Professional Documents
Culture Documents
Director
Centre for Carbon Water and Food
12 May 2016
Dear Minister,
Open Letter about Wye River and Separation Creek
th
Similarly, some officials believe it reasonable that the State Government and the COS may impose
BAL ratings that assume all private property is unmanaged with respect to fuel loads, yet at the same
time have the power and responsibility to fine people who do not manage their fuels!! The State
Government and the COS are clearly not happy enough with imposing costs on private land owners
via BAL ratings, they wish to also impose costs via fines by ignoring their own regulations for the BAL
ratings. Joseph Heller would be proud. The illogicality of this situation is a prime candidate for legal
action by any whom may have pockets deep enough to fight local or state governments.
The two recent reports by Leonard et al., and by Dexter et al. point further to the hypocrisy and
illogicality that is doing the rounds within government circles at present.
The Leonard et al report shows clearly that the current BAL ratings had little effect in protecting
houses (or people). In fact, they (the BAL ratings) were an extraordinarily weak predictor of house
survival or destruction. Fuels of all kinds vegetation, houses, tanks, stored materials, retaining walls
are the major risk factor that can be mitigated. However forcing owners to bear huge costs to
reduce the risk to adjoining houses, via costs of construction of their house, is utterly pointless and
illogical if the vegetation fuels remain unmanaged! How on earth can anyone justify an argument for
BAL ratings based on an assumption of unmanaged fuels?
The Dexter et al. report, disturbing as it is on the subject of the overall fire management, yet again
shows how important it is for the state government to manage fuels on public land. No serious
scientist that I know of disputes that back-burning is a legitimate fire-fighting tactic. However, unless
fuels have been systematically reduced PRIOR to a fire starting, the legitimacy, effectiveness and
safety of all fire-fighting efforts, including back-burning, is seriously compromised.
Finally, I must question the use of taxpayer funds to produce a video clip. Undoubtedly the argument
from government is that this clip provides information to those affected. My view is that it is rather
more propaganda for the governments position.
The clip is perplexing in the use of comparisons between Wye/Sep and Anglesea or Aireys Inlet. One
speakers point seems to be that Wye River and Separation Creek have wetter climates and
vegetation, and steeper terrain, and that they are thus more difficult places to create an APZ. This is
so obvious as to not be worthy of comment. Putting aside the fact that the climate and terrain at Wye
River is very similar to that at Lorne (which has been far better protected by fuel reduction), the fact
remains that governments state and local - have the responsibility to protect life and property of ALL
communities equally.
Difficult as fuel management may be around Wye and Sep, it was the Andrews Government that has
decided to step away from the recommendations of the 2009 Victorian Bushfires Royal Commission,
and instead create a zoned approach to fuel management. Difficult is now seemingly code for
inaction. As noted by the Dexter et al. report, the terrain around Wye and Sep is no more difficult than
that in many other parts of the state. If difficulty becomes a criterion for decisions about where APZ
will, and more importantly, will NOT be created, then the Victorian Government is making decisions
that will seriously and negatively affect the value of hundreds, and probably thousands, of houses.
This too will become a matter for lawyers and class actions that will cost tax-payers dearly.
It is governments that zone and sub-divide land, that collect stamp duty upon sale, and levy rates.
There is no warning when you buy property in Wye River and Separation Creek to the effect: watch
out, the government will not support you the way it supports those elsewhere!
The present approach is a highly retrospective imposition of costs on private land-owners, and
unequally and unfairly imposes those costs. That your Government should seek to financially punish
those who lost houses, seems completely at odds with public statements about fairness and about
support for communities.
As I concluded in my last communication to you, I am very concerned that the approach being
followed pits neighbor against neighbor, and removes the right of all those that own property anywhere
in Victoria, to expect they will be treated equally when it comes to protection of their assets.
It has been disturbing to listen to government officials try to rationalize the proposed BAL ratings on
the basis that these are all about keeping people and property safe. As noted above, the current
BAL ratings are inadequate for that purpose. Moreover, the BAL ratings are not going to be applied to
existing houses which, as so clearly set out in the Leonard et al report, pose very real risks to
adjoining houses. Many of the most fire-prone and flammable houses remain intact. Imposing costs
on those seeking to rebuild does nothing about the risks to life and property created by the existing
housing stock. This extraordinarily inequitable situation might be somewhat justified if it were the case
that the BAL ratings, consistently applied over coming years and decades, were to reduce risk overall.
Sadly, the BAL system is so flawed that this argument evaporates. In its place we have a one-off,
retrospective cost imposition being unequally and unfairly applied, for little or no net gain for the safety
of the community.
Property owners have been recently advised that the government will now adopt a street-by-street,
property-by-property approach to the BAL ratings. There is some merit in this, but only if the Victorian
State Government and the COS accept their responsibility to manage fuels, either directly in the form
of a declared and properly maintained APZ for all of Wye and Sep, and in properly maintaining fuels
on roadside and other reserves, and indirectly managing fuels via private land owners.
Unless state and local governments accept their proper role(s) and costs in managing fuels on public
land, and acknowledge publicly that they will bear those costs, the present approach could be
characterized as a cynical exercise in cost-shifting and divide and rule politics, that will ultimately fail
the real test have governments served well their people?
Yours faithfully,
CSIRO L AN D AN D WATER
2 Country Fire
Authority
Report EP16924
April 2016
Report to the Victorian Country Fire Authority
ii
Citation
Leonard et al, Wye River / Separation Creek - Post-bushfire building survey findings (2016) CSIRO Client
Report EP16924
Important disclaimer
CSIRO advises that the information contained in this publication comprises general statements based on
scientific research. The reader is advised and needs to be aware that such information may be incomplete
or unable to be used in any specific situation. No reliance or actions must therefore be made on that
information without seeking prior expert professional, scientific and technical advice. To the extent
permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for
any consequences, including but not limited to all losses, damages, costs, expenses and any other
compensation, arising directly or indirectly from using this publication (in part or in whole) and any
information or material contained in it.
Acknowledgements
Victorian Country Fire Authority (CFA), Department of Environment, Land, Water and Planning (DELWP),
Emergency Management Victoria and the Colac Otway Shire Council
iii
Executive summary
Following the Wye River fire event the CFA commissioned CSIRO to perform a review and provide a report
on house losses in the Wye River fire footprint. This report identifies factors that led to the loss, damage
and survival of houses and illustrates these factors by presenting examples of houses within the fire
footprint. This report has a strong focus on houses built under Bushfire Management Overlay (BMO)
provisions.
This study found seven examples of houses built to the regulatory standards, which have been in place
since 2010, that were impacted by fire. Of these seven houses, four were lost to fire and three survived.
Although the number of buildings in this sample is small, it does suggest a higher survival rate than the 80%
loss rate experienced in the region affected by fire. The fourteen houses built to planning and building
regulatory standards between 2003 and 2010 fared much better; three were lost to fire and eleven
survived. It is clear from the study that the house loss rate would have been significantly higher if
suppression was not provided by ground and aerial crews.
The townships of Wye River and Separation Creek experienced a broad scale surface fire spread through a
combination of dried grasses and litter fuels originating from extensive established tree coverage. The main
impact of the fire on houses was through ignition of heavy fuel elements that were adjacent to or under
buildings. These include adjacent houses (house-to-house ignition), combustible retaining walls,
combustible decking, combustible stairways, vehicles, stored equipment, plastic water tanks and firewood.
It is clear that buildings built to the bushfire regulatory standards as well as buildings that were not built to
the bushfire regulatory standards were subject to these heavy (domestic) fuels, and many were not capable
of withstanding the radiant heat or flame contact from these fuels. This appears to be a key reason why
such a large house loss rate was experienced. There were few examples of fire spread through surface fuels
that were sufficiently intense to provide radiation exposure levels capable of compromising houses directly,
these were limited to a few areas within the township.
Surface fuels within the township were readily ignited by embers which rained down on the township, this
meant that many parts of the township were simultaneously ignited. Subsequently the fire spread up
through surface fuels in the township in multiple fire fronts. During this time people outside their houses
would have been presented with life threatening exposure levels of radiant heat, flame and smoke. These
surface fires spread would have seamlessly transitioned into heavy fuel ignition throughout the township
involving a number of houses. The combustion of these heavy fuels would have also provided life
threatening levels of radiant heat, flame and smoke to anyone in the area that was not able the shelter in a
house or vehicle including fire fighters. The steep terrain and challenging access within the township meant
that risk to life of occupants forced to leave burning houses would have been high. Roadways cut into steep
slopes and house access via steep driveways meant that any egress during the fire by foot or by vehicle
would be difficult and life threatening. Very few clear areas for refuge were found within the fire-affected
areas of the townships. Despite these difficulties, no lives were lost, which is testament to the warning
given to occupants prior to the arrival of the fire and the willingness of occupants to heed these warnings.
iv
Executive summary
iv
2.1
Terrain .......................................................................................................................... 2
2.2
2.3
2.4
Occupancy ..................................................................................................................... 6
2.5
2.6
House maintenance......................................................................................................... 7
2.6.1 Subfloor areas ....................................................................................................... 7
2.6.2 Town gas supply..................................................................................................... 7
12
16
23
27
7.1
Suppression ................................................................................................................. 27
7.2
Water supply................................................................................................................ 29
7.3
Extent of loss................................................................................................................ 29
8.1
Regulation ................................................................................................................... 30
8.2
8.3
References
A.1
A.2
30
32
A.3
A.4
A.5
A.6
vi
attended the fire-affected area as part of a multi-agency (CSIRO, CFA) survey initiative
gathered information from various sources to describe the nature of the fire and fire
suppression activities around these selected houses
gathered information (from the local council and CFA) on the nature of building regulatory
standards for the selected houses
gathered information around the likely state of the houses prior to fire arrival
identified issues likely to have contributed to the loss, damage or survival of houses.
The report is intended to be used by the CFA to inform the provision of community information, education
and engagement activities.
Surveys were carried out from 6 to 8 January 2016 to examine the remains of the destroyed, partially
damaged and unaffected houses as well as their surroundings within the fire perimeter.
The surveys aimed to provide a better understanding of the mechanisms of house failure or house survival
and is based on previous post bushfire surveys [3, 4]. Detailed information on individual properties was
collected by geo-referencing and attributing relevant elements of structures, surrounding objects and other
pertinent information (e.g. degree and cause of damage, site details and combustible elements). This
report provides the findings of the review of this data combined with other available information that
relates to weather, fire arrival and spread observations to gain a better understanding of structural design,
building regulatory standards and planning issues.
2.1 Terrain
Both Wye River and Separation Creek are built on steep south-eastern facing slopes. The townships extend
from the coast up these slopes (Figure 2). The ridges are covered predominantly by native bushland. Some
housed areas of Wye River and Separation Creek reach gradients higher than 30 degrees (Figure 3).
Because of the steep terrain, many road and driveway cuttings require extensive land stabilisation. House
and other building access is typically via narrow gravel roads and driveways, which makes access for fire
suppression and rapid egress difficult.
Figure 2: Digital elevation model for the townships of Wye River and Separation Creek based on Shuttle Radar
Topography Mission (SRTM) data.
Figure 3: Distribution of slopes for parcels of land in the Wye River and Separation Creek townships.
Figure 4: Example of land stabilisation using timber sleepers in proximity to other combustible features within Wye
River.
Mature native trees provide important land stabilisation, shade and wind attenuation to the township.
However, they also deposit extensive leaf, bark and twig debris on buildings and the surrounding landscape
(Figure 6). This litter, in combination with cured grasses, provides a near continuous surface fuel layer,
broken only by areas of hard surfacing or roadways where wind, run off and/or local traffic shifts the
surface fuels.
2.4 Occupancy
The township has a small proportion of permanent residents. The majority of houses are owned as a
secondary residence, such as holiday houses or investment properties. During the summer period, some
houses may be vacant or occupied by a non-owner.
Early on 25 December, warnings encouraged residents of Wye River and Separation Creek to evacuate. It is
understood that impacted houses were vacated prior to the arrival of the fire and no lives were lost during
the fire event.
Figure 7: Planning zones and overlays relevant to the Wye River and Separation Creek townships.
These provisions are likely to interact with the ability to implement various bushfire mitigation measures.
6
Figure 8: A house on Wallace Street, Wye River which was not impacted by fire showing under house fuel loading.
The steep terrain also means retaining walls are prominent throughout the fire-impacted area. A large
proportion of these are constructed using timber sleepers (Figure 4).
40
90
35
80
30
70
60
25
50
20
40
15
30
10
20
24/12/2015
R.H.(%)
23:30:00
21:30:00
19:30:00
17:30:00
16:00:00
14:30:00
13:01:00
12:00:00
9:00:00
Temp (C)
10:30:00
7:30:00
6:00:00
4:00:00
2:00:00
0:00:00
22:00:00
20:00:00
18:00:00
16:00:00
14:00:00
12:00:00
8:00:00
10:00:00
6:00:00
0
4:00:00
5
2:00:00
10
Temperature (C)
100
0:00:00
The month of December 2015 was the sixth-warmest month for nationally-averaged mean temperature on
record in Australia, according to the Australian Bureau of Meteorology (BOM) monthly weather review [5].
The nearest automatic weather station (AWS) to the Wye River and Separation Creek townships, with
available data, is at Aireys Inlet, 25 km to the east and Cape Otway AWS, 41 km to the west. For each
station, half-hourly weather records are available. The relative humidity and temperature on the day
preceding and on the day of the fire (24 and 25 December) is shown in Figure 9 (Aireys Inlet) and Figure 10
(Cape Otway).
25/12/2015
40
90
35
80
30
70
60
25
50
20
40
15
30
10
20
10
24/12/201
Temperature (C)
100
0:00:00
2:00:00
4:00:00
5:39:00
7:30:00
9:30:00
11:30:00
13:30:00
15:30:00
17:00:00
18:30:00
19:30:00
21:00:00
22:30:00
0:30:00
2:00:00
3:31:00
5:16:00
6:30:00
8:00:00
10:00:00
11:21:00
13:00:00
14:30:00
15:33:00
17:18:00
18:30:00
20:30:00
22:30:00
Figure 9: Half-hourly observations of relative humidity and temperature recorded at the Aireys Inlet AWS (90180)
on 24 and 25 December 2015.
25/12/201
R.H.(%)
Temp (C)
Figure 10: Half-hourly observations of relative humidity and temperature recorded at the Cape Otway AWS (90015)
on 24 and 25 December 2015.
Both Figure 9 and Figure 10 show cool, damp early morning conditions on 24 December with the humidity
approaching 95%, followed by moderate daytime conditions where temperatures approached 25C and
50% relative humidity. As the day progressed into evening, the conditions persisted until just after midnight
when the temperatures at both stations rose and humidity levels dropped. This coincides with a swing in
the wind direction to a northerly and an elevation in wind speed. The conditions persisted for the rest of
the morning, with a further increase in temperature and reduction in relative humidity as the day
progressed. With the warm and dry overnight conditions preceding the fire event, fuel elements within the
townships would not have regained the moisture lost during the daylight hours of 24 December and would
have continued to dry throughout the day of 25 December prior to the arrival of the fire. The wind speed
and direction profiles are shown in Figure 11 and Figure 12.
360
300
240
180
120
0:00:00
1:30:00
3:00:00
4:30:00
6:00:00
7:30:00
9:00:00
10:30:00
12:00:00
13:30:00
15:00:00
16:30:00
18:00:00
19:30:00
21:00:00
22:30:00
0:00:00
1:30:00
3:00:00
4:30:00
6:00:00
7:30:00
9:00:00
10:03:00
11:30:00
13:00:00
14:30:00
16:00:00
17:30:00
19:00:00
20:30:00
22:00:00
23:30:00
60
Figure 11: Half-hourly observations of wind direction recorded at the Cape Otway AWS (90015) and Aireys Inlet
AWS (90180) on 24 and 25 December 2015.
45
40
35
30
25
20
15
10
0
0:00:00
1:30:00
3:00:00
4:30:00
6:00:00
7:30:00
9:00:00
10:30:00
12:00:00
13:30:00
15:00:00
16:30:00
18:00:00
19:30:00
21:00:00
22:30:00
0:00:00
1:30:00
3:00:00
4:30:00
6:00:00
7:30:00
9:00:00
10:03:00
11:30:00
13:00:00
14:30:00
16:00:00
17:30:00
19:00:00
20:30:00
22:00:00
23:30:00
Figure 12: Half-hourly observations of average wind speed recorded at the Cape Otway AWS (90015) and Aireys
Inlet AWS (90180) on 24 and 25 December 2015.
Based on AWS data from Aireys Inlet (90180), Cape Otway (90015) and Lorne (90185), the rainfall for the
Wye River region was well below (-42%) the December state average (see Figure 13 and Table 1). In the
calculation of long-term fuel dryness, only rainfall in excess of 5 mm is considered enough to increase longterm fuel moisture. 1 The first 5 mm of any rainfall event is assumed to be lost due to canopy interception
and subsequent evaporation. None of the AWS rainfall data indicated an event exceeding 5 mm in the
month leading up to the bushfire. The AWS data resulted in a KeetchByram [6] Drought Index (KBDI) of 80
at Aireys Inlet (90180) and 83 at Cape Otway (90015) on the day of the fire. These levels are considered
serious with respect to fuel ignition potential 2 and led to a drought factor of 9.3 for input into Forest Fire
Danger Index (FFDI) calculations. These low soil moisture levels mean that fuel elements adjacent to or in
contact with the soil will also be at low moisture levels.
Daily Rainfall (mm)
30.0
25.0
20.0
15.0
10.0
5.0
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
December 2015
Cape Otway (090015)
Lorne (90185)
Aireys Inlet
(90180)
22.6
Cape Otway
(90015)
35.4
Lorne
(90185)
23.4
39.8
13.8
18.7
20.6
35.8
56.4
72.4
120.2
52.9
2.6
14.3
20.3
45.8
90.1
106.4
153.7
71.4
15
21.3
23.4
65.6
126.7
154
204.4
10
On 25 December, the temperature in the Wye River area reached approximately 36C at around 16:30, with
a relative humidity of 17% (Figure 9). Wind was blowing from a north/north-westerly direction with gusts of
nearly 33 km/h. Based on temperature, relative humidity, wind speed and KBDI data, the FFDI reached a
peak at 16:30 of 49 in Aireys Inlet and 43 in Cape Otway (Figure 14). At a FFDI of 49 the weather condition
are considered to be at a fire danger rating of Very High.
Figure 14: FFDI on 25 December 2015 based on the Aireys Inlet and Cape Otway AWS data.
11
Figure 15: Time of arrival as derived from thermal line scan imagery.
Separation Creek was the first area impacted by fire. The fire arrived in the Separation Creek township by a
number of point ignitions from airborne embers. The fires that developed from the ignition points grew
within the township with a tendency to spread upslope and back towards the advancing fire front.
At 15:07:40 five spot fires were active within Separation Creek. One of these developed below the corner of
Bass Avenue and Harrington Street, and a further four developed in the vicinity of Mitchell Grove. These
spot fires are indicated by the blue arrows in Figure 16.
12
Figure 17 is a similar image one minute later showing the development of these spot fires and a further
break out (marked with a blue arrow) in a vacant block on Olive Street approximately 400 metres ahead of
the advancing fire front. The fire below the corner of Bass Avenue and Harrington Street was in a densely
vegetated gully. The fire spread rapidly upslope towards the approaching main fire front. It appears that
the smoke from each of these fires was drawn east towards the approaching fire front.
13
Figure 18 shows the same scene seven minutes later. All of the spot fires developed rapidly and were
drawn back towards the advancing fire front. The fires near Mitchell Grove appears to also have developed
rapidly. A new spot fire has developed adjacent to Bass Avenue and is marked by a blue
arrow. The spot fire identifed by the blue arrow in Figure 17 adjacent to Olive Grove has developed the
most rapidly most likely due the a cobmination of slope and vegetation structure.
14
15
Figure 20 is the second line scan taken 55 minutes later. The scan shows that the fire front moved past
Separation Creek, with heat signatures evident from many of the burning houses in Separation Creek. Five
additional spot fires seem to have developed within Separation Creek that appear to have led to some
house damage but no complete losses, possibly due to the fire-fighting activities around these isolated spot
fires. The main heat signatures are from the combustion of houses within Separation Creek, indicating that
nearly all houses lost in the region are at this time ignited. The two regions where extensive house losses
occurred would most likely have been too difficult to access due to the large number of simultaneous
house fires in close proximity.
16
This line scan (Figure 20) also shows the Wye River township in the middle of a fine fuel burn through
phase. The combination of ground observations and this image suggests that the area was impacted with
embers emanating from behind the township, creating many simultaneous ignitions of surface fuels
throughout the fire-affected areas of the Wye River township. This mass ignition and rapid fine fuel spread
would have presented a major hazard to fire fighters and residents (if present) in this region. The foreshore
area below Iluka Avenue (marked with a blue circle) appears to have developed from a series of spot fires
that merged and ran upslope into and past houses along the street.
Figure 21 is a pre-fire image of this region highlighting the extent and separation distance between the
houses and vegetation. The slope in this coastal vegetation is such that flame fronts formed against the
prevailing wind with enough intensity to generate radiation levels and possible flame contact sufficient to
ignite houses.
17
Figure 21: Image showing the location of Iluka Avenue and coastal vegetation ignited by embers.
Figure 22 shows an FLIR image take at 18:00, approximately one hour and fifteen minutes later. This image
was used in the absence of any available line scan data during this period. The image shows extensive heat
signatures in the areas of house loss in the Separation Creek and Wye River townships. The completion of
fine fuel burn out appears to have occurred in some areas of the Wye River township, while other areas
show broad scale fire activity. The region below and along Iluka Avenue shows strong fire activity,
suggesting that all houses were well alight while fuels along the foreshore were still burning out. It should
be noted that the FLIR image in Figure 22 provides a more sensitive heat signature than in the line scan
images provide before (Figure 20) and after (Figure 24).
Figure 22 also identified fire progression to involve some additional houses in the westerly extent of the
image and new spot fires forming in this region, indicated by a rectangle.
The FLIR image in Figure 23 provides a close up of a spot fire within the blue circle in Figure 22. This isolated
spot fire did not develop further to impact houses and may have been supressed by aerial or ground crews.
Figure 23 also shows a number of airborne embers, which had already travelled more than two hundred
metres from nearby fire fronts.
18
Figure 23: 18:01 Greyscale forward looking infrared (FLIR) imagery showing a spot fire igniting behind a row of
houses. Darker colours are cooler (e.g. water bodies) and white colours represent significant heat.
19
Figure 24 was taken one hour later at 19:00. The scan shows clear heat signatures from nearly all burning
houses in Separation Creek but the more defined and isolated patches indicate a reduction in overall
intensity.
The fine fuel burnout phase had also passed the Wye River township except in the westerly extent, where
there is a heat signature over a broad area including areas where houses were lost; these may have already
ignited or were in the process of being ignited. At this stage nearly all of the fire affected houses in Wye
River appear to be either ignited or were in the advanced stages of burning out.
Figure 25 was take at 19:32, 32 minutes later. This shows that all fine fuel burn out was complete with heat
signatures present for nearly all houses that were lost in the fire event. These heat signatures continued to
reduce in size except for those in the westerly extent of the image.
Figure 26, Figure 27 and Figure 28 show the progressive completion of burn out of all structures.
20
21
Figure 29: Digital elevation model for the townships of Wye River and Separation Creek showing house damage
(based on Shuttle Radar Topography Mission (SRTM) data).
Fire ignited by spotting spread uphill and more slowly spread laterally across the slope, supported by
surface and near surface fuels. Leaf freeze observations throughout the affected areas confirmed that the
dominant fire direction was upslope, indicating that below the canopy the convection force of the fires was
stronger than ambient wind effects. There were localised areas below Iluka Avenue Wye River and along
23
creek lines where flames involved elevated fuels, supporting significant flame heights. Other than these
areas, the average solid body flame heights within the township appeared to be typically less than
0.4 metres. As detailed in Section 5 the fine fuel burnout phase followed a mass spotting event hence
rapidly involved an extensive area without any clear indication of fire arrival and progression. This
circumstance would make it particularly hazardous for a resident to plan an exit path for ground crews to
safely or effectively manage fire ignitions and subsequent spread.
Ground observations identified that most houses were impacted by either surface fire or the subsequent
heavy fuel element burn out. The ground fire spread involved surface fuels and near surface heavy fuel
elements. Generally, the solid body flame height appears to have been up to 0.4 metres from these surface
fuels. Some secluded gullies within the township burnt with higher intensity, but these fire runs did not
impact buildings. The houses along Iluka Avenue appear to have been the only houses affected by a higher
fire severity from adjacent vegetation. As the extensive fine fuel fires ignited heavier fuel elements and
houses the risk of flame impact, radiation exposure and smoke exposure would not have abated and in
many areas would have increased in severity.
Many examples of destroyed timber retaining walls surrounded impacted houses (Figure 30). Combustible
materials were an obvious factor in their low survival rate, as were the low soil moisture levels described in
Section 2. It is likely that combustible retaining wall elements in contact with soils were also low in
moisture content at the time of fire arrival, leading to a higher likelihood of ignition, more severe
combustion and more extensive timber burn out.
Where timber (in particular treated pine) retaining walls were in contact with or within a few metres of a
building, their combustion is likely to have contributed to house loss. Figure 30 provides an example of a
large timber retaining wall that was burnt out.
Figure 30: Driveway off Karingal Drive, Wye River. An example of burnt out timber retaining wall sleepers.
24
Where treated pine is used there is also the risk of toxic smoke emissions during the fire and toxic ash
residue on the ground and blown by the wind after the fire. Timber retaining walls also provided a direct
threat to buildings, or subsequent ignition or heat exposure to other adjacent elements, such as LPG
pressure vessels. LPG pressure vessels are common in the townships and generally were observed to have
vented, as they are designed to do when overheated. There were a few instances where these cylinders
appear to have vented towards structures and are likely to have contributed to the loss of the house. An
example is provided in the case study in the appendix in Section A.2.4, where pressure vessels vented
towards the house.
Strong evidence existed of house-to-house fire spread. This is not unusual once a fire impacts a township
e.g. [7]. However, despite the generally large distances between buildings in the fire-affected area, houseto-house ignition appears to have still been a significant factor. There was clear evidence of damage and
loss where houses up to 12 metres apart, including indications of high radiant heat levels. These distances
are greater than is typically found in other post-bushfire studies. A key factor in house-to-house spread at
these greater distances was due to the low wind speeds in the townships during and after the fire event.
Houses at these distances emit radiant heat to adjacent houses, causing them to increase in temperature.
The ambient winds act to cool the houses and limit the peak temperature of the adjacent houses. The
lower wind speeds in this fire event were less effective at providing this cooling effect.
Fire in the heavier fuels stored under and adjacent to the houses subfloor area such as plastic water tanks,
building materials, small garden sheds, boats and kayaks, also appears to have been a significant factor in
many of the losses.
While extensive surface litter provided a near continuous flammable fuel bed, the built elements within the
township represented the vast majority of the fuel load. Some of these built elements ignited by the initial
spread of fire through surface litter fuels. The built elements then continued to burn for many hours,
igniting other built elements. The extensive number of houses burning in the fire affected area would have
meant that these areas would have been life threatening for any residence present as well as fire fighters.
To summarise, the primary progression of spread mechanisms involved:
long distance spotting from the fire front outside the perimeters of the townships
ignition of surface fuels below structures within the townships
surface fire spread through fine surface fuels and near surface vegetation onto built elements.
It is interesting to note that localised ember spread within the townships was not as prevalent as other
surveyed bushfire events involving house losses of more than 100 houses. This may be due to the relatively
low wind speeds within the townships at the time of fire activity. The low wind speeds also appeared to
exacerbate the prevalence of house-to-house ignitions at distances previously considered sufficient. These
spread mechanisms supported the initial progression of fire within the townships and provided flame
contact as follows:
interaction between fine surface fuels and heavy fuel elements adjacent to houses
interaction between fine surface fuels and combustible elements on the houses themselves
interaction between fine surface fuels and LPG pressure vessels providing the potential for gas
flares and explosions.
Heavy fuel elements then interacted with each other in the advanced stage of fire development within the
township through the following mechanisms:
flame or radiant heat transfer to LPG pressure vessels providing the potential for a gas flares and
explosions.
25
The interaction of fire with established tall trees also increased the risk of tree and branch strike because
fire weakens knots and flaws in trees. Branch or tree strike, either on buildings or across roadways, is a
common risk during and after bushfire arrival. While there was no direct evidence of tree strike onto
houses subject to recent regulatory standards, evidence of direct tree strike either during or in the days
following the fire event were apparent with some houses (Figure 31).
Figure 31: A house on Wallace St Wye River, subject to tree strike cause by fire interaction at the base of a tree.
A detailed set of case studies of fire impacted regulated houses is provided in Appendix A
26
There was also evidence of extensive aerial water bombing activity throughout the affected areas.
Indications included water wash marks (Figure 33 and Figure 34) on the landscape surrounding the houses
and limited examples of house gutter deflection and other minor roof damage due to a sudden deluge of
water.
27
Figure 33: Silt washed onto decking from an adjacent water drop
28
Figure 34: Vegetation between houses showing signs of water impact from water bombing
As detailed in Section 5, the conditions within the township presented a risk to any occupants present both
during the initial fine fuel based fire spread and the following heavy fuel burn out phase. This risk would
have provided many challenges to fire crews active in the township. The fine fuel based fire spread
provided risks to crew of crew vehicle burnover on the narrow roads cut into the terrain, and the heavy fuel
burn out phase involved an extensive number of houses with life-threatening radiant heat levels and toxic
smoke at ground level. Steep slopes promoted smoke plume attachment to the ground, presenting a risk to
fire crews attempting to use roads above houses that were actively burning.
29
8.1 Regulation
For context the paragraph below is the opening paragraph of the forward for AS3959-2009 [8].
This standard is primary concerned with improving the ability of building in designated bushfire-prone areas
to better withstand attack from bushfire thus giving a measure of protection to the building occupants (until
the fire front passes) as well as to the building itself.
There were approximately 80 planning referrals to the CFA in Wye River and Separation Creek, from as
early April 2003. Both building and planning requirements have evolved significantly from 2003 to the
present. There is a reasonable level of complexity in categorising these referrals to determine whether they
resulted in an actual building and/or planning outcome. The nature and stringency of these outcomes has
evolved over time, ranging from the requirement for stored water supplies to integrated sets of building
and planning requirements.
These requirements can be broadly categorised into two categories: those that were built since the
introduction of AS3959-2009 [8] revisions and those prior. To gauge the relative effectiveness of the two
categories, a review of past building approvals was undertaken to determine which houses were
completely built or upgraded to meet these regulatory standards. The review identified seven houses built
to the AS3959-2009 [8] standard and fourteen to prior regulatory standards which were complete to the
point of obtaining a certificate of occupancy. Other houses may have been largely complete but are not
included in the analysis as minor construction details can have a major impact on building performance in
bushfires. Of particular interest is the apparent performance of houses in these two categories as shown
Table 2 below.
Table 2: Survival outcomes for AS3959-2009 and houses built to prior regulatory standards.
AS3959-2009
Prior to AS3959-2009
Destroyed
4
3
Surviving
3
11
For both categories there appears to be a marginal improvement over the 80% loss rate experienced across
the fire-affected area generally. The AS3959-2009 [8] category had a 57% loss rate and the category prior to
these regulatory standards had a 21% loss rate. Given the low numbers in the sample this should be treated
as a rough indication of the survival prospects of houses built to the recent regulatory standards under
these exposure conditions, rather than as a definitive measure of performance. It is recommended that a
detailed review of the changes around the construction provisions of subfloor and other near ground
building elements be conducted to understand whether this could have affected the outcomes for these
buildings.
BAL-29 allow combustible stumps, bearers, flooring, decking, stair and balustrades within close proximity to
the ground. These elements were either directly threatened by fire spread through typical levels of fine fuel
and grasses within the townships or ignited by typical heavy fuel elements that resided under or adjacent
to the buildings. The typical elements included retaining walls, stored materials, vegetation, plastic water
tanks and vehicles.
Some of these weaknesses are only specifically addressed in BAL-40 and BAL-FZ (flame zone) construction
levels, which specify that heavy fuel elements should not be located under or adjacent to BAL-40 and BALFZ buildings.
Resistance to heavy fuel element combustion is not formally recognised as a fire load in current bushfire
related building codes. BAL-FZ construction is the most likely to resist this heavy fuel exposure, and the
higher BAL level in the newly adopted National Association of Steel-Framed Housing (NASH) building
standard offers some design redundancy to improve resistance to this exposure.
other shelter in place contingencies such as personal fire shelters with extended periods of
occupancy
alternative ways of specifying likely sight exposure for the given terrain and landscape factors.
the separation distances between buildings within the township to limit structure-to-structure
spread
31
9 References
1. Community Update, Lorne Jamieson track fire, Issued: 21/12/15 14:05.
http://emergency.vic.gov.au/public/event%2Fincident%2F1552223%2Fwarning%2F21655327.html (accessed
March 2016)
2. "Great Ocean Road fire: Number of homes lost in Christmas Day blaze rises to 116". ABC News (ABC). 26
December 2015. Retrieved 26 December 2015.
3. Blanchi, R., & Leonard, J. (2006). A study of the interaction between bushfire and community in the South
Australian Wangary fire 2005. Melbourne: CSIRO report to Bushfire CRC.
4. Leonard, J., Blanchi, R., Leicester, R., Lipkin, F., Newnham, G., Siggins, A., Barwick, M. (2009). Building and Land
use planning research after the 7th February 2009 Victorian bushfires. Preliminary findings. Melbourne: Interim
report USP2008/018 - CAF122-2-12 . Retrieved from http://www.bushfirecrc.com/managed/resource/bushfirecrc-victorian-fires-research-taskforce-final-report.pdf (accessed March 2016)
5. Bureau of Meteorology, Monthly Summary for Australia, Product Code IDCKGC1A00, Tuesday 5 January 2016
(2015), http://www.bom.gov.au/climate/current/month/aus/archive/201512.summary.shtml (accessed March
2016)
6. Keetch, John J. and George M. Byram. "A drought index for forest fire control". (1968). U.S. Department of
Agriculture, Forest Service, Southeastern Forest Experiment Station. 35 p.
7. Leonard, J., & Blanchi, R. (2005). Investigation of bushfire attack mechanisms involved in house loss in the ACT
Bushfire 2003. CSIRO Manufacturing & Infrastructure Technology.
8. Standards Australia. (2009). AS 3959-2009 (Amend. various) Construction of buildings in bushfire prone areas.
Standards Australia.
32
Appendix A
A.1
A.1.1
KEY LEARNINGS
This house on Karingal Drive was completed in February 2013 to BAL-40 construction requirements. It
survived with damage to its decking and decking support structure. The main threat to the house and
decking was from the combustion of treated pine retaining walls adjacent to and below the structure and
deck. The houses steel support structure and non-combustible subfloor, cladding, window frames and
doors were effective in resisting ignition in combination with aerial suppression activities.
Even though the fire arrival context was far below the worst-case exposure scenario assumed for BAL-40
construction, the fuel elements adjacent to the building may have been sufficient to cause building loss if
aerial suppression did not occur.
A.1.2
HOUSE LOCATION
The house location is indicated by the yellow dot at the tip of the blue arrow (Figure 35).
33
A.1.3
HOUSE DESIGN
This house was built to the BAL-40 construction requirement. It is of modular construction, utilising steel
for its supports and internal framing. The cladding is also metal, as is the window framing material. The
underfloor area of the house is not clad, and instead uses non-combustible James Hardie Scyon flooring
sheets (see Figure 36).
Figure 36: Under floor system, steel and James Hardie Scyon flooring.
A.1.4
This house was exposed to a low-level surface fire, which followed the slope moving up to and past the
building (Figure 37). This location was impacted between 15:30 and 16:45 by a fine fuel burn through.
34
Figure 37: Fire arrival upslope, involving some dead tree stumps.
This low-level surface fire spread ignited various heavy fuel elements such as retaining walls adjacent to the
decking as shown in Figure 38.
35
During active burning of these retaining walls, aerial water bombing drops washed down over the retaining
walls and under the building. This water bombing appeared to be effective at supressing the burning of the
retaining walls and limiting the duration and intensity of flame exposure on the buildings and attached deck
(Figure 39).
Figure 39: Deckretaining wall interaction with evidence of water wash from aerial suppression.
The BAL-40 compliant decking was made of a plasticwood composite with fire retardant additives. This
showed signs of charring where flames had impinged on the decking boards. The boards did not appear to
have encouraged flame spread. The deck was supported by galvanised steel bearers and posts, which were
effective in supporting the decking structure and building throughout the fire event (Figure 40).
36
Other than the retaining walls, the only other significant combustible materials found under the house
were two full polyethylene water tanks (Figure 41). While the PVC piping that was not full of water
suffering some deformation, the polyethylene tanks were not sufficiently close to other stored materials or
the combustible retaining wall to receive direct flame contact.
37
Some ember attack was evident as shown in Figure 42. The decking and support structure appeared to be
effective in retarding flame development from the ember attack.
The BAL-40 compliant windows utilised a flexible pull down fire shutter as shown in Figure 43. These
shutters were not deployed during fire exposure but radiation levels were not sufficient to cause damage
to the exposed windows.
The inclusion of gutter guards and a simple roof profile also appeared to limit the likelihood of a roof
ignition (Figure 44).
38
39
A.2
A.2.1
KEY LEARNINGS
This house may have ignited from a variety of processes, if ember attack, including low-level fire spread, did
not directly ignite the house it is likely that the combustion of the decking and retaining walls, and LPG
pressure vessel flaring, would have caused ignition of the house.
A.2.2
HOUSE LOCATION
This house was located at the tip of the blue arrow in Figure 46.
A.2.3
HOUSE DESIGN
This house was completed in August 2005 to Wildfire Management Overlay (WMO) requirements after
being categorised as Medium level of Bushfire Attack in accordance with AS3959-1999. 5 The key
requirements are a static water supply of at least 10,000 litres and vegetation management in an inner and
outer zone in accordance with the WMO. This vegetation management consists of an inner zone forming a
Standards Australia (1999) Australian Standard AS 3959 Construction of buildings in bushfire-prone areas, Standards Australia
40
10 metre perimeter around the house clear of elevated fuels no greater than grass less than 100 mm, and
litter fuels less than 20 mm in depth. At least 50% of the outer zone must be clear of elevated fuel.
Figure 47 provides and indicative photo of the house design and surrounding vegetation elements in the
years preceding the fire event. The house appears to have had a non-combustible facade and metal clad
roof. It also appears to have had treated pine retaining walls running up to and under the house.
Figure 47: Indicative photo from Durimbil Avenue at February 2010 (Google Street View).
A.2.4
This house was exposed to a low-level surface fire, which followed the slope moving up to and past the
building. This location was impacted between 15:30 and 16:45 by a fine fuel burn through. The fine fuel
spread appears to have occurred below the shrubs, which received some leaf scorch as it passed (Figure 48
and Figure 49).
41
Figure 48: House on Durimbil Avenue showing intact elevated vegetation adjacent to building.
This fine fuel burn through appears to have ignited treated pine retaining walls and timber decking
elements adjacent to the house (Figure 50).
42
These retaining walls also burnt in close proximity to four 65 kg LPG pressure vessels (Figure 51). These
pressure vessels stood on a poured concrete slab against a metal frame, which supported the copper pipes
attached to the pressure vessels. The fire in the adjacent retaining walls was sufficient to heat the pressure
vessels, causing the pressure to increase and flare from the safety release of some of the pressure vessels.
43
The flaring from these pressure vessels pointed directly towards the house (Figure 52).
Figure 52: House on Durimbil Avenue showing house and location of pressure vessels.
44
A.3
A.3.1
KEY LEARNINGS
This house appeared to have not been exposed to an extensive attack from adjacent vegetation, instead it
was subject to surface fire approaching 0.4 meters in height and ember attack. The house was required to
be built to BAL-40 construction requirements. The approach taken to comply to BAL40 was unusual and
may represent a key weakness in AS3959-2009 [8]. The construction approach was to externally cladding
the houses timber frame in plasterboard to achieve a compliant fire rated wall system. An additional
external cladding of stained cedar was then used over the plasterboard to create a very aesthetically
pleasing a exterior finish. The use of combustible cladding over the fire rated wall appears to be allowed by
the standard [8] and is currently opening promoted as a compliance solution on various websites including
the following:
http://www.5startimbers.com.au/downloads/Design_Guide_04_Building_Timber_Bushfire_Areas_3-9_MB.pdf
A.3.2
HOUSE LOCATION
This house was located at the tip of the blue arrow in Figure 53.
45
A.3.3
HOUSE DESIGN
This house was completed in October 2014, was subject to BAL-40 construction requirements. It comprised
steel posts, aluminium framed windows, a non-combustible clad subfloor and cypress pine timber cladding.
BAL-40 compliant houses dont not typically have timber cladding due to the inherent ignition risk during a
bushfire. This house however was constructed using an approach which consists of timber frame which is
then clad in plasterboard in a way that achieves a fire rated construction level for the wall. A cladding of
timber (cedar) was then placed over the plasterboard with the assumption that this would not compromise
the fire rating of the plasterboard wall system. It appears that this system does meet the compliance
requirements of AS3959-2009 [8] but unfortunately is not likely to survive a bushfire given that the
combustion of the wall cladding will provide sustained flame contact on all adjoining building elements. At
BAL40 these adjoining elements are rated to withstand radiant from a passing bushfire and not this level of
sustained flaming. Pre-fire Photo examples of this house could be found at the following links at the time of
writing this report:
https://www.facebook.com/tuckerhousewyeriver/ (accessed March 2016)
Combustible elements below the subfloor appeared to consist mainly of treated pine retaining walls and an
old tree stump (Figure 54).
46
Figure 54: House on Koonya Avenue showing an old tree stump under the floor on the right of the picture.
The floor system including the decks was supported by steel piers and major steel beams. Timber minor
beams were evident under the non-combustible subfloor cladding (Figure 55). The decking surface was
non-combustible.
The predominant ground cover adjacent to the retaining walls was a combination of leaf litter and wood
chip mulch. The walls and ground cover extended up to and under the house. There were numerous points
of potential interaction between the retaining walls and the combustible facade. The combustible fine fuels
in the various crevices would have been readily ignited by embers and surface fire spread. This, in turn,
could have readily ignited the retaining wall sleepers and building facade. These two elements would be
able to transmit heat to each other and increase the rate of combustion until the facade was burnt
through. The fire could then enter the house via the windows, doors roof system and ignite the building
47
contents. If the house was occupied at the time of fire impact it is possible that the house could burn
externally with sufficient severity to prevent the occupants from safely leaving the house. As the houses
internals ignite, the house may have provided an entrapment for the occupants if they were present. The
non-combustible decking surface would have improved the prospects of escape however the combustible
cladding and decorative combustible panelling could have provide a life threatening level of heat to those
trying to escape.
Observations of the wreckage following the fire offers no definitive clues to the cause of initial fire ignition
and spread. It does provide sufficient evidence to rule out other process such as tree impact and vehicle-tohouse fire interactions. Figure 56 shows a view of the house following the fire, this view reveals the
connection between the combustible building facade and combustible elements at ground level.
A.3.4
This house was exposed to a surface fire of flame height approaching 0.4 metres, which followed the slope
moving up to and past the building (Figure 37). This location was impacted between 15:30 and 16:45 by a
fine fuel burn through. Fine fuels close to, under and against the house mainly consisted of bark mulch,
dried grasses and windblown fine litter fuels. This ground fire attack would have provide a likely ignition of
combustible elements adjacent to and under the house.
48
A.4
A.4.1
KEY LEARNINGS
This house contained many design aspects that could resist radiation and ember attack from an
approaching bushfire. The actual bushfire approach was a surface fire of less that 0.8 meters that did not
directly impact the house. The main threats to house loss would have been presented by the extensive use
of timber retaining walls under and adjacent to the house as well as the combustion of the neighbouring
house which was less than 3.5 meters away.
A.4.2
HOUSE LOCATION
This house was located at the tip of the blue arrow in Figure 57.
49
A.4.3
HOUSE DESIGN
The house was completed in 2012 and built to BAL40 construction requirements under AS3959-2009 [8].
The house was timber framed and clad in Colorbond steel. The roof was also clad in Colorbond steel. Decks
adjacent to windows were supported by timber posts and bearers and clad with a composite decking
product. At the driveway entrance a deck comprising of steel bearers and timber cladding was used, this
deck was made to support vehicle access to the garage and served as a parking area.
Extensive use of timber retaining wall was evident around and under the building see Figure 58 & 59. A
polyethylene water tank was also present under the building immediately adjacent to timber retaining
walls (Figure 60). This tank and adjacent retaining walls had collapsed and burnt out. This may have been
either a cause or a result of the house fire. If it occurred prior to the house burning it would have present
two risks to the building:
-
By providing a direct fire impingement to the houses sub floor and wall system
By providing a direct impact to the facade and windows immediately below the tank as the tank
ruptured
The driveway timber deck and adjacent polyethylene water tanks burnt extensively and may have present
an egress issue if the house was occupied as it appear that the only way to exit the property onto the street
is via to pass over of past these elements Figure 61 & 62.
Figure 58: House on Iluka Avenue Wye River, showing vegetation condition below structure
50
Figure 59: 23 Location of burnt out timber retaining walls below and beside structure
Figure 60: 23 Burnt out retaining walls and tank adjacent to house.
51
Figure 62: 23 Interaction between the timber driveway deck and a polyethylene water tank. It is possible that the
rupture water tank split out over the timber driveway and prevent the complete combustion of the deck.
52
A.4.4
This house may have been subjected to multiple fire approaches. Figure 20 shows a line scan at 16:45
indicating that fire had approached for the east via spotting of surface fire spread along the line of houses
below Iluka Avenue. Figure 22 provides a FLIR image which indicated that the coastal scrub below this
house was burning at the time the image was taken. This house was also build within 3.5 meters of its
nearest neighbour to the west which was a house that also burnt and appear to have been built prior to the
period where building or planning requirements for bushfire would have been present.
The coastal scrub below the house showed indications of a surface fire which was sufficient to scorch the
foliage but not remove it in the upper portion of the coastal scrub. This suggested a mean flame height of
0.8m arrived in the adjacent vegetation. The maximum distance the fire could run up this slope was 41m
being the distance between the road below and the edge of the adjacent vegetation below the house. The
estimated separation distance between the continuous vegetation below the house was 3-4 meters.
53
A.5
A.5.1
KEY LEARNINGS
This house had a range of features (Figure 65 to 72) which could have supported ignition and development
to totally involve the house. Even though the subfloor area was readily accessible to ember entry and
contained elements which were readily ignitable there were no signs of ember arrival in this sub-floor area
of immediately adjacent to it. The surface fire present a significant threat of direct ignition of the
combustible near ground features, however surface fire spread did not reach the building, most likely
because of suppression activities. The relatively flat ground on all sides of the building in combination with
a concrete apron around two sides of the building could have provided an area for ground crews to stand
and defend the building. There were a number of signs that suppression crews defended this property,
indicated by water wash marks and supressed surface fire (Figure 70 & 71).
A.5.2
HOUSE LOCATION
This house is located at the tip of the blue arrow in Figure 63.
54
A.5.3
This house was completed in April 2006 and was categorised as Medium. The building is clad with two
different systems, Colorbond steel and painted timber weatherboards. The house has a timber subfloor
that is semi enclosed with gapped timber boards (Figure 66 to 68). The gaps between the battens is in
excess of 6mm.
Figure 64: Google Street View image of the house some time before the fire event.
55
Figure 66: Concrete apron, house to surface fuel orientation, and surface fuels present after the fire.
56
57
Figure 69: House to tank proximity and adjacent combustible elements including rubbish bins and firewood.
Figure 70: Some fire interaction between a burning tree trunk and an adjacent polyethylene water tank. The ground
surface and tree bark condition indicate the suppression had occurred in this area.
58
Figure 71: 33 Bare earth interface adjacent to sub-floor area. The dried earth on the sub-floor boards indicate water
suppression in the area.
Figure 72: 33 Elevated deck with no clear signs of radiation or ember attack.
A.5.4
This property allotment experience a surface fire of less than 0.4 meters over the majority of the allotment,
there were also extensive patches of unburnt surface fuels most of which appeared to be a result of
suppression activities. There were no burnt patches against the house suggesting that the ground crews as
well as aerial suppression may have been present.
This location was impacted between 19:00 and 19:32 by a fine fuel burn through.
59
A.6
Airborne thermographic (or infrared) cameras are used operationally by fire agencies to see through smoke
to observe and reconnaissance on the fire location, size, path and localised hotspots. Instead of the 400
700 nanometre range of the visible light camera, infrared cameras operate in wavelengths as long as
14,000 nm (14 m). The resulting image is greyscale where cooler radiant temperatures are darker and
warmer radiant temperatures are lighter (Figure 73). Note that some of the very bright features in the
image are not related to the fire. These include heated bare ground such as roads, beaches and breaks in
vegetation. Very bright regions that do relate to fire include active flame, smouldering combustibles and
heated air.
Images are stored in 8-bit format, meaning that values range from 0 to 255. The image values are often
referred to as digital numbers. The surface measurement is derived through calibrating an image to
radiance values, in this case it would be kilowatts of radiance per meter squared. Calibration of these
digital numbers to black body radiant heat (Kw/m2 ) values is complicated due to the influence of many
environmental effects. The altitude of the aircraft alone has a significant effect on brightness temperature.
Radiometric calibration of thermal infrared requires corrections relating to atmospheric absorption and
emission (Figure 74) and surface emissivity effects. Regression analysis and field sampling would be ideal
for calibrating the radiance received by the sensor and thermal radiance at the surface. The nature of a fire
makes it unsafe and impractical to take a representative measurements of surface radiant temperatures at
the time that the image is captured.
60
Figure 74: Effects of the atmosphere and some of it components on waves of electromagnetic radiation. Source:
http://wattsupwiththat.com/2011/02/28/visualizing-the-greenhouse-effect-atmospheric-windows/
High altitude fire line scan imagery was captured by the Victorian Aviation Services Unit at irregular times
to observe the progression and nature of the fire in southern Victoria on 25 December 2015. An analysis of
this imagery found that using a colour threshold was the best way to determine the spatial and temporal
characteristics and nature of the fire approach on the towns of Separation Creek and Wye River. We found
that most information came from the saturated pixels; these indicate a high heat emittance (Figure 75).
61
Figure 75: Saturated pixels of a line scan image captured 25/12/2015 15:50 overlayed on a visible colour aerial
image.
Artefacts such as smouldering fire and clouds of hot air from convection are eliminated from the image
when only saturated pixels are shown. Evidence of this finding is shown in Figure 76, where it is visible
where highly combustible elements in the environment (such as houses) continue to burn while the fine
fuels have since burnt out.
62
Figure 76: Comparison of greyscale (left) and only the maximum image values (right) overlayed on visible colour
aerial imagery
A helicopter mounted fire line infra-red (FLIR) video recorder was also flown during the fire. Frames from
this footage were extracted and where possible, georeferenced to observe in detail any footage that
captured the progression and nature of the fire. Due to the high resolution of these images it was deemed
unnecessary to threshold the colour to saturated pixels. See Figure 77 as an example of a georeferenced
frame from the FLIR footage captured.
63
Figure 77: Example of a georeferenced frame from the FLIR footage capturing the entry of the fire into Separation
64
65
AN EVALUATION OF INSPECTOR-GENERAL
FOR EMERGENCY MANAGEMENT [IGEM]
REVIEW OF THE WYE RIVER JAMIESON TRACK
FIRE AND RELATED MATTERS.
WHAT ARE THE REAL LESSONS LEARNT?
4 April 2016
Athol Hodgson AM. BScFor. DipFor. has more than 60 years experience in fire management and
forest fire research in Australia, USA, Canada, France, Spain and China. He was formerly
Commissioner of Forests, Forests Commission of Victoria and Chief Fire Officer Department of
Conservation, Forests and Lands, Victoria. He was a Member of the Board of the Country Fire
Authority and a Member of the State Disaster Committee. He was awarded a Winston Churchill
Fellowship to study fire management in North America and is a graduate from the National
Advanced Fire Behaviour School, Marana, Arizona. He was made a Member (AM) of the Order of
Australia in 2012 for service to forestry science, particularly the development of land management
and bushfire risk reduction strategies, to emergency service organisations and to the community of
rural Victoria.
R A (Rod) Incoll, BASocSci, GradDipBus, DipFor (Vic), AFSM, developed expertise in the use of
prescribed fire and in bushfire management with the Forests Commission, Victoria in six forest
districts across the State. Following a period with the State Electricity Commission 1984 - 1989, Rod
was appointed Chief Fire Officer of the Department of Conservation and Natural Resources, Victoria.
He was a member of the Board of the Country Fire Authority, the State Emergency Services Council,
and a foundation Director of the Australasian Fire Authorities Council. He worked closely on aviation
support for forest fires with North American agencies. Rod is currently an independent risk
consultant specialising as an expert witness in bushfire and forest injury matters.
Summary
This evaluation of the Inspector-General for Emergency Managements review of the Wye RiverJamieson Track fire exposes on-going serious problems concerning Victorias policy and practice for
forest fire management (the prevention, suppression and planned use of fire in the landscape) that
were not effectively addressed in the IGEMs Review.
The evaluation draws attention to:
Terms of reference did not meet the test of a truly independent review;
There were no independent contributors to the review who had extensive experience in forest
fire management on public land;
Key underlying circumstances that influenced the conduct of the bushfire, which are within the
IGEMs charter under section 64 of the Emergency Management Act2013, were not considered;
and
Fundamental lessons from previous bushfire response; involving matters of - culture, leadership,
organisation and priorities governing fire suppression, reinforced by the Wye River-Jamieson Track
fire, were either ignored or dismissed.
The evaluation considers why the initial attack on the bushfire failed, the consequence of which
produced a serious threat to all priorities under the States Strategic Control Priorities.
Notwithstanding the important redeeming feature that there was a highly successful evacuation of
threatened Surf Coast communities and that the number one priority of preservation of life was
fortuitously met, the evaluation concludes that the fire was a disaster that most likely could have
been avoided.
The real lessons to be learnt are documented together with recommendations for addressing them.
Contents
About the authors. ............................................................................................................................ 2
Summary
1.
Preface. ......................................................................................................................... 5
2.
3.
Evaluation of IGEM Review of the Wye River- Jamieson Track Fire and Related Matters. 8
3.1.
Ministerial request to the IGEM for a review of the bushfire including the scope and
terms of reference/key matters to be considered. ......................................................... 8
3.2.
IGEMs responsibilities under Section 64 of the Emergency Management Act 2013 and
other matters influencing the conduct of fire suppression. ......................................... 9
3.3
3.4
Discussion .................................................................................................................... 14
3.5.
Conclusions. ................................................................................................................. 17
4.
5.
6.
Appendixes. ................................................................................................................. 21
Appendix 1. Ministers request for IGEM Review of Wye River-Jamieson Track fire. .............. 22
Appendix 2
Appendix 3
Appendix 8. Extract from Changes in forest fire management in south-eastern Australia: for
better or worse? Mike Leonard. October 25, 2014 in Part 3 of Fire Management on
Public Land - Victoria Burns While Its Bureaucracy Fiddles. Will Recent Legislative
Changes to Crisis and Emergency Management Improve Fire Management on Victorias
Public Lands? Dexter, B.D; A. Hodgson AM. January 2015. ISBN 978-0-9942531-0-1 ... 45
1.
Preface.
Since 2002 nearly half of Victorias native forest on public land (over 3 million hectares or some 15%
of the State) have been burnt; an unprecedented amount over such a short period in the States
bushfire history since European settlement. Many fires which originated on public land have had
substantial impacts on urban, regional and rural communities at the public/private land interface.
The complexity of forest fire management (the prevention, suppression and planned use of fire in
the landscape) has significantly increased over the past three decades. Notwithstanding the wideranging nature of the Royal Commission into the 2009 Victorian bushfires and governments
acceptance of the Inquirys recommendations, seven years on, fundamental problems have not been
addressed, effectively or efficiently.
For example:
Successive governments have demonstrably failed to maintain forest fire management as a
primary core business of the State;
The level of expertise and resources available to combat the bushfire threat remains in a
worse condition that at any time since the early 1980s;
Numerically depleted/strategically located work centres and lack of experienced and
accredited personnel have diminished fire prevention, particularly fuel reduction burning
and bushfire first attack to quickly bring bushfires under control.
Inappropriate development in naturally and artificially created bushland settings; without proper
regard to the known, but largely ignored, risk created by high fuel loads and highly flammable
vegetation on adjacent public land and dwellings closely intermingled with bushland on private
property, life and property is in grave danger of incineration.
Rampant empire building continues as Government Ministers and the responsible Department
scramble to reinvent themselves in response to public concern and damning internal and
independent reports.
Over the past three decades it is estimated that costs associated with fire suppression on public land
have increased about sevenfold while the effectiveness and efficiency of forest fire management
have progressively declined. It is estimated that the 2015/16 Surf Coast fires will rack up a total cost
of some $450 - 500 million just to contain a bushfire that eventually covered about 2,500ha.
The predecessors of the IGEM have previously reported on major bushfires at the then
governments request, viz:
3
4
The coroner declined to hold an inquiry and the State government continued to refute the need
for an inquiry.
This audit commenced prior to the fires of summer 2002-2003, and did not examine suppression operations. The audit focused on
the planning, prevention and preparedness measures that can prevent or reduce the severity of Victorias seasonal wildfires a nd on
whether those essential planning and prevention measures are being effectively implemented.
a condition, with respect to time, place, manner, agent etc., which accompanies, determines, or modifies affairs surrounding and
affecting an agent Macquarie dictionary. Revised Third Edition.
Independent Investigation of the Lancefield-Cobaw Fire. Prepared by the Independent Lancefield-Cobaw Fire Investigation Team.
Murray Carter (Lead Investigator)
The Forests Act 1958 requires the Department (Secretary) to carry out proper and sufficient
work in these areas to prevent and suppress bushfire...(pg 4)
Many previous inquires and reviews have raised concerns about the aging workforce of fire
and land management practitioners and the relatively small pool of experienced operational
burn practitioners.....the Investigation Team holds strongly to the view that fire
management and planned burning on public lands requires a strong and well integrated
single agency able to bring to bear a diverse body of science, technical expertise, personnel,
physical resources and infrastructure that also services broader land management functions
and responsibilities. Fire management cannot be separated from land management activities
(pg. 27).
Consequently, there is much interest in the IGEMs findings on the Wye River-Jamieson Track Fire.
Will history repeat itself? Have vital lessons from previous inquiries been applied?
The Coroner has also announced that it is in the public interest to hold an inquiry into the Wye River
fires but the scope and form of Inquiry/Inquest will be informed by the IGEMs report to avoid
unnecessary duplication. This seems sensible but much is riding on the IGEMs analysis and
subsequently, the Coroners findings to ensure Victoria has once again, effective and efficient forest
fire management as a major core business.
2.
IGEM was established in Victoria on 1st July 2014 under Part 7 of the Emergency Management Act
2013.
Its objectives are to:
Provide assurance to the Government and the community in respect of emergency management
arrangements in Victoria;
Foster continuous improvement of emergency management in Victoria.
Arrangements for crisis and emergency management include those arising from Victorias major
natural hazard bushfire.
IGEM operates independently of Emergency Management Victoria, the new overarching statutory
body responsible for co-ordinating emergency management policy and the implementation of
emergency management reform. However, the policy and practice of forest fire management
[prevention, suppression and planned use of fire, in the landscape] on public land is a matter of
public policy for State Government, not IGEM; but under its mandate through the Emergency
Management Act IGEM has the duty to independently evaluate the capacity, capability and
performance of the emergency management sectors in responding to the lightning strikes in the
hinterland above the Surf Coast on 19th December 2015.
3.
The authors acknowledge that they have drawn extensively and directly from the IGEMs Review.
The scope of this evaluation addresses:
3.1. Ministerial request to the IGEM for a review of the bushfire including the scope and terms of
reference/key matters to be considered.
3.2. IGEMs responsibilities under Section 64 of the Emergency Management Act 2013 and other
matters influencing the conduct of fire suppression.
3.3. Detection of and capacity/capability to control lightning-caused fire in the first 55 hours.
3.4. Discussion.
3.5. Conclusions.
Each of these matters is considered in the following text.
3.1.
Ministerial request to the IGEM for a review of the bushfire including the scope
and terms of reference/key matters to be considered.
The Ministers formal request (undated), reference CD/16/2127, is reproduced in Appendix 1. It was
prompted by internal DELWP documents leaked to the media, questioning media reports, local
people, United Firefighters Union and institute of Foresters of Australia calling for an independent
inquiry into the conduct of the fire. The request was received in the IGEMs office on 5th January
2016.
The IGEM review overall was to consider and incorporate good practice for 5 matters (see App.1).
This evaluation considers item 1: Detection of the lightning strike and initial attack on the Wye
River-Jamieson Track Fire which occurred on the 19th December 2015.
The Minister, on her understanding that the IGEM and some staff had previously visited the going
fire, also requested an update and initial draft report by the 21 st January 2016.
This report was prepared without the benefit of the IGEM interviewing several key personnel who
were still engaged in suppressing the fire. The final report was to be submitted by 19 th February
2016 and was released to the public on 25th February 2016.
Adverse publicity about the conduct of the fire rattled the government and most likely prompted the
requirement for an (ill-timed) interim report and restrictive terms of reference (TofR) for the final
report.
TofR are deficient. Consequently, the Review is deficient as the TofR severely limited the IGEM to
identify key potential learnings as well as areas of good practice.
Most truly independent inquiries include a TofR such as: any other matters that the IGEM deems
appropriate in relation to the Surf Coast fires.
This is standard practice although not without political wrangling as was the case for TofR for the
2009 Victorian Bushfires Royal Commission.
These matters are canvassed under Section 3.2 and Section 3.4.
8
3.2.
The IGEMs brief stems from S.64(c) viz: at the request of the Minister, provide advice to, or prepare
a report for, the Minster on any matter relating to the functions of the IGEM. [Authors emphasis]
The IGEMs website Review of 2015 Wye River-Jamieson Track fire, states that the conduct of this
review is within the normal scope of the IGEMs responsibilities [under S.64 of the Act] to foster
continuous improvement in Victorias emergency management arrangements.
The functions of the IGEM under S.64 are reproduced below.
[S.64]
(1)
(a)
develop and maintain a monitoring and assurance framework for emergency management, including
outcome measures, against which the capacity, capability and performance of the emergency
management sector is to be assessed; and
(b)
(c)
at the request of the Minister, provide advice to, or prepare a report for, the Minister on any matter
relating to the functions of the Inspector-General for Emergency Management; and
(d)
evaluate state-wide training and exercising arrangements to maintain and strengthen emergency
management capability; and
(e)
monitor and report to the Minister on the implementation of the Strategic Action Plan by
(i) responder agencies; and
(ii) Departments; and
(ii) the Emergency Services Telecommunications Authority; and
(iv) Emergency Management Victoria; and
(f) monitor and investigate the performance (in matters that are not financial matters) of the Emergency
Services Telecommunications Authority in relation to the provision of services by the Emergency
Services Telecommunications Authority to emergency services and other related services organisations;
and
(g) make recommendations to the Minister about matters arising from any monitoring or investigation of the
Emergency Services Telecommunications Authority; and
(h) perform any other functions conferred on the Inspector-General for Emergency Management by or under
this or any other Act.
(2)
In making recommendations when performing functions under this section, the Inspector-General for
Emergency Management must have regard to the resources that agencies have to implement the
recommendations.
(3)
In performing the functions conferred on the Inspector-General for Emergency Management, the
Inspector-General for Emergency Management may, subject to sections 65(1) and 67, consult with
relevant stakeholders.
Under these functions the IGEM had a duty to evaluate, in the context of the known risk, (such as
very dry seasonal conditions, long unburnt accumulated forest fuel, dwellings in the bushland,
tourist season and previous fire history):
DELWPs record in meeting its obligations in law under the Forests Act 1958 in matters of fire
prevention, fire suppression and the planned use of fire on public land.
9
DELWPs (include PV) level of preparedness statewide and specifically in the region as at 19th
December 2015. This would include human resources accredited for fire duty (incl. fire ground)
total number and available numbers. The Government establishes a state of readiness for each
fire season. Standards for levels of readiness are described in joint or agency Standard Operating
Procedures (SOP).
The number of project fire fighters (PFF) State and local and their level of training.
Equipment x type and location including aircraft on standby.
Compare and contrast actions on the Jamieson track and Delaneys Road lightning strikes. Detail
why initial suppression action was successful on one and not the other the lessons learnt.
19th December 2015 was a day of Total Fire Ban, so the emergency management sector would have
been at maximum readiness.
What numbers of DELWPs resources were available for immediate deployment and how many were
already deployed on going fires for which DELWP was responsible?
Incident documents used by the IGEM in preparing the Review are listed in Appendix 2.
3.3
This analysis is based on the following points either brought out by or considered to be inadequately
covered/omitted in the IGEMs Review.
(i) 19 December 2015 was a day of total fire ban state-wide. Given the preceding prolonged dry
conditions, previous and on-going fires, long unburnt fuel in the forest and approaching
Christmas/peak tourism; the emergency management sector should have been at maximum
state of preparedness. The Review needs to document the actual level of preparedness as at
December 2015.
(ii) An incident control centre at Colac was activated. Pre-positioned crews in the Otway ranges
with slip-on units [water storage and pumps] were patrolling to respond to bushfire.
(iii) Numerous fires across the State and locally were said to create high competing demands for
resources. Firefighters in the Barwon South-West Region were resourcing [numbers and units
not specified] the Scotsburn fire near Ballarat.
(iv) Upon detection of smoke near Jamieson Track by the Mt Cowley fire tower, crew/s were also
despatched to the Delaney Road lightning strike [time of detection and time resources arrived
not specified the resources were despatched late afternoon from where?].
(v) Resources assigned for first attack were said to be appropriate [with strong qualification];
...given the restrictions of the difficult terrain and forest environs, concerns for firefighter
safety and the competing demands of concurrent serious fires.
(vi) The Incident Management Team (IMT), ground crews, aerial resources and plant required for
direct attack was increased as the fire grew in size and complexity, yet the fire was unable to
be contained.
(vii) On 21 December 2015 the incident/bushfire was upgraded from a Level 2 to Level 3 status.
With hindsight it was later considered that a Level 3 status would have been appropriate from
the onset.
(viii) The IGEM observed (No1 OBS) in relation to initial attack:
Available resources were allocated appropriately for a Level 2 incident;
Safety of firefighters was the foremost consideration given the extreme danger of the
forest environment;
10
(ix)
(x)
IGEM supported the assertion of incident control personnel that the fire could not have
been resourced more during the initial attack without an unacceptable escalation of risk
[not detailed].
The IGEM also observed (No2 OBS) that the date and time markers for some incident and
planning documentation was inconsistent or absent; concluding this had the potential to
impact on both firefighter safety and effective utilisation of available resources.
IGEM later noted that, as the fire escalated, an extensive and exhaustive options analysis was
a key part of control strategy planning; and that it was highly likely this contributed to the
successful outcome of preserving life and minimising damage.
In view of the above, it is therefore highly questionable that the IGEMs interim report and final
review could categorically conclude, as did the government and the EMC, that under the prevailing
circumstances everything was done that could have been done and that this led to a successful
outcome. The ill-timed interim report of 21 Jan16, while the fire was still going, pre-ordained the
findings of the final Review.
Thankfully, there was no loss of life and the evacuation of threatened Surf Coast communities was
very well accomplished by Victoria Police. The success of other aspects of the fire fight is
questionable and requires a truly independent investigation.
The contributors [see Appendix 3] to the IGEMs Review did not include any truly independent
persons/organisations with extensive forest fire management experience. It also appears that when
originally factoring in the potential consequences of failing to quickly bring the fire under control,
insufficient attention was paid to the much greater potential threat to firefighters safety,
communities and assets when trying to control the fire under adverse fire weather that inevitably
would occur (on 25 December 2015).
Initial attack on the Wye River-Jamieson Track Fire Paper work first action follows.
The IGEM reports that incident planning during the period 19-25 December 2015 was documented
in the Incident Shift Plan (ISP) to direct operations on the fireground.
ISPs were prepared by the Colac IMT [presumably on the basis of reports from the fire boss who
actually saw what was happening at the fireground]. However, standing protocols in relation to
incident documentation require immediate preparation of an Incident Action Plan (IAP) [authors
emphasis].
Apparently, DELWP has traditionally produced ISPs rather than IAPs and for the Wye River-Jamieson
Track Fire guided operations with regular ISPs.
This over-preoccupation with process rather than outcomes reflects the absurdity of the situation
especially when the IGEM reported inconsistencies or absent incident and planning documentation.
11
12
The initial strategy of direct attack, preparation of containment lines and patrolling increased in
intensity; despite this the fire continued to grow in size and was spotting.
Construction of containment lines was slow in the difficult terrain, with reports indicating little
progress between 1922 December. There were significant risks to the safety of firefighting
personnel engaged in establishing the containment lines around the fire on steep slopes, which
meant that increasing the onground resources was not feasible.
Controllers also commented that two other fires requiring significant resourcing at Portland, and in
the Strathdownie area of western Victoria, had occurred in the previous month. This ongoing
demand meant firefighting personnel were becoming fatigued. [Authors emphasis]
The ICs expectation late on 19 December was containment of the fire by midday 20 December.
On the afternoon of 21 December, the RC deployed a Level 3 IC to assess the fire management
arrangements, and concluded that with extreme fire weather predicted on 25 December, there was
significant risk of the fire developing into a major incident.
The RCs evolving view of the fires management on 20 and 21 December was that the fire should be
managed with a more focussed approach.
The RC subsequently discussed escalation of the incident with the DELWP Chief [Fire?] Officer, and
SRC leading to a decision to upgrade the incident to Level 3. The formal transition of incident control
occurred at 4.18pm on 21 December.
By 21 December, containment was not expected until late 24 December, due to increased fire activity
on 21 December, including a spot fire to the east of the main fire.
IGEM considers that the available resources were allocated appropriately (according to the incident
Level) to implement this strategy, following due consideration of the fire; safety concerns relating to
the extreme danger of the forest environment; steep terrain, thick vegetation; and the weather. It is
clear that the safety of firefighters was the foremost consideration from the onset of the fire at all
levels of control.
IGEM notes that the date and time markers for some incident and planning documentation was
inconsistent, or absent.
The chronological sequence of detection and initial attack is given in Appendix 5.
13
3.4
Discussion
The Wye River-Jamieson Track fire was a fire prevention and fire suppression failure that likely ran
up a total bill for the taxpayer of some 450-500 million dollars; probably the highest cost for such a
fire in Australias history.
The only redeeming features were the successful evacuation of some Surf Coast Communities by
VicPol and fighting the fire without loss of life.
The IGEM takes the view; IGEM recognises the appropriately scaled approach to control and
resource allocation throughout this fire. The effectiveness of strategy planning and implementation
were highly likely to have contributed to the successful outcome of preserving life and minimising
further losses. This achievement of the most important strategic control priority is a successful
outcome for the community and the emergency management sector. IGEM supports the intent,
rationale, decision process and implementation of the fire control strategies in order to reduce risk to
communities; notwithstanding making some significant observations and recommendations
concerning problems with conduct of the fire which were then claimed to have had no material
effect on the outcome.
The claimed overall successful outcome did not meet the States Strategic Control Priorities, viz:
The States operating policy framework for emergency response, the State Emergency Response
Plan (SERP), is set out in the Emergency Management Manual Victoria (EMMV) Part 3. The SERP
defines the priorities and set out the organisational arrangements for managing response to
emergencies.
The intent of the priorities is to minimise the impact of emergencies and enable affected communities
to focus on recovery as early as practicable.
Priority
Comment
Protection and preservation of life is paramount. This includes:
safety of emergency services personnel
Success
safety of community members, including vulnerable community members
and visitors/tourists located within the incident area.
Issuing of community information and community warnings detailing incident
information that is timely, relevant and tailored to assist community members
Contentious
make informed decisions about their safety
Protection of critical infrastructure and community assets that support
Failure
community resilience
Protection of residential property as place of primary residence
Contentious
Protection of assets supporting individual livelihoods and economic production
Failure
that supports individual and community financial sustainability
Protection of environmental and conservation assets that considers the
Failure
cultural, biodiversity and social values of the environment.
The Otway Ranges present no more difficulties in fire suppression than many other parts of
Victorias mountain forests.
Over many years, experienced, well led forest fire fighters have a demonstrable record in quickly
containing bushfires burning under moderate to high forest fire danger index [FFDI]. This requires:
early detection of the fire, highly trained, well equipped resources strategically located for quick
deployment, aggressive first attack in the shortest possible time to bring the bushfire under control
and then keep safe.
This is the most effective and efficient way to achieve all of the States Strategic Control Priorities.
14
Fire prevention.
There was a heavy accumulation of fuel in the long un-burnt forest. Prolonged dry conditions made a
greater proportion available for combustion. Some fuel reduction burning [FRB] was scheduled in
March 2015, but was put-off when questions were raised about the safety of koalas and
subsequently adverse weather apparently precluded FRB before the start of the 2015/16 bushfire
season. In principle, where there is heavy fuel accumulation, FRB is usually applied over two of more
seasons to gradually reduce various fuel components in a way that minimises environmental damage
and greatly reduces the chance of fire escaping from the planned burn area.
Tolhurst and McCarthy (2016)9 report on the importance of FRB, viz: Effect of prescribed burning on
wildfire severity: a landscape-scale case study from the 2003 fires in Victoria. The following is an
extract from the Abstract: this study examined the effect of previous fuel reduction burning (FRB) on
the severity of the >1 million ha 2003 Alpine Fire in Eastern Victoria. ... a number of strong trends
were evident.
... fire or FRB age and percentage of north-western aspect, best explaining the variation in measured
fire severity data. The most important finding was that the reduction in fire severity and suppression
assistance effects of previous fuel-reduction burning started to decline substantially when the Forest
Fire Danger Index [FFDI] exceeded 50. Above FFDI 50, landscape-scale fire became weatherdominated and variation in fuel and topography became less important to continued fire spread. The
greatest effects of previous FRB in reducing wildfire severity and in assisting fire suppression occurred
when (1) the FFDI fell to 25 or less (late in the evening and overnight); (2) the age of the FRB was less
than 3years (i.e. when all three components of fuel surface, bark and elevated material were still
substantially reduced). Some fire-severity reduction effects were still evident for FRBs up to 10 years
old, but there was almost no evidence of FRBs older than 10 years having any effect on fire severity.
FRBs up to 10 years also had measurable effects on increasing burnt area patchiness and decreasing
canopy loss, both of which have [positive] ecological implications.
The IGEM is acutely aware of this critical matter having recently conducted a review of FRB targets
(published April 2015); examining a risk-based approach to bushfire management against the then
existing hectare-based performance target recommended (and accepted by Government) by the
2009 Victorian Bushfires Royal Commission.
The Government now accepts the IGEMs Recommendation 1, viz:IGEM recommends a risk
reduction target as the most effective form of performance target for bushfire fuel management on
public land to protect life and property and guide investments in fuel reduction burning.
IGEM also placed some significant caveats on the adoption of this new approach, viz:
However, such measures are unlikely to be meaningful to communities, other stakeholders and
government the key users of performance information. ...
Measures of ecosystem resilience [poorly developed at the moment] do not relate directly to
some of the observable results of fire.
Such measures do not capture the quality of engagement with communities.
Additional measures of performance should be included within a broader suite of performance
measures for the Bushfire Fuel Management Program.
which are conveniently ignored.
Furthermore, a risk-based approach and an annual rolling target of 5 percent minimum of public
land, as recommended by VBRC, are not mutually exclusive; having a defined target is also widely
9
Effect of prescribed burning on wildfire severity: a landscape-scale case study from the 2003 fires in Victoria. Australian Forestry,
2016. Vol.79 No.1, 1-14.
15
recognised as essential for holding organisations accountable against their business plans/statutory
obligations under Acts of Parliament.
Not surprisingly, many people experienced in forest fire management are concerned the
Government will significantly reduce the previous annual FRB target of minimum 390,000ha (which
DELWP was slowly attempting to achieve) at the expense of landscape-scale FRB.
These concerns have been realised with the Governments release (19 November 2015) of a major
communications exercise titled: Safer together A new approach to reducing the risk of bushfire in
Victoria.
Fuel quantities directly impact on firefighter and community safety, fire intensity and on the conduct
of fire suppression.
Consequently, IGEM should have reviewed:
The location and type of any fuel modification, including FRB, for mitigating bushfire risk to
communities, assets and landscapes in the Surf Coast and hinterland environs over the past
decade;
Timeframe for what areas and what vegetation types were treated and what were those
considered untreatable and why.
Further details are presented in Appendix 6; Extracts from and comments on a Government
communications exercise titled: Safer together A new approach to reducing the risk of bushfire in
Victoria; published 19 November 2015. (See www.delwp.vic.gov.au)
Fire Suppression.
A chronological sequence of detection and initial attack is given in Appendix 5.
Given the known risk; well established El Nino weather pattern, significant rainfall deficit, high fuel
levels available for combustion, an early start to the fire season with warning of fire severity; the
responsible agency should have been at maximum state of preparedness (capacity and capability).
The Bureau of Meteorology would have recorded and reported the incidence of lightning on
19/12/15. Fixed fire tower [Mt Cowley] supplemented by aerial reconnaissance should have ensured
the earliest possible detection of a lightning strike and triggered a rapid response to quickly quell any
outbreak of fire.
DELWP is a complex mega-department. Previous reports reveal that there is only a small
number/proportion of staff available for fireground duty and a deficit (emergency-sector State-wide)
of Level 2 and Level 3 incident controllers, particularly with extensive experience in forest fire
suppression.
IGEM is acutely aware of this situation and yet provided only scant information on DELWPs capacity
and capability for 2015/16 fire season as it affected resources available for the fire. There was no
mention of any seasonal firefighters [PFFs] engaged on the fire nor numbers of Parks Victoria and
VicForests personnel.
Data in Appendixes 5 and 7 indicate that opportunities appear to have been missed in containing the
fire under relatively benign fire weather and periods of low rate of fire spread [Figure 2, page 10].
16
It appears that first attack took too long to mobilize, access and focus on directly attacking the fire
day and night. It is not clear what resources were specifically allocated to each of the Jamieson Track
and Delaney Road fires and why the initial attack was successful on one lightning strike and not the
other.
The deployment of firebombers was belatedly put en train on 19 and 20/12/15 and the productivity
of some of the dozers preparing containment lines also appears to be low.
Operational factors critical to a successful attack are documented in Appendix 7.
These matters are now considered in Conclusions and What are the Real Lessons Learnt?
3.5.
Conclusions.
(i)
Real time data provided to the IGEM does not substantiate his claim (and previous
announcements by the Minister and EMC following the IGEMs ill-timed interim report) of an overall
successful outcome in combating the fire, notwithstanding there was, fortunately, no loss of life.
The IGEMs observations and Recommendation 2:
Throughout this Review there were a number of times where important activities were
identified as having occurred for which there were either no documents, or inadequately
documented results...Whilst there is no evidence to suggest that this negatively influenced
operational decision making [and consequently, fireground operations] during the response to
this fire, the potential exists for this to become an issue in future emergencies
are at best, disingenuous.
These matters obviously require further scrutiny with the benefit of independent people
experienced in forest fire suppression,
The Regional Controller [a person with considerable forest fire management experience]
belled the cat. The RCs evolving view of the fires management on 21 and 22 December was
that the fire should be managed with a more focused approach.
(ii)
A truly independent investigator considers all the relevant facts and the underlying
circumstances; each having a role in formulating conclusions and recommendations.
Terms of reference for the Review were deficient, but the IGEMs charter under Section 64 of the
Emergency Management Act (2013) provided the necessary scope.
The IGEM failed to comprehensively examine key circumstances mainly confining them to: extreme
danger of the forest environment ... steep terrain, thick vegetation and weather; fires in other
areas of Victoria requiring large ground deployments... (draw) heavily on firefighting resources;
controllers commented that two other fires (that) had occurred in the previous month ... meant
firefighting personnel were becoming fatigued.
This reflects on an organisation that has been allowed to progressively run down and relegate, what
should be one of its major core businesses, to a much lower priority.
The Otway Ranges present no more difficulties in fire suppression than many other parts of
Victorias mountain forests.
17
Notwithstanding there are some very good people in DELWP experienced in forest fire management,
the fact is they are now few in number immersed in a highly dysfunctional mega-department with,
at best, indifferent leadership. Independent reports (referenced in this evaluation) confirm this state
of affairs. Further, a top down command and control system under Emergency Management
Victoria, largely untested until the 2015/16 fire season, resulted, at least, in congestion and
apparently some confusion in decision-making.
IGEM avoided properly documenting these circumstances, all of which adversely influenced the
conduct of the fire.
(iii)
Opportunities were lost in combating the fire in the first 55 hours resulting in failure to
control the fire when it was small. This created a much greater potential threat to firefighters and
community safety and assets when attempting to control the fire under blow up severe fire
weather that inevitably, did occur (25/12/15) in the height of summer.
(iv)
The appropriate use of bulldozers in forest fire suppression is seriously undervalued,
compared with the huge escalation in use of firebombers, particularly as access within State forests
and national parks has been significantly reduced over the past several decades.
Contracting and placing bulldozers on standby each fire season with means for their quick
deployment is just as important as aircraft and suitably configured, they make a major contribution
to firefighter safety and productivity. The rehabilitation of their operations is a legitimate cost of fire
suppression.
(v)
IGEM Observation 1 Initial attack: ... it is clear that the safety of firefighters was the
foremost consideration from the onset of the fire at all levels of control ... the fire could not have
been resourced more during the initial attack without an unacceptable escalation of risk.
This is the elephant in the room that must be addressed if the States Strategic Control Priorities
are to be met. The first point in Observation 1 is correct; the second point is disputed. Better
deployment of the available resources with due regard to safety had a high probability of success
thus effectively and efficiently meeting all Victorias Strategic Control Priorities.
Fighting forest fires is a high risk occupation, but no more so than many others. It must be
considered as akin to warfare where risk is minimised (but cannot be completely eliminated) by
intensively trained experienced, well led and well equipped troops. Clearly there is a deficit of such
personnel in DELWP.
It is regrettable that keeping forest firefighters safe is often used as an excuse for not directly and
aggressively attacking bushfires. This is not a reflection on the troops, but on the culture and
leadership of the organisation and its level of preparedness. Great emphasis is rightly placed on
hazardous trees precluding fire suppression by ground crews until they are identified and dealt with.
However, as a cause of death and injury this is rarely put in perspective with other dangerous
occupations. It is a relatively rare event in forest fire suppression.
18
4.
(i) The complexity of fire suppression was greatly increased and firefighter and community safety
compromised because of high fuel loads in long unburnt forest.
There is unequivocal evidence that risk can be reduced, firefighter safety increased and fire
suppression aided by fuel reduction burning at the public/private interface and at landscapelevel in the forests.
(ii) Every window of opportunity under benign fire weather must be taken to quickly and
aggressively attack bushfire. It is the most effective and efficient way of keeping communities
and assets safe and protecting the environment.
There is no better example in the relatively brief history of initial attack on suppressing
bushfire in Australia than that of the disastrous 2003 Canberra firestorm. Coroner Marie
Doogan found that fire agencies failed to take advantage of the window of opportunity for
several days following lightning-caused fires on 8th January 2003. Subsequently, fires
coalesced and were unstoppable on 18th January 2003. Leonards10 summary of the four
inquiries into the fires is reproduced in Appendix 8.
(iii) The matter of firefighter safety must be addressed [training, equipment and modus operandi in
the field]. It is a dangerous occupation, the more so if the workplace is used as an excuse not to
directly attack fires when they are small.
Failure of initial attack usually results in a long and very costly campaign fire which exposes
firefighters, community and the environment to much greater risk of death and destruction.
(iv) The application of the current emergency management model with its top down and control
system and role of incident controllers in forest fire suppression needs re-evaluation.
This requires the aid of independent people with extensive experience of forest fire
management, including on the fireground.
Fire suppression is analogous to warfare, drawn to attention by Leonard 11.
In other evidence given to the [Bushfire] Royal Commission, Professors Herman Leonard
and Paul t Hart, and Major General (Rtd.) Jim Molan made what were arguably some of
the most pertinent comments about the nature of wildland fire-fighting, and particularly
first-attack. Molan for example, suggested:
If there are a vast number of fires on a particular day, the best people to handle it
and the way they know to handle it is through a decentralised system within an
overall general guidance from the top ... being command centric doesnt mean the
whole organisation is centralised and only ever one person makes a decision. It
means that the commander at each level is the boss and he has as many obligations
going up as he has going down. 12.
10
th
Leonard, M.L. (October 2014) Forest fire management in south-eastern Australia: for better or worse? Oct 25 2014. See Part 3. FIRE
MANAGEMENT ON PUBLIC LAND - VICTORIA BURNS WHILE ITS BUREAUCRACY FIDDLES. Will Recent Legislative Changes to Crisis and
Emergency Management Improve Fire Management on Victorias Public Lands? Dexter, B.D; A. Hodgson AM. January 2015. ISBN 9780-9942531-0-1
11
Ibid.
12
2009 Victorian Bushfires Royal Commission evidence transcript April 28th 2010 (140th day) and April 30th 2010.
See: http://www.royalcommission.vic.gov.au/getdoc/fbcb1aac-5749-4b38-8c30-439422740c84/Transcript_VBRC_Day_140_28-Apr2010.PDF.
19
5.
for
the
The State must recognize that forest fire management on public land cannot be separated from
other land management activities.
The Government to commission an external, independent panel experienced in organisations and
management, including high level expertise in forest fire management to provide three (3) options
identifying with reasons, their preferred option for structural arrangements to most effectively and
efficiently meet the States forest fire management obligations under the Forests Act (1958) and the
State Strategic Control Priorities set out in the Emergency Management Manual Victoria Part 3.
The panel would be informed by recent independent reports and Government policy that a strong,
well integrated single agency is required to bring to bear a diverse multi-disciplinary body of science,
technical expertise, personnel, physical resources and infrastructure that also services broader land
management functions and responsibilities. It must recognise that forest fire management on public
land cannot be separated from other land management activities.
The task would involve a thorough consideration of the required leadership skills and experience,
definitive job descriptions and key selection criteria for proposed staff in head office and regional
Victoria.
The analysis of the options would also identify the total number of accredited, fire-skilled categories
required, including for the fire ground, their employment conditions, specific training requirements
and base capital and recurrent expenditure for effective and efficient forest fire management and
other management activities on public land.
The options would also identify:
13
Number and location of work centres to minimise travel time for first attack crews and other
field management activities;
Appropriate number of regions and work centres to better serve communities and for staff to
have more intimate knowledge of the land and resources.
The authors are aware that DELWP is currently again [some 8 reorganisations since 2009, 4 in the last few months] attempting to
reorganise itself as a consequence of the Minister accepting all recommendations in the Carter Report (Nov.2015). The authors
consider DELWP is too dysfunctional to undertake this task which should be done by an experienced independent panel.
20
6.
Appendixes.
1) Ministers request for IGEM Review of Wye River-Jamieson Track fire.
2) List of incident documents IGEM used to conduct review.
3) Key contributors to the IGEMs review.
4) Organization of Otways forest fire fighting in the 1960s.
5) Chronological sequence of detection and initial attack.
6) Extracts from and comments on a Government communication exercise titled: Safer
together - A new approach to reducing the risk of bushfire in Victoria published 19 11 15.
[See www.delwp.vic.gov.au]
7) What actually did/did not happen on the fireground?
8) Extract from Changes in forest fire management in south-eastern Australia: for better or
worse? Mike Leonard. October 25, 2014 in Part 3 of FIRE MANAGEMENT ON PUBLIC LAND VICTORIA BURNS WHILE ITS BUREAUCRACY FIDDLES. Will Recent Legislative Changes to
Crisis and Emergency Management Improve Fire Management on Victorias Public Lands?
Dexter, B.D; A. Hodgson AM. January 2015. ISBN 978-0-9942531-0-1
21
Appendix 1. Ministers request for IGEM Review of Wye River-Jamieson Track fire.
22
23
Appendix 3
IGEM worked with the following bodies and individuals in preparing this report:
Colac Otway Shire Council
Country Fire Authority (CFA)
Department of Environment, Land, Water, and Planning (DELWP)
Emergency Management Commissioner (EMC)
Emergency Management Victoria (EMV)
Incident Controllers (ICs) and members of Incident Management Teams (IMTs) involved in
managing the Wye River Jamieson Track fire
Members of the state and regional control teams
Surf Coast Shire Council
Victoria Police (VicPol).
24
Fig 4.1:
District Organisation
In the Otway ranges in the 1960's, the Forest Act 1958 and Regulations were administered through Gellibrand
and Eastern Otways forest districts. Eastern Otways district office was located at Forrest. Subdistrict centres
were Lorne and Anglesea. Gellibrand office was located at Gellibrand with subdistrict centers at Colac, Beech
Forest and Apollo Bay. Each district was managed by a district forester supported a staff of foresters, forest
overseers, admin staff and a full time work crew.
Forest fire incidence
The western Otways is largely wet forest due to the high winter rainfall and rich soils. The presence of
mountain ash (E. regnans) forests that rely on high intensity fire for regeneration is an indicator that high
intensity fire has been experienced in the west Otways over geological time. The wet forests graduate into dry
forest and extensive heaths north of Anglesea in the rain shadow.
Beech Forest (Weeaproinah) is the highest rainfall station in Victoria due to orographical lift off the sea.
However the fuels quickly dried out through spring and summer and a record of fast moving fires started by
arson, lighting and rural activity meant a vigorous local response was required to suppress fires before
extensive fire spread occurred.
25
Resources
The following local resources were available for firefighting each summer. These figures varied over time, but
are indicative of Otway fire fighting resources during the 1960's.
Tankers 4x4
Location
Foresters
Overseers
Admin
Work Crew
Forrest
12
Lorne
Anglesea
Gellibrand
Beech Forest
Apollo Bay
Colac
Lavers Hill
Total
11
3000
400
4
1
1
2
1
10
1
3
1
1
27
The Divisional Office at Ballarat managed fire fighting support between the six forest districts of Southern
division and could quickly send additional resources for "early start next day" crews. The Division of Forest
Protection in Melbourne was responsible for statewide coordination of resources and could readily deploy the
"Summer Crew" of some 20 firefighters and resources from other Divisions as the need arose. Fast backup
from fully equipped, well led and skilled forest firefighters could be relied upon.
The CFA was not established in the northern Otways, but had a presence in Anglesea, Lorne and Apollo Bay.
Lorne and Anglesea brigade personnel were experienced forest firefighters and Lorne CFA rakhoe crews were
used to good effect in the forest during the Lorne fires in the mid 60's.
Bell helicopters were deployed via the Melbourne office for fire reconnaissance from about the mid 60's. Fire
bombers were in their infancy in the 1960's and were not used in the Otways at this time.
Principles
Fire fighting was core business and was recognised by everyone on the team as our most important activity.
We trained for an annual Otways fire school each year in which various competitive tanker, pumper and trail
raking activities were held and a high degree of speed and skill was developed. The admin staff were not
exempt and took their place on the summer standby roster.
Our mantra was that fires must be suppressed during the first work period at an area of less than five acres
(2 ha).
In the event of failure to achieve this objective, the Chief of the Division of Forest Protection required a
detailed, sometimes verbal as well as written, "please explain".
I was posted to Eastern Otways forest district during 1963-1969 as a forester, living at Forrest in a
departmental residence. I quickly gained my stripes as a fireline supervisor and was soon appointed as a
"fireboss" under the tutelage of veteran district forester Kevin Roderick O'Kane. I had developed a fire
administration system to make this task easier.
26
27
Next minute there was a great clatter as the dozer seemingly dived vertically into space and the control line
was duly constructed.
We were not "gung-ho" or swashbuckling in our approach, but we were confident of and expected success and
persisted until we achieved it. Fires were just another day at the office.
There were no stringent safety restrictions apart from the wearing of protective dress. There was no need. All
personnel were competent in the bush, knew what the hazards were and how to avoid them. In my time as a
forester there were no deaths or serious injuries of firefighters despite some epic, if not heroic, firefighting
feats carried out over long periods under the most difficult circumstances.
Why was the Forests Commission an effective fire fighting organisation?
Looking back, the main factors in the success of the organisation in putting out fires fast seem to me to be:
Fire prevention and suppression was a core function of the Commission with the Chief of Forest Protection
reporting direct to the Commission itself;
The Chief had directed that all fires be suppressed in the first work period at less than two hectares,
leaving no doubt about priorities (or consequences);
Fast, vigorous and determined first attack was essential to achieve this objective and was always our No 1
priority;
Locally employed people who knew the country made up the fire crews and there was a real sense of
team spirit within and between crews;
Light and heavy dozers with skilled forest operators and all other forestry tools were immediately
available from our local resources;
The sawmilling industry provided experienced plant operators and heavy dozers in most forested areas of
the State including the Otways;
Crew leaders, sector bosses and fire bosses were all competent leaders who knew what they had to do,
and who had the respect of the crews.
Sure, it is easy to talk up the 'good old days' from the rocking chair but the five factors listed above are not so
obvious in fire crews today and to my knowledge have not been replaced by equivalent "new order" values.
Current impediments
Here are some of the impediments acting against effective forest fire suppression:
Many personnel deployed to fight forest fires do not have an in depth knowledge of bushcraft or fire
fighting in local forests and may have limited (or no) forest fire fighting training or experience;
Many crew leaders do not have an in depth (or any) knowledge of bushcraft or fire fighting in local forests
and may have limited (or no) actual forest fire fighting training or experience;
Many line managers do not have an in depth (or any) knowledge of bushcraft or fire fighting in local
forests and may have limited (or no) actual forest fire fighting training or experience;
Fire crews may be dispatched to fires in unfamiliar territory without local knowledge or an understanding
of tracks and vegetation types, leading to inefficient use of time and effort and possible safety issues;
28
The objective of suppressing all forest fires in the first work period has long been abandoned, and the
principle of fast, vigorous and determined first attack has gone with it, leading to the development of
larger and more intense, and damaging fires that cannot be suppressed;
Logging contractors do not operate in many forested areas and their knowledge and equipment is no
longer available for firefighting;
Contract plant operators hired in may have no knowledge or experience of fireline or forest track
construction;
There is undue reliance on aircraft for trail building, especially in steep forested country. While this might
make good copy for politicians it is ineffective and very expensive. There is no substitute for skilled boots,
wheels, and tracks on the ground in achieving forest fire control;
There is a perceived reluctance, especially in parks, to attack small fires rapidly with machinery, causing
delays that result in the extensive spread of fires on the next hot, dry, windy day;
There is a perceived anti-fire suppression doctrine abroad based on "natural" or "let burn" fires that might
be OK in the Everglades, but is a certain recipe for disaster when applied in eucalypt forest and especially
in the Otways;
There is an excessive level of bureaucracy in fire fighting with cumbersome lines of communication and
delays in decision making, in headquarters located long distances from incidents.
extensive and repetitive damage to forest and watershed values that no-one seems to give a toss
about:
potential for (regrettably) large loss of life in outer urban forested areas;
Rod Incoll
20/01/16
29
30
31
32
33
34
35
36
Table 4.1
Contrasting the Line of Control for Bushfire Suppression on Public Land in 2014/15 to that of the Department of State Forests in 1982/83.
Emergency Management Victoria 2014/15
STATE
STATE
Forests Commission [3 Commissioners. Displan may be requested to have come into force for a major emergency to provide and co-ordinate
additional support for fire suppression.]
STATE
37
precluded more resources [ground rather than air?] being deployed. The IMT now adopted a
direct attack strategy using medium helitack and large air tankers to suppress the fire. Three
dozers continued to work on establishing bare earth containment lines.
1536hrs. Fire area was assessed at 65ha; an increase of 37ha since the morning report.
By late on 20 Dec15, crews had established 2km of containment line on 7km fire perimeter.
It is not clear what the proportion of manual [if any] versus dozer control line had been
constructed.
21 12 15
1103hrs. The fire area had increased to 86ha [+21ha in 20 hours], indicative of milder
weather conditions.
DELWP now substantially increased the resources on the fire; 40 on the fireground plus a
further 30 staff in the ICC [total in ICC not specified but likely to outnumber those on the
fireground]. Seventeen slip-on units were supporting crews directly working on the fires.
21 Dec15. Milder conditions allowed firefighters to continue building containment lines.
AM 21 Dec15. Crews detected a spot fire 200m from the main fire and reported continuing
difficulties in accessing the fire in dense forest and steep terrain.
1526hrs. The spot fire had grown to 3ha and aerial reconnaissance identified increased fire
activity. The main fire was now 99ha, an increase of 13ha in 4hrs 23mins. On 21 December
the fire was upgraded from a Level 2 to Level 3 incident.
During the night of 21 Dec15, the fire burnt to with 10m of Jamieson Track. Fire crews
continued to patrol and work on the containment line where it was safe to do so.
22 12 15
1426hrs. The fire area had grown to 141ha, an increase of 42ha in 23 hours and by late 22
Dec15 was estimated at 180ha revised down to 163ha on 23 Dec15 [an increase of about
1ha an hour under the prevailing conditions].
The inability to contain and bring the fire under control in relatively mild fire weather under the
combination of terrain, forest type and heavy fuel load and constraints on deployment of resources
had now reached the point where it was necessary to change strategy.
Fire suppression strategy was radically changed over 22/23 Dec15 to one of back-burning in an
attempt to contain and control the fire before the onset of predicted severe fire weather by 25
December 2015. This involved a massive upgrade in resources, viz: 138 Personnel, 17 tankers, 61
slip-on units, 8 dozers and 6 aircraft in support.
On 24 December 2015 the fire area was 271ha [no time specified] and at approximately 1100hrs on
25 Dec15, strong gusty winds led to multiple spot fires and the fire area increased to 2,080ha by
2226hrs that night, destroying 116 dwellings in Wye River [98] and Separation Creek [18].
39
40
Additional measures of performance should be included within a broader suite of performance measures
for the Bushfire Fuel Management Program.
Communities need down to earth, everyday life parameters, not esoteric or nebulous concepts.
2. What does risk look like?
In 2002, bushfire risk in Victoria was almost 90 percent. Following a decade of drought and fire which included
the 1.5 million hectare Alpine fires in 2002-2003, the 1.0 million hectare Great Divide fires in 2006-07 and the
0.5 million hectare 2009 Black Saturday fires, bushfire risk fell to below 60 percent in 2010.
With the recovery of our forests and parks from the impacts of these fires, and as fuel loads increase, bushfire
risk has risen. It is now at 65 percent. Without effective intervention, it is likely to return to 2002 levels (almost
90%) by 2020.
Clearly DELWP is counting hectares burnt in bushfires as complementing hectares fuel reduced by planned
burning. DELWP and its predecessors have been warned about this folly on several occasions, including the
Premier and relevant portfolio Ministers in January 2015.
There is incontrovertible evidence14 that in some places, fuel created by severe fires since 2002 was the main
driver of 2006-2007 fires. This problem is not recognised by DELWP; in fact they perpetuate a lie given the
number of prior alerts. Certainly, there are sections within large areas destroyed by bushfire that have had low
to moderate fire intensity. It is legitimate to include these areas with those that have been successfully [to a
15
standard] fuel reduced with a planned burn . Remote sensing and ground truthing must delineate such areas
for their inclusion in meeting planned burn targets.
DELWP has repeatedly failed to rectify this problem.
The Minister should direct Secretary DELWP to immediately address these problems.
The IGEM should have included these matters in the Review of the Wye River-Jamieson Track fire.
14
Such matters were extensively reviewed and recommendations made to Government in January 2015. See Fire Management on Public
Land Victoria Burns While Its Bureaucracy Fiddles. Will Recent Legislative Changes to Crisis and Emergency Management Improve Fire
Management in Victorias Public Lands. Dexter,B.D. and A.Hodgson AM January 2015. ISBN 978-0-9942531-0-1
At the invitation of the IGEM, this report was one of two submissions made by these authors to the IGEMs review of the performance
targets for fuel management on public land. It can be found on the IGEMs website.
15
Kevin G. Tolhurst & Greg McCarthy (2016) Effect of prescribed burning on wildfire severity: a landscape-scale case study from the 2003
fires in Victoria, Australian Forestry, 79:1, 1-14, DOI: 10.1080/00049158.2015.1127197
41
Appendix 7.
Figure 7.1#. Timeline of selected key events for 2015 Wye River-Jamieson Track fire (18 to 22
December). # modified extract from IGEM Review.
Resources deployed between 19-22 December 2015 [modified from Appendix 3 of IGEM Review].
42
20 12 15
0658hrs. IMT requests aircraft at fireground ASAP. [Why so late given first light was
about 0600hrs?]
0823hrs. IMT requests further aircraft as Jamieson Track fire had breached control lines.
0900 -1330hrs. Aerial response increased to 7 fire bombers [4 water bombers and 3
helitaks].
21 12 15
Comment:
19 12 15. There were about 5 hours of daylight from time of detection to attacking the lightning
strikes. Only 1 medium helitack was deployed to service both Jamieson Track and Delaney Road
fires. Why werent the 2 other available firebombers, carrying foam, re-configured to water
bombers? A critical opportunity lost when the fire was a spot.
20 12 15. Some 3 hours of daylight was lost in belatedly requesting firebombers which slowly
built up to 7 firebombers over another 4.5hours. Another opportunity lost in supporting ground
crews, which were under-resourced [14 personnel attacking 2 fires].
1815hrs. One dozer to prepare 1.2kms of access from Jamieson Track to the fire.
Time not specified. Dozers started work to directly attack the fire.
20 12 15
21 12 15
Comment:
McCarthy et al (2003)16provide data on fireline conditions and fireline construction rates collected
from 103 fire events spanning the 1997-98 to 2000-01 fire seasons in Victoria.
The following comments draw on data in Research Report (No. 56) viz:
Table 2, page 13. Fireline construction rate summary;
16
Prediction of firefighting resources for suppression operations in Victoria's parks and forests. Dept. of Sustainability & Environment
Research Report No. 56. Gregory J. McCarthy, Kevin G. Tolhurst and Michael Wouters. FOREST SCIENCE CENTRE. Orbost, Creswick a nd
Mildura. December 2003. ISBN 1 74106 798 7
43
Figure2, page 14. Model of firefighter line construction rates based on restriction due to the
presence of elevated fuel and terrain.
Figure 3, page 16. Model for D4 bulldozer construction rates based on effect of adverse terrain
and debris.
Figure 4, page 18. Model for D6.
Note, IGEM Review does not specify the size/capacity of each of the 4 bulldozers employed, nor
their actual hours of preparing containment lines.
In dense timber stands with elevated fuel, progress can be slow when side cutting on slopes is
required. D6 class is the preferred machine. However, the bulldozer is a vital tool when used to do
the initial clearing work so that a ground crew can be significantly more productive in cutting a
mineral earth containment line.
It is assumed that one bulldozer took about 3 hours to construct 1.2km of access from Jamieson
Track to near the fire [from 1815 2130hrs on 19/12/15] and worked an additional 3.5hrs [to
0100hrs 20/12/15] preparing 500 700m of containment line on/near the fire.
It is perplexing that 3 bulldozers were reported [1025hs, 20/12/15] as continuing to work on
establishing bare earth containment lines and by late on 20/12/15 had established [only] 2km of line
on the estimated 7km fire perimeter. Three [or 4] dozers working for some 15hrs apparently had a
combined productivity of only 170m/hr.
The IGEM Review presents no specific data on these matters which are determinant on the success
or failure of the initial attack, highlighting the need for another truly independent public
inquiry/inquest into the Wye River Jamieson Track fire.
44
24. This most recent case involved civil litigation and extended over ten years. It ultimately involved
plaintiffs suing the NSW Government on the basis that its Rural Fire, and Parks and Wildlife
Services were negligent in their handling of the fires.
25. In his judgement, Judge Higgins found that the conduct of the NSW fire-fighting authorities fell
below the standard of reasonable care that could or should be expected from a fire-fighting
agency, but due to various provisions of NSW law, the State was not liable to pay damages.
17
Electro Optic Systems Pty Ltd v The State of New South Wales; West & West v The State of New South Wales [2012] ACTSC 184 (17
December 2012).
45
26.
This is the fourth Inquiry to formally examine the initial management of the fires which
started in NSW. The earlier Inquiries were:
July 2003: The Inquiry into the Operational Response to the January 2003 (Canberra)
Bushfires (Mr Ron McLeod);
September 2003: NSWs Deputy State Coroners Inquiry into the Circumstances of the
Fire(s) that started in NSW/ACT Border which burnt during January 8-28, 2003 (Carl
Milovanovich); and
December 2006: The Canberra Firestorm Inquests and Inquiry into the Four Deaths
and Four Fires between 8 and 18 January 2003 (Maria Doogan, Coroner).
27. Each Inquiry went to some lengths to examine the actions of the NSWs IMT, over the first few
days after the fires ignited. Three of the four Inquiries were subsequently critical of actions
taken by NSWs authorities during this period viz:
The Inquiry is of the view that the fires, started by lightning strikes, might have been
contained had they been attacked more aggressively in the 24 or so hours after they broke
out. Nevertheless, the dryness of the vegetation after a prolonged, severe drought and the
high volume of flammable fuel that had accumulated over timecoupled with weather
conditions that were extremely conducive to firemeant that once the fires gained a hold
they proved extremely difficult to contain or suppress. Indeed, the fires on 18 January were
accurately described as unstoppable (McLeod ACT Commissioner, pg. iii);
From the accompanying report you will see that I conclude that the failure to aggressively
attack the fires in the first few days after they ignited on 8 January was one factor that led to
the firestorm on 18 January 2003, which resulted in four deaths, many injuries (some of them
extensive and permanent) and property losses values at $600 million to $1 billion (Doogan
ACT Coroner covering letter);
The question is, assuming, as Ms Crawford (the Incident Controller) conceded, her
judgment on strategy was so flawed as to fall below the standard to be expected of a
competent IC and, hence, the judgment of those who were aiding and, in the case of Mr
Hunt, assuming those duties, was similarly flawed, did that amount to negligence and, if so,
did that negligence cause the break-out of fire which happened?
I accept that the strategy particularly to the south east and the south west was so flawed.
(Judge Higgins ACT Supreme Court paragraphs 223 224).
28. An alternative view of the NSWs IMTs actions was taken by the NSW Deputy Coroner viz:
I am of the view that to have sent fire-fighters into the Brindabella Ranges on the
afternoon of the 8th January, 2003, would be contrary to all the basic firefighting knowledge
and would have placed professional and volunteer fire-fighters in potentially grave danger.
The decisions that were made on the afternoon of the 8th January, 2003, must be examined
in light of what was known then and not what is now known as having taken place on the
18th January
(Milovanovich NSW Deputy Coroner pg. 15).
46
8 April 2016
Dear Minister,
Open Letter about Wye River and Separation Creek
I write from my professional position but acknowledging I also have a private stake in this matter,
being part of one of more than one hundred families, partnerships, cooperatives and others who lost
houses at Separation Creek and Wye River, in the Christmas Day fires, last year.
I write following the public meeting held last evening (7th April) at the Treasury Theatre. I thank the
Colac Otway Shire and the Victorian Government and its agencies, for the opportunity to hear about
progress and future plans for our communities.
I will also state for the record that I have served as a bushfire authority for numerous state and
federal governments, in a range of capacities. I have published widely on the subject, including the
book Burning Issues that was published by CSIRO Press in 2012. Perhaps most pertinent in regard of
this letter were my roles as an expert witness to the 2009 Victorian Bushfires Royal Commission, and
last year (2015) as a member of an advisory panel for your department (Department of Environment,
Land, Water and Planning) on bushfire fuel management.
The substance of my letter can be summarized as four points:
1. There are inaccuracies and inconsistencies in the Terramatrix assessment of fuels, and in
part slopes, in its recommendations for Bushfire Attack Level (BAL) ratings. Your government
has seemingly accepted the Terramatrix assessment and declared it to be uncontestable.
2. A bewildering decision to refuse to take the steps required (including, if necessary, acquisition
of private land) to establish an Asset Protection Zone (and a related Bushfire Moderation
Zone) around the Wye River and Separation Creek townships. This is clearly and directly
opposed to your governments stated approach to bushfire risk management* and its
application in Victoria. It is directly opposed to the findings of the 2009 Royal Commission.
3. The clear shifting of the financial burden of risk mitigation from the public purse to the private
purse. This includes your governments application of BAL ratings that will incur very large
additional costs for those seeking to rebuild, whilst refusing to incur costs associated with
managing the fuels responsible for those ratings.
4. An overall approach to the post-fire planning and implementation that will, in my view,
seriously damage if not destroy the social fabric and cohesion of these townships, by pitting
neighbor against neighbor. As one example, the costs incurred by private landowners are
disproportionately shared amongst the community. Those with property around the perimeter
Centre for Carbon Water and Food
The University of Sydney
380 Werombi Road
Camden
NSW 2570, Australia
of the townships will carry a greater financial burden as a result of your governments refusal
to properly manage fuel loads. As another example, there will be streets where some
residents will have to build to a BAL rating of FZ, while others will build to a less expensive
rating. Given good evidence that burning adjoining house(s) contributed to fire risk for most of
the houses that were destroyed on Christmas Day, those who are forced to spend much more
to build to a FZ rating around the perimeter, could be argued to be protecting houses with
much lower ratings, closer to the centre of the townships. My point here is that such
retrospective changes to BAL ratings create inequity between residents that was not there
before.
I have provided a little more detail in the following pages.
In closing, it brings me no joy to write. I have for decades worked with governments to improve the
management of forests and their sustainability, and the safety of peoples lives and property as these
relate to bushfire. I hope the government reconsiders its overall approach, as well as the contributing
aspects noted here.
Yours faithfully,
* Last year the Victorian Government decided to reject the Royal Commission recommendation of
hectare-based targets for fuel reduction (the only recommendation to be so rejected) in favour of a
Risk Management approach.
1.
Whilst understanding that any blanket approach to BAL ratings is bound to throw up anomalies, the
current plan, as presented at the meeting yesterday, has so many inconsistencies and inaccuracies
that its value must be seriously questioned. As just one example, there are many blocks where only
one side has exposure to unmanaged (no fuel reduction) forested land, yet are given a FZ rating,
while others, or parts of others, of equally steep slope, escape with a BAL 40 rating. It seems no
account has been taken of the fuel reduction that is required by local and state government of private
landowners. In addition, there seems to have been little account taken of the actual distances of
house sites from the fuels. There are many examples where distances from forest fuels exceeding
40m still result in FZ ratings. Numerous well-qualified people have noted that the mapping used in the
report is inaccurate with respect to locations of boundaries and roads, and slopes have seemingly
been assigned uniformly over large areas, with little checking at the block level.
2.
This is an insidious issue, seldom recognized. In this case, it is perhaps a little more obvious than
many. The BAL ratings are in large part derived from fuel loads. The BAL ratings determine a very
significant portion of the costs of building in fire-prone landscapes. However, it is the government that
is responsible, overwhelmingly, for fuel loads. In part, local and state governments are responsible for
ensuring private land owners manage the fuels on their property. Most Wye River and Separation
Creek landowners are very familiar with the yearly or twice-yearly reminder letters to this effect. We
are warned we could be fined heavily if we fail to comply. Sadly, no one fines the government for its
failure to do likewise in the tens of thousands of hectares of forested land to the north and west of Wye
River and Separation Creek! In addition, successive governments have done nothing to ensure that
fuels were properly managed on the large areas of private forested land that immediately surrounds
much of Wye River and Separation Creek.
To suggest that because there is private land immediately adjacent to the townsites, nothing can be
done about the fuels and fire risk, beggars belief. The Victorian Government has more than sufficient
powers, and financial capacity, to either acquire the land compulsorily or reach an agreement with the
landowner, to ensure fuel loads are managed as in all other Asset Protection Zones.
Insurance costs are yet another example of cost shifting. There is no doubt whatsoever that insurance
costs at Wye River and Separation Creek will rise, and rise quickly and substantially. Insurance
companies will note that the State Government has refused to introduce an Asset Protection Zone,
and adjust their premiums accordingly.
4.
Overall approach
It seems to an interested observer that the Victorian Government was quick to absolve itself of
responsibility for the destruction, and costs to individuals, caused by the Christmas Day fires.
Beginning immediately after the fire had passed, and with the Premier and his Ministers, the
Government has defended its handling of the fire, and assured everyone that they would be helped to
rebuild.
Now that the emergency has passed, the truth seems different. Big costs are to be borne by private
individuals, with government imposing a significant proportion of those costs. Fuel management costs
are being avoided by the Victorian State Government. Private landowners must face increased
building and insurance costs.
The tragedy of the Christmas Day fires is all the more so given this was not the big one as noted
by Craig Lapsley, the Emergency Management Commissioner - this was not a roaring crown fire
sweeping down into the townships of Wye River and Separation Creek, chock full of people between
Christmas and New Year. The Christmas Day fire was of relatively low intensity, and had for days
prior been a very small fire, that the government and its agencies might normally have been expected
to easily contain.
Sadly, with the history of failure of successive governments in fuel reduction, and the refusal even
now, to accept the responsibility for reducing risk through creation and implementation of Asset
Protection and Bushfire Moderation zones, the Victorian Government encourages the grim possibility
of a larger, even more destructive fire. No amount of fuel reduction can fully prevent catastrophic
bushfires there will always be residual risk. But willful ignorance of its own policy and strategy
seems an odd way for the Victorian Government to approach the problems of Wye River and
Separation Creek.
All governments - local, state and commonwealth have responsibilities to protect their citizens and
support communities. At present, Wye River and Separation Creek are being failed on many levels. It
is governments that decide which land is available for building, governments that decide how such
land should be serviced with roads and power etc. Governments determine, largely, the financial
resources available for fire fighting, and fire risk mitigation. Governments are responsible for how
community costs, including aspects of fire risk, should be shared within the community.
I also find the overall approach puzzling at a social and economic level. It would seem to be in the
interests of local and state government that people want to build and live and holiday in Wye River and
Separation Creek. Right now, the agenda seems to be to make it as difficult (and still as dangerous)
as possible.