Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Predicting the integrity of ecosystems in the face of species loss

Scarff, Fiona (2002) Predicting the integrity of ecosystems in the face of species loss. PhD thesis, Murdoch University.

PDF - Whole Thesis
Available Upon Request


There are too many kinds of organisms to be able to study and manage each; yet loss of a single species could lead to the unravelling of an ecosystem. Chapter 1 (Introduction) opens with this problem and frames the central question of this thesis - can focus on a minority of target organisms safeguard ecosystem integrity? The question is refined, and I discuss possible answers in view of recent directions in theoretical ecology. Finally, an empirical model is proposed for distinguishing between various competing hypotheses.

In Chapter 2 (The Oxidation Pond), the question is tested according to this model in a microcosm of aquatic microbes and small invertebrates. It is discovered that no single species extinction can disrupt the integrity of the pond. There arc thus no critical species that invite targeted management attention. In terms of its ‘naturalness’, the oxidation pond is as legitimate a system to study as any, irrespective of its size and containment. However, it is only one system, and a review of the literature reveals that the proportion of critical species - hence the suitability of the system to a target species approach - varies between ecosystems.

Identifying what proportion of the species in a system are indispensable would appear to be essential for deciding on an effective management approach for a species-rich ecosystem. However, the manipulative techniques used to acquire this information, both in this chapter and in the literature, are not suited to most large scale systems of management interest.

Chapter 3 is entitled ‘Ecological redundancy as a framework for critical species’. and in it I consider the species property of being indispensable to an ecosystem, as a reflection on its ecological uniqueness. To be indispensable is to be so unique that no other species in the pool of recruits can compensate for the loss of the critical entity. This perspective provides the foundation for the ‘fusewire species’, which I develop as a new concept in ecology and conservation. As a metaphor the fusewire concept overcomes some previous difficulties associated with the use of ‘keystone species’ as priority organisms for conservation and management. At the end of the chapter I draw on recent literature to develop a model for measuring ecological redundancy in large scale systems, with a view to identifying critical ‘fusewire’ species.

The next two chapters relate how this method was used in the analysis of two natural systems. A computer model was written to scan ecological data sets for instances of uniqueness, and this approach is then applied to The impact of birds on a eucalypt forest (Chapter 4) and Redundancy in South African grasslands (Chapter 5).

In the first case, data on the foraging ecology of an assemblage of forest birds were analysed, taking the view that it is mainly through their foraging activity that most of these birds impact their ecosystem. The results suggest that most unique functions arc performed by a relatively small proportion of the bird assemblage, so that a target species approach may be suitable. The Emu, a large frugivorous ratite, was perhaps most conspicuous in its unique contributions to ecosystem function.

The second analysis focussed on assemblages of graminoids in South Africa’s grassland biome. Data on a range of phonological, physical, reproductive and physiological attributes were examined. This chapter is concerned less with identifying appropriate priority species in these particular communities, than with exploring some of the properties of the redundancy matrix method. The results indicate that the method can be sensitive to both the number and identity of functional attributes chosen for analysis. The attributes need to be deliberately chosen and justified with reference to specific management goals.

In Chapter 6 (Synthesis) I briefly review how measuring redundancy can assist in predicting how ecosystems will respond to local species loss. Some practical and theoretical problems with trying to measure redundancy are then discussed. I also consider how the narrow treatment provided in this thesis could be broadened; both conceptually (to include evolutionary as well as ecological process), and taxonomically (to measure the ecological contribution of bacteria as well as eukaryotes). Links are drawn with some recent theoretical developments: the study of scale-free networks, and the unified neutral theory of biodiversity and biogeography. Some directions are suggested for further research into ecological redundancy. The thesis concludes with a note of caution on the interpretation of ecological redundancy as a working concept in conservation. Like the organisms we study, conservation biologists also face inescapable dilemmas of resource partitioning. The problems associated with trying to manage complex systems by measuring and understanding all their components, are reiterated - in the context of recent research that emphasises that local extinctions may precipitate irreversible ecosystem change. In view of such findings, I argue that whereas conservation dilemmas involving choosing one species over another arc inevitable, the measurement of patterns of redundancy offers the possibility of making more informed decisions in managing complex biological systems.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): Division of Science and Engineering
Notes: Note to the author: If you would like to make your thesis openly available on Murdoch University Library's Research Repository, please contact: Thank you.
Supervisor(s): Bradley, Stuart
Item Control Page Item Control Page