Remote Australia Online

The purpose of this synthesis is to present an overview of the main findings of the ‘People, communities and economies of the Lake Eyre Basin’ project. This synthesis is designed as a stand-alone document and as the keynote chapter in the edited compilation of project reports. This three-year study has developed from the stance that successfully caring for the country and the communities of the Lake Eyre Basin (LEB) depends on four principles. First, it is crucial to better understand the social and economic landscape of the LEB and its resources. Second, effective management of the LEB is underpinned by building effective engagement processes between residents, management agencies and policy makers Third, it is necessary to identify and explore the key human dimensions specific to remote regions that affect the management of the LEB now and into the future. Finally, the fourth is to develop a mechanism to underpin long-term social learning to support the effective governance of the LEB in the form of a monitoring process. In contrast to monitoring frameworks used for day-to-day compliance monitoring, we emphasise that the framework developed in this project focuses on supporting continual learning that needs to be built into ongoing management activities to facilitate long-term care of the LEB and its communities. For organisations that are faced with the unique management challenges of remote regions, and are already stretched in terms of available resources, additional funding may be required. The synthesis begins with an introduction and background to the project presented in section 1, followed by a description of the project methods in section 2. This is followed by an overview of the four main components of the research project: a regional profile of the LEB, a toolkit of success factors of NRM organisations, case studies of successful natural resources management (NRM) engagement processes and a framework for monitoring successful engagement processes in the LEB.

The main aim of this project was to raise awareness of the soil degrading effects of accelerated wind erosion and to provide tools with the potential to increase land productivity sustainably and reduce some of the environmental uncertainties of pastoral production in the Australian desert. The project outputs include: - A representative and well-connected group of research user stakeholders - A fully functioning DustWatch wind erosion monitoring network - Maps of wind erosion activity produced from: meteorological records of dust storm occurrence, a numerical wind erosion assessment model and an intelligent knowledge-based system model - Prototype methodologies for: quantitatively discriminating accelerated wind erosion rates, and soil condition indicators for use in property-based Environment Management Systems (EMS).

The dust storm of 23 October 2002 covered most of eastern Australia and carried one of the largest recorded dust loads in Australia. In the 6 months leading up to the event, severe drought conditions in eastern Australia, plus above average maximum temperatures resulted in high potential evapo-transpiration rates, producing severe soil moisture deficits and reduced vegetation cover. Although increased wind speeds associated with a fast moving cold front were the meteorological driving force, these winds speeds were lower than those for the previously documented large dust storms. The dust storm was 2400 km long, up to 400 km across and 1.5–2.5 km in height. The plume area was estimated at 840,860 km2 and the dust load at 0900 h was 3.35–4.85 million tones (Mt). These dust load estimates are highly sensitive to assumptions, regarding visibility–dust concentration relationships, vertical dust concentration profiles and dust ceilings. The event is examined using meteorological records, remote sensing and air quality monitoring.The dust storm of 23 October 2002 covered most of eastern Australia and carried one of the largest recorded dust loads in Australia. In the 6 months leading up to the event, severe drought conditions in eastern Australia, plus above average maximum temperatures resulted in high potential evapo-transpiration rates, producing severe soil moisture deficits and reduced vegetation cover. Although increased wind speeds associated with a fast moving cold front were the meteorological driving force, these winds speeds were lower than those for the previously documented large dust storms. The dust storm was 2400 km long, up to 400 km across and 1.5–2.5 km in height. The plume area was estimated at 840,860 km2 and the dust load at 0900 h was 3.35–4.85 million tones (Mt). These dust load estimates are highly sensitive to assumptions, regarding visibility–dust concentration relationships, vertical dust concentration profiles and dust ceilings. The event is examined using meteorological records, remote sensing and air quality monitoring.

The objectives of this study were to predict the potential distribution, relative abundance and probability of habitat use by feral camels in southern Northern Territory. Aerial survey data were used to model habitat association. The characteristics of ‘used’ (where camels were observed) v. ‘unused’ (pseudo-absence) sites were compared. Habitat association and abundance were modelled using generalised additive model (GAM) methods. The models predicted habitat suitability and the relative abundance of camels in southern Northern Territory. The habitat suitability maps derived in the present study indicate that camels have suitable habitat in most areas of southern Northern Territory. The index of abundance model identified areas of relatively high camel abundance. Identifying preferred habitats and areas of high abundance can help focus control efforts.

The objectives of this study are to: Determine the impact of harvesting on camel population dynamics. This includes a sensitivity analysis to determine the relative importance of age-specific mortality and fecundity to the rate of increase. Further, the study assessed the harvest required to reduce population increase rate to zero and to reduce populations to particular target densities. Determine environmental predictors of camel distribution in the Northern Territory. Identify areas in the Northern Territory where commercial harvesting is feasible and the resultant potential reduction in densities is possible. To achieve these objectives we used models to examine strategies for managing wild camels in the Northern Territory.

Attracting and retaining skilled and professional staff is a problem not limited to remote, or even rural and regional locations in Australia. There is strong evidence to suggest that it is increasingly a global problem and organisations throughout the world are seeking innovative strategies to attract and develop new talent and develop other strategies to retain that talent. The attraction and retention of skilled and professional staff to a community is critical if it is to remain vibrant and self-sufficient, and hence able to sustain itself into the future. Attracting skilled and professional staff to Australian remote conditions has not been made easier by the Australian government policies which have been oriented to stringent neo-liberal economic policy for the last three decades. This has meant that services and infrastructure have been rationalised based on efficiency rather than equity. The under-provision of infrastructure is a hallmark of non-metropolitan Australia and its lack both pushes people out of communities and turns people off going there for any length of time. It was apparent in this project that attraction and retention strategies continually fall to community and industry groups.

One-humped dromedary camels (Camelus dromedarius) were first introduced to Australia in 1840 from the Canary Islands. Thereafter the majority came from the Indian subcontinent (McKnight 1969) and were used to open up the arid interior for European settlement (McKnight 1969). Camels were well suited to working in remote dry areas and were used as a means of transport, for freight and as draught animals (McKnight 1969). It is thought that over a 27-year period between 1880 and 1907 approximately 20 000 camels were imported. It was at least 40 years after the first importation that quarantine regulations were developed and enforced; it is probably simply fortuitous that the common camel diseases found in the Middle East and south Asia are not present in Australian camels today. The replacement of the camel by the motor vehicle as a mode of transport in the early 20th century resulted in large numbers of camels being released into the wild and the subsequent establishment of a feral population. Records show that the number of captive camels registered in 1941 was 2300, compared with 12 649 in 1920 (McKnight 1969). There is no reliable estimate of the number of camels that were released into the wild, although it is thought that it may have been between 5000 and 10 000 camels (Edwards et al. 2004).

Organisational complexity continues to hide the bordering processes that determine the boundaries of Martu people’s representation in Aboriginal affairs in the Western Desert of Australia. This paper describes participatory action research that sought to understand why Martu people are not represented in the institutions underpinning service provision. Key findings are that there is a complex inter-cultural space where mining companies and government create a multitude of Martu representative boundaries, most of which are not recognised by Martu people. A community-of-practice approach towards improved Martu representation is recommended, based on three principles: firstly, enabling Martu governance and representation; secondly, expanding dialogue and deliberation; and thirdly, recognising the role of boundary spanners. The paper outlines the relevance of these principles to the practical reality of enhancing representation of Martu people in service delivery.

Across Australia, Aboriginal and Torres Strait Islander people have experienced severe and systematic disempowerment with devastating health and social impacts. Despite Australia’s world-class health system, the life expectancy of Aboriginal and Torres Strait Islander men is 11 years lower than that for Australian males (67 years compared to 78) and for Aboriginal and Torres Strait Islander women it is 10 years less (73 compared 83 years) (SCRGSP 2009). Concerted action can make a difference (as demonstrated by the reduction in the life expectancy gap between Indigenous and non-Indigenous people in North America and New Zealand to around seven years) (SCRGSP 2009). But in Australia, in some important respects, the circumstances of Aboriginal people appear to have either deteriorated or regressed or health inequity has increased (Gary Banks, quoted in Steering Committee for the Review of Government Service Provision 2005). In the Northern Territory, for example, all-cause mortality rates for Aboriginal people declined overall and for all age groups over a long period (1967– 2000). Declines in Aboriginal mortality however ‘did not keep pace with the relative decline for the total Australian population’ (Wilson et al. 2007) There is now an evidence base which indicates that interventions that empower socially excluded populations across psychological, organisational and community levels have achieved improved health outcomes and quality of life of disadvantaged groups (Wallerstein 2006). Empowerment is a social action process that promotes participation of people, organisations and communities towards the goals of increased individual and community control, political efficacy, improved quality of community life, and social justice (Wallerstein 1992). Increased control and mastery means that people are better able to deal with the forces that affect their lives, (Syme 2003) and have greater capacity to deal with dayto- day challenges of life without being overwhelmed by them (Syme 1998). Aboriginal people have described empowerment as ‘Healing – coming to terms with the past and present situation, dealing with the pain; gaining control; becoming strong; finding your voice; participating in change; and working together for a strong community’

Key resources in arid and semi-arid systems are highly variable in time and space. Biophysical variability is mainly driven directly or indirectly by rainfall. Arid-dwelling people are also exposed to variability – indirectly from their biophysical environments, and directly as a result of sparse, mobile populations and variable, distant signals from policy and markets. The tradeoffs between the benefits and costs of increasing the scale of resource use are comparatively more critical in shaping system dynamics in arid systems than mesic environments. Understanding the strategies employed to scale resource use is vital for arid-system management. We contribute to this discussion using a conceptual model, based on modern portfolio theory, to explore how variability is managed by diversifying access to resources across space and time. The model is developed using ecological ideas and our Australian case knowledge, but demonstrates general learning on how optimal investment in strategies changes with resource distribution, and with the physical characteristics of the entity (e.g. plant, animal) and resource (e.g. water, nutrients) in question. Using pastoral enterprise, arid-system administration and policy cases, we argue that people and organisations need variability-coping strategies that are comparable with biophysical responses, and that strategy behaviour needs greater acknowledgement in managing arid systems.

Life in desert regions of Australia (and indeed worldwide) is affected by a combination of factors: variability in climate, markets and policy that is outside the control of local people; sparse and often mobile populations; natural, human and cultural resources that are patchy in space and time; and relatively poor scientific understanding despite significant local and traditional knowledge, among other factors. These `desert drivers' affect not only ecology and natural resource management, but also service delivery, enterprise development, governance systems and many other aspects of desert living. This paper outlines moves towards consolidating these issues into an integrated, inter -sectoral new discipline - the Science of Desert Living - and the benefits this could deliver.

While research-article impact is routinely judged by citation counts, there is recognition that a much boarder view is needed to better judge their true values. This paper applies a developing framework based on the application of network theory, where the network consists of journal articles on arid-systems research which are listed on ISI Web-of-Science. Keywords were used to identify articles related to arid-systems research. Linkages between articles were defined by citations, and we bound our analysis by focusing on how the Australian subsample contributes to the international arid-systems literature. The analysis showed that impact based how on articles contribute structurally to the flow of knowledge within the literature offers an alternative metric to citation counts. The analysis also presented a partitioned view of the Australian-arid literature. This showed that there exists some citation-based structure within the literature, and we showed this structure better describes the literature than a partition based on which journals articles are published in. What this paper most contributes, is a discussion about the true value of our research, and how we may strategise to seek impact in more meaningful ways than merely being cited.

Social actors in arid regions must develop strategies to respond to available resources, which are scarce, variable, patchy and unpredictable relative to other regions. We explore our observations of relationships amongst people and organisations in Australian deserts using a stylised network model of the structure of social networks in arid systems. Results suggest that temporal resource variability drives increased network density, but with fewer strong ties; sparse populations drive a relatively higher proportion of strong ties, and that networks develop a hub configuration as resource endowments become more patchy spatially. These ideas highlight some issues that warrant improved understanding by actors seeking to enhance livelihoods and local resilience in these extreme environments.

Arid systems are markedly different from non-arid systems. This distinctiveness extends to arid-social networks, by which we mean social networks which are influenced by the suite of factors driving arid and semi-arid regions. Neither the process of how aridity interacts with social structure, nor what happens as a result of this interaction, is adequately understood. This paper postulates three relative characteristics which make arid-social networks distinct: that they are tightly bound, are hierarchical in structure and, hence, prone to power abuses, and contain a relatively higher proportion of weak links, making them reactive to crisis. These ideas were modified from workshop discussions during 2006. Although they are neither tested nor presented as strong beliefs, they are based on the anecdotal observations of arid-system scientists with many years of experience. This paper does not test the ideas, but rather examines them in the context of five arid-social network case studies with the aim of hypotheses building. Our cases are networks related to pastoralism, Aboriginal outstations, the ‘Far West Coast Aboriginal Enterprise Network’ and natural resources in both the Lake-Eyre basin and the Murray–Darling catchment. Our cases highlight that (1) social networks do not have clear boundaries, and that how participants perceive their network boundaries may differ from what network data imply, (2) although network structures are important determinants of system behaviour, the role of participants as individuals is still pivotal, (3) and while in certain arid cases weak links are engaged in crisis, the exact structure of all weak links in terms of how they place participants in relation to other communities is what matters.

Livestock mobility is practised by pastoralists to cope with some of the variability and unpredictability of limited forage resources and because a diverse portfolio of strategies is needed to manage risk. The global trend towards rangeland privatisation, fragmentation and land-use intensification is eroding many of the institutions that have traditionally facilitated pastoral mobility. While Australia’s pastoral industry was developed as a European private-property system, livestock mobility has recently been increasing, indicating an important response to variability regardless of a nation’s wealth or development. This paper discusses how opportunistic movements of livestock over large scales by trading grazing rights between enterprises are effective but imperfect. Knowledge about the trustworthiness of individuals and local environments is often limited and poorly monitored. There is scope for policy to support mobility by targeting these institutional failures. The Australian system of trading grazing rights can inform efforts to maintain spatial flexibility in the industrial era.

In this study, we analysed the development of institutions under a recent draft strategy for managing water resources in Alice Springs. We used the Institutional Analysis and Development (IAD) framework, which has been developed and tested internationally for managing common-pool resources Water is a critical resource for the economic, social and ecological sustainability of desert towns. Effective and efficient water management is required, yet strategies are diverse even among the towns of Alice Springs, Coober Pedy, Roxby Downs, and the Goldfields of Western Australia, which share a similar type of groundwater resource The objectives of our study are to: 1. describe the recently developed Alice Springs Water Resource Strategy (ASWRS), and analyse its potential impact on desert water resource management institutions and outcomes 2. examine the validity and utility of the IAD framework as a guide for policy analysts and decision makers in describing and evaluating water resource strategies of desert towns 3. suggest simple methods for using the IAD framework to adapt and design institutions.

Outback settlements are markedly different from their coastal counterparts in terms of their sociocultural, economic, demographic and environmental characteristics. Differences are also strong between outback settlements, although they share some basic characteristics. The population size distribution of outback settlements is markedly clumped. There are a small number of major service centres (population 20,000 to 30,000) as well as a much larger number of settlements that are two orders of magnitude smaller (i.e. population <200–500) and which have very different socioeconomic characteristics. Standard regionalisation procedures mask this pattern and its significance for analysis. Standard procedures also take state/territory borders as hard boundaries, disguising the significance of cross border similarities and relationships. These differences have been a major challenge to regional policy making, research and investment decisions. In this study, we have classified and mapped settlements into socio-regions. The study has considered a UCL (Urban Centre or Locality – a settlement with 200 or more people) as a basic unit for analysis. There are 222 UCLs in the rangelands excluding Darwin. A preliminary hinterland boundary or socio-region was created for each UCL as spatial mapping units. This process was primarily based on similarities and differences of key demographic and socio-economic indicators. Indicators of ecological characteristics were also considered in mapping the socio-regions.

A comprehensive understanding of vegetation and fire dynamics and their interactions is important for effective land management in central arid Australia, the majority of which is vegetated with natural and semi-natural (grazed) vegetation. The potential loss of fire-sensitive populations of tree and shrub species and the impact of invasive grasses on fire regimes are of concern. This paper synthesises current knowledge, from published literature and expert opinion, about vegetation–fire interactions in central arid Australia. An initial conceptual framework for vegetation–fire interactions is presented for exploring the relationship between average fire frequency and potential fuel load in different vegetation types. This framework is intended to encourage debate and stimulate further research regarding these issues. The paper also considers the influence of both fire management and other drivers of changed fire regimes, such as climate change and the spread of introduced pastoral grasses, in particular buffel grass (Cenchrus ciliaris).

Desert landscapes of central Australia have inspired various narratives for nation building. These narratives, based upon discourses of land as a commodity for the colonial project, include the inferior peripheral wastelands of the colonial centre; the wild and pristine haven of ‘noble aborigines’; and the frontier home of heroic explorers and pastoralists. These narratives continue to influence land management policy and practice. However they do not reflect the diverse contemporary realities of individuals living in the region. In this paper I juxtapose these narratives and realities to reveal the multitude of ways in which desert landscapes are known and understood. A case study of a fire management project that involved Aboriginal, pastoral and conservation land managers from the southern Tanami region of central Australia provides evidence as to why desert landscapes should be re-conceptualized as cultural, contested and dynamic. To conceive of these spaces and places in this way can facilitate open discussion and dialogue between land managers from the region. It is the first step in working towards the process of cultural hybridity that I argue is necessary for equitable and sustainable environmental governance and management in Australia.

There are ongoing debates in the contemporary environment and development literature regarding the role of scientific, local and indigenous participation in sustainable development initiatives. The debates have been critical of the supremacy of western scientific knowledge in such initiatives, with some academics asserting that science can be imperialistic, and its application can sometimes lead to social inequity and exclusion. In response, local and Indigenous knowledges have been offered as providing a panacea for all environment and development problems. This thesis argues that in Australia the meta-narrative of ecologically sustainable development is in fact unsustainable because it perpetuates the intra- and inter-generational inequalities that it is supposedly meant to overcome. This is because the meta-narrative of ecologically sustainable development separates ways of knowing the world into dichotomies of self/other and universal scientific knowledge versus place-based local knowledge. The thesis argues that equitable and sustainable ecologically sustainable development is dependant upon moving beyond these dichotomies. The research questions what lies between the complex sets of knowledge of best practice environmental management at the local environmental management and community development interface in Australia. An investigation is conducted into the knowledge synergy that is, or indeed is not, occurring between government organisations, non-government organisations, local community groups and individuals involved with two environmental management and community development projects in Australia. One project works across interest groups to protect and enhance threatened species habitat in Victoria. The other project considers what it means to manage fire across different land tenures in the Northern Territory. These case studies act as points of access into the localised knowledge networks surrounding environmental governance and management in Australia. They give life to the thesis critique and relevance to the practical outcomes. Part 1 grounds the thesis within the discipline and practice of critical human geography. Part 2 locates the thesis within the contemporary environment and development literatures. It demonstrates an applied peoples’ geography to consider the power and the potential of local spaces of environmental management and community development. Part 3 practices this applied peoples’ geography. It illustrates how the political, cultural, knowledge and social landscapes of any environmental management and community development project reflect the diverse knowledges of the environment in Australia. Understanding these complex landscapes provides the means for moving beyond ‘them and us’ dichotomies inherent to the meta-narrative of ecologically sustainable development in Australia. Part 4 demonstrates that knowledge of best practice environmental management move across and between multiple, networked and entangled local spaces of environmental governance. Individuals choose to engage with these local spaces of management for a variety of social livelihood reasons. As such, projects have a responsibility to deliver both tangible management outcomes and intangible social processes for ecologically sustainable development. Part 5 illustrates that the success of any environmental management and community development project is contingent upon acknowledging the value, the power and the limitations of local community knowledge. Equally, project success depends upon recognising the many varieties of local knowledge at the environmental management and community development interface. Equitable and successful ecologically sustainable development in Australia depends upon improved knowledge sharing and knowledge synergies between these many knowledge cultures. Part 6 demonstrates that improving knowledge sharing at the local environmental management and community development interface is contingent upon four applied principles. Engaging in activities that complement these principles will result in the social action necessary to celebrate what lies between the different ways of knowing and managing the environment in Australia. This research contributes to debates in the environment and development literature. It demonstrates what it means to move beyond the above mentioned dichotomies. It informs contemporary environmental governance and management policy and practice in Australia. It highlights the importance of diversity in overcoming issues of social and environment degradation. It sheds light on the local spaces of power and knowledge of two environmental management and community development projects. It illustrates the potential of and the necessary processes for increased knowledge sharing and synergy at the local environmental management and community development interface.

This paper presents an overview of the Australian Indigenous housing research identifying strengths and gaps in the body of literature published between 1970 and 2006. The findings of the review of the Community Housing Infrastructure Programme (CHIP), entitled Living in the Sunburnt Country, are cast against this overview in order to illustrate housing research strengths and weaknesses. The CHIP review introduced major reforms to Indigenous housing and makes a distinction between Indigenous housing in remote and non-remote locations. The CHIP review proposes to counter the Indigenous homelands movement by encouraging Indigenous migration from smaller settlements to those with greater access to services, a move from ‘country’ to the ‘big smoke’. This paper illustrates the significance of existing housing research to the development of Indigenous housing policy and programs and it identifies ways that housing research can be reshaped to inform current Indigenous housing reforms.

This paper is concerned with the relationship between Indigenous mobility and the sustainability of remote Indigenous communities in Australia. Indigenous people in remote and rural Australia are frequently moving between places. Such movement is a key to the maintenance of Indigenous relationships to places and to kin. This cultural maintenance continues despite a period of 100 years or more when the government employed strategies to disrupt traditional Aboriginal social and geographic patterns. Recently there have been renewed attempts by government to question the sustainability of remote Indigenous communities based on the economics of remote service provision. Given the widespread reporting of high mobility in the ethnographic and Indigenous housing literature, and the recent political call to reconsider the role of small-scale remote Aboriginal settlements, this paper addresses the need to understand both the role of mobility in sustaining and expressing the attachment of Indigenous people to their places and kin and the relationship between mobility and service needs. The paper examines case studies of the mobility of two remote Indigenous communities, Alpurrurulam in the eastern Northern Territory, and Dajarra in northwest Queensland. The paper outlines the implications of Aboriginal mobility for service provision in this study region. The paper reveals a policy challenge to balance the forces for continuity of remote and rural lifestyles, and those of changes in service needs and supplies.

Environmental reporting and understanding of aeolian processes at different spatial and temporal scales require both extensive temporal and spatial data sources. The Bureau of Meteorology (BoM) SYNOP data have been used for environmental reporting in Australia by McTainsh and colleagues since 1998; however, with the current trend to replace meteorological observers with automated surface observation systems such as weather stations, these data sources are not sustainable. DustWatch was established to fill some of the gaps in the BoM network by strategically adding a community network of observers and instruments. Satellite imagery is also utilized when it is available. This paper outlines the nature of the DustWatch network, describes some of the educational and awareness outputs, outlines the types of data gathered and gives examples of how the data are being used both for environmental reporting and research into aeolian processes and activity.

We analysed the movement of seven female camels collared in the Anangu Pitjantjatjara Yankunytjatjara Lands of South Australia. Understanding the movement patterns of feral camels and subsequently where high densities are likely to threaten biodiversity and cultural assets, provides land managers and government agencies with information and decision support tools to manage camel impacts. Accordingly, we tested if there were any seasonal changes in camel movement, any measurable separation between home range and migration, and any relationship between broader camel landscape use, rainfall and mountainous terrain. We fitted ARGOS GPS satellite collars to seven female camels in South Australia during August 2007 and found evidence to suggest that over a 12-month study period, some camels had returned to locations that they had previously visited. This cyclic movement pattern was more regular up to ~50 days, however, one collared individual returned to a previous location after 300 days. Despite only having a small sample size, we did find camels moving into areas that received higher rainfalls in the warmer months and possibly some attraction of these camels to steep mountainous terrain over this period.

The fire mosaic hypothesis is intuitively appealing to scientists and land managers due to its perceived potential to deliver favourable outcomes for biodiversity conservation and fire management. However evidence in support of the biodiversity benefits of fine-scaled fire mosaics is scant. In this thesis I investigate the key assumptions of the fire mosaic hypothesis using a model system, the mulga woodland/mulga bird community of central Australia. Mulga woodland is an ideal model system for this question because it is structurally and floristically simple, yet supports a rich avifauna. I tested how avian diversity (variety and number) was influenced by 1) time-since-fire; 2) patch size; and 3) the boundary between burnt and unburnt mulga woodland (pyric edge). An investigation of time-since-fire is crucial for testing the fire mosaic hypothesis. If there is no effect of time-since-fire on biodiversity, then the spatial arrangement of different times-since-fire is irrelevant and the definition of habitat patches and habitat edges based on time-since-fire is not valid. Patch size and edge effect are potential mechanisms by which a fine-scale fire mosaic may support greater avian diversity than a coarse-scale fire mosaic. For this to be the case, avian diversity must increase with decreasing patch size; or be greater at pyric edges than in the interior of habitats.