Modeling of AU Grassland Fires

Modeling Australian Grassland Fires

The purpose of this web page is to facilitate the development and validation of fire models for vegetative fuels, in this case the surface fuel of grasslands. A number of grassland experiments were conducted in Australia in the 1980s. Both data and findings regarding fire behavior gathered during these experiments will be use for model validation.

Contents
Experiment F19

physical conditions / procedures
experimental data

Experiment C064

physical conditions / procedures
experimental data

EXPERIMENT F19


Physical and thermal characteristics of fuel and atmosphere in experiment F19 properties are from Australian grassland experiments unless otherwise noted
fuel property	value
moisture, %	5.8
fuel height, m	0.51
fuel loading, kg/m²	0.313
bulk density, kg/m³	0.614
surface-to-volume, m^-1	12240
fuel element density, kg/m³	512^A
packing ratio	0.0012
char yield, %	20^B
volatile heat of combustion, MJ/kg	15.6^C
specific heat c_p, kJ kg^-1 K^-1	0.01 + 0.0037 T^D
initial temperature, C	34

atmosphere characteristics
ambient wind speed @ 2 m height ~200 m upwind of fire plot, m s^-1	3.2 - 4.0 (see figure below)
averaged plot-corner wind speed @ 2 m height	4.8 (see figure below)
averaged mid-flame wind speed	4.5
temperature profile, C	34 (constant)
relative humidity, %	20

Experimental procedures and other info Initiation of line fire: The experimental line fire is initiated by two men with drip torches starting at the center and moving away from each other (perpendicular to the prevailing wind) for 56 s at which time they were separated by 175 m. The fire spread downwind in a 200 m x 200 m grassland plot bordered by firebreaks. The head fire reaches the downwind fire break at 138 s.
Notes: A) from Rothermel ... B) based on average of cheat grass and Idaho fescue obtained from Susott, R.A., "Characterization of the thermal properties of forest fuels by combustible gas analysis," Forest Sci. 28: 404-420, 1982 C) per kilogram of gaseous fuel average of .... as in B above D) W.J. Parker, "Prediction of the heat release rate of Douglas fir," Fire Safety Science -- Proceedings of the Second International Symposium, pp. 337-346, 1989

Experimental Data for Fire19 (supplied by Phil Cheney)
photos
fire_1_56s.jpg	fire_2_86s.jpg	fire_3_121s.jpg	fire_4_138s.jpg	fire_5_nas.jpg time not known
Rectified (to plan view) fire perimeters based on the above fire 1,2,4 photos
The distance between the vertical red line on the left-hand-side and the right-hand-side of the figure is 212.8 m. The scaling is consistent throughout the figure. The vertical red line on the left-hand-side is the beginning of the grassfuel (i.e. the road on the photos above that bordered ignition line is not included - all the other roads are included). The redlines are the position of the flame front which have been interpreted to allow for flames leaning over the fuel at times. The black lines are the interpretation of the flame depth which is defined as the distance of continuous flaming behind the flame front. Data files (ascii) containing the position of the flame front (red lines on figure at left): time = 56 s : fire19_56s_perimeter.txt time = 86 s : fire19_86s_perimeter.txt time = 138 s : fire19_138s_perimeter.txt
Wind time histories
corner wind plots, 2m above the ground, 5 s between measurements This data was taken at the corners of the 200 m x 200 m grass plot and can include the fire induced winds. Note, in the photos and rectified data plot North point downward. So NE is the bottom left-hand-side corner, etc.
Clarke mast measurements 200 m upwind of fire plot; 2 m and 10 m above the ground, 1 minute between measurements Note, these wind measurements are less likely to be affected by the fire induced wind and may therefore more accurately represent the ambient wind speeds.

EXPERIMENT C064


Physical and thermal characteristics of fuel and atmosphere in experiment C064 properties are from Australian grassland experiments unless otherwise noted
fuel property	value
moisture, %	6.3
fuel height, m	0.21
fuel loading, kg/m²	0.283
bulk density, kg/m³	1.35
surface-to-volume, m^-1	9770
fuel element density, kg/m³	512^A
packing ratio	0.0026
char yield, %	20^B
volatile heat of combustion, MJ/kg	15.6^C
specific heat c_p, kJ kg^-1 K^-1	0.01 + 0.0037 T^D
initial temperature, C	32

atmosphere characteristics
ambient wind speed @ 2 m height ~200 m upwind of fire plot, m s^-1	3.75 - 4.0
averaged plot-corner wind speed @ 2 m height	4.6
averaged mid-flame wind speed	2.8
temperature profile, C	32 (constant)
relative humidity, %	35

Experimental procedures and other info Initiation of line fire: Procedure was similar to experiment F19 but less well documented. The experimental line fire is initiated by two men with drip torches starting at the center and moving away from each other (perpendicular to the prevailing wind) for 25 m for an igntion line of 50 m long. In the WFDS simulations the ignition procedure is assumed to take 25 s. The fire spreads downwind in a 108 m x 104 m grassland plot bordered by firebreaks. The head fire reaches the downwind fire break at about 100 s.
Notes: A) from Rothermel ... B) based average of cheat grass and Idaho fescue obtained from Susott, R.A., "Characterization of the thermal properties of forest fuels by combustible gas analysis," Forest Sci. 28: 404-420, 1982 C) per kilogram of gaseous fuel average of .... as in B above D) W.J. Parker, "Prediction of the heat release rate of Douglas fir," Fire Safety Science -- Proceedings of the Second International Symposium, pp. 337-346, 1989

Experimental Data for Fire C064 (supplied by Jim Gould)
photos
27 s	53 s	85 s
Rectified (to plan view) fire perimeters based on the above fire photos
The redlines are the position of the flame front from the above photos which have been interpreted to allow for flames leaning over the fuel at times. Note there is no photo for the last perimeter, at t = 100 s. Data files (ascii) containing the position of the flame front (red lines on figure at left): time = 27 s : firec064_27s_perimeter.data time = 53 s : firec064_53s_perimeter.data time = 85 s : firec064_85s_perimeter.data time = 100 s: firec064_100s_perimeter.data
Wind time histories
corner wind plots, 2m above the ground, 5 s between measurements
Clarke mast measurements 200 m upwind of fire plot; 2 m and 10 m above the ground, 1 minute between measurements

contact: ruddy@nist.gov; 301-975-4797 ; 310-975-4052 (fax)

Funded by: Building and Fire Research Lab of the National Institute of Standards and Technology US Forest Service North Central Research Station
Research Group: William (Ruddy) Mell, NIST, USA Ron Rehm, NIST Mary Ann Jenkins, York University, Canada Ruiyu Sun, University of Utah Steve Krueger, University of Utah, USA Mike Zulauf, University of Utah		Collaborators: Jay Charney, USDA Forest Service Phil Cheney, CSIRO, Australia Jim Gould, CSIRO, Australia