Proceedings of the conference held 8-9 October 1994, Footscray, Melbourne
Biodiversity Series, Paper No. 8
Department of the Environment, Sport and Territories, 1996
2. Effects of fuel reduction burning on fuel loads in a dry sclerophyll forest
Dept. Conservation & Natural Resources, Victoria
This study looked at the extent and duration of reduced fire hazard following fuel reduction burning in a dry sclerophyll foothill in north-central Victoria. Fuels were divided into surface litter, coarse dead fuels, elevated live and dead fuels and bark on trees. Comparisons were made of the effects of spring and autumn burning on the effectiveness of fuel reduction.
Surface litter fuels reaccumulated to pre-burn levels after two to four years. Elevated fuels were slower to recover and may take ten years or more to return to pre-burn conditions. Coarse fuels were not significantly affected by a single low intensity fuel reduction burn. Bark on trees will take an estimated 15-25 years to recover to pre-burn conditions. Litter levels reaccumulated quicker on the spring burnt sites. Fire protection works should be based on a measure of total fire hazard not just litter fuel loads.
Key words: fuel reduction, fuel loads, sclerophyll forest, Victoria, low-intensity burns, fire protection strategy
Low-intensity burning is used in Victoria as a fire protection measure and as a tool to conserve fauna and flora. Approximately 150 000 ha of native vegetation on public land are burnt annually for these purposes, with most of that area being burnt as a fire protection measure. The main objective of burning for fire protection is to reduce fuel levels during mild weather conditions and thereby reduce the intensity and damage of any subsequent wildfire burning under hot, windy conditions.
The amount of fuel reduced, and its rate of recovery to pre-fire levels, is of particular relevance to management because the basis of a fire protection strategy is to keep fuels below specified levels. In spite of the importance of fuel loads to fire protection operations, there is a paucity of fuel load or fuel accumulation information for Victoria.
In this study we are measuring fuel loads in burnt and unburnt forest and quantifying the accumulation rates after fires. Humus, litter, twigs, branches, fallen trees, shrubs and bark on trees are being assessed in spring and autumn. A complete account of the methods and results presented in this summary is given in Tolhurst & Flinn (1992).
Surface fuel loads at the beginning of the study are shown in Table 1.
The low-intensity fires used in this experiment reduced the amount of litter, twigs, shrubs and wire-grass by about 60% and bark on trees by about 30%, but had no significant effect on the amount of humus or coarse fuel components. This was attributed to the difference in drying patterns of the various fuel components; during the mild spring and autumn conditions, only the elevated and loosely compacted litter fuels were dry enough to burn.
|Twigs (6 to 25mm)
|Branches/Logs (>25 mm)
|Control||2.9 (0.6)||10.3 (1.1)||2.2 (0.5)||77.8 (31.8)|
|Autumn||3.0 (0.4)||10.8 (2.0)||1.9 (0.7)||73.5 (28.8)|
|Spring||3.4 (0.5)||11.1 (1.9)||3.1 (0.9)||62.1 (22.3)|
|Average||3.1 (0.3)||10.8 (0.8)||2.4 (0.4)||71.1 (11.5)|
Ninety-five percent confidence interval (±) shown in parentheses.
Total fine fuel is the combined weight of humus and litter.
Coarse fuels (twigs, branches and fallen trees) are not important to the rate of spread of a fire. However, they are important in the mop-up stage of fire control, they influence the degree of stem damage to trees, and they provide habitat for reptiles, small mammals and invertebrates. Coarse fuels which remain unburnt provide islands of remnant fauna and micro-flora which can recolonise the surrounding area after burning. Coarse fuels that burn near the base of trees can sustain a heat load long enough to cause stem damage (Cheney et al. 1990; Buckley & Corkish 1991). By leaving coarse fuels unburnt, adverse biological effects have been minimised, but it has not reduced the difficulty to fire fighters of mopping-up wildfires.
Humus is in an advanced stage of decomposition and therefore contains many decomposing organisms. The drying of the humus layer, brought about by the removal by burning of the overlying litter, the subsequent increased exposure to sun and wind, and redistribution by rain splash following burning, can be expected to reduce the levels and activity of decomposing organisms, and hence the rate of decomposition, until the litter layer is re-established (Baker & Attiwill 1985).
In this study, the rapid accumulation of litter after burning (Fig. 2.1) can be attributed to the non-combustion of 35 per cent of the preburning litter load, the additions from annual litterfall, and the interruption to invertebrate and fungal decomposition of litter for one to two years after the fires (Neumann & Tolhurst 1991). This rapid rate of litter accumulation after burning has been reported in other forest types (e.g. Raison et al. 1983) and is related to the forest age, productivity, amount of scorching, effects of burning on decomposing microbes, and seasonal patterns. The interplay of these various factors has not been described adequately enough to enable accurate prediction of fuel accumulation rates in most forest stands (Walker 1979). Attiwill et al. (1978), working in the Wombat State Forest, found that litterfall, including twigs up to 2 cm diameter, varied between 4.4 and 6.1 t ha-1 yr-1 over a three year period; and that in the Mt Disappointment forest, litterfall averaged 3.6 t ha-1 yr-1 over a two year period with approximately 30 per cent of this litter being twigs. Similar rates of litterfall could be expected in the present study which would indicate that very little of the residual or new litter decomposed during the two years after burning.
Figure 2.1: Variation in litter fuel on the forest floor with time since burning compared with the unburnt control. Spring burning treatment was measured in spring and autumn burning treatment was measured in autumn.
This rapid rate of litter accumulation means that trigger levels for fine fuel of 8 t ha-1 for Protection Priority 1 Zones and 12 t ha1 for Protection Priority 2 Zones used in public land fire management plans in Victoria (O'Bryan 1988) will be exceeded in about two and four years respectively, after a fuel reduction burn. Priority 1 Zones adjoin value assets such as townships and pine plantations, whereas Priority 2 Zones form strategic barriers to the spread of large wildfires. These fire management plans specify the areas in which these trigger levels are to apply, and require that these areas be repeatedly burnt to keep fine fuel loads below the trigger levels. Emphasis on the litter fuel component of the fuel complex in the plans, however, may be overemphasised when it is considered that the two major objectives of fuel reduction burning are to make fire control easier and to minimise fire severity should a wildfire occur. Other fuels that affect the severity of the fire and the ability of fire fighters to control it are wire-grass, shrubs and bark on trees. These other fuel components accumulate much more slowly after a fire, and so the effectiveness of fuel reduction burning from a fire protection point of view may be underestimated if only litter fuels are considered.
Elevated fine fuels such as wire-grass and shrubs take much longer to return to pre-burn cover and height. Based on the initial recovery rates for elevated fine fuels measured in the present study, shrub height may take at least ten years to return to pre-burning conditions and wire-grass height may take at least four years. Cover can be expected to return to pre-burn levels much faster than height, but overall the structure of both of these elevated fuels can be expected to remain significantly altered for at least ten years. This expectation is consistent with Fox et al. (1979), who found that the height of understorey shrubs increased at a constant rate for at least ten years after burning, and with Van Loon (1977), who found that shrub height increased for at least 25 years.
The bark of the messmate stringybark trees burnt readily because of its fibrous nature, vertical arrangement and deep fissures. In isolated instances, the bark did not burn, especially in mild conditions, because there was insufficient surface fuel to carry the fire to the base of the tree. As most of the study areas had not been burnt for 30 to 50 years, the bark on the trees was quite thick (up to 10 cm). Burning significantly reduced both the amount of bark and the fissure depth on the trees in the first rotation fires. As a result of this and the slow growth rate of the bark, most trees from which bark was burnt in the first fire did not burn again in the second rotation fires. The estimated 7 t ha-1 of bark that was burnt from the trees in the first rotation fires added significantly to the amount of short-distance spotting, which increased the difficulty of keeping these fires controlled. No such problems were experienced with the second rotation fires, demonstrating that the fuel reduction burn had given good protection from short-distance spotting for the three years studied, and probably for several more years to come. Under drier and windier conditions, bark, via spotting, may be more important to the forward rate of spread of the fire than was found under the mild conditions during this study and is certainly a key limitation to fire control under most weather conditions. At the present growth rate of 1.26 mm yr-1, bark thickness would take 15 to 25 years to return to pre-burning conditions on the overstorey trees (>30 cm DBHOB). Given the significant bark reduction by low-intensity burning, there is likely to be a period of about ten years during which short-distance spotting will be reduced and fire control made easier should another high- or low-intensity fire occur.
Fine fuel quantities on the control (unburnt) treatments tended to be greater in autumn than in spring. This was to be expected because the greatest period of litterfall is in summer and autumn because this is the driest part of the year when the trees are subject to moisture stress (e.g. Attiwill et al. 1978). Being dry, there is also unlikely to be much litter decomposition at this time. This seasonal difference is exaggerated during extended droughts as found by Simmons and Adams (1986) in 1982/83, when fuel loads were 4.5 t ha-1 above the levels observed in a normal season. Pook (1985) also found that a severe drought caused the levels of leaf shedding in a eucalypt forest to be 3.5 times greater than average, with up to 97 per cent of the total tree leaf area being shed.
Low-intensity fires in this study reduced the litter, twig, elevated fuels and bark on trees. A single fire did not significantly affect the coarse fuels or the fragmented humus fuels. Both spring and autumn burning were similar in this regard.
Leaf and twig litter accumulated quickly after burning, and the levels were not significantly different from those in the unburnt areas within a two to four year period of the fires. Elevated fuels, such as the shrubs and wire-grass, were much slower to recover, and may take ten years or more to return to unburnt conditions. Bark on trees will take an estimated 15-25 years to recover to pre-burn conditions, and it was found that almost no burnt trees had enough bark to support a second fire within three years. The effectiveness of the low-intensity fires in reducing the fire hazard therefore persists for longer than the effect on litter fuels alone would indicate. Indeed, litter fuel load underestimates the fire hazard by ignoring elevated fuels and bark on standing trees. The current trigger levels used in Priority 1 and Priority 2 areas should therefore be replaced with a measure of fire hazard which includes litter fuel, shrub fuels, bark on standing trees and other elevated fuels.
Litter loads in autumn were generally greater than those in spring. Fire intensities in late summer and autumn could therefore be expected to be marginally greater than they are in late spring and early summer in any given area under a given set of weather conditions. This difference would be more marked in drought years.
I wish to thank the staff and field crews of the Department of Conservation and Natural Resources for their cooperation in the planning of this experiment and for logistic support in performing the burning operations. I also thank Don Oswin, Amanda Ashton, Chris Norman, Tony Morris, Kevin Brooker and the other staff of the Creswick Forest Research Station for their enthusiasm and physical support in conducting this research.
Attiwill, P.M., Guthrie, H.B. & Leuning, R. 1978, 'Nutrient cycling in Eucalytpus obliqua (L'Herit) forest. I Litter production and nutrient return', Australian Journal of Botany, vol. 26, pp. 79-91.
Baker, T.G. & Attiwill, P.M. 1985, 'Loss of organic matter and elements from decomposing litter of Eucalyptus obliqua L'Herit and Pinys radiata', Australian Forest Research, vol. 15, pp. 309-319.
Buckley, A.J. & Corkish, N.J. 1991, Fire Management Branch, Research Report No. 29. Department of Conservation and Environment, Victoria. 28 pp.
Cheney, N.P., Gould, J.S. & Hutchings, P.T. 1990, 'Eucalypt Regrowth Management', Department Conservation & Environment, Victoria, Technical Report, No 12.
Fox, B.J., Fox, M.D. & McKay, G.M. 1979, 'Litter accumulation after fire in a eucalypt forest', Australian Journal of Botany, vol. 27, pp. 157-165.
Neumann, F.G. & Tolhurst, K.G. 1991, 'Effects of fuel reduction burning on epigeal arthropods and earthworms in dry sclerophyll eucalypt forest of west-central Victoria', Australian Journal of Ecology, vol. 16, pp. 315-330.
O'Bryan, D. 1988, Rationale for selecting trigger levels for priority 1 and 2 zones – Address to Regional Fire Protection Officers 19/7/88, Department Conservation, Forestry & Lands, File 85/627, (unpub.).
Pook, E.W. 1985, 'Canopy dynamics of Eucalyptus maculata Jook III. Effects of drought', Australian Journal of Botany, vol. 33, pp. 65-79.
Raison, R.J., Woods, P.V. & Khanna, P.K. 1983, 'Fuel dynamics in recurrently burnt eucalypt forests', Institute Foresters of Australia, 10th Triennial Conference, Melb. pp. 59-64.
Simmons, D. & Adams, R. 1986, 'Fuel dynamics in an urban fringe dry sclerophyll forest', Australian Forestry, vol. 49, no. 3, pp. 149-154.
Tolhurst, K.G. & Flinn, D.W. eds., 1992, Ecological impacts of fuel reduction burning in dry sclerophyll forest: First progress report. Forest Research Report No. 349, Department Conservation & Envirnonment, Victoria.,
Van Loon, A.P. 1977, Bushland fuel quantities in the Blue Mountains – litter and understorey. N.S.W. Forestry Commission, Research Note No. 33,pp.1-22.
Walker, J. 1979, Aspects of fuel dynamics in Australia, CSIRO, Institute of Earth Resources, Division of Land Use Resources, Technical Memorandum 79/7, pp.1-26.