III III International International Confer. Confer. on on Forest Forest Fire Fire Research Research
14Conference on on Fire Fire and and Forcat Forcat MeteorololY MeteorololY 14- Conference
VOL 1998 VOL I. I. pp. pp. 83 83 -112.Lwo.16120 -112.Lwo.16120 November November 1998
EXAMPLE AN EXAMPLE
OF MULTI-FACETED WILDLAND WILDLAND FIRE OF MULTI-FACETED
FIRE RESEARCH: THE THE INTERNATIONAL INTERNATIONAL CROWN RESEARCH: CROWN FIRE MODELLING EXPERIMENT EXPERIMENT MODELLING
ALEXANDERCI), BJ. STOCKS(2), WOTTON ) and RA. LANOV1LLE LANOV1LLE(3)) M.E. ALEXANDERCI), STOCKS(2), BM. WOTTON2) 2
(3
(I)Canadian Forest Service, Northern Forestry Centre, 5320-122 Street, Edmonton. Alberta,
Canada, T6H T6H 3S5; E-mail:
[email protected] [email protected] Canada, (2 (2)Canadian Forest Service, Service, Great Great Lakes Lakes Forestry Forestry Centre, Centre, P.O. Box Box 490. 490. Sault Ste. Marie, Marie, Ontario, Ontario, )Canadian Forest
P6A 5M7; E-mails:
[email protected] E-mails:
[email protected]@nrcan.gc.ca Canada P6A Canada and
[email protected] 3loepartment of ((3loepartment of Resources, Resources, Wildlife Wildlife and and Economic Economic Development, Development, Forest Forest Management Management Division, Division,
Box 7, Fort Fort Smith. Smith. Northwest Northwest Territories, Territories, XOE XOE OPO; OPO; P.O. Box E-mail:
[email protected] [email protected] E-mail:
SUMMARY SUMMARY
The International Crown Fire Modelling Experiment (ICFME) constitutes a major, Fire cooperative, global undertaking involving coordination by the Canadian Forest Service Fire Research Network and the Government of of the Northwest Territories' Forest Management Research of collaborating scientists and operational operational fire fire Division combined with participation of personnel. principally from Canada and the USA, but with representation from several other countries as well. The initial impetus for the ICFME was oriented towards the testing and of a newly developed physical model for predicting the spread rate and flame front calibration of of crown fires. fires. However, the ICFME has also provided the opportunity to examine intensity of other aspects or implications of crown fire behaviour, without comprising this primary objective, including linkages to fire-fighter safety/personal protective equipment (PPE) and wildland-urban interface or intermix issues as well as certain ecological and environmental impacts or effects, including concerns about atmospheric chemistry from biomass burning. are providing The five experimental crown fires that have taken place in the last two years are valuable new data and insights into the nature and characteristics of crowning forest fires 83
and opportunities that will be be affecting needed for dealing with the fire management problems and both people and ecosystems in the coming century.
INTRODUCTION
is a major field The International Crown Fire Modelling Experiment (lCFME) is
campaign being carried out in part under the auspices of the International Boreal Forest Research Association's (IBFRA) Stand Replacement Fire Working Group (SRFWG). The SRFWG was, as the first working group of the IBFRA, IBFRA, established in 1992 to to foster cooperative research with the general goal of better understanding global climate change apd ecosystem function with respect to the role of fires of lethal intensity in northern circumpolar
boreal forests (Fosberg 1992; Goldammer and Furyaev 1996, 1996, pp. 516-517). The ICFME also represents a core research activity of the International Geosphere GeosphereBiosphere Programme's (lGBP) International Global Atmospheric Chemistry (IGAC) project
dealing with the Biomass Burning Experiment (BIBEX) (BIBEX) (Goldammer 1994; Goldammer and Furyaev 1996, pp. 1-20 and 518-524). The first burning associated with the mFRA-SRFWG's experimental program of fire behavior was the Bor Forest Island Fife Experiment which took "! place in north-central Siberia in July 1993 and involved a 49-ha stand of mature Scot:. -'-' ::.... "!
(Pinus sylvestris) sylvestris) as documented and reported on by the Frre Research Campaign Asia-North (Pinus (FIRESCAN) Science Team (1994, 1996).
A third third project initiated in 1995, referred to as the FROSTFIRE involves the burning of 809 ha of boreal conifer and hardwood forests in the Caribou-Poker Creeks Research Watershed
located
40 40
km km
north
of
Fairbanks,
Alaska,
USA
(see (see
http://www.fsl.orstedulhomelusfslgepp/alaskalfrstfire.htm): aim of all all three three of http:// www .fsl.orstedulhomelusfslgepp/alaskalfrstfire.htm): The broad aim these experimental burning projects is to investigate the behavior, ecological and atmospheric chemical effects of high-intensity. stand replacement or stand killing fifes in boreal forest ecosystems.
The purpose of this paper is to provide a broad overview of the ICFME and to current status thereby updating previously published accounts (Stocks and summarize its summarize its current Alexander 1996; Alexander et al. 1998) of this global wildland fire research effort for broader
As a matter of possible interest, .. distribution via inclusion in these conference proceedings. As
84 84
several 1997; Andersson Andersson several popular popular articles articles dealing dealing with with the the ICFME ICFME have have also also appeared appeared (Shilts (Shilts 1997;
1998, Anon. 1998; Holloran 1998; Klasseen 1998; SamoilI998). 1998, Anon. 1998; Holloran 1998; Klasseen 1998; Samoil I 998).
BACKGROUND INFORMATION INFORMATION
Wildland fIre fIre research research scientists scientists in in Canada Canada and and the the United United States States have have been been working working for for Wildland many years years on on the the development development of of fIre fIre danger and fIre fIre behavior behavior prediction prediction systems, systems, many danger rating rating and which are are now now in in widespread widespread use use across across North America and and overseas overseas (Andrews (Andrews 1991; 1991; which North America Alexander 1996). These daily by et aI. aI. 1996). These systems systems are are used used daily by fIre fIre management management agencies agencies in in these these Alexander et
fIre control and fIre fIre use use countries, forming forming the the basis basis for for decision decision support support systems systems pertaining pertaining to to fIre countries, control and activities. be activities. Although Although much much progress progress has has been been made made to to date, date, much much more more remains remains to to be accomplished, accomplished, including including the the formulation formulation of of aa predictive predictive physical-based physical-based model model of of wildland wildland fIre fIre behavior behavior that that can can encompass encompass the the full full range range of of the the types types and and intensities intensities of of free-burning free-burning fIres fIres
encountered in in nature nature for for any any combination combination of of fuel, fuel, weather weather and and topography. topography. encountered Despite the the fact fact that that high-intensity crown fIres fIres currently currently account account for for an an overwhelming overwhelming Despite high-intensity crown proportion the subsequent proportion of of the the annual annual area area burned burned in in Canada, Canada, aa full full understanding understanding and and the subsequent
to model model the the initiation, initiation, propagation propagation and and spread spread of of crowning crowning forest forest fIres fIres remains an ability to ability remains an elusive goal goal for elusive for fIre fIre research research scientists scientists in in this this country country and and throughout throughout the the world. world.
The The
Canadian Canadian Forest Forest Service Service (CFS) (CFS) approach approach to to the the prediction prediction of of crown crown fIre fIre phenomena phenomena has has been been largely 1990). An largely empirical empirical in in nature nature (McAlpine (McAlpine et et al. al. 1990). An extensive extensive experiment experiment burning burning program program
fIres to to high-intensity high-intensity carried out out in in several several major major fuel fuel types, types, involving involving both both gentle gentle surface surface fIres carried crown crown fues fues (Alexander (Alexander and and Quintilio Quintilio 1990), 1990), coupled coupled with with observations observations obtained obtained from from monitoring monitoring, selected selected wildfIres wildfIres (e.g., (e.g., Alexander Alexander and and Lanoville Lanoville 1987), 1987), created created aa large large data data base base , that Canada that is is the the foundation foundation for for the the system system of of quantitatively quantitatively predicting predicting fire fire behavior behavior used used in in Canada
today (Forestry (Forestry Canada Canada FIre FIre Danger Danger Group Group 1992; 1992; Taylor et aI. aI. 1997). 1997). Conversely, Conversely, the the today Taylor et of systems systems for for rating rating development development of
fIre danger and predicting predicting fIre fIre behavior behavior in in the the United United fIre danger and
. .
: : . .
States States has has been been based based largely largely on on aa semi-theoretical, semi-theoretical, laboratory-based laboratory-based fIre fIre spread spread model, model, which which
has has proven proven limited limited in in the the prediction prediction of of crown crown fIre fIre behavior, behavior, resulting resulting in in the the necessity necessity to to resort resort to to
empiricalism empiricalism as as well, well, at at least least for for the the interim interim (Rothermel (Rothermel 1991). same time, time, CFS CFS fIre 1991). At At the the same
behavior behavior researchers researchers have have come come to to realize that it it would would be be virtually virtually impossible realize that impossible to carry carry out a series range of of experimental experimental fIres fIres over over aa range of burning burning conditions in all all the the important fuel types types series of conditions in important fuel found found in in Canada Canada for for aa number number reasons reasons (e.g., (e.g., time, time, expense, expense, logistics, logistics, shortage shortage of of personnel). personnel). 85
Within this this context, CFS CFS fIre fIre behavior behavior researchers researchers began began cooperating with with Dr.Frank Dr.Frank A. A. Within of the the Department Department of Mechanical Mechanical and and Industrial Industrial Engineering, at at Montana Montana State Albini of Albini University in in Bozeman, Bozeman, Montana, Montana, aa leading leading wildland wildland fIre fIre modeUer, modeUer, on the the extension extension of aa University some-what limited limited crown fire fire spread model model developed and and tested tested in in the the mid-80s mid-80s (Albini (Albini and some-what Stocks 1986). 1986). This This process, process, initiated initiated in in late late 1992, 1992, was was to to involve involve aa large large amount amount of of theoretical theoretical experimental fIres fIres carried carried out in in aa fIeld fIeld setting for model development development coupled with with specifIc experimental model testing and and validation validation purposes. purposes. testing
This initiative initiative was was financially financially as as well well as as intellectuall intellectually This y
supported by by the the CFS CFS using using Green Green Plan Plan funds funds (Anon.l (Anon.l994) and by by 994) and
the USDA USDA Forest Forest the
Service's Intermountain Intermountain FIre FIre Sciences Sciences Laboratory Laboratory at at Missoula, Missoula, Montana. Montana. The The final report report for Service's
or.
Albini's research on aa "physical modell modell of crown crown fire fire behavior" behavior" was was completed completed in May or. Albini's the research effort effort have have also also been been separately reported reported on (Albini (Albini 1997. Component Component parts parts of the 1997. 1996; 1996; Call Call 1997; 1997; Call Call and and Albini Albini 1997) 1997)
STUDY AREA DESCRIPTION The ICFME study area, located about 40 Ian Ian northeast of the town of Fort Providence in
(NWT) (61.6 ON ON latitude, 117.20 W longitude), was selected Canada's Northwest Territories (NWT)
following a reconnaissance of potential candidate sites by CPS and Government of NWT of Resources. Resources. Wildlife Wildlife and and Economic Development Development (DRWED) (DRWED) personnell personnell in June June Department of Department the study study area area is is flat flat (elevation: (elevation: 159 159 m m above above MSL). MSL). A A preliminary preliminary fuel fuel 1994. The The terrain terrain in 1994. in the survey of the the area area was was carried out in September 1994. 1994. The The fuel fuel complex consists of aa 65survey year-old jack jack pine pine stand stand (Pinus year-old
banksiana) with with the the overstory canopy averaging averaging 12 12 m m in in banksiana)
height and and 4100 stemslha with with an an understory understory (4600 (4600 stemslha) of of black black spruce (Picea height (Picea mariana). and and was was viewed viewed as as ideal ideal for supporting high.:.intensity high.:.intensity crown fires fires in normal mariana).
on experimental experimental burning burning experience experience in in other other comparable fuel fuel summer weather weather based based on summer complexes (Stocks (Stocks 1987. 1987. 1989). 1989). The The study study area area is is surrounded by by shrub-dominated meadows meadows varying from from - 0.5 - 11 km km in in width width that that were were burnt burnt off in in May May 1995 1995 and and portions portions of it it again in varying May May 1997 1997 with with aa view view to to begin begin securing the the area area from from the the possibility possibility of "escapes" or excursions excursions from from the the planned planned experimental experimental fires. fires.
86
INTERNATIONAL CROWN FIRE FIRE MODELLING MODELLING EXPERIMENT EXPERIMENT INTERNATIONAL CROWN NORTHWEST NORTHWEST TERRITORIES, TERRITORIES, CANADA CANADA . Study Study Area Area Plot Plot Layout Layout Boo Dw..t Dw..t Birch Birch Boo (Bumt .... .... 41116) (Bumt 41116)
@ @
NFPA NFPA 299 299 Standard Standard
Clearing (WUI) (WUI) Clearing Existing Existing Cut Cut Unea Unea - - Plot Plot Boundary Boundary
5
-
-
--
_
_
FIreguIrd FIreguIrd
Area Boundry Boundry Area
s~-----~ 400 o 200 400 200 o L-' --r.. ''--..,.. ...... '...l --,~.I.. r--.l.. ---Wi ---. . 100 100
300 300
600 600
m
m
Figure 1: Plot layout for the International Crown Fire Modelling Experiment (ICFME) study area. area. To reference the location of the study area area in in relation to the geography of the Northwest Territories visit the ICFME homepage. ::.'.
'
Ten experimental burning plots were surveyed in the summer of 1995 (Figure 1). Eight of the ten original plots are 150 x 150 m in in size; the other.two are smaller (75 x 75 m and 100
to account for some likely variation variation in x 100 m). Plot orientation has been purposely varied to predominate wind directions during the "burning window", which was selected on the basis of day length, local climatological knowledge, and an analysis of historical fire danger records available for the DRWED fire fire weather station in in Fort �vidence. ~vidence. In 1996. three additional plots were established adjacent to the primary plots in in order to analyze various aspects of fuel 87
management treatments on potential fire behavior such as the effects of thinning/pruning and the value of trembling aspen (Populus tremuloides) stands as a fuelbreak. Following the first phase of ICFME burning, four additional plots were established in July 1997 to further
(Putnam 1995, 1996; Roth 1997) on a examine the effectiveness of: (i) protective fire shelters (Putnam seismic line line and in small clearings as opposed to a cleared break such as represented by the firegaurds (plots Sl and S2 S2)) and (ii) the NFPA NFPA (1991) 299 Standard for community horne
protection from wildfires (plots 11 and 12 12 complete with simulated houses).
METHODOLOGY
Preburn sampling of the ground, surface and crown fuel characteristics load, depth and bulk density (cf. Walker and Stocks 1975) was undertaken during the summers of 1995 and 1996 using standardized techniques (e.g., Stocks 1980; Nalder et al. 1997). Construction of
flre 50-m (minimum) wide fireguards around each of the plots in order to facilitate access and flre control considerations (Le., to compensate for the lack of surface water in the area) was begun in 1995 and completed in in the fall of 1996. Two ICFME preburn planning meetings were held in in Calgary, Alberta, Canada, during December 1995 and again in January 1997 in order to
bring together the participating researchers from Canada and the United States together with DRWED staff and native community leaders from Fort Providence. These meetings resulted in the development of logistical and research plans that were acceptable to all parties involved in the ICFME project
A World Wide Web homepage (address: http://www.nofc.forestry.calfirelfmnlnwtl) for A in scope and the ICFME initially developed in the spring of 1997, has gradually expanded in
attained a certain degree of notoriety within the wildland fire community (Hogenbirk 1998).. This use of the internet's capability has greatly facilitated effective communication at various levels and aided in certain operational aspects of the project as wen. constitutes another
to advance wildland fire research globally example of how computer technology has help to
A fully-instrumented fire weather station was established adjacent to the plots (Weber 1995). A soon after snow-free cover in the large, open meadow area south of the plots (Figure 1) as soon daily 1300 MDT MDT fire fire weather observations from this station each spring as possible. The daily
permitted the calculation of the six standard components of the Canadian Forest FIre FIre Weather
88
Index Index System System (Van (Van Wagner Wagner 1987) 1987) which which were were in in turn turn used used along along with with the the fire weather forecast forecast to to determine determine the the suitability suitability of burning burning on any given given day. Spot fire weather forecasts for the ICFME were prepared twice daily at around 0800 and 1500 MDT (Figure 2) and posted on the ICFME homepage. The 0800 forecast forecas t predicted the weather for the curre nt day including hourly predictions of temperature, relative humidity, current probability of precipitation, wind speed and direction as well the sky condition. The 1500
issue was was the the forecast forecast for for days' 2 to to 4 and included maximum maximum temperature, temperature, minimum relative humidity, probability of precipitation and predominate wind speed and direction.
Forecast at 0815 1997, Forecast 1997 Forecast Issued Issued at 0815 MDT MDT 04 04 July July 1997, Forecast for for Friday Friday 04 04 July July 1997 Discussion Synopdc Discussi on At the surface
High pressure area over the Beaufort with a ridge extending to Great
Slave Lake. Lake. Mostly sunny skies and light winds accompany the ridge. At 500 mb
Upper high over southern Alaska with ridge eastward to extreme western
sections sections of the the southern southern Mackenzie. Mackenzie. Upper Upper low low along the the Arctic Arctic coast with with a trough trough southeastward to northern northern Saskatchewan. Moderate Moderate northwest northwest flow flow over the the southern southern southeastward to Mackenzie.
Time
Sky
MDT
RH RH
Temp. Direction! Speed
(%)
�C) ~C)
(kmIh) (kmIh)
1300
Sunny
30
20
Light + variable
1400
Sunny
28
211 2
L+V
1500
A few clouds 25 25 A
22
L+V
1600
A few clouds 25 25 A
22 22
L+V
1700
A few clouds 25 25
22
L+V
,.
,
1800
A few fewcJouds A cJouds 28
.. 21 21
L+V
1900
Sunny
28
·21
L+V
2000
Sunny
30
20
L+V
89 89
.
Comments and Confidence Confidence Co mments and Winds... .light northeast this morning becoming light and variable by noon. Best guess as to
direction would be NE NE this morning becoming E or SE this afternoon. Wind-speed will be light but variable as well with winds up to 15 kmIh kmIh on occasion. One model- the Canadian south (100 Spectral Model has a high pressure cell developing just south
Jem) of the fire fire site this Jem)
If this happens afternoon. afternoon. If happens then outflow from the high could could push the direction to a more
southerly direction. direction.
Probability of prec..... 0%
Steering Flow (500 mb) .... Moderate
northwest Day 2-4 ued at 1415 MDT Friday 04 2-4 forecast iss issued 04 July 1997
DfscussionIRemarks nIRemarks Synoptic Dfscussio Upper air air features ... Upper high and ridge in Alaska moves west Ridge over Alberta -
southern Mackenzie weakens and moves southeast This This results in a broad trough extending Alaska. from a low over the Arctic Islands to a low in the Gulf of Alaska.
At the the surface .... High High pressure pressure area area moving moving into into Saskatchewan Saskatchewan with with aa ridge ridge extending extending to to Great Great Slave Lake on on At surface.... Slave Lake Saturday. Low pressure pressure area area developing developing over over the the extreme extreme southwest comer of of the the Mackenzie. Mackenzie. Saturday. Low southwest comer
Saturday.....in .increasing with aa chance chance of of an an afternoon Monday.... cloudy with with aa Saturday creasing cloud cloud with afternoon shower. shower. Sunday Sunday and and Monday .... Mainly Mainly cloudy
of an afternoon shower. chance of chance an afternoon shower.
Saturday.AO% pop Temp 23C Min RH RH 40% 40% Winds SE 20 kmIh kmIh Sunday....40% 40% pop Temp 23C Min RH RH 40% Wmds Wmds E 15 kmIh kmIh RH 40% 40% Wmds Wmds E 15-20 kmIh Monday ..AO% pop Temp 23C Min RH kmIh
Forecasts prepared by Mike Flannigan
phA03 435-7338 435-7338 (office) 403436-1626 (home) 403436-1626 e-mail
[email protected] Figure 2: Sample Sample weather weather forecast issued daily during the burning p phases the International International Crown Crown forecast issued hases of of the Figure rrre Modelling Experiment
rrre
90 90
For both both forecasts a synoptic discussion of the key features in in the upper air air and surface was included. The forecasts were prepared by a CFS meteorologist based in Edmonton, Alberta, using information obtained from a variety of internet sites. Surface analysis, upper air air charts, skew-T thermodynamic diagrams (tephigram) from three three nearby upper air air stations rman Wells and Fort Nelson), satellite images (GOES and NOAA, visible, (Fort Smith, No Norman
infrared, infrared, enhanced and animations (loops», and hourly surface observations were all all available
on the internet in a timely fashion. fashion. Also, available on the ICFME internet site were output (ECMWF, MRF MRF and CMC products) as as from a wide variety of numerical weather models (ECMWF, well as forecasts prepared by the Canadian Atmospheric Environment Service. No radar data was used as there is no radar operating in this region. In 1998, a portable radiosonde will be used on potential burning days to provide a detailed vertical structure of the atmosphere over the ICFME site for both forecast purposes and for quantifying the upper air air component of the fire environment associated with each experimental crown fire. Updated forecasts and consultations were provided as required on potential burning days via communication through
link right at the ICFME study area. a satellite phone link The experimental plots were ignited as a 'line source' using a truck-mounted pressurized flame thrower or "terra-torch" (Bradshaw and Tour 1993). Typically, the length of a 150 m plot edge could be be "fired up" in less than a minute. Permanent DRWED staff or contractors and fire suppression crews on contract to DRWED provided all all the necessary logistical support for conducting the experimental fires, including the establishment of a temporary base camp at the study area. A vast arr ay of ground-. tower- and helicopter-based instrumentation was employed to array quantify numerous fire characteristics in the most detailed manner possible.
All of this All
equipment worked extremely well, permitting measurements of smoke chemistry, flame size arid geometry, radiant heat flux. gas temperatures and fire spread using visible and infrared
cameras, radiometers and thermocouples (e.g., Butler 1994� 1994~ Kautz 1997).
ICFME PHASE I -
-
1997 PROGRESS REPORT
After three years of planning and preparation. the first phase of ICFME burning took 1997. place in June-July 1997.
data Although designed primarily to develop knowledge and data 91 91
essential physical behavior fires, essential to to predicting predicting the the physical behavior and and impacts impacts of of high-intensity high-intensity crown crown fires, ICFME-I together a a diverse of environmental environmental and wildland fire scientists as as ICFME-I brought brought together diverse group group of and wildland fire scientists
well managers from from various various parts North America America and and overseas truly cooperative well as as managers parts of of North overseas in in a a truly cooperative research undertaking. research undertaking.
A total total of A of
individuals from from outside outside of NWr participated, participated, for 45 individuals of the the NWr for
varying lengths the first first field field phase phase of the ICFME. ICFME. This This included group of varying lengths of of time, time, in in the of the included a a group of
20
from Canada Canada (principally (principally members CFS Fire Fire Research Research Network Network based based at at Edmonton, Edmonton, from members of of the the CFS Alberta, Sault Ste. Ste. Marie, and Victoria, as well well as as representatives representatives Marie, Ontario Ontario and Victoria, British British Columbia, Columbia, as Alberta, Sault
from the from Canadian fire management management agency), agency), aa group from Canadian universities universities and and a a provincial provincial fire group of of 23 from the United States (principally from from federal agencies although United States (principally federal government government agencies although universities universities and and private private were represented represented as grassland fire fire scientists scientists from industry (see industry were as well), well), and and two two grassland from South South Africa Africa (see ANNEX!). ANNEX !). After wet spring spring in in the the Fort Fort Providence researchers began arriving on on June area, fire fire researchers began arriving June After aa wet Providence area,
19 in weeks of Over the three in preparation preparation for for three three weeks of experimental experimental burning burning in in 1997. Over the next next three weeks, gradual drying trend produced burning conditions ICFME site weeks, a a gradual drying trend produced some some ideal ideal burning conditions at at the the ICFME site (Figures (Figures
The weather this three-week three-week period 3 and and 4). The weather experienced experienced during during this period was was unusual unusual in in
several Typically, in atmosphere (500 mb) experimental site several respects. respects. Typically, in the the upper upper atmosphere mb) the the experimental site is is on on the the of North east that extends along the west coast North America. America. However, However, in in east side side of of a a upper upper ridge ridge that extends along the west coast of
the Gulf 1997 the the 500 mb mb level level was was dominated dominated by by two two well-developed well-developed upper upper lows, lows, one one over over the Gulf of Alaska and Arctic Islands, Islands, which which resulted trough over of Alaska and the the other other over over the the Arctic resulted in in a a broad broad trough over the the Southern Mackenzie the surface, surface, a a trough usually extends extends along along the the Mackenzie Southern Mackenzie District. District. At At the trough usually Mackenzie
River valley year. During During the River valley at at this this time time of of year. the
1997 burning window, surface pressure highs burning window, surface pressure highs and and
ridges Mackenzie District. District. Daily Daily rainfall rainfall was ridges were were common common over over the the Mackenzie was very very light light through through this this mm on June three-week period. Rainfall totals (24-h) exceeded exceeded 0.5 mm on June three-week period. Rainfall totals (24-h)
26/97 (5.3 mm) mm) and and again again
on July 1 1112 ((10.1 10. 1 mm). cold front was responsible responsible for for the the rain rain on on June while an an upper upper on July 11112 mm). A A cold front was June 26 while
disturbance (cold low) brought rain rain to July 11112. typical summer to the the site site on on July 1 1112. In aa typical summer airmass airmass disturbance (cold low) brought showers/thunderstorms by daytime daytime heating heating are are common. common. Winds were light light during during showers/thunderstorms triggered triggered by Winds were most of of the the period, period, with with wind wind speeds speeds over over 20 krn/h krn/h observed observed on on only only three days. most three days. After few small small test test fires, the first full-scale crown crown fire was ignited ignited on on July After aa few fires, the first full-scale fire was July
1 Canada Canada
Day! This This was additional successful fires on July 4 and Day! and was followed followed by by two two additional successful high-intensity high-intensity crown crown fires on July July July
A number number of potential burning .9. A of other other potential burning days days had had to to be be cancelled cancelled due due to to largely largely
unfavourable wind The unfavourable wind directions. directions. The
1997 burning rains burning window window effectively effectively ended ended with with rains
The burning with the fires are are beginning July 10. The beginning on on July burning conditions conditions associated associated with the three three 1997 fires documented Table 1. Postburn Postburn fuel up was July 12. documented in in Table fuel sampling sampling and and clean clean up was completed completed by by July
92
35 35
Internatfonal Crown Fire Modelling Experiment, Northwest Territories, Canada· Phase II
100 100
8830 0..- 25 25 0..-
30
j20 j20 �~ 15 15
.$ .$ � 10 ~ 10
5 5
Relative humidity humidity Relative
o
o
o
o
··40 ··40
360 360
nd direct! Wnd direct! n n W
315 315
270~ �
270
2259: 2259:
1801 1801 g 135g
-
135 go...g.
-
go...g.
o
o
45 45
,May , May
1997 1997
June June
July Ju ly
o
o
3: Daily trends in in 1300 in Figure 3: 1300 MDT fire weather observations prior to and during the first phase of the International Crown Fire Fire Modelling Experiment in 1997. 1997.
International Crown Fire Modelling Experiment,Northwest International Crown Territories, Canada I International Crown Fire Fire Modelling Modelling Experiment, Experiment, NorthwestTerritories, Territories,Canada Canada- -Phase Phas
e I
50 50 50
.•
.•
[i[ 40 [i[ Eff 40
II 30 II
40 30
111020 11 1 30
20 20
~� 1 1000
0
�eo i~eo iMa° e l: 60 jjl: tf
tf a
1
00 100 00
500 500
500
400§ 400
�12 i20 ~i200 oooo &00 d !!d[eo [eo
400 § O~ 0 O� 30 ~ 30 � 200 E 200:8 200 :8
DC* B
4° 20
30
•
100 100
100
000
0
FFMC
�eo80 �~~eo
20 20
20
&In
40
20
20
000
May May
May
_
r4n 3
16
e0
�~U., I ° 40
;
; 16 15 i i 10 I 10 I 55
5
June 1997June
1997 1997
July
July July
000
Figure 4: Daily trends in rainfall and the six standa rd comp onents of Daily trends trends in in rainfall rainfall and and the the six six standard standard components ofthe theCanad IndexSyste of the the Figure 4: Daily components of the Canadian Forest FireWeath Weather System priortoto and during the Phase II of ian Fores t Fire er Index m prior Intern ational Crow n Fire International Crown FireMode Modelling Experiment in1997. 1997. Modelling Experiment in 1997. lling Exper iment in
and during the Phase I of the
ICFME PHASE II II - 1998 PROGRESS REPORT
during roughly roughly the the same same period period as as in in The second second phase phase of of ICFME ICFME burning burning in in 1998 during The
1997. For For the the second second year in in a a row, row, the the ICFME brought brought together together aa diverse diverse group group of of scientists scientists the United States as well as Russia, Russia, France and Australia. and managers from Canada and the
ICFME-ll participants participants began began arriving arriving in in Fort Fort Providence Providence on on June June 22 22 in in preparation preparation for for three three ICFME-ll
weeks of experimental burning; a smaller advance advance group arrived arrived on June 18. A total total of 59 research and and operational operational staff staff from from outside outside the the Northwest Northwest Territories Territories participated, for varying research
lengths of of time, time, in in ICFME-ll. ICFME-ll. This This included included a a group group of of 28 28 from from Canada, Canada, a a group group of of 27 from from lengths the United States, States, one one Australian Australian bushfire bushfire research research scientist, scientist, two two Russian Russian fire fire scientists scientists and and a a the United
PhD student (see ANNEX ANNEX 1). 1). student from from France. France. (see PhD
Furthermore, film film crews crews from from Austria Austria and and Furthermore,
England gathered imagery imagery for various upcoming wildland fire documentaries on various aspects of of wildland wildland fire. fire. aspects
Snow-free cover occurred about two weeks earlier than normal in the the spring of 1998.
The main main weather weather station at the ICFME site was established on May 7 (as compared to May The
14 in in 1997). In In contrast contrast to to 1997, the the spring spring of of 1998 was was relatively relatively dry dry (the (the Duff Duff Moisture Moisture
of 84 on June June 10). As As participants participants arr arrived ived on June June 22 22 they were were greeted greeted Code reached reached aa high high of by a downpour in during the the next next several days a gradual by a downpour in Fort Fort Providence Providence during the evening. evening. Over Over the several days a gradual
trend produced produced conditions suitable suitable for for initiating initiating crown fires fires (Figures (Figures 5 and and 6). After aa drying trend few small fires, the crown fire on the Independence Day USA's Independence Day few small test test fires, the first first full-scale full-scale crown fire was was ignited ignited on the USA's by an crown fire holiday, holiday, July July 4 (plot (plot 8), followed followed by an additional additional successful successful high-intensity high-intensity crown fire o� o~ July July
5 (plot ten minutes before igniting "rogue" thunderstorm thunderstorm dumped dumped 6.5 (plot 7). On On July July 6, ten minutes before igniting Plot Plot II, aa "rogue" mm mm
of This effectively of rain rain directly directly on on the the plot plot in in the the space space of of about about 20 minutes. minutes. This effectively ended ended any any
chance chance of of burning burning within within the the remainder remainder of of the the 1998 1998 window window (scheduled (scheduled to to end end July July 13) although although hope hope was was held held out out until until as as late late as as July July 9 when when the the project project had had to to be be terminated terminated due due to to the The fire the territorial territorial wildfire wildfire situation. situation. The fire weather weather and and fire fire danger danger conditions conditions associated associated with with
the two two 1998 1998 fires fires are are given given in in Table Table 1. 1. the
95 95
Table I: I: Fire weather observations and fire danger indexes associated with the various fires
documented in the International Crown Fire Modelling Experiment to date.
ICFME ICFME plot
Date Date
of buming
Time Time
. Dry-bulb Relative of ignitio temperature humidity n
No
A A
(dd1rnm/yy)) (MOl) (MOl) (dd1rnm/yy
fC) fC)
(%) (%) 28
S S
0I/f17197
1426
22.3
04/(J7197 04/(J7197
1730 1730
20.6
6
09/07197
24.0.
8
04/(J7/98
7·
05107/98
1406 1604 1604 1642 1642
1 1
..
30.4
29.2
37
44 44 24 24
41
100m Days Canadian Forest Fire Canadian open Weather Index � Weather Index ~ wind since System components speed (kmIb) rain2 rain2 FFMC FFMC DMC DMC DC DC lSI lSI BUI BUI FWI FWI (kmIb) 5 15.9 15.9 91.8 12.2 56 27 3S 348 12.2 12.6 8 12.6 43 370 7.4 67 21 21 89.4 13 89.9 59 410 9.9 86 29 17.2 5 9.4 91.9 37 343 8.8 58 22 6 16.4 92.5 42 352 13.6 13.6 65 41
fire behaviour indexes comprising the FWI FWI The three fuel moisture codes and three fire
System are defined below (from Canadian Forestry Service 1984): - A numerical rating of the moisture content of litter Fine Fuel Moisture Code (FFMC) and other cured fine fuels. This codes is an indicator of the relative ease of ignition and flammability flammab ility of fine fuel. - A numerical rating of the average moisture content of Duff Moisture Code IDMC) loosely compacted organic layers of moderate depth. This code gives an indication of fuel consumption in moderate duff layers and medium-sized woody materials. (DC) - A A numerical rating of the average moisture content of deep, Drought Code (DC)
This code is a useful indicator of seasonal drought effects on forest compact, organic layers. This fuels, and amount of smouldering in deep duff layers and large logs. It rate of fire fire spread. It Initial Spread Index asn - A numerical rating of the expected rate
asn -
combines the effect of wind and FFMC on rate of spread without the influence of variable quantities of fuel.
(BUD - A numerical rating of the total amount of fuel available for Buildup Index (BUD combustion that combines DMC and DC. DC.
- A numerical rating of fire intensity that combines lSI and FIre Weather Index
