4.
Live leaves and twigs (0-114 inch)
F.
Percentage of area covered by shrubs
(0-100%)
G.
Oils and waxes (circle one)
Yes
No
LITTER FUEL DATA ENTRY FORM
P r e v i o u s l y I n v e n t o r i e d F u e l Data
A.
Loads ( t o n s l a c r e )
1.
1-HR (0-30)
2.
10-HR (0-30)
3.
100-HR (0-30)
B.
Depth (0-5 f t ) ( f t = cm
C.
Area coverage (0-100%)
i
30.48)
New F u e l Data
A.
L i t t e r source ( 1
1.
2.
3.
B.
3)
Conifers
Hardwoods
Both, b u t a t l e a s t 30% of l e s s e r t y p e
Needle l e n g t h i f c o n i f e r s o r b o t h ( 1
1.
2.
C.
-
2)
Mediumllong--e.g.,
lodgepole o r
ponderosa p i n e
Short--e.g.,
Douglas- fir
L i t t e r compactness ( 1
1.
2.
3.
-
-
3)
Loose ( f r e s h l y f a l l e n )
Normal
Compact ( o l d e r compressed l i t t e r )
D.
T o t a l l i t t e r l o a d (0-100 t o n s l a c r e )
E.
L i t t e r d e p t h (0-5 f t ) ( f t = cm
F.
P e r c e n t a g e of t o t a l l i t t e r l o a d i n e a c h s i z e
c l a s s . E n t e r a s whole p e r c e n t i l e (must
t o t a l 100%)
G.
1.
1-HR (0-114 i n c h )
2.
10-HR (114-1 i n c h )
3.
100-HR (1-3 i n c h e s )
30.48)
P e r c e n t a g e of a r e a covered by l i t t e r
(0-100%)
SLASH FUEL DATA ENTRY FORM
I.
11.
P r e v i o u s l y I n v e n t o r i e d Fuel Data
A.
Loads ( t o n s l a c r e )
B.
Depth (0-10 f t )
C.
Area coverage (0-100%)
New Fuel Data
A.
Logging method ( 1
1.
2.
3.
Major
species
-
3)
Commercial t i m b e r c u t , h i g h l e a d s k i d d i n g
Commercial t i m b e r c u t , ground l e a d s k i d d i n g
Precommercial t h i n n i n g
B.
Age (0-5 y r )
C.
T o t a l component l o a d (0.01-100 t o n s l a c r e )
D.
T o t a l 10-h l o a d (0.01-30 t o n s l a c r e )
Crown
class
(1-Dom)
(2- Int)
Ave
d.b.h.
Species
% foliage
retention
% by s p e c i e s
i f C or D
G.
Average percentage twig retention,
H.
Area c o v e r a g e (0-100%)
E
species
10-HR l o a d
per acre
species
F
species
intercepts
p e r foot
MULTIPLE 1-HOUR DATA ENTRY FORM
I. LITTER COMPONENT
Loads (0-30 t o n s p e r a c r e )
S/V r a t i o s (800-3,500 f t 2 / f t 3 )
Depth (0-5 i t )
Area coverage (%)
11. SLASH COMPONENT
Loads (0-30 t o n s p e r a c r e )
S/V r a t i o s (800-3,500 f t 2 / f t 3 )
1-HR
1-HR
10-HR
100-FIR
Depth (0-10 f t )
Area coverage ( % )
111. SHRUB COMPONENT
Loads (0-30 t o n s p e r a c r e )
s/V
r a t i o s (800-3,500 f t 2 / f t 3 )
1-HR
1-HR
10-HR
100-HR
Live woody
Depth (0-10 i t )
Area coverage ( % )
Waxes o r o i l s
I V . GRASS COMPONENT
Loads (0-30 t o n s p e r a c r e )
1-HR
1-HR
10-HR
Live herbaceous
Depth (0-10 f t )
Area covered (%)
Model type
(dynamiclstatic)
S/V r a t i o s (800-3,500 f t 2 / f t 3 )
COMMON
DATA ITEMS
Four items that occur in several places on the data forms will be
defined prior to subsequent use in the detailed explanation of data
entries for each fuel component.
Total component load.--This is the total load for an individual fuel
component ( g r a s s , s h r u b , litter, or s l a s h ) . It can be any combination of 1-, l o - , and 100-h dead fuels, live herbaceous material, and
the leaves and 114-inch o r smaller twigs of live shrubs. This fuel
generally occurs within 6 feet of the F layer surface. Record in tons
per acre.
Individual live and dead loads.--These loads are most commonly
available from existing inventory data. Record in tons p e r acre for
each of the following loads that should be included in the fuel model:
Dead fuels:
1-h ( less than 114-inch diameter)
10-h (114- to 1-inch diameter)
100-h (1- to 3-inch diameter)
Live fuels:
Leaves and live twigs less than 114-inch diameter.
Enter zero for those that a r e inappropriate.
Percent of the loads in individual classes. --When using the New
Fuel Data" portion of the s h r u b and litter forms, estimate as necessary the percentage of the total load in the 1-h, 10-h, 100-h, andlor
live fuel classes. These percentages are used to break the total load
into individual live and dead loads. Record the percentages of live
and dead fuels to the nearest whole percentile. The percentages
must sum to 100 for each component.
Depth.--Record the average depth of the fuel model component in
feet. If the litter component is shallow, i t may be measured in centimeters, then converted to feet. Review the definition of depth in the
section "General Field Observation Concepts" for "Grass and Shrubs"
if there is any question about what depth is. See also figures 5 and
6. Experience has shown that 70 percent of the maximum depth gives
a reasonable estimate of depth for g r a s s , s h r u b s , and slash, while
maximum depth is more appropriate for fallen litter fuels that a r e
lying horizontally.
A V E R A G E D E P T H OF AREA
11111-1
Figure 5.--Concept of grass and s h r u b depths. Average grass
o r s h r u b depth is about 7 0 percent of the maximum leaf o r stalk
height. I t can be visualized as the average h e i g h t of a pliable
sheet draped over the fuel particles.
AVERAGE
SLASH
DEPTH
AVERAGE L I T T E R
DEPTH
F i g u r e 6.- -Concept of slash and l i t t e r d e p t h . L i t t e r d e p t h is t h e
v e r t i c a l distance from the t o p of t h e F layer t o t h e general u p p e r
surface o f t h e L l a y e r . Slash depth is about 7 0 percent of t h e
distance from the t o p of the n a t u r a l l i t t e r layer t o t h e average h i g h
i n t e r c e p t . I t can be visualized as t h e average depth of a pliable
sheet draped over t h e fuel particles.
Percentage of t h e area covered b y each fuel component. --Initial
fuel load estimates a r e based on t h e assumption t h a t 100 percent of
the area i s covered b y the fuel component in question. If your
inventory procedure was to sample the entire a r e a , both where fuel
existed and where i t did not e x i s t , e n t e r 100 for t h e percentage of
area covered. Then your inventoried load will not be reduced. If
you used the inventory procedure presented h e r e for collecting "newt'
fuel data, e n t e r your estimate of t h e percentage of the area actually
covered b y each fuel component. Then fuel loads will b e reduced
from the assumed 100 percent coverage to actual coverage.
Estimating bulk density classes for g r a s s e s and shrubs.- - Appendix
A provides photo s e t s to help visualize bulk densities for different
g r a s s and s h r u b t y p e s , ranging from fine to v e r y coarse.
First select t h e photo s e t t h a t best r e p r e s e n t s t h e morphology of
t h e g r a s s or s h r u b t y p e t h a t will most effectively c a r r y t h e fire.
Then select the photo within t h a t t y p e that b e s t r e p r e s e n t s i t s bulk
density. If t h e g r a s s e s o r s h r u b s occur in clumps, select a photo
t h a t b e s t r e p r e s e n t s t h e bulk density of a typical clump, r a t h e r than
trying to estimate the average bulk density t h a t would exist if all t h e
vegetation in the clumps were spread evenly over the entire area.
Once t h e bulk density for grasses and s h r u b s ( o r both) has been
estimated, then either their average loads or depths must be determined. Grass and s h r u b loads p e r acre can be estimated by clipping
and weighing 3-inch diameter and smaller material from sample plots of
known size, ovendrying i t , weighing i t , and expanding the average
sample plot load t o a per- acre basis. In this process i t must be
assumed t h a t the g r a s s e s and s h r u b s cover 100 percent of the a r e a ,
even if that is not t r u e . NEWMDL reduces these loads for t h e percentage of the area you s t a t e is actually covered b y g r a s s e s o r
shrubs.
Estimating the g r a s s o r s h r u b load from i t s depth i s a much faster
procedure. The depth of any fuel component i s t h e vertical distance
from the bottom of t h e fuel component layer to the appropriate height
at which the bulk density begins to rapidly decrease; o r alternatively
about 70 percent of t h e average maximum leaf o r stalk height. Figure
5 illustrates this definition f o r g r a s s and s h r u b components. Depth
must be estimated with t h e assumption t h a t t h e s h r u b o r g r a s s t y p e
u n d e r consideration covers 100 p e r c e n t of t h e area. From t h a t ,
NEWMDL will f i r s t estimate t h e load p e r acre based on 100 percent
area coverage, then r e d u c e t h a t load f o r actual a r e a coverage.
Estimating load and d e p t h for l i t t e r a n d slash.--Bulk density
photos f o r l i t t e r a r e impractical; therefore t h e bulk densities a r e
based on litter source (hardwoods, conifers, o r b o t h ) , conifer needle
length ( l o n g , medium, o r s h o r t ) , a n d litter compactness (loose,
normal, o r compact). These data a r e used b y NEWMDL, along with a
d e p t h value, to determine l i t t e r load. Litter d e p t h i s d e f i n e d a s t h e
vertical distance from t h e top of t h e F layer t o t h e general u p p e r
s u r f a c e of t h e L l a y e r . Scattered p r o t r u d i n g fuel particles a r e to b e
ignored. Figure 6 illustrates t h e definition of d e p t h for slash and
litter.
By f a r t h e most r e s e a r c h h a s been done on s l a s h , s o t h e
relationships developed in t h e s e s t u d i e s have been u s e d to simplify
field observations for estimating slash loads and d e p t h s . The
r e q u i r e d information includes logging method, slash a g e , a n d one of
several expressions of slash load. If estimating slash load i s diffic u l t , t h e data s h e e t s which accompany photo s e r i e s often provide an
excellent source of information. These data a r e most conveniently
recorded a s "previously inventoried fuel d a t a . " A partial list of
available photo s e r i e s i s included in t h e "References" section.
A note of caution i s advised when using photo s e r i e s . T h e
1-h load given on t h e data s h e e t will probably not account for needles
still retained on s l a s h . This i s because t h e s t a n d a r d fuel inventory
technique u s e d to develop t h e s e data (Brown 1974) does not include
measurements on needle loads. Brown recognizes this a n d h a s
provided multiplying ratios t o calculate needle quantity based on
estimated b r a n c h wood weight. These ratios a r e p r e s e n t e d in his
appendix I11 for several species of western conifers. Modification of
t h e 1-h load p r e s e n t e d in a photo s e r i e s i s appropriate for " r e d
needle" slash.
Alternatively, fuel loads for l i t t e r o r slash may be determined
directly using inventory techniques described b y Brown (1974). His
publication provides excellent documentation and detailed i n s t r u c t i o n s
t h a t need not b e repeated h e r e . NEWMDL does not r e q u i r e an invent o r y a s described b y Brown, b u t u s e of h i s p r o c e d u r e s will provide
all t h e load and d e p t h information r e q u i r e d f o r l i t t e r a n d slash loads.
Again, remember t o account for needle load when inventorying " r e d "
slash. If you have n e v e r measured fuels, some practice will be
helpful in understanding and utilizing t h e methods described h e r e .
SPECIFIC DATA
Grase Component
Specific i n s t r u c t i o n s for completing t h e data forms for individual
fuel components follow.
I.
Previously inventoried fuel data
A.
Model t y p e - Record whether t h e model i s to be
dynamic o r static.
B.
Total g r a s s load - Record total g r a s s load (live a n d
d e a d ) in tons p e r a c r e .
C.
Depth - Record adjusted g r a s s d e p t h in feet.
D.
Maximum percentage t h a t can b e live - For dynamic
fuel models, indicate t h e g r e a t e s t proportion of t h e
total g r a s s load t h a t i s live a t any time d u r i n g t h e
y e a r , r e g a r d l e s s of how green t h e g r a s s may be a t t h e
p r e s e n t time. Accumulation of dead g r a s s from previous seasons will generally keep this number below 50
p e r c e n t . Leave blank if you a r e building a s t a t i c fuel
model.
E.
C u r r e n t percentage live - For static fuel models e n t e r the
proportion of the g r a s s , b y volume, t h a t i s live at the
time of year for which the model i s being designed. I t
can b e estimated by clipping a few pounds of g r a s s , separating all t h e live material into one pile and all dead
material into a number of piles equal in size to t h e pile of
live material. Then the percentage value to e n t e r is:
100
(total number of piles)
Make no e n t r y if you a r e building a dynamic model.
F.
11.
Shrub Component
I.
11.
Area coverage - Record percentage of area covered by
grass.
New fuel data
A.
Model t y p e - Record "dynamic" o r "static" a s explained
u n d e r I-A above.
B.
Grass t y p e - Compare each page of g r a s s t y p e ( 1 , 2 , 3,
and 4 ) photos with your field situation. Record t h e
number of the g r a s s t y p e which is most similar morphologically. The purpose of this s t e p i s to just select a
general g r a s s t y p e category.
C.
Bulk density class - The bulk density is defined by
matching bulk density photos of t h e appropriate g r a s s
type with your field observations. Record t h e density
class number (1- 6).
D'.
Total g r a s s load - Record if available from a "clip and
weigh" inventory, otherwise leave blank.
E.
Grass depth - Record adjusted depth in feet.
70 percent of maximum depth. See figure 5.
F.
Maximum percentage live - See I-E above.
G.
C u r r e n t percentage live - See I-F above.
H.
Area coverage - Record percentage of the area covered
by g r a s s .
That i s ,
Previously inventoried fuel data
A.
1- , l o - , and 100-h dead fuel loads, leaf and live twig
loads - Record t h e load for each of these fuel categories
that should b e included in t h e fuel model. Enter zero
for those that a r e inappropriate.
B.
Depth - Record the adjusted s h r u b depth in feet.
C.
Area coverage - Record percentage of t h e area covered
by shrubs.
D.
Oils and waxes - Some s h r u b s contain oils and waxes
t h a t significantly increase t h e contribution of t h e live
foliage to t h e fire intensity and also increase the moist u r e content at which these fuels will b u r n . Record
whether such material i s o r i s not p r e s e n t in t h e s h r u b s .
New fuel data
A.
S h r u b type - Compare each page of s h r u b type ( 1 , 2 ,
3, 4 , and 5) photos with your field situation. Circle the
number of the s h r u b type which is most similar morphologically.
B.
Bulk density - Select b y matching bulk density photos
of the appropriate s h r u b type with t h e field situation.
Record the bulk density class number (1- 6).
,
/
- Record total s h r u b load in tons per
acre if available from a clip-and-weigh inventory,
otherwise leave blank.
C. Total s h r u b load
D.
Depth - Record s h r u b depth in feet.
E.
Percentage of s h r u b load in each size class - Estimate to
nearest whole percentile.
F.
Area coverage - Record percentage of area covered by
shrubs.
- Review I-D for s h r u b s if necessary;
then record yes or no.
G. Oils and waxes
Litter Component
I.
Previously inventoried fuel data
A.
1-, l o - , and 100-h loads - Record in tons p e r acre for
each of those fuel categories that should b e included in
the fuel model. Enter zero for those that are
inappropriate.
B.
Depth - Record average litter depth in feet. If the
litter is shallow i t may be measured in centimeters,
then converted to feet b y dividing by 30.48.
C. Area coverage - Enter percentage of area covered by
litter.
11.
New fuel data
A.
Litter source - Record whether t h e litter results from
hardwoods, conifers, o r both.
B.
Needle length - Needle length affects the bulk density
of conifer litter, with medium- to long-needle species
such as lodgepole o r ponderosa pine producing a litter
bed having a lower bulk density than short-needle
conifers such a s larch or Douglas-fir. Record as
mediumllong or as s h o r t .
C. Litter compactness - NEWMDL will use different bulk
densities f o r loose, normal, or compact litter. Hardwood
litter particularly i s most likely to b e loose o r fluffy
when it first falls, b u t compact after it has been on t h e
ground for at least one winter.
D.
Total litter load - Record total litter load in tons p e r
acre if available from an inventory. Skip this entry if it
is unknown.
E.
Depth - Record litter depth in feet. If the litter i s
shallow it may b e measured in centimeters, then converted to feet by dividing by 30.48.
F.
Percentage of litter load in each size class - Estimate to
nearest whole percentile.
G. Area coverage
-
Record percentage of area covered by
litter fuels.
Slaeh Component
I.
Previously inventoried fuel data. Data obtained by comparing photo series with the field situation should be
entered here.
A.
1-, l o - , and 100-h loads - Record in tons p e r acre for
each of those fuel categories to b e included in t h e fuel
model. Enter zero for those that a r e not appropriate.
B.
Depth - Record slash depth in feet.
C. Area coverage
by slash.
-
Record percentage of the area covered
.
11.
New fuel data
A.
Logging method - Record a s 1, 2 , o r 3 to define the
slash origin a s follows:
1.
Commercial timber cut, high lead skidding
2.
Commercial timber c u t , ground lead skidding
3.
Precommercial thinning
B. Age - Record slash age a s number of winters it has
existed.
Slash load data can be recorded in the most convenient form a s
expressed by C, D , E, o r F below. In any case, record the
major species comprising the slash, the crown class code
(dominant [ l ] o r intermediate [ 2 ] ) , and the average d.b.h.
of each species. You may record the percentage of foliage
retention by species if you would rather use your own data
than have the program make these estimates for you.
C.
Total slash load - If the total slash load is available,
enter a s tons per acre; otherwise leave blank. If
entered, record percentage of slash contributed by
each species.
D.
Total 10-h load - If the total 10-h load is known,
enter a s tons p e r acre; otherwise leave blank. If
entered, record percentage of the slash contributed by
each species.
E.
Species 10-h load p e r acre - Record the major
species comprising the slash and the 10-h load p e r
acre for each species. Enter a s tons p e r acre. Entry
of percentage slash by species is not required.
F. Species intercepts per foot - Record the species name
and the number of 10-h intercepts per foot for each
major species comprising the slash. Entry of percentage slash by species is not required.
G. Average twig retention, all species.
Enter the percentage of twigs less than 114-inch diameter still
retained on the slash. Estimate an average value for
all the slash, r a t h e r than for each species.
H. Area coverage
slash.
-
Enter percentage of area covered by
Multiple
1-Hour Fuels
Although the familiar 3-dead-class, 2-live-class fuel model should
be adequate for most fuel modeling jobs, there may be situations
where two distinctly different sizes of 1-h fuels exist. One example
might b e dead leaves and twigs on frost- o r drought-killed s h r u b s ;
another example i s r e d coniferous slash such as ponderosa pine where
the needles have a much smaller average size than the twigs.
The NEWMDL program contains a section that will accept data on
the load and surfacelto-volume ratios for two sizes of 1-h fuels plus
the 10-h, 100-h, and live fuels. This is called a 4-dead-class,
2-live-class fuel model. You are given the option of selecting this
capability early in t h e NEWMDL program when you are asked whether
you want to build a model with one or two sizes of fine fuels. Select
the option for two sizes of fine fuels if you have the data for the
"Multiple 1-Hour Data Entry Form." Appropriate data can be obtained
from option 2 of the DEBMOD program (Puckett and Johnson 1979), or
from a-fuels inventory you conduct in t h e field to get the data. This
section of NEWhlDL requires that you have the data on hand for
direct e n t r y . The program will not give any tutorial assistance on
values to enter.
On completion of data e n t r y , the program will change your 4-deadclass, 2-live-class model to a 3-dead-class, 2-live-class model so that
it will b e compatible with the r e s t of the BEHAVE system and the
TI-59. The 1-h load and fuel bed depth must b e altered in this
process to preserve the fire behavior characteristics of t h e model, so
do not be concerned about that. The resultant 3-dead-class, 2-liveclass model should be tested with the TSTMDL program where you can
make any necessary adjustments.
The "Multiple 1-Hour Data Entry Form" is simple enough that
detailed ex p lanation should not be necessary. J u s t record and enter
the data for those components that contribute significantly to the fuel
model. Remember, this section of the program expects direct e n t r y
of your data. It will not suggest values to e n t e r .
Estimating surface-area-to-volume ratios. --When using N EWMDL to
enter your data, you will be asked for surface-area-to-volume (SIV)
estimates. The following tabulation presents three broad ranges of
h
SIV ratios for g r a s s , broadleaf, a n d coniferous plants. ~ l t h o u ~the
specific p l a n t ( s ) you are concerned with may not be listed, you
should be able to find a plant similar enough to select among the
three SIV ratio ranges. The midpoint of the appropriate ran g e would
b e a good initial value. You may adjust this value later when using
the TSTMDL program to modify your initial fuel model.
Estimating heat content.--Heat content estimates are requested
when you enter your fuel model data into NEWMDL. Guidelines a r e
provided by the program and will not b e repeated here.
Surface-Area-to-Volume Ratio Ranges f o r Various Plants
500-1,500 f t 2 / f t 3
1,500-2.500 ft 2 1ft 3
More t h a n 2,500 f t 2 / f t 3
Grasses
Jamaica s a w g r a s s
(Moriscus jamaicensis)
Fountaingrass
(Pennisetum setaceum)
Molassesgrass
(Melin is minu tiflora)
Yellow beadlil y
(Clin tonia borealis)
Sonoma manzanita
(Arctos taphylos densiflora)
Palmetto
(Sabal s p p . )
Common pearleverlasting
(Anaphalis margaritacea)
Gallberry
Medusahead
( Toeniatherum asperum)
Cheatgrass
(Bromus tectorum)
Pinegrass
(Calamagrostis rubescens)
Idaho fescue
(Fes tuca idahoensis)
C r e s t e d wheatgrass
( A gropyron spicatum)
Broomsedge
(Andropogoti v i r p i n i c u s )
Broadleaved p l a n t s
Spreading dogbane
(Apocynum androsaenifolium)
Bigleaf a s t e r
( A s t e r macrophyllus)
Marsh peavine
(Lathyrus palustris)
Interrupted- fern
(Osmunda claytoniana)
Eucalyptus
(Eucalyptus obliqua)
Wild sarsaparilla
(A ralia nudicaulis)
B u n c h b e r r y dogwood
(Cornus canadens i s )
Brackenfern
(Pteridium aquilinum)
Serviceberry
(Amelanchier s p p . )
Roundleaf dogwood
(Cornus rugosa)
Willow
(Salix s p p . )
Showy mountainash
(Sorbus decora)
Ninebark
(Physocarpus molvaceus)
Oceanspray
(Holodiscus discolor)
Mountain alder
( A l n u s sinuata)
Menziesia
(Menziesia f e r r u g i n e a )
Snowberry
(Symphoricarpos albus)
Blue h u c k l e b e r r y
( Vaccinium globulare)
Quaking aspen
(Populus tremuloides)
Red maple
(Acer r u b r u m )
White oak
(Quercus alba)
Scrub oak
(Quercus dumosa)
Oregon- grape
(Berberis repens)
Conifer needles
Eastern hemlock
( Thuja canadensis)
Northern white- cedar
( Thuja occidentalis)
Jack pine
(Pinus banksiana)
Balsam fir
(A bies balsomea)
Ponderosa p i n e
l Pinus ponderosa)
Engelmann s p r u c e
(Picea engelmannii)
Lodgepole pine
(Pinus contorta)
Douglas-fir
(Pseudotsuga menziesii)
Grand fir
(Abies grandis)
Loblolly pine
( Pinus taeda)
Western r e d c e d a r
(Thuja plicata)
Eastern white pine
(Pinus s t r o b u s )
Western white pine
(Pinus monticola)
Western hemlock
(Tsuga heterophylla)
Western larch
( L a r i x occidentalis)
PROGRAM TSTMDL
General Concept
The purposes of TSTMDL a r e to: (1) provide a means to examine
t h e fire behavior characteristics of the initial fuel model under a
variety of environmental conditions, and ( 2 ) provide a convenient
method to examine the effect on fire behavior when individual fuel
model parameters a r e modified. A l t h o ~ g hthe NEWMDL and TSTMDL
programs systematize fuel modeling, it is far from a mechanical
process t h a t produces incontrovertible results.
I t is extremely
important to t e s t e v e r y fuel model for t h e broadest r a n g e of
environmental conditions to which i t may b e applied. Otherwise you
may find, for example, that a fuel model that works well for low fuel
moistures or windspeeds produces unrealistic fire behavior for high
moistures or windspeeds. These t e s t s can and should be performed
with the TSTMDL program, b u t you a r e also encouraged to t e s t any
new model with the BURN program to verify t h a t it will not produce
spurious results when used operationally.
The initial verification of a fuel model r e s t s upon your judgment of
whether the r a t e of s p r e a d , flame length, and other values a r e
reasonable for a range of environmental conditions. Field verification
can only be attained by using the model and comparing i t s predictions
with actual observations. Rothermel and Rinehart (1983) define
techniques for observing fire behavior that can be used to assess
whether your fuel model produces reasonable values.
TSTMDL has both a "normal" and a "technical" version. The
program defaults to the normal version when you first begin. The
normal version is for those situations in which a model can be built
r a t h e r easily, without a need for extensive examination. I t provides
t h r e e g r a p h s and a table. The g r a p h s are: (1) r a t e of spread
v e r s u s midflame windspeed, ( 2 ) flame length v e r s u s midflame windspeed, and ( 3 ) the fire characteristics chart (Andrews and Rothermel
1982). Rate of spread and flame length are graphed for either one o r
t h r e e values of 1-h fuel moisture over a midflame windspeed range of
0 to 18 milh. This chart enables comparison of your fuel model's
behavior characteristics plots to one or two of the 13 NFFL fuel
models for currently defined environmental conditions.
The tabular output i s identical in both the normal and technical
versions. I t allows you to assign t h r e e values to any environmental
parameter, then lists the fuel model and the values calculated for
five fire behavior parameters: (1) r a t e of s p r e a d , ( 2 ) flame length,
(3) reaction intensity, ( 4 ) heat p e r unit a r e a , and ( 5 ) fireline
intensity.
The technical version provides additional graphic output. I t allows
you to place any fuel or environmental parameter on the x-axis and
examine i t s affect on any appropriate fire behavior parameter. Thus
t h e technical version provides a great deal of flexibility, and a
powerful means to examine the influence of the fuel model parameters
on fire behavior calculations. The interactions between the fuel
model, fire model, and environmental parameters a r e exceedingly
complex. You will undoubtedly get some mystifying plots, b u t the
educational value of this program lies in understanding them.
Program Structure
The TSTMDL program has three sections, each controlled b y keywords. The f i r s t section is the "control," which permits task selection and general program control; the second section is the "fuel and
environment manipulationtt section for changing values of individual
parameters, and the third section is the "fuel a n d environment modification" section, which provides for data e n t r y and listing (fig. 7 ) .
PROGRAM TSTMDL
I
C
O
N
T
R
O
L
S
E
C
t
FUEL
MANIPULATION
FUEL
MODIFICATION
SECTION
ENV lRONMENT
SECTION
ENVl RONMENT
T
I
*
t
PARAMETER
O
N
MANIPULATION
SECT l ON
t
MODIFICATION
SECT ION
PARAMETER
Figure 7.--General flow of program TSTMDL. The TSTMDL program
has three sections: control, fuel o r environment manipulation, and
fuel o r environment modification. Keywords associated with each
section provide user control.
When you a r e at the "control" section, you get to the "fuel" or
"environment" manipulation section by entering keyword FUEL or
ENV, respectively. Then, e n t r y of keyword CHANGE takes you to
t h e t h i r d section, the "fuel modification" or "environment modification"
section. Each e n t r y of kevword QUIT moves you up one section.
Thus you QUIT section t h r e e to get to section two and also QUIT
section 2 to get back to t h e "control" section. Entering QUIT from
t h e "control" section terminates operation of the program.
The keyword method of program control permits much flexibility in
program operation. For example, whenever prompted for a keyword,
you can e n t e r any keyword belonging to the section where you a r e .
Thus program flow does not follow a s t r i c t p a t t e r n , b u t allows you to
perform tasks defined for each section in any sequence. This capability i s symbolized by the dot and short line leading to each keyword. Note that only t h e keywords FUEL, ENV, CHANGE, and QUIT
will move you from one section to another.
A list of keywords and their functions in program control and
manipulation of fuels and environmental data i s provided in table 1.
Table 2 provides a list of keywords for selecting an environmental
variable to which additional values can b e temporarily assigned for
tabular i n p u t , and a list of variables t h a t can be assigned to the X
and Y axes when using the technical version's graphics.
Table 1.--TSTMDL keywords and functions
Control Section
KEYWORD
FUNCTION
KEY
TERSE
WORDY
NORM
TECH
FUEL
ENV
GRAPH
TABLE
RENUMBER
RESTART
FILE
TI59
QUIT
Prints this keyword list
Set terse mode for minimal prompting
Set wordy mode for full prompting
Implement "normal" version of program
Implement "technical" version of program
Go to "fuel manipulation" section
Go to "environment manipulation" section
Request graphic output of computed results
Request tabular output of computed results
Renumber fuel model and select dynamic or
static
Start program at beginning again
Access the fuel model file
List fuel model and TI-59 registers
Quit this session with TSTMDL
-
Fuels and Environment Manipulation Section
Fuels
KEYWORD
NEW
NFFL
CHANGE
FUNCTION
Enter new fuels
data
Enter a fire
behavior model
Go to "fuel modification" section
Enviroment
KEYWORD
NEW
STD
CHANGE
LIST
List fuel model
LIST
QUIT
Go to "control"
section
QUIT
FUNCTION
Enter new environmental data
Enter standard
environmental data
Go to "environment
modification"
section
List environmental
data
Go to "control"
section
Fuels and Environment Modification Section
Fuels
KEYWORD
SA1
SAH
SAW
DEPTH
HEAT
EXTM
L1
L10
LlOO
LH
LW
KEY
QUIT
FUNCTION
Change the:
1-HR S/V ratio
Herb S/V ratio
Woody S/V ratio
Fuel bed depth
Heat content
Extinction moisture
1-HR fuel load
10-HR fuel load
100-HR fuel load
Herbaceous load
Woody load
List these keywords
Go t o "fuel manipulation" section
Environment
KEYWORD
M1
M10
MlOO
MHERB
MWOOD
WIND
SLOPE
QUIT
KEY
FUNCTION
Change the:
1-HR fuel moisture
10-HR fuel moisture
100- HR fuel moisture
Live herb moisture
Live woody moisture
Midflame windspeed
Percent slope
Go to "environment
manipulation"
section
List these keywords
Table 2 . --TSTMDL keywords for tabular
and graphic output
Tabular Output Keywords
KEYWORD
KEY
M1
M10
MlOO
MHERB
MWOOD
WIND
SLOPE
FUNCTION
Print this keyword list
1-HR fuel moisture
10-HR fuel moisture
100-HR fuel moisture
Live herb fuel moisture
Live woody fuel moisture
Mid flame windspeed
Slope
Graphic Output Keywords
KEYWORD
KEY
SA1
SAH
SAW
L1
L10
LlOO
LH
LW
DEPTH
EXTM
HEAT
A41
M 10
MlOO
MHERB
MWOOD
WIND
SLOPE
MEANING
Print this keyword list
1-HR S / V ratio
Herb S/V ratio
Woody S / V ratio
1-HR fuel load
10-HR fuel load
100-HR fuel load
Herb fuel load
Woody fuel load
Fuel bed depth
Extinction moisture
Heat content
1-HR fuel moisture
10-HR fuel moisture
100-HR fuel moisture
Herb fuel moisture
Woody fuel moisture
Midflame win dspeed
Percent slope
TSTMDL Technical Version Y-axis Keywords
FLINT
RATE
REAC
FLAME
H/A
PACK
RSFL
Program Operation
Fireline intensity
Rate of spread
Reaction intensity
Flame length
Heat p e r unit area
Packing ratio
Rate of spread to
flame length ratio
The specific procedure for accessing your computer and the
TSTMDL program must b e obtained from your computer specialist.
When you begin, the first message will indicate that you a r e using
the fuel model testing program and ask you to e n t e r your last name.
A maximum of 20 characters is allowed. Then you will b e asked if
you are using a h a r d copy device such a s a printing terminal. The
purpose of this question i s to indicate whether pauses a r e necessary
in the flow of output, as when a CRT screen i s filled.
..
Your next response will be to indicate whether you want the
TERSE mode. ~ n s w e r"No" unless you a r e an experienced u s e r .
You will then be asked whether you will b e creating a new fuel
model o r loading a previously built model from your ,fuel model file.
After making this choice you will either be asked to e n t e r a number for
your proposed new model o r for the previously built model to be selected
from the fuel model file. If you a r e creating a new model you will also
be asked to e n t e r a name for t h e model and whether it is to b e "dynamic"
o r "static. "
The next question i s whether you want a list of keywords and
their functions. Because keywords control the program, this i s a
good time to list them for reference, or you may decline the list.
If you a r e using the WORDY version, the next program prompt i s
a suggestion to e n t e r NORM o r TECH to get the version you want.
This prompt is not printed in the TERSE version. The NORMAL version is t h e default, so if this i s what you want, keyword NORM does
not have to b e entered, b u t doing so will print a message indicating
t h a t t h e NORMAL version i s set. You can get the TECHNICAL
version only by asking for i t .
The next prompt i s "CONTROL SECTION. KEYWORD?". Whenever
this prompt =ppears, you can enter any keyword in t h e control
section keyword l i s t , although you will get e r r o r messages if the
wrong ones a r e entered f i r s t . Such messages will not cause t h e
program to "c r a s h , " but r e t u r n control to the point where you can
enter another keyword. The general approach should be to:
1. Define the fuel model. Keyword FILE will give you a chance
to get a custom model from the fuel model file. Otherwise keyword
FUEL will give you the opportunity to select a fire behavior model,
input new fuel model data, change, o r list all fuel model data.
2 . Define the environmental data. Keyword ENV will allow you to
e n t e r , change, o r list the environmental data. You can either assign
your own values to the environmental parameters, o r select one of the
"standard" conditions.
3 . Define t h e t y p e of output you want; that i s , graphic (keyword
GRAPH), or tabular (keyword TABLE). In either case you will be
asked a few questions required to set up the graph o r table.
After your first time t h r o u g h , in which you set up t h e fuel and
environmental d a t a , you have complete freedom to use the keywords
in any order. For example, you can enter keyword FUEL or ENV,
change the value of one o r more fuel o r environmental parameters,
then output another graph o r table. You can also switch between t h e
TERSE and WORDY modes or t h e NORMAL and TECHNICAL versions
whenever "CONTROL SECTION. KEYWORD?" is printed.
It is not necessary to e n t e r decimal points unless your intention i s
to e n t e r a decimal fraction. They a r e not required for integer
numbers.
To obtain the list of TI-59 registers and numbers needed to record
this fuel model on a magnetic c a r d , e n t e r keyword TI59.
Like NEWMDL, TSTMDL i s designed to be a friendly and "difficult
to crash" program, so you a r e encouraged to explore i t s capabilities
until you a r e completely familiar with i t s operation.
Remember that although fuel models can be created with t h e
TSTMDL program by entering the data directly, i t s primary purpose
i s for testing models initially built with the NEWMDL program.
FUEL MODELING CONCEPTS
Introduction
The Fire Spread Model
Interactions between fuel model, topography, and environmental
parameters, and t h e mathematical fire spread model are so numerous
gnd complex t h a t attempting to present -all t h e possible r e s u l t s would
be an unreasonable task. Yet a basic understanding of the relationships provides valuable insight to t h e fuel modeling process.
Therefore this section is presented for those who a r e interested in
examining in detail the concepts most important to fuel modeling.
The mathematical fire model developed by Rothermel (1972) and
amended b y Albini (1976) provides a means to estimate t h e r a t e at
which a fire will spread through a uniform fuel a r r a y t h a t may
contain fuel particles of mixed-sizes. I t is basically- a r a t e of- spread
model, b u t it also computes an intensity t h a t can b e i n t e r p r e t e d into
t h e more familiar fireline intensity and flame length developed by
Byram (1959).
--
The theoretical basis for t h e fire spread model was developed by
Frandsen (1971). The terms of Frandsen's equation could not b e
solved analytically, however, so i t was necessary to define new
terms, reformulate the equation, and design experimental methods to
evaluate t h e individual terms. The final form of the r a t e of s p r e a d
equation, derived b y Rothermel (1972), which will b e examined in
depth is:
where
R
Ir
i s t h e forward r a t e of spread of the flaming f r o n t , in
feet p e r minute.
i s the reaction intensity--a measure of t h e e n e r g y release
r a t e p e r unit area of fire front ( ~ t u l f t ~ l r n i n ) .
5
('ks;) i s t h e propagating flux ratio--a measure of t h e
proportion of t h e reaction intensity that heats adjacent
fuel particles to ignition.
Ow
(f; wind) is a dimensionless multiplier t h a t accounts for
the effect of wind in increasing the propagating flux ratio.
4,
(f; slope) is a dimensionless multiplier t h a t accounts for
t h e effect of slope in increasing the propagating flux ratio.
Pb
E
(ro) is a measure of t h e amount of fuel p e r cubic foot of
fuel bed ( l b / f t 3 ) .
( e p ' s -1on) i s a measure of the proportion of a fuel particle
t h a t is heated to ignition temperature at the time flaming
combustion s t a r t s .
is a measure of t h e amount of heat required to ignite 1 pound
Qig of fuel ( B t u l l b ) .
Basically t h i s equation shows t h a t t h e r a t e at which fire s p r e a d s i s a
ratio of the heat received by the potential fuel ahead of the f i r e , to
t h e heat r e q u i r e d to ignite t h i s fuel. T h u s if fire can b e thought of
a s a series of ignitions, i t will p r o g r e s s through a fuel bed at t h e
r a t e at which adjacent potential fuel can b e heated to ignition tempera t u r e . Only a small portion of the heat produced in the flaming front
of a wildland fire reaches nearby unignited fuel. The majority of t h e
i
,
heat i s c a r r i e d upward b y convective activity o r is radiated in o t h e r
directions. The numerator of t h e above equation r e p r e s e n t s t h e
amount of heat actually received b y t h e potential fuel, while t h e
denominator r e p r e s e n t s t h e amount of heat r e q u i r e d to b r i n g t h i s fuel
to ignition temperature.
Definition of Terms
in t h e S p r e a d Equation
This section p r e s e n t s a detailed explanation of how fuels, weather,
and topographic i n p u t s affect t h e s e terms. Your fuel modeling
capabilities will b e improved b y u n d e r s t a n d i n g these relationships.
We will explain t h e concept of t h e s p r e a d equation b y f i r s t defining t h e individual terms a n d briefly discussing what t h e y r e p r e s e n t .
Then we will look at t h e terms in g r e a t e r detail to examine how fuels,
weather, a n d topography affect them.
REACTION INTENSITY (IL.)
Reaction Intensity ( I ) i s a measure of t h e e n e r g y release r a t e ,
p e r unit area of t h e fir: f r o n t . The u n i t s assigned to it a r e :
~ t u l f t ~ l r n i n .I t i s affected b y :
1 . Size of t h e individual fuel particles. Fuel particle size
strongly influences fire s p r e a d and intensity. In almost all f i r e
situations, t h e fire front advances t h r o u g h fine fuels such a s g r a s s ,
s h r u b foliage, o r l i t t e r . Both t h e size of t h e particles a n d t h e i r
compactness a r e important. The f i r e model u s e s a description of t h e
fuel particle surface-area-to-volume ratio a s t h e i n p u t describing
particle size. The smaller t h e particle, t h e l a r g e r i t s surface- areato-volume ratio. This can be visualized b y cutting a fuel particle in
half, lengthwise. The total volume of material remains t h e same, b u t
additional s u r f a c e area is contributed by each of t h e two c u t
s u r f a c e s . T h u s t h e surface-area-to-volume ratio increases. This
process i s amplified a s more c u t s a r e made, producing e v e r smaller
particles b u t more s u r f a c e a r e a .
For long, cylindrical objects s u c h a s conifer needles, twigs, a n d
g r a s s e s , t h e a r e a of t h e e n d s can be neglected, s o t h e surface- areato-volume ratio can b e found b y dividing t h e diameter i n t o t h e
number 4. For flat objects such a s leaves t h a t have v e r y little a r e a
on t h e i r e d g e s , t h e surface-area-to-volume ratio can b e found b y
dividing t h e thickness into t h e number 2 . The unit of feet is u s e d
for all measurements. For example, 114-inch diameter sticks have a
surface-area-to-volume ratio of 192 f t 2 1f t 3 . The units a r e often simplified to l l f t o r f t - l .
Expressing diameter and thickness of small
fuels in feet is awkward, b u t avoids t h e problem of wondering what
units were u s e d in various p a r t s of t h e model. T h e mathematical
symbol u s e d to r e p r e s e n t surface-area-to-volume ratio i s t h e small
Greek l e t t e r , sigma, a .
When a fuel a r r a y i s composed of different size particles, t h e f i r e
model u s e s their individual s u r f a c e a r e a s , a n d t h e proportion of t h e
total s u r f a c e a r e a contributed b y each size class, to a r r i v e a t a c h a r acteristic size t h a t r e p r e s e n t s t h e a r r a y . I t i s then assumed t h a t t h e
a r r a y would b u r n a s if it were composed of only fuel particles of t h e
characteristic size.
The timelag concept u s e d in t h e National Fire-Danger Rating System
(Fosberg and Deeming 1971) for describing fuel particle size of dead
fuels i s also used in NEWMDL a n d TSTMDL. Only t h e foliage a n d fine
stems of living fuels a r e considered. These a r e described a s e i t h e r
herbaceous" for shallow-rooted g r a s s e s and herbaceous p l a n t s , o r
"woody" for deep-rooted s h r u b s . For woody p l a n t s , only t h e foliage
and twigs less t h a n 114-inch diameter a r e considered.
2 . T h e compactness of t h e fuel b e d , which i s e x p r e s s e d as t h e
packing ratio. At t h e two extremes, a fuel b e d may contain no
fuel--packing ratio i s 0--or i t may b e a solid block of wood--packing
ratio i s 1. T h u s , e x p r e s s e d a s a p e r c e n t a g e , t h e packing ratio i s
t h e percentage of t h e fuel bed t h a t i s composed of fuel, t h e remaind e r being air space between t h e individual fuel particles.
A very
compact fuel bed b u r n s slowly because airflow i s impeded, and t h e r e
a r e so many particles to be heated to ignition in a given length of the
bed. A very open or porous fuel bed b u r n s slowly because the individual fuel particles a r e spaced so f a r apart t h e r e is little heat transfer between them. That i s , each particle in t h e fuel bed would b u r n
as an individual. The maximum reaction intensity occurs at some
intermediate packing ratio. The effect of fuel particle size and
packing ratio upon the reaction intensity is incorporated in an important intermediate term called the reaction velocity. The reaction
velocity is a ratio of how efficiently the fuel will be consumed to the
burnout time of the characteristic fuel particle size. Therefore, fine
fuel a r r a y s arranged to b u r n most thoroughly in the shortest time
have the largest reaction velocity. Fine fuel particles have higher
reaction velocity in fuel a r r a y s that a r e very loosely packed, whereas
larger fuel particles need to be closer together to b u r n well.
Each size fuel particle has an optimum packing ratio. In the absence
of wind the optimum packing ratio for any particle size is determined
by a mathematical expression in the fire model. This relationship is
illustrated in figure 8. In t h e presence of wind, the optimum packing
ratio shifts to less tightly packed fuel a r r a y s . The reaction velocity
is depicted in figure 9 for a r a n g e of particle sizes and packing
ratios. Note the s h a r p reduction in reaction velocity on either side
of the optimum packing ratio.
Because the reaction intensity depends directly upon reaction veloci t y , it has the same dependence upon fuel particle size and packing
ratio just described for reaction velocity.
FUEL PARTICLE SURFACE AREA1 VOLUME RATIO
(Fr21 Fr')
Figure 8. -- Optimum packing ratio.
Fuel particle surface-area-to-volume
r a t i o determines the optimum packing
r a t i o for any fuel a r r a y .
3 . Moisture content of the fuel. Higher moisture contents reduce
reaction intensity because more of the heat released during combustion is required to evaporate the moisture. Less heat i s available to
raise the next fuel particle to ignition temperature.
4. Chemical composition. Although the quantity and type of
inorganic material in the fuel affects t h e r a t e a t which it b u r n s , our
primary concern i s the heat content--the Btu's of heat released d u r ing combustion of 1 pound of fuel. The heat content i s lowest for
those fuels with few volatiles--oils and waxes--and higher for those
F i g u r e 9 . --Reaction velocity. The reaction velocity decreases sharply when the packing r a t i o is shifted from its
optimum value for any given surface-area-to-volume r a t i o .
with more of them. Fuels having higher heat contents have more heat
available p e r pound of fuel. The r a t e at which this heat will b e
released depends on t h e particle size, the packing ratio, t h e moisture
content, and the mineral content of the fuels. At this time t h e effect
of inorganic materials o r minerals associated with salts in the fuel i s
not adjusted in NEWMDL o r TSTMDL although it is variable in t h e fire
model. The total salt content for all fuels i s assumed constant at
5.55 percent for all fuel models and the effective salt content is
assumed constant a t 1.0 percent (Rothermel 1972).
To examine some of these points graphically, figure 10 illustrates
t h a t a s t h e size of t h e individual fuel particles increases (surface- tovolume ratio g e t s smaller), they must b e packed more tightly to maximize t h e reaction intensity. That i s , t h e maximum reaction intensity
for fine fuels occurs at a packing ratio of about 0.03 ( 3 percent of
t h e fuel bed i s wood), while it occurs at a packing ratio of about
0.08 for 114-inch sticks and 0.10 for 112-inch sticks. The packing
ratio producing the maximum reaction intensity for a particular size
fuel particle is called the optimum packing ratio. At t h e optimum
packing ratio, the fuellair mixture i s optimized for efficient combustion. Figure 10 also illustrates t h a t reaction intensity decreases
----
FINE FUEL
'-FUEL
,
1 /2'- FUEL
1/4
Cor@6
F i g u r e 70.- - Reaction i n t e n s i t y . T h e moximum r e a c t i o n
i n t e n s i t y o c c u r s a t h i g h e r p a c k i n g r a t i o s f o r l a r g e r fuel
p a r t i c l e s t h a n f o r small ones. T h e r e a c t i o n i n t e n s i t y
decreases when t h e p a c k i n g r a t i o is e i t h e r less t h a n o r
g r e a t e r t h a n optimum f o r a n y g i v e n fuel p a r t i c l e size.
when the packing ratio varies from i t s optimum value for any given
fuel particle size.
From a fuel modeling standpoint, it is important to know that
although the reaction intensity is maximized at the optimum packing
ratio, this does not necessarily hold for r a t e of spread and flame
length. Altering load and depth to adjust the packing ratio also
affects the amount of heat required to ignite the fuel, a s expressed
by the denominator of the spread equation and the proportion of heat
transferred to the fuel ahead of the fire as expressed by the propagating flux ratio 5 . Thus r a t e of spread and reaction intensity
do not peak at the same packing ratio.
Tabular output from TSTMDL provides both the packing ratio for
t h e model and a result labeled PRIOPR. The PRIOPR value is the
ratio of actual packing ratio to optimum packing ratio. It is less
than 1 if the packing ratio of the fuel model is less than optimum,
1 if they are equal, a n d greater than 1 if the fuel model packing
ratio exceeds the optimum value. There is no rationale for attempting
to adiust loads and d e ~ t huntil PRIOPR eauals 1. In fact. it normally exceeds 1 for compact "horizontally oriented" fuels such a s
needle litter, but is usually less than 1 for vertical fuels such as
g r a s s . This number will indicate how tightly your fuel model i s
packed should you want to make this comparison with one of the more
familiar NFFL fuel models. Division of packing ratio by the PRIOPR
value yields the optimum packing ratio.
The heat content is the only chemically oriented fuel model parame t e r u s e r s can change. Increasing the heat content always produces
a "hotter" fuel model, while decreasing it reduces the calculated fire
behavior.
Remember that reaction intensity ( I ) is t h e total heat release rate
r
p e r unit area of fire front, and includes heat convected, conducted,
and radiated in all directions, not just the direction of the adjacent
potential fuel. The next term discussed serves to adjust this total
energy release r a t e down to t h a t portion which i s effective in propagating the fire.
The propagating flux is t h a t portion of the total heat release rate
from a f i r e , which is t r a n s f e r r e d and absorbed by the fuel ahead of
the fire, raising i t s temperature to ignition. The propagating flux is
calculated under the assumption that the fire i s burning on a flat
surface and in calm air (no wind, no slope). Effects of wind and
slope are discussed later.
The parameter 5 in the r a t e of spread equation represents a ratio
between this no-wind, no-slope propagating flux [ ( I ) ] and the
reaction intensity ( I r )
.
P
0
Mathematically it is defined -as:
It
expresses what proportion of the total reaction intensity ( I )
r
actually heats adjacent fuel particles to ignition. Propagating flux
ratios can vary from zero--no heat reaches adjacent fuels--to 1--all of
the heat reaches adjacent fuel. Realistically, and expressing the
propagating flux ratio in percentage, typical values range from about
1 percent to 20 percent. Multiplying t h e f i r s t two terms in the
numerator of the spread equation--reaction intensity times propagating
flux ratio ( I E)--produces the propagating flux, ( I p ) which is an
r
estimator of the r a t e of heat t r a n s f e r that would drive the fire forward in a no-wind, no-slope situation.
The propagating flux ratio is affected b y :
1. The average size of t h e fuel particles in t h e fuel bed, that i s ,
the characteristic surface-to-volume ratio.
2 . The packing ratio, o r fuel bed compactness a s explained previously.
Figure 11 shows the effect of both packing ratio and average fuel
size on the propagating flux ratio. Note that at a constant packing
ratio--0.04 i s highlighted--the propagating flux ratio is g r e a t e r for
fine fuels than for coarse ones. As shown by figure 11, the
propagating flux ratio tends to increase with increasing packing ratio,
b u t the effect is much more pronounced in the finer fuels.
This implies that if fuel bed depth is kept constant and the dead
fuel load (1-h, 10-h, and 100-h) is increased, thereby increasing the
packing ratio, then a g r e a t e r proportion of the heat produced by t h e
fire will be effective in preheating the adjacent unburned fuel. This
effect i s more pronounced in the finer fuels. Remember, however,
the reaction intensity is also strongly affected by the packing ratio.
Reaction intensity will decrease if the fuel bed is either too tightly
packed, or too loose. Similarly, the amount of fuel that must be
heated to ignition is increased as fuel load is increased, t h u s illustrating that it is not easy to guess how fuel changes will affect fire
behavior.
F i g u r e 1 7 . - - Propagating f l u x r a t i o . The p r o p a g a t i n g
f l u x r a t i o increases much faster for f i n e fuels than
coarse ones, as the packing r a t i o increases. B u t a t
any packing r a t i o , the propagating f l u x r a t i o is
h i g h e r f o r the f i n e r fuels.
WIND COEFFICIENT (@")
In t h e discussion of t h e no-wind propagating flux ratio ( 5 ) it was
assumed t h e r e was no ambient wind and the t e r r a i n was flat ( f i g .
1 2 ) . When this i s not t h e case, wind and slope coefficients ( 4 ) and
W
( Q s ) a r e used by t h e f i r e model t h r o u g h t h e expression ( I + @+u$ ) .
W
S
-INTERNAL
RADIATION
& COP
F i g u r e 12.--Schematic of a no- wind f i r e .
Consider t h e no-slope case. The wind coefficient increases rapidly
with windspeed in loosely packed fine fuels, t h u s greatly increasing
s p r e a d r a t e . This occurs because wind tips t h e flame forward a n d
causes direct flame contact with t h e fuel ahead of t h e f i r e a s well a s
increased radiation from t h e flame to t h e fuel. This greatly increases
t r a n s f e r of radiant and convective heat t o u n b u r n e d fuel ahead of t h e
fire (fig. 1 3 ) .
.--
-
SOL1 D M A S S TRANSPORT
A
d
WIND
Yw -
/ CONVECT I ON
-
INTERNAL RAD I AT I ON
/
& CONVECT ION
F i g u r e 7 3. - - Wind- driven f i r e . Increased radiant and convective
bed transfer contributes to faster spread rates in wind- driven
fires.
The wind coefficient i s affected b y :
1. The fuel bed's characteristic surface-area-to-volume ( S I V )
ratio. Figure 14 illustrates t h e effect of increasing the characteristic SIV ratio of a fuel bed whose packing ratio i s half t h e optimum.
Note that increasing the characteristic SIV ratio increases the wind
coefficient, and that the effect is greater at higher windspeeds. A
similar b u t less pronounced effect occurs for fuel b e d s with higher
packing ratios.
2 . The packing ratio of the fuel bed. For this discussion, a
relative packing ratio i s introduced. I t is the ratio of t h e actual
packing ratio divided by the optimum packing ratio. I t s value is 1.0
when beds a r e packed optimally in the no-wind case. Figure 15 illust r a t e s t h e effect of increasing packing ratio in a fuel bed whose characteristic S / V ratio i s 1,500. Note that the wind coefficient decreases
rapidly as the fuel bed i s more tightly packed, and that t h e effect i s
more pronounced a t low packing ratios. A similar b u t more pronounced effect occurs with finer fuels.
Figure 7 4 . --Effect of fuel particle surface-area-to-volume
r a t i o on wind coefficient. The effect of wind on f i r e
increases more r a p i d l y for fine fuel than for coarse fuel.
F i g u r e 15. --Effect of packing r a t i o on the wind
coefficient. Wind has a greater effect on fires
in loosely packed fuels than t i g h t l y packed
fuels, w i t h this effect being more pronounced a t
low packing ratios.
3. The windspeed. Obviously an increase in windspeed will produce an increase in the wind coefficient. Even here there can be a
limit which will be discussed soon.
The wind coefficient is increased by increasing the S / V ratio of
the 1-h, live herbaceous, o r live woody fuels. Reducing the packing
ratio by either reducing the fuel load o r increasing the fuel bed
depth also increases the wind coefficient. Remember, however, that
packing ratio also affects reaction intensity. So decreasing the
packing ratio will increase the wind coefficient, but if the packing
ratio falls below optimum, the reaction intensity will decrease even
though the wind coefficient may be r a t h e r large.
Before leaving this discussion of wind's effect on fire behavior
modeling, one note of caution i s in o r d e r . That i s , while wind generally increases fire spread rate and intensity, there i s a limit to this
effect. McArthur (1969) measured r a t e of spread on heading grassland fires in Australia and found that excessive wind actually
reduced the spread r a t e (fig. 16). Although the fire model does not
predict reduced spread r a t e at high windspeed, it does identify
when maximum spread is reached. F u r t h e r increases in windspeed
will not give higher spread r a t e s ; the model will continue to predict
the maximum for those fuel conditions. The effect i s caused by t h e
wind forces being stronger than the convective forces of the fire.
This will occur when the effective windspeed (milh) equals 11100
of the reaction intensity ( ~ t u / f t ~ / m i n )Effective
.
windspeed i s
the no-slope midflame windspeed that produces the same spread rate
as for a fire burning upslope and upwind. Effective windspeeds
having a magnitude greater than 0.011 will not increase the
r
calculated r a t e of spread. This wind limit may also be expressed
a s 9/10 of the reaction intensity when the windspeed is in feet p e r
minute. This effect i s most likely to be noticed with fuel models that
represent s p a r s e fuel types. For example, at 1 percent fuel moist u r e , NFFL model 1 (short g r a s s ) produces a maximum spread r a t e
when the effective windspeed i s 1 2 milh, while a 42 milh effective
wind is required to reach the windspeed limit for NFFL model 3 (tall
g r a s s ) at the same moisture content.
Source of data
Kongorong Fire. 5 . A.
x Geelong Fires
Vic.
Longwood Fire. Vic.
0 Tasmanian Fires
AVERAGE WINDSPEED (MI1 H)
F i g u r e 16. - - Reproduction of M c A r t h u r l s (1 969) r a t e of spread data for
grass. The windspeed was measured a t a h e i g h t of 33 feet above the
g r o u n d in the open.
SLOPE COEFFICIENT (4,)
The effect of slope i s introduced by the coefficient ( 4 ) in the
S
expression ( 1 + 4
+ $ s ) . Wind is eliminated from this discussion by
W
assuming the wind coefficient ( 4 ) i s zero. Then a s the slope
W
increases from 0 p e r c e n t , where it does not affect s p r e a d r a t e , to
some larger value, the r a t e of spread steadily increases. The
mechanism producing this effect is the same as for wind--improved
heat t r a n s f e r because the flames a r e closer to unburned fuels on
steeper slopes (fig. 1 7 ) . The effect, however, is not a s pronounced
as it i s with wind.
The slope coefficient i s affected b y :
1. Slope steepness. The slope coefficient increases a s slope
steepness increases. Negative slopes a r e not accepted by the model.
A discussion of backing fires on slopes and cross-slope fire s p r e a d is
given b y Rothermel (1983).
2 . The packing ratio of the fuel bed. A s for the discussion on
the wind coefficient, the effect of packing ratio i s illustrated (fig.
18) from half to twice the optimum. The slope coefficient was determined for fine fuels, which a r e largely responsible for fire s p r e a d .
The packing ratio of a fuel model will slightly influence i t s sensitivity to slope steepness. This effect, however, i s small relative to
the magnitude of other effects produced b y changes in packing ratio
and so need not be of great concern to the fuel modeler. Changing
fuel particle size does not affect the slope coefficient. Wind and
slope are both recognized b y the fire model, b u t t h e r e i s no consideration of interactions between them.
I
-
F i g u r e 77.--Schematic of a f i r e on a slope.
F i g u r e 78.--Effect of packing r a t i o on the slope
A lthough fires spread faster upslope
coefficient.
as slope steepness increases, the effect is much
less than t h a t of wind.
The slope coefficient is
affected l i t t l e b y packing ratio.
BULK DENSITY (P,,)
Bulk density i s the first term to be discussed from the denominator of the r a t e of spread equation. Remember the denominator
expresses the amount o f heat required to bring the fuel to ignition
temperature; that i s , it represents a heat sink. Bulk density i s the
ovendry weight of fuel per cubic foot of fuel bed. The units a r e
l b l f t 3 . I t is determined by dividing the fuel load ( l b l f t 2 ) b y the fuel
bed depth ( f e e t ) . Bulk density can b e increased by increasing the
fuel load or by decreasing the fuel bed depth. I t serves as a-basis
for quantifying how much fuel is potentially available, per cubic foot
of fuel bed. to act a s a heat sink. Not all the fuel i s necessarilv
heated to ignition; this is discussed in the section on the effective
heating number.
I t is important to realize the significance of having the bulk
density in the denominator of the r a t e of spread equation. Increasing
the bulk densitv tends to decrease the r a t e of s ~ r e a dbecause the
total heat sink, as expressed by the denominator, i s increased. This
1
effect, however, is altered by the influence of fuel load on the
reaction intensity, and bulk density on the propagating flux ratio.
Therefore, no absolute statement can b e made with regard to the
effect of altering fuel load or bulk density.
'
EFFECTIVE HEATING
NUMBER (E)
When large logs b u r n , the center of the log may be cool, relative
to the surface that is on fire. That i s , only the outer shell of the
log has been heated to ignition temperature (320° C ) . The effective
heating number ( E ) provides the means to define what proportion of
an individual fuel particle is heated to ignition temperature at the
time flaming combustion s t a r t s . This proportion depends on the size
of the fuel particle. Figure 1 9 shows that nearly the entire fuel
particle for fine fuels i s heated to ignition temperature at the time of
ignition, while a relatively small proportion of larger fuels i s heated
to this degree. Multiplication of t h e bulk density by t h e effective
heating number quantifies the amount of fuel, p e r cubic foot, that
must b e heated to ignition temperature a s the fire progresses. That
i s , this product defines t h e amount of material in the heat sink.
SURFACE AREA- VOLUME RATIO (FT21 FT3)
F i g u r e 79.--Heating number. As fuel particle size decreases,
a greater portion of the fuel particle is heated to ignition
temperature a t the time flaming combustion s t a r t s .
HEAT OF
PREIGNITION (Q
it3
)
Heat of preignition (Q. ) quantifies the amount of heat required to
'g
raise t h e temperature of 1 pound of moist wood from ambient tempera t u r e to t h e temperature at which it will ignite. In this process,
first the water is evaporated from the wood, then t h e d r y wood itself
is heated. The amount of heat required to raise 1 pound of d r y wood
from air temperature to ignition temperature is a reasonably constant
value that can be calculated in advance. The moisture content of
wood, however, is not constant and it strongly affects t h e amount of
I
-
-
heat required to d r y the fuel particle. Figure 20 shows t h a t the heat
of preignition increases steadily a s the moisture content of the wood
increases. Notice t h a t even at zero percent moisture content, 250
B t u t s a r e still required to heat each pound of absolutely d r y wood to
ignition.
Although the product of bulk density times effective .heating
number ( p E ) quantifies how much fuel weight, p e r cubic foot of fuel
b
b e d , must be heated to ignition temperature, the heat of preignition
quantifies how much heat i s required to do t h i s , p e r pound of moist
fuel. T h u s t h e units for Q a r e Btullb. Then the product (pbcQig)
ig
i s the total amount of heat ( B t u t s ) p e r cubic foot of fuel bed that
must be supplied by the propagating flux.
The many interactions produced when fuel parameter values a r e
changed preclude an exact description of how any particular change
mav affect predicted fire behavior. The technical version of TSTMDL
was developed to provide an easy way to examine these changes
graphically. You a r e strongly encouraged to use the technical
graphics section of TSTMDL.
This completes a first look at each term in the r a t e of spread
equation; however, additional fuel modeling insight can be gained
from looking at some of these terms in greater detail, and from
examining the method of weighting the influence of the various fuel
size classes.
W E I G H T E D FUEL MOISTURE ( P C T )
F i g u r e 20. --Heat of p r e i g n i t i o n . The
amount of heat r e q u i r e d t o i g n i t e woody
fuels increases as t h e i r moisture con tent
increases.
Weighting of
Fuel Size Classes
Even though a fuel model may contain several fuel size classes,
each having a different surface-area-to-volume ( S I V ) ratio, a, the
mathematical fire model requires t h a t just one SIV ratio value represent the entire fuel complex being modeled. The method of calculating this value weights the importance of each fuel class by i t s
surface a r e a , thus emphasizing the smaller fuels, which have the most
effect on spread r a t e . A brief discussion of the weighting procedure
may clarify some of the g r a p h s produced by TSTMDL. Several tabulations will be used to help illustrate the weighting procedure, by
placing an unusually large load in successive fuel classes. For these
tabulations, the S/V ratio of each fuel class will be assigned these
constant values :
= 2,000 f t 2 / f t 3
1-h S / V ratio ( u l h )
10-h S / V ratio ( u l O h )
=
109 f t 2 / f t 3
100-h SIV ratio ( u l O o h )
=
30 f t 2 / f t 3
Live herbaceous S/V ratio ( u h b )
= 1,800 f t 2 / f t 3
Live woody SIV ratio ( u
= 1,500 f t 2 / f t 3
wd
)
The fuel model loads for the six example cases will be:
Case
number
1-h
Fuel model load (tons / acre)
Live
Live
10-h
100-h
herbaceous
woody
The first s t e p in the weighting procedure is to determine t h e
square feet of fuel surface area p e r s q u a r e foot of fuel b e d for each
fuel size class. These values are determined for each size class by
dividing the fuel particle density into the product of fuel particle S/V
ratio times the ovendry load of that class. That i s , the surface area
of any given fuel class, p e r cubic foot of fuel bed, obtained by canceling equivalent units of measure is:
/ f t 2 of fuel surface area)
\
f t 3 of fuel volume
/
) - (ft
Ib of fuel
2
\
of fuel b e d )
lb of fuel
- /ft2
\
of fuel surface area\
f t 2 of fuel bed
f t 3 of fuel volume
These surface areas will be r e f e r r e d to as:
Alh
A l ~ h
= f t 2 of 1-h fuel surface area per f t 2 of fuel bed
= f t 2 of 10-h fuel surface area p e r f t 2 of fuel bed
= f t 2 of 100-h fuel surface area p e r f t 2 of fuel bed
A l ~ ~ h
Ahb
= f t 2 of live herbaceous fuel surface area p e r f t 2
of fuel bed
Awd
= f t 2 of live woody fuel surface area p e r f t 2 of fuel bed.
Then the surface areas for all t h e fuels in the dead category and the
surface areas for all t h e fuels in the live category a r e summed separately:
Adead = A l h
+
A l ~ h
+ A
l ~ ~ h
)
From these two s e t s of numbers, individual fuel class weighting
factors a r e calculated by dividing t h e surface area in each fuel class
by the total surface area in i t s category (live or d e a d ) :
flh
= Alh/Adead
-
f l ~ h- AlOh/Adead
The f i r s t three factors define t h e proportions of t h e total dead
h e 1 surface area t h a t a r e contributed by t h e 1- , l o - , and 100-h fuel
classes, while t h e last two define the proportions of t h e total live fuel
surface area t h a t a r e contributed b y the live herbaceous and woody
fuel classes.
The magnitudes of these weighting factors for the six sample fuel
models a r e shown in t h e listings below. Note t h a t t h e heavily loaded
fuel component has been underlined in each case.
-
Fuel class weighting factor
w
Case number
*lh
f l ~ h
f l ~ ~ h fhb
fwd
Because t h e S/V ratio for 1-h fuels is much greater than t h e S / V
ratio for 10- and 100-h fuels, f l h will generally b e much l a r g e r than
Thus t h e 1-h fuels dominate t h e dead fuel category.
f l ~ Or
h
100h'
Live herbaceous a n d woody fuels often have similar S / V ratios, howe v e r , s o f h b and f w d may b e nearly equal. Note t h a t t h e sum of t h e
ratios in t h e live and dead categories of each case i s 1.
The fuel class weighting factors a r e then used to determine a
weighted S/V ratio for t h e dead and live categories b y summing t h e
products of t h e weighting factors for each class times t h e SIV ratio
defined for t h a t class.
u
live
- fhb*"hb
+
fwd*"wda
The weighted S/V ratios for t h e dead and live categories of t h e six
sample fuel models are:
Weighted S / V ratios b y fuel category
Case number
u
dead
'live
To complete the discussion on calculation of a single fuel particle size
o r S / V ratio t o represent the entire fuel b e d , a final set of factors is
calculated to define the proportion of t h e total fuel bed surface area
that i s contributed by each fuel category (dead and live).
For the sample fuel models, these a r e :
Fuel category weighting factors
Case number
dead
flive
Then the weighted S / V ratios for the dead and live categories a r e
combined into a "characteristictt S/V ratio for the entire fuel complex.
This i s accomplished by adding the products of the weighting factor
for each category times the weighted S/V ratio for that category:
3
=
*
fdead 'dead
+
flive*'live'
The t'characteristictt SIV ratios for the fuel model examples a r e :
Case number
Characteristic S/ V ratio
for the fuel model
The fire model assumes t h a t fuel complexes composed entirely of particles having a "characteristictt S / V ratio of 8 would b u r n the same as
the actual fuel complex being modeled, which usually contains several
different fuel size classes.
From a fuel modeling standpoint, the "characteristic" S/V ratio, 8 ,
is used numerous times in t h e fire model. In general, larger values
suggest a faster combustion r a t e , therefore f a s t e r spread r a t e ,
greater flame lengths, increased response to wind ahd slope, etc.
The "characteristic" S/V ratio i s printed in tabular output of
TSTMDL.
The most useful concept to remember from this discussion is that
the relative magnitudes of t h e individual fuel class weighting factors
greatly affect the response of a site specific fuel model to changes in
fuel moisture. These weighting factors a r e primarily affected by the
SIV ratios and loads of 1 h , live herbaceous and live woody loads, all
of which can be varied in TSTMDL.
,
d
Response of Fuel
Models to Fuel
Moisture
Live and dead fuel moistures, live and dead moistures of extinction, and quantities of fine dead and live fuels all influence t h e
response of a fuel model to fuel moisture changes.
As was described in the previous discussion on S/V weighting of
fuel size classes, just one t'characteristictt S / V ratio must r e p r e s e n t
the entire fuel complex. Similarly, a single ttcharacteristicttdead fuel
.
moisture is determined to represent the average moisture content of
the three dead fuel classes. The weighting procedure to determine a
"characteristic" dead fuel moisture utilizes the same fuel class weighting factors ( f x ) as described for the SIV weighting. Therefore the
1-h fuel moisture obviously dominates the "characteristic" dead fuel
moisture because of the large SIV ratio associated with i t .
For any fuel type, there exists a dead fuel moisture of extinction
which is the lowest average dead fuel moisture at which a fire will not
spread with a uniform front. By this definition, fuel will only burn
if the actual moisture is less than the moisture of extinction. As the
actual fuel moisture increases and approaches the moisture of extinction the fire will b u r n less vigorously. When dead fuels are dry
enough to produce sufficient heat to desiccate and ignite the live
fuels, these too contribute to the predicted fire intensity.
Fuel moistures affect both the numerator and denominator of the
spread equation. The denominator is altered by changes in the heat
of preignition (Q. ) ; higher moistures increase Pig, lower values
lg
decrease it. Fuel moistures modify the numerator by altering the
reaction intensity through a multiplier called the moisture-damping
coefficient. As the "characteristic" dead fuel moisture approaches
the dead moisture of extinction, the moisture-damping coefficient
approaches zero, thus reducing the reaction intensity. Figure 2 1
illustrates the general shape of the moisture-damping coefficient
curve. Graphs having this general shape are often produced by the
FUEL MOISTURE RATIO (M,I M,
F i g u r e 27.--Moisture damping c u r v e . Fuels typically
have an intermediate moisture range over which t h e i r
s e n s i t i v i t y to changes in fuel moisture is minimized.
-
technical version of TSTMDL when r a t e of s p r e a d o r flame length a r e
plotted for a r a n g e of 1-h fuel moistures o r loads. Increasing t h e
dead moisture of extinction will lengthen t h e "flat" c e n t e r portion of
t h e c u r v e indicating the fuel t y p e being modeled will b u r n well u n d e r
relatively high fuel moistures. The converse i s t r u e for lower dead
fuel extinction moistures.
Dynamic fuel models r e a c t v e r y differently from static models if
t h e y include a significant load of live herbaceous material. In
dynamic models, material i s t r a n s f e r r e d between t h e live herbaceous
and t h e 1-h classes a s t h e herbaceous moisture content r a n g e s
between 30 and 120 p e r c e n t . This a l t e r s not only t h e load, b u t also
t h e weighted moisture content of t h e live and dead fuel categories.
The general r e s u l t i s t h a t r a p i d changes in fire behavior predicted b y
static models for critical moisture r a n g e s a r e l e s s likely in dynamic
models.
For fuel modeling, the most important concepts r e g a r d i n g fire
behavior r e s p o n s e to fuel moisture a r e :
1. Fuel classes having t h e highest S / V ratio (1-h, live herbaceous, a n d live woody) dominate t h e fuel moisture effects.
2 . If t h e fuel t y p e being modeled b u r n s well a t a relatively high
moisture content, t h e model should have a high dead fuel moisture of
extinction. If t h e fuels do not b u r n well a t high moistures, t h e
model should have a low moisture of extinction.
3. When combustion of t h e dead fuels p r o d u c e s enough heat to
desiccate a n d ignite t h e live fuels, they too will a d d to t h e total fire
intensity; otherwise they s e r v e a s a heat s i n k .
4 . The dead moisture of extinction defines t h e "characteristic"
moisture of dead fuels a t which f i r e will not s p r e a d with a uniform
f r o n t . Increasing t h e moisture of extinction will increase predicted
f i r e behavior a t all moisture levels--for example, fuels t h a t b u r n well
a t high moisture levels should be given high values of moisture of
extinction, 30 p e r c e n t o r more.
5. The f i r e behavior response of a fuel model t o changes in fuel
moistures i s strongly affected b y t h e relative loads in t h e fuel
classes.
6 . For dynamic models, herbaceous fuel moisture changes in t h e
r a n g e of 30-120 p e r c e n t produce fuel load t r a n s f e r s between t h e 1-h
a n d t h e live herbaceous classes, t h e r e b y altering t h e moisture damping c u r v e . T h e resulting f i r e behavior may be quite different t h a n a
similar static model.
General Techniques
for Adjusting
Fuel Models
This discussion section e n d s with general guidelines on how to
adjust t h e f i r e behavior characteristics of a fuel model. I t must b e
emphasized, however, t h a t guidelines only can b e provided. I n t e r actions of t h e fuel model a n d environmental parameters with t h e f i r e
model a r e s o complex t h a t "cookbook r u l e s " cannot b e s u b s t i t u t e d for
a basic u n d e r s t a n d i n g of t h e fuel modeling p r o c e s s a n d examination of
t h e models with TSTMDL. Fuel models should f i r s t b e adjusted to
perform well a t low fuel moistures, t h e n t e s t e d a t h i g h e r fuel moist u r e s to see if t h e y respond properly t h e r e . The s t a n d a r d environmental conditions in t h e TSTMDL program provide a convenient means
to s e t u p low-, medium-, and high-moisture situations. If a fuel
model must b e adjusted to r e s p o n d properly a t high moistures, check
t h e low-moisture response again to e n s u r e t h a t it i s reasonable. All
new fuel models should be well t e s t e d a t all possible environmental
conditions for which they may b e u s e d . This will help eliminate any
undesired s u r p r i s e s in operational situations.
A common fuel-modeling problem i s having t h e s p r e a d r a t e about
r i g h t , b u t t h e flame l e n g t h too low, o r vice v e r s a . T h e technical
version of TSTMDL provides an opportunity to determine whether
changing a particular fuel model parameter h a s a g r e a t e r effect on t h e
s p r e a d r a t e o r t h e flame l e n g t h . This can b e accomplished b y plotting the ratio of s p r e a d r a t e to flame l e n g t h for a r a n g e of any fuel
model o r environmental parameter. Such a plot will show whether
s p r e a d r a t e will increase f a s t e r than flame length ( r i s i n g c u r v e ) o r
slower (descending c u r v e ) a s t h e value of t h e selected parameter
changes ( f i g . 2 2 ) . Modifying t h e fuel model parameters in t h e following o r d e r i s a reasonable way t o proceed.
1.
2.
3
4.
5.
Adjust loads.
( a ) 1-h timelag
( b ) live herbaceous
( c ) live woody
( d ) 10-h timelag
( e ) 100-h timelag
Adjust fuel bed d e p t h .
Adjust surface-area-to-volume ratios.
( a ) 1-h timelag
( b ) live herbaceous
( c ) live woody
Adjust t h e extinction moisture for dead fuels.
Adjust t h e heat content.
F i g u r e 22.- - Relative effect of 1-h timelag load o n r a t e of s p r e a d vs
flame l e n g t h f o r t h i s sample model. From A t o B t h e r a t e of s p r e a d
increases f a s t e r t h a n flame l e n g t h . From B to C flame l e n g t h
increases faster t h a n r a t e of s p r e a d .
CHANGING FUEL LOAD
Fuel loads have both direct a n d indirect effects on e v e r y variable
in t h e s p r e a d equation. Therefore because t h e load can b e changed
for any of t h e t h r e e dead and two live classes, a wide variety of
r e s p o n s e s can b e produced. Usually an increase i n fuel load will
cause reaction intensity to increase more t h a n r a t e of s p r e a d . In
fact, t h e r a t e of s p r e a d may actually decrease because more fuel must
be r a i s e d to ignition temperature. Addition of live herbaceous o r
woody fuels increases fuel model sensitivity to seasonal moisture
changes in living vegetation. The general effect of live herbaceous
fuel in dynamic models i s t o produce somewhat more i n t e n s e fire
behavior t h a n static models when t h e live herbaceous moisture i s
between 30 and 120 p e r c e n t . T r a n s f e r of "fine" fuel between t h e
herbaceous and 1-h classes accounts for t h i s .
T h e sensitivity of a fuel model to wind a n d slope can b e increased
b y reducing t h e fuel load, t h e r e b y decreasing t h e packing ratio.
Because of t h e complex effects fuel load changes can produce, it i s
s u g g e s t e d t h a t t h e technical version b e used to plot r a t e of s p r e a d
and flame length over a wide range for any fuel load class you a r e
investigating.
Increasing the 1-h load will generally increase the r a t e of spread
and flame length until the fuel model becomes too tightly packed, then
the r a t e of spread will decrease. Additional 10- o r 100-h loads will
generally decrease t h e r a t e of s p r e a d , but the flame length may
either increase or decrease.
CHANGING FUEL
BED DEPTH
Increasing fuel bed depth reduces the packing ratio, making a fuel
model more sensitive to both wind and slope. Increasing depth also
reduces the bulk density, which in t u r n reduces the heat sink
(denominator of the sprkad equation), t h u s tending to increase the
r a t e of s p r e a d . Increasing depth increases t h e reaction intensity if
the packing ratio is greater than optimum (PRIOPR in TSTMDL tabular output is greater than l ) , but decreases it if the packing ratio i s
less than optimum. Because both r a t e of spread and reaction intensity affect flame length, no absolute statements can be made about
how depth changes will affect i t .
CHANGING S/V RATIOS
In loosely packed fuels, increasing the SIV ratio of 1-h, live
herbaceous, or live woody fuels will increase the r a t e of spread and
flame length, and also increases the sensitivity of the fuel model to
wind, but not to slope. Increasing t h e SIV ratio in tightly packed
fuels, however, may decrease the spread r a t e and flame length.
CHANGING DEAD FUEL
MOISTURE OF EXTINCTION
The greater the difference between the weighted dead moisture
of the 1-, l o - , and 100-h fuels, and the dead fuel moisture of
extinction, the more intense the predicted fire behavior. Dead moist u r e of extinction not only defines the weighted moisture content for
dead fuels at which predicted fire behavior is zero, b u t also influences the fire intensity predicted a t all fuel moisture levels. Increasing dead fuel extinction moisture produces a "hotter 1 ' fuel model at all
moisture levels and increases the moisture at which the fire i s predicted to stop spreading. Changes in moisture of extinction will produce more pronounced fire behavior response at high fuel moisture,
however, than a t low fuel moisture.
CHANGING HEAT CONTENT
Heat content affects all fire behavior outputs directly; higher
heat content produces more intensive fire behavior, lower heat
content reduces i t . Because the effect of heat content i s direct and
predictable, it provides a means to "fine tune" a fuel model.
RECORDING AND USING SITE-SPECIFIC
FUEL MODELS WITH THE TI-59 CALCULATOR
After developing, refining, and testing a fire behavior fuel model
with the NEWMDL, TSTMDL, and BURN programs of the BEHAVE
system, it can be recorded on a magnetic card for use in the field
with a TI-59 calculator containing a fire behavior CROM. To obtain
the values for a fuel model and the TI-59 registers in which to e n t e r
them, use program TSTMDL to first "load" the fuel model, either from
your fuel model file, or by entering it directly. Entry of keyword
TI59 in the "CONTROL" section of TSTMDL will list t h e values to
e n t e r in the TI-59 registers. A sample listing is shown in figure 23.
Figure 2 4 provides a form on which you can record the values for
your fuel model if you a r e not using a hard-copy terminal.
TI-59 Data for Static (Dynamic) Model XX Model Name
Model parameter
1-HR
10-HR
100-HR
Live herbaceous
Live woody
Parameter value
TI Reg. No.
0.0689
0.0460
0.0115
0.0459
0.0688
1-HR
10-HR
100-HR
Live herb
Live woody
Heat content
ROS for IC
Ext moisture
Depth
M WS constant
8000.
999999.
30.
0.20
1.
To use static models in the TI-59, the live herbaceous and
live woody loads have been combined in t h e live woody
class, and the live herb load was s e t to zero. You must
also e n t e r the live herb and live woody SIV ratios a s
shown in the above listing, even though the h e r b load is
zero.
Figure 23. --Sample TS TMDL listing needed to
produce a fire behavior fuel model c a r d for
the T 1-59 calculator.
-.
-
TSTMDL TI-59 OUTPUT
Model number
File name
Wind reduction factor
for fully exposed fuels
Model parameter
Parameter value
TI Reg. No.
Live herbaceous
Live woody
---S/ V ratio--1-HR
10-HR
100-FIR
Live herbaceous
Live woody
---- Others---Heat content
Rate of spread for
ignition component
Extinction moisture
Depth
M WS constant
Figure 24.--Site-specific fuel model recording form for T I - 5 9 .
Modifying the
Keyboard Overlay
The fire behavior keyboard overlay was designed to define only
one key for e n t r y of LIVE fuel moisture. This key will continue to
be used for e n t r v of live fuel moisture for the 13 NFFL fuel models
a n d for all static fuel models. For static models, a single average
moisture is entered to represent both t h e live herbaceous and live
woody fuels. To use dynamic fire behavior fuel models, however, the
keyboard overlay must be modified to label a key for entering live
herbaceous fuel moisture. Place the label HERB above the INV key
(fig. 25). Live herbaceous moisture can b e entered by keying t h e
moisture value into the display, then pressing SBR HERB. It can be
recalled by pressing SBR 2nd HERB. Live woody fuel moisture can
be entered using the key labeled LIVE.
F i g u r e 25. --Modify the f i r e
behavior keyboard overlay b y placing the label
H E R B above the IN V key
on the calculator.
Recording a
Fuel Model
To record a site-specific fuel model on a TI-59 magnetic c a r d ,
s t a r t with your calculator OFF to e n s u r e all data r e g i s t e r s are
zeroed. Then perform the following s t e p s :
1.
T u r n the calculator ON, then p r e s s 2nd PGM 2 SBR RIS.
A - 4. will appear in the display. Successively e n t e r t h e values of the
parameters listed for your fuel model into the display and s t o r e them
in t h e indicated r e g i s t e r s . For example, to s t o r e t h e 1-h timelag
load illustrated in figure 23, e n t e r .0689 in t h e display, and p r e s s
S T 0 11. After all values for your model have been s t o r e d , put a -4
in t h e display, then p r e s s 2nd RIS and r u n a magnetic s t r i p through
the readlwrite slot in the calculator.
If your fuel model is static, t h a t i s it has no herbaceous load
( r e g i s t e r 15 is zero) t h e fuel model may b e used a s though i t were
one of the 13 NFFL models. Live fuel, when it occurs in static
models, i s stored in register 16 a s live woody material. In this
situation, s t e p 2 does not apply.
/
2 . If t h e fuel model is dynamic ( r e g i s t e r 15 i s not z e r o ) , p r e s s
RST LRN and e n t e r t h e following program:
Step
Code
000
00 1
002
003
004
005
006
007
008
009
010
011
012
013
014
015
016
017
0 18
019
020
021
76
53
43
11
42
94
43
15
42
98
91
76
54
43
94
42
11
43
98
42
15
91
Keystrokes
2nd SBR
(
RCL
11
ST0
94
RCL
15
ST0
98
RIS
2nd SBR
)
RCL
94
ST0
11
RCL
98
ST0
15
RIS
Then p r e s s LRN 1 2nd R I S , t u r n t h e magnetic s t r i p end for e n d , and
r u n i t t h r o u g h t h e readlwrite slot again. At this point you have t h e
fuel model recorded on one side of t h e card and t h e above program
on the other. Label t h e card.
Using a
Fuel Model
1. To load any previously recorded fuel model with t h e TI-59-static o r dynamic--press 2nd PGM 2 SBR RIS. A - 4 . will appear in
t h e display.
2 . Run the fuel model side of t h e card t h r o u g h t h e card r e a d e r .
3. If t h e model i s s t a t i c , t h e following s t e p s do not apply; just
use t h e model a s though it were one of t h e 13 NFFL models.
4 . If t h e model i s dynamic, p r e s s 1 and then r u n t h e program
side of t h e card through t h e card r e a d e r .
5. P r e s s RST SBR ( and ignore t h e number t h a t a p p e a r s in t h e
display.
6 . Press 2nd PGM 2 SBR RIS.
7. Enter o r change environmental inputs--including herbaceous
moisture.
8. Run t h e fire behavior program.
9 . P r e s s RST SBR )
10. To do another r u n , go back to s t e p 6 .
REFERENCES
Albini, Frank A . ; Brown, James K . Predicting slash depth for f i r e
modeling. Res. Pap. INT-206. Ogden, UT: U . S. Department of
Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1978. 2 2 p .
Albini, Frank A . Computer-based models of wildland fire behavior :
a u s e r ' s manual. Ogden, UT: U . S. Department of Agriculture,
Forest Service, Intermountain Forest and Range Experiment Station;
1976. 68 p .
Anderson, Hal E. Aids to determining fuel models for estimating f i r e
behavior. Gen. Tech. Rep. INT-122. Ogden, UT: U.S. Department
of Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1982. 2 2 p .
Andrews, Patricia L. BEHAVE: fire behavior prediction and fuel
modeling system--BURN subsystem. Gen. Tech. Rep. Ogden, UT:
U.S. Department of Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station. Review d r a f t .
Andrews, Patricia L. ; Rothermel, Richard C. C h a r t s for interpreting
wildland fire behavior characteristics. Gen. Tech. Rep. INT-131.
Ogden , UT : U S . Department of Agriculture, Forest Service,
Intermountain Forest and Range Experiment Station; 1981. 2 1 p .
.
.
Brown, James K Handbook for inventorying downed woody material.
Gen. Tech. Rep. INT-16. Ogden, UT: U.S. Department of Agricult u r e , Forest Service, Intermountain Forest and Range Experiment
Station; 1974. 2 4 p .
Brown, James K . Weight and density of crowns of Rocky Mountain
conifers. Res. Pap. INT-197. Ogden, UT: U.S. Department of
Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1978. 56 p .
Brown, James K . ; Oberheu, R. D. ; Johnston, C. M . Handbook f o r
inventorying surface fuels and biomass in t h e interior West. Gen.
Tech. Rep. INT-129. Ogden, UT : U S. Department of Agriculture,
Forest Service, Intermountain Forest and Range Experiment Station;
1982. 48 p .
.
Burgan, Robert E. Estimating live fuel moisture for t h e 1978 National
Fire-Danger Rating System. Res. Pap. INT-226. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station; 1979. 17 p .
Burgan, Robert E. Fire d a n g e r l f i r e behavior computations with the
Texas Instruments TI-59 calculator: u s e r s 1 manual. Gen. Tech.
Rep. INT-61. Ogden, UT: U . S. Department of Agriculture, Forest
Service, Intermountain Forest and Range Experiment Station; 1979.
25 p .
Byram, George M . Combustion of forest fuels. In: Davis, Kenneth P . ,
e d . Forest fire control and use. New York: McGraw-Hill; 1959:
90-123.
Cohen, Jack. FIRECAST u s e r ' s manual. Riverside, CA: U . S. Department of Agriculture, Forest Service, Pacific Southwest Forest a n d
Range Experiment Station, Forest Fire Laboratory. Review d r a f t .
Fischer, William C. Photo guide for appraising downed woody fuels in
Montana forests: interior ponderosa pine, ponderosa pine-larchDouglas-fir , larch-Douglas-fir , and interior Douglas-fir cover
t y p e s . Gen. Tech. Rep. INT-97. Ogden, UT: U . S. Department of
Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1981a. 133 p .
Fischer, William C. Photo guide for appraising downed woody fuels in
Montana forests: lodgepole pine, and Engelmann spruce- subalpine
fir cover t y p e s . Gen. Tech. Rep. INT-98. Ogden, UT: U . S.
Department of Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station; 1981b. 143 p .
Fischer, William C. Photo guide for appraising downed woody fuels in
Montana forests: g r a n d fir-larch-Douglas-fir, western hemlock,
western hemlock-western redcedar, and western redcedar cover
t y p e s . Gen. Tech. Rep. INT-96. Ogden, UT: U . S. Department of
Agriculture, Forest Service, Intermountain Forest and Range
Experiment Station; 1981c. 53 p .
Fosberg, Michael A.; Deeming, John E. Derivation of the 1- a n d
10-hour timelag fuel moisture calculations for fire-danger rating.
Res. Note RM-207. Ogden, UT: U.S. Department of Agriculture,
Forest Service, Intermountain Forest and Range Experiment Station;
1971. 8 p .
F r a n d s e n , William H . Fire s p r e a d t h r o u g h porous fuels from t h e
conservation of e n e r g y . Combustion a n d Flame. 16: 9-16; 1971.
Hough, W . A.; Albini, F. A . Predicting fire behavior in palmettogallberry fuel complexes. Res. Pap. SE-174. Asheville, NC : U . S.
Department of Agriculture, Forest Service, Southeastern Forest
Experiment Station; 1978. 44 p .
Koski, W . H . ; Fischer, W . C. Photo s e r i e s for appraising thinning
slash in n o r t h Idaho western hemlock, g r a n d f i r , a n d western
r e d c e d a r timber t y p e s . Gen. Tech. Rep. INT-46. Ogden, UT: U . S .
Department of Agriculture, Forest Service, Intermountain Forest
and Range Experiment Station; 1979. 50 p .
Maxwell, W . G . ; Ward, F. R. Photo s e r i e s for quantifying forest
r e s i d u e s i n t h e coastal Douglas-fir hemlock t y p e a n d coastal
Douglas-fir-hardwood t y p e . Gen. T e c h . Rep. PNW-51. Portland,
OR: U.S. Department of Agriculture, Forest Service, Pacific
Northwest Forest and Range Experiment Station; 1978a. 103 p .
Maxwell, W . G . ; Ward, F. R. Photo s e r i e s for quantifying forest
r e s i d u e s in the ponderosa pine a n d associated species t y p e , a n d
lodgepole pine t y p e . Gen. Tech. Rep. PNW-52. Portland, OR: U.S.
Department of Agriculture, Forest Service, Pacific Northwest Forest
a n d Range Experiment Station; 1978b. 73 p .
Maxwell, W . G . ; Ward, F . R. Photo s e r i e s for quantifying forest
r e s i d u e s in t h e S i e r r a mixed confier t y p e , Sierra t r u e f i r t y p e .
Gen. T e c h . Rep. PNW-95. Portland, OR: U.S. Department of Agric u l t u r e , Forest Service, Pacific Northwest Forest a n d Range
Experiment Station; 1979. 79 p .
Maxwell, W . G.; Ward, F. R. Photo s e r i e s for quantifying forest
r e s i d u e s in common vegetation t y p e s of t h e Pacific Northwest. Gen.
Tech. Rep. PNW-105. Portland, OR: U.S. Department of Agricult u r e , Forest Service, Pacific Northwest Forest a n d Range Experiment Station; 1980. 230 p .
McArthur, A. G . The Tasmanian b u s h f i r e s of 7th F e b r u a r y , 1967,
and associated fire behavior characteristics. I n : Technical
cooperative programme mass fire symposium proceedings. Vol. I ,
Defense S t a n d a r d s Lab. , Maribyrnong , Victoria; 1969. 23 p .
P u c k e t t , John V. ; Johnston, Cameron M . Users' guide to d e b r i s p r e diction a n d hazard appraisal. Revised. Missoula, MT : U . S . Department of Agriculture, Forest Service, Northern Forest Fire
Laboratory, Intermountain Forest a n d Range Experiment Station a n d
Fire & Aviation hlanagement, Northern Region; 1979 J a n u a r y . 37 p .
Rothermel, Richard C. A mathematical model for fire s p r e a d
predictions in wildland fuels. Res. Pap. INT-115. Ogden, UT: U.S.
Department of Agriculture, Forest Service, Intermountain Forest
a n d Range Experiment Station; 1972. 40 p .
Rothermel, Richard C . How t o predict t h e s p r e a d a n d intensity of
forest a n d r a n g e f i r e s . Gen. Tech. Rep. INT-143. O g d e n , UT:
U . S . Department of Agriculture, Forest Service, Intermountain
Forest and Range Experiment Station; 1983. 161 p .
Rothermel, Richard C. ; R i n e h a r t , George. Field p r o c e d u r e s for
verification a n d adjustment of fire behavior predictions. Gen. Tech.
Rep. INT-142. Ogden, UT : U . S . Department of Agriculture, Forest
Service, Intermountain Forest a n d Range Experiment Station; 1983.
25 p .
.
. .
Rothermel, R. C ; Philpot , C W Predicting changes in chaparral
flammability. J. For. 71 (10): 640-643; 1973.
APPENDIX A: GRASS AND SHRUB FUEL TYPES
The photos in this appendix a r e meant to illustrate the general
morpholog y for broadly different t y p e s of g r a s s e s and s h r u b s . That
i s , any set ( p a g e ) of g r a s s or s h r u b photos r e p r e s e n t s a l a r g e
variety of g r a s s or s h r u b species. One must select t h e photo that
b e s t fits t h e actual conditions at h a n d .
To help you visualize t h e general plant morphology each g r a s s and
s h r u b t y p e i s meant to r e p r e s e n t , t h e specific species photographed
a r e listed below:
Photo page
Grass Type 1
Grass Type 2
Grass Type 3
Grass Type 4
S h r u b Type 1
Shrub Type 2
S h r u b Type 3
S h r u b Type 4
S h r u b Type 5
Species photographed
Morphology r e p r e s e n t e d
Cheatgrass
Bromus tectorum
Rough fescue
Fes tuca scabrella
Fountaingrass
Pennisetum ruppeli
Sawgrass
Mariscus s p p .
Fine g r a s s e s
Huckleberry
Vaccinium s p p
Ninebark
Physocarpus s p p
Ceanothus
Ceanothus s p p
Chamise
Adenostoma s p p .
hlanzanita
Arctostaphylos s p p .
Fine stems, thin leaves
.
.
.
Medium coarse g r a s s e s
Coarse g r a s s e s
Very coarse g r a s s e s
Medium stems, thin
leaves
Medium stems, thick
leaves
Very dense, fine stems
and leaves
Thick stems and leaves
-,
GRASS TYPE 1
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
GRASS TYPE 2
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
.
\
58
GRASS TYPE 3
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
GRASS TYPE 4
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
SHRUB TYPE 1
DENSITY 1
DENS ITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
SHRU B TYPE 2
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
62
SHRUB TYPE 3
DENSITY 1
DENS IN 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
SHRUB TYPE 4
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
SHRUB TYPE 5
DENSITY 1
DENSITY 2
DENSITY 3
DENSITY 4
DENSITY 5
DENSITY 6
APPENDIX B: EXAMPLE NEWMDL SESSION
This NEWMDL session provides examples of various ways data can
be e n t e r e d when building a new fuel model. Not all possible d a t a
e n t r y combinations a r e p r e s e n t e d , b u t first-time o r occasional u s e r s
should find t h i s listing helpful.
In this session, data have been e n t e r e d for most of t h e fuel components in more t h a n one way. This i s t o illustrate several proced u r e s so you can r e f e r t o those of i n t e r e s t . I t i s not intended t h a t
you sign on t o a computer a n d duplicate this session, although t h a t
may certainly be done.
The only fuel model file p r o c e d u r e used in t h i s session i s adding a
model to t h e file. Extensive file manipulations a r e p r e s e n t e d in t h e
TSTMDL session (appendix C ) . Lines t h a t begin with a prompt
c h a r a c t e r ( > ) were t y p e d b y t h e u s e r . All o t h e r lines were p r i n t e d
b y t h e computer.
Page
I.
11.
111.
IV.
V.
VI.
VII.
VIII.
IX.
X.
XI.
....................
Entering l i t t e r data . . . . . . . . . . . . . . . . . .
A. By load a n d size class . . . . . . . . . . . . . .
B. By load only . . . . . . . . . . . . . . . . . . .
Entering g r a s s data . . . . . . . . . . . . . . . . . .
A. By load a n d d e p t h . . . . . . . . . . . . . . . .
B. By depth only . . . . . . . . . . . . . . . . . .
Entering s h r u b data b y d e p t h only . . . . . . . . . .
Entering s h r u b d a t a . . . . . . . . . . . . . . . . . .
A. Direct e n t r y of inventoried d a t a . . . . . . . . . .
B. E n t r y of total slash load . . . . . . . . . . . . .
C. E n t r y of total 10-h l o a d . . . . . . . . . . . . . .
D. E n t r y of 10-h load b y species . . . . . . . . . . .
E. E n t r y of 10-h i n t e r c e p t s p e r foot b y species . . .
Entering surface-area-to-volume ratio data . . . . . .
Entering heat content d a t a . . . . . . . . . . . . . .
Printing d a t a e n t e r e d f o r a l l f u e l c o m p o n e n t s . . . . .
Printing y o u r completed fuel model . . . . . . . . . .
Adding t h e fuel model to t h e file . . . . . . . . . . .
Building and filing a model with two sizes
of fine fuel . . . . . . . . . . . . . . . . . . . . .
S t a r t of session
67
68
68
69
70
71
72
73
74
75
76
78
80
83
85
86
87
87
88
89
-
.,
[..d 1.1el'l' fl1;;: (: 1;;: $J '1' (::I 1::' 'r I,.\:1
R A N(1;: E ::.I;)'l'
'1 fS (1
> 1";(j
1::'
:[ C;
..: ri
... I,,) 1;:... 19 1:':...rj
. ...
%(-
$;:... /.,I
I:.!1.,j y....) !..:;').
f.\
.::)I,.
I;;1".
fi S t.1 ::;I;: (:; 'T' :I: [:I $) ,
+y[:j I..!
1
:I:N I.' I.!'T Y Cl I..)R :I: I\IV I; :
.
t:'
(3i..)
:
N 'I' !::iI:<:I: 1;: D
:!) A;' A
!:j
N I...(3 fi fi F 'f
!; :[ 2 1;:
S I..! 1) 1': 1': "1'I..] '[ i\,t I::'1;: f.:"r
... ')
:::: (3
f:j .I; (1
f< A N G E:
-\.
.' L..
,
'r
J::l
LJ1-1 '1' 13 :1...I:< ("1... :1...pJ '7' !:j ;1: '1' l..j1;:
t.4 fi i',) ;:1 ti: :::: [I '1'!::I -1 (1
:> .<$ (j
.:3~,.$1$I-l I...[::I(~,:[)::; < -f,..'f?j
1 1-111
1 2 (:I
I-.
I
)
[:,
12 ;1: gl :I: :;< ...-..
rj!,.\ :1... :1.....
r, I:...{ -1' 5; I... $, !;; /..I'2
(.%
C:I... A 9 :;j
t ..1)
.~. n... 1::.' ..p .I.I../
!:; 1::1 1.: I-. .[ 1.-(1.
............ ::.,.I i\liAMli:
.....................................................".................
;(
{a i\J1) :1: 12
lAj1;:$j'1' 1;:12 j',/
I... 1:; 1-1
1;: N1;(. If::L. A NN I;; 1': I:<\ j(:'; 1:.:
c; 1. 1.I{
.,$. 1. .p. -1.. pd
. .1:.:. 1::' 1: 1.3
1.-[:I :[j [{ 1' (:I I...1;;: 13 11: PJ 1;:
:1...c,: > -1.1::
...I., t.4 hJ1-1:I: 'T 1:; P :I:t-?1;:
w 1-1 11: .I. E: GI I:< 12 11: pd 1;:;
IJE:c; -f :1 [<iq 12 :1. 1
.:) . .".:1...j.1PII:<
p f:) iq I,/;
la:[:I & p :[ jq 1:.
:[I [:I
!.J I... @I:;; :1 :I: I?
'
a
;:[
W 15: <:;'f' E:la N 1-1 1;: I...(11(:: I(
:;13 1::.... (-' .. E;: c;
[.; f.1 1::.
........
........................................... ....
.C
G ':F
W,I
ti:: S
($1::'
I...1:)
WFS
WB
kl f:
p I:'
:1
W 1-1
I",;'1: 1;: (: 11: E:: C i N pj E:
................ .... ............................"..............
[:.; 12. A N1) '1: 11: 12
1:....
: 7 :1... 1
1 A ):,j:I[ I. !
1;: h!(:; t:: I.. PI P%!pj :; p 1.4 1.1 (:; 1:;:
$; \ J
1. .'1: . . .fd. .1::. . 1:.: 11I?
I...[I1) G E I" (11 I.. :1; I:' :I: b! 1;:
W::.1 S 'f' :1: W N W 1-1 11: "1'1.: 'F :I: i'.f 1;:
!d 1-1 :I:"f'E: If.:?
(3 t.2 I.; 1:) :[ N:1:
14I::::;-rE:I?i~ R I:::1) [:; E:I[: f?l 12
F3 I:]
J'i X) E 1I (:) f5 A 1': :I:N1;:
1:)0 IJ[:.; I...A S I::. 1I?
w ti::5 T E: ta N 1-1 I:::r/i I...r:s r: K
.
.
I
I
.
.
"f'
[:I [:; 1:; :(} 1:' 1.j I:<*r1-1 1 : 1.2 j'(::I I..) \ J:s .r 1-1 (4 'J1:;: 1::' I..)1:; I... 11: k-,! f:' [:I I:?iq '1' :I: [:I i.4 11: f..)
1':; [:I \ Jrq 1: [jr$
-r!s ; !::I I:.! 1;: I..,::;: ;1 1;: PJ 1) [:; (1) t.4 : :I:1)1;: 'r 11: f:) d'i f-]1::' ${ I...el<!;1-1
I. , 'T' C)'T' A I... s I. A it; 1-1 I... 0A 1) :I:i..! "T'[:I N !i; '1: E I:? A I::14. I :
2 , I.. C] 61I} [::IF I. [I 1 1 1-4 ;1: \..II:::I...$2 fl bJ I...'(
p.j 'r p.) $;
E: v! (A (:; / i c::
3 , 1 0 1-1 I:? I... [:I A 1:) :I: f4 '1' (J iq ${ 1.J 1;: I? elC I?[;: ; Y{:. '{ ::i 1:) 1;: (:: 1: E: $;
4 , fi.)!.! i'fj1.) 1;: [:j 1:': 1 (1 l..l[q -.1..~xJ
. .-r. .1': . . .[". .1.:' 1:) 'I' :;; 13 :1 14 1::' (:] 1.'.1 'r >
'y' C; 1:) 1;: ('., ..[. 1;: :;{
!:.j,
:1; PJ 1) 5 I.. 6)$1-1
h,! 13 1-1
SJ ]:lf-1'F I:< 1.i)-7'el 1-1 &\ 1), 13:[ I-.{) I...;1:
!::I
[ J14 1:;
[:]
,
\
d
.............. ........ ....-. .........-.........................................
'71:.
.....". ........
I.
..I If if
i...
3 13 0 0
'.!;:.! () (1
................
3
I. 5ii
,.
-.[ !,Y.i (1 (1
4
,!!I 0 0
1::.
..I
20 0 0
%5fl(.t
;! 5 f j
................
(J
%y5(j
15 5 (1,
7
.r14 :1. . . .1 .1.5.:1:. .I? ..r M A i;.4 \.J A I-. :I:I:'
................ "..
,.I
r
I;;'
<:%
'3
1I1
;! 1
1:.I:
....
I. A
................................................................
2 0 [I 0
2 1:;: (1 []
................
................
2000
;! (1 !I
1!.!j (1 [:I
!.:; (1 (1
................
.... -.. ........
................
2 5 0 !I
iV Cl I?1: I! 1;: '1.fi :I: I...1; 1) ;' i
I./'
1) fi '1' R :I: 5 N1.:i iii: 1) If:l!,
3
APPENDIX C: EXAMPLE T S T M D L S E S S I O N
This TSTMDL session provides brief examples of most of the capaIt illustrates how to manipulate fuels and
bilities of this program
environmental data. obtain graphic and tabular output. use both t h e
normal and technical versions. manipulate t h e fuel model file. and
obtain a fuel data listing for the TI-59
Although the session can be duplicated a s .presented. i t i s s t r u c t u r e d for easy reference to specific activities s u c h as changing values
of fuel model parametem. doing technical version graphics. etc
Lines t h a t begin with a prompt character ( > ) were typed by the
All other lines were printed b y t h e computer
user
.
.
.
.
.
Page
I
.
I1
.
94
Fuel
96
.........
B . Listing t h e fuel model . . . . . . . . . . . . . . .
C . Getting an NFFL model . . . . . . . . . . . . . .
D . Changing the fuel model . . . . . . . . . . . . . .
E . Getting a fuel model from your file . . . . . . . .
Environmental section . . . . . . . . . . . . . . . . .
A . Getting s t a n d a r d data . . . . . . . . . . . . . . .
B . Entering new data . . . . . . . . . . . . . . . . .
C . Changing the environmental data . . . . . . . . .
D . Listing t h e environmental data . . . . . . . . . . .
Obtaining tabular output . . . . . . . . . . . . . . .
Obtaining graphic output . . . . . . . . . . . . . . .
A . Normal version graphics . . . . . . . . . . . . . .
1. Standard scaling . . . . . . . . . . . . . . .
2 . Calculated scaling . . . . . . . . . . . . . . .
B . Technical version graphics . . . . . . . . . . . .
Fuel model file manipulations . . . . . . . . . . . . .
A . Getting a model from t h e file . . . . . . . . . . .
B . Listing the models in the file . . . . . . . . . . .
C . Changing the fuel file header . . . . . . . . . . .
D . Adding a model to the file . . . . . . . . . . . . .
E . Replacing a model i n t h e file . . . . . . . . . . .
F . Deleting a model from the file . . . . . . . . . . .
Obtaining TI-59 fuel model card data . . . . . . . . .
A
I11
IV
V
.
VI
.
.
.
VII
.
....................
section . . . . . . . . . . . . . . . . . . . . . .
S t a r t of session
.
Entering fuel model data directly
96
97
97
98
99
100
100
101
102
102
103
105
106
106
108
110
114
114
115
116
117
119
120
121
.....................".................................................
I. 1-1 I:?
1 !I 1-1
:! l'i ii 1.4 F?
I..,:I:1%) t;: t.11:;: p x{
i...:I:'21;: 114 CIc:l:Ll 1''
1. [j , 2 [J
8 , :t <>
tj ,
1. , ;I. 'y
0 , I:!
a:.:;
....................................................................
i. 1-1 I:?
I... :I:1%) 1;: l..~
1:.:......
1.)
I.. I J I I ! 1 1:I'
1(7L.
:-> (:..
,
!,v.i!:I(:I ,
I3 0 f 3
,
........................................................................................
......
1j :1 1:) !..I
(11 , {3 :!
1.. 1:: 6, ....+......
.1
:1 +.!
tj 4 4 (1 ,
1;:x -1. i.f f:j :Cf:;; t.1 12 15:
23,
-r
"r
*r1-1 1;:
:1
%' (
(11 14 1) :!: 1::. [:I I:< '7' 1-1 :;1:
I< 1:;: '7' ,.)i [:I [<Ij
.' 1:......['
).'
Pi I.
j, [:I
M 1 I1 II
i"i I../ i...1::' I:<'fj
ivi 1h.I[:I [:I!1:~
I!,' :I: N:Q
5; 1 [:I p :1;
1%)
1;:;p..)
I.,)
:I: I:< [:!t.J 1:: p.1 '1. I.,.
L)
I:{ 1: A E I.. 1.:
p 1;. .I. ?.j-(' -(I..! y. . .. I{ i;:y $J (:I I!;
-1. \-\f>'
\..I f!: I..,
(1 :I: ::i.r I, J
r:
:{:)
:r [::
I.., :I:
c;
.! !I i-I I? 1:; I.!1;;: I.. PI(:I :I:S 'T 1.J 1.2 I!!:
-1, (1 (1
I? :1 1.J I;: I... PI!:] C; Lj I?E:
1.. 11: \.,' 1;;: 1.11;;: 12 1.) 1::. \..I1:; I... p\ f:] :I: ::;I-I..!:1: 1;:
I.. :I: !.j1;: bj [:I [:j El .'-( 1:; I..]1;;:I-.. P j !:]:I: !;{ -1'I...II:? 1:...:.
fi 11: 1.1
. . 1: 1... p!1;;: J,! 11: t,~1:) C; 1:) 1;:1:...,..j
1:) 1;;:I<i-.
.., 1:...
:- gj 'r c; I,. [:I 1:) 1;:
6,
:[ if.) :c; I*-,
1;;:$;
{\ I.!
;
1:; N I..) 1: 1.4 [:I i'.! PI 1: p.1 'r a 1.f)f?,'r'A
................ .. .................... .....-.................................
1. 1.1I:< :1 pj
4,
C:'
:I. (1 14g ':1 pj
,.J
1!f l1 1-4 R I::' M
Cr ,
I*,.1: I,.) : 1 1 1 1;: F{ I{ 1 v,
C.? (1 ,
I... :I:v if: w CI!:i1)Y F:' t.'i
12 ,
1.: . .I:<
. :1. . I{1;:I..![q 1%) '1.....
. ('1 I:? 12 1;:C; I,.! -1.
........................................................................................................................................................................................
1%) $1
1-.(1. .FYI:
. . :.' ( % )
1.4 1.J
d'i [:I -1'1-1 E: f<
L
1a.J
:>N
[::I $
., 5 ,
..3
Ef I...1:; ?
<
(1
(j 112
...,
1...
'3
-fa
I
1
1;.:.:
d+ i.2
c.7 (5 .(I ,
Y
a:.
r.!
I::.
.
Y
,
2; I.! ,
::>
,,
.(I i.2 I.
,
4,
c.7 (:I .(+ ,
.1 ..fa /-. ,
t
............
". ..
I
.q ,
.<I2 j. 2 ,
....I..,
S-1
?. !I
r
............
13
A
...I
<./
%
(.
............
.I. ,! i'-q I;
1 ( 1: )
-1. ;:1 ( 1.) .r !J
, (;It:..r , PI
1-1 >', (!q ( 1.) .I*1.1 /'<;;q: :1 y j
i::.1. . . . .(.1.2. . .1 .1 /./ 1::. .r . ,.I(2 1::...
' [:; j
f:'
.. ...
'r
:I::[:I 1:.I...g, pl E: l!j :I: p.! 1:)
!:I ,
....................................................................
8
Cj;
1:..[ 1;:
:I: xi I...: :1
..Y
..;I
,
[:I :::, ,
:1
"I'
/*
!
!
2
.-.I
a:. .
0
i'l
:I:
N
,
(1 (j
.>:.
.I..
.;<.
.{..
.;..
#
.y;
.x.
.+.
.+
.+. (:i 3
I..>
!
.y;
.'A
.(..
.+.
.4.
.+.
: X J2
..
>
..;
.+.
.J..
.+
!
X
x X X
> i % X
)'(':>;
X ........ ?/ .............'L''\ ........ 'L'.'\ ........ :jf;............ ..&........ ):;.('
.....................................+..................................... .+ .....................................+........................................i
.I.....
;:::
, (:I !I
3 , !!
[:, , fl
<;;I , ij
.I. 2
:!
I*\ 11: 1) :1 I.
1;: lAl 11: t..J !:i
1:) 1;: :1......
'('j
( tq 11: 1.1
:
<',
II...
A
pi
1::
.J. (1
!
!
j
,)
.;:
. . . ..,.
.
..,
..I.
II
: , (j
N
!:::
:
, (1
1-1
-9.
!
!
.+.
!
.x.
....
.;q
.x
.... 2 ,[j
.+
.+.
1::.
) ' X
!
'I'
.
.+.
.+.
.
.+.
,/.,
;.'
.+
.{..
,,
.+.
X
:>;
,K.
X
.x
.{.,
.b.
.x
.{..
.+.
.
.A.
.+.
7%.
'
.
,..'. . . . . 'x.
.+.
.I..
>;
..,
.;<
....
X
:< .!. :.
ii
.A.
)<
Y:;
X
. ...........................................................................................................................................
...................................................4......................................
i.) it,
:
X
.+.
X
.x.
.>;.
>..;
.p;.
1
.+. [i!::l
.. I...'
.Y.
!
!
I...
1:..
!
..<.
.
N
.+............. [:I 13 ........ .....+............. ............ ....
, !] {j
8
If (1
...!!! !I
I:r
,Lj [f
................................ .....+..... .... ........ ............ .........+ .... .... .... .... .... ........ .... .... .{.. .....................................I.,
:I, !.:; (:I
..:I :: f..
...,
;.:.! [:I [j
i... ..I .! I:{
.J ff ff t i
\.j p.1 11: .y ,$ 1.4 1;:; & ( 13
. 1.1 ,,.I c; [;I 1::- -i)
I (1 I:! i j
I../1;: 6,-1. :1 1.4
1s:
-fa
:I"(j j'
.-> ...{
W fi PJ'I'
!::I I.!
(;:
::1 A' : 1
].I
,... .... .:1 I...6,iq' :;1
'::I
I., 1;:
p.j
;:c
'-f.]-l'>
< 'y
(::I 112
t,.! 1 ,
:C [::I Y (::II.j !,(.I~2~p~-[. '1' [:! (,j :; :;1 :
:\ j $j'1' ci p.) :[) :1 1j
?.I.; } (1; ,q.I (1; I..!.I pi 1;: 1:)
!;; [:I . I i'.J[:.; ['I VY "f' I.! :;1
....f?i 7: '1. Cj ' j
%
'{
1;: N "T 1;: 12 1
\<
.-j
i
i...
fJ
I:< ;.?
'{r) I .I p .I. 1-1 {I: -:1 jq :;;
1:) (1) 'y [::I I..! !.,I PI p!, '1' '1.(1) I ~ JI..
.\
y
1: 1 :; -f
I:< ft p.) G:;:1
t.)
::::
GI I...I.,! 1;;
;i
,
.t. ,
[:)
'r
[:I I..!
!...I
:;1 !ii ::1 (:I i!?
(:) 1:j 1;:
\:
1-1 12.
j+\ {"I
... '1... $yi
.- '.( \ ..1 112 1;:' ..!..:; '<
:1
( '..(
f
(11 ,
...><:)
'...'
.-'':1...["(1
.. PJ.1....1
$I
..i
A fl f:.; 1;:
> .J. ;.:!
,(-..I (:I
{j
:(
(1
::::
I.,. (.) 1;;: ?'
2 , '1' (11
{!, ,.I
.T 1(;
(11
,
1:) 1-1 :;
,
1
1
')
'f' (:]
f,] j
....
':1 1 : I:?1::...'
("1... !..I pi
1:i el[:; '1' 1::'
:I:K; '1' :I:[:; ;; (1; I..! 1:; '1. f?
< '.f
(::I I:<
,J
:I ,
i
N
,[ ]
.+ .... -.. .............................+ ..................................... I" .....................................+......................................{.. .................................... .+. .................................... ,A.
5.8
:! %
i. 5
:. 111
yeI::
L.. *I
:.3 (:I
I 1-1I?!.:I 1:;:I," I...[:I 1) ( [::I?.J :;;.,f' fil 1: I:<1;:j
ii1;: 'l' !.! 11,'p.) 7' [:I I:;(1) i'.! "l'1: p.) 1.1 1;:
I. , 0
1:' I:,)1:;
::;
j 1
' I I.. I . 2 , .
f
1:': N'1' 1;: R x > "i' j C! I?
'
j j'
(;
1
4
(;I !.I :I: 'I'
:; 1;:
1 1.. 1;
:
,
1: ;
;
1 : Y iiy..... .t-.:, [:I I:?
!.! :I:,I' ?'
.%i.
:v.. .v. .v.
I.. C,
1..
,..
. . .Y.
. . ..1..
.'<.
i..
2...
-
<: i$:{:l:s
1I. . '.!(1
~<~~~:"~(l~~li::i~:<:(j'~
/o
........ .... ................ .... .... .- ........ ......-................. .... .... .... .....................^.-. ....,... ........
1) 5: ;1
[:; [:I PJ 1) i::
[:I 1:) 1
' ' [j 1::'
:[ :!<
(:] I j 1;:I...
ij
0 I.! I? 1)I:? (1 1:) !::I::$1;:r) ;1: \.)I:;:I... pi [::I 1) 1;:I... 1) 'f P-J j!?!
11: [:; < :! ;I
[::If? $;
'r
C, ..
'[ i.':
... . L... .'j '2
1j 'f iq &. :I:;[ pi (J 1) 1;: I...j!
1) J { J !;; 'r 1: (:] :;1;:Ci ?;;[::I P.j el I... 1) f.! )' :I: p.)(:.; '1' 1-1 I<
[:] I..![:.; 1.1 'r I:<GI p,! c; 1.: 1:: 14
[:I 1:: 1.. '[.. &IE: 1-1 :I K I{fill:: 1;:[:I
I..) $; I...[:jCl 1) E{1 : '1' 1.4:;1 :1: $,! I... :I ),1 1;: 1-11;: I:? :fj $1 (:; 1;;:(111.J :;
pJ 1) 1. i-l(:)I..!12
-7' 1: Pi 1 : I...{a lie; l-.
I...(1 {: :;; 1;;: !!i>
[::I:1
1 1 1;;: 14. [> 1.; 1;:[:j I..! ::; 1: 1.1 1;:I... P j [] :I: 5; .r 1J. 1:t 1;: ,
$; 6) :1 IJPJ [: .i 11: KI
j's4
:[e{l
d
D (3 Y [::I1.1 W rlN 'I' A I. I !; '1'
> iq
(I1
F' 1:;
I.. Cl A %> ( 'T' ./ A C :I
....-..
.................................."".. ........
1 1.4 I?
1,02
I 0 1-1 Fl
4,UO
1 (1 0 1-1 11
.*
20
I. . :t ?I! 1;: 1-1 1:;: ri :[3
0 , !I 0
.
.
.
I
.
.
.
.
.
.
.
.
.
.
I
I.;,
1
.I J 1;;:.1
i:) fi E: 1.
I< 1 : )'W [:I
12 X) !I; ?
{
.I
i.*
[:) R
N)
APPENDIX D: FUEL MODEL FILE STRUCTURE
The fuel model file s e r v e s a s the basic means of communication
between t h e programs of t h e BEHAVE system. The s t r u c t u r e of t h e
file is:
1. A "header1' record containing the u s e r ' s password and a general
description of the models in the file.
2 . One record for each fuel model i n t h e file.
3. An end of file mark,
If fuel models have been deleted from a file, you may find some
extraneous records after t h e f i r s t end of file mark. They should not
be a cause for concern. With some computers you may see these
records if you look at t h e file with t h e editor. Other computers may
delete them.
The records of the file a r e described in detail below.
" Header"
Column ( s )
Record
Data recorded
User's password
Blank
File description
Blank
A letter of t h e alphabet
The letter in column 80 will be used to check whether or not the
fuel model has the c u r r e n t format. When BEHAVE i s implemented this
l e t t e r will be A . If the format changes in t h e f u t u r e , the letter will
be changed to B , then C , etc.
Fuel Model Card Records
Column ( s )
Data recorded
1 3
4 36 40 44 48 52 56 60 65 67 71 -
2
35
39
43
47
51
55
59
64
66
70
74
75 - 78
79
80
Fuel model number
Wind reduction factor
Fuel model name
1-h load
10-h load
100-h load
Live herbaceous load
Live woody load
Fuel bed depth
Heat content
Extinction moisture
1-h SIV ratio
Live herbaceous SIV
ratio
Live woody S I V ratio
Letter
Dynamic ( I ) ,
static ( 0 ) code
The formats used to write and read these records a r e :
'!Header1! record:
Write format (A4,2X, l8A4, l X , A l ) )
Read format (A4,2X,18A4, l X , A l ) )
Fuel model records:
Write format (I2 , I l , 3 2 A l , 6I4,I5,I2,3I4, A 1 , I l )
Read format (F2.0,11,32Al,6F4.2 ,F5.O,F2.0,3F4.0 ,A1 , I 1 )
APPENDIX E: WEIGHTING PROCEDURES USED IN PROGRAM NEWMDL
Field data a r e usually collected for more than one of t h e fuel cornponents--litter, g r a s s , s h r u b s , o r slash. The data collected for each
component will differ. For example, t h e 1-h S / V ratio for litter will
not likely be the same as for s h r u b s or g r a s s . And t h e heat content
may b e different for slash than for t h e live leaves and twigs of
s h r u b s . Therefore, while t h e NEWMDL program will accept t h e diversity of data collected on the various fuel components, i t must eventually b e condensed to "average" values t h a t r e p r e s e n t t h e entire fuel
complex. This appendix describes t h e weighting procedures used to
calculate average heat content, 1-h S / V ratio, dead fuel extinction
moisture, and fuel bed depth for t h e " f i r s t cut" fuel model produced
by t h e NEWMDL program.
Heat C o n t e n t
1.
Calculate the mean total surface area of fuel in t h e j t h
the
( 0 ) ,j
dead category:
class of
(W0) l j
A,, =
and the
(o),j(w0)2j
live category:
A
2j
=
(pp) 2j
where
ratio of t h e jth
= surface-area-to-volume
o
class of
the dead fuel category
= ovendry load in t h e j t h
Wo
class of t h e dead fuel
category
p
2.
P
= particle density (32 l b / f t 3 )
Calculate t h e mean total surface area of t h e
dead category:
A
lj
=
C A
j=l l j
and t h e
2
live category:
A
2j
=
C A
j=l
2j
and t h e mean total surface area of t h e complex
3.
Determine t h e fraction of t h e total surface a r e a in t h e
dead category:
f,
=L
live herbaceous class :
-
A
291
f2,1 - -
live woody class:
4.
f2,2 -
A
2,2
-
Az
Calculate the weighted heat content for all fuel classes
and categories
-
Hw - f l H 1 , l
+
f2,1H2,1
+
f2,2r-12,2
where
H
1,1
1
H
292
= dead fuel heat content ( B t u l l b )
= live herbaceous heat content ( B t u l l b )
= live woody heat content ( B t u i l b )
One-Hour Timelag Surface-to-Volume Ratio
1.
Calculate weighting factors for each component
where
Wi
ai
2,
= ovendry load of each component
= 1-h S / V ratio of each component
Calculate the "characteristic" 1-hour S/V ratio for the
fuel complex
Dead Fuel Extinction Moisture and Fuel Bed Depth
1.
Convert total load of each component from tons p e r acre to
pounds p e r s q u a r e foot
2.
Calculate the packing ratio for l i t t e r , g r a s s , and slash
components a s
where
= component packing ratio
B c ~
w = component load ( l b / f t 2 )
CP
6
= component depth
cP
Calculate t h e extinction moisture ( % ) for l i t t e r , g r a s s , and slash
components
where
hl
cP
= component extinction moisture
Component extinction moisture (Al
) estimates are based on t h e
XCP
relationship of extinction moisture to packing ratio for the 13
NFFL fuel models (fig. 26). These models can b e separated into
two groups:
- s h r u b s and tall coarse g r a s s (models 3-7)
- s h o r t e r , finer grasses (models 1 and 2 ) a n d fuels t h a t
a r e primarily horizontal (models 8-13)
The two groups were considered separately. The extinction
moisture of t h e f i r s t group is s e t , in subroutine SHRUB, as 0.35
if t h e leaves a r e said to contain oils a n d waxes, 0 . 2 0 if not.
The extinction moisture of t h e second group i s calculated using
t h e regression line fitted to t h e points plotted for models 1 - 2 ,
a n d 8-13.
Calculate extinction moisture for t h e fuel model
where
w
0
= total ovendry load
Depth for t h e fuel complex is similarly calculated
0.01
0.02
PACKING RATIO
126
0.03
0.04
F i g u r e 26. --Moisture of
e x t i n c t i o n is assigned f o r
s h r u b - t y p e fuels (models
3-71, b u t calculated from
t h e ex t i n c t i o n moisture
equation for o t h e r fuel
types (models 1-2 and
8-13).
* U. S. GOVERNMENT PRINTINGOFFICE:1986776-03211067 REGION NO. 8
B u r g a n , R o b e r t E.; Rothermel, R i c h a r d C. B E H A V E : f i r e
b e h a v i o r p r e d i c t i o n a n d f u e l modeling system--FUEL
s u b s y s t e m . General T e c h n i c a l R e p o r t I NT-167. O g d e n ,
U T : U. S. D e p a r t m e n t o f A g r i c u l t u r e , F o r e s t S e r v i c e ,
I n t e r m o u n t a i n F o r e s t a n d Range E x p e r i m e n t S t a t i o n ;
1984. 126 p.
T h i s manual d o c u m e n t s t h e f u e l modeling p r o c e d u r e s o f
BEHAVE--a s t a t e - o f - t h e - a r t w i l d l a n d f i r e b e h a v i o r p r e d i c t i o n system. D e s c r i b e d a r e p r o c e d u r e s f o r c o l l e c t i n g
f u e l data, u s i n g t h e d a t a w i t h t h e p r o g r a m , a n d t e s t i n g
a n d a d j u s t i n g t h e f u e l model.
KEYWORDS: f i r e , f u e l s , f i r e b e h a v i o r p r e d i c t i o n
The lntermountain Station, headquartered in Ogden, Utah, is one
of eight regional experiment stations charged with providing scientific knowledge to help resource managers meet human needs and
protect forest and range ecosystems.
The lntermountain 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 consumption. They also provide recreation opportunities for millions
of visitors each year.
Field programs and research work units of the Station are maintained in:
Boise, ldaho
Bozeman, 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, ldaho (in cooperation with the University of
Idaho)
Provo, Utah (in cooperation with Brigham Young University)
Reno, Nevada (in cooperation with the University of
Nevada)