Biodiversity publications archive

Country in flames

Proceedings of the 1994 symposium on biodiversity and fire in North Australia - Biodiversity series, Paper no. 3
Deborah Bird Rose (editor)
Biodiversity Unit
Department of the Environment, Sport and Territories and the North Australia Research Unit, The Australian National University, 1995

Why the skillful use of fire is critical for the management of biodiversity in Northern Australia

David Bowman
Conservation Commission of the NT, Darwin


This paper is divided into two parts. The first section presents the argument in support of my contention that skillful fire management is critical for the management of biodiversity in Northern Australia. The second part considers problems standing in the way of skillful fire management, and philosophical issues associated with the destruction of biodiversity due to a failure to adequately manage landscapes.

The argument

Biodiversity is often defined as the diversity of living things on Earth. But just as it would be senseless to attempt to regulate the global economy by making a collection of all the different coins in the world, so it is senseless to attempt to understand how to manage biodiversity by ignoring ecological process and fixating on biological inventories. Both ecosystems and economies are driven by processes. In order to manage biodiversity one must influence the nature of ecological processes. Understanding processes necessarily involves the consideration of time, particularly rates of change.

A superb example of ecological processes are the consequences that follow the combustion of vegetation by fire. These consequences occur across a wide range of time scales: from events that occur during the passage of flames through to much longer-term effects such as the evolution of species with features that enable them to survive particular frequencies and intensities of wildfire. Given this huge cross-section through time, it is obvious that a comprehensive and perfect knowledge of the effects of fire on ecosystems is impossible given our short lifespans, our inability to travel through time, and, in the case of the European colonists of Northern Australia, their brief acquaintance with this region's environment.

Nonetheless, these difficulties have not deterred many non-Aboriginal people with an interest in land management from holding strong views about the effect of fire on Northern Australian landscapes. For example, the following ideas have wide currency: that North Australia was originally covered by a huge rainforest, and that the North would return to this original pristine condition if it were not for fire. One version of this idea is that the reason for the minuscule size and fragmentary distribution of rainforests in Northern Australia is that Aboriginal people destroyed the original rainforests through their use of fire (Kershaw 1985). Although these are interesting academic hypotheses, they are supported by a weak chain of inference: rainforests were once widespread rainforests currently have a fragmentary geography rainforests are destroyed by fire Aborigines use fire to burn landscapes therefore Aborigines destroyed the rainforests.

Recent research has cast grave doubt on the above model of Aborigines as destroyers of the great rainforest. It is now doubtful that rainforests were ever widespread in Northern Australia. Indeed there is some biogeographic evidence to suggest that in fact fire-adapted savanna may be very ancient (Bowman, Woinarski & Menkhorst 1993). Be that as it may, what is clear is that the absence of fire in Eucalyptus savannas does not result in their conversion to rainforest (Bowman & Panton 1993c). Further, the fragmentary distribution of rainforests may actually reflect the constituent species' superb ability to disperse and colonise vacant, albeit fire protected, niches in the landscape (Russell-Smith & Dunlop 1987). Perhaps the most damning of the above model is evidence that the current destruction of fire sensitive vegetation types is almost certainly the result of the cessation of skillful Aboriginal burning (Bowman & Panton 1993a, Price & Bowman in press, Bowman 1994).

Figure 1. The results of a computer simulation study that shows the response of a stand of Callitris intratropica to various fire frequencies and intensities over the past 300 year period.

Figure 1. The results of a computer simulation study that shows the response of a stand of <em>Callitris intratropica</em> to various fire frequencies and intensities over the past 300 year period.

The histograms are symbolic portrayals of different size-class distributions that are produced by the various fire regimes. For more details see Price and Bowman (in press).

Both field research and mathematical modelling have shown that on sites without topographic fire protection, healthy populations (that is forests with a balance of seedlings, saplings and trees) of northern cypress pines (Callitris intratropica) can develop only under a very specific fire regime of frequent mild fires (Fig. 1). Infrequent fires result in cypress pine forests that are densely stocked with juveniles. This situation has occurred in an area in western Arnhem Land where a forestry program stopped all fires in an area dominated by Callitris intratropica (Bowman, Wilson & Davis 1988), and in southern Australia where grazing has resulted in a cessation of wildfires.

Conversely very frequent burning ultimately eliminates all size classes and results in the local extinction of Callitris (Fig. 1). The very specific fire regimes required to support healthy populations of cypress pines would not occur under natural conditions. Lightning strikes at the end of the dry season are a common cause of intense bushfires which are highly destructive of cypress pine forests (Bowman 1988).

The local extinction of Callitris is currently occurring in vast areas throughout the Top End. The decline of the cypress pines is occurring on all land tenures: Aboriginal land, pastoral land and national parks (Bowman & Panton 1993a). This tree species is most probably like the miners' canary, signalling that fundamental ecological changes are occurring in response to the breakdown of traditional Aboriginal land management and a shift to intense fires, many of which are deliberately lit. If this conclusion is correct then the conservation of biodiversity in Northern Australia will hinge on land managers returning to fire regimes that approximate those used by Aboriginal people.

The return to land management that approximates that of traditional Aborigines is not an easy matter given the ecological changes that have occurred over the last 150 years of European colonisation. Large areas of land have not been burnt by traditional methods due to the galaxy of impacts associated with Europeans (eg decline of Aboriginal populations due to disease, prohibition of traditional burning by pastoralists, and the adoption of more sedentary lifestyles by many Aboriginal people). Further, since European colonisation, ecological processes have changed and remain dynamic, owing to the introduction of feral and domestic megafauna like cows, horses, donkeys, etc, and the spread of weeds. For example, there is evidence that some species (such as native annual Sorghum species or spear grass) have become more abundant (Bowman & Panton 1993b). These species, and the exotic Pennisetum polystachion or mission grass, are highly flammable and their increased biomass results in more frequent, severe and destructive fires (Bowman & Wilson 1988, Panton 1993).

Technological approaches to land management may not be able to mimic traditional Aboriginal burning practices. Setting fires from aircraft cannot simulate the effect of traditional Aboriginal people burning their country during the course of their annual round. The cost of paying people to burn landscapes on foot may be prohibitively expensive, although this method is practical for specific localities (eg burning around camping grounds, rainforests, rock art sites and along bushwalking tracks).

There is no doubt that the arrival of Aboriginal people some 60 000 years ago also caused substantial changes to the ecology of Australia: the early Aboriginal settlers probably made mistakes, and caused some extinctions. However, I believe the diversity of life that existed in Australia 200 years ago was the direct product of skillful traditional Aboriginal land management. Because the ecology of Northern Australia has changed, and because the way people use Northern Australian landscapes has changed since European colonisation, I argue that Aborigines and non-Aborigines alike must now learn (or re-learn) to manage the North Australian landscapes of today. We can do this by informing ourselves of traditional Aboriginal ecological knowledge, applying the scientific method and learning from our mistakes: a process of 'adaptive management'.

The rich natural bounty of living things unique to Northern Australian is both a monument to one of the greatest ecologically sustainable cultures, and a challenge to our technological civilisation. Without skillful management we may look forward to landscapes populated by weeds, vermin and feral animals.

Issues arising from the argument

Fundamentally land management is about making choices. Land managers need to be clear about objectives and must operate within tight budgetary constraints. The achievement of some objectives is easy while others are impossible. For example earlier this century it was decided to eradicate the Tasmanian tiger because it was perceived to be an agricultural pest: since 1936 there have been no definite records of this creature and it is presumed to be extinct. The objective of ridding Tasmania of a pest was apparently achieved, although in hindsight many grieve for this remarkable creature. Pigs are a serious agricultural and ecological pest on the Australian mainland, and feral pigs may become reservoirs of diseases that could damage the nation's livestock industry. There is a clear need to exterminate feral pigs. To date, however, our civilisation has been unable to control this feral animal. Thus although land managers can identify their goals, they may be unable to achieve them. Similarly, ideal land management for the whole of Northern Australia might be a return to traditional Aboriginal fire regimes, but because of ecological changes associated with European colonisation such an objective would be impossible to achieve for the entire region. Finally, a paradoxical feature of land management is that not making a decision is a decision in itself. For example if we do not attempt to control fire, then those species that require fire management will be disadvantaged.

Where do we go from here? In a sense, science has met a boundary, as further refinement of the above argument will do little to improve things in the North Australian landscapes. Of course applied research is required to determine what are the most effective fire-management strategies to achieve particular ends, but this is technological rather than scientific research. It is now the time for society to demand better fire management, be that through education, increased funding for land management agencies, or changes in administrative relationships between government bureaucracies, pastoralists and Aboriginal Land Councils. Fundamentally the problem of how to manage fire to conserve biodiversity is not a scientific matter. The difficult choices regarding what will be managed and what will not, need to be made by society through 'the political process' (eg agitation of community groups, political leadership, or administrative reform). This leaves scientists in a perplexed state, for in a sense the scientists have dealt themselves out of the pack. Championing the need for land-management, or indeed becoming a land manager, are noble activities but these tasks result in the cessation of the one activity which motivates a scientist the practice of science.

But what is science? For me it is a rational dialogue with mystery: our lives are so cluttered by technology that we forget how little we know of our origin and our destiny. Similarly ecological science can be thought of as a way of 'talking to country'. Over the last ten years I have been privileged to interact with North Australian landscapes. I first learnt to overcome my initial fear (a blend of alienation and cultural revulsion?) of the strange vegetation and discordant bird calls perhaps it was the light that seems to burn away all colour to leave a midday delirium. But then you learn of a light that calms before darkness falls, a light so soft that it reveals subtle folds of land, and makes small details (a rock, a shrub, a pool) resonate perfection. Country 'speaks' in many ways to those who are prepared to pay attention.

Through science I have learnt some of the landscape's story. I have heard a landscape crying out for management. Therefore I feel obliged to communicate that message. The observed has influenced the observer: like the quantum physicists, ecologists fundamentally lose absolute objectivity, and like the original atomic physicists some ecologists may comprehend sin. The decisions we make today not only affect the continued existence of long lineages of life forms, but in a sense those decisions directly affect us. Why else does extinction bother us so much?


Bowman DMJS, 1988. Stability amid turmoil? Towards an ecology of north Australian eucalypt forests, Proceedings of the Ecological Society of Australia 15, 149 158.

Bowman DMJS, 1994. Preliminary observations on the mortality of Allosyncarpia stems on the Arnhem Land Plateau, northern Australia, Australian Forestry 57, 62 64.

Bowman DMJS & Panton WJ, 1993a. Decline of Callitris intratropica in the Northern Territory: implications for pre- and post-European colonisation fire regimes, Journal of Biogeography 20, 373 381.

Bowman DMJS & Panton WJ 1993b. Differences in the stand structure of Eucalyptus tetrodonta forests between Elcho Island and Gunn Point, northern Australia, Australian Journal of Botany 41, 211 215.

Bowman DMJS & Panton WJ, 1993c. Factors that control monsoon-rainforest seedling establishment and growth in north Australian Eucalyptus savanna, Journal of Ecology 81, 297 304.

Bowman DMJS & Wilson BA, 1988. Fuel characteristics of coastal monsoon forests, Northern Territory, Australia, Journal of Biogeography 15, 807 817.

Bowman DMJS, Wilson BA & Davis GW, 1988. Response of Callitris intratropica to fire protection, Murgenella, northern Australia, Australian Journal of Ecology 13, 147 159.

Bowman DMJS, Woinarski JCZ & Menkhorst KA, 1993. Environmental correlates of tree species diversity in Stage III of Kakadu National Park, northern Australia, Australian Journal of Botany 41, 649 60.

Kershaw AP, 1985. An extended late Quaternary vegetation record from north-eastern Queensland and its implications for the seasonal tropics of Australia, in MG Ridpath & LK Corbett (eds), Ecology of the Wet-Dry Tropics, Proceedings of the Ecological Society of Australia, Vol. 13, Ecological Society of Australia, Canberra, 179 189.

Panton WJ, 1993. Changes in post World War II distribution and status of monsoon rainforests in the Darwin area, Australian Geographer 24, 50 59.

Price O & Bowman DMJS, in press. Fire-stick forestry: a matrix model in support of skilful fire management of Callitris intratropica by north Australian Aborigines, Journal of Biogeography.

Russell-Smith J & Dunlop CR, 1987. The status of monsoon vine forest in the Northern Territory: a perspective, in The Rainforest Legacy: Australian National Rainforests Study, Volume 1. Special Australian Heritage Publication Series No. 7, AGPS, Canberra, 227 304.