Paintings & Photography

19 pages
7 views

Relationship between Severe Landscape Dryness and Large Destructive Fires in Victoria

Please download to get full document.

View again

of 19
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Based on the results of an initial climatological study (Gellie, 2009) of the ACT region, seasonal and inter-annual variation in daily soil water deficit (SWD) was investigated and associated with landscape susceptibility to large destructive fires
Transcript
  VI International Conference on Forest Fire Research D. X. Viegas (Ed.), 2010 Relationship between Severe Landscape Dryness and Large Destructive Fires in Victoria  Nicholas Gellie  Department of Sustainability and Environment, Office of Land and Fire, Ovens District, PO Box 235, Ovens, VIC 3737, Australia, nicholas.gellie@dse.vic.gov.au Kelsy Gibos  Department of Sustainability and Environment, Office of Land and Fire, Ovens District, PO Box 235, Ovens, VIC 3737, Australia, kelsy.gibos@dse.vic.gov.au Ken Johnson Fenner School of Environment and Society, The Australian National University, Canberra, ACT 0200 ken.johnson@anu.edu.au  Abstract Based on the results of an initial climatological study (Gellie, 2009) of the ACT region, seasonal and inter-annual variation in daily soil water deficit (SWD) was investigated and associated with landscape susceptibility to large destructive fires (LDFs) in Victoria. Point-based indices of SWD at key forest locations in southern Victoria were linked to levels of general landscape dryness, which can then be related to drought-affected fuel hazard levels and extreme fire behaviour in major fire events in Victoria. Daily soil water deficit (SWD) was estimated from long-term weather datasets for dry and wet sclerophyll forest types near the Melbourne ranges using the Mount Soil Dryness Index (MSDI). Threshold values of SWD in excess of 140 and 200 mm during the recognised fire danger period (January to February) were used to define acute periods of SWD for each forest type. Fire susceptible  periods that overlapped in time for both dry and wet sclerophyll forests identified fire seasons with large destructive bushfires within 100 km of Melbourne. Resulting increased fine fuel loads and critically low dead and live leaf moisture contents throughout the forest structure can then lead to much higher combustion and energy release rates on severe fire weather days that what can be predicted using current drought indexes. The combination of extremely dry landscapes and severe fire weather conditions on Black Saturday led to rapid development of crown fires and ember spotting. Massive energy releases, ranging from200 to 4000 GW, produced convection columns 9000–11,000 m in height, accompanied by cyclonic winds in downdrafts, numerous fire whirls and gaseous explosions. This fire climatological study illustrates that critical ranges of SWD and heat–wave temperatures can  be used to improve estimates of the potential for large destructive fires. Coupling landscape level dryness information with local moisture content and loads for various vegetation–fuel complexes may  provide better estimates of energy release during LDFs. Keywords: Fire climatology, soil water deficit, landscape dryness, vegetation flammability, fuel combustibility, energy release 1  VI International Conference on Forest Fire Research D. X. Viegas (Ed.), 2010  1 Background and Rationale 1.1 Fire History in Victoria South–eastern Australia has long been subject to large and sometimes catastrophic fire events since European settlement, and even before then. Prolonged drought, searing heat waves, days of extreme fire with temperatures exceeding 43 o C and gale force winds are the first order factor in fire size and destructive potential. Once seasonal very dry landscape conditions are established, the occurrence and potential size of a large destructive fire 1  (LDF) depends on sources of fire ignition and the daily nature of fire weather (Gill 2005:67). Large destructive fires occurred in the central region of Victoria in the fire seasons of 1851–52, 1897– 98, 1905–06, 1913–14, 1925–26, 1926–27, 1938–39, 1943–44, 1961–62, 1964–65, 1968–69, 1982– 83, 1997–98, 2002–03 and 2006–07 (Foley, 1947, Council of Australian Governments, 2003). The interaction of the seasonal patterns of daily rainfall and potential evaporation (E P ) largely influences seasonal landscape dryness and the related potential for an LDF. Extreme fire events in 1982–83 and 2008–09 can be related to deeply dried landscape and contributing fire weather, in particular heat waves and strong turbulent winds up to 5000 m deep in the atmosphere. The Ash Wednesday fires of 16 February 1983 burnt 202,500 ha and resulted in 73 deaths and 2894 properties lost. The Black Saturday fires of 7th February 2009 burnt 360,000 ha with 173 lives lost and 4200 homes destroyed. Both events had a destructive impact on local communities, as well as forests, wood, biodiversity, and hydrological processes. 1.2 Assessing Landscape Dryness The temporal nature of a landscape’s susceptibility to LDFs is generally predicted with drought indexes derived from daily weather recorded at representative points in a landscape. An assumption is made that when identified values of a drought index at these representative points are exceeded, key vegetation factors such as continuity across vegetation types, fuel bed combustibility 2  and live and dead fuel flammability 3  have also reached critical levels. This means that moist areas in the landscape such as sheltered wet forests traditionally not expected to burn, have now become dry and available to  burn in a LDF. However, the ability of current drought indexes to predict and represent the availability of both fine and coarse dead fuels for a representative vegetation type has never been formally quantified by a properly conducted scientific study. Common drought indexes, such as the Keetch–Byram Drought Index (KBDI) (Keetch and Byram, 1968), the Mount Soil Dryness Index(MSDI) ((Mount, 1972)) or the Fowler soil water balance model (F–SWDBM)(Fowler, 1992, Fowler and Adams, 2004) take the form of a daily soil water balance model (DSWBM):  TOA QQE-PMS      (mm d -1 ). Equation 1.1 where  MS is the change in soil moisture store, P is the daily precipitation, E A  is actual total evaporation; Q O  is the runoff via overland flow above the soil surface; and Q T  is the runoff via through-flow within the soil and the immediate sub-soil. The daily addition or subtraction of MS     from a previous level of soil moisture is termed soil water deficit (SWD) and is defined as ‘ the amount of effective rainfall required to bring the soil back up to 1  A LDF is defined here as a fire that burns a significant proportion of a regional landscape (somewhere between 20 and 100% of the area) with intense fire effects. The potential for an LDF in a region is assumed to be governed by the duration and intensity of levels of seasonal soil and vegetation dryness, paired with the potential for dangerous fire weather conditions. 2  Fuel combustibility refers to the ease at which dead forest fuels will burn during flaming combustion. 3  Fuel flammability refers to the ease with which live forest fuels (particularly leaves and small twigs) can ignite and burn upon being heated by flames (Anderson, 1970). 2  VI International Conference on Forest Fire Research D. X. Viegas (Ed.), 2010 field capacity’ (Mount, 1972:4). A drought index in effect tracks the daily rise and fall in SWD as a seasonal index of soil dryness. Drought indexes are key components of the Australian, Canadian, and American fire danger rating systems (FDRS). The first two FDRSs use a drought index to determine the seasonal availability of forest fuels during a fire season. In the Australian Forest Fire Danger Index (FFDI), the drought factor (DF) model represents the availability of fine dead surface fuels at the forest floor, which is based on a simple curvilinear function of a drought index, essentially SWD in either the KBDI or the SDI, and the number of days since rain (McArthur, 1967, Matthews, 2009). Once the KBDI or SDI exceeds a SWD value of 100 mm, there is no further increase in fire danger rating from the contribution of coarser woody fuels or the live (and now potentially dead) plants in the near-surface, understorey, and overstorey. The increased availability of larger dead fuel or fine live fuel components on crown fire  behaviour and energy release is therefore largely not accounted for in the Australian and Canadian FDRSs. With the possible exception of the American FDRS (Deeming et al ., 1978, Bradshaw et al ., 1983), neither the FFDI nor the Canadian FDRS (Van Wagner, 1987) can accurately estimate the combustibility and potential energy release based on all (live and dead) fuel components in the forest. This largely reflects the srcins of both FDRSs as they were srcinally intended to estimate surface fire spread based on fine dead fuels. Classification systems are now evolving to better quantify all available fuels in a forest with attributes that can be linked to models of fuel availability, combustibility and flammability, as well as fire combustion and energy release rates (Sandberg et al ., 2001, Ottmar et al. , 2007). 1.3 Rationale and Objectives In long unburnt dry and wet sclerophyll forests consumed by the Black Saturday fires, there is strong field evidence that fine and coarse woody debris were totally consumed, indicating that dead fuel moisture contents were between 3 and 15%, depending on fuel particle size. There is also anecdotal evidence of wilting and dead understorey vegetation and increased leaf litter accumulation rates that would not be accounted for in the drought indexes currently used in Australia. The hypothesis  presented in this study is that near-extreme levels in landscape dryness exacerbated fuel availability, combustibility and flammability, which ultimately caused significant increases in rates of spread, fire intensity and energy release on Black Saturday and in other severe LDFs summarised in Section 1.1. After evaluating different drought indexes for Australia, the SWD derived from the MSDI or F– SWDBM were found to best correlate with the occurrence of large destructive fires, as they could be calibrated for use in local fuel types (Gellie, 2009). Given that there has been negligible research on the seasonal relationship between soil water deficit in forests and the potential available heat energy from both live and dead fuels, this paper presents an exploratory discussion of the relationship  between characteristics of landscape level dryness and the extreme fire behaviour events that unfolded on Black Saturday, 7 February 2009. A conceptual framework is presented on how to represent landscape susceptibility to large destructive fires using soil water deficit as a surrogate for seasonal fuel availability and combustibility, and then relating it to potential maximum energy releases during a LDF. Specific objectives were to:    develop long-term time series of SWD in both dry and wet sclerophyll forest to identify  periods of peak landscape susceptibility to LDFs in the last 120–150 years in Victoria, and to compare the severity of landscape dryness conditions of 2008–09 fire season with other significant historical fire seasons;    summarize existing published data knowledge of the live and dead forest–fuel array in the more widespread ecologically mature eucalypt forests in Victoria    discuss landscape dryness and its relationship to fuel bed ignitability, combustibility and vegetation flammability for significant fire events in Victoria fuel types; and 3  VI International Conference on Forest Fire Research D. X. Viegas (Ed.), 2010    suggest relationships between landscape dryness and resulting fire energy release, spread, spotting and convection column development 2 Study Area, Vegetation, and Climate The study area encompasses the hills and ranges 40 to 100 km north and north-east of Melbourne (latitude: -37.84 o S; longitude: 144.98 o ) where the two largest of the Black Saturday fires occurred (the Kilmore East and Murrindindi Fires) (Figure 1). Driven by extreme fire weather conditions on the day, these two fires burnt with catastrophic impact through three dominant broad vegetation–fuel types: grasslands, and dry and wet sclerophyll forests. MelbourneWallaby CreekMelbourne Airport BendigoKilmoreWarragul AlexandraMarysville Victoria SydneyCanberraMelbourne Kilmore East-Murrindindi fire complex   Figure 2 South eastern Australia, the areal extents of the February 2009 fires. The fires and their extents on the day are depicted in an orange-red colour. Source: Landsat ETM image from Geoscience Australia (Geoscience Australia, 2003). Vegetation on the basalt plains around Melbourne were and still are grasslands and patchy grassy woodlands whereas grasslands in the hills and ranges were once largely forests and woodlands, mostly cleared since European settlement on the flatter and more undulating terrain. On the foothills and more exposed west and north-facing slopes, the less productive dry sclerophyll forests of Red Box (  Eucalyptus polyanthemos ) and Broad-leaved Peppermint (  E. dives ) dominate in shallower infertile soils derived from Permian shales and mudstones; they have a low shrubby understorey and prefer areas with annual average rainfall less than 700 mm. Upslope on the mountain escarpments and  plateaux, taller dry sclerophyll forests of Messmate Stringybark (  E. obliqua ) and Narrow-leaved Peppermint (  E. radiata ssp. radiata ) forests between 25 and 50 m in height occur. The understorey in these dry sclerophll forests comprise bracken fern ( Pteridium esculentum)  and herbaceous species in areas with rainfall between 800 and 1200 mm, occupying north and west-facing slopes with deeper yellow and red podzolic soils. Mountain Ash (  E. regnans ) forests to 70 m in height are found on the opposing more sheltered south and eastern-facing slopes. The understorey is usually multi-layered with tall shrub species such as Blackwood (  Acacia melanoxylon ), Silver Wattle (  Acacia dealbata ), and a range of other mesophytic shrubs, such as Blanket Bush (  Bedfordia arborescens ), Musk-Daisy Bush ( Olearia argophylla ), and Hazel Pomaderris ( Pomaderris aspera ), mixed in with tree fern (  Dicksonia 4  VI International Conference on Forest Fire Research D. X. Viegas (Ed.), 2010 antartica ), overlying a vine-herb layer closer to the forest floor. These wetter type forests prefer deeper red kraznozem soils, and the understorey increases in density as elevation increases. The type of vegetation responds strongly to climatic conditions as they change through different elevations. For this study, the range in mean monthly rainfall and temperature are represented by two weather stations (Figure 2). At a lower elevation, the Melbourne station rainfall is distributed evenly across the year, with a small dip in January and February. The mean monthly temperature range is  broad (from 13.0 to 25.6 o C), reflecting the passage and influence of hot dry continental air masses in summer and moist air masses in winter (Figure 2a). The higher elevation station at Wallaby Creek shows a more marked variation in rainfall of almost 60 mm between winter and summer. The mean monthly temperature ranges from 8 to 22 o C corresponds to more cool temperate conditions found in Mountain Ash forest (Figure 2b). (a) 020406080100120140    J  u   l   A  u  g   S  e  p   O  c   t   N  o  v   D  e   J  a  n   F  e   b   M  a  r   A  p  r   M  a  y   J  u  n Month    M  e  a  n  m  o  n   t   h   l  y  r  a   i  n   f  a   l   l   (  m  m   ) 0.05.010.015.020.025.030.0    M  e  a  n  m  o  n   t   h   l  y   t  e  m  p  e  r  a   t  u  r Rainfall Temperature  (b) 020406080100120140    J  u   l   A  u  g   S  e  p   O  c   t   N  o  v   D  e   J  a  n   F  e   b   M  a  r   A  p  r   M  a  y   J  u  n Month    M  e  a  n  m  o  n   t   h   l  y  r  a   i  n   f  a   l   l   (  m  m   ) 0.05.010.015.020.025.030.0    M  e  a  n  m  o  n   t   h   l  y   t  e  m  p  e  r  a   t  u  r Rainfall Temperature   Figure 3 Mean monthly rainfall and mean daily temperature at (a) Melbourne (34 m ASL) and (b) Wallaby Creek (488 m ASL) Source: Rainfall data are based on Melbourne (1855-2010) and Wallaby Creek (1885-2010) and temperature data is based on Melbourne (1908-2010) and Wallaby Creek (1953-2010), (Bureau of Meteorology, 2009) Figure 3 gives a long term perspective of the variation in annual rainfall at both stations. 0200400600800100012001400160018002000    1   8   5   5   1   8   6   0   1   8   6   5   1   8   7   0   1   8   7   5   1   8   8   0   1   8   8   5   1   8   9   0   1   8   9   5   1   9   0   0   1   9   0   5   1   9   1   0   1   9   1   5   1   9   2   0   1   9   2   5   1   9   3   0   1   9   3   5   1   9   4   0   1   9   4   5   1   9   5   0   1   9   5   5   1   9   6   0   1   9   6   5   1   9   7   0   1   9   7   5   1   9   8   0   1   9   8   5   1   9   9   0   1   9   9   5   2   0   0   0   2   0   0   5   2   0   1   0 Year     A  n  n  u  a   l   R  a   i  n   f  a   l   l   (  m  m   ) Melbourne Wallaby Creek   5
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x