Self-organised instability and megafaunal extinctions in Australia

FORUM FORUM FORUM FORUM is intended for new ideas or new ways of interpreting existing information. It provides a chance for suggesting hypotheses and for challenging current thinking on ecological issues. A lighter prose, designed to attract readers, will be permitted. Formal research reports, albeit short, will not be accepted, and all contributions should be concise with a relatively short list of references. A summary is not required. Self-organised instability and megafaunal extinctions in Australia Michael A. Forster, PO Box A175, Sydney South, NSW 1275, Australia (micf@rpb2.com). It is generally known that during the Quaternary there were wide spread extinctions of fauna on all continents except for Antarctica. The coinciding of these extinction events with the arrival of humans on each of the respective continents has led to the hypothesis that humans hunted this fauna into extinction (Flannery 1994). Termed the blitzkrieg or overkill hypothesis (Martin 1984), such was invoked by Flannery (1994) to explain the extinction of the megafauna in Australia. However, the applicability of the overkill hypothesis to Australia was brought under intense and serious scrutiny due to problems with the timing of the extinctions and human arrival (Grayson 1990), and the lack of any mass-kill sites such as has been found in New Zealand and North America (Owen-Smith 1989). Opponents to the human caused extinction have often cited the fluctuating climate characteristic of the Pleistocene as the causal agent in the demise of the Australian megafauna (Horton 1977). Problems with correlating extinction events with severe climatic conditions have produced incongruence in the climateinduced hypothesis (Flannery 1994). Consequently, a third hypothesis, the anthropogenic disruption hypothesis, which states that Aboriginal burning modified habitats enough to bring about a change in the ecology of the megafauna and their eventual demise, has been put forward (Miller et al. 1999). Due to the fact that the megafaunal extinction debate in Australia has reached a certain stalemate relative to the debate in other regions of the world, a rather different, if not heuristic, hypothesis is proposed here to stimulate an alternative line of inquiry. The hypothesis has been largely developed from recent advancements in the study of system complexity. I propose that an increase in immigration of fauna from south-east Asia and speciation of the existing local fauna in response to the increasing aridity of the Pliocene led the faunal assemblage of Australia to a level of self-organised instability, as defined by Solé et al. (2002a). The arrival of humans into Australia increased significantly the level of connectivity between OIKOS 103:1 (2003) the faunal elements in the Australian faunal system. Subsequently, the fauna of Australia was inherently susceptible to an extinction event. The purpose of this paper is to outline the logic behind this hypothesis, from here to be termed the self-organised instability (SOI) hypothesis of megafaunal extinctions, and the conditions that allow it to be applicable to Australia. Self-organised instability SOI was conceptualised by Solé et al. (2002a) as an extension of Bak (1996) self-organised criticality (SOC). It was determined that certain facets of SOC do not apply to ecological systems, such as the influence of external perturbations, and therefore SOI was proposed (Solé et al. 2002a). Solé et al. (2002a) explained SOI as follows: Under the constant immigration of species, diversity increases until a critical number of resident species is reached. Now instability acts as a barrier preventing further increases in diversity (although species turnover is observed). In other words, species number will reach a certain level in any given system and will subsequently be maintained through the immigration and/or speciation of new species and the extinction of existing species (Solé et al. 2002a). Extinction is a consequence of the increase in connectivity (amount of interaction) between species brought about by the assumption that as the number of species increases the number of interactions within the system also increases. In effect, the feedback between the parameters immigration, diversity, connectivity and extinction lead an ecological system to SOI. SOI implies that ecosystems may be close to saturation and this is a point explored by Levin (1999). Evidence for SOI SOI is an ecological theory still very much in its infancy and as such there is not widespread evidence for it. 235

Nevertheless, Solé et al. (2002a) simulated data to show that an exponential relationship exists between species diversity and connectivity. As mentioned above, SOI is an extension of SOC and there have been studies showing ecological communities exhibiting this latter phenomenon. A study by Keitt and Marquet (1996) claimed to demonstrate SOC in the avifauna of Hawaii. Through examining historical records for the timing of the arrival of exotic avifauna to the Hawaiian Islands, a critical point was found where the addition of one more species to the system triggered an extinction event among the native avifauna. Such a process has been envisioned for the extinction of the megafauna in Australia. For this theory to be applicable in Australia two conditions need to be met. Firstly, it must be shown that Australia went through a period of immigration and speciation prior to the megafaunal extinction event that could have driven the system to SOI. And secondly, a high level of connectivity between the fauna and other biological organisms in Australia, which increases the susceptibility of elements within the system to an extinction event, needs to be demonstrated. Immigration Between approximately 40 mya and 4.5 mya Australia was geographically isolated from the rest of the world. Although bats and birds managed to immigrate to Australia during this period there is no evidence for a sustained immigration of terrestrial fauna. Around 4.5 mya there were the first indications of terrestrial placental mammals, members of the family Rodentia, having immigrated to Australia most likely on debris from south-east Asia (Archer and Wade 1976). Following this first wave of immigration, Hand (1984) has cited up to two subsequent waves of faunal immigration. As for the immigration of humans, the exact timing of human arrival into Australia is still largely debated. Current consensus lies between 40 000 and 60 000 years before present (Horton 1991, Lourandos 1997) and this approximately coincides with the extinction of the megafauna (Roberts et al. 2001). Speciation While there was unprecedented immigration during the Pliocene and Pleistocene this coincided with the extensive development of the Australian arid zone (Bowler 1982, Martin 1998). The climate of the Pliocene and Pleistocene has been characterised by increasing aridity and variability (Truswell 1993) and hence the Australian biota needed to adapt to those changing conditions. The Pliocene was particularly a period of faunal speciation as the Australian biota adapted to the drying conditions. The arid and semi-arid zone faunal assemblage was largely formed during this period coincident with further speciation in the Macropodoidea that had commenced during the Miocene (Flannery 1984). Westerman and Krajewski (2000) discuss speciation of bandicoots in the genus Isoopodon during the Pliocene and Pleistocene whereas Kirsch et al. (1997) list the radiation of the genus Petrogale (rock-wallabies) at 3 mya as a significant event in marsupial evolution. There are many other examples of faunal speciation in Australia during this period. Connectivity Connectivity, in this context, is a rather ambiguous term that can have much variability between as well as within species. In spite of this ambiguity connectivity is defined as the level of interaction between species (Solé et al. 2002a). Interactions in ecology are largely seen in the form of competition, predation, parasitism, mutualism and detritivory (Begon et al. 1996). In Australia, congruent with these ecological interactions, it is increasingly being recognised that many interactions are found in the forms of co-operation within species and between species. However, in regards to the SOI hypothesis of megafaunal extinctions, it is important to note the high level of connectivity that humans have with other organisms. Interactions between Australian Aboriginals and the Australian biota were extensive with many organisms perhaps relying on Aboriginal ecology for their existence (Flannery 1994). In the Blue Mountains west of Sydney, for instance, Aborigines may have consumed no less than 72 plants, 45 mammals, 235 birds, 6 fish, 16 reptiles, 29 frogs and numerous invertebrates (Merriman 1993). Hynes and Chase (1982) hypothesised that Aboriginal domiculture structured vegetation communities in the Cape York region of Queensland. Extensive interactions have also been noted by Latz (1995) in central Australian Aborigines. Therefore, just as modern humans have some form of connectivity with every other organism on the planet, either directly or indirectly, it is very conceivable that the early Aboriginals had a similar connectivity with Australian organisms. The SOI hypothesis of megafaunal extinctions The justification for each of the parameters outlined above having been made, a scenario that may best describe the extinction of the Australian megafauna in light of the SOI hypothesis of megafaunal extinction will now be proposed. 236 OIKOS 103:1 (2003)

During the Pliocene and Pleistocene an increasing number of species were immigrating to Australia largely due to its closer proximity to south-east Asia. Coupled with speciation of these immigrants and speciation in the existing faunal stock the Australian faunal system was led to a level of diversity characteristic of SOI. This diversity also led to an increase in the interactions, or connectivity, of the species concerned. During the Pliocene and Pleistocene Australia was evolving to a state close to the critical point of SOI that would leave the system increasingly susceptible to a mass extinction event. In the meantime there was probably a degree of species turnover, depicted by Solé et al. (2002a), as further immigration and speciation placed pressure on the faunal system. The arrival of the highly adaptable and versatile human immigrant, with their ability to form many connections and interactions with all other biota in Australia, pushed the system into a super-critical state where extinctions could occur relatively easily. The Australian system was poised for a mass extinction event after humans arrived because of the broad generality in human ecology. There did not have to be extensive interactions between organisms before the arrival of humans but the arrival of humans caused a large increase in the subsequent interactions. By this reasoning, humans did not have to hunt megafauna per se. Rather there just had to be some form of interaction, benign, malign or otherwise, with the biota of Australia. The mere presence of humans in Australia, because of the unique ecology of humans in general, ultimately caused the extinction of the megafauna. Conceptually this process is captured in Fig. 1(a,b). Fig. 1a shows the general model of SOI outlined by Solé et al. (2002a). Fig. 1b, on the other hand, depicts the SOI hypothesis of megafaunal extinction in Australia proposed here. With the additional element of humans in the model the number of interactions between organisms in Australia increased dramatically leaving the Australian faunal system in a super-critical state susceptible to a mass extinction event. Discussion The SOI hypothesis of megafaunal extinctions has the potential to explain a wide range of unresolved issues in the Australian megafauna debate. The hypothesis can explain extinctions without resort to the overkill hypothesis. It is most likely that the early Aboriginals were probably just learning how to survive in their new environment, building interactions, rather than hunting fauna on a grand scale into extinction. There is also no need for climate to be invoked, however the hypothesis of anthropogenic disruption does explain to a certain degree the extinctions in light of the SOI hypothesis. SOI, as a unifying ecological theory, also has the potential to explain a range of ecological phenomena from species abundance relations to patterns in species diversity. The hypothesis developed here has similarities to those developed for the mass extinctions such as occurred at the end of the Permian (Lewin 1993). Conceptually the pieces appear to fit rather conveniently in regards to the extinction of the Australian megafauna but the SOI hypothesis of megafaunal extinctions should be judged with extreme caution. There are a number of reasons for this. Firstly, SOI as a theory in itself is in its infancy and it is unclear at this stage whether or not it has widespread applicability to real ecological systems. Solé et Fig. 1. a) The self-organised instability model after Solé et al. (2002a) and; b) The self-organised instability hypothesis of megafaunal extinctions in Australia where the added element of humans increased interactions so the system was in a super-critical state susceptible to a mass extinction event. OIKOS 103:1 (2003) 237

al. (2002a) acknowledge the oversimplifications of reality made within their model. Specifications such as physiology and energetics of organisms were not explicated and these could have certain ramifications for the implications inherent in the model. Solé et al. (2002b) propose an alternate ecology-based model that explicitly includes a speciation and external perturbation component that, in an evolutionary context, is present in the hypothesis presented here. Similarly, Alroy (2001) has undertaken modelling on the American megafaunal extinction event. Secondly, the SOI hypothesis of megafaunal extinctions has been conceptualised using the results found in the Hawaiian avifaunal study of Keitt and Marquet (1996). These authors express concern at validating SOC in real ecological communities due to, among other factors, sampling issues (T. Keitt, pers. comm.). The SOI hypothesis of megafaunal extinctions may eventuate to be merely a theoretical curiosity. The major issue with the hypothesis is not so much its fundamental logic but how it can be validated with rigorous scientific data. The patterns found in the Hawaiian avifauna could be discerned because of the detailed historical, biological and ecological records gathered. Such detailed information for the megafauna of Australia may never be forthcoming due to the ambiguities in the fossil record generally and the antiquity of the extinction event. In fact, along with issues of timing, this is perhaps the major reason why there has been no consensus or synthesis between the overkill, climate-induced and the anthropogenic modification hypotheses in the Australian megafaunal extinction debate. There are also doubts as to the overall influence of immigration and speciation during the evoked period. There are only three periods in the prehistory of Australia where the fossil record is extremely detailed: the Miocene, early Pliocene and Pleistocene (Archer 1984). The immigration and speciation events discussed here may merely be an artefact of the available fossil record. Proving the exact nature and influence of immigration and speciation may be near impossible with the data currently available. However, the SOI hypothesis of megafaunal extinctions does provide grounds for mathematically derived models and computer simulations. Choquenot and Bowman (1998) undertook such an approach in regards to the overkill hypothesis through the development of a predator-prey model. This model found that is was unlikely that Aborigines hunted the megafauna into extinction. A similar course of research may lead to the validation or invalidation of the SOI hypothesis of megafaunal extinctions in Australia. Acknowledgements The author would like to acknowledge Nadia Toppler for her continued support and constructive criticism on a draft copy of this paper and Bob Forster for his assistance on a revised version. References Alroy, J. 2001. A multispecies overkill simulation of the end-pleistocene megafaunal mass extinction. Science 292: 1893 1896. Archer, M. 1984. The Australian marsupial radiation. In: Archer, M. and Clayton, G. (eds), Vertebrate zoogeography and evolution in Australasia. Hesperian Press, pp. 633 808. Archer, M. and Wade, M. 1976. Results of the Ray E. Lemley Expeditions, Part 1. The Alingham formation and a new Pliocene vertebrate fauna from northern Queensland. Mem. Qld. Mus 17: 379 397. Bak, P. 1996. How nature works: the science of self-organized criticality. Copernicus. Begon, M., Harper, J. L. and Townsend, C. R. 1996. Ecology: individuals, populations and communities. Blackwell Science. Bowler, J. M. 1982. Aridity in the late Tertiary and Quaternary of Australia. In: Baker, W. R. and Greenslade, P. J. M. (eds), Evolution of the flora and fauna of arid Australia. Peacock Publications, pp. 35 44. Choquenot, D. and Bowman, D. M. J. S. 1998. Marsupial megafauna, Aborigines and the overkill hypothesis: application of predator-prey models to the question of Pleistocene extinction in Australia. Glob. Ecol. Biogeogr. 7: 167 180. Flannery, T. 1984. Kangaroos: 15 million years of Australian bounders. In: Archer, M. and Clayton, G. (eds), Vertebrate zoogeography and evolution in Australasia. Hesperian Press, pp. 817 836. Flannery, T. 1994. The future eaters: an ecological history of the Australasian lands and its people. Reed New Holland Publishers. Grayson, D. K. 1990. Commentary on Flannery (1990). Archaeology in Oceania 25: 60 61. Hand, S. 1984. Australia s oldest rodents: master mariners from Malaysia. In: Archer, M. and Clayton, G. (eds), Vertebrate zoogeography and evolution in Australasia. Hesperian Press, pp. 905 912. Horton, D. R. 1977. Extinction of the Australian megafauna. Aust. Inst. Aborig. Stud. News 7: 72 75. Horton, D. R. 1991. Recovering the tracks: the story of Australian archaeology. Aboriginal Studies Press. Hynes, R. and Chase, A. 1982. Plants, sites and domiculture: Aboriginal influence upon plant communities in Cape York Peninsula. Archaeology in Oceania 17: 38 50. Keitt, T. and Marquet, P. 1996. The introduced Hawaiian avifauna reconsidered: evidence for self-organized criticality? J. Theor. Biol. 182: 161 167. Kirsch, J. A. W., Lapointe, F. J. and Springer, M. S. 1997. DNA-hybridisation studies of marsupials and their implications for metatherian classification. Aust. J. Zool. 45: 211 280. Latz, P. 1995. Bushfires and bushtucker: Aboriginal plant use in Central Australia. Iad Press. Levin, S. 1999. Fragile dominion: complexity and the Commons. Perseus. Lewin, R. 1993. Complexity: life at the edge of chaos. Phoenix. Lourandos, H. 1997. Continent of hunter-gatherers: new perspectives in Australian prehistory. Cambridge University Press. Martin, H. A. 1998. Tertiary climatic evolution and the development of aridity in Australia. Proc. Linn. Soc. NSW 119: 115 136. Martin, P. S. 1984. Prehistoric overkill: the global model. In: Martin, P. S. and Klein, R. G. (eds), Quaternary extinctions: a prehistoric revolution. Univ. of Arizona Press, pp. 354 403. 238 OIKOS 103:1 (2003)

Merriman, J. 1993. Difficult, rocky, sandy, stoney, flowery: aboriginal ecology in the Blue Mountains. In: Stockton, E. (ed.), Blue Mountains dreaming: the Aboriginal Heritage. Three Sisters Productions, pp. 82 113. Miller, C. H. F., Magee, J. W., Johnson, B. J. et al. 1999. Pleistocene extinction of Genorynis newtonii: human impact on Australian megafauna. Science 283: 205 208. Owen-Smith, R. N. 1989. Megafaunal extinctions: the conservation message from 11 000 years BP. Conserv. Biol. 3: 405 412. Roberts, R. G., Flannery, T. F., Ayliffe, L. K. et al. 2001. New ages for the last Australian megafauna: continentwide extinction about 46,000 years ago. Science 292: 1888 1892. Solé, R. V., Alonso, D. and McKane, A. 2002a. Self-organized instability in complex ecosystems. Philos. Trans. R. Soc. Lond. B 357: 667 681. Solé, R. V., Montoya, J. M. and Erwin, D. H. 2002b. Recovery after mass extinction: evolutionary assembly in largescale biosphere dynamics. Philos. Trans. R. Soc. B 357: 697 707. Truswell, E. M. 1993. Vegetation changes in the Australian Tertiary in response to climatic and phytogeographic forcing factors. Aust. Sys. Bot. 6: 533 557. Westerman, M. and Krajewski, C. 2000. Molecular relationships of the Australian bandicoot genera Isoodon and Perameles (Marsupialia: Peramelina). Aust. Mamm. 22: 1 8. OIKOS 103:1 (2003) 239