Knox

Knox & Clarke, Fire response of shrubs in grassy woodlands Fire response syndromes of shrubs in grassy woodlands in the New England
Tableland Bioregion
Kirsten J.E. Knoxa and Peter J. Clarke Botany, School of Environmental Sciences and Natural Resources Management, University of New England, Armidale, New South Wales 2351, AUSTRALIA. a current: Yorke and Mid-North Region, SA Department for Environment and Heritage, PO Box 822, Clare, South Australia, 5453. Email: knox.kirsten@saugov.sa.gov.au Abstract: In fire-prone grassy woodlands, fire response and time to reach reproductive maturity are two traits that can be
used to provide an indication of the minimum interval between fires needed to maintain biodiversity. This study examined
the effects of fire intensity and adult size on shrub mortality together with the primary and secondary juvenile periods of
shrub species in the New England Tableland (NET) Bioregion. Most shrub species resprouted via basal lignotubers
following fire, irrespective of fire intensity and shrub size. The primary juvenile period of most species was found to be
greater than four years and the secondary juvenile period for most resprouting species was less than four years. These
results suggest that a minimal interval between fires of eight years may be needed to maintain shrub species in grassy
woodlands in the NET Bioregion, and that repeated fires at intervals of less than 8 years should be avoided. The time taken
for shrubs in the grassy woodlands of the NET Bioregion to reach reproductive maturity appears to be longer than conspecifics
in other Bioregions. Caution is needed when using data collected from outside a Bioregion to determine minimum fire
frequency thresholds.
Cunninghamia (2004) 8(3): 348–353
Introduction
attain fire-tolerance (obligate seeders). Knowledge of patternsof fire-tolerance and percentage mortality in different size- Knowledge of how plant species respond to fire regimes is classes allows greater understanding of how species will fundamental for management of biodiversity. In fire-prone react under different fire regimes, especially when communities, studies of the fate of the standing population, communities are dominated by species that resprout.
and of the time taken to reach reproductive maturity post-fire, are of two-fold importance. Firstly, when no The time taken to reach reproductive maturity following fire quantitative data are available about the effects of different is important for both obligate seeders and resprouters, as it is fire regimes on communities, species attributes can be used directly related to generation length and will affect the to predict qualitative changes in species composition (Noble capacity of a population to increase in size (Whelan 1995).
& Slatyer 1980). Secondly, the continued study of factors The length of the primary juvenile period (the time taken to affecting the post-fire survival of populations can be used to reach reproductive maturity from seed) is particularly important for obligate seeders as population decline or localextinction could potentially occur if the interval between fires The fire response of an individual plant will either be death is shorter than the time taken to reach reproductive maturity or survival and at a population level this dichotomy is often and to accumulate an adequate seedbank (Cary & Morrison applied to describe species that survive fire (resprouters) and 1995, Keith 1996). The time taken for resprouting those killed by fire (obligate seeders). In reality, however, a individuals to flower is referred to as the secondary juvenile continuum from 0–100% mortality of individuals within a period. Population decline may occur if the time between fires population exists among species (Morrison 1995, Bond & is less than the secondary juvenile period of resprouters, Van Wilgen 1996, Morrison & Renwick 2000).
particularly those species that do not generally display 100% Characteristics of a particular fire, distribution of size-classes post-fire survival (Keith 1996). Age at first reproduction has and the physiological and anatomical features of a species been found to vary among species within a community will affect the percentage mortality of a population post-fire (Benson 1985, Bradstock & O’Connell 1988), between (Whelan 1995). Bond and Van Wilgen (1996) formalised size- populations (Benson 1985) and between individuals within a specific post-fire survivorship into four survival curves. In population (Carthew 1993). Resprouting species generally summary, the first group (type I) acquires fire-tolerance take longer to flower from seed than obligate seeding species quickly, and has little post-seedling mortality; the second (Abbott 1985, Bell 2001). The secondary juvenile period of group (type II) becomes increasingly fire-tolerant with resprouters, however, tends to be shorter than the primary increasing size; the third group (type III) attains fire- juvenile period for resprouters and obligate seeders (Zammit tolerance during juvenile stages, but loses fire-tolerance with age, and the final group (type IV) are those species that never Knox & Clarke, Fire response of shrubs in grassy woodlands Knowledge of the fate of the standing population and the time Eucalyptus melliodora and Eucalyptus viminalis. The to reach reproductive maturity after a fire event has increased dominant species in the herbaceous layer were Poa sieberiana steadily over the past decade for many community types in var. sieberiana and Themeda australis. Australia. However, little is known about the fire ecology of Booroolong Nature Reserve (865 ha in size) is 30 km NW of woody plants in grassy woodlands in eastern Australia (see Armidale (National Parks & Wildlife Service 2002). No review by Clarke 2000). In particular there is a dearth of specific data are available on the climate of Booroolong knowledge about the shrub species occurring in the cool Nature Reserve, although the climate is similar to that of temperate regions in the New England Tableland Bioregion nearby Guyra, average maximum temperature in January of northern NSW (NET Bioregion). Some fire response and (summer) is 24.6oC; average minimum temperature in July maturation observations from coastal grassy communities (winter) is 0.6oC, average yearly rainfall is 884 mm). Areas have been made but it is not known if it is appropriate to burnt in experimental fires ranged from 1290–1310 m extrapolate data from conspecifics in different climates for elevation, on yellow podsolic soils, overlying a metasediment lithology. The areas were generally flat or had a slight to This study addresses four questions: (i) Which species medium slope. No fires have been recorded at Booroolong resprout following fire and where does resprouting arise? (ii) NR for the past 20 years (National Parks & Wildlife Service Does plant size influence post-fire survival for those species 2002) and the area may not have been burnt for more than 50 that show some degree of resprouting? (iii) Does fire years, based on the lack of fire scars. Vegetation structure at intensity affect mortality for those species that show some Booroolong NR ranges from woodland to open forest with a degree of resprouting? (iv) What is the length of the primary sparse to medium density of shrubs in the understorey and a and secondary juvenile periods of shrub species in grassy dense herbaceous layer. The dominant tree species in the study sites were Eucalyptus caliginosa, Eucalyptus dalrympleanasubsp. heptantha, Eucalyptus laevopinea and Eucalyptus radiata. Poa sieberiana var. sieberiana and Themeda australiswere common in the herbaceous layer.
The New England Tableland Bioregion (NET Bioregion) Fire response traits of shrub species covers 3 004 080 ha., about 40% of which, native The fire response of shrub species was determined following vegetation remains (National Land & Water Resources Audit three experimental burns at each Nature Reserve.
2001). About 7.5% of extant native vegetation is within Experimental burning occurred at Imbota NR in Spring 1999, conservation reserves (Benson 1999). Grassy woodlands once and at Booroolong NR in Autumn 2000. Each burn site was covered extensive areas of the NET Bioregion (Benson & approximately 50 × 50 m and had an average fine fuel load of Ashby 2000) but, have been extensively cleared for approximately 8 tonnes/ ha. Before each fire, fine fuel (straw) pastoralism. Remnant patches of grassy woodland occur in was added to half the area of each burn site in order to conservation reserves, travelling stock reserves, roadside increase the fuel loads to at least 16 tonnes/ ha so as to reserves and on private land. Clarke (2003) demonstrated that determine the effects of different fire intensities on plant under pastoralism, shrub abundances decline, although the mortality. Recording the intensity of fires is inherently herbaceous layer remains relatively intact.
difficult (Whight & Bradstock 1999) and intensity was not Fire response studies were conducted in two grassy measured in the current study as the rate of spread could not woodland conservation areas: Imbota Nature Reserve and be determined due to the way the fires were lit (along control Booroolong Nature Reserve. Imbota Nature Reserve (218 ha lines and then burnt towards the centre). Kitchin (2001) used in size) is 10 km SE of Armidale (National Parks & Wildlife comparable fuel loads to that in the high fuel load areas of Service 2001). The average maximum temperature in this study, and achieved fires of moderate intensity, January (summer) in Armidale is 27.1oC, while the average indicating that moderate intensity and low intensity fires minimum temperature in July (winter) is 0.3oC. Average yearly would have been expected in our high fuel load and low fuel rainfall is 790 mm. Areas at Imbota NR where experimental burns were conducted ranged from 1010–1040 m elevation.
Shrubs were tagged before each fire within each burn area, Experimental burns were conducted in areas with yellow and in adjacent unburnt areas, and the size (measured as podzolic soils and metasediment lithology on sites that were basal girth) of individuals was recorded. Three years after generally flat or with a slight to medium slope. The last the fire individuals were assessed for evidence of resprouting, recorded fire in Imbota NR was a small fire in 1969; no records and the position of resprouting noted. Species were exist for the fire history before this (National Parks & classified as obligate seeders when less than 30% of Wildlife Service 2001). Vegetation structure at Imbota NR individuals survived 100% leaf scorch, and as resprouters ranges from woodland to low open forest with a sparse when more than 70% of individuals survived 100% leaf scorch understorey of shrubs and a near-continuous herbaceous layer.
The dominant tree species in the study sites were Eucalyptusblakelyi, Eucalyptus bridgesiana, Eucalyptus caliginosa, Knox & Clarke, Fire response of shrubs in grassy woodlands Table 1. The post-fire response, method of resprouting of shrub species examined. For survival, numbers are the percentage
survival (sample size). n.a. indicates that the life-history trait is not applicable as this species is an obligate seeder.
Fire response
Survival %
Mode of resprouting
Acacia buxifolia subsp. buxifolia Brachyloma daphnoides subsp. glabrum Leucopogon lanceolatus var. lanceolatus Lissanthe strigosa subsp. strigosa Primary juvenile and secondary juvenile periods Observations on primary and secondary juvenile periods were Post-fire response of species and location of dormant buds made in August 2002 and September 2003. Data for theprimary juvenile period of species were collected from Of 31 shrub species recorded (Table 1) 27 species classified naturally recruiting populations and from seed sown in the as resprouters, including 24 species exhibiting 100% survival.
field for another study. Shrubs were examined for evidence Four species were classified as obligate seeders — Acacia that they had reached reproductive maturity at a number of dealbata, Acacia ulicifolia, Cassinia leptocephala and burn sites with different times-since-fire. At Imbota NR, three Cassinia quinquefaria. No mortality of individuals outside sites were 3 years post-fire, three sites were 2.5 years post- fire, three sites were 2 years post-fire and three sites were 1.5 All but three resprouting species resprouted exclusively from years post-fire, when sampled in 2002. At Booroolong NR, basal (lignotuberous) buds. Jacksonia scoparia and Acacia three sites were 2.5 years post-fire, three sites were 2 years implexa resprouted via root suckers and Acacia filicifolia post-fire and three sites were 1.5 years post-fire when resprouted via root suckers and epicormic buds.
Achieving 100% leaf scorch of the mat-forming shrub Pultenaea microphylla proved difficult as the dense prostrateplant appears to be nearly fire resistant. With increased fuel Formal data analyses were only undertaken for species that loads, 100% leaf scorch was induced and resprouting from displayed some variation in post-fire mortality. Plant mortality at the two different fire intensities was comparedusing analyses of deviance (with a binomial error structure) Effects of fire intensity and plant size on mortality using GLMStat (Beath 2001). The relationship between plant Formal analysis was conducted on only those species that mortality and plant size was investigated by assigning displayed variation in mortality — Acacia filicifolia, Cassinia individuals to one of three size-classes based on stem quinquefaria and Olearia viscidula (Olearia sp. aff. elliptica diameter before fire (small: 0–10 cm, medium: 11–20 cm, did display some variation in mortality, but too few and large: > 21 cm diameter) and analyses of deviance (with individuals were available to examine the effects of plant size a binomial error structure) using GLMStat (Beath 2001).
Knox & Clarke, Fire response of shrubs in grassy woodlands Table 2. The post-fire response time to flowering of shrub species examined. * Indicates that no data are available. n.a. indicates
that the life-history trait is not applicable as this species is an obligate seeder
Fire response
Primary juvenile period
Secondary juvenile period
Acacia buxifolia subsp. buxifolia Brachyloma daphnoides subsp. glabrum Hakea laevipes subsp. graniticola Leucopogon lanceolatus var. lanceolatus Lissanthe strigosa subsp. strigosa and fire intensity). Neither fire intensity nor plant size Discussion
significantly affected mortality of Acacia filicifolia (F < 0.001, P > 0.05; F = 2.29, P > 0.05) or Olearia viscidula Fire response of species and the influence of fire intensity (F = 0.123, P > 0.05; F = 1.77, P > 0.05). For Cassinia quinquefaria, fire intensity did not significantly affect The majority of species in this study was found to resprout = 6.67, P > 0.05), but plant size did after fire, with basal resprouting being the most common significantly affect mortality, with the smallest plants having regeneration mode. Most resprouting species exhibited 100% greatest survivorship (F = 14.02, P < 0.01).
survival, and neither fire intensity nor initial size of plants was found to influence mortality of most species. These Primary and secondary juvenile periods of a selection of results differ from previous studies where mortality within a species has been found to be influenced by fire intensity and Acacia ulicifolia (primary juvenile period 3 years), Indigofera plant size (e.g. Moreno & Oechel 1993, Morrison & Renwick australis (4 years) and Hardenbergia violacea (4 years) were 2000). Plant mortality within a species may increase with the only species that had a primary juvenile period of four fire intensity, especially in the smallest size-classes (Morrison years or less (Table 2). Other species examined were not & Renwick 2000). The lack of a relationship between size flowering and evidently have a primary juvenile period greater and mortality may be due to all individuals being of roughly than four years though how long is unknown. Most equivalent age or at least having had ample time to develop resprouting species were found to have a secondary juvenile fire-tolerance. If germination events for shrub species period of less than four years (Table 2).
generally occur after fire, then only a few individuals in thesmallest size-classes would be expected in this study as fireis thought to have been excluded from both reserves for morethan 30 years.
Knox & Clarke, Fire response of shrubs in grassy woodlands Few studies have examined the percentage post-fire survival of co-occurring species in Australian vegetation Results from this study can be used to formulate a baseline communities and of those, few have found such a for the minimum fire interval for grassy woodlands in the preponderance of species with 100% post-fire survival (i.e.
NET Bioregion. The primary juvenile period is a particularly type I species as defined by Bond & Van Wilgen 1996).
important life-history attribute of obligate seeders, as Hodgkinson (1998) examined shrub species in semi-arid population decline may occur if the interval between fires is woodlands of NSW and found survival ranged from 18–90%.
shorter than the time taken to reach reproductive maturity.
Similarly, in open forests in the Sydney region, shrubs and The primary juvenile period for most obligate seeders in this small trees ranged from 0–83% survival under low intensity study was found to be greater than 4 years, suggesting that fires and 0–27% survival under high intensity fires (Morrison population decline will occur in the grassy woodlands if the interval between successive fires is less than 5 years. Keith et Our findings are consistent with a previous landscape scale al. (2002) recommended that the minimum threshold in fire study in the NET Bioregion (Clarke & Knox 2002) that found frequency should also include three reproductive seasons, in 81% of grassy woodland shrub species had a resprouting fire order to accumulate an adequate seedbank. The minimum response. Clarke and Knox (2002) found grassy woodlands threshold for these woodlands would therefore be a fire-free and wet heaths to have a higher proportion of resprouters interval of 8 years. However the minimum threshold is likely than in shrubby forest communities. Various models were to be greater than 8 years, as it is not known when the examined to account for these patterns although none was obligate seeders will reach reproductive maturity; continued able to account for landscape scale differences in resprouting.
monitoring in the future is necessary. The maximum fire-free An interesting feature of the mode of resprouting in our study period could not be determined from this study as data on the was the prevalence of species with basal resprouting longevity of shrubs were not collected. However the very compared with epicormic shooting. Bellingham and Sparrow low densities of shrubs in remnant sites may reflect the (2000) have suggested that such patterns are a result of exclusion of fire from some of these areas for more than 50 selection under frequent intense disturbance. This theory years. Hence the upper limit for exclusion of fire may be in needs to be tested by obtaining better information on the the order of 20–40 years, for species which have fire- disturbance frequencies in grassy woodlands, although this may be difficult to reconstruct given the fragmented nature One species, Acacia filicifolia, was found to have less than of grassy woodland in the present landscape.
100% post-fire survival, and a secondary juvenile periodgreater than 4 years. These results suggest that unless some inter-fire intervals are greater than the time taken for Acacia The primary juvenile period of most species was found to be filicifolia to re-reach reproductive maturity (secondary greater than four years, conversely, the secondary juvenile juvenile period) then population decline might occur.
period was generally found to be less than four years, results The historic fire regimes of grassy woodlands are poorly consistent with previous studies (e.g. Zammit & Westoby known and are difficult to reconstruct (Lunt 2002). There is a dearth of contemporary and historic data on fire regimes in There is limited but growing knowledge about the length of the NET Bioregion. Benson and Ashby (2000) suggested that the primary juvenile period for Australian species, and the grassy woodlands of the NET Bioregion were subjected to primary juvenile period of some of the species in the current relatively frequent burning by Aboriginal people, though, study has been recorded elsewhere. The primary juvenile given the length of primary juvenile periods for obligate period of Hardenbergia violacea has been recorded as one seeders in our study, a history of high fire frequency is year in the Sydney Region (Benson & McDougall 1996); in unlikely for these places. In addition a field survey of the the current study, it was four years. Wark (1997) found the distribution and abundance of shrub species across the primary juvenile period of Indigofera australis to be two years region showed that obligate seeding species were present in in Victoria, but in the current study it appears to be more than 60% of the sites (Knox, unpublished data) suggesting that four years. These differences may be related to climate, with inter-fire intervals were not generally shorter than 8 years.
the NET Bioregion experiencing a shorter growing season, For better management of shrub species in the remaining and plants taking longer to reach reproductive maturity. This vestiges of the NET Bioregion grassy woodlands, further suggestion seems reasonable, as many obligate seeders within quantification of primary juvenile periods and fire responses shrubby woodlands of the NET Bioregion have primary for a broader range of species, including the rare and juvenile periods of greater than 5 years (Clarke, unpublished threatened taxa, is needed. Season of burn may also affect data). Caution is needed when using data about reproductive mortality in a population and this factor needs to be examined experimentally. Studies of the effects of fire Most of the shrubs in this study had a secondary juvenile regimes on other life history stages are also needed to period of less than four years and are similar to a range of develop a clear understanding of population dynamics of shrub species within shrubby woodlands in the NET Bioregion shrubs under different fire regimes. Finally, to test Knox & Clarke, Fire response of shrubs in grassy woodlands suggested minimal fire frequency thresholds in these Clarke P. J. & Knox K. J. E. (2002) Post-fire response of shrubs in the woodlands, manipulative landscape studies of the effects of tablelands of eastern Australia: do existing models explain habitat differences? Australian Journal of Botany 50: 53–62.
Gill A. M. & Bradstock R. A. (1992) A national register for the fire responses of plant species. Cunninghamia 2: 653–660.
Acknowledgements
Hodgkinson K.C. (1998) Sprouting success of shrubs after fire — height dependent relationships for different strategies. Oecologia We thank the staff at Armidale National Parks and Wildlife Service for implementing the experimental fires detailed in Keith D. (1996) Fire-driven extinction of plant populations: a this study. Financial support was provided to K.J.E.K. by an synthesis of theory and review of evidence from Australian Australian Postgraduate Award, NCW Beadle Scholarship and vegetation. Proceedings of the Linnean Society of N.S.W 116: 37– by the Armidale National Parks and Wildlife Service.
Keith D.A., Williams J.E. & Woinarski C.Z. (2002) Fire References
management and biodiversity conservation: key approaches andprinciples. In Flammable Australia: the fire regimes and Abbott I. (1985) Reproductive ecology of Banksia grandis biodiversity of a continent. Bradstock R.A., Williams J.E. & Gill (Proteaceae). New Phytologist 99: 129–148.
A. M. (Eds) (Cambridge University Press: Cambridge, New York) Beath K.J. (2001) GLMStat User Manual Version 5.5. Sydney, Kitchin M.B. (2001) Fire ecology and fire management for the Bell D.T. (2001) Ecological response syndromes in the flora of south- conservation of plant species and vegetation communities in a western Western Australia: fire resprouter versus reseeders. The National Park in Northern NSW, Australia. PhD Thesis, Botanical Review 67: 417–440.
Bellingham P.J. & Sparrow A.D. (2000) Resprouting as a life Lunt I.D. (2002) Grazed, burnt and, cleared: how ecologists have history strategy in woody plant communities. Oikos 89: 409–416.
studied century-scale vegetation changes in Australia. Australian Benson D. & McDougall L. (1996) Ecology of Sydney plant species Journal of Botany 50: 391–407.
part 4: Dicotyledon family Fabaceae. Cunninghamia 4: 553–746.
Moreno J.M. & Oechel W.C. (1993) Demography of Adenostoma Benson D.H. (1985) Maturation periods for fire-sensitive shrub fasciculatum after fires of different intensities in southern species in Hawkesbury sandstone vegetation. Cunninghamia 1: California chaparral. Oecologia 96: 95–101.
Morrison D.A. (1995) Some effects of low-intensity fires on Benson J.S. (1999) Setting the scene: The native vegetation of New populations of co-occurring small trees in the Sydney region.
South Wales. Background Paper No. 1. Native Vegetation Proceedings of the Linnean Society of NSW 115: 109–119.
Morrison D.A. & Renwick J.A. (2000) Effects of variation in fire Benson J.S. & Ashby E.M. (2000) Vegetation of the Guyra 1: 100 000 intensity on regeneration of co-occurring species of small trees in map sheet New England Bioregion, New South Wales.
the Sydney region. Australian Journal of Botany 48: 71–79.
National Land & Water Resources Audit (2001) Australian Native Bond W.J. & Van Wilgen B.W. (1996) Fire and Plants (Chapman & Vegetation Assessment 2001. National Land and Water Resources Bradstock R.A. & O’Connell M.A. (1988) Demography of woody National Parks & Wildlife Service (2001) Imbota Nature Reserve Draft plants in relation to fire: Banksia ericifolia L.f. and Petrophile Plan of Management. Unpublished Report, NSW National Parks pulchella (Schrad) R. Br. Australian Journal of Ecology 13: 505– National Parks & Wildlife Service (2002) Booroolong Nature Reserve Carthew S.M. (1993) Patterns of flowering and fruit production in a Draft Plan of Management. Unpublished Report, NSW National natural population of Banksia spinulosa. Australian Journal of Noble, I.R. & Slatyer, R.O. (1980) The use of vital attributes to Cary G.J. & Morrison D. A. (1995) Effects of fire frequency on plant predict successional changes in plant communities subject to species composition of sandstone communities in the recurrent disturbances. Vegetatio 43: 5–21.
Sydney region: combinations of inter-fire intervals. Australian Wark M.C. (1997) Regeneration of some forest and gully Journal of Ecology 20: 418–426.
communities in the Angahook-Lorne State Park (north-eastern Clarke P.J. (2000) Plant population processes in temperate woodlands Otway Ranges) 1–10 years after the wildfire of February 1983.
of eastern Australia — premises for management. In Proceedings of the Royal Society of Victoria 109: 7–36.
Temperate eucalypt woodlands in Australia : biology, conservation, Whelan R. J. (1995) The ecology of fire (Cambridge University Press: management and restoration. Hobbs R. J. and Yates C. J. (Eds).
(Surrey Beatty and Sons: Chipping Norton NSW) pp. 248–270.
Whight S. & Bradstock R. (1999) Indices of fire characteristics in Clarke P. J. (2003) Composition of grazed and cleared temperate grassy sandstone heath near Sydney, Australia. International Journal of woodlands in eastern Australia: patterns in space and inferences in time. Journal of Vegetation Science 14: 5–14.
Zammit C. & Westoby M. (1987) Seedling recruitment strategies in obligate-seedling and resprouting Banksia shrubs. Ecology 68:1984–1992.

Source: http://www.rbgsyd.nsw.gov.au/__data/assets/pdf_file/0004/58918/Cun8Kno348.pdf