KATANNING NATIONAL MALLEEFOWL FORUM 2007

PROCEEDINGS

S.J.J.F. DAVIES (Editor)

1 2

CONTENTS Page Preface v

Acknowledgements v

National Malleefowl Forum 2007 – Programme vi

Resolutions of the Forum x

1. Sandell, P. Aims of the National Malleefowl Forum 1

2. Benshemesh, J. Progress towards a national system of monitoring 4 and adaptive management for Malleefowl

3. Stokie, A. Towards a national system: the power of the people 17

4. Ewin, P. Malleefowl in New South Wales: review of past 21 and future actions

5. Stokie, P. Victorian state round-up – achievements since 32 Mildura 2004

6. Gillam, S. Malleefowl conservation in South Australia: 40 achievements from 2004 to 2007

7. Danzi, C. Achievements in Western Australian Malleefowl 47 conservation since the forum at Mildura in 2004

8. Ward, M. & Bellchambers, K. Status, monitoring and management 57 of Nganamara in the southern Alinytjara Wilurara Natural Resource Management region

9. Partridge, T. Nganamara in South Australia‘s Anangu Pitjantjatjara 70 Yankunytjatjara Lands

10. Gillespie, J. Captive breeding and management of Malleefowl 74 (Leipoa ocellata) at Western Plains Zoo, 1989 to present

11. Mulder, R., Dunn, P. & Donnellan, S. Conservation genetics of 81 Malleefowl

12. Sinclair, J.R. Are we doing enough? A review of the conservation 84 status of megapodes and the implementation of their action plans

13. Mauro, I. Bruijn‘s Brush-turkey Aepypodius bruijni: field discovery, 107 monitoring and conservation of an enigma

14. Jones, D. N. Megapodes in northern Australia: a summary of 120 recent research on the Orange-footed Scrubfowl and Australian Brush-turkey

15. Parsons, B. Malleefowl in the Western Australian Wheatbelt; 126

iii an ecological study informed by community knowledge

16. van der Waag, J. How to support the recruitment of Malleefowl 130 Leipoa ocellata in small remnants: an overview of project findings

17. Coombes, C.J., Dehaan, R. & Wilson, A.L. Post-release 137 survival of captive-reared Malleefowl in western new South Wales

18. Benshemesh, J. Dennings, S., Danzi, C. & Saunders, A. Malleefowl 142 searches at Yeelirrie Station in arid Western Australia: a valuable collaboration between industry and volunteers

19. Stokie, P. Re-searching malleefowl monitoring sites – 153 a Victorian Malleefowl Recovery group perspective

20. Priddel, D., Wheeler, R. & Copley, P. Does the integrity or 158 structure of mallee habitat influence the degree of fox predation on Malleefowl?

21. Scheltema, M. & Jones, P. Malleefowl at Mt. Gibson minesite 162

22. Reichelt, R. W. Little Desert Nature Lodge presentation 166

iv

PREFACE

The National Malleefowl Forum 2007 at Katanning in Western Australia was a project of the National Malleefowl Recovery Team and an integral part of the Natural Heritage Trust‘s Multi- regional Malleefowl Monitoring, Population Assessment & Conservation Project. It was the second national forum, a review of the progress of Malleefowl conservation across Australia. In addition three papers reviewed conservation issues in other species of megapode, from within Australia and in the adjoining Indo-Pacific region. It was always the intention of the Recovery Team that a permanent record should be prepared of the proceedings of the forum and by publishing this volume in the Curtin University of Technology, Environmental Biology Bulletin series it will be available both as a printed book and on line. We are grateful to the Department of Environmental Biology for enabling this to occur.

The forum attracted 22 papers from a wide range of sources, universities, state agencies, volunteer groups, consultants, zoos, mining companies and private naturalists, providing a comprehensive review of issues and achievements in the conservation of Malleefowl and other megapodes. These contributions should provide a useful baseline against which future progress can be assessed. All speakers were asked to provide a manuscript at the forum, either as a hard copy or in an electronic medium. In the event only three of the 22 did so; manuscripts have dribbled in slowly but for those who did not provide manuscripts by the date of printing, a summary has been included based on a transcript of their spoken presentation. A number of poster displays were also presented at the forum. Abstracts of these were published in the programme but have not been included in these proceedings.

The National Malleefowl Recovery Team is grateful to the Steering Committee that organised the forum. The members of the Steering Committee were: Dr. Stephen Davies (Chairman), Sally and Wally Cail, Carl Danzi, Lynda Green, Natalie Holland. Liz McClelland, Kat Miller, Judy O‘Neill, Blair Parsons, Alice Rawlinson, Ann and Peter Stokie and Jessica van de Waag.

ACKNOWLEDGEMENTS

The National Malleefowl Recovery Team, and the Forum Steering Committee in particular, are most grateful to the numerous and generous sponsors who supported the event, the Australian Government (Natural Heritage Trust), the Avon Catchment Council, the Commonwealth Scientific and Industrial Research Organisation, the Department of Environment and Climate Change (NSW), the Government of Western Australia (Department of Environment and Conservation), the Great Southern Area Consultative Committee, Iluka Resources, Lotterywest, the Mallee Catchment Management Authority (Victoria), Mount Gibson Iron, Parks Victoria, Perth Convention Bureau, the World Wide Fund for Nature (WWF) and the Yongergnow Australian Malleefowl Centre.

The Steering Committee would also like to express its thanks to those who so ably chaired the sessions and kept the forum on time, and to those who looked after the registration desk during the event. The forum was held at the Katanning Leisure Centre where the staff members were most helpful. Catering was provided by Sally Boxall Catering and, on Tuesday, by the Yongergnow Australian Malleefowl Centre, both of whose contributions were greatly appreciated.

v

NATIONAL MALLEEFOWL FORUM 2007 – PROGRAMME

The forum was held at the Katanning Leisure Centre from September 7-10, 2007, and at Yongergnow Australian Malleefowl Centre on September 11, 2007

Friday September 7

6-8 pm Registration in the foyer of the Katanning Leisure Centre. WWF Australia (WA), in celebration of Threatened Species Day, hosted a reception for delegates in the Foyer during registration.

Saturday September 8

8-9 am Registration

9 am Welcome by Peter Sandell, Chairman of the National Malleefowl Recovery Team

9.15 am Welcome on behalf of the local Noonyar Community by Shane Woods Welcome on behalf of the Shire of Katanning by President, Cr. Phil Rae.

9.30 am Peter Sandell – Aims of the National Malleefowl Forum

9.45-10.30 Dr. Joe Benshemesh – Progress towards a national system of monitoring and adaptive management for Malleefowl.

10.30-11 am Morning tea

11-11-30 am Review of the implications for monitoring of mounds: detailed discussion of NHT Project

Chair: Ann Stokie, NHT Project coordinator

State by state round-up: achievements since Mildura

Chair: Ron Wiseman, Chair, Victorian Malleefowl Recovery Group

11.30-12 Peter Ewin – Malleefowl in New South Wales: Review of past and future actions.

12-1.30 pm Lunch

1.30-3 pm State by state round-up: achievements since Mildura

Chair: Ron Wiseman, Chair, Victorian Malleefowl Recovery Group

Peter Stokie – Victoria

Sharon Gillam – Malleefowl Conservation in South Australia – Achievements from 2004 to 2007.

3-3.30 pm Afternoon tea

vi 3.30-5 pm Chair – Peter Mawson, Senior Zoologist, DEC, WA

Carl Danzi – Achievements in Western Australian Conservation since Mildura Forum, February 2004.

Thalie Partridge and Matt Ward, DEH, SA – Nganamara in South Australia‘s Aboriginal Lands.

Judith Gillespie – Captive Breeding and Management of Malleefowl (Leipoa ocellata) at the Western Plains Zoo, 1989 to present.

Raoul Mulder – University of Melbourne – Genetics of Malleefowl

7 pm Launch of the National Malleefowl Monitoring Manual followed by the Conference dinner

Sunday September 9

First Western Australian Megapode Symposium

Review of current Megapode work Chair: Len van der Waag, Board, Yongergnow Australian Malleefowl Centre

9-10 am Ross Sinclair. Leibniz – Megapode work in Indonesia, the Philippines, the Pacific and Indian Oceans

10-10.30 Iwein Mauro – Bruijn‘s Brush Turkey Aepypodius bruijnii: Field Discovery, Monitoring and Conservation of an Enigma.

10.30-11 Morning tea

Megapode work in Australia - Chair: Stephen Davies, Curtin and Murdoch Universities

11-12 noon Darryl Jones - Megapodes in Northern Australia: a Summary of Recent Research on the Orange-footed Scrubfowl and Australian Brush-turkey.

12-1 pm Lunch

Recent studies of Malleefowl – Chair: Professor Darryl Jones, Griffith University

1-1.30 pm Blair Parsons, University of Western Australia – Analysis of sighting data

1.30-2 pm Jessica van der Waag, - How to Support the Recruitment of Malleefowl Leipoa ocellata in small Remnants: An Overview of Project Findings.

2-2.30 pm Chris Coombes – Post-release Survival of captive-bred Malleefowl in Western New South Wales.

2.30-3 pm Joe Benshemesh, Susanne Dennings and Carl Danzi – Malleefowl searches at Yeelirrie Station in arid WA: a valuable collaboration between industry and volunteers.

3-3.30 Afternoon tea

Further studies of Malleefowl – Chair: Susanne Denning

vii

3.30-3.50 pm Peter Stokie – Researching Malleefowl Monitoring Sites – A VMRG perspective.

3.50-4.10 pm Robert Wheeler – Release of captive bred Malleefowl in South Australia.

4.10-4-30 pm Martine Scheltema – Malleefowl at Mt Gibson Minesite.

4.30-4.50 pm Raymond Wimpy Reiicheldt – Little Desert Nature Lodge Malleefowl Presentation.

5-6 pm Meeting of the National Malleefowl Recovery Team

Monday September 10

Management of Malleefowl populations I – Chair: Natalie Holland, WWF

8.30-9 am Jessica van de Waag, University of Western Australia – Issues in Malleefowl conservation

9-10.30 Discussion on improvements to Malleefowl populations in large areas and arid lands.

10.30-11 am Morning tea

Management of Malleefowl populations II – Chair: Mr. Terry Waldron, JP, MLA, Member for Wagin

11-12.30 Discussions on improvements to management of Malleefowl on small areas and isolated remnants

12.30-1.30 Lunch – Poster Presentations

Management of Malleefowl populations III – Chair: Jason van Weenen, Dept. of Environment and Heritage, SA

1,30-3 pm Discussion on the contribution of captive breeding and reintroduction programmes.

3-3.30 pm Afternoon tea

Conclusion – Chair: Professor Stephen Davies, Curtin and Murdoch Universities

3.30-4.30 Review of discussions

4.30-6 pm Plenary session:

Tuesday September 11

10-10.30 am Morning tea at Yongergnow Australian Malleefowl Centre

10.30-11 Presentation on Yongergnow – Melissa Savage, Manager, Yongergnow

11-12.30 Disperse in small groups to inspect Yongergnow and Foster Road Malleefowls

12.30-1.30 Lunch

viii

1.30-3 pm Disperse in small groups to inspect Yongergnow and Foster Road Malleefowls

3 pm Bus leaves for Katanning

ix

RESOLUTIONS OF THE FORUM

1. The Katanning National Malleefowl Forum endorses the maintenance of the National Malleefowl Recovery Team.

2. The Katanning National Malleefowl Forum 2007 strongly endorses the Adaptive Management approach to Malleefowl conservation and seeks continuing support for it from Federal and State agencies.

3. The Katanning National Malleefowl Forum 2007 endorses monitoring and data management to national standards and seeks financial support from government agencies to continue the monitoring program.

4. To implement these resolutions the Katanning National Malleefowl Forum 2007 recommends supporting the goals of the National Malleefowl Recovery Plan through a tenured national coordinator.

5. The Katanning National Malleefowl Forum 2007 supports the development of a National Malleefowl website.

x

1. AIMS OF THE NATIONAL MALLEEFOWL FORUM

Peter Sandall, Chair, National Malleefowl Recovery Team Environmental Program Manager (Mallee District) Parks Victoria, Mildura, Victoria

On behalf of the National Malleefowl Recovery Team, I would like to welcome everyone to the second National Malleefowl Forum.

Malleefowl are a unique species from a number of perspectives:

Their unique mound building behaviour Their capacity to survive European settlement whereas a number of other ground- dwelling birds and mammals have disappeared from our mallee regions The groundswell of community support that this ‗icon species‘ attracts. It has a public following like perhaps no other species from our southern inland

Why is this? Is it because:

Their nests provide conspicuous evidence of their ongoing presence? Their value as indicators of the broader health of our mallee environments? Their non-aggressive behaviour? Their protestant work ethic? Their poor parenting skills?

Whatever it is, the formation of community groups to promote the conservation of this species from Victoria to Western Australia (Victorian Malleefowl Recovery Group, Malleefowl Preservation Group etc.) is clear evidence of the appeal of Malleefowl. But, once formed, what can these groups do of a concrete nature to help secure the future for this species? They can:

Promote the cause of Malleefowl in the wider community such that landholders defer on actions such as grazing or clearing of remnant mallee, or government agencies play a greater role in managing the ongoing risks to Malleefowl within parks and reserves, e.g. from fox predation and inappropriate fire regimes. We don‘t need to look much further than nearby Ongerup to see a very good example of regional promotion of Malleefowl conservation

Collect long term data on the status of the species within local / regional / multi- regional areas to understand the nature and level of risks to the species. Monitoring and research are critically important to the conservation of the species at a national level. As Julie Kirkwood said at our last forum; ‗lack of data is an important impediment to species recovery‘

Get involved in regional planning undertaken by catchment management authorities. These plans tend to be important in influencing government spending on conservation activities

It is a very big ask for volunteer groups to perform all of these tasks across the full range of the species. Hence we need to give consideration to where our skills lie, what is most achievable at a local level, and what might make a significant contribution at a national level.

Monitoring is a function which has not generally been well-performed by government agencies. This is often because the impetus to monitor relies upon key personnel who may have moved on, or the cost cannot be justified in the short term. Volunteer groups have a number of advantages over government agencies in performing this function.

1 There is often longevity of involvement by key personnel in the group providing continuity of corporate knowledge and support for newcomers

Volunteers obviously can conduct monitoring for a much lower cost than can a paid government workforce. In Victoria, it is estimated that the use of volunteers to monitor has resulted in a 90% reduction in the cost

The quality of data collected tends to be high because the volunteers have a keen interest in the subject matter

Where volunteers need backup tends to be in the area of analysis of information collected and interpretation of those data at a regional and national level. This is where the National Malleefowl Recovery Team can play a role, and is, of course, what is being achieved through the multi-regional project. More about that later from Joe and Ann.

The National Recovery Team has approx. 4 representatives from each of the 4 states over which the Malleefowl ranges. Community groups and state government agencies are represented. Malleefowl are a nationally listed species under the Environmental Protection and Biodiversity Conservation Act, but the recovery team receives no funding except that state agencies allow time for their staff to be involved. The Commonwealth has supported a number of Malleefowl initiatives such as the Natural Heritage Trust funded project the ‗Malleefowl Monitoring, Population Assessment & Conservation Project‘. The Department of Environment and Water r Resources has supported Malleefowl work through the WWF in Western Australia.

The Commonwealth has also funded (through the South Australian Department of Environment & Heritage) a recent review of the national recovery plan. A key objective of the National Recovery Team is to promote and support the implementation of the recovery plan. The recovery plan is one that recognises the role of community.

So who actually implements the plan? This is really a collaboration between communities and the land management agencies in each state. Volunteers can play a role by, if you like, keeping the agencies honest. This could involve holding meetings with land managers at a local or regional level to ascertain whether certain actions required of the agencies have been implemented.

To be effective in this role, volunteers need to be armed with the best available information. Have Malleefowl numbers declined in your area? What is the evidence for this decline? What does research or monitoring locally or from other regions tell us about the current risks to Malleefowl? Armed with this sort of information, volunteers are in a stronger position to influence positive outcomes.

I would encourage everyone to take the opportunity of this forum to absorb the most recent information on Malleefowl presented at this forum so that we can all apply it to good effect on our return home.

I would also like to thank Stephen Davies and the Steering Committee for organising our second National Malleefowl (and first WA Megapode) Forum. They have developed a very comprehensive program which offers something for everyone with an interest in this intriguing species and family.

Aims

At our last national forum, Denis Saunders spoke about the significance of fragmentation of habitat within semi-arid Australia and of the likelihood of species ‗relaxation‘ within isolated remnants. If we look at the national situation with regard to mallee habitats, we see that the situation varies considerably from state to state. In Western Australia and South Australia, the clearing of mallee tended to be more extensive than in Victoria and New South Wales. However, in these latter two states, remnant mallee has been routinely leased for stock

2 grazing. This has changed relatively recently when there has been an increase in the level of reservation of mallee vegetation with an associated reduction in stock grazing.

The remaining habitat for Malleefowl can hence be categorised as being:

In those areas where mallee vegetation was most extensively cleared, Malleefowl survive in relatively isolated remnants that may or may not have a history of stock grazing, or in vegetation on the fringes of the mallee belt that was considered unsuitable for agriculture. These marginal habitats may now be managed for pastoralism, mining, or be under Aboriginal ownership

In those states where mallee vegetation was not as extensively cleared, larger remnant blocks remain on soils that were marginal for agriculture, e.g. lowan sands in Victoria, or perhaps where clearing did not take place for other reasons, e.g. western division of New South Wales. These areas were generally subjected to prolonged stock grazing which may have ceased in recent years to allow them to be managed for conservation, as has been the case with the Victorian Mallee parks declared in 1991

The National Recovery Team suggested to the Forum Steering Committee that conservation of Malleefowl in these contrasting circumstances (i.e.small remnants of better habitat and larger blocks of marginal habitat) would provide a good basis for the discussion sessions on Monday. One, or perhaps both, of these circumstances will be relevant to everyone in this room.

So what are the broader aims of the National Forum? From my perspective they are primarily about renewing our collective enthusiasm for Malleefowl conservation. The Forum should allow us to focus our efforts to best effect armed with new information and a new understanding. This time round there is also the opportunity to hear about research into other megapodes from northern Australia, Indonesia, and the Phillipines.

Secondly, the Forum provides the ideal opportunity to socialise with those from other regions and compare notes and celebrate the progress made since our last National F0orum in Mildura in 2004. And there have been a number of achievements when we reflect back on the checklist from Mildura.

We now have a documented national monitoring framework thanks to the work of Ann, Joe and others. This work formed part of the Natural Heritage Trust-funded ‗Malleefowl Monitoring, Population Assessment & Conservation Project‘. Thanks need to go to the Mallee Catchment Management Authority in Victoria for their work in initiating and coordinating this project

Following a meeting earlier this year in Adelaide, there are positive steps towards development of a national database

The approach of adaptive management advocated in the recovery plan will provide the basis for development of a broader understanding of the steps we need to take collectively to conserve this species

How this will be coordinated remains unclear, and it seems unlikely that we will have funding for a position of national coordinator

Communication between groups is as good as it has ever been thanks mainly to the multi- regional project and to the re-emergence of ‗Around the Mounds‘

I believe that this forum will build on the achievements since 2004 and further strengthen our national coordination and communication. By the time of the next forum I expect that we will be in a position to clarify the status of Malleefowl at a national level based on rigorous

3 ongoing data collection. A lack of data will no longer be an impediment to Malleefowl recovery.

4 2. PROGRESS TOWARDS A NATIONAL SYSTEM OF MONITORING AND ADAPTIVE MANAGEMENT FOR MALLEEFOWL

Joe Benshemesh La Trobe University, School of Biological Sciences, Bundoora, Vic.

Abstract

One of the main aspirations voiced at the last Malleefowl Forum in Mildura was to standardise and consolidate the monitoring and move toward a more dynamic phase in which monitoring is used to assess management in terms of its benefit to Malleefowl. Three years on, there is much to report and we are well on our way to reaching this goal. In particular, a multi-regional project was developed to standardise and improve the monitoring and to analyse the data collected so far. These data are impressive, totalling nearly 600 site-years at over 60 sites across Australia, and were analysed by Professor Richard Barker and Ryan Macfarlane (University of Otago, NZ) in regard to environmental factors such as rainfall, landscape, and management practices such as fox baiting. The results confirm that Malleefowl are declining nationally even in reserves set aside for their protection and challenge long-held views about management practices such as fox control which we found has not been effective at benefiting Malleefowl. The analysis has demonstrated that we do not know as much as we thought about how best to manage Malleefowl populations, and highlights the need for better ways of identifying management practices that are beneficial and effective. Improving, standardising and analysing the monitoring data over the past few years have provided a firm foundation on which to build such an ‗adaptive management‘ system. The system being designed will be underpinned by the monitoring conducted by community groups and, by linking management and research, will provide a core action for Malleefowl recovery for many years to come. Current studies are providing a framework for this adaptive management system and developing the necessary databases. These developments represent a shared vision among community groups, managers and scientists, and have been fuelled by enthusiasm and a sense of purpose as much as funds. Further development will require even greater opportunities for community involvement in Malleefowl conservation and a high degree of collaboration between state governments and numerous Natural Resource Management organisations

Introduction

Within the past century the range of Malleefowl has contracted, particularly in arid areas and at the periphery of its former range, and severe declines have occurred in southern agricultural areas due to the clearing of the mallee for wheat and sheep production (see Benshemesh 2006c). The fate of Malleefowl within the remaining habitat is uncertain and declines have been described in many protected areas across Australia (Brickhill 1985, 1987, Priddel 1989, 1990, Priddel & Wheeler 1995, 2003, Benshemesh 2007b, Priddel et al. 2007), and the species is regarded as threatened in every state in which it occurs and is listed as Vulnerable nationally. While land management agencies and individuals grapple with ways of reversing these declines, few techniques have actually been proven to be effective and there remains considerable uncertainty about how best to benefit the species.

Against this worrying backdrop, the role of monitoring has become central to the Malleefowl conservation effort. Formal Malleefowl monitoring programs started in most states in the late 1980s and early 1990s, and provide fundamental information on trends in Malleefowl abundance at a number of sites across Australia. This information is needed in order to assess the conservation status of the species across their range and to identify areas in which the species is declining. Perhaps even more importantly, monitoring provides a means of measuring the effects of naturally occurring events and the effectiveness of management actions on Malleefowl numbers.

Monitoring populations involves obtaining reliable and repeated measures of their numbers in order to measure changes in population size. Malleefowl are shy and elusive birds and counts of the birds themselves are very difficult, but their mounds are conspicuous and provide a reliable means of measuring the abundance of breeding birds in areas where they

5 are moderately common (Benshemesh 2004). This is because Malleefowl tend to renovate old mounds rather than construct new mounds afresh each year (Frith 1959), so that each old mound is a potential site for breeding. Annually checking the known mounds each spring thus provides a good estimate of the trends in breeding numbers at each carefully delineated site. New mounds are occasionally built by the birds and a thorough re-search of monitoring sites is required every few years to record these and ensure accurate estimation of breeding numbers.

Monitoring Malleefowl is well suited to volunteer involvement, and volunteers have made, and continue to make, an enormous contribution to Malleefowl conservation through monitoring programs. In fact, most monitoring that occurs across Australia is undertaken by volunteers and in many areas volunteers are responsible for all aspects of organizing and conducting the monitoring, including data storage, vetting and analysis. Numerous volunteer groups, as well as government and non-government agencies, and individuals are involved in this effort and maintaining standards and efficiency in the face of this diverse interest has become a major challenge.

Here, I wish to overview some of the most important developments that have occurred in regard to Malleefowl monitoring since the last National Malleefowl Forum held in Mildura (Vic) in 2004 (Victorian Malleefowl Recovery Group 2004). At that time it was clear that the monitoring effort was severely fragmented across Australia and that monitoring methods varied, making comparisons difficult. Moreover, although there were nominally nearly 100 monitoring sites across Australia, there was no central list, let alone data that could be readily accessed, and after several decades of dedicated effort by volunteers and government agencies much of the data was not collated, verified or analysed.

In acknowledgment of this growing problem, one of the main aspirations voiced at the Malleefowl Forum in Mildura in 2004 was to standardize, consolidate and analyse the monitoring at a national scale and to move toward a more dynamic phase in which monitoring is used to assess management in regard to its benefit to Malleefowl. In response to this wide community support, an application to Natural Heritage Trust was developed by Julie Kirkwood of the Threatened Species Network in collaboration with community groups, state and regional authorities and the Malleefowl Recovery Team. The application addressed much of Action 9 of the National Malleefowl Recovery Plan and was successful, with two years of funding being granted for a range of office-based and on-ground works. Work on the ―National Malleefowl Monitoring Population Assessment & Conservation Action Project‖, which simply became known as the ‗multi-regional Malleefowl project‘, started in 2006 guided by a national steering committee on which all monitoring groups were represented, and administered by the Mallee Catchment Management Authority (Victoria) in collaboration with the Victorian Malleefowl Recovery Group. Its main aims were to:

improve monitoring and develop a consistent national monitoring system collate and analyse monitoring data in regard to management and environmental variables develop the monitoring program so that management actions that are most beneficial to Malleefowl can be identified, and advise regional Natural Resource Management bodies on how best to promote Malleefowl conservation within their region

This paper is largely an outline of the technical achievements of this project over the past two years towards the development of a national monitoring system for Malleefowl. However, it should be noted that I do not attempt to outline all the achievements of the project, which included on-ground works such as fencing, extensive training workshops and national round- table discussions (Ann Stokie, this volume, discussed some of these). It should also be noted that a large number of people from volunteer groups and agencies collaborated on this project and much of the success of this project was due to their guidance and the high degree of cooperation and enthusiasm with which they contributed. Indeed, the renewed spirit of

6 cooperation and common aims across the country is perhaps one of the greatest achievements of the multi-regional Malleefowl project.

Improving the monitoring

Collating and over-viewing the data sets

One of the first actions of the tasks in the multi-regional project was to collate and review the data that had been collected to date. Collating the data turned out to be surprisingly difficult (Benshemesh 2006a) despite the willingness of data custodians, because data sets were often fragmented and were not readily accessible even within individual states or regions. Much of the data existed only on paper and was entered onto databases for the current project by teams of volunteers, while some original records could not be located at all. These difficulties in collating the monitoring data were frustrating, but most importantly also demonstrated the need for improvements in data management and strengthened the resolve of all those involved in the hunt for data to work toward a national database for Malleefowl monitoring records.

A ‗gap analysis‘ of the collated data was also performed (Benshemesh 2006b) which examined the gap between the available data and the data we expected or would reasonably have liked to have for the two main tasks ahead: reviewing the effectiveness of the data routinely collected in the Malleefowl monitoring program, and analysing the trends in Malleefowl in regard to environmental variables. While over 20,000 mound visits had been recorded across Australia, the gap analysis basically looked at what was missing and how useful the data sets in their various forms were likely to be. For example, a relatively common problem in some datasets was that they were incomplete because observers only visited some of the mounds known in an area during a particular season. This meant that the actual number of breeding Malleefowl at a site was uncertain, and in some severe cases this rendered the data set useless for trend analyses. Overall, the assessment of the strengths and weaknesses of the collated data proved helpful for developing national standards (see below).

Apart from systemic problems in some monitoring data sets, it was also clear that information on past management practices (such as fox control) were difficult to obtain from state agencies and that it would be worthwhile for the Malleefowl monitoring program to obtain these records each year, and perhaps also monitor important environmental factors such as the abundance of predators and herbivores, food availability and habitat changes.

Review of the routinely collected data: streamlining and justifying the monitoring

The primary aim of the Malleefowl monitoring program is to track changes in the number of birds inhabiting specific areas. Observers (mostly volunteers) examine and categorise all the known mounds at each site as either ‗active‘ (currently used as an incubator) or not active (Patford et al. 2004). To enable vetting of records and the detection of errors in judging the activity of mound, the size, shape and appearance of mounds is also described each time a mound is visited. These descriptors have been defined previously (Benshemesh 1997) and the resulting protocols have been used in both Victoria and South Australia from the early 1990s, and in Western Australia since 2005.

No substantive changes have been made to the Malleefowl monitoring protocols since the early 1990s, although some new data fields have been added and others have been more rigorously defined to reduce confusion. This conservatism was necessary to enable comparisons through time, but as the usefulness of the data had never been assessed, some data collected each year may have been unnecessary and provided little useful information. Also, new technologies over the past decade or so have changed the ways that data may be validated and reduced the need for some types of descriptive data. GPS, digital photography, and digital data capture using Palm handheld computers and Cybertracker software have all been introduced to the monitoring program since 2001. These recent changes, and the goal of a unified national Malleefowl monitoring approach, made a review of the usefulness of monitoring data timely and provided an opportunity to consider improvements. The review

7 (Benshemesh 2007c) focused on the data collected in Victorian since 1995 which provided a consistent and detailed data set and a good basis from which to examine the merits of the data routinely collected. The steering committee identified several questions which provided a focus for the review:

Are the descriptors useful in validating/vetting records of active nests?

As anyone involved in monitoring Malleefowl mounds will know, there are a lot of questions to answer in the field regarding the appearance of mounds and the necessity of these data is sometimes queried. Descriptor variables enable data-vetting, but the need for detailed descriptions has decreased over recent years with the availability of digital photos and accurate GPS technology. In the review, a statistical assessment of the descriptive data collected routinely at mounds showed that about half were valuable for corroborating breeding counts and should be collected annually, whereas other data fields were of less value and need only be used occasionally, such as to describe new mounds that are added to the monitoring program. The ensuing simplification and streamlining of the monitoring process would result in further emphasis being placed on the most important measures, and the increased efficiency would save time and effort that could be used more productively to measure other variables of interest such as habitat quality or predator numbers. We expect the streamlined protocols to be in use for the 2008/9 monitoring season.

Do the descriptors tell us much about how long it‟s been since a mound was active?

As mounds age they accumulate crust, herbs and moss, and lose eggshell and height. We found statistically significant relationships between the number of years since a mound was last active and each of these variables, but using these relationships to predict when a mound was last active results in fairly crude estimates of limited application. There seems little need to collect these data every year, although describing mounds in regard to these features would be useful when mounds are newly found and added to the monitoring lists for a site.

Could the efficiency of the monitoring program be improved by omitting very old mounds, or would this compromise the accuracy of the monitoring?

People involved in monitoring occasionally request very old and derelict mounds to be removed from monitoring lists as well, as structures that have been dubiously identified as mounds, and this is usually done in an ad hoc manner if the requests persist. This is a minor issue at most monitoring sites, but in a few sites very old mounds seem to be common and may provide an unnecessary burden on monitoring. To examine this question, we looked at the probability of mounds with different characteristics becoming active again in later years. We found that objective criteria could be used to identify mounds that were unlikely to be used again and that these mounds could be omitted with little sacrifice in the accuracy of monitoring breeding numbers, provided the number of mounds omitted was small. Nonetheless, it seems prudent to be selective and cautious in removing any mound from annual monitoring lists, and mounds that are identified as not requiring annual inspection should still be visited occasionally in order to check for changes.

How often should sites be re-searched?

This question has vexed the monitoring from the beginning as it is known that Malleefowl occasionally make new mounds, although the rate at which these are created has been unclear, especially at sites where Malleefowl populations are stable or increasing. The question also has a large bearing on the number of sites that can be adequately monitored by a group or agency because thoroughly searching a site requires at least ten times the effort compared to simply visiting the known mounds. Fortunately, it seems that re-searches of sites is not usually as important as previously supposed. Analysis of data from 30 re- searches of monitoring sites which resulted in the addition of over 170 mounds showed that re-searches tended to discover mounds that were actually old and inconspicuous, and had probably been missed during previous searches rather than having been newly constructed during the intervening years. There was no evidence to suggest that the number of newly found mounds was related to the number of years since sites were last searched, or with the

8 number of times a site had been searched. At long-unburnt sites, previously known mounds were three times as likely to become active in later seasons as mounds that were newly found, although this was not the case at sites that were burnt 20-30 years previously where newly found mounds were more likely to be active in later years than known mounds. In light of these findings, re-searches every 3-5 years may be appropriate in sites burnt 20-30 years previous and wherever it is likely that there maybe a shortage of old mounds suitable for renovation at a time when Malleefowl numbers were increasing. At long-unburnt sites, routine re-searches every 5-10 years would seem to provide a reasonable compromise between effort and necessity.

Are data on animal signs at mounds useful?

The frequency of prints and scats of animals at Malleefowl mounds is information that is incidental to the main objective of monitoring, but which is nonetheless collected because it is potentially useful for interpreting changes in Malleefowl populations. These animals include predators (foxes, cats, dogs) and possible competitors (kangaroos, rabbits, goats, emus) of Malleefowl, and large changes in their abundance may affect Malleefowl populations. The abundance of fox sign is especially important because of the direct threat that this introduced predator may pose to Malleefowl. The frequency of fox scats on mounds has been used as an indirect means of assessing trends in fox abundance and appears sensitive to control efforts (see below), suggesting it does provide a useful index of fox numbers. However, it is generally unclear whether signs of other animals at mounds also provide an index of their abundance. In any case, given the importance of monitoring other species in order to understand Malleefowl trends, other more conventional techniques such as sand-pad monitoring are warranted at Malleefowl monitoring sites and these could be carried out by volunteers with suitable training.

Developing a national manual

A major achievement in the past year has been the development and publication of a thorough national Malleefowl monitoring manual written by representatives of monitoring groups from around Australia (Natural Heritage Trust National Malleefowl Monitoring Project 2007). This democratic process was guided by Peter and Ann Stokie (see elsewhere in this volume) who brought representatives together in round-table discussions and collated and circulated their contributions. The production of the manual tapped into the collective experience of a large number of people, as well as the analyses outlined above, and was achieved by consensus. While discussions were always lively and many viewpoints were expressed, the group managed to achieve this goal with an extraordinary spirit of mutual purpose and harmony.

Improving data management

The introduction of electronic data capture in the field, using handheld computers and Cybertracker software, was a major improvement in data collection. However, the difficulty in collating monitoring records from around Australia for the multi-regional project showed that data management is a major weakness of the monitoring system. On a more routine level, data custodians in each state still struggle every year with the same issues of entering, downloading, screening, summarizing and filing the data collected. Given this degree of duplication and effort each year, and the inadequacies in data retrieval, a centralised database for the Malleefowl monitoring project is desirable and is currently being planned with additional funding support from Australian Government Department of the Environment and Water Resources. Richard and Margaret Alcorn, whom many people have met in national meetings, will develop the database over the next few months under the guidance of representatives from each state who have been involved in data management.

The database is being modelled on basic functionality of the current Victorian Malleefowl database, but it will also be designed to conduct many of the routine tasks that are currently done manually. We plan for it to be web-based to provide easy access from anywhere in Australia. Its main purpose is to provide a service to coordinators of monitoring in each state by assisting them in managing the large volume of data that is routinely collected while

9 maintaining the standards and operational procedures of the monitoring program. Where possible, the intention is to enable volunteers to manage the data they collect and to download information required to monitor Malleefowl in their area.

Another important task of the new database will be in providing a means of feedback and reporting in a secure environment. By default, data will be confidential and access to sensitive data, such as the locations of mounds, will be controlled by the people who collect and submit data. The focus will be on developing an efficient system of moving monitoring data from the field to a central database where the data can be inspected for errors, corrected if need be and stored for later use. Another priority will be to build a feedback system into this for the volunteers involved in monitoring, as well as more general reporting on Malleefowl trends at the site, regional, state and national scales. We envision a series of structured reports, and possibly even individualized reports, so that people get the data in the form that is most useful to them.

In short, the database will be designed to service users across Australia while maintaining the integrity of the data and providing community volunteers, managers, and state wildlife authorities with the data they require in the form they require it.

Analysis of monitoring data: Are Malleefowl populations declining, and if so, what environmental factors might be responsible?

A major aim of the multi-regional Malleefowl project was to collate and analyse the data on Malleefowl trends that had been collected to date. This data had accumulated since the late 1980s when monitoring programs were initiated in most states, but had never before been collated let alone collectively analysed. This rich treasure of data describes the trends of Malleefowl at numerous monitoring sites in New South Wales, South Australia, Victoria and Western Australia over many years, and thus shows the influence that environmental factors (such as rainfall) and management actions (such as fox control) may have had on Malleefowl populations. After thoroughly vetting and screening all the data on Malleefowl trends from the national monitoring effort, we were left with high-quality data from over 60 sites across Australia spanning up to four decades and representing 590 counts of Malleefowl breeding (and over 20,000 mound inspections). However, the data-set was also complex due to the variable amount and quality of data from each site, and consequently sophisticated statistical analysis was required. This expertise was provided by Professor Richard Barker and his student Ryan Macfarlane (University of Otago, New Zealand) who undertook the analysis using a statistical modelling technique known as hierarchal Poisson (loglinear) regression.

While the main body of data comprised annual counts of mounds that were used for breeding each year for each of these monitoring sites, we also collated information regarding fox control efforts, rainfall, landscape fragmentation, and fire history in order to examine the effects of these variables on Malleefowl breeding numbers. Many people contributed to this pooling of information, including Sharon Gillam and Peter Sandell (fox control), Dr Kate Callister (fire history), Roman Urban (fragmentation), Dr Graham De Hoedt (rainfall modelling), as well as numerous rangers, land managers/holders and volunteers. A few of the monitoring sites considered in this study were within landscapes that are known to have sizeable goat and/or kangaroo populations, but we were unable to obtain reliable information on the abundance of these grazers/browsers at many sites and were unable to examine the possible effects in our analysis. Most of the monitoring sites are within reserves, and none was believed to be routinely grazed by sheep.

On the basis of previous observations and studies of Malleefowl ecology reported in the literature, we expected breeding densities to gradually decline where foxes were abundant and where the landscape was highly fragmented, and to decline more suddenly during dry seasons and following wildfires. Conversely, we expected to see Malleefowl counts increase in areas where fox baiting was more intensively practised, and in the decades following an area being burnt.

The results of the analysis confirmed that, on the whole, Malleefowl have declined nationally even in reserves set aside for their protection. This downward trend was most evident in

10 South Australia (142 counts of breeding numbers at 23 sites) where the decline was statistically significant, and Western Australia (36 counts at 6 sites) where it was not significant (possibly due to low sample sizes). In Victoria (365 counts at 29 sites) no clear trend was evident despite the large number of sites and monitoring records. In New South Wales (47 counts at 6 sites) we found a significant positive trend in Malleefowl, although it should be noted that we only obtained monitoring data from two large reserves in the south- west corner of New South Wales (Tarawi and Mallee Cliffs). Elsewhere in central and western New South Wales several studies have documented declining breeding numbers, especially in very small (<500 ha) isolated remnants (Brickhill 1985, 1987; Priddel 1989, 1990; Priddel & Wheeler 1995, 2003) and it would appear that the Malleefowl trends in Tarawi and Mallee Cliffs are an exception in New South Wales and not representative of the remainder of that state.

Of greater interest were the results of the analysis of the environmental factors that were associated with these population trends. Populations go up and down for a variety of reasons, and the large amount of data that was collated for this project provided a unique and powerful opportunity to identify which factors were most responsible for the Malleefowl trends at the 64 sites in the analysis. This number of sites and seasons is necessary to distinguish with any confidence between different factors operating in differing geographic and climatic regions.

Fire

As expected, we found that fire had a negative effect on Malleefowl numbers and that Malleefowl tended to increase thereafter. These findings broadly agree with previous studies on the short and longer term responses of Malleefowl to fire; that Malleefowl breeding densities are lower in areas burnt within the last few decades than in old-growth mallee has been established in several studies (Woinarski 1989, Benshemesh 1990, Clarke 2005), and the mitigating effects of patchy or incomplete fires has also been documented (Benshemesh 1990). Fire is clearly a threat to Malleefowl, often occurring on a vast scale in mallee when it may depress breeding numbers for several decades. However, despite the strong effects of fire shown in this study, fire does not provide a satisfactory explanation for the declines evident in Malleefowl populations as very few sites in this study were burnt during the course of monitoring.

Landscape configuration

We used two site variables to investigate the effect that landscape configuration may have on Malleefowl populations: the size of the block of habitat in which the monitoring sites was located (Patchsize), and the proportion of cleared land within five kilometres of the centre of the monitoring sites (5kClr). Neither of these variables was significantly related to trends in Malleefowl populations. Thus, contrary to our expectations, there was no evidence in our analysis that small isolated patches were more prone to Malleefowl declines than larger patches in the short to medium term. In the long term, small and isolated populations are subject to genetic deterioration and chance extinctions due to low numbers, and it seems likely that populations in these small reserves would inevitably decline without careful management to ensure that Malleefowl populations are not completely isolated.

Another reason for the lack of measurable effect of patch size is that Malleefowl may obtain some benefit from cleared land, especially crop lands that usually surround small habitat patches. Malleefowl are frequently observed feeding on herbs and fallen grain on cleared land at the edge of reserves, and this additional food source may be especially important during droughts and times of food shortage and mitigate the negative effects of small patch size and fragmentation. It is also worth noting that sites near cleared land may also naturally support higher numbers of Malleefowl: clearing was concentrated toward more fertile soils in higher rainfall areas suitable for agriculture, and this is also where Malleefowl densities tend to be highest. In our analysis, the amount of cleared land within five kilometres of monitoring grids had a positive effect on Malleefowl trends, although this effect was not significant.

11 That Malleefowl have persisted over the medium term in small patch sizes and in fragmented landscapes is an encouraging result and suggests that, with appropriate management to avoid population and genetic bottlenecks, such sites will continue to be of importance to Malleefowl conservation.

Rain

Winter rain has a pronounced effect on Malleefowl breeding numbers and this was evident in our analysis. We found significant positive effects of winter rain on Malleefowl, and this was expected as the birds often skip breeding in dry years. What was more surprising was that there were significant lag effects for the following two, three and four years (the one year lag effect was weakly positive but not significant), meaning that the number of Malleefowl breeding at a site is influenced by the winter rainfall not just in the current year, but in the previous four years (at least). Exactly why this is the case is unclear and may be related to food production, recruitment of young into the breeding population, or both of these factors.

The importance of this finding can hardly be overstated. Lower than expected winter rainfall has characterised most monitoring sites over the past decade or so, and may provide an explanation for the declines in Malleefowl described in this study. More than 80% of the sites in this study experienced lower winter rain over the past 10 years, and 95% over the past five years, compared with long term averages between 1961-90 (a period which is accepted as a recent meteorological standard). Given the lag effects for winter rain on Malleefowl of up to four years shown in this study, the average four-year winter rainfall deficit (Figure 1) is especially pertinent to Malleefowl and graphically shows that while winter rain was higher in the early 1990s than the average for the previous 30 years, since 1996 the four year deficit has increased steadily and in 2004-5 was nearly 20% below that expected.

Given the significant relationships between winter rain and Malleefowl numbers, current predictions of climate change for Australia (Pittock & Wratt 2001) provide considerable cause for concern. The predictions show a decline in winter rain in most semi-arid habitats that support Malleefowl, while the results of analysis of monitoring records suggest that each relatively dry winter will have negative ramifications on Malleefowl breeding populations for at least four years. If current climate predictions are correct, and if climate change is not arrested, substantial declines in Malleefowl populations are likely in the future.

20% 15% 10% 5% 0% -5% -10%

-15% 4-year(%) deficit winter-rain

-20%

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year

Figure 1. Average four-year winter rainfall deficit for the monitoring sites in this study. For each site and year, the deficit was calculated as the average winter (May-September) rainfall over the past four years, minus the expected rainfall (average winter rain for the period 1961- 90), divided by the expected rainfall.

12

Foxes

The most surprising result from our analysis concerned the effect of fox control and fox abundance on Malleefowl breeding populations. While there is no doubt that foxes eat Malleefowl, the degree to which predation by foxes influences Malleefowl numbers has long been controversial and unresolved. This study represented the first attempt to examine this question at the population level across multiple sites. As baiting with 1080 is generally used to control foxes, we estimated the number of baits laid each year within a 100km2 area with the monitoring site at its centre to provide an index of fox control intensity. Such an area is generally regarded as a minimum for effective fox control. To estimate the effect of fox control on fox abundance, we used data on the frequency of fox scats on Malleefowl mounds; these data were routinely collected in South Australia and in Victoria since the early 1990s.

Our analysis showed a strong and significant negative effect of fox control (i.e. baiting) on the incidence of fox scats on mounds, suggesting that baiting did indeed lead to a decline in fox numbers and that more baiting was associated with fewer foxes. However, we found no evidence at all of a significant effect of fox control on Malleefowl breeding numbers either in the year that baiting occurred, or for the following four years.

Although, the four-year lag effect did approach statistical significance, the direction of this effect was positive (i.e. more foxes resulting in more Malleefowl) rather than negative, suggesting that baiting in the longer term might actually be associated with Malleefowl decline rather than recovery (although this was not statistically significant). Thus, while baiting did appear to result in a decline in fox numbers, there was no evidence of a benefit to Malleefowl breeding numbers or amelioration of declines.

This result was surprising as fox predation has been considered a major cause of Malleefowl declines, and the conservation of Malleefowl is one of the most common reasons managers provide for fox baiting campaigns (Reddiex et al. 2004). However, the evidence that suggests fox predation is linked to Malleefowl declines is far from compelling (Benshemesh 2005), and is based largely on three lines of argument: (1) the fact that foxes eat Malleefowl eggs, chicks, juveniles and adults, (2) the vulnerability of captive reared Malleefowl to foxes, and 3) an analogy with medium sized mammals whose populations have been shown to be sensitive to fox numbers.

Our study has been the first to examine this issue directly by comparing trends in Malleefowl with fox-control intensity, as well as measures of fox abundance. We failed to find any evidence that fox control, as it has been practiced over the past two decades around Malleefowl monitoring sites, has been of benefit to Malleefowl abundance. It should be noted, however, that most baiting that occurs in and around Malleefowl monitoring sites is low intensity. Since 1990, about half the sites in this study were baited to some degree, but only about 10% of sites were baited at an intensity which is generally regarded as a minimum for an effective reduction of fox predation on native mammals (>4 baits/km2). Nonetheless, the failure of any clear response by Malleefowl to fox baiting indicates that we have placed too much emphasis on fox control and not enough effort has been directed at discovering the true causes of decline or to find ways of benefiting wild populations.

Conserving Malleefowl in light of management uncertainties: Why we need adaptive management

So where does all this leave us in terms of management? Superficially, it may seem that the multi-regional project has merely concentrated on strengthening the monitoring effort rather than on management actions that might reduce threats to the species. Even the analysis may seem of questionable value given that it has refuted the reasonable and popular idea that fox control is necessary for Malleefowl conservation and provided no clear direction. However, both these positions miss the point and misunderstand the central role that monitoring plays in the recovery of threatened species. Monitoring provides the crucial feedback we need to assess the effectiveness of management actions, as well as the conservation status of the species. Without this feedback, we would be locked into applying management indefinitely,

13 based on conviction rather than empiricism, and there would be little opportunity for testing or improving our practices. The failure of fox-baiting at mitigating Malleefowl declines is a case in point and indicates that we should change our practices.

While the monitoring has already proven to be of great value to Malleefowl conservation, to fulfil its potential the monitoring system needs to be integrated with management and research. Ideally the network of monitoring sites would be used to conduct management experiments with appropriate levels of replication, randomisation and experimental control. Carefully designed management experiments provide more reliable information than descriptive studies (such as our analysis) and the current Malleefowl monitoring system would seem well-suited for such experiments, particularly as the monitoring information is already being collected by agencies and an army of dedicated volunteers. This could be achieved by adopting an active adaptive management approach using the monitoring sites to provide a framework for the monitoring effort at a national level and to better integrate monitoring, management and research. Adaptive management is a pragmatic and collaborative process of ‗learning by doing‘ that confronts uncertainties in management and seeks to gain reliable knowledge through experimental management. Key components of the adaptive management approach include experimental design and modelling, field management treatments and monitoring, structured in such a way that the success of management alternatives can be evaluated with confidence (Walters and Holling 1990). Management actions that are proven to be effective are adopted, and in the case of Malleefowl may be applied broadly or at least where they are most needed.

Adaptive management is an approach well-suited to Malleefowl conservation for a number of reasons. Firstly, as demonstrated in this study there is considerable uncertainty about the effectiveness of management actions in reversing Malleefowl declines and in the role of environmental factors. This uncertainty is likely to increase dramatically if climate change predictions are correct. Adaptive management embraces such uncertainty and provides a means of identifying best management practice in a coherent and statistically meaningful way. Secondly, Malleefowl still occur over much of their uncleared range, providing opportunities for replicating management treatments and controls (non-treatment sites). The current network of monitoring sites represents a tiny proportion of this range and varying management treatments at these sites is unlikely to compromise the conservation of the species. Thirdly, there is already a strong community involvement in Malleefowl conservation at a national scale and a general enthusiasm for collaboration with agencies and land managers. That most of these sites are monitored by an army of volunteers each year is especially valuable and relieves agencies of this critical, but time-consuming and potentially expensive task. Indeed, monitoring is often the most expensive part of carefully designed and replicated management experiments.

Further collaboration between community volunteers, land managers, scientists, and other stakeholders will be required across political boundaries to develop and implement an adaptive management plan for Malleefowl management. We are currently collaborating with scientists and mathematical modellers from the Applied Environmental Decision Analysis research hub and Arthur Rylah Institute (Melbourne) to develop a suitable framework for adaptive management using the monitoring sites. Once this framework is complete, we hope to involve all stakeholders in discussion, planning, and implementation.

While these national plans are underway, it is essential that the many Natural Resource Management bodies across Australia are aware of the importance of the monitoring effort and that they need to look beyond their regional borders and collaborate on a national adaptive management program in order to obtain the maximum benefit for Malleefowl. Natural Resource Memanagement bodies (called Catchment Management Authorities or Catchment Councils) have recently become major conduits for conservation funds flowing from commonwealth agencies, and while there are advantages of this regional focus, there is also a danger that large scale conservation programs will suffer unless Natural Resource Mandagement bodies are made aware of these programs and work collectively and collaboratively toward common goals. In a sense, bureaucratic and administrative fragmentation threatens Malleefowl conservation as much as geographic fragmentation.

14 To this end, the final phase of the multi-regional project was to provide a concise document advising Natural Resource Management bodies of the national plans and providing them with brief summaries of Malleefowl records, monitoring sites and major issues in each region (Benshemesh 2007a). There are about 15 Natural Resource Management regions that contain Malleefowl monitoring sites, out of about 20 in which Malleefowl occur, and the document also advises these organizations of plans of the Malleefowl Recovery Team to implement adaptive management across monitoring sites (Benshemesh 2005) and the need to collaborate if adaptive management is to be effective.

Concluding comments

The past three years since the last Malleefowl Forum has been a busy time for those involved in Malleefowl conservation, particularly in regard to the multi-region project which has developed national standards, disseminated protocols, reviewed practices and begun to establish central databases, analysed past data, and generally facilitated the monitoring effort on a national scale through meetings, workshops and documents. Here, I have attempted to give an overview of the technical achievements of the past three years which have been largely guided by community aspirations and were only made possible by the dedication of volunteers and agencies in collecting, questioning and sharing the monitoring data over the past two decades.

The future looks bright too, and the shared vision of conserving Malleefowl effectively and efficiently has gained momentum over recent years and, in turn, has provided those involved in monitoring with a greater sense of purpose and necessity. The community of volunteers in particular has proven to be a major driving force in Malleefowl conservation and seems to grow stronger each year. As important as volunteers are to the monitoring effort, volunteers cannot implement management treatments, and the active collaboration, involvement and commitment of agencies and land managers will be essential to manipulate the management of monitoring sites according to national plans. Likewise, we need the involvement of ecological modellers and statisticians to tell us how to structure management at monitoring sites so that we progressively gain reliable knowledge of how best to conserve Malleefowl.

Further progress will, then, depend largely on the willingness of Natural Resource Management bodies, state authorities and land managers, as well as community groups, researchers and academics, to work collaboratively toward improved conservation outcomes for Malleefowl.

References

Benshemesh, J. (1990). Management of Malleefowl with regard to fire. pp 206-211 in Noble, J. C., Joss, P. J. and Jones, G. K. (eds). The Mallee Lands, a Conservation Perspective. CSIRO, Melbourne.

Benshemesh, J. (1997). Manual for monitoring Malleefowl. Parks Victoria and Department of Conservation and Natural Resources, Mildura.

Benshemesh, J. (2004). Monitoring malleefowl: Options, problems and solutions. pp 128-134 in Proceedings of the National Malleefowl Forum 2004, Mildura. Victorian Malleefowl Recovery Group, http://www.malleefowlvictoria.org.au/seminar.html.

Benshemesh, J. (2005). National Recovery Plan for Malleefowl (draft). Malleefowl Recovery Team and Department of Environment and Heritage, Adelaide.

Benshemesh, J. (2006a). Data collation report. Unpublished report to the Victorian Malleefowl Recovery Group and the Mallee Catchment Management Authority, Mildura.

15 Benshemesh, J. (2006b). Gap analysis of monitoring and associated data and an update on data collection. Unpublished report to the Victorian Malleefowl Recovery Group and the Mallee Catchment Management Authority, Mildura.

Benshemesh, J. (2006c). National Recovery Plan for Malleefowl (draft). Malleefowl Recovery Team and Department of Environment and Heritage, Adelaide.

Benshemesh, J. (2007a). Advice to Natural Resource Management regions regarding management and monitoring of malleefowl. Unpublished report to the Victorian Malleefowl Recovery Group and the Mallee Catchment Management Authority, Mildura.

Benshemesh, J. (2007b). Malleefowl Monitoring in Victoria: 2006/7. Unpublished report to Victorian Malleefowl Recovery Group, Melbourne.

Benshemesh, J. (2007c). Review of data routinely collected in the national malleefowl monitoring program. Unpublished report to the Victorian Malleefowl Recovery Group and the Mallee Catchment Management Authority, Mildura.

Brickhill, J. (1985). An aerial survey of nests of Malleefowl Leipoa ocellata Gould (Megapodiidae) in central New South Wales. Aust. Wildl. Res. 12:257-261.

Brickhill, J. (1987). The conservation status of Malleefowl in New South Wales. M.Sc. thesis, University of New England, Armidale.

Clarke, R. (2005). Ecological requirements of birds specialising in mallee habitats: modelling the habitat suitability for threatened mallee birds (draft). Department of Zoology, La Trobe University, Melbourne.

Frith, H. J. (1959). Breeding of the Mallee Fowl, Leipoa ocellata Gould (Megapodiidae). CSIRO Wildl. Res. 4:31-60.

NHT National Malleefowl Monitoring Project. (2007). National manual for the Malleefowl montoring system. Victorian Malleefowl Recovery Group and Mallee CMA, Melbourne.

Patford, R., Stokie, A & Stokie, P. 2004. Ordinary people doing extraordinary things. Pages 146-153 in Proceedings of the National Malleefowl Forum 2004, Mildura. Victorian Malleefowl Recovery Group, http://www.malleefowlvictoria.org.au/seminar.html.

Pittock, B. & Wratt, D. (2001) Australia and New Zealand. In: Pittock, B. & Wratt, D. (Eds.) Climate Change 2001: Impacts, Adaptation, and Vulnerability. Cambridge University Press, Cambridge, UK.

Priddel, D. (1989). Conservation of rare fauna: the Regent Parrot and the Malleefowl. pp 243- 249 in Noble, J. C. and Bradstock, R. A. (eds). Mediterranean Landscapes in Australia: Mallee Ecosystems and Their Management. CSIRO, Melbourne.

Priddel, D. (1990). Conservation of the Malleefowl in New South Wales: an experimental management strategy. pp 71-77 in Noble, J. C., Joss, P. J. and Jones, G. K. (eds). The Mallee Lands, a Conservation Perspective. CSIRO, Melbourne.

Priddel, D., & R. Wheeler. (1995). The biology and management of the Malleefowl (Leipoa ocellata) in New South Wales. NSW National Parks and Wildlife Service, Sydney.

Priddel, D., & Wheeler, R. (2003). Nesting activity and demography of an isolated population of malleefowl (Leipoa ocellata). Wildlife Research 30:451-464.

16 Priddel, D., Wheeler, R. & Copley. P. (2007). Does the integrity or structure of mallee habitat influence the degree of fox predation on Malleefowl (Leipoa ocellata)? Emu 107:100- 107.

Reddiex, B., Forsyth, D. M., McDonald-Madden, E. M., Einoder, L. D., Griffioen, P. A., Chick, R. R. & Robley, A. (2004). Review of existing red fox, wild dog, feral cat, feral rabbit, feral pig, and feral goat control in Australia. I. Audit. Arthur Rylah Institute for Environmental Research, Department of Sustainability and Environment, Melbourne. [Internet: http://deh.gov.au/biodiversity/invasive/publications/feral-audit/pubs/audit- final.pdf].

Victorian Malleefowl Recovery Group. (2004). Proceedings of the National Malleefowl Forum 2004, Mildura. Victorian Malleefowl Recovery Group, http://www.malleefowlvictoria.org.au/seminar.html.

Walters, C. J., & Holling, C. S. (1990). Large-scale management experiments and learning by doing. Ecology 71:2060-2068.

Woinarski, J. C. Z. (1989). The vertebrate fauna of Broombush Melaleuca uncinata vegetation in north-western Victoria, with reference to effects of broombush harvesting. Aust. Wildl. Res. 16:217-238.

17

3. TOWARDS A NATIONAL SYSTEM: THE POWER OF THE PEOPLE

Ann Stokie NHT Malleefowl Project Convenor Victorian Malleefowl Recovery Group

A review of the implications of mound monitoring: detailed discussion of the Natural Heritage Trust Multi-regional Malleefowl project

Where have we been?

The application for funding of a multi-regional Malleefowl project arose directly from the recommendations of the 2004 Mildura National Malleefowl Forum. It is significant and extremely important to be able to bring the outcomes of the project to this Katanning National Malleefowl Forum.

The 2004 Forum recommendations specifically requested a number of actions to bring the monitoring efforts across Australia into some order so data could be adequately assessed and standardised. Actions to be undertaken included:

The development of a national register of Malleefowl Monitoring programs A commitment to a National Monitoring Framework The adoption of standard protocols for monitoring A review of the adequacy of existing monitoring programs The importance of the co-ordinating role of the National Malleefowl Recovery Team

The 2004 Forum considered that the best way to achieve these actions was through a national co- ordinator and recommended the securing of funding for a National co-ordinator to facilitate this fairly ambitious agenda.

The planning for funding for the actions plans and the National co-ordinator was largely undertaken under the direction of the National Recovery Team, coordinated by Julie Kirkwood from WWF who was a Victorian member of the Recovery Team.

The funding application to the Natural Heritage Trust was widely circulated in its formative stages, and supported by a large number of organisations large and small across Victoria, South Australia and Western Australia. Significantly, Government and Non Government organisations and volunteer groups were all part of the supporting group, and these groups continued to work together throughout the whole two year project.

The funding application was successful, with the Natural Heritage Trust announcing the outcome in June 2005. The biggest sticking point with the project was that there was no funding for a National co-ordinator to undertake the organisation of the tasks, and insufficient funds for a consultancy firm to take on the job without compromising the scope of the project. The concept of a volunteer organisation taking on the running of the project was canvassed, and the Victorian Malleefowl Recovery Group was approached. This group has good credentials, having organised the 2004 National Forum, and having managed the Victorian monitoring since 2000. The committee structure and the experience in the group were considered to be more than adequate to manage the project in a professional way. The Victorian Malleefowl Recovery Group accepted the challenge and therefore the Natural Heritage Trust Malleefowl Project was managed by a volunteer group.

The funding agreement was signed late in 2005. The Mallee Catchment Management Authority was the lead organisation and the Victorian Malleefowl Recovery Group undertook to manage the project, to allocate monies for consultancy and to ensure the project goals were met in a timely and professional fashion.

Where are we now?

18

The last two years have seen an extraordinary movement towards developing a National approach to address the conservation needs for Malleefowl, an agreed National System for the monitoring of Malleefowl, and agreement on National Standards.

What has been particularly striking is the involvement of such a large number of people both land managers and volunteers. The activities have included:

The collection of historical data The entering of data onto the database The data analysis The use of the data analysis to refine monitoring National meetings in Melbourne and Adelaide The National Monitoring Manual

The technical and theoretical understandings and findings of these activities are addressed in other places in this forum, particularly in the papers presented by Dr Joe Benshemesh. Whilst the Victorian Malleefowl Recovery Group has been an integral part of bringing these matters to a satisfactory scientific conclusion, I would like to address in particular the human element provided by volunteers that has made these findings possible.

1) The collection of data

The first task that was required was to find out who was monitoring, who was holding data and how that data was stored. The collection of historical data was sought from holders of such data, requiring them to trust the integrity of the Victorian Malleefowl Recovery Group. We were asking holders of data very dear to the collectors to allow the national project to have access to and use their data. In many cases the request meant hours of searching, but in the spirit of utmost cooperation the data were generally found and passed on.

It was easy to find out who was monitoring as members of the Victorian Malleefowl Recovery Group were invited to workshops in both Western Australia and South Australia in the early stages of the project. These workshops covered most geographic areas where monitoring was occurring, and were well attended by the volunteers who were conducting the monitoring. This initial face to face contact was invaluable as the project proceeded, particularly in bringing volunteers together for national meetings and conducting correspondence either by phone or email. We all knew whom we were dealing with, and a friendship and trust across Malleefowl states grew as the project grew.

2) The entering of data onto the database

In many cases historical data was just too difficult or too numerous for individuals to manage in a way that was going to be useful for the national database. In another acceptance of trust, holders of data were prepared to allow experienced Victorian volunteers to deal with this problem. Data from both Western Australia and South Australia were sent to the Victorian Malleefowl Recovery Group to be entered onto the database. Feedback and assurances that the data were useful helped to build up the spirit of cooperation that was going to be needed for the later tasks of producing a National monitoring protocol and National monitoring manual.

3) The data analysis and the use of the data analysis to refine monitoring

The sharing of findings and the distribution of discussion documents was a necessary step to ensure volunteers had sufficient background and information to assist them to contribute to the development of National monitoring manual. This process relied on the use of emails, in particular, and willingness on everybody‘s part to read and respond to sometimes difficult and challenging materials.

4) National meetings in Melbourne and Adelaide

The most important aspect of developing a national monitoring manual was the need to gain consensus from all parties involved in monitoring Malleefowl across Australia. Two national meetings, one held in Melbourne and one in Adelaide, were attended by more than 25 people at each,

19 representing mainly volunteers from State volunteer groups, but also representatives of Government Environment Departments and non government environment groups. At these meetings further trust and acceptance developed and the level of contribution was outstanding. These meetings also generated Malleefowl stories from all over the place, all contributing to our understanding of the species. In fact, it confirmed that the more we know about Malleefowl the less we know about them. All points of view and individual comments were recorded and used in the subsequent development of the National Manual. At the end of both meetings initial drafts of the manual were ratified for further development.

5) The National Monitoring Manual

The National Manual had a long period of input, right from the first meetings at volunteer training sessions, but essentially serious input commenced at the National meetings. However without the use of emails containing the latest drafts moving regularly between volunteers from each state and the Victorian Malleefowl Recovery Group it would not have been possible to come up with a final consensus driven document. It was important that all those who wanted to contribute had every opportunity to do so, until there was no further need to have these discussions. The end result was a truly consensus document which outlined agreed processes and practices for the future.

An enormous spirit of co-operation has prevailed and now we have a National Malleefowl Monitoring Handbook. It was officially launched at the Katanning Forum. We have commitment from three states to monitor using National standards. We have available data from across Australia gathered and analysed.

We have demonstrated how much we can achieve when we work together. We have demonstrated that we can work across different jurisdictions to achieve a common purpose.

We have made substantial progress in building trust. We know a lot more about Malleefowl populations than we did three years ago. We have successfully managed a multi-regional project and we are in a good position to apply for more funds if we need to.

Malleefowl have even made it onto the main stage by recently being adopted as one of the eight flagship species of WWF- Australia.

Where are we going?

The challenge for this Forum is to decide where we go to from here.

The Forum is a good time to celebrate achievements. We need to reflect, to create recommendations for the future. We need to visit this on the last day of the Forum, but we need to start to think about this now.

I submit the following reflections that arise from my position as the project manager of the Natural Heritage Trust National Project.

We need to act nationally, so we need to address whether we need a National Co-ordinator? (NB. This was a recommendation from the 2004 Forum)

There has been untold value arising out of the meetings in Melbourne and Adelaide. Do we need to set up funding and commitment to hold an annual meeting of state representatives, particularly volunteers, to renew national contacts and keep us up to date?

We need a useable National database which holds data very securely, but which can be accessed at different levels so that Australian as well as State pictures can be obtained

We need to establish adaptive management strategies so that better management can occur

We need to continue to develop the National Malleefowl Recovery Team, and investigate its status and funding. Does it need to be more than a body of volunteers who represent each state?

20

We need a National website, or do we?

It would seem appropriate that we consider establishing an Australian Megapodes Group or Association

I look forward to discussing these points and all other relevant points in the plenary session of this Forum.

21 4. MALLEEFOWL IN NEW SOUTH WALES: REVIEW OF PAST AND FUTURE ACTIONS

Peter Ewin Regional Biodiversity Conservation Officer Department of Environment and Climate Change P.O. Box 318, Buronga, 2739, NSW

Abstract

One of the key challenges facing Malleefowl conservation in New South Wales is the ability to increase the volunteer effort in the monitoring and management of the species. Compared with the other Malleefowl states, New South Wales has relatively little volunteer involvement, apart from some assistance from the Victorian Sunraysia Malleefowl Preservation Society and the Goonoo Bigfoot Malleefowl Project. Landholders in the Lower Murray Darling Catchment Management Authority area are playing an active role in Malleefowl conservation, but more participation is needed across the range of the species, which spans seven Catchment Management Authorities. Although the Malleefowl‘s distribution spans from the south western corner to as far east as the Goulburn River National Park, the core populations occur in the south western corner of New South Wales (on both National Parks and leasehold land) and the central New South Wales mallee around Yathong, Nombinnie and Round Hill Nature Reserves and adjoining leasehold lands. The main recovery actions on Department of Environment and Climate Change estates have been aerial and ground-based fox-baiting and monitoring via aerial survey and some ground based monitoring. Control of feral goats is another important activity that will assist in the protection of Malleefowl habitat. Some preliminary results of the aerial surveys are presented here, with mound to mound surveys indicating an increase in breeding activity following fox-baiting. Opportunistic monitoring of the one of the eastern most populations in the Goonoo State Conservation Area (formerly Goonoo State Forest) near Dubbo, has been undertaken by members of the Goonoo Bigfoot Malleefowl Project, but unfortunately about 25000 hectares (of the 63000 hectare reserve) including much of the core habitat was burnt in a wildfire in 2007. A successful captive breeding program has been operating at the Western Plains Zoo at Dubbo since 1990. In that time, over 500 juvenile birds have been released at Yathong and Nombinnie Nature Reserves to supplement the wild population. It is hoped that with the recent creation of Catchment Management Authorities in New South Wales that more extensive off-park work may be undertaken in New South Wales, particularly in the south-west corner, and this may increase the amount of volunteer work that is undertaken.

Introduction In recent decades, much of the effort undertaken in the conservation of the Malleefowl (Leipoa ocellata) in New South Wales has been undertaken by government agencies. The lead agency has been the New South Wales National Parks and Wildlife Service, which now constitutes one of the divisions of the Department of Environment and Climate Change. The main actions undertaken include: Fox baiting and monitoring Aerial Survey Captive breeding/release of chicks Habitat Management (including fire) Unlike the other states with Malleefowl populations there has been very little coordinated conservation work on the species undertaken by volunteers. For example, in New South Wales there are no Malleefowl monitoring grids that are surveyed using volunteers, despite large numbers of grids being established and monitored by volunteers within the Mallee region of Victoria. Some of the projects in New South Wales that have involved volunteers in the past include: Baseline data within Mallee Cliffs National Park and Tarawi Nature Reserve established from transects walked by Sunraysia Malleefowl Preservation Society and Venturer/Rover Scouts

22 Community involvement (coordinated by Sunraysia Malleefowl Preservation Society to put road signs along the Sturt Highway (and Malleefowl Rest Area) because of birds being hit by vehicles Goonoo Bigfoot Malleefowl Project Ad hoc programs by landholders, such as fox-baiting around active mounds Some landholders within the Lower Murray Darling Catchment Management Authority area have been actively involved in Malleefowl conservation actions such as fox control and grazing removal as part of incentive programs or offsets for clearing approvals.

New South Wales Population National Parks and Wildlife Service (2002) lists the state population as less than 500 pairs based on data from the mid-1990s (Priddel & Wheeler 1995). This follows estimates for the mid-1980s of around 750 pairs (Brickhill 1987). Although the exact size of the current New South Wales population is not known, there is no doubt that the overall population has declined since European settlement (Benshemesh 2000). Figure 1 shows the records within the Department of Environment and Climate Change Atlas of New South Wales Wildlife (2007) for three periods (pre 1990, 1990-2000 and since 2000) and shows the concentration of records is restricted to relatively few areas (chiefly Department of Environment and Climate Change conservation reserves). These figures have been influenced by the way that the data have been collected, particularly with earlier surveys obtaining records from ―private‖ lands that have not been surveyed comprehensively again. Other sites, such as Goonoo State Conservation Area, are biased by larger survey effort (particularly by amateur birdwatchers from Sydney) while other areas, particularly leasehold land in the south west corner of New South Wales have been particularly poorly surveyed. Approximately 80 percent of the New South Wales population occurs in the south western corner of the state within the Lower Murray Darling Catchment Management Authority area (Figure 2). Department of Environment and Climate Change reserves within this area that have Malleefowl are Mallee Cliffs National Park (58000 ha with approximately 45 pairs) and Tarawi Nature Reserve, which with the adjoining Scotia Sanctuary (managed by Australian Wildlife Conservancy) and Nanya Station (University of Ballarat) has approximately 127000 ha with an estimate of 30 pairs. Over 300 pairs are estimated to occur on leasehold land much of which retains native vegetation and includes properties such as ―Petro‖, ―Lethero‖, ―Wamberra‖, ―Arumpo‖ and ―Wampo‖ Stations. Large areas of habitat suitable for Malleefowl in the Lower Murray Darling Catchment Management Authority area has been set aside for conservation (usually as an offset for vegetation clearing approval) in such reserves as Southern Mallee and Property Vegetation Plan Reserves (Figure 2). Domestic stock grazing is excluded from these reserves, and the leaseholder has other management requirements, such as removal of feral goats. The situation for the remaining 20 percent of the New South Wales population is significantly different. This portion of the population occurs on the central west slopes and plains within an area bounded by Narrabri, Dubbo, Temora, Griffith, Ivanhoe and Cobar (Figure 3). Though the species probably occurred in suitable habitat throughout this area, much has now been cleared for agriculture so most local populations are restricted to larger areas of remnant vegetation. These include Yathong, Nombinnie and Round Hill Nature Reserves (total area of 237000 ha and approximately 30 pairs), Goonoo State Conservation Area (until recently Goonoo State Forest) which is over 63000 ha which in 2000 was estimated to have approximately 20 birds (New South Wales Department of Primary Industry 2003),

23

# #% % # #% # # # # # $# # # # # % % # # # # # # # %%# % %%#%%$%$# ## %# $%$#%% # $%%$ %# $% ## # # # # # ## # $###%# # # # %% # # # ## % ### # # #%# $ ## %# % # # # % # $ $%$#%# %% # % $$$## # $%$%# # $# $%$ # # %% # $ # #%%% ## # # # %# % % # # $% ##%# # % # $# %# # % ## # # % ## # %# # # % # % %$ %# # %## % # $%$% ## %# ## $% %$ $ # $#% %#%# % # $%$ # # # # # # $ %#$#%%%% $ %$%%$ #% % $ # # % # %$ # $%%$#%$ # $% %$%#%$%% $$ # % $ $%

%# Malleefowl (Atlas)

# Pre 1990 % Since 1990 $ Since 2000 % Towns

% %% $

% % % % %%%%$%$% %$$%$%%%$ % $% % $%% % $ % % $ % $ $%%% %% % $$%$ $%$% % $$$%$ %%% $%%% % $% % %% % % % %$ % % % % $% % $%$$% $%% % $ $%$ %% $ $%%%% $ $ %$%%$%% % $ %$ $%% $% %$%%$$%% $$ % $ $% % Malleefowl (Atlas)

# Pre 1990 % Since 1990 $ Since 2000 % Towns

$

% $ % $$$$$ % $ $ $ $$$$$ $$ $$$$ $ $ $ % $ $ $ $$ $ % $ $ $$ $ $$ % $ $$ $$$$ $$ $ $$%

% Malleefowl (Atlas)

# Pre 1990 % Since 1990 $ Since 2000 % Towns

Figure 1. Malleefowl records for New South Wales for three time periods (pre 1990, 1900- 2000 and post 2000). Source Department of Environment and Climate Change (2007).

24 and the Pilliga State Forest/State Conservation Area which possibly has about five pairs. Until about 20-30 years ago, small Malleefowl populations still persisted in many of the smaller Nature Reserves throughout this region though they are now extinct in most (Big Bush, Pulletop, The Charcoal Tank and Gubbata) or have tiny populations that are probably not sustainable in the long term (Loughnan and Buddigower). Populations also occur in small State Forests, such as Jimberoo and Kolkibertoo, and crown land at Tallimba, but like other populations in small remnants, the long term future of these populations would appear to be bleak. It is predicted to be extinct by 2008 at Yalgorgrin where much research was undertaken in between 1986 and 1999 with the population at the onset of research estimated to be 32 birds (Priddel & Wheeler 2003). Birds may still persist on freehold and leasehold land within this area, but the numbers are not known, and past clearing and mallee harvesting may mean that these birds are isolated and so may not be viable in the future. In the more arid environments around Cobar, the species is known to persist at low densities, but little survey has been undertaken in this area.

# #

# # #

%% %$ $ %% # # % # %%%%% %% # %% # # # % % %$%

% #

# # # # % # %# $ % % % $ % # $ # %$ % %$ $ $ # $% % $ $% $ # %%% %%% % %# % $ % % %$ % Malleefowl (Atlas) %$% ## # # Pre 1990 $ $ Since 1990 % % % % # $ %%%%$%$% $ Since 2000 $ # %%%$%% %%% $$$ # %$% $$ DECC (NPWS) Reserves % $ % Other Land Managed for Conservation %%% State Forests % Towns Lower Murray Darling CMA

Figure 2. Malleefowl records and reserves within the Lower Murray Darling Catchment Management Authority area. Source Department of Environment and Climate Change (2007).

Management Activities

Fox baiting has been undertaken on Department of Environment and Climate Change reserves since the early 1980s, with more detailed data collection (such as baiting effort and take rates) being undertaken since the mid 1990s. Much of the recent activity has been undertaken through the Fox Threat Abatement Plan (National Parks and Wildlife Service 2001). The Malleefowl was one of 30 species to have targeted fox control to try and determine whether baiting is having an impact on numbers. Six

25

# # %%% % # # # # #

# # $ # # #

# # # % % # # # # # # $%%# % %% %%$# % %$%$#$%%%% # # # $%# % $ # %# # $% # # # # # # # # # # # %# # # $ # # % % # # % # # ## # # $ # % # %# % # # # % %$%# $ #$#%# $ $%#$# # # % #$ $%$ # $ # # # # # # # # # #$%# # # # % # % # # # # # % Malleefowl (Atlas) # %# ## %# # %# ## ### $%% # # Pre 1990 # # # %# %# # # # # # # % Since 1990 $ $ Since 2000 % # # DECC (NPWS) Reserves # State Forests # % Towns CMA Boundary

Figure 3. Malleefowl records and reserves within the Central West Slopes area. Source Department of Environment and Climate Change (2007) priority sites were identified for baiting: Goonoo State Conservation Area Yathong Nature Reserve Nombinnie and Round Hill Nature Reserves Mallee Cliffs National Park Abbots Tank leasehold land (―Sunnyside‖) Tarawi Nature Reserve Baiting has been undertaken at all these sites, except on ―Sunnyside‖ which was suspended in 2005 at the landholder‘s request. Most baiting is usually ground-based but frequent and extensive aerial baiting has been undertaken in the past at Yathong Nature Reserve. Within the Department of Environment and Climate Change reserves in the south west (Mallee Cliffs National Park and Tarawi Nature Reserve) between 200 and 300 baits are laid at set bait stations four times a year. The bait stations are monitored for at least the next two days and additional days if bait take rates are high. Additional monitoring of fox numbers is done by the use of spotlighting transects, and more recently at Tarawi Nature Reserve through the use of sandpads. A small number of Malleefowl Monitoring grids were established on Mallee Cliffs National Park and ―Sunnyside‖ in 1999. The latter grids have not been monitored since 2004, while the grids on Mallee Cliffs National Park are monitored as part of the aerial survey undertaken annually by Department of Environment and Climate Change. Ray Dayman, Ranger, National Parks and Wildlife Service, Lower Darling Area, has made these data available to the National Malleefowl Monitoring database. None of these grids have been re-surveyed to locate any new mounds. The majority of monitoring done in New South Wales is through aerial surveys. There are two methods that are utilised. The first, ―mound to mound‖, involves flying in a helicopter to a set number mounds that have been marked with numbered white aviation runway marker cones

26 that aids detection from the air. As each mound is visited it is recorded as either active (i.e. nest mounded up, litter trails leading to mound, extensive soil and litter disturbance, birds seen actively digging), leaf litter present (where that has been opened up and contains leaf litter but where nesting activity appears to have not taken place) or inactive (where no activity has taken place). This has been undertaken on Mallee Cliffs National Park and Tarawi Nature Reserve with about 150 mounds being surveyed at the former location since 1989. This method is significantly cheaper than surveying randomised aerial transects as much less flying time is required, but assumes that activity of a fixed set of mounds will be correlated with the number of active mounds across the entire site (i.e. recruitment into the population will be reflected by a greater number of the existing mounds being active). The other method of aerial surveys, ―aerial transects‖, involves flying 100 randomly selected transects within a 10 kilometre by 10 kilometre (10000 ha) area and recording any mounds on these transects. The transects are 100 metres wide by 10 kilometres long, giving a total area surveyed as 10000 hectares and are flown at 80km/hr at 60 metres altitude. Once a mound is located the helicopter hovers over the site and the mound is classified using the

Mallee Cliffs active mounds fox control 35 y = 4.2619x - 8519 R2 = 0.7865 30

25

20

15

10

5 y = 0.8857x - 1760.3 R2 = 0.0785 0 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

Figure 4. Preliminary analysis of mound to mound survey of active mounds within Mallee Cliffs National Park between 1989 and 2006. same criteria of active, leaf litter present and inactive. This methodology has been used to monitor the Fox Threat Abatement Program baiting sites plus the property ―Wamberra‖ as a control site that has not been baited for the last five years. Some preliminary analysis on data from Mallee Cliffs National Park has shown that the two methods give a similar estimate of the breeding population present on the park (about 45 pairs in 2006). Figure 4 shows the results from the mound to mound survey for Mallee Cliffs National Park. The first four years of data have been excluded as it is uncertain whether all mounds were counted. The first line of best fit (between 1993 and 1998) shows a slow increase in the breeding numbers in this period. Intensive fox baiting was commenced in 1995 and the data was analysed between 1999 and 2006 (allowing a four year lag to reflect recruitment after baiting started) showing a significant increase in the number of active mounds present. A captive population of up to sixteen pairs has been maintained at Western Plains Zoo, Dubbo since 1988. The birds are sourced from the central west slopes population with most from Yalgogrin and a few from Loughnan Nature Reserve and Goonoo State Forest. Over 500 sub-adult birds have been released, mainly into Yathong Nature Reserve but more recently also Nombinnie and Round Hills Nature Reserves. There has been some limited monitoring, though this has been more intensive in recent years including the use of radio

27 transmitters. Despite the intensive on-ground management of these reserves, the population of Malleefowl appears to be only achieving minor increases in size. Because of concerns that the programme may not be assisting the recovery of the populations at Yathong, Nombinnie and Round Hills Nature Reserves, the breeding and releasing of birds is currently suspended, but the breeding population has been maintained in case future releases are allowed and also for educational purposes. The results of genetic analyses may mean that releases may not be restricted to the central west source population but may be used to supplement populations outside this area. Native vegetation clearing is now managed in New South Wales through the Native Vegetation Act and the development of Property Vegetation Plans using the Native Vegetation Assessment Tools (formerly the Property Vegetation Plan Developer). Since the introduction of this legislation there has been a significant decrease in the amount of Malleefowl habitat sought and approved for clearing. It also has meant that approximately 51000 hectares of Malleefowl habitat within Lower Murray Darling Catchment Management Authority is now being managed for conservation (including stock grazing removed, feral goat management and some fox control amongst others) and another 15000ha under negotiation. This is in addition to the 87000ha that were reserved as part of the Southern Mallee Vegetation process, though these reserves were offsets for larger areas of clearing than has been approved using Property Vegetation Plans. Management of introduced herbivores, particularly goats, also has impacts on Malleefowl habitat quality. Studies are also being undertaken on Tarawi Nature Reserve, Mallee Cliffs and Mungo National Parks looking at total grazing pressure (including goats, rabbits and kangaroos), which may provide information that will improve habitat management for Malleefowl. Fire management is also important for the ongoing preservation of Malleefowl habitat. Within Department of Environment and Climate Change reserves the main fire management method is the creation of fire breaks using strip burning which aims to prevent large wildfires burning an entire reserve. This should also allow the creation of a mosaic of different fire ages within the reserves, which will benefit the various threatened bird species that occur in the same habitat that each use different post-fire age vegetation (Clarke 2005). Fire management on leasehold land is even more difficult, with the main goal being suppression as quickly as possible to prevent large areas being burnt. Fire Management was excluded from Southern Mallee Reserve management actions, but is included within the actions required for a Property Vegetation Plan reserve. It is hoped that the Mallee Fire and Biodiversity Project being undertaken by La Trobe and Deakin Universities will provide further information on post-fire habitat requirements and hence appropriate fire management for threatened (and other) species across all tenures.

Future Actions There are three additional activities that are currently underway that have management implications for the Malleefowl in New South Wales. These are: The development of Conservation of Regional Environments/Biodiversity Forecasting Tool by Department of Environment and Climate Change in conjunction with Lower Murray Darling Catchment Management Authority The Priorities Action Statement for threatened species conservation The capacity for Catchment Management Authorities to develop programs with landholders and volunteers to assist Department of Environment and Climate Change in threatened species management

The Conservation of Regional Environments/Biodiversity Forecasting Tool is a Geographic Information System based analysis tool that has been developed by Department of Environment and Climate Change for use in biodiversity management at a regional scale. It uses attributes such as preferred vegetation community and its condition, patch connectivity and fragmentation and entity behavioural attributes such as foraging and dispersal distance, home range and minimum viable habitat area (dispersal rates and reproductive rates generate Minimum Viable Habitat) to generate values such as occupancy and persistence of fauna entities. The Lower Murray Darling Catchment Management Authority (and the previous Southern Mallee Vegetation Committee) has provided funding and resources for the project and so currently the south west corner of New South Wales is one of the areas of

28 focus of the spatial analysis. Currently 21 fauna entities (individual or groups of species sharing common characteristics) have been modelled for the Lower Murray Darling Catchment Management Authority area, including the Malleefowl (which has been modelled as an individual species). Figure 5 shows one of the outputs (predicted occupancy based on current landuse) for the Malleefowl. Predicted occupancy is given as the percentage of likely occurrence (the higher the number the higher the likelyhood the species occurs in that grid cell). Comparing Figure 5 with Figure 3 shows that the biggest factor driving the likelihood of occupancy is the condition of the vegetation that is calculated in the model from grazing pressure by stock. When grazing is removed, the suitability of the habitat increases. The lack of records in Mungo National Park and the ―Petro‖ Southern Mallee Reserve to the north (Figure 3) may represent lack of survey effort or may indicate a difference in habitat that has not been included in the model. Targeted survey in this and other identified areas would be a useful task to help refine the model (and gain more information on the species‘ actual distribution). An important factor to remember is that the model is a landscape feature model and hence does not include environmental variables such as rainfall or fire. There is scope to include these inputs and more detailed information gained from research and survey into the models to try and increase the accuracy of the predictions. The New South Wales Priorities Action Statement was drafted in 2006 to identify actions that can be undertaken to achieve recovery of threatened species and can be done collaboratively with various stakeholders (including government agencies, Catchment Management Authorities, Non-Government Organisationss, (research organisations and volunteers) (Department of Environment and Conservation 2006). For over 800 threatened species, populations and ecological communities a number of actions (currently over 14000) have been prepared that can be used in programs that aid in their recovery and can also be used as a benchmark to measure how successfully recovery programs are progressing. The Priorities Action Statement has replaced the requirement for a Recovery Plan for this species in New South Wales, which was at draft stage in 2002 (National Parks and Wildlife Service 2002). Table 1 lists the actions that have been identified for the Malleefowl in the Priorities Action Statement. Many of these actions have been conducted by Department of Environment and Climate Change and other parties for a number of years. It is anticipated that each Catchment Management Authority will have a Threatened Species Implementation Plan developed for species which are considered as high priority within their area (high numbers of records or locally endemic) that will allow the identification of recovery actions and projects for future funding.

29

Figure 5. Predicted current occupancy of Malleefowl generated using CORE/BFT.

The Catchment Management Authorities (the equivalent of other states Natural Resource Management Boards) are a relatively new creation within New South Wales. They are important as they provide a strong link between government and land managers and also are a distribution point for federal government funds (such as through Natural Heritage Trust). It is hoped that strong links between Department of Environment and Climate Change and Catchment Management Authorities can be maintained and through such processes as the Priorities Action Statement that a coordinated approach to Malleefowl management with private land managers can be achieved. This may be through actions such as incentive- based agreements to remove domestic stock and feral goats, more intensive fox control programs, further research into the impacts of grazing on Malleefowl habitat and through establishing monitoring programs. Potential programs for the latter that are in preliminary

30 Table 1. Actions listed in the Priorities Action Statement for Malleefowl.

No. Strategy Priority Action Recovery Team and Represent Department of Environment and Climate Change on 1 High Plan Coordination National Malleefowl Recovery Team Implement fox control programs in conservation reserves and Habitat encourage and support (through advice, joint programs, resources, 2 Management: Feral High etc.) the control of feral cats and foxes in other areas of suitable Control habitat. Habitat Use management agreements and incentives on private and Rehabilitation/ leasehold land to manage total grazing pressure through such 3 High Restoration and/or actions as removal of artificial water points and feral and native Regeneration herbivore control. 4 Monitoring High Undertake annual aerial monitoring of established survey sites. Using current records and literature on habitat requirements 5 Monitoring High determine the current distribution and identify areas of potential habitat to undertake survey and management actions. If considered appropriate, establish ground-based Malleefowl 6 Monitoring High monitoring grids within suitable habitat outside Department of Environment and Climate Change reserves. Captive Husbandry 7 or ex-situ collection/ Medium Maintain a captive Malleefowl population at Western Plains Zoo. propagation Produce and distribute information and provide technical support Community and on the species to assist land managers and field-based support landholder liaison/ (such as LandCare and Catchment Management Authorities) to 8 Medium awareness and/or implement actions that assist in the species' recovery (including education grazing strategies, vegetation rehabilitation and general information on the species). Determine, through research, the ecological requirements of the Habitat 9 Medium species, particularly in regard to fire, and incorporate the Management: Fire information into Regional and Local Fire Plans. Survey/Mapping In reserves at the edge of the species' range, undertake survey to 10 and Habitat Medium establish size and ongoing viability of populations. Assessment

discussion are establishing a formal grid within Goonoo State Conservation Areas, and formalising a monitoring program on some of the leasehold land in south west New South Wales.

Acknowledgements Damon Oliver, Ray Dayman, Peter Christie, Paul Mahon, Andrew Deane, Robert Wheeler, Else Foster, John McLaughlin, Marc Irvin (all Department of Environment and Climate Change) and Noel Hayward (Lower Murray Darling Catchment Management Authority) all provided comments on the original presentation. Damon Oliver and David Parker (Department of Environment and Climate Change) provided further comments on this document.

References Benshemesh, J. (2000) National Recovery Plan for Malleefowl. Report for the National Malleefowl Recovery Team. National Parks and Wildlife SA, Adelaide and Department of Environment and Heritage, Canberra.

31 Brickhill, J. (1987) The conservation status of Malleefowl in New South Wales. M.Sc. Thesis. University of New England, Armidale. Clarke, R. (2005) Ecological requirements of birds specialising in mallee habitats: modelling the habitat suitability for threatened mallee birds. Draft report, Department of Zoology, La Trobe University, Victoria. Department of Environment and Conservation (2006) Introducing the Threatened Species Priorities Action Statement (PAS). Report accompanying exhibition of the draft PAS. Biodiversity Conservation, Department of Environment and Conservation, New South Wales, Sydney. Department of Environment and Climate Change (2007) The Atlas of NSW Wildlife. Data extracted August 2007. National Parks and Wildlife Service (2001) Threat Abatement Plan for Predation by the Red Fox (Vulpes vulpes). New South Wales National Parks and Wildlife Service, Hurstville. National Parks and Wildlife Service (2002) Malleefowl (Leipoa ocellata) Draft Recovery Plan. New South Wales National Parks and Wildlife Service, Dubbo. New South Wales Department of Primary Industry (2003) The Malleefowl of Goonoo State Forest. http://www.forest.nsw.gov.au/malleefowl/default.asp Accessed 1/9/06 Priddel, D. & Wheeler, R. (1995) The biology and management of Malleefowl (Leipoa ocellata) in New South Wales. New South Wales National Parks and Wildlife Service, Sydney. Priddel, D. & Wheeler, R. (2003) Nesting activity and demography of an isolated population of Malleefowl (Leipoa ocellata). Wildlife Research, 30:451-464.

32

5. VICTORIAN STATE ROUND-UP – ACHIEVEMENTS SINCE MILDURA 2004

Peter Stokie Victorian Malleefowl Recovery Group

Introduction

The theme of my presentation is Victorian groups working together for Malleefowl. In Victoria we place significant emphasis on collecting data by monitoring Malleefowl breeding density and using the analysed data to develop land management strategies.

Monitoring & Land Management is conducted by a range of groups comprising:

• Volunteer organisations • Government organisations • Catchment (Natural Resource Management) authorities • Non Government organisations

Malleefowl monitoring sites in Victoria

The Victorian Malleefowl Recovery Group is responsible for collecting all data from monitoring sites in Victoria. Malleefowl are found almost exclusively in the National Parks of northwest and central Victoria and smaller Fauna and Flora Reserves close to these National Parks. The Victorian map (Figure 1) shows the complex of national parks and reserves of Murray Sunset National Park, Hattah Kulkyne National Park, Wyperfeld National Park, Little Desert National Park close to the South Australian/Victorian border, and a small isolated remnant of habitat at Wychitella Fauna and Flora Reserve near Bendigo. The second map (Figure 2) indicates the approximate area of the monitoring sites within the National Parks and Reserves as at 2004.

Figure 1. Victorian National Parks and Reserves

33

Figure 2. The location of Victorian Malleefowl Recovery Group monitoring sites in 2004 are shown by red spots

Victorian Malleefowl Recovery Group

History

The Victorian Malleefowl Recovery Group has as its core purpose the monitoring of Malleefowl sites, but the role of the group has developed and changed significantly since Mildura 2004. A brief outline of these changes is important to explain the development of Malleefowl conservation in the past three years since Mildura. The Victorian Malleefowl Recovery Group is a totally volunteer group of approximately 90 members drawn from rural, farming and city communities. The common interest of the members is the conservation of Malleefowl, and the common activity is the monitoring of Malleefowl sites in Victoria. The common shared values of the group are dedication to accurate collection of data and a firm commitment to volunteerism.

The history of monitoring Malleefowl has changed considerably over time:

• 1960s The breeding densities of Malleefowl were measured at several sites by communities who were motivated to reserve these habitats, eg: • – Wandown in 1969 by the Mid Murray Field Naturalists – Wathe in 1963 by local farmer Angus Torpey – Wychitella in 1969 by Bob Johnson and other interested local residents – Several sites by Government authorities in 1960s (data lost) • 1980s 12 sites were set up for monitoring as part of PhD studies • 1990s Government expanded the number of sites and the community started to assist with monitoring • 2000+ Community take over all aspects of monitoring of about 30 sites in Victoria (about one thousand nest sites)

The Victorian Malleefowl Recovery Group was the community group that formed in late 1990s with monitoring as its core business with the following operation:

34

• Focus on rigorous and efficient operations • Funded through Parks Victoria community grants and occasionally other sources • Data provided to government agencies • Victorian Malleefowl Recovery Group make a major in kind contribution to conservation and saves the government time and money

Monitoring 2004 – 2007

In 2004 the Victorian Malleefowl Recovery Group was monitoring 25 existing sites, including a constant 22 sites that form the basis of data analysis for trends in Malleefowl breeding numbers. As the number of members grew, it became possible to expand the number of monitoring sites.

By 2007 nine additional sites were added to the monitoring list of the Victorian Malleefowl Recovery Group, increasing the total number of sites monitored from 25 to 34. The additional sites were set up in unrepresented areas in:

» Little Desert » Western Big Desert » Wychitella

Searching Sites 2004 – 2007

The National Recovery Plan for Malleefowl clearly indicates the need to maintain the scientific integrity of monitoring data by re-searching sites every five years. The Mildura Forum recommended this as a priority action. The Victorian Malleefowl Recovery Group had until this point in time concentrated upon collecting data, but realised that if re-searching were to occur then it was the organization in Victoria to make it happen.

Sites in Victoria were last searched in 1997/8 when Department contractors were undertaking monitoring. The Victorian Malleefowl Recovery Group was faced with the daunting task of needing to search 25 sites in quick time and creative thinking was required to manage this task to find resources outside its own members to achieve this task.

In the three-year period 2004 – 2007 16 existing sites searched and 9 new sites established. The following groups were used:

– Greencorps – Technical and Further Education Natural Resource & Management students – Parks Victoria Staff – Community Groups • Hopetoun Bikeriders • Victorian Malleefowl Recovery Group members • Wedderburn Conservation Management Network • Hopetoun Kindergarten parents • State Emergency Service Nhill Unit

The principles applied to determine the location of new sites were:

• Under represented areas, eg – Western Big Desert – Little Desert • Community interest • Conservation interest - Wychitella • Specific purpose to test impact of cropping on Malleefowl breeding • Site establishment where data existed, but no monitoring was occurring

Training 2004 – 2007

35

The Victorian Malleefowl Recovery Group places great emphasis on training volunteers for two purposes. It is important that volunteers are well prepared to collect scientific data, and secondly volunteers have a sound understanding of safety requirements whilst undertaking monitoring tasks.

• Victorian Malleefowl Recovery Group Monitoring Training has occurred annually, with the expectation that all those who intend to monitor will have attended training • Search Teams needed special training, education packages and on ground training, as well as the development of Palm Pilot sequences to record details of the searches

Education 2004 – 2007

The role of the Victorian Malleefowl Recovery Group has expanded to include the promotion of Malleefowl conservation as the success of the group‘s monitoring program has become more widely recognized within the community. The group has initiated numerous activities to meet its educational objectives since the Mildura Forum, including providing speakers, working with schools and developing publications.

• Speakers – Birds Australia – Bird Observers Club of Australia – Landcare – Community Groups • Schools – Education Package • Publications • Newspapers • Journals • Newsletters • Website • Track and Scat Manual

The most significant publications have been the website, www.malleefowlvictoria.org.au and the Track and Scat Manual.

Political Lobbying

The Victorian Malleefowl Recovery Group has been surprised at the amount of land clearing that has been proposed in habitat suitable for Mallefowl in the past three years. The group never expected that it would need to take political positions to defend Malleefowl from this encroachment, but several key campaigns have been conducted in defence of Malleefowl habitat. As a consequence the Victorian Malleefowl Recovery Group has become well versed in the Environmental Protection and Biodiversity Conservation Act, Controlled Actions and Environmental Effects Statement Panel hearings. • The Victorian Government proposed to place a Toxic Waste Dump at Nowingi in Malleefowl habitat and in the centre of an important corridor. The hearings were held over a record 57 days and the proposal was finally rejected • A sandmining proposal will also impact on Malleefowl habitat, and preparatory details are being gathered by the Victorian Malleefowl Recovery Group to oppose this development • Horticultural clearing of thousands of hectares for olive and almond plantations have impacted on existing Malleefowl habitat especially in and near the best breeding sites, and the Victorian Malleefowl Recovery Group are pursuing offsets for this development • State Fire Plans have recently been opened up for public scrutiny and submission, and the Victorian Malleefowl Recovery Group have responded to the plans with the

36 aim of ensuring that planned actions are beneficial rather than detrimental to Malleefowl • The Victorian Biodiversity White Paper recently released for public submissions provided the opportunity for the Victorian Malleefowl Recovery Group to argue for changes to land use policy and land management to improve the prospects for the long term conservation of Malleefowl

Malleefowl Monitoring Reports 2004 – 2007

Annual reports on Malleefowl Monitoring have been produced for each year 2004 – 2007. The detailed reports contain information on breeding trends, data analysis, fox presence analysis, data collection and individual site trends.

The following general observations for the Monitoring Reports 2004–2007 are worth noting:

• 2002/3 – Cybertracker introduced • 2004/5 – Lowest result on record since 1996 (except 2002/03 drought year) • 2005/6 – Best results since 1996 • 2006/7 – Another drought year, but better breeding mounds – (85 active mounds compared to only eight active mounds in 2002/3) • 2006/7 - Cybertracker used for all mounds

The following chart, Table 1, lists the number of mounds monitored each year, the additional mounds added to the monitoring list each year, the number of active mounds identified through the monitoring, and the number of active mounds in the original 22 set sites on which the annual reports are based (Figure 3).

Table 1. Data collected from annual reports 2004 - 2007

Year Number of Increase Number of Number of active mounds mounds in mounds active mounds in set sites

2002/03 850 8 active 8 active in 22 sites

2003/04 945 + 95 107 active 83 active in 22 sites

2004/05 960 + 15 93 active 68 active in 22 sites

2005/06 997 + 37 123 active 93 active in 22 sites

2006/07 1043 +46 90 active 85 active in 22 sites

Each report contains detailed information, too lengthy to be presented in this paper, so a summary of the 2005 Monitoring Report is included as a typical example of what each report contains.

The 2005 Monitoring Report in summary

• Highest record of active mounds ever recorded in Victoria • Greencorps searched 6 sites – 28 newly described mounds – 2 active • Trend details (Figure 3) – Positive trend, but still below mid 1990‘s – Higher in north west, but still two thirds of 1996/7 – Eastern Big Desert below numbers prior to 2002 drought – North east unusually high numbers

37

Trends in breeding numbers 1994 - 2007

60

50

40

30

20

Mounds Active No. 10

0

F

1993/4 1994/5 1995/6 1996/7 1997/8 1998/9 1999/0 2000/1 2001/2 2002/3 2003/4 2004/5 2005/6 2006/7

Figure 3. Trends in Malleefowl breeding numbers at 22 set sites over the past 11-14 years. The graph represents three distinct monitoring areas Eastern Big Desert comprise 6 sites over 14 years (triangles) North East comprise 4 sites over 13 years (shaded squares) North West comprises 12 sites over 11 years (solid circles. (Note:1994/5, 2002/3, and 2006/7 were major drought years).

Land Management

Land Management is undertaken by four organisations and a summary of their responsibilities and current actions are presented in this paper.

Wedderburn Conservation Management Network Parks Victoria Department of Conservation and Environment Catchment Management Authorities

Wedderburn Conservation Management Network

• Established September 2003 • A network of remnant vegetation patches, their managers and other interested people • A partnership and cooperative arrangement between public and private land managers • Aims to improve the conservation management of the Wedderburn-Wychitella area, especially the Wychitella Fauna and Flora Reserve • The Malleefowl is a Flagship Species for the group

The group was in the formative stages when the Mildura Forum was held, and a list of their achievements since then is remarkable.

• Erosion control of creeks in the Flora and Fauna Reserve • Revegetation with Malleefowl food plants of 100 hectares of cleared land adjacent to Malleefowl habitat • Weed control of Bridal Creeper and Wheel Cactus

38 • Monitoring of grazing in bush land, especially kangaroos • Fox control through an extensive baiting program • Line searches for malleefowl mounds • Fencing of 70 hectares for stock exclusion especially in corridors • Working with the Wedderburn Secondary College to assist students in developing a greater understanding of their local environment

The aim of all of these works is to release captive reared Malleefowl into the Reserve.

The Bush Heritage Trust has purchased 800 hectares adjacent to Skinners Flat in the Wychitella Reserve, and is conducting rabbit control on its land as well as on the Reserve.

Parks Victoria

• Responsible Public Land Managers • Pest and Plant Management – Fox – Kangaroo and Goat • Partnership and funding body for Victorian Malleefowl Recovery Group monitoring • Individual membership of Victorian Malleefowl Recovery Group

Department of Sustainability and Environment

• Overall Land Managers • Responsible Private Land Managers • District Fire Protection Plans • Fire operations to minimize risk of landscape scale fires • Support for Wedderburn Conservation Management Network

Catchment Management Authorities

• Mallee Catchment Management Authority – Greencorps Site Searches – Fencing of remnant reserves – Publications promoting threatened species – Project Manager for Natural Heritage Trust Malleefowl Project

• Wimmera Catchment Management Authority • North Central Catchment Management Authority – Funding for fencing of remnant reserve – Funding for fencing and erosion control contractors

Non-Government Organisations

During the past three years a number of Non-Government organisations have worked with Volunteer Groups and Government organisations in initiatives to assist with Malleefowl conservation. WWF, The Wilderness Society, Victorian National Park Association, Bush Heritage Australia and Trust for Nature have given invaluable support to:

• Promote threatened species initiatives • Advise the Victorian Malleefowl Recovery Group • Assist with publications • Provide support for Victorian Malleefowl Recovery Group initiatives • Arrange and support grants approvals • Campaign to protect Malleefowl Habitat from clearing

39

National Malleefowl Recovery Team

The structure of the representation on the National Recovery team has been unique to the team as it is based on collaboration between each group involved in Malleefowl conservation. The four allotted representatives are drawn from:

Government organisations – Parks Victoria Catchment Management - Mallee Catchment Management . Authority Non Government Organisations - WWF Volunteer groups - Victorian Malleefowl Recovery Group

The next three years

• Little Desert will be a priority to establish monitoring sites • Adaptive Management strategies will be implemented • Education will be an increasing priority of the Victorian Malleefowl Recovery Group • Support for the National system and monitoring will continue • Interaction with the National Database will be a priority

Concluding Remarks

Positives

• The success of the long term scientific study of Malleefowl • The recognition of the value of volunteers and the contribution they make to the study of Malleefowl • Public lands are well managed in Victoria

Areas for improvement

• There are obvious gaps in knowledge in public lands • Issues for managing Malleefowl on private lands have not been addressed • Co-operation between stakeholders can be strengthened

The cooperative arrangement for conducting Malleefowl monitoring and research in Victoria is outstanding. Malleefowl and Victoria are indeed fortunate to have a situation where scientists, volunteers and Land Managers (especially Parks Victoria) and Non-government Organisations‘s such as WWF, The Wilderness Society and Victorian National Parks Association operate together on public lands in mutual support.

Long may this continue and be strengthened

40

6. MALLEEFOWL CONSERVATION IN SOUTH AUSTRALIA

Achievements from 2004 to 2007

Sharon Gillam Department for Environment and Heritage, GPO Box 1047, Adelaide SA 5001.

Abstract

In South Australia, Malleefowl continue to persist in isolated areas of remnant mallee, in both public reserves and on private land; however, the long-term conservation requirements for Malleefowl have been unclear. In 1989, a Malleefowl survey and monitoring program commenced based on monitoring the activity of Malleefowl mounds located within permanently marked grids, to assess persistence and changes in breeding activity over time.

During the period 2004 – 2007, considerable progress has been made in Malleefowl conservation efforts in South Australia. Significant changes, beginning in 2004, were made to the monitoring system. This included the adoption of an electronic monitoring method developed by Benshemesh (1997), and collecting, validating and converting all historic Malleefowl data to electronic format. Other initiatives include the funding of contract project officer positions in the Murraylands, South East and Adelaide to support the Malleefowl recovery effort; fencing of priority Malleefowl habitat; monitoring workshops for volunteers; and a research project into Malleefowl genetics involving the South Australian Museum. In addition, community groups, individuals and agency staff have become actively involved and shown great enthusiasm in furthering conservation efforts for Malleefowl.

Introduction

Malleefowl are considered threatened across all remaining areas of their range in South Australia , and are listed as Vulnerable. Records collected through biological surveys, landowner questionnaires and opportune sightings show that Malleefowl occur in four relatively discrete areas in South Australia, which include: the Murraylands and Upper South East; the southern portion of Yorke Peninsula; Eyre Peninsula and the Far West; and the Aboriginal Lands in the northwest of South Australia, in both public reserves and on private land. Throughout much of this range, the species persists in isolated areas of remnant mallee or Acacia/mulga shrublands; however, the long-term conservation requirements for Malleefowl have been unclear.

Since 1989, studies were instigated in South Australia to determine accurately the distribution, status and apparent threats to Malleefowl. The aim of the studies was to ascertain how to best direct conservation efforts to benefit Malleefowl within the state. A Malleefowl survey and monitoring program commenced in 1989, using a system established in Victoria by Benshemesh (1989), where Malleefowl mounds were located and mapped within permanently marked grids. Monitoring the activity of mounds provides the most accurate indication of persistence and changes in breeding activity over time, and is now undertaken regularly throughout the current range of habitats occupied by Malleefowl. Apparent declines in breeding activity recorded over the past 15 years may indicate reduced viability of these populations over time, and may also provide a direct link between breeding activity and population trends. Mound monitoring may also provide a tool for measuring the effectiveness of current and future management actions. The establishment and monitoring of grids has been accomplished largely through the contributions and enthusiasm of community groups and individuals.

41

In South Australia, the process of mound monitoring has continued over many years; however, limited funds and coordination have led to sporadic monitoring in some grids; inconsistencies in data collection; lack of data transfer, analysis and subsequent reporting. Despite this, a strong body of data has been collected, and improvements have occurred along the way. GPS units were used to replace the compass, making relocating mounds a lot easier and quicker, and improvements were made in data collection sheets. Beginning in 2004, existing survey methods in South Australia were updated to reflect available GPS and hand-held computer (Palm Pilot) technology, as per the monitoring manual produced by Benshemesh (1997).

This paper outlines the considerable progress that has been made in Malleefowl conservation efforts in South Australia from 2004 to 2007.

Achievements over the past three years: 2004/05 to 2006/07

2004/05

In 2004 the National Malleefowl Recovery Team was reformed, and a National Malleefowl Forum was held in Mildura in Victoria – both of these events played an important role in reigniting community and agency enthusiasm to conserve the Malleefowl.

Thirty-five Malleefowl monitoring grids existed in some form in four regions in South Australia at the beginning of 2004: the Murraylands, South East, Yorke Peninsula and Eyre Peninsula. The majority of these grids had been set up in the Murraylands (Mallee District and Riverland District), once thought to be the stronghold of Malleefowl in South Australia.

The increase in the number of grids monitored (from 10 out of 35 grids in 2003) was motivated by a Monitoring Workshop, conducted by the Victorian Malleefowl Recovery Group in Victoria in October 2004. This workshop was designed to train new volunteers and refresh the skills of regular volunteers for monitoring Malleefowl nests during the approaching nesting season. In addition, the workshop introduced people to the electronic monitoring technique used by the Victorians, which provided for a very organised, sophisticated and efficient system of data collection and storage, whilst maintaining a high standard in data quality, that simplified analysis and reporting. This is the preferred method for monitoring breeding densities of Malleefowl across all states, so that (a). data collection methods are standardised at a national level, (b). records may be readily transferred and stored centrally, and (c). data can be easily combined to provide regional impressions of Malleefowl trends. Having a national standard also provides a more efficient basis for maintaining community support and analysing the resulting data.

In mid 2004, collating and sorting of historic Malleefowl records began in the Murraylands, with the appointment of contractors by the Department for Environment and Heritage, Berri Office, to coordinate this project. This signified a major turning point in renovating the Murraylands‘ monitoring ‗system‘. The 2004/05 season saw the start of the Victorian Mallefowl Recovery Group monitoring method being adopted in South Australia, with plans to extend this method to all grids next season, and establish a coordinated community based monitoring program. Six people from the Riverland (Department for Environment and Heritage staff and others) attended the Victorian Mallefowl Recovery Group training workshop, and equipment was purchased to begin the set up of monitoring ‗kits‘. Monitoring grids on Calperum and Taylorville Stations (located in the Bookmark Biosphere Reserve north of the Murray River) also picked up pace through Community Land Management Group volunteers.

In 2004, four new grids were established on Calperum Station by Community Land Management volunteers, giving a total of 39 grids in South Australia. Of these, 28 grids were monitored in the 2004/2005 breeding season – 26 of these in the Murraylands (Table

42 1). In this season, of the total 573 mounds monitored, 23 were active and 550 were inactive (Table 2).

Table 1. Number of grids monitored in the 2004/05 Malleefowl breeding season in each region in South Australia.

Region Grids Monitored Total Grids Murraylands 26 31 Eyre Peninsula 2 5 Yorke Peninsula 0 1 South East 0 2 Total 28 39

Table 2. Number of active and inactive mounds per region in South Australia, in the 2004/05 Malleefowl breeding season.

Region Inactive Mounds Active Mounds Total Mounds Murraylands 455 15 (3%) 470 Eyre Peninsula 95 8 (7%) 103 Yorke Peninsula 0 0 0 South East 0 0 0 Total 550 23 (4%) 573

An aerial survey method was trialled over the Gawler Ranges region in March 2004, to assess the effectiveness of helicopter surveys for monitoring Malleefowl mounds in areas where mound densities are very low and vegetation relatively dense. The results of the trial indicated, as with similar trials in southwest New South Wales, that this method is useful in locating mounds in large areas of vegetation and could be effective in directing further ground-based search efforts. However, on calibrating the aerial survey technique against ground surveys on an established 2km2 Malleefowl grid at Lock, it was found that 50% of mounds recorded from the ground were missed from the air, and double counting may also have occurred. Further investigation of this method may be warranted where there are large tracts of vegetation with low mound densities. `

2005/06

Three significant events occurred in 2005 to further the recovery effort for Malleefowl: at a national level, funding was secured for a major project aimed at implementing key components of the National Malleefowl Recovery Plan, and at a state level, funding became available for a centrally-based position to aid in Malleefowl conservation initiatives. At a regional level, a part-time Malleefowl Project Coordinator was also appointed in the South East, to drive monitoring activities in that region.

Multi-Regional Project

A two-year, multi-regional project “National Malleefowl Monitoring, Population Assessment & Conservation Action Project” commenced, funded by the Natural Heritage Trust and managed by the Mallee Catchment Management Authority in Mildura. This project was allocated ~$291,000 from 2005/06 to 2006/07, and involved South Australia, Western Australia and Victoria. Main aims of the project were to:

Analyse existing monitoring data to interpret Malleefowl breeding density trends and impacts of land management and environmental variables on Malleefowl conservation Develop a consistent national monitoring system, database and volunteer manual for Malleefowl, with the view to fostering an ongoing and self sufficient national monitoring system for use by volunteer groups

43 Integrate the monitoring system into outcomes for habitat conservation across the landscape in Victoria, South Australia and Western Australia, on both public and private land, involving targeted implementation of on-ground actions from the National Recovery Plan for Malleefowl

In the first year of this project (2006), funding was available to South Australia for fencing to protect significant Malleefowl habitat; for holding two training workshops; and for establishing and monitoring two new grids (or re-establish two old grids). Subsequently, two hundred and sixty hectares of Malleefowl habitat on northern Eyre Peninsula was fenced to exclude stock and hence conserve precious habitat, and expand and improve the connectivity of existing habitat. This was achieved through the hard work of six landholders. Two volunteer training workshops were held, one in Lock on Eyre Peninsula, and the other in Berri, in the Riverland. These were the first workshops of this kind held in South Australia, and followed the model set up by the Victorian Malleefowl Recovery Group, back in 2001. One new grid was established in Gum Lagoon Conservation Park in the South East of the state – the Naen Naen grid - and two grids that had been partly established some years ago, were reinstated into properly functioning grids – the Innes (Yorke Peninsula) and Mt Scott (South East) grids. Statewide

In October 2005 the position of Threatened Fauna Project Officer commenced with the Department for Environment and Heritage based in Adelaide, with a primary focus on Malleefowl.

The main aim of the position was to help facilitate the integrated and coordinated approach to Malleefowl conservation in South Australia, in line with the National Recovery Plan for Malleefowl. Initially, a state-based strategy for Malleefowl data collection, interpretation and dissemination was developed (in line with the Victorian monitoring method). Key players in each region were identified with regards to Malleefowl monitoring, and what each region had achieved in te0rms of managing and monitoring grids, methods of data collection and storage, and volunteer networks and training, was established. All data on Malleefowl monitoring across South Australia was gathered and made available for incorporation into a state and national database. The process of data collection for the 2005 monitoring season was facilitated, and a comprehensive report detailing the status of Malleefowl monitoring in South Australia was compiled. A statewide meeting was convened in April 2006, to review the previous season‘s (2005) Malleefowl monitoring and to set future directions.

A part-time position of Malleefowl Recovery Program Coordinator was created, also in October, in the South East of the state, with a focus on coordinating the monitoring efforts in that region. This included the task of reinstating one partially set-up grid and thoroughly re- searching another.

The Malleefowl Project Coordinators located in the Riverland continued to implement and expand the national monitoring system in the region, with many tasks successfully achieved. Many issues were resolved including reprojecting mound coordinates (from original surveys) into the correct datum (WSG84) and subsequent ground truthing, providing accurate mound locations and maps.

The results of the 2005 monitoring effort are summarised in Tables 3 and 4, showing that all 39 grids in South Australia were monitored, most of them via the new electronic monitoring method – a significant achievement. A total of 819 mounds was monitored across the regions, up from 573 the previous season – 51 of these were active.

Table 3. Number of grids monitored in the 2005/06 Malleefowl breeding season in each region in South Australia.

Region Grids Monitored Total Grids Murraylands 31 31 Eyre Peninsula 5 5

44 Yorke Peninsula 1 1 South East 2 2 Total 39 39

Table 4. Number of active and inactive mounds per region in South Australia, in the 2005/06 Malleefowl breeding season.

Region Inactive Mounds Active Mounds Total Mounds Murraylands 521 18 (3%) 539 Eyre Peninsula 174 18 (9%) 192 Yorke Peninsula 20 6 (23%) 26 South East 53 9 (15%) 62 Total 768 51 (6%) 819

2006/07

Much has been achieved in implementing recovery actions for Malleefowl at a statewide level over the last 12 months. Considerable enthusiasm and effort was shown by all stakeholders, including staff from various agencies, landholders, and a large contingent of volunteers.

Multi-Regional Project

This year (2007), under the initiative of the multi-regional “National Malleefowl Monitoring, Population Assessment & Conservation Action Project”, more than 5000 ha is in the process of being fenced, this time including the Murraylands, South East and Yorke Peninsula, as well as Eyre Peninsula. Great support has once again been shown by various landholders across the regions.

The collection and collation of all monitoring data from around the state – a huge task completed with the help of many – resulted in a review and reports by Benshemesh, including a Trend Analysis Report, which found that the breeding activity of Malleefowl has been declining by 2-3% per annum, across southern Australia (Benshemesh 2006). Monitoring information from South Australia was collected from 23 monitoring sites over an average of 6.1 years of data (range: 1 – 15 years).

While on-ground training will continue through the regional coordinators and ideas for more training workshops are on the drawing board, all those interested are welcome to attend the training weekend held by the Victorian Malleefowl Recovery Group in Wyperfeld National Park (northwest Victoria) in September or October each year. Production of a training video is also on the agenda.

Two new grids (Bandon and Ettrick) were set up on Heritage Agreement properties in the Murraylands, while two grids were re-searched – the Karte grid (Riverland) and Innes grid (Yorke Peninsula). Protocols for establishing and re-searching grids have been drafted, so that these tasks are undertaken according to standards that apply at a national level. These will be included in a National Malleefowl Monitoring Manual, another large project currently underway and coordinated by the Victorian Malleefowl Recovery Group.

Statewide

There are now 43 operational grids in South Australia. This season (2006) all grids were monitored via the standard electronic methods, using a Palm Pilot and GPS. All data were subsequently validated by relevant coordinators in each region and made available for the national and state databases. Whilst all (national) data is presently held by Joe Benshemesh, procedures are underway to develop a national database. Tables 5

45 provides an outline of the total amount of grids per region and those that were monitored in 2006, while Table 6 shows the total of active and inactive mounds per region, also for 2006.

Table 5. Number of grids monitored in the 2006/07 Malleefowl breeding season in each region in South Australia.

Region Grids Monitored Total Grids Murraylands 26 34 Eyre Peninsula 5 5 Yorke Peninsula 1 1 South East 3 3 Total 35 43

Table 6. Number of active and inactive mounds per region in South Australia, in the 2006/07 Malleefowl breeding season.

Region Inactive Mounds Active Mounds Total Mounds Murraylands 662 27 (4%) 689 Eyre Peninsula 179 15 (8%) 194 Yorke Peninsula 35 9 (20%) 44 South East 75 9 (11%) 84 Total 951 60 (6%) 1011

Funding was approved to begin a major research project into Malleefowl genetics. This is a collaborative project involving the University of Melbourne, University of Wisconsin (USA) and South Australian Museum. The research will provide information on how habitat fragmentation and isolation is affecting Malleefowl, and will also provide the first comprehensive Australia-wide picture of the state of genetic variation in Malleefowl populations.

A statewide Malleefowl meeting was held in March, and attended by staff and volunteers from all regions except Eyre Peninsula. This meeting provided an opportunity to review and discuss last season‘s monitoring efforts, look at future directions, and share any ideas and concerns.

Two volunteer workshops (one in Melbourne, one in Adelaide) were organised by the Victorian Malleefowl Recovery Group earlier this year, and were well attended by representatives from South Australia. These workshops were convened to review and refine the monitoring methodologies, as the basis for the production of a National Malleefowl Monitoring Manual.

A Nest Excavation Project was coordinated by Sharon Gillam during February to April, and occurred across three regions. All (34 in total) of the active mounds in eight monitoring grids were excavated towards the end of the breeding season to determine the number of eggs laid and numbers hatched. Whilst the report for this project is in progress, initial findings showed 2006 to be a year of below-average rainfall across all regions, with particularly low winter rainfall, which in turn affected the breeding efforts of Malleefowl. Results showed an overall reduced clutch size compared to known average clutch sizes, with productivity in the South East region particularly low. The final report will be available soon.

Once again, most of the conservation initiatives underway for Malleefowl could not be realised without the much valued input and enthusiasm given by all volunteers. The hours put in this year were outstanding, and a total figure is given for each region in Table 7.

Table 7. Total volunteer hours per region for Malleefowl conservation in 2006/2007. (DEH = Department for Environment & Heritage, CLM = Community Land Management)

46 Region Yorke Pen. Eyre Pen. South East Murraylands Murraylands Total DEH CLM Hours 255 313 163 715 376 1822

Conclusion

Significant headway into Malleefowl conservation has been made in South Australia during the past three years, demonstrating strong will and determination to improve the long-term prospects for this species.

A major focus has been on the implementation of the electronic Malleefowl monitoring system; the collation of data; and putting in place measures to continue the nationally standardised monitoring process. Whilst the monitoring system provides a whole range of valuable information which has been recognised in the results of Benshemesh‘s (2006) report, it has the potential to be explored further in conjunction with research and management, which is recommended under an Adaptive Management framework in the National Recovery Plan for Malleefowl (Benshemesh 2007).

Whilst the conservation initiatives described above all contribute to assisting in the long-term survival of Malleefowl in South Australia, it is imperative that agencies continue the commitment to develop and implement actions outlined in the National Recovery Plan for Malleefowl (Benshemesh 2007), and to harness the enthusiasm shown by all stakeholders so that the recovery process may continue to generate the best conservation outcomes.

Acknowledgements

The conservation initiatives over the past three years would not have been accomplished without the time, effort and keen interest of many people, including community volunteers and groups, landholders, Greencorp teams, Conservation Volunteer Australia teams, and agency staff. The continued support of Joe Benshemesh and the Victorian Malleefowl Recovery Group is also acknowledged.

References

Benshemesh, J. (1989) Report of the establishment of sites for the monitoring of Malleefowl populations: Operation Raleigh 1989. Unpublished report. Department of Conservation, Forests and Lands, Melbourne

Benshemesh, J. (1997) Manual for monitoring Malleefowl. Parks Victoria and Department of Conseravation and Natural Resources, Mildura.

Benshemesh, J. (2006) Trend analysis of Malleefowl monitoring – Draft. Milestone 3 Report to the Mallee Catchment Management Authority, Victorian Malleefowl Recovery Group and multi-regional ―National Malleefowl Monitoring, Population Assessment and Conservation Action Project‖ steering committee.

Benshemesh, J. (2007) National Recovery Plan for Malleefowl 2006-2010. Department of Environment and Heritage, Adelaide, South Australia.

47

7. ACHIEVEMENTS IN WESTERN AUSTRALIAN MALLEEFOWL CONSERVATION SINCE THE FORUM AT MILDURA IN 2004

Carl Danzi Western Australian Malleefowl Network Facilitator

Background

Malleefowl inhabit suitable habitat in the southern half of Western Australia, but their range and population trends in most of these areas are largely unknown. For these reasons Malleefowl monitoring in the next decades will be vital in deciding their management.

Figure 1 shows location of Malleefowl records and major Malleefowl groups in WA. The dark orange on the map shows the location of the Western Australian wheatbelt; most of this area has had its natural vegetation cleared for agriculture. Previous to

Figure 1. Location of Malleefowl records and major Malleefowl groups in WA. AWC – Australian Wildlife Conservancy; BHA – Bush Heritage Australia; FONEM – Friends of the North-eastern Malleefowl; MPG – Malleefowl Preservation Group; NCMPG – North Central Malleefowl Preservation Group

48 land clearing this area provided large amounts of suitable habitat for Malleefowl but they are now found here only in and around larger areas of uncleared land. The area to the north and east of the Western Australian wheatbelt is largely uncleared. The area to the south and west is a mix of cleared and uncleared land.

Blue dots on the map represent areas where Malleefowl were or are currently recorded as present. The eastern half of the wheatbelt has the highest concentration of Malleefowl records. North and east of this, records are sparse, probably due to this area being marginal for Malleefowl. South and west of the wheatbelt Malleefowl records are also sparse, and most of these are also old records and Malleefowl are now not recorded in much of this area.

Also shown on Figure 1 are areas where major Malleefowl groups are based, shown as green dots. In the southern part of the map is Ongerup, the home of the most well known and biggest Malleefowl group in Western Australia, the Malleefowl Preservation Group, established in 1992. The Malleefowl Preservation Group has done a lot of work in Malleefowl conservation since 1992. The Malleefowl Preservation Group has been heavily involved in Yongergnow that is based in Ongerup also. Farther north is Merredin where a new group the Merredin Malleefowlers are based. North west of Merredin is the north/eastern wheatbelt where Friends of North Eastern Malleefowl are based in an area where many Malleefowl records exist. North/west again another 200km or so is Wubin where North Central Malleefowl Preservation Group is based. The North Central Malleefowl Preservation Group started not long after Malleefowl Preservation Group in the early to mid nineties and have achieved many good outcomes for Malleefowl since then. Two conservation Non-Government Organisations in Bush Heritage Australia and Australian Wildlife Conservancy have also been involved in Malleefowl conservation in recent years. The Australian Wildlife Conservancy at their Mt. Gibson Sanctuary and Bush Heritage Australia at their Charles Darwin and Eurardy reserves. The Department of Environment and Conservation are also involved in Malleefowl conservation at Shark Bay in Western Australia and are also often involved in surveys and other Malleefowl work with all groups.

Malleefowl Conservation - Informed & integrated community action

In August 2004 WWF-Australia commenced a project to assist in Malleefowl conservation in Western Australia titled ‗Malleefowl Conservation - Informed & integrated community action‘. This project is funded by Natural Heritage Trust and delivered by WWF-Australia through Avon Catchment Council. The project works in four natural Resource Management regions in Western Australian being Avon, South Coast, Northern Agricultural and southern parts of Rangelands, these being the areas where Malleefowl are present. The project is currently scheduled for completion in June 2008.

WWF-Australia have appointed a Western Australian Malleefowl Network Facilitator to manage the project.

Major achievements of the project to date are as follows:

Set up the Western Australia Malleefowl Network. In Western Australian there are many stakeholders involved in Malleefowl conservation. These stakeholders include community groups, individuals, Non-government Organisations, Natural Resource Management bodies, consultants and private industry such as mining companies. The Western Australian Malleefowl Network is a conduit for all stakeholders to be involved in Malleefowl conservation. It is a way of sharing experiences and learning together rather than every stakeholder reinventing the wheel. The Western Australian Malleefowl Network holds meetings quarterly which stakeholders are encouraged to attend.

The project has put together the ‗Western Australian state action plan for Malleefowl‘. This plan was put together by the Western Australian Malleefowl Network members by referring to the National Recovery Plan for Malleefowl and pulling out of it the aspects most relevant to Western Australia.

49

A major part of the facilitators role currently and until the end of the project is to help stakeholders be involved in setting up sites and monitoring them to national standards in line with the ‗National Manual for the Malleefowl Monitoring System‘ which was released at the Katanning National Malleefowl Forum. Also to provide training to the stakeholders to be able to achieve this.

Community is a big part of Malleefowl conservation in Western Australia. As was seen earlier from other forum speakers, community volunteers can save about ninety percent of the funds compared with using paid expertise to achieve on-ground Malleefowl outcomes. Because of the volunteers‘ interest and passion for the cause of Malleefowl conservation, the quality of the work they provide can also be better than paid people. The Western Australian Malleefowl Network always welcomes new volunteers to be part of the programme.

The project has been partly used to fund research into Malleefowl conservation in Western Australia; Blair Parsons research is funded by the project.

Research

There have been two main research projects in Western Australia since the Mildura Forum as follows:

Blair Parsons Research

Blair Parsons has been working on a PhD project at the University of Western Australia in collaboration with the Malleefowl Preservation Group, Wildlife Research and Management Ltd. and CSIRO. Blair has analysed old and recent sightings data from Western Australia. He has set up a link on the Malleefowl Preservation Group website to allow anyone with internet access to provide their current sighting to him quickly, easily and reliably. Many other sources have also been used by Blair for analysing sightings data, such as Western Australian Museum and North Central Malleefowl Preservation Group. Blair has also been doing research on the relationship between fire and Malleefowl, as well as developing models to predict the occurrence of Malleefowl in different areas.

Jessica van der Waag’s Research

Jessica has been working on a project on how to support the recruitment of Malleefowl in small remnants. This project is a PhD project at the School of Animal Biology in the University of Western Australia. The fieldwork for this project has been in Ongerup and the work has been undertaken very closely with Malleefowl Preservation Group on one of their sites.

Education – Malleefowl Magic

Malleefowl Magic is a curriculum based education package that is aimed at primary schools; the package can be altered such that it is suitable for high school also. The program has been running for about 6 years and to date over 130 schools in Western Australia over the entire Malleefowl range area have been visited. Figure 2 shows location of schools visited.

Malleefowl Magic was developed by Malleefowl Preservation Group and introduced by Sir David Attenborough. Susanne Dennings who has been heavily involved in the Malleefowl Preservation Group has delivered the program since its beginning. Susanne has done many Malleefowl Magic tours since the Mildura Forum. The package consists of a small book, a large book, a teacher‘s manual and a CD that includes a Malleefowl song. Funding support for Malleefowl Magic includes the Natural Heritage Trust, Lotterywest and BHP Billiton.

50

Figure 2. Location of schools visited by malleefowl magic education package

Surveys and Monitoring

Figure 3 shows Malleefowl sites targeted to have ongoing yearly Malleefowl monitoring to national standards by 2007/08 in Western Australia shown in green. Possible future sites are shown in red.

Starting with the green dots at the bottom of the map, Ongerup is where the Malleefowl Preservation Group is based and they have five sites in this area. Carl Danzi worked with the Malleefowl Preservation Group to get four of these sites monitored to national standards for the 2007/08 monitoring season and by doing so, the Malleefowl Preservation Group are trained to monitor these sites without assistance next season.

North from Ongerup on Figure 3 is Merredin. A survey by Malleefowl Preservation Group of a site of approximately 307 ha was conducted in Merredin at the Merredin Peak reserve in June 2006. This survey discovered about 31 Malleefowl mounds in the area surveyed of which nearly 10 had Malleefowl activity. Later that year members of the Merredin Malleefowlers group revisited the mounds with Malleefowl activity in June to discover all but one of the mounds were still being worked by Malleefowl. In late October 2007 Carl Danzi revisited and monitored all the mounds with the Merredin Malleefowlers. The mounds have therefore been monitored this year and the Merredin Malleefowlers should be able to complete the monitoring unassisted next year.

Mt. Jackson is a site outside the Western Australian wheatbelt which is being mined for iron ore. The mining company (Portman Ore) as part of their environmental works in the area is doing some Malleefowl work. The Malleefowl Preservation Group has done Malleefowl surveys coordinated by Susanne Dennings in the area each year since 2004. Each year a new area in close proximity has been surveyed. Carl Danzi is working with Bamford Consulting (Portmans Environmental Consultant) to get Mt. Jackson monitored to national

51 standards. Bamford Consulting envisages they will be the environmental consultant for the life of the mine, approximately 20 years.

Figure 3. 2007/08 targeted Malleefowl monitoring sites (green) and possible future sites (red). AWC – Australian Wildlife Conservancy; BHA – Bush Heritage Australia; MPG - Malleefowl Preservation Group; NEWROC – North Eastern Wheatbelt Regional Organisation of Councils; NCMPG – North Central Malleefowl Preservation group.

Wubin is where the North Central Malleefowl Preservation Group is based. They currently have six sites in the area, last year the North Central Malleefowl Preservation Group monitored all six of their sites to national standards and forwarded the data to the National Recovery Team. This is the only area in Western Australia that currently is being monitored to national standards and the group has plans to do this each year.

Yeelirrie in central Western Australia is an area that the Malleefowl Preservation Group again has had heavy involvement in surveying over the last 6 years. Susanne Dennings again has been the coordinator of these surveys. Parts of Yeelirrie are mined by BHP Billiton and they have been involved in the surveys. Last year monitoring data from Yeelirrie was forwarded to the National Recovery Team by George White who has been involved in many Malleefowl Preservation Group surveys; it is hoped this will be ongoing yearly monitoring.

Eurardy is a Bush Heritage Australia property. In August 2007, Paul and Leanne Hales (Eurardy managers) conducted a Malleefowl survey at Eurardy with the guidance of Carl Danzi. This was brought about by a number of Malleefowl sightings on Eurardy earlier in the year. The survey was very successful in many ways including community involvement. Three areas on Eurardy were surveyed in an attempt to find an area that Malleefowl used for breeding. Unfortunately only two old mounds were found in the area of the three with the most Mallefowl activity; this search also uncovered many fresh Malleefowl footprints in the area. The managers are keen to monitor the two mounds found in the area of 80 ha. Hopefully this area will be expanded next year to about 400 ha to bring it in line with national standards and uncover additional mounds in the process.

52

The red dots on the map of Figure 3 are possible future sites, some of these have already had surveys done and other preliminary work.

Eyre Bird Observatory in the east has had about 1200 ha surveyed and has interest but has the tyranny of distance and mandatory relatively expensive accommodation to contend with. There is talk of Eyre being included as a site as soon as 2008/09.

Kalgoorlie, Mullewa and Lake Magenta are sites the facilitator knows little about other than that they have been areas that in the past have been targeted as having Malleefowl sites. There are some old data from Lake Magenta and if there are still Malleefowl there it would be a good site being so close to the Malleefowl Preservation Group and also being such a large area of natural vegetation. All three of these sites are positioned well to help make a complete monitoring program in Western Australia.

Lake Muir is an interesting site in that its vegetation is different from other areas in Western Australia that support Malleefowl; it is mainly Jarrah forest based. According to the Department of Environment and Conservation Manjimup, Malleefowl have been sighted in the area occasionally and because of this they are keen to organise a survey of the area. The Department of Environment and Conservation and Susanne Dennings are working to do a Malleefowl survey around Lake Muir in May 2008. If this survey is successful in finding recent Malleefowl mounds the area can be included as a monitoring site.

Wongan Hills is an area where Malleefowl were found in the past and recent sightings have been made there. The Wongan Ballidu Bushcare group have shown interest in setting up a site in the area and earlier this year in May did a preliminary survey of a site at Wongan Hills. This survey of about 20 ha uncovered a few old mounds. It is hoped they will continue their interest and if a suitable site can be located in the area, it will be a monitoring site.

North-eastern Wheatbelt Regional Organisation of Councils is in the north/eastern wheatbelt of Western Australia and is where Friends of the North-eastern Malleefowl are based. There are many sightings of Malleefowl in the area and there is strong interest in this area for Malleefowl work from Natural Resource Management, Friends of the North-eastern Malleefowl and individuals. This area will be targeted to set up at least one monitoring site in the 2008/09 season.

North Central Malleefowl Preservation Group have been working with the Australian Wildlife Conservancy at Mt. Gibson Sanctuary and Bush Heritage Australia at Charles Darwin Reserve on Malleefowl conservation. These are both in areas outside the Western Australian wheatbelt and reportedly with strong Malleefowl presence, they will both be targeted as areas for future Malleefowl monitoring sites.

The final red dot is at Shark Bay in an area from which foxes have been almost eradicated but cats are still plentiful. Malleefowl were reintroduced to the area about 10 years ago. The North Central Malleefowl Preservation Group surveyed an area of 400 ha at Shark Bay in July 2006 and found only two mounds, one of which was being worked by Malleefowl. The area surveyed was good for finding Malleefowl tracks and many tracks were found. This is an area that may be a possibility for a monitoring site in the future.

Highlights of Western Australian Malleefowl Groups

Malleefowl Preservation Group

Table 1. Summary provided by Malleefowl Preservation Group showing its milestones and achievements since 2004.

2004 Membership reaches 650 National Malleefowl Forum (Mildrua) – by Malleefowl Preservation Group key speaker presented paper and 5 posters

53 WA Premier launches Yeelirrie Report – Albany South West Station Country Survey – (Bullfinch area) planning commenced Eyre Bird Observatory Monitoring -partnership with Birds Australia Malleefowl Matter newsletter – 33rd edition in March Garden Week Ecotopia – 5 day display at Perry Lakes Mt. Jackson Survey (north of Bullfinch) completed “Malleefowl Believers” film documentary sub-committee formed Strategic Planning – three day workshop held to review by Malleefowl Preservation Group future directions Conservation Council Committee Visit Rossmoyne High School Bushrangers –visit to office and malleefowl site Foster Road Botanical Survey – associated habitat report completed (Anne Rick and by Malleefowl Preservation Group volunteers) Gnowijow Primary School Certificate – awarded to Logan Smith, Ongerup Primary School

2005 Committee workshop – implementing the new strategic plan (replacing Susanne‘s roles) State Conservation Conference “Empowering the Conservation Community – March (paper presentation) Tertiary Support – University of Western Australia PhD students, Jessica van der Waag and Blair Parsons (in partnership with CSIRO) project given by Malleefowl Preservation Group backing. Bullfinch Survey – completed for Portman Ore, May 2005 Research Information Evening – by Malleefowl Preservation Group joint projects with CSIRO and University of Western Australia presentations – Ongerup June 7th “BHP Billiton Nickel West Malleefowl On the Nullarbor” – joint project with Birds Australia -30 volunteers participated over two weeks- June 2005 Launch of “Give a Chick a Chance” campaign – tax deduction fund raiser to help PhD research. AGM Camp out Weekend – Centre for Sustainable Living, Denmark Malleefowl Magic – All students from Hyden Primary School participated International Megapode Conference – by Malleefowl Preservation Group provide support for next Western Australian conference - appoint representative to coordinating committee Peniup Survey Site mounds checked – two active this year Gnowijow Primary School Certificate awarded to Ongerup Primary School student, Cathy Elliott International Exposure – Jessica van der Waag‘s paper included in Megapode Newsletter December 2005

2006 State Monitoring Workshop – PALM instruction workshop (Peter Stokie Victorian Malleefowl Recovery Group). Attended by 4 Malleefowl Preservation Group members (January) National Survey Data – Carol Nicholas collating all survey site results for National data base (Joe Benshemesh) Birds Australia WA Report –Partnership Project presentation Malleefowl Magic Education Extension – amongst others, Augusta, Margaret River, Nannup, Nyindamurra, Karridale, Cowaramup, Bridgetown, Pemberton, Denmark, and Walpole Primary schools were visited. BHP Billiton – Nickelwest – Yeelirrie Malleefowl survey Portman Iron Ore Pty Ltd – Mt. Jackson Malleefowl survey Fox baiting – coordination of fox baiting program in the region Malleefowl Magic Education Extension – Augusta, Bridgetown,

54 Cowaramup, Denmark, Karridale, Margaret River, Montessori, Nannup, Northcliffe, Nyindamurra, Pemberton, St Thomas More, Walpole, Cascade, Condingup, Esperance Christian, Esperance, Grass Patch, Jerdacuttup, Jerramungup, Munglinup, Our Lady Star of the Sea, Ravensthorpe, Salmon Gum, Scaddan & Wangatha Aboriginal Schools were visited

2007 Malleefowl Magic Education Extension February/March– Broomehill, Yealering, Tammin, Wyalkatchem, Wongan Hills, Cadoux, Kalannie, Dalwalinu, Wubin, Coorow, Carnamah, Morowa, Perenjori, Buntine, Bencubbin, Mukinbudin Christian Community School, Mukinbudin District High School, Westonia, Quairading, Babikin, Corrigin, Lake Grace & Dumbleyung schools were visited Opening of Yongergnow Australian Malleefowl Centre – attended by 300 participants – relocation of Malleefowl Preservation Group Office Portman Iron Ore Pty Ltd – Malleefowl survey at Mt. Jackson Fox baiting – coordination of fox baiting program in the region, conclusion of 2006 project Malleefowl Magic Education Extension – July – Southern Rangelands schools visited

North Central Malleefowl Preservation Group

The group has had three major highlights since Feb 2004 as follows:

In 2006 North Central Malleefowl Preservation Group monitored their six Malleefowl monitoring sites to national standards. They are the first group to do this in Western Australia and now have the knowledge and skills to continue monitoring their own sites and, if they have time, help other stakeholders especially near them to do the same.

North Central Malleefowl Preservation Group had a successful application in round 8 of the Australian Government Natural Heritage Trust, WWF-Australia Threatened Species Network community grants program for 2005-06 titled Signage and Baiting in the Shire of Dalwallinu for Malleefowl Conservation. The signage consisted of three interpretive display boards erected in Dalwalinu, Wubin and Perenjori. The baiting part of the grant was for fox baiting, and North Central Malleefowl Preservation Group have been very successful in getting a high level of community involvement in the Northern Wheatbelt.

Successful application in Natural Heritage Trust, WWF-Australia Threatened Species Network grants program in 2003-04 titled Recovering the Malleefowl in the Northern Western Australian Wheatbelt. As part of this grant North Central Malleefowl Preservation Group were able to survey two sites and then record the mound information using Palm technology used for national standards. There was an urgent need to review the status and conservation requirements for Malleefowl in the northern wheatbelt where Malleefowl had not yet been properly studied. The group worked with local farmers, Natural Resource Management groups, the CSIRO, Birds Australia (WA) and Department of Environment and Conservation to determine the abundance, distribution, habitat use, movement and management actions needed to recover Malleefowl in this landscape.

Friends of the North-eastern Malleefowl

55 The Friends of North Eastern Malleefowl, is a non-incorporated group of interested individuals, shires and others from Western Australia‘s north-eastern Wheatbelt, that are working together to protect and promote Malleefowl and their conservation in Western Australia, between Wyalkatchem and Westonia, and from Trayning to Tampu.

In 2003, with the support of the ‘Better Bencubbin‘ Progress Association, the Shire of Mt. Marshall and the Threatened Species Network, members erected 12 Malleefowl Crossing signs along main roads that cut though known Malleefowl habitat across the shire of Mt. Marshall.

Since then Friends of the North-eastern Malleefowl have also supported a number of local projects aimed at fencing Acacia shrublands where Malleefowl are known to occur, more than 200Ha has been fenced. As a small group, Friend of the North-eastern Malleefowl has so far focused on awareness raising, collecting sighting data and supporting local projects that benefit Malleefowl. They have developed a sightings database for the North East wheatbelt area.

In the future, Friends of the North-eastern Malleefowl hopes to establish a site in the northeast Wheatbelt, and train locals to monitor a large (about 400 ha) patch of Malleefowl habitat to national standards.

Working with the Western Australian Malleefowl Network, the group hopes to provide useful data on Malleefowl activities from this little studied area & coordinate these activities.

Please contact Mick Davis for More information – 0429 473 567

Private Land Conservancies

As stated previously in this manuscript two private land conservancies have been involved in Malleefowl conservation in Western Australia as follows:

Bush Heritage Australia at their Charles Darwin and Eurardy reserves. North Central Malleefowl Preservation Group have been involved with Bush Heritage Australia at their Charles Darwin Reserve on surveys for malleefowl. Being an organisation protecting nature Bush Heritage Australia have been involved in fox baiting not only for Malleefowl but also for the benefits it has for other species. Fire management is a priority on nature conservation land especially when large areas of natural vegetation are to be protected. The priority is not to let large hot fires happen. At their Eurardy property Bush Heritage Australia did a Malleefowl survey in August 2007 with the plan to monitor the mounds found annually to national standards. Two very old mounds were found in an area of 80 ha. These two mounds are to be monitored and the site expanded to be large enough to reach national standards.

Australian Wildlife Conservancy at their Mt. Gibson Sanctuary. As can be seen from the map on Figure 3 Mt. Gibson is relatively close to Charles Darwin Reserve, these are both in similar habitat on uncleared land. Because of this and the fact that the managing organisations have similar aims their management is similar in terms of factors such as fire management and fox control. There have been bush fires on Mt. Gibson Sanctuary over the last ten years but again, similar to Charles Darwin, the aim is to reduce large hot fires. Over the last three years the Australian Wildlife Conservancy have been implementing a fox control program on Mt. Gibson and it seems this may be having a good effect on Malleefowl numbers. North Central Malleefowl Preservation Group have been assisting Australian Wildlife Conservancy with surveying areas on Mt. Gibson for Malleefowl mounds for the purpose of setting up a monitoring site. The facilitator has mentioned setting up a site at Mt. Gibson to Australian Wildlife Conservancy but this has not been possible to date.

56

Future Challenges in WA

The Western Australian Malleefowl Network has been set up and has been instrumental in helping achieve meaningful outcomes for Malleefowl in Western Australia for the last three years. The challenge now is to keep the Western Australian Malleefowl Network alive into the future and to go on helping to achieve the most meaningful outcomes for Malleefowl in Western Australia.

Western Australia is in the early stages of getting a state-wide nest monitoring program to national standards under way. It will be a challenge to do everything required to get this to happen in Western Australia from establishing and surveying sites to getting monitoring of the sites happening on an annual basis. Also to get enough of these sites set up to be able to get reliable data representative of all areas in Western Australia that support Malleefowl.

Once the state-wide monitoring program is set up it will also be a challenge to take the data and work out what is happening with the birds in Western Australia. Are they declining? Why are they declining? How do we stop them declining? These are all questions of developing trends and using adaptive management to help Malleefowl.

Probably the biggest challenge for Malleefowl conservation in Western Australia is how to get people involved and how to keep them involved in the work the Western Australian Malleefowl Network is doing to help Malleefowl survive in the wild in Western Australia. This is a big challenge for any environmental group or for that matter any community group at all with all the various options people in Australia now have to occupy their spare time.

Acknowledgments

Many people and groups have provided information and other help in putting together this manuscript. Thank you to the following as well as all others for their help: Malleefowl Preservation Group, North Central Malleefowl Preservation Group, Friends of the North- eastern Malleefowl, Sally Cail, Judy O‘Neill, Claudine Deering, Blair Parsons, Jessica van der Waag, Susanne Dennings, Mick Davis. Thank you also to all stakeholders for their part in Malleefowl conservation in Western Australia since February 2004.

57

8. STATUS, MONITORING AND MANAGEMENT OF NGANAMARA IN THE SOUTHERN ALINYTJARA WILURARA NATURAL RESOURCE MANAGEMENT REGION

Dr Matthew Ward Department for Environment and Heritage, South Australia GPO Box 1047 Adelaide, South Australia, 5001

Keith Bellchambers Ardeotis Biological Consultants 74 Arkaba St. Hawker, South Australia, 5434

Abstract In the Maralinga Tjarutja Lands, there have been numerous historic records of Nganamara (Malleefowl) and Nganamara signs; however no regular monitoring of Nganamara sites has occurred. A recent survey of animal tracks across the Maralinga Tjarutja lands found six locations with fresh Nganamara tjina (tracks), confirming the continued occurrence of Nganamara in the region. The habitat of these signs varied from dense mallee to open mulga woodland. Subsequently, a survey took place in August 2007, involving Anangu and ecologists, targeting known Nganamara mounds and searching areas for new Nganamara sites. The results of this survey are presented here, and will form the basis of future monitoring and management of Nganamara in the southern Alinytjara Wilurara Natural Resource Management region. This monitoring should be aligned with monitoring occurring in the Anangu Pitjantjatjara Yankunytjatjara Lands and in Western Australia.

Background The Alinytjara Wilurara Natural Resource Management region in South Australia, which covers 26% of South Australia, is contiguous with the Western Australia border and stretches from the Northern Territory south to the Great Australian Bight (Figure 1). It encompasses the Anangu Pitjantjatjara Yankunytjatjara Lands, the Maralinga Tjarutja Lands, and Yalata Indigenous Protected Area, all Aboriginal managed lands. The region is one of the most intact and pristine wilderness areas in Australia, as most of the area has been untouched by the ravages of pastoralism and agriculture. As such, many important species of conservation significance still persist in the region, such as the Marsupial Mole Notoryctes typhlops, Tjakura (Great Desert Skink) Egernia kintorei; Princess Parrot Polytelis alexandrae, Sandhill Dunnart Sminthopsis psammophila and Nganamara (Malleefowl) Leipoa ocellata.

Despite the vast tracks of contiguous and potentially suitable habitat across the arid zone, Nganamara numbers are thought to have declined markedly in central Australia since the onset of pastoralism and the spread of foxes. In the 1990s, surveys revealed that the species still occurred in the Anangu Pitjantjatjara Yankunytjatjara Lands, and subsequent surveys have recorded numerous sites in the Anangu Pitjantjatjara Yankunytjatjara Lands. This suggested that the species could potentially occur sporadically across the vast landscape, primarily within the Great Victoria Desert bioregion.

Historic records of Nganamara within the southern portion of the Alinytjara Wilurara region (Maralinga Tjarutja Lands, Yumburra, Yellabinna and Yalata) are few in number, as summarised in Benshemesh (2007) and the Department for Environment and Heritage Biological Survey Database records (Figure 1). The scarcity of Nganamara records in the southern Alinytjara Wilurara region is no doubt in part due to the low densities of Nganamara, the logistical difficulties involved in surveying for Nganamara and the lack of biological surveys conducted in the area. Furthermore, there is little

58

Figure 1. Previously recorded Nganamara sightings in the southern Alinytjara Wilurara Natural Resource Management region, pre July 2007. recent information about the status of known Nganamara mounds / records within the region. Recently, however, Nganamara signs were detected at six locations in the Maralinga Tjarutja Lands, during an Animal Track survey of the Maralinga Tjarutja Lands designed to establish some ecological baseline data on the frequency of occurrence across the region of different native and non-native animals (Southgate et al. 2007). This indicated that Nganamara were still present in the region and that further information could be gathered on their status.

In line with Action 10.1 of the National Malleefowl Recovery Plan (Benshemesh 2007), data on the distribution, abundance and habitat preferences of Nganamara in the remote areas of South Australia and Western Australia needs to form the basis of future monitoring and management at a number of intensities. While the density of Nganamara in these areas may be low, the enormous areas involved suggest that they may be of great importance for Nganamara conservation (Bellchambers 2007).

Therefore, a survey for Nganamara was conducted for the southern portion of the Alinytjara Wilurara region. The primary aim of the survey was to establish known locations of Nganamara mounds to form the basis of future monitoring. This was deemed important and timely, because there is currently a considerable injection of funding for Natural Resource Management work in the Maralinga Tjarutja Lands, providing an opportunity for local community members to conduct Nganamara monitoring in the region. The basis of the 2007 survey was revisiting existing and reliable record locations (primarily recent records with high precision) from the Department for Environment and Heritage Biological Survey Database and areas where Nganamara tracks were recorded in the Animal Track survey.

The core objectives of the survey were to: 1. Determine the status of previously recorded Nganamara nest mounds in the southern parts of the Alinytjara Wilurara Natural Resource Management region, in particular Yellabinna, Yumbarra, Tallaringa and Mamungari Conservation Parks and Maralinga Tjarutja Lands,

59 2. Locate new Nganamara nest mounds in the southern parts of the Alinytjara Wilurara Natural Resource Management region, in particular Yellabinna, Yumbarra, Tallaringa and Mumungari Conservation Parks and Maralinga Tjarutja Lands, 3. Establish monitoring of located Nganamara nest mounds in accordance with National Nganamara Recovery guidelines, 4. Engage Anangu consultants, in particular from Yalata and Oak Valley, in the searching and monitoring of Nganamara nest mounds.

The following report discusses the methods and results of the survey, and the monitoring and management requirements for Nganamara in the southern Alinytjara Wilurara Natural Resource Management region, as reported in detail in Bellchambers (2007) and Ward & Clarke (2007). The Nganamara survey was conducted in August 2007 by Ardeotis Biological Consultants (Keith Bellchambers) and Desert Wildlife Services (Rachel Paltridge), organised by the Department for Environment and Heritage Aboriginal Lands Regional Ecologists and funded through the Department for Environment and Heritage and the Alinytjara Wilurara Natural Resource Management Board. Ardeotis arranged the survey for Nganamara mounds, while the primary role of Desert Wildlife Services was to facilitate Anangu employment in the survey. A detailed scientific report (Bellchambers 2007) and a plain language document and summary of Anangu involvement (Paltridge et al. 2007) is available from the Department for Environment and Heritage.

Survey methods and results

A database of previous records of Nganamara from the survey area was compiled to provide a starting point on which to base search effort. These records included nest mound locations, sightings of birds and observations of the Nganamaras‘ characteristic tracks. Records were sourced from Department for Environment and Heritage biological databases, Birds Australia Atlas database, and previous surveys in the area such as Greenslade et al. (1986), Southgate et al. (2007) and Robinson et al. (1990).

Given the size of the area of interest and the probable sparse nature of the distribution and abundance of the species within it, the process of creating a database of known locations enabled the decision to be made to concentrate survey effort on the accessible portions of the Yellabinna, Yumbarra, Tallaringa and Mamungari Conservation Parks and Maralinga Tjarutja Lands.

Previous observations from the survey area were relocated with a GPS. Most of these sites were re-visited and searched during this survey. In addition, new sites where nest mound, sightings of birds, observations of the Nganamaras‘ characteristic tracks and opportune sites in suitable habitat were also searched during this current survey. Some of these sites were located during opportune searches in suitable habitat whereas others were located while slowly driving along roads and tracks in the area.

The general search protocol was to spend 1 hour actively searching at a site for any sign of Nganamara. The effort involved in this active searching varied from 2 man-hours for most sites up to 8-10 man-hours when Anangu consultants from Oak Valley were on site. These searches involved walked random transects or up and back transects through an area of at least 500m by 500m around a central point.

When nest mounds were located, a set of standardised data were collected, as outlined in the National Nganamara Monitoring Manual (Natural Heritage Trust National Malleefowl Monitoring Project 2007). A photopoint was also established for future reference. An area of at least 500m by 500m around each mound was also searched in an effort to locate other mounds that may be nearby or other sign of Nganamara. Nganamara Monitoring Grids (Gillam 2005, 2007) were not set up at any mounds located during this survey due to the remoteness of the area and the sparse nature of the distribution and abundance of the species within it.

60 When Nganamara or their tracks were located, an area of at least 500m by 500m around each sighting was searched or tracks were followed wherever possible in an effort to locate nest mounds that may be nearby or other sign of Nganamara.

In addition to the sites where Nganamara had been recorded during this or previous surveys, a number of opportune sites in suitable habitat were searched during this survey. An area of at least 500m by 500m around each point was searched for an hour in an effort to locate nest mounds or any other sign of Nganamara.

Major Results of Nganamara Survey A total of 53 sites were assessed for the presence of Nganamara during this current survey, including 11 of 28 previously recorded Nganamara sites, and an additional 42 new sites (Figure 2).

The presence of Nganamara was detected at 20 sites during this current survey. Nest mounds were located at 8 sites, tracks were located at 11 sites and birds were observed at 5 sites. Several sites had a combination of Nganamara sign.

Figure 2. Sites where Nganamara were searched for in August 2007 survey in the southern Alinytjara Wilurara Natural Resource Management region.

The status and location of the 8 nest mounds located during this survey are summarised in Table 1 and Figure 3. Two nest mounds were known prior to this

61

Figure 3. Location of all known mounds (including active, inactive and mounds of unknown status) and sites with Nganamara tracks in the southern Alinytjara Wilurara Natural Resource Management region, January 2008.

current survey with the other 6 being located during this survey. Only 1 of the mounds was active at the time of the survey with birds raking litter into the nest in preparation for egg laying at a later date. Another mound had been active over the past couple of years (Robert Sleep pers. comm. - Department for Environment and Heritage Ceduna senior ranger), but inactive in this survey, whilst another had newly built into a spoil heap but inactive. The remaining mounds had not been in use for some time. One mound was located as a result of information from an Anangu

62 Table 1. List of sightings recorded in this 2007 Nganamara survey in MT Lands. Sighting Comments Mound New mound, not previously used. Low open mallee over shrubs with Spinifex groundcover. AWMF 1 Sandy soil. Tall dunes and narrow swales. Mound Inactive, used for past 2 years. Low open mixed species mallee over shrubs with Spinifex AWMF 2 and annual grasses. Sandy soil. Tall dunes and narrow swales. Old fire scar area. Mound Inactive, not used for >2 years. Low open mixed species mallee over shrubs with Spinifex AWMF 3 and annual grasses. Sandy soil. Tall dunes and narrow swales. Old fire scar area. Mound Active, being prepared for use. Open mixed species Acacia shrubland with few emergent AWMF 4 mallee over annual grasses. Sandy soil. Complex dunes. Mound Inactive, not used for 5 years. Low open mixed species mallee over sparse shrubs with AWMF 5 Spinifex groundcover. Sandy soil. Low dunes and broad swales. Mound Inactive, not used for >5 years. Low open mixed species mallee over sparse shrubs with AWMF 6 Spinifex groundcover. Sandy soil. Low dunes and broad swales. Mound Inactive, not used for >5 years. Tall Mulga shrubland over shrubs within open woodland AWMF 7 vegetation. Hard red soil. Broad plain. Mound Inactive, not used for >5 years. Low open mixed species mallee over sparse shrubs with AWMF 8 Spinifex groundcover. Hard sandy soil. Low dunes and broad swales. Tracks 1 day old. 1 bird. Low open mixed species mallee over shrubs with Spinifex groundcover. Sandy soil. Tall dunes and narrow swales. Old fire scar area. Tracks 1 day old. 1 bird. Low open mixed species mallee over shrubs with Spinifex groundcover. Sandy soil. Tall dunes and narrow swales. Tracks 1 day old. 1 bird. Open mixed species shrubland with few emergent mallee over annual grasses. Sandy soil. Broad plain. Tracks 1 day old. 2 birds. Associated with AWMF 4 and recent sightings nearby. Open mixed species Acacia shrubland with few emergent mallee over annual grasses. Sandy soil. Complex dunes. Tracks 1 week old. 1 bird. Tracks also recently observed in area. Low open mixed species mallee over shrubs with Spinifex groundcover. Hard sandy soil. Broad plain. Tracks Today up to 1 week old. 2 birds. Bird sighted today and bird and tracks also recently observed in area. Low open mixed species mallee over shrubs with Spinifex groundcover. Hard sandy soil. Broad plain. Tracks 1 week old. 1 bird. Low open mixed species mallee over sparse shrubs with Spinifex groundcover. Sandy soil. Low dunes and broad swales. Tracks Today up to 1 week old. 2 birds. Low open mixed species mallee over sparse shrubs with Spinifex groundcover. Sandy soil. Low dunes and broad swales. Tracks 1 week old. 1 bird. Low open mixed species mallee over sparse shrubs with Spinifex groundcover. Sandy soil. Low dunes and broad swales. Tracks Today up to 1 week old. 2 birds. Associated with AWMF 7 and sightings today. Tall Mulga shrubland over shrubs within open woodland vegetation. Hard red soil. Broad plain. Tracks 1 day old. 2 birds. Open mixed species Acacia shrubland with few emergent mallee over annual grasses. Sandy soil. Complex dunes. Birds Seen last week. 2 birds. Associated with AWMF 4. Open mixed species Acacia shrubland with few emergent mallee over annual grasses. Sandy soil. Complex dunes. Birds Seen last week. 2 birds. Low open mixed species mallee over shrubs with Spinifex groundcover. Hard sandy soil. Broad plain. Birds Today. 1 bird. Tracks of 2 birds sighted today and bird and tracks also recently observed in area. Low open mixed species mallee over shrubs with Spinifex groundcover. Hard sandy soil. Broad plain. Birds Seen 2 months ago. 1 bird. Low open mixed species mallee over sparse shrubs with Spinifex groundcover. Sandy soil. Low dunes and broad swales. Birds Today. 2 birds. Associated with AWMF 7. Tall Mulga shrubland over shrubs within open woodland vegetation. Hard red soil. Broad plain.

63

consultant who indicated that she had visited it approximately 5 years ago and removed 12 eggs from it (Brown pers. comm.). It appeared not to have been used since that time. Most mounds were located in low open mallee over shrubs with spinifex groundcover on sandy soil in dune and swale systems, however mounds were found in a variety of vegetation types and land systems.

The 11 locations at which Nganamara tracks were observed during this survey are summarised in Table 1 and their locations are shown in Figure 3. Observed track patterns varied from single birds foraging over large areas where tracks could be followed for several kilometres over the terrain to relatively small, high use areas by several birds, associated with nest mounds or feeding habitat. None of the sites where tracks were observed during this survey were associated with previous observations of Nganamara from the survey area.

The 5 locations at which Nganamara were recorded during this survey are summarised in Table 1 and their locations are shown in Figure 3. Two of the observations were made while on the field survey and the other three were recent reports from locals familiar with the species. Two of these reports were from the Oak Valley store truck driver who frequently travels through the area and the other reported sighting was made by the Land Management Officer for Maralinga Tjarutja based in the Oak Valley community (Dodd pers. comm.).

The single bird observed during the survey was quietly feeding amongst low shrubs and in the litter under low mallee. There were numerous tracks of 2 birds in the area and a bird had previously been sighted in the area (Ashton pers. comm., Joseph and Pedler pers. comm.). The pair of birds observed during the survey was quietly feeding amongst thick litter under tall Mulga shrubland and were close to an inactive mound. There were numerous tracks of 2 birds and feeding sign in the area. None of the sites where Nganamara were observed during this survey were associated with previous observations of Nganamara from the survey area.

In summary, 53 sites were assessed for the presence of Nganamara during this survey. 20 sites had some sign (nest mound, tracks or sighting) and 33 sites had no sign of Nganamara. When combining old sighting with new sightings, there are now a total of 44 sites within the search area from the southern part of the Alinytjara Wilurara Natural Resource Management region at which Nganamara have been recorded.

Distribution

The results obtained from this survey confirm that the Nganamara is widespread but at very low densities in the southern part of the Alinytjara Wilurara Natural Resource Management region. The number of sites at which Nganamara have been recorded in the survey area was nearly doubled at the completion of this survey, however, there are still less than 50 sites at which the species has been recorded.

Nganamara observations from the region have been recorded by explorers, amateur ornithologists, site specific biological surveys, track surveys and this current survey yet little is known about the species away from the few areas that can be accessed. It is assumed that the species is also widespread but at very low densities in those enormous areas of country away from the road and track system. Given the size of the area, there does not need to be a high density of birds for there to be a potentially large population of Nganamara in the region.

The historical database combined with the results of this and other recent fauna surveys in the region indicate that, while the Nganamara is distributed widely throughout the region, its abundance in the region is highest in the south and around the Maralinga area. There are fewer records from the Vokes Hill Corner and Anne Beadell Highway section of the region and then another grouping of records in the north-west of the state in the Anangu Pitjantjatjara Lands.

It is not known if this reflects survey effort or is a measure of habitat preferences by the Nganamara. The results of this current survey and previous observations indicate that the

64 species utilises a number of habitat types but the south of the region is dominated by low open mixed species mallee over shrubs with spinifex groundcover with sandy soil on dunes and swales. Farther north in the region the mallee becomes more open and the dunes become lower and the swales become broader. North of Maralinga and Oak Valley the vegetation becomes dominated by open mixed species Acacia shrubland with few emergent mallee over annual grasses with harder sandy loam soils on broad plains or complex dunes. Photographs of Nganamara mounds at Walalkara even farther north show low very open Acacia shrubland (Acacia minura) over Spinifex on harder sandy soils.

Monitoring Nganamara in the Alinytjara Wilurara Natural Resource Management region Traditional monitoring of Nganamara, which involves grid based searches, is not appropriate in the Alinytjara Wilurara region because of the sparse nature of Nganamara mounds, the low density of birds, and the fact that many records of Nganamara are not of mounds but simply of tracks and traces. With such low numbers and the difficulties in managing such a vast landscape, there are questions as to whether monitoring of Nganamara is necessary in this landscape. Below we outline why monitoring of Nganamara should occur in the Alinytjara Wilurara region.

Legislative

Recently, the South Australian Government endorsed No Species Loss; A Nature Conservation Strategy for South Australia (Department for Environment and Heritage 2007), which has an aspirational goal of losing no native species to extinction from the state. Nganamara is listed as Vulnerable, both nationally (Environment Protection and Biodiversity Conservation Act 1999), and in South Australian (National Parks and Wildlife Act 1972). These acts require the Alinytjara Wilurara Natural Resource Management Board and Department for Environment and Heritage to undertake efforts to conserve the species.

There is also an imperative to monitor Nganamara within a Natural Resource Management framework. Short-term Management Action Target 9.1 for the Maralinga Tjarutja Lands in the Alinytjara Wilurara Integrated Natural Resource Management Plan is to ―ensure adequate monitoring programs are in place for native plant and animals by June 2005‖ (Aboriginal Lands Integrated Natural Resource Management Group 2004).

There is, therefore, a legislative mandate to monitor Nganamara populations to be able to determine population trend, better clarify conservation status and determine whether active management is necessary.

Conservation / Ecological

Current trends in Nganamara populations across Australia indicate a decline in numbers. Nganamara populations have declined by at least 20% over the past three generations (estimated at 15 years each), and it is likely that populations may decline by at least another 20% over the next three generations (Benshemesh 2007). Monitoring of Nganamara in the southern Alinytjara Wilurara is, therefore, important in determining whether Nganamara are following the national trend of decline.

Monitoring of Nganamara is also an essential component of the adaptive management framework adopted by the National Malleefowl Recovery Team at the 2007 National Malleefowl forum. By monitoring and establishing baseline information on persistence of Nganamara, conservation managers will have more information at hand should the implementation of different management scenarios be deemed appropriate or necessary.

In the southern Alinytjara Wilurara Natural Resource Management region, monitoring of Nganamara will provide a focus around which other ecological monitoring could occur, therefore providing a broader scope of baseline ecological information for the region. For

65 example, the monitoring of Nganamara should be combined with regular track transects and 2-ha sand-plot animal track surveys (Southgate et al. 2007), which will provide information on predator dynamics over time.

Social

Nganamara hold their own place within Aboriginal society of central desert regions, featuring in Aboriginal mythology and associated with certain ‗Tjukurpa‘ or ‗Dreaming‘ sites and trails in central Australia (Benshemesh 2007). Nganamara were also an important food and protein source - their eggs (ngampu) were harvested as a traditional food source and birds would probably also have been harvested occasionally. Traditionally, individual Nganamara mounds were under the custodianship of individual Traditional Owners (Donald Fraser, pers. com.). After the broader conservation plight of Nganamara was explained to Traditional Owners in the Anangu Pitjantjatjara Yankunytjatjara Lands, local communities and Traditional Owners have demonstrated a serious conservation concern and custodianship for Nganamara.

Given this, Anangu in the Alinytjara Wilurara region need to be given every possible opportunity to take responsibility for the custodianship and monitoring of Nganamara (i.e. appropriate resources and incentives, both financial and logistical). Anangu have many skills and knowledge to contribute to the monitoring of Nganamara in the region, and should be consulted in all aspects of future monitoring and survey. For example, capturing indigenous knowledge regarding the location of Nganamara mounds was an initial step in the recent Nganamara surveys. Also, Anangu have generations of experience in looking after country, and should always be consulted and engaged in land management work associated with Nganamara. This consultation should continue throughout and beyond the lifespan of projects, to ensure that the most up-to-date knowledge of the region and of Nganamara is at hand.

The monitoring of Nganamara also has the potential to deliver numerous positive social benefits for local indigenous communities. As discussed already, Nganamara surveys in the Maralinga Tjarutja lands provided an opportunity for members of the Oak Valley community to take part in land management work. As a result, significant employment was gained through the survey (a total of 332 employment days, or roughly $11,000 of employment). The monitoring of threatened species can provide a focal point for land management work, through which a number of other positive social outcomes are generated, including: -• training in land management skills (e.g. GPS, Cybertracker); Formatted: Bullets and Numbering -• consolidation of natural history skills (e.g. tracking); -• intergenerational transfer of skills and indigenous ecological knowledge; -• employment; and -• positive health benefits.

Landscape monitoring of Nganamara persistence and area of occupancy It is difficult to measure and determine the persistence of Nganamara over a landscape such as the Alinytjara Wilurara Natural Resource Management region, because of the vast areas of potentially suitable habitat, the scarcity of animals and the difficulty in finding animals. Furthermore, finding tracks in an area does not necessarily indicate a sedentary pair, as often tracks are those of dispersing birds.

Because of these difficulties, it will be useful to report the persistence and status of Nganamara in the Alinytjara Wilurara region (both southern portion and northern Anangu Pitjantjatjara Yankunytjatjara portion), in terms of the ―area of occupancy‖.

The ―area of occupancy‖ is a criteria used by the International Union for the Conservation of Nature. It is defined as the area within its ‗extent of occurrence‘ that is occupied by a

66 taxon, excluding cases of vagrancy. It is important to note that the area of occupancy will be a function of the scale at which it is measured, and should be at a scale appropriate to relevant biological aspects of the taxon, the nature of the threats and the available data (International Union for the Conservation of Nature 2001).

One way of determining area of occupancy is to establish consistent grid cells of a certain size, and each year report on the extent of occurrence and area of occupancy of Nganamara (and other animals) within these networks of grid cells. A recent project initiated by the Department for Environment and Heritage for the Northern and Yorke Region is attempting to determine the status of threatened species within that Natural Resource Management region by determining the area of occupancy of taxa using a 10km x 10km grid of cells developed for the entire state for extent of occurrence, and 1km x 1km cells for area of occupancy. Given the aforementioned project, it may be timely to implement a similar monitoring and reporting structure for biodiversity (in particular animals) across the Alinytjara Wilurara Natural Resource Management region.

Although there are issues with scale (are 10 km x 10 km cells appropriate for the Alinytjara Wilurara region, given its size?), using a grid consistent with other parts of the state would be valuable. Furthermore, the monitoring and reporting of threatened species area of occupancy, such as Nganamara, could be restricted to those grid cells which occur close to main tracks. This is where, realistically, the majority of the information about threatened species occurrence would occur.

Broad objectives for Nganamara monitoring in southern Alinytjara Wilurara region, 2008-2012 Monitoring of Nganamara in other parts of Australia has taken place for many years. As such, specific parameters such as breeding density of Nganamara pairs can be quantified, as well as the population trend in a particular region (i.e. increasing / decreasing).

Monitoring of Nganamara in the southern Alinytjara Wilurara region is in its very early stages, and there are many logistical and social difficulties that must be considered in conducting such work. Therefore, the aims of Nganamara monitoring in the southern Alinytjara Wilurara region should be relatively simple, and not necessarily all scientific. For example, land managers and government agencies need to initially demonstrate that monitoring can actually take place in a regular and consistent manner.

The broader aims of Nganamara monitoring in the southern Alinytjara Wilurara region for 2008 - 2012 should be to: 1)• Establish monitoring guidelines for Nganamara Formatted: Bullets and Numbering 2)• Communicate these with relevant staff (Alinytjara Wilurara project staff, land management staff, Department for Environment and Heritage rangers) 3)• Establish a database for Nganamara monitoring in the region 4)• Establish capacity of land management authorities to conduct monitoring, including equipment, resources and work plans 5)• Conduct consistent annual monitoring of known Nganamara sites 6)• Conduct surveys for Nganamara in new areas of the region; and 7)• Review monitoring data

Management of Nganamara in the southern Alinytjara Wilurara region Management of Nganamara in other parts of Australia often focuses on potential threats to Nganamara, as listed in the National Recovery Plan: clearing; fragmentation and isolation; grazing; predation; fire (wildfire and intentional burns); disease and inbreeding; and climate change (Benshemesh 2007). Of these, predation and wildfire are realistic threats to Nganamara in the Alinytjara Wilurara Natural Resource Management region.

67 Direct ‗management‘ of Nganamara in the southern Alinytjara Wilurara Natural Resource Management region is difficult, because of the massive area over which animals occur. Any consideration of management activities must consider the following questions: • Is it financially sustainable over a long time and large spatial area? • Is it logistically possible over a long time and large spatial area? • Can we measure its potential impact and intended outcome?

Predation

Research has demonstrated that Nganamara chicks of any age suffer heavy mortality as a result of fox predation. Foxes are most common in mallee near agricultural land where high densities may be maintained by the ready availability of rabbits, mice and sheep carrion (Saunders et al. 1985).

That said, there is little clear evidence that Nganamara populations have increased following fox control operations, even though fox control is widely practised in areas where Nganamara conservation is a concern (Benshemesh 2007). This is due to varying effectiveness of baiting programs, the fluctuating nature of Nganamara breeding effort and success, the varying effect of predation in different habitat types, and importantly, the lack of adequately controlled and replicated studies (Benshemesh 2007). It must also be noted that recent analysis of Nganamara monitoring data has indicated that variation in rainfall has a significant effect on Nganamara breeding success.

For arid areas north of the dog fence, it must also be noted that foxes are not the only ground predator of Nganamara, and the nature of the relationship between foxes, cats and dingoes / dogs is very important. For example, there is evidence that dingoes might suppress the numbers of foxes and cats (Glen et al. 2007), and any baiting may lead to a localised increase in fox or cat activity. Furthermore, there is little chance of controlling foxes over an area as vast and inaccessible as the Maralinga Tjarutja Lands.

Given the preliminary nature of Nganamara and other ecological monitoring in the Alinytjara Wilurara region, the sole target for management of predation should be to collect regular and consistent information on predator dynamics in areas associated with Nganamara. This is in line with the discussion paper produced by Department for Environment and Heritage on 1080 baiting in the Alinytjara Wilurara Natural Resource Management region (Clarke & Ward 2007). In the past, regular monitoring of predators has yet to be achieved in the Alinytjara Wilurara Natural Resource Management region. In 2007, however, the Warru (Black-flanked Rock-wallaby Petrogale lateralis MacDonnell Ranges race) Recovery Team began to collect good information on predator dynamics around a remnant colony, but only through a dedicated and continuous project focus.

The most effective and consistent way of collecting information on predator dynamics associated with threatened species sites (e.g. Nganamara, Sandhill Dunnart) will be to monitor dedicated track transects and 2-ha sand-plots (see Southgate et al. 2007), as described in Ward & Clarke (2007).

Such predator dynamic monitoring is critical to adopting a successful adaptive management framework for Nganamara and other threatened species in the region. This adaptive management framework has been adopted by the Natural Heritage Trust National Malleefowl Monitoring Project. Any future effort to assess and address fox predation pressures should be in line with a national approach. Recent analysis of the Nganamara data across the country has indicated that fluctuations in breeding activity of Nganamara are also related to rainfall, and, therefore, declines in breeding activity are not always tied to predation pressures.

Patch-burning In central Australia, traditional burning practices created a mosaic of different aged habitats. A heterogeneous post-fire habitat would have benefited Nganamara (and biodiversity in general) by providing long-unburnt habitat required by Nganamara for nesting and shelter

68 (Benshemesh 1990). These burning practices were interrupted by pastoral activities, as well as the move away from a traditional lifestyle by Aboriginal inhabitants of central Australia. As a result, a single large wildfire has the potential to temporarily remove large tracts of suitable habitat. The effects of large wildfires on biodiversity are worse when compounded with long periods of low rainfall and explosions of feral herbivores such as rabbit and camels.

Information on the historic frequency and nature of patch burning in the southern Alinytjara Wilurara Natural Resource Management region, however, is currently not readily available. Before fire management is prescribed across the region, a well researched, comprehensive and well resourced fire strategy for the region needs to be developed. For threatened species such as Nganamara, this should aim to capture the knowledge of traditional burning practices around Nganamara mounds from Traditional Owners. There also needs to be the consideration of whether fire is to be used for site protection and threat mitigation, or to promote a heterogenous vegetation mosaic that provides a range of food sources, as their prescription could be quite different.

In the short-term, patch burning around Nganamara sites should be limited and aimed simply at protection of Nganamara sites. This should be done on a case by case basis, depending on the habitat surrounding the mound. It is very important that this is carried out at an appropriate time of year by Anangu with a lot of knowledge and experience in patch-burning, to prevent larger wildfires. There have been instances where inappropriate burning in the Alinytjara Wilurara region has lead to large wildfires that have burnt for months.

Conclusions As a result of two recent widespread surveys in the southern Alinytjara Wilurara Natural Resource Management region, we can confirm that Nganamara continues to persist in the vast tracts of mallee and mulga scrub in the Maralinga Tjarutja Lands, and the Yumburra and Yellabinna conservation reserves. However, Nganamara occur across this vast landscape in low densities and are sparsely distributed, and only one known mound was active in 2007. However, Anangu, land managers and scientists now have the capacity to begin monitoring of Nganamara in the southern Alinytjara Wilurara Natural Resource Management region, and across the Great Victoria Desert bioregion as a whole. Monitoring of Nganamara in the southern Alinytjara Wilurara region will be based around Nganamara ―sites‖, including mounds, and known location of tracks and birds, and will attempt to measure persistence over time by using area of occupancy measures.

References

Aboriginal Lands Integrated Natural Resource Management Group (2004) Integrated Natural Resource Management Plan for the Aboriginal Lands Integrated Natural Resource Management Region of South Australia. Aboriginal Lands Integrated Natural Resource Management Group, Inglefarm, South Australia.

Bellchambers, K. (2007) Nganamara surveys in the southern Alinytjara Wilurara Region. Report for Department for Environment and Heritage, South Australia, Adelaide.

Benshemesh, J. (1990). Management of Malleefowl - with regard to fire. In 'The Mallee Lands. A Conservation Perspective'. Noble, J. C., Joss, P.J. and Jones, G.K. (eds.) pp. 206-211. CSIRO: Melbourne.

Benshemesh, J. (2007) National Recovery Plan for Malleefowl 2006-2010. Department for Environment and Heritage, South Australia, Adelaide.

Clarke, A., & Ward, M. J. (2007). Alinytjara Wilurara NRM Region 1080 Baiting Issues Discussion Paper. (Department for Environment and Heritage: Adelaide.)

Gillam, S. (2005). An update of Malleefowl monitoring in South Australia – project progress report. Department for Environment and Heritage, South Australia.

69 Gillam, S. (2007). Establishing and Re-searching Malleefowl Monitoring Grids. Draft Unpublished Document. Department for Environment and Heritage, South Australia.

Glen, A. S., Dickman, C. R., Soule, M. E., & Mackey, B. G. (2007). Evaluating the role of the dingo as a trophic regulator in Australian ecosystems. Austral Ecology 32, 492- 501.

Greenslade, P. Joseph, L. & Barley, R. (eds.) (1986). The Great Victoria Desert. Nature Conservation Society of South Australia Inc, Adelaide.

IUCN (2001) IUCN Red List Categories and Criteria: Version 3.1. IUCN, Gland, Switzerland and Cambridge.

NHT National Malleefowl Monitoring Project. (2007) National manual for the Malleefowl monitoring system. Standards, protocols and monitoring procedures. National Heritage Trust.

Paltridge, R., Bellchambers, K. & Carpenter, G. (2007). Looking for Nganamara on the Maralinga Tjarutja Lands. Department for Environment and Heritage, South Australia.

Robinson, A.C., Casperson, K.D. & Copley, P.B. (1990). Breeding records of Malleefowl (Leipoa ocellata) and Scarlet-chested Parrots (Neophema splendida) within the Yellabinna Wilderness Area, South Australia. South Australian Ornithologist 31(1) 8-12.

Saunders, G., Coman, B., Kinnear, J. E., & Braysher, M. (1985). Managing Vertebrate pests: Foxes. (Australian Government Publishing Service: Canberra.)

Southgate, R., Paltridge, R., & Eldridge, S. (2007) Investigation of animal tracks in the Maralinga Tjarutja lands in the Alinytjara Wilurara NRM (AW) region. Department for Environment and Heritage, Adelaide.

Ward, M.J. & Clarke, A. (2007) Monitoring Nganamara in the southern Alinytjara Wilurara NRM region 2008 – 2012. Department for Environment and heritage, Adelaide.

70

9. NGANMARA IN SOUTH AUSTRALIA‟S ANANGU PITJANTJATJARA YANKUNYTIJATARA LANDS

Thalie Partridge Threatened Species Officer Anangu Pitjantjatjara Yankunytjatjara Land Management

Abstract

Nganamara mounds in the Anangu Pitjantjatjara Yankunytjatjara Lands have been monitored since the early 1990s by Anangu, Anangu Pitjantjatjara Yankunytjatjara Land Management and visiting ecologists, primarily within the Walalkara and Watarru regions. Twenty-six mounds are recorded in the Anangu Pitjantjatjara Yankunytjatjara Land Management database. No new active mounds have been recorded since 2005, some mounds have become inactive during the period of monitoring. Reasons for this remain unclear. Some concern has been expressed by traditional owners about the effect of predators; however more systematic monitoring of Nganamara and their predators will be undertaken before predator control is considered. Foot surveys led by Anangu from Watarru have recorded Nganamara tjina (tracks) in habitat to the east of recorded mounds and surveys will continue in this area and other areas with suitable habitat.

Introduction

The Anangu Pitjantjatjara Yankunytjatjara lands, in the north-west of South Australia, are a very large area to manage. There are quite a number of aboriginal communities of different sizes in the lands, and the lands are used in various ways. Cattle grazing, particularly in the north, but to the south, where it is extremely remote country, there is a limited amount of traditional land management, mostly in the Walalkara and Watarru areas. Sand dunes are common but there are some areas of mallee-spinifex. A biological survey was conducted in the area in 1995 and some further research work has taken place since then.

Survey Results

Nganamara (Pitjantjatjara name for the Malleefowl, Leipoa ocellata) mounds in the Anangu Pitjantjatjara Yankunytjatjara Lands have been monitored since the early 1990s by Anangu, Anangu Pitjantjatjara Yankunytjatjara Land Management and visiting ecologists. Records exist primarily from within the Walalkara and Watarru Indigenous Protected Areas and nearby country. Anecdotal records derived from Indigenous knowledge, rare sightings and the distribution of suitable habitats suggest that the species was once more widely spread throughout the Anangu Pitjantjatjara Yankunytjatjara lands. Thirty-eight mounds (Benshemesh 2007) are recorded in the Anangu Pitjantjatjara Yankunytjatjara Land Management database (Figure 1).

Breeding success has been monitored (Figure 2) through the presence of eggs (ngambu), egg shells (nyiri) and tracks of young (tjitji). In 1998/1999 eggs were removed from mounds and measured. The size of eggs was considered to be within the normal range (Benshemesh 1999). Foot surveys (Figure 3), led by Anangu from Watarru, have recorded Nganamara tjina (tracks) in habitat to the east of recorded mounds. Future surveys will continue here and in other areas of habitat recognised as suitable for Nganamara by Anangu. An active mound was recorded in 2005 by Anangu from Walalkara after following tracks first found at an inactive mound.

Anangu have detailed knowledge of Nganamara habitat and food plants (eg. Acacia ramulosa, A. kempeana, Senna artemisioides, Waitzia acuminata). Robin Kankanpakantja (senior custodian from Walalkara) has been able to confirm that Nganamara were not in an area because food plants were either not available or not currently flowering/seeding. Anangu also continue to conduct patch-burning activities in Triodia (spinifex) grassland surrounding

71 Nganamara habitat. However, resources currently limit Anangu Pitjantjatjara Yankunytjatjara Land Management‘s ability to support and develop this activity.

Figure 1. Malleefowl mounds recorded in the Anangu Pitjantjatjara Yankunytjatjara Land Management database (Benshemesh 2007).

No new active mounds have been recorded since 2005. Some mounds have become inactive during the 15 year period of monitoring. Reasons for this remain unclear. Some concern has been expressed by traditional owners about the effect of predators. In March 2007 tracks at a mound showed that a dingo had been sitting on the mound and, when a Ngananmara approached, the dingo chased it and it flew away. In 2002 and 2003 baiting for predators, dingos, cats and foxes, was trialled at the mounds and in the surrounding country. This was stopped due to concerns that it was actually attracting predators to the area. Lack of resources prevented adequate monitoring. More systematic monitoring of Nganamara and their predators will be undertaken before predator control is considered in the future.

Future work

Aims for future monitoring include: • Develop more systematic monitoring of Nganamara activity across the Anangu Pitjantjatjara Yankunytjatjara lands • Annually assess active and inactive mounds for signs of returning Nganamara • Conduct extensive foot surveys into remote Nganamara habitat • Develop and conduct predator monitoring at active and inactive Nganamara mounds and in country where other signs of Nganamara have been recorded • Utilise Cybertrackers to assist Anangu with independent data recording

72

Figure 2. Anangu Pitjantjatjara Yankunytjatjara Land Management monitoring a Nganamara mound at Walalkara, 2007.

Figure 3. Anangu Pitjantjatjara Yankunytjatjara Land Management conducting a foot survey for Nganamara tracks at Watarru, 2007.

Acknowledgements

I would like to acknowledge all of our Anangu Rangers, Anangu Pitjantjatjara Yankunytjatjara, current and past staff of Anangu Pitjantjatjara Yankunytjatjara Land Management, Joe Benshemesh, the Department of Environment and Heritage, Alinytjara Wilurara, Threatened Species Network and the Indigenous Land Corporation.

73

References

Benshemesh, J. (1999) Letter to Anangu Pitjantjatjara Land Management

Benshemesh, J. (2007) Malleefowl monitoring data collected in the Anangu Pitjantjatjara Yankunyjatjara Lands since 1990. Report to Department of Environment and Heritage SA and Anangu Pitjantjatjara Yankunytjatjara Land Management.

74

10. CAPTIVE BREEDING AND MANAGEMENT OF MALLEEFOWL (Leipoa ocellata) AT WESTERN PLAINS ZOO, 1989 TO PRESENT

Judith Gillespie BSc Western Plains Zoo, Obley Rd, Dubbo, NSW, 2830. e-mail: [email protected]

Abstract

The Zoological Parks Board of New South Wales has been breeding Malleefowl (Leipoa ocellata) in captivity at Western Plains Zoo for almost 20 years to assist the New South Wales Malleefowl Recovery program. Successful breeding can be achieved through appropriate aviary design and husbandry resourcing. The captive program provides a valuable process for developing husbandry skills and highlights the valuable role captive institutions offer as sites for insurance-populations in case of catastrophic declines of birds in the wild. Continuation of the breedng for release program at Western Plains Zoo provides the opportunity for maintaining husbandry skills as well as opportunities for research that can assist managing birds for release.

Introduction

Western Plains Zoo has been involved in the captive breeding and reintroduction of Malleefowl in New South Wales for close to 20 years. The program began in 1988 when the Zoological Parks Board of New South Wales partnered with the New South Wales National Parks and Wildlife Service (now Department of Environment and Climate Change) as part of the Malleefowl recovery effort in the state. The Zoological Parks Board‘s role in the program was to take eggs collected from the wild by Department of Environment and Climate Change and incubate and rear them at Taronga Zoo. The chicks were then returned to Department of Environment and Climate Change for wild release and some were also sent to Western Plains Zoo to set up the nucleus of a breeding program to produce offspring for ongoing annual release.

In 1991 Western Plains Zoo built a series of aviaries to accommodate the expanding program and since then over 500 chicks have been released in three Nature Reserves in the central far west of New South Wales, these being Yalgogrin, Yathong and most recently at Nombinnie/Round Hill.

Along the way much has been learned about caring for Malleefowl in the captive environment. The development of husbandry skills provides insurance for the species when in the event of a serious decline of birds in the wild, captive breeding can be used as recovery mechanism. As part of the aims of the Recovery Plan a husbandry manual is in production. At present further breeding of Malleefowl at Western Plains Zoo is on hold at the request of Department of the Environment and Climate Change and the zoo is maintaining a core captive breeding group.

Facility Design

In the right conditions Malleefowl have proven relatively easy to breed. At Western Plains Zoo the Malleefowl facilities are comprised of 16 off-display breeding aviaries as well as a Walkthrough Aviary on public display. The breeding aviary complexes, located in a remote corner of the zoo, are composed of seven separate units, these units being either paired aviaries or triplets of aviaries.

The main characteristics required of facilities are that they are spacious, dry and well drained. In managing aviaries, it is essential to provide good service access for keepers as each

75 autumn they have to barrow in large qualities of leaf litter for the birds to commence preparing their mounds (Figures 1 nd 2).

Breeding aviary dimensions and features are: Two aviary units – 10 m x 20 m x 2.5 m, joined length-wise with an external airlock Five aviary complexes - each 20 m x 20 x 2.5 m with an internal airlock that separates two 10 m x 10 m aviaries Mesh size used over the aviaries is 12.5 mm x 12.5 mm x 1.2 mm gauge and there is predator proof wire mesh (chain link) extending vertically below the ground and also horizontally out from the aviaries for half a metre. There is also a layer of course pebbles around the outside of the aviaries to deter digging Aviary perimeters have shade cloth over the lower portion to provide a visual barrier that assists the chicks to remain calm and prevent males from seeing each other The aviary complexes were renovated in 2003, being completely rewired, support cabling and posts replaced and new shade cloth erected, at cost of $40,000. To minimise solid roof framework that can be hazardous to birds, staff came up with the idea of using plough discs welded to the top of posts to support roof cabling, to which the mesh is then attached Complexes are set up with a compatible pair in the large aviary, with an empty pen opposite to house chicks. Breeding pairs should not be housed in adjacent enclosures where males can see each other Chicks, once hatched, are moved to an adjoining aviary to form crèches and during this process are wing feather-clipped and micro-chipped. There are two further catch-ups, one at around six months for sexing, and at final release with health checks The Walkthrough Aviary is 40 m x 20 m with a sloping roof to a maximum height of 4 metres and a minium height of 1.2 m, and it houses the most productive breeding Malleefowl pair. The mound overspills onto the public walkway providing the visitors with a wonderful opportunity to see the pair working the mound as well as a very intimate view of mound chamber when it is open

At Western Plains Zoo there is a full time keeper to care for eight Malleefowl pairs and their chicks as well as the other birds in the Walkthrough Aviary. There is also a back-up keeper to cover leave and there are two veterinarians on staff to undertake any medical care of the birds. Adequate staff resources are important to enable behavioural observations of individuals, for the labour intensive mound management during the breeding season and to enable conditioning activities with chicks.

Out of the breeding season birds are provided with a basic seed mix (millets) with sorghum, wheat and cracked corn, supplemented with mealworms, fly pupae and fresh alfalfa sprouts. This diet is further supplemented in the breeding season by the addition of a small quantity of ‗soft bill‘ diet, (blended boiled egg, insectivore dry mix, biscuit powder), and green lucerne.

Husbandry observations

The flighty nature of Malleefowl offers a challenge in captivity and several techniques have been employed to assist managing them in the confines of an aviary. Initially chicks were feathered clipped on one wing to reduce their ability to launch upwards into the netting. More recently keepers have instituted regular conditioning of birds to their presence, using mealworms as a positive reinforcement has proven an effective strategy for luring birds into the smaller airlock areas where they can be more effectively confined and restrained. This method significantly lowers the risk of injury to the birds.

One of the most time-consuming husbandry activities is the collection of large volumes of natural leaf litter, and sometimes soil, which needs to be provided on a regular basis to the breeding pairs between May and September. Each mound needs anywhere between 15 to 50 barrow loads of litter over this period, which provides the necessity for easy service access into the aviaries. This material is sourced from around the zoo‘s grounds.

While managing Malleefowl in captivity Western Plains Zoo keepers have observed some interesting behaviours in their birds.

76

Figure 1.. Malleefowl breeding aviary

Figure 2. Mesh support and suspension technique

• There‘s no guarantee that pairs set up together will be compatible. Some pairings have been unproductive and these birds have usually been repaired after two years if breeding has been unsuccessful

Adults can be aggressive to their own chicks as well as to other bird species. One adult male has had a tendency to repeatedly attack chicks and this behaviour has been managed by building a ‗creep‘ in which newly hatched chicks can shelter until they are safely removed to another aviary

77 In the walkthrough aviary large ground birds such as Bush-stone Curlews and Banded Plovers were not tolerated around the mound and Apostlebirds were removed after being observed harassing Malleefowl chicks. Quail, finch and softbill activity in the walkthrough does not appear to elicit any negative responses by the Malleefowl Contrary to the general observation that the male undertakes the initial mound building there has been one female that was one of the earliest mounders Malleefowl have proven to be very determined in temperament, in that no matter how inconveniently they may locate a mound, they will not relocate even after keepers have prepared an alternative site A four month old Malleefowl chick was observed ‗surrendering‘ to ants. The chick was found sitting in a group with other chicks with eyes closed and covered in ants. This individual had been seen two hours prior, walking around the aviary. On becoming aware of the keeper‘s presence the chick arose and shook off the ants (H. Findlay pers.comm.) Malleefowl have been observed flying around their aviaries (apparently free of external motivating stimuli) (H. Findlay pers.comm.)

Breeding and Development Results

Breeding results have been variable between individual pairs and from year to year. As with wild Malleefowl, the natural climatic conditions influence the breeding activity in captivity with little or no mounding in extreme drought periods.

At Western Plains Zoo females have commenced laying at a range of ages, starting as early as two years old and as late as 10 years old. Males are generally recorded to commence mound building activity around three to four years of age and occasionally as early as two years.

WPZ Individual Female Malleefowl chick production 1992-2006

30

25

751098 20 761009 061098 651079 15 861009

No. No. chicks 361098 10 561098 A00060

5

0 92-93 93-94 94-95 95-96 96-97 97-98 98-99 99-00 00-01 01-02 02-03 03-04 04-05 05-06 breeding season

Figure 3. Seasonal chicks results for individual Malleefowl females

The sizes of clutches (based on total number of chicks hatched) have ranged from nought to 27 and the highest average clutch size (number of hatched chicks) for a pair was 19 and the lowest was six (Figure 3).

78

700

600

500

400

300 (g) Weight

200

100 See Figure 5.

0 0 20 40 60 80 100 120 140 Age (days)

Figure 4. Chick weight after hatching

Chicks that were artificially incubated weighed on average 112 grams on hatching. Note that chicks were weighed regularly to develop growth data and this activity may have had a negative impact on development, based on their general flighty disposition and thus should be taken into account when considering the growth data. Chicks were observed to lose a couple of grams post hatching and then to gain approximately 2 -3 grams per day over the next two weeks (Figures 4 and 5). An average adult weight is two kilograms.

Average chick weight after hatching

130.00 125.00 120.00 115.00

Weight (g) Weight 110.00 105.00 0 1 2 3 4 5 6 7 8 9 Days after hatching

Figure 5. Chick weight after hatching

Overall 62% of chicks that hatched survived for eventual release. A peak in egg laying and chick production occurred in the summer of 1999 /2000. Losses of chicks can be attributed to a range of causes, with most losses recorded in the first month of life. These causes include: Limb and wing trauma sustained in the aviary Nutritional secondary hyperparathyroidism (metabolic bone disease), resulting in bone, foot or beak abnormalities that would impact chick survival on release. This condition arises when juvenile birds are raised on diets with an inappropriate calcium / phosphorous ratio

79 Predation or trauma by eagles, goannas, adult conspecifics Fungal pneumonia - aspergillosis and mucormycosis respiratory infections. An outbreak of fungal pneumonia occurred following the use of steam-processed bulk Melaleuca leaf for nesting material, combined with extremely dry environmental conditions that assisted the fungal spores to be dispersed in the air. Aspergillosis is normally a very rare condition in the arid environment of Dubbo, although occasional cases have occurred in individuals Coccidiosis – this occurs sporadically in malleefowl under four weeks of age Erysipelas – this is caused by Erysipelothrix rhusiopathiae, and the condition was observed in birds after periods of heavy rain. Vaccination has been used to treat birds; however the efficacy of the vaccination remains unknown (Blyde & Woods 1999)

Malleefowl Release Program With Department of Environment and Climate Change

Malleefowl chicks are generally released to the wild at between six months and 10 months of age. There have been several older individuals (between two and 10 years old) that have also been released. Birds are maintained in their enclosure crèches until a release is co-ordinated. The birds selected for release are usually caught up the evening before, given a veterinary examination and then individually boxed in large cardboard fruit boxes (540 mm x 340 mm x 300 mm high) that are then stored in the veterinary centre. Very early the next morning the birds are driven approximately fours hours to the appointed release site. The release sites have been State Nature Reserves that are managed by the Department of Environment and Climate Change. Both Yathong and Nombinnie Reserves contain large areas of suitable mallee habitat that have undergone highly successful, extensive long term fox baiting programs and already contain existing Malleefowl populations that are aerially monitored by the Department of Environment and Climate Change.

The release program commenced in 1989 and built to a peak in 2000. In all, 98 chicks originally collected as wild eggs have been returned to both Yalgogrin and Yathong Nature Reserves. A further 368 captive bred chicks have been released at these two sites and 90 chicks have been released at Nombinnie/ Round Hill Nature Reserve. In the years that releases have not been planned to occur the breeding pairs have been retained and eggs removed from mounds as a means of controlling breeding.

In 2004 Western Plains Zoo was approached by Charles Sturt University, Wagga Wagga, to assist with a proposed study by Chris Coombes involving monitoring the survival and dispersal behaviour of chicks at a release site. An initial pilot study occurred at the zoo, investigating radio tracking attachment techniques and potential impact on Malleefowl behaviour. This study was followed by the actual chick release and monitoring.

Table 1. History of Malleefowl releases from Western Plains Zoo animal records

80 Year # released from # released from Total # captive hatched captive bred released (ie wild eggs) (ie captive bred birds) Site

1989 9 (TZ incubated) 9 Yalgogrin 1990 54 54 Yathong 1993 27 1 28 Yathong 1994 9 9 Yathong 1996 4 22 26 Yathong 1997 11 11 Yathong 1998 13 13 Yathong 1999 4 74 78 Yathong 2000 97 97 Yathong 2001 92 92 Yathong 2002 12 12 Yathong 2003 37 37 Yathong 2005 56 56 Nombinnie 2006 34 34 Nombinnie Total 556

The Future

The Zoological Parks Board of New South Wales, along with Zoos South Australia, financially supported the development of the National Recovery Plan for Malleefowl and Western Plains Zoo will continue to support the recovery efforts in collaboration with the Department of Environment and Climate Change through maintaining a breeding group of Malleefowl. The current collection includes nine adults hatched in 1989 from eggs originally collected from the wild at Yalgogrin. With these birds in their 18th year, recruitment of F1 generation birds to maintain a breeding nucleus is a priority. Western Plains Zoo has made a major commitment in resources to Malleefowl conservation through the release program and recognises that managing their captive population and breeding program is important for maintaining husbandry skills and providing opportunities for research into Malleefowl that can inform the long term recovery of the species. A husbandry manual is expected to be available by the end of 2007 following circulation of a draft to the Recovery Team and captive holders.

Acknowledgements

Thanks are extended to Judith Denby, Hayley Findlay and Dr Rebecca Spindler for their advice and contributions, and also to the Zoological Parks Board of New South Wales for their support in enabling attendance at the 2007 Malleefowl Forum at Katanning, Western Australia.

Reference

Blyde, R. & Woods, R. (1999), Eriysipelas in malleefowl. Australian Veterinary Journal 77:434-5.

81

11. CONSERVATION GENETICS OF MALLEEFOWL

Raoul Mulder1, Peter Dunn2 and Steve Donnellan3 1 Department of Zoology, University of Melbourne, Parkville, 3010. 2 University of Wisconsin, Milwaukee, Wisconsin, 53201, USA. 3 South Australian Museum, North Terrace, Adelaide, 5000, SA.

Abstract

Defining appropriate genetic units for the management of Malleefowl is one of the key objectives of the Malleefowl Recovery Plan. We recently obtained funding from the Australian Research Council's Linkage Grants Scheme to examine genetic variation in populations of Malleefowl throughout their range, estimate genetic diversity and isolation, define biologically relevant management units, and assess the risk of extinction. The information will enable us to develop management approaches that incorporate genetic considerations when considering strategies for minimised extinction risk. The project is a collaboration between the Universities of Melbourne and Wisconsin (USA), the South Australian Museum, the South Australian Nature Foundation, the Victorian Department of Sustainability and Environment and the Malleefowl Preservation Group, and will run from 2007 to 2011. We are keen to encourage collaboration and participation from additional participants, in order to maximise the quantity and diversity of genetic material that will be available for analysis.

Introduction

Over time we have become more and more interested in using genetic tools to study either animal behaviour or populations of animals. We have worked mainly on mating systems of birds and recently on conservation genetics of bird species. Malleefowl seem to be one of those groups that are crying out for some sort of genetic study and doing genetic work is a priority that has been identified in the Recovery Plan from the very start. What we want to do today is to advertise what the project is about and to appeal to you to get involved where possible in the project.

Malleefowl Populations Today

As we all know, one of the big problems to do with the conservation of Malleefowl is that their range has contracted and particularly that contraction has had a consequence that populations of Malleefowl are small now and also very isolated. We know that small populations, composed of few individuals, are likely to be vulnerable to reduced genetic variation. That is a problem because inbreeding with closely related individuals, the whole population mating together, can lead to weakened offspring and the population can be weakened through random loss of allees, which happens over time in any small population. The problem is trying to work out whether or not this is an issue for Malleefowl because we know nothing about genetic variation in Malleefowl populations. The aim of this study, we define, as addressing that gap in our knowledge and trying to contribute something that will be useful for the future national conservation of Malleefowl. So although we are new faces at this particular Forum, we hope very much that by the time the next Forum rolls along in three years time, we shall have been able to do a lot of work on this genetic variation, and will be able to discuss some of these conservation issues with a bit more information about genetics.

So why are these genetic data useful, why are they, we would say, essential to understand what the current status of Malleefowl is? Just to give you some examples of how knowledge of genetic structure could be useful, it will indicate how isolated, for example, populations are from each other. Now we know that populations are geographically isolated and how far they are from each other, but that does not show us how genetically different they are. Genetic structure will allow us to tell how genetically isolated those populations are from each other. We also know that we have reasonably good population sizes for some locations, but just because we have population sizes, that does not actually tell us what the effective population size is, nor does it tell us how many individuals could effectively breed in the population. The genetic aspect will provide us with a window to try to find out whether or not small populations

82 are genetically viable. All this should add to management options. We have talked about examples where in particular locations numbers are going down and where one management option, for example, might be to reintroduce birds from other areas, or eggs from other areas. Now if we do that without any understanding of what the genetic profile is of the birds we are going to reintroduce, we are flying blind when it comes to addressing that genetic problem unless we know whether or not the birds we are going to introduce are genetically very different, and that might be one of our aims, or genetically very similar and that might be an alternative management strategy. In terms of population management genetic information is likely to be useful in captive breeding programs, in pointing out how we should insert habitat corridors, etc.

There are a few things that make Malleefowl interesting in relation to genetic questions in addition to those general sorts of issues. The recent set of population viability analyses conducted on a whole range of threatened species, 19 species in all, estimated that if you get small populations, populations that number less than a thousand individuals, and there are many Malleefowl populations in that boat, on average it takes about 43 years for those threatened populations to go extinct and that is not counting the effects of potential threats, like foxes and other predators, issues of rainfall and climate. Just by virtue of being a small population, 43 years on average is about what it takes for the population to disappear. That is if there is no inbreeding; if there is inbreeding then that reduces by about 15% to 37 years. Malleefowl have strong site fidelity, they tend to stay in one location. Now the young obviously disperse but that site fidelity does suggest restricted dispersal, so the males have restricted gene flow, and maybe you get quite genetically structured populations that are a bit different between one site and another. The other thing that is interesting to me is that we assume that Malleefowl are monogamous. Having studied the mating systems of a number of different bird species that were all presumed to be monogamous, including things like black swans and fairy wrens, we have to say we are monogamy sceptics. But we think it would be great to use some of these genetic tools to actually find out exactly what the mating system of Malleefowl is, and to show that they are genetically monogamous. Or conversely, If they are not entirely monogamous, but certain males are managing to mate with more than one female, the fact that males are not monogamous is also going to have genetic consequences, because its going to increase genetic drift.

The Project

We are not sure if you are familiar with Australian Research Council Linkage Projects, so we will tell you a little bit about what they are and how they operate. The idea is that these are a set of grants that are set up to try and support collaboration between university and industry partners, and the industry partner can be defined as anything from a Non-Government Organisation, a state organisation, or a community organisation, where there is some kind of mutual benefit because both parties are trying to work towards a particular goal. The idea is that the industry partners, as they are called, make a contribution to the project, and that contribution can be in terms of funding, cash for the project, or an in-kind contribution, supporting the project by helping in various ways, and that the Australian Research Council also provides funding, so that it becomes a value-added set up. This project involves collaboration between three of us, Steve Donnellan at the South Australian Museum, Raoul Mulder, at the University of Melbourne, and Peter Dunn at the University of Wisconsin. You might ask why is the University of Wisconsin, involved in Malleefowl research. Well, the reason is that Peter Dunn has a lot of experience working on prairie chickens in the United States, and prairie chickens are an interesting conservation story. It has a few parallels with the Malleefowl situation, although it is a bit more dire in the case of the prairie chicken, the Illinois population was down to about 50 individuals, Peter led a study that was conducted as an exact parallel to this genetic study, on the prairie chicken, and established priorities for relocations to the point where the prairie chicken in Illinois is now doing really well. That is a good sign and his work is a model for what we might be able to do with Malleefowl. Steve Donnellan at the South Australian Museum has been interested all along in this project, and has started to accumulate a range of genetic material for reference. The industry partners involved so far, and we‘d love to have more partners involved in the project, the South Australian Museum, the SA Nature Foundation, the state Department of Environment in Victoria and the Malleefowl Preservation Group.

83

Let us give you an overview of the four broad goals that the project has. Obviously the first thing we want to do is try and estimate genetic variation across Malleefowl populations in Australia, and our hope is that this will be a national project. It will be a shame if we have to try to focus it in one or other state. That would ignore the fact that Malleefowl do not respect state boundaries, and we want to be able to say something about genetic variation across the range of the species. We shall be using a couple of different genetic markers for estimating mitochondrial or nuclear DNA to find out more about genetic variation. Those markers are useful as a way of being able to work out mating system of Malleefowl. As we mentioned before we should like to find out more about the genetic mating system of Malleefowl. With that information we should then set up a whole series of models that are asked to look at what are effective populations of Malleefowl, how big the populations should be and how big the gene flow might be between different populations; and how isolated are they from each other genetically. And from there, what we can do is to find biologically relevant management protocols based on that kind of structure. And we shall also try to model at what rate populations might be going extinct, how likely they are to disappear. One of the nice things is that there are Malleefowl samples in a whole range of different museums around the world and we are slowly and systematically going through those collections to get as many samples as we can. That will give us some idea of what genetic variation might have existed in historic populations of Malleefowl. When we look at contemporary populations it is sometimes hard to say whether or not the level of genetic variation we find is relatively low or relatively high. If we can compare what we find in contemporary populations with birds that have been found more than fifty or a hundred years or even longer ago, then that too will give us a window on whether or not there has been a loss of genetic variation in particular parts for a very long time. In terms of our sampling what we have at the moment is what has been collected at the South Australian Museum, about 210 tissue samples, that come from 34 sites, a couple in New South Wales, mostly sites in South Australia, a few in Victoria and Western Australia. Obviously we are hoping that this is a tiny sub-set of the range that we are going to find. One of the useful things about the monitoring network that has been set up, is that it is an important resource in terms of trying to get genetic samples from many different sites across the range of Malleefowl. If we had done the project five years ago we do not think we should have been in nearly the same position in terms of trying to find out the genetic variation that we are in now with the monitoring set up. If you think that you might be able to be involved in helping to gather samples then we would like to hear from you.

The great thing about doing DNA analyses is that there are a lot of very useful samples that we can get from material that is easy to collect and not invasive. For example feathers that are found on mounds are a great source of DNA. We have talked to a number of people at the meeting already who have revealed to me that they have treasure troves of bags of feathers, individually marked and categorised by mounds; that is a gold-mine of information. Feathers are easy to collect and great for DNA analysis if just wrapped in an envelope. Obviously where people are involved in captive breeding programs, for example zoos, it might be possible to get a blood sample from a captive bird or chick that can be stored in alcohol. And even road-kills or decayed left over fox kills that you might find, still can provide us with useful DNA if you get a bit of muscle from them. So it is early days; we literally only just found out that we have got funding for this project and we shall be working on getting a systematic approach to the collection of samples. That involves sending out information sheets that tell people exactly what we want, to try to get information, to integrate into the monitoring system, and to set up kits for people to use for the collection of samples. The first step for you is to get in touch with us and we shall try and get as many people as possible too. We are very open to any ideas, suggestions or advice that people have, and to receive sample collections for the study.

84

12. ARE WE DOING ENOUGH? A REVIEW OF THE CONSERVATION STATUS OF MEGAPODES AND THE IMPLEMETATION OF THEIR ACTION PLANS.

J. Ross Sinclair Leibniz Institute for Zoo and Wildlife Research, Postfach 601103, D-10252 Berlin, GERMANY. Wildlife Conservation Society, Global Conservation Programs, 2300 Southern Boulevard, Bronx, NY 10460, USA. e-mail: [email protected]

Abstract

Over half of the megapode (Aves: Megapodiidae) species that once existed are now extinct. Almost half of the 22 extant species are classified as threatened with extinction, and all threatened species are declining in numbers. This vulnerability among megapodes is shaped by four characteristics of the family: large ‗nests‘, large clutches, large eggs, and independent young. Conspicuous nests with many nutritious eggs leads to over-harvesting of eggs. Mound-builders are trapped and hunted while spending long periods at their mounds. Introduced predators kill birds and eggs, with dispersing chicks particularly vulnerable. Megapode habitat is being degraded by industries such as logging, oil palm plantations and farming. To facilitate the conservation of threatened megapodes, the World Conservation Union has produced two Megapode Conservation Action Plans that assess the conservation status of species and detail a 5-year plan of action. Of the 37 conservation actions recommended in the Megapode Conservation Action Plans only 24% were fully implemented. Case studies show when actions are implemented they produced positive conservation outcomes. The deterioration of nesting fields of the maleo (Macrocephalon maleo) have been reversed; closed seasons have been shown to dramatically increase the production of chicks in the Melanesian megapode (Megapodius eremita), and research into mound-building megapodes in New Guinea has produced management prescriptions that are useful to local people. Despite these few successes, not enough work is being done to conserve megapodes. If the actions identified in the Megapode Conservation Action Plans are not implemented, more species will be added to the list of extinct megapodes.

Introduction

Megapodes (Aves: Megapodiidae) are prone to extinction. Over half of the megapode species that existed when people first colonised Oceania are now extinct (Steadman 1999; Steadman 2006), and significantly more species in the family are threatened with extinction than would be expected by chance (Bennett and Owens 1997). There were perhaps 45-55 species of megapodes when humans arrived in the Pacific (Steadman 2006); there are now 22 species (Jones et al. 1995), nine of which are classified as threatened with extinction and eight of which are apparently declining (IUCN 2006, Table 1). Throughout the range of the family - from Niuafo‘ou Island in the east to the Nicobar Islands in the west - there is cause for concern (Jones et al. 1995): megapode habitat is being degraded, megapode eggs are over- harvested, and megapode adults and chicks are killed by introduced predators (Dekker et al. 2000) and humans. The survival of many species is now beyond what nature alone can do; active management is needed to protect some megapode populations and species.

In this paper I will first give an overview of the distribution, behaviour and ecology of the megapodes. I will then outline the conservation status and threats to species, and review the two World Conservation Union (IUCN) Conservation Action Plans that have been published for megapodes to determine how successful we have been at implementing conservation actions. I will then end by outlining three case studies

(a) Micronesian megapode (EN) 4 Moluccan megapode (VU) LC 1 Bruijnii brush-turkey (VU) NT 6 Biak megapode (VU) LC NT1 85 5 1 1,3, 8-10 LC NT LC NT1 Nicobar 2 1 NT NT megapode (VU) LC7

Figure 1. The distribution and status of megapodes (Aves : Megapodiidae). (a) The present distribution of the family (dotted line) and the conservation status of species. The (IUCN 2006) risk categories are given in Table 1: LC1 = wattled brush-turkey (Aepypodius arfakianus); LC2 = Australian brush-turkey (Alectura lathami); LC3 = New Guinea megapode (M. decollatus); NT1 = Sula megapode (M. bernsteinii); LC4 = Tabon megapode (M. cumingii); LC5 = Melanesian megapode (M. eremita); LC6 = dusky megapode (M. freycinet); LC7 = orange-footed megapode (M. reinwardt); NT2 = Tanimbar megapode (M. tenimberensis); LC8 = red-billed brush-turkey (Talegalla cuvieri); LC9 = black-billed brush-turkey (T. fuscirostris); LC10 = brown-collared brush- turkey (T. jobiensis). (b) The present distribution of megapodes (dotted line) and past distribution (dashed lined) in the SW Pacific showing extant and extinct species: EX1 = Megapodius undescribed sp. A; EX2 = M. undescribed sp. B; EX3 = M. undescribed sp. C; EX4 = M. undescribed sp. D; EX5 = M. undescribed sp. E; EX6 = M. undescribed sp. F; EX7 = Unknown genus undescribed sp.; EX8 = M. amissus; EX9 = M. altirostris; EX10 = M. molistuctor; EX11 = M. alimentum; EX12 = Sylviornis neocaledoniae. (Data from Jones et al. (1995); IUCN (2006); Steadman (2006). that highlight conservation management of megapodes: conservation of the maleo1 (Macrocephalon maleo) in Sulawesi, the experimental management of nesting fields on islands in northern Melanesia, and the management of mound-building megapodes in New Guinea.

Distribution and biology of Megapodes

1 I follow Jones et al. (1995) for the names of megapodes, except Megapodius cumingii, which I call the Tabon megapode.

86

Megapodes are of Gondwana origin and probably evolved on the Australian Plate before dispersing east into Oceania and west though the Indonesian Archipelago (Dekker 2007). Their present range is restricted to the Indo-Pacific region (Figure 1), with all but the Nicobar megapode (Megapodius nicobariensis) and some populations of the Tabon megapode (M. cumingii) occurring east of Wallace‘s Line (Jones et al. 1995). Wallace‘s Line is a zoogeographical boundary between fauna with affinities to Asia and those of Australasia (Whitmore 1982). Two theories have been proposed for the western limit of the megapodes: competitive exclusion by the pheasants (Phasianidae) (Olson 1980) or predation by cats (Felidae) and civets (Viverridae) (Dekker 1989b), while the eastern limit of Samoa (Steadman 2006) is probably due to the increasing distances among islands being a physical barrier to dispersal.

Most genera of megapodes have a restricted range with the monotypic genera Alectura and Leipoa restricted to Australia, Macrocephalon to Sulawesi and Eulipoa to the Moluccas Islands, while two species of Aepypodius and three species of Talegalla are found only on New Guinea and surrounding islands. Conversely, the polytypic ‗super-genus‘ Megapodius has 13 species distributed across the range of the family.

Habitat use and diet

Megapodes typically inhabit tropical rainforests, with only the atypical malleefowl (Leipoa ocellata) inhabiting arid regions of southern and Western Australia (Frith 1956a). They occur in habitat ranging from primary forests through to early secondary regrowth, gardens and even monocultures like coconut plantations (e.g., Megapodius spp.)( Broome et al. 1984; Sinclair et al. 1999), although less disturbed areas may be required for successful breeding (Sinclair et al. 2002; Gorog et al. 2005). It seems that in general members of the genus Megapodius are most able to persist in areas that are degraded, and in a study of three sympatric genera they were found to use a wider range of habitat for breeding than either Talegalla or Aepypodius (Sinclair 2002).

Megapodes are generalist omnivores taking a wide range of plant and animal material by foraging on the ground (Jones et al. 1995).

Heat sources and incubation sites

Megapodes are different from all other bird families in using heat from the environment to incubate their eggs and in having young that are entirely independent after hatching (Frith 1956a). This freedom from brooding has shaped much of the behaviour and ecology of the family (Jones & Birks 1992), which in turn has contributed to their vulnerability and desperate conservation status.

Megapodes use three sources of heat to incubate their eggs: geothermal heat, solar radiation and heat generated by the microbial decomposition of organic material (Frith 1956a); in the case of the malleefowl the latter two heat sources are used in tandem at mounds (Frith 1956b) but this is atypical of mound-builders in the family (Jones et al. 1995).

The most common incubation method is mound-building where piles of organic material and soil are raked together on the forest floor. These mounds require specific conditions for successful incubation (e.g. a critical mass of material before the mound becomes a stable homeotherm (Seymour & Bradford 1992)) and are located non-randomly (e.g., under a closed canopy that may reduce temperature fluctuations and desiccation, Jones 1987; Sinclair 2002). In addition to building mounds, some megapode species burrow into existing sources of microbial heat, like the decaying roots of dead trees, and these sites are also located non- randomly (e.g., at very large trees that may provide a sufficient biomass of dead roots to generate enough heat for incubation, Sinclair et al. 2002). Sites for both mounds and burrows into decaying roots are located in forest that is generally less disturbed than random points in the same habitat (Jones 1987; Sinclair 2002; Sinclair et al. 2002).

87 Burrows in dead trees contain few eggs and few egg chambers, and pairs may use more than one such incubation site (Sinclair et al. 1999). Mounds, on the other hand, generally have one chamber that can contain many eggs (Jones 1988a; Sinclair 2001b) and usually only one mound is used by a male or pair (Jones 1990a; Birks 1997). Both mounds and burrows in the decaying roots of dead trees are widely scattered, and may be in the core area of the home range of territorial species (e.g., Crome & Brown 1979).

Geothermal or solar burrows differ in many respects to mound sites. Rather than being dispersed like microbial sites, they are highly clustered at localised sites in the form of large colonies of ‗nests‘2, often called ‗nesting fields‘. Within these fields, however, burrows may be located non-randomly. For example, geothermal sites of the Melanesian megapode (M. eremita) on Lihir Island, Papua New Guinea differed significantly from random points in having a more closed canopy and a higher proportion of thicket species in the vicinity (Samson 2007). In addition, at these sites the conditions in burrows were also different from random points, with smaller substrate particles in burrows and higher soil temperature.

Mating systems and breeding behaviour

Without the energetic input of brooding, females of all species of megapode seem to pursue a similar strategy of maximising their clutch size (Jones 1994): they lay an unusually large clutch of 10-20 eggs over an extended period of about 3 months (Jones et al. 1995). Males, on the other hand, pursue one of two strategies, they either monopolise the reproductive output of a mound (e.g., the resource-defence polygyny of the Australian brush-turkey, Alectura lathami) or that of a female (e.g., the female-defence monogamy of the orange- footed megapode, M. reinwardt) (Jones 1992). In the former system, the male guards the mound so other megapodes do not have access to it, while unattended females may copulate with more than one male (Jones 1990a). In a study of the Australian brush-turkey 23% of eggs in mounds were fertilised by males other than the mound ‗owner‘ (Birks 1997). In the latter system, the male and female are most often together as the male ‗guards‘ the female, so incubation sites are not defended and are often used by more than one pair (Crome & Brown 1979; Sankaran & Sivakumar 1999) or more than one species (Sinclair 2002). As with so many other aspects of megapode biology, the exception to this is the malleefowl, where within a monogamous mating system (c.f., Weathers et al. 1990) the male stays close to the mound and does not ‗guard‘ the female (Frith 1959), leading to this being described as resource-defence monogamy (Jones et al. 1995).

In the brush-turkey genera Aepypodius and Alectura the male alone builds and maintains the mound and the female comes to copulate and lay only (Jones 1990b; Table 1. The risk categories, population trends, threats and activities detailed in IUCN Conservation Action Plans for megapode species (Aves : Megapodiidae) that are threatened or near threatened with extinction. Red Risk categories are Endangered (EN), Vulnerable (VU), Near Threatened (NT), and Least Concern (LC) and threats are egg harvesting (E), hunting (H), introduced predators (P), habitat degradation (D), natural disasters (N), fire (F), grazing by introduced herbivores (G), disturbance at nesting grounds (I), natural predators (B), competition with introduced animals (C), tourism (T), and invasive vegetation (V). The trends in megapode populations are either declining (), stable () or not quantified (NQ). Activities in action plans were either implemented (I), partially implemented (P) or not implemented (N). Two species have recovery plans so did not have actions listed in the 2000-2004 action plan. Risk1 Trend IUCN Conservation Action Plan Species name Threats3 19952 20061 (1995-1999)2 (2000-2004)3 Protect fields (P) Protect fields (P) Maleo EN   Awareness (P) Awareness (P) (Macrocephalon maleo) E, D, I, B, V Survey (I) Harvest plan (N) Micronesian megapode EN (Megapodius) E, H, P, N, G, ?  Monitoring (I) (Recovery plan) laperouse) T

2Megapode nesting fields are commonly referred to as ‘communal’ but share little with avian communal nesting systems in the classic sense and are more adequately described as ‘colonial’.

88 Monitoring (N) Monitoring (N) Polynesian megapode EN Awareness (P)   Awareness (P) (M. pritchardii) E, H, P, N, C Translocation (N) Translocation (N) Research (N) Bruijn's brush-turkey EN4 Survey (I) NQ NQ Survey (P) (Aepypodius bruijnii) H, F5 Manage. plan (N) Monitoring (I) Malleefowl VU   Research (I) (Recovery plan) (Leipoa ocellata) H, P, F, G Manage. plan (I) Protect fields (P) Moluccan megapode VU   Survey (P) Awareness (N) (Eulipoa wallacei) E, D, I, B Harvest plan (P) Biak megapode VU Survey (N)   Survey (N) (M. geelvinkianus) E, H, P, D Manage. plan (N) Protect habitat (N) Survey (I) Nicobar megapode VU Awareness (P)   Research (I) (M. nicobariensis) E , D, H Monitoring (I)

Research (P)

Vanuatu megapode VU Harvest plan (P) /  Survey (P) (M. layardi) E, H, P, D, N, Survey (P) F Tanimbar megapode NT 2 /  Survey (N) None (M. tenimberensis) E,H, P, D Sula megapode NT 2 /  None None (M. bernsteinii) E,H, P, D E, H P, D 7/10 10/10 Overall (spp./spp.) I = 7, P = 6, N = 5 I = 2, P = 9, N = 8 9/11 7/11   1IUCN (2006) 2Dekker et al. (1995b) 3Dekker et al. (2000) 4 Up-graded from Vulnerable to Endangered in 2008 (BirdLife International and IUCN in prep.), while all other species remain unchanged. 5Mauro 2006.

Birks 1997). For all other species of mound-builders the male and the female are both involved in building or maintaining the mound or both (Jones et al. 1995).

The use of some mounds by more than one species has been described for megapodes (e.g., Dwyer 1981; Sinclair 2002; Rand & Gilliard 1967) and assumed to be a form of brood parasitism (Diamond 1983; del Hoyo et al. 1992), although it is not clear that there is a cost to the ‗host‘ and if this relationship is best described as parasitic (Sinclair 2000a).

Hatching success and survival

Megapodes have higher hatching success (about 65%) than most tropical birds (about 30%) because their large ‗nests‘ and asynchronous laying and hatching reduce the likelihood that predators can take the entire clutch (for a review see Sinclair in prep.). This is counterbalanced by megapode chicks suffering extremely high mortality with very few chicks surviving to 1 month (e.g., Priddel and Wheeler 1990; Göth & Vogel 2002), with almost no chicks surviving in the presence of introduced mammalian predators (a review in Sinclair in prep.; Figure 3).

Megapodes are long-lived birds with captive individuals known to have lived for up to 20 years (D. Bruning pers. comm. in Dekker & Wattle 1987). That said, the longevity of birds in the wild is not known and is likely to be considerable less (e.g., in one study all 12 New

89 Guinea megapodes (M. decollatus) marked as adults all had disappeared from their mounds within 5 years; RS unpubl. data).

Threats to megapodes

According to the IUCN (Dekker et al. 1995b; Dekker et al. 2000; IUCN 2006) the threats faced by8 megapodes range from predators of eggs and young to disturbance at nesting sites1 from things like tourism (Table 1, Figure 2). 7 0.9 0.8

6 0.7

5 0.6

4 0.5 Rate azard

0.4 No. chicks No. 3 0.3 2

0.2 h Cumulative 1 0.1 0 0 6 18 30 42 54 66 78 90 102 114 126 138 150

Time since release /h

Figure 2. Cumulative hazard rates (probability of mortality) for Melanesian megapode (Megapodius eremita) chicks dispersing from nesting fields of the megapode on Simbo Island. Bars are the number of chicks killed by introduced mammalian predators (hollow), died by other causes (shaded) and missing (hatched). Chicks faced a 98% risk of being preyed upon by an introduced predator in the first week after emergence and a significantly higher risk from cats (Felis silvestris catus) and dogs (Canis lupus familiaris) (hollow circles) than from all other causes combined (filled circles). (Reproduced with permission from Sinclair (in prep.)

The most serious threats to megapodes are in many ways shaped by four interesting characteristics of the family: (1) large ‗nests‘ - incubation mounds are the largest nesting structures built by birds (Jones 1989); (2) nesting fields can cover several hectares (e.g., Broome et al. 1984); (3) a large ‗clutch‘ of big yolk-filled eggs - eggs weigh 75-232 g (Dekker & Wattle 1987) and have among the highest yolk contents of any avian egg (about 50-70% by weight, Brom & Dekker 1992); and (4) independent young.

Hunting and harvesting eggs

Hunting and harvesting of eggs are the most common threats cited for megapodes with a conservation status (Table 1). Megapode ‗nests‘ are large and conspicuous and are therefore easily found by humans and other predators. Furthermore, because megapode eggs are large, nutritious, and often in great numbers at incubation sites, harvesting is a worthwhile expenditure of time and energy for a predator. As a result, eggs of all species of megapodes are harvested by humans, except perhaps Bruijn's brush-turkey (Aepypodius bruijnii, Mauro 2006) and the three species occurring in Australia (Dekker et al. 1995b). For the colonial nesters, harvesting can be highly efficient with almost all eggs collected (MacKinnon 1978; Heij 1997; Sinclair 2001a) and unsustainable harvesting has led to abandonment of nesting fields and local extinctions (Argeloo and Dekker 1996; Baker & Butchart 2000). For example, in the Tangkoko-DuaSudara Nature Reserve in Sulawesi a nesting field disappeared within 6 years of a village being established nearby (MacKinnon 1981), and continued over-harvesting of eggs at the remaining fields in the reserve is believed to be responsible for sending the local population to the verge of extinction (O'Brien & Kinnaird 1996). Furthermore, because many species spend long periods of time at their incubation sites (Jones et al. 1995), snaring and hunting of adult birds is widespread (Baker & Butchart 2000; Mack & West 2005;

90 Sivakumar 2007), although some communities that collect eggs do not hunt megapodes (Sankaran 1995; Sinclair 2001a).

Introduced predators and habitat degradation

The next most common threats cited for megapodes are introduced predators and habitat degradation (Table 1). Megapodes and their eggs are preyed upon by a bewildering array of introduced predators; both eggs and birds are taken by pigs (Sus spp.)(Kisokau 1976; MacKinnon 1981a; Heij et al. 1997), dogs (Canis lupus familiaris)(Kisokau 1976; Jones 1988b; Benshemesh 1992; Heij et al. 1997), foxes (Vulpes vulpes)(Frith 1959; Brickhill 1986; Priddel & Wheeler 1994; Priddel & Wheeler 1996; Göth & Vogel 2002), and cats (Felis silvestris catus)(Weir 1973; Jones 1988b; Benshemesh 1992; Priddel & Wheeler 1994; Heij et al. 1997; Göth & Vogel 2002). Although in a 5-year study of Malleefowl, foxes raided 48% of nests (n = 71) and took 37% of eggs (n = 1094) (Frith 1959), introduced predators are not generally a major threat to the hatching success of megapode eggs for the reasons detailed above. These predators are, however, a serious threat to the birds. The natural ranges of megapodes and mammalian predators do not widely overlap and the vulnerability of the former to the latter has been used to explain this allopatry (Dekker 1989a).

Growing human populations in developing countries modify increasing amounts of habitat, but these subsistence activities pale in comparison to rampant logging and oil palm plantations in the tropical lowlands, mining in the mountains of places like New Guinea, and land clearance for industrial agriculture in places like Australia.

The remaining threats are not as widely applicable across the family, but are often highly significant to individual species or populations.

Catastrophes, weeds and tourism

Catastrophes and natural disasters poise a serious threat to species of megapode with small populations or highly restricted distributions. For example, it is estimated that the 2006 Asian tsunami wiped out nearly 70% of the Vulnerable Nicobar megapode (Sivakumar 2007). Fire threatens the malleefowl (Benshemesh 1999) and Bruijn's brush-turkey (Mauro 2006), severe weather events threaten island populations of the Vanuatu (M. layardi, Foster 1999) and Micronesian megapodes (M. laperouse) (Dekker et al. 2000), while volcanic eruptions pose a threat to the Polynesian megapode (M. pritchardii, Dekker et al. 1995b). Small populations of some species are also considered to be at threat from natural predators, such as varanids preying on eggs in remnant nesting fields (e.g., the maleo, MacKinnon 1981) and Moluccan megapode (Eulipoa wallacei, Heij 1997). Invasive vegetation is covering some nesting fields and reduces the area available for nesting (e.g., the maleo, MacKinnon 1981; Gorog et al. 2005). Disturbance from tourism is also cited as a threat (e.g., Micronesian megapode, Stinton & Glass 1992); one maleo nesting field may have been abandoned due to disturbance from tourism (Gorog et al. 2005).

The conservation status of megapodes

The result of the multitude of threats facing megapodes is that the family is at risk. Four species of megapodes are classified as Endangered, five as Threatened and two as Near Threatened (Table 1). Among these species, declines seem to be continuing unabated. When the first IUCN Megapode Conservation Action Plan was published in 1995 (Dekker et al. 1995b), seven of 10 species were considered as either declining (4 spp.) or stable/declining (3 spp.), with only three species considered stable (IUCN 2006; Table 1). The seemingly perilous situation in 1995 had deteriorated to the point where by 2008 all megapodes that are threatened or Near-Threatened for which there are data are declining (Table 1); at least half of all extant megapode species are in decline (IUCN 2006). Furthermore, local people in Papua New Guinea and the Solomons Islands report recent declines in the numbers of Talegalla spp., wattled brush-turkeys (Aepypodius arfakianus) and Melanesian megapodes (Sinclair unpubl. data), all species that are classified as Least

91 Concern and considered widespread and abundant (IUCN 2006). Declines in such ‗safe‘ species are likely to be common in the family, meaning several further species will require reclassification as threatened if sustainable management is not implemented.

From 2000-2008 three species of megapode have been reclassified: the Polynesian megapode was downgraded from Critically Endangered to Endangered, while both the maleo and Bruijn's brush-turkey were upgraded from Vulnerable to Endangered (IUCN 2006) (see species accounts below for details).

Megapode species threatened with extinction

In this section I will give a brief overview of each species of megapode that is threatened with extinction, detailing the estimated size of the global population, the main threats to populations, and their conservation status. I will not discuss the two threatened species for which there are recovery plans because they are not dealt with by the Megapode Conservation Action Plan. (For details of the recovery plans see Benshemesh (1999; 2000) for malleefowl and USFWS (1998) for the Micronesian megapode.)

Polynesian megapode – Endangered

The Polynesian megapode lays in geothermal areas and once occurred on many island in Tonga (Steadman 1993; 2006). At the beginning of both Megapode Conservation Action Plans it was confirmed to have a population of less than 1000 occurring on only Niuafo‘ou Island, a small island in northern Tonga (Göth and Vogel 1995; 1999). The remnant population on Niuafo‘ou Island was threatened by volcanism, hunting and competition from pigs (Table 1), although the harvesting of eggs and predation by cats were seen as the main threats (Göth & Vogel 1995). Between 1991-1993 about 100 eggs and chicks were transferred to Late and Fonualei islands. Although anecdotal evidence suggested some birds survived on the islands (e.g., Rinke 1993), there were no formal surveys until 2003. In 2003 a survey of Fonualei Island found a population of 300-500 birds (Watling 2003), significantly reducing the risk to this species and resulting in it being down-graded to Endangered (IUCN 2006).

Maleo – Endangered

The Maleo is endemic to Sulawesi and nearby islands in Indonesia, and lays eggs in geothermal and solar colonies (Jones et al. 1995). The global population is estimated to be 8,000-14,000 (Butchart & Baker 2000) but is declining rapidly (Dekker et al. 2000). Although probably given more attention from conservationists than any megapode outside Australia, the maleo remains in serious trouble. It faces habitat degradation, disturbance at nesting fields, where invasive weeds are also problem, and the over harvesting of eggs (Table 1). Some fields are now so small, predation by natural predators may be contributing to local declines (MacKinnon 1981). There have been several thorough surveys of nesting fields of maleo (i.e., Dekker 1990; Argeloo 1994; Baker & Butchart 2000; Butchart & Baker 2000; Gorog et al. 2005) that have unfortunately shown dramatic declines in the quality and number of nesting fields - only 3% of the 142 fields known are not threatened (Baker 2002) - while wildlife surveys have shown associated declines in the bird population (O'Brien and Kinnaird 1996). Gorog et al. (2005) found nesting fields most likely to be abandoned if they were in coastal areas, near human settlements, or not in protected areas, and connectivity of fields to forest was the best predicator of abandonment.

Bruijn’s brush-turkey – Endangered

Bruijn‘s brush-turkey is a mound builder that is restricted to cloudforest above 620 m on Waigeo Island, west of New Guinea (Mauro 2006). Until recently this species was virtually unknown, with the most recent record from 1938 and no observations in the wild despite at least 15 expeditions to find it (Jones et al. 1995; Dekker 2000). Given its very restricted distribution and a population assumed to be less than 2,500, Bruijn‘s brush-turkey was originally classified as Vulnerable in both Megapode Conservation Action Plans, and hunting and mining were seen as potential threats. Recent surveys (Mauro 2005; 2006) have

92 estimated a global population of less than 1000 and the major threat to its small and fragmented habitat is fire (Table 1), resulting in it being up-graded to Endangered in 2008 (IUCN 2006). See Mauro (this volume) for a detailed description of the threats and status of this species.

Moluccan megapode – Vulnerable

Moluccan megapodes lay their eggs in solar heated soil at colonial sites (Dekker et al. 1995a), with one field on Halmahera Island and another on Haruku Island used by most birds (Dekker et al. 1995b). Unlike other megapodes, this species lays at night, with lunar synchrony in laying; more birds visit nesting fields on bright nights (Baker & Dekker 2000). Moluccan megapodes face habitat degradation, disturbance at nesting fields and high levels of egg predation from natural predators (Table 1), but egg collecting is the major threat to the species (Heij et al. 1997; Dekker et al. 2000). Unlike many areas where traditional control of egg harvests have broken down, this system is still enforced for the Moluccan megapode but may still not prevent over-exploitation of eggs (Argeloo & Dekker 1996). The global population is estimated at about 10,000 and status as Vulnerable (IUCN 2006).

Nicobar megapode – Vulnerable

Nicobar megapodes are mound-builders endemic to the Nicobar Islands, where 90% of incubation sites are found within 100 m of the coast (Sankaran 1995). At least 65% of mounds are used by more than one pair, and a strong hierarchy exist among the pairs that use the mound (Sankaran & Sivakumar 1999). This species is at threat from over-harvesting of eggs, the hunting of birds and habitat degradation (Table 1). On-going declines in the population and habitat fragmentation has led to Nicobar megapode being classified as Vulnerable (IUCN 2006). The population was estimated at 4,000-8,000 (Sankaran 1995), but was reduced by about 70% by the 2004 Asian Tsunami, after which habitat degradation and hunting of adults has increased (Sivakumar 2007).

Biak megapode – Vulnerable

The Biak megapode (M. geelvinkianus) is endemic to Biak and surrounding islands, and is very poorly known (Jones et al. 1995). Although the threats to this species are not documented, they are assumed to be the harvesting of eggs, hunting of birds, introduced predators and habitat degradation (Table 1). These possible threats and the assumption of a small and declining population result in a classification of Vulnerable (IUCN 2006).

Vanuatu megapode – Vulnerable

The Vanuatu megapode is endemic to Vanuatu, and displays considerable behavioural plasticity in using all the heat sources and all the types of incubation sites described for the family (Jones et al. 1995). At colonial nesting sites eggs are heavily harvested (Bowen 1996), and this is considered a major threat (Foster 1999), as is logging (O'Brien et al. 2003). Other threats are hunting, introduced predators, habitat degradation, natural disasters and fire (Table 1). Although it occurs in two protected areas (Kratter et al. 2006), this species is classified as Vulnerable due to a population that is assumed to be small and declining (IUCN 2006); the latter trend being reported by local people (Bowen 1996; Foster 1999; O'Brien et al. 2003).

Conservation action plans

Conservation Action Plans are produced by specialist groups like the Megapode Specialist Group that are comprised of volunteers affiliated to the Species Survival Commission of the IUCN (Fuller et al. 2003). Plans assess ―… the conservation status of species and their habitats, and [specify] conservation priorities.‖ (Dekker et al. 2000, back cover). Action plans contain an overview of the conservation issues facing the group, a summary of their

93 conservation status, accounts for each threatened species with information on range, population, ecology, threats, conservation and targets. They also detail a 5-year plan of action (e.g., see Dekker et al. 2000).

The changing conservation status of species and the implementation of actions can be measured against the plans (Fuller et al. 2003). Although action plans have been criticized (e.g., Collar 1994), a review of three plans - including the 1995-1999 Megapode Conservation Action Plan - found considerable conservation oriented activity had occurred that was related to the recommendations in the plans (Fuller et al. 2003). One criticism of the plans has been their emphasis on surveys and research rather than practical actions. The 1995-1999 Megapode Conservation Action Plan followed this pattern with 72% of recommendations (n = 18) being surveys, research or monitoring (Table 1). Given at this time our lack of knowledge of almost all species was profound (Jones 1999), this emphasis was reasonable. The next Megapode Conservation Action Plan had only 32% of such recommendations (n = 19), indicating that as our knowledge increased, recommendations became more focused on practical actions. Given my experience working on six species of megapodes from four genera in three countries, the recommendations in the two Megapode Conservation Action Plans were both sensible and needed if megapodes are to be conserved. In that sense, my analysis of our success in implementing conservation actions for megapodes over the past 12 years and two Megapode Conservation Action Plans is not a comment on the plans themselves, but on our effectiveness at implementing their recommendations.

Megapode Conservation Action Plans 1995-1999 and 2000-2005

Not enough work has been done to conserve megapodes. Of the 18 conservation actions recommended for 1995-1999 (Dekker et al. 1995b) about a third each were implemented, partially implemented, and not implemented (Table 1). This rate of implementation did not become any better for the period of the second Megapode Conservation Action Plan for 2000-2005 (Dekker et al. 2000), with only two of 19 actions implemented and 10 partially implemented, while six were not implemented (Table 1). Review of the actions in both Megapode Conservation Action Plans suggests that for most, partial implementation is not sufficient to ensure the conservation of the species involved. The most positive outcome over the period was the implementation of recovery plans for two species (i.e., the malleefowl and Micronesian megapode, Dekker et al. 2000).

For the Polynesian megapode, the 1995-1999 Megapode Conservation Action Plan called for research into sustainable harvest of eggs, while both Megapode Conservation Action Plans called for monitoring, translocations of more eggs to islands and a public awareness campaign. Only the later was partially implemented (e.g., Rinke et al. 1993; Table 1).

The first Megapode Conservation Action Plan for maleo called for surveys, which were implemented (Dekker 1990; Argeloo 1994; Baker & Butchart 2000; Butchart & Baker 2000; Baker 2002; Gorog et al. 2005), while the harvesting plan recommended in the second Megapode Conservation Action Plan was not developed. Both Megapode Conservation Action Plans called for protection of nesting fields and awareness campaigns, which were partially implemented (Table 1). Below I will discuss the conservation actions being taken for this species.

Both Megapode Conservation Action Plans for Bruijn‘s brush-turkey called for surveys, which were implemented (Mauro 2005; 2006), while no management plan has yet been developed (Table 1).

The surveys recommended in the first Megapode Conservation Action Plan for the Moluccan megapode were partially implemented (Anonymous 1995; Dekker et al. 1995a; Heij et al. 1997). Several nesting fields of this species are formally managed by local people (Argeloo and Dekker 1996; Heij et al. 1997), so in that sense the protection of nesting fields and harvesting plan recommended in the second Megapode Conservation Action Plan were partially implemented, while an awareness campaign has not been implemented (Table 1).

94 The surveys and research recommended in the first Megapode Conservation Action Plan for the Nicobar megapode were implemented (Sankaran 1995), as was the monitoring called for in the second Megapode Conservation Action Plan (Sivakumar 2007), but not the protection of nesting habitat. The pubic awareness and research into population dynamics and social organisation (Sankaran & Sivakumar 1999) were only partly implemented (Table 1).

Neither the surveys nor management plan for the Biak megapode recommended in the Megapode Conservation Action Plans have been implemented (Table 1).

The surveys and harvest plan (Bowen 1996; Foster 1999; O'Brien et al. 2003) recommended in the Megapode Conservation Action Plans for the Vanuatu megapode have both been only partially implement (Table 1).

Three case studies of megapodes conservation

In this section I detail three case studies of conservation projects on megapodes. I start by detailing the background to the species and its conservation not covered above, then give the rationale and objectives for the project, before detailing the methods used and the results obtained.

Maleo in Sulawesi - enforcement, protection of nesting fields, use of hatcheries

There are several conservation projects for the maleo (e.g., Summers 2007; Tasirin 2007; van As 2007). These projects are based around the enforcement of current laws, the protection of nesting fields and eggs, the enhancement of nesting fields and the use of hatcheries to increase the production of megapode chicks.

The maleo is a large (1.3-1.7 kg) monogamous species found only in the lowland and lower montane rainforests of Sulawesi, Indonesia (Jones et al. 1995). Males and females dig burrows together in solar or geothermal fields where females lay 8-12 eggs per year (Dekker 1990; c.f. ~ 30, MacKinnon 1978). Although illegal, eggs are heavily harvested throughout the range (Baker & Butchart 2000) with harvesters collecting up to 100% of eggs in some areas (MacKinnon 1978). There is a trend of deteriorating quality of nesting fields across the entire island (Baker and Butchart 2000).

The rationale for conservation of maleo is that it is classified as Endangered, is fully protected under Indonesian law (MacKinnon 1981), and that it is in serious decline (IUCN 2006). Maleo eggs are not the major protein or income source for local people at nesting fields (Argeloo & Dekker 1996), so reducing the harvest will not result in economic hardship or reduced food security. Furthermore, local people close to some fields are supportive of conservation efforts (e.g., Summers 2007; Tasirin 2007; van As 2007) and active management has been shown to be successful for this species (e.g., MacKinnon 1981)

The three projects are located at Bogani Nani Wartabne National Park (Christy and Lentey 2002; Tasirin 2007, John Tasirin in litt. 2007) and Tangkoko (van As 2007) in North Sulawesi, and Taima in Central Sulawesi (Summers 2007). They have the following broadly similar objectives: to increase the maleo population, to raise awareness of the importance of conservation, to increase the involvement of the local community in conservation, and to promote a broader understanding of conservation (John Tasirin in litt. 2007).

Protection of nesting fields is the main recommendation made by those who study maleo conservation (e.g., MacKinnon 1981; Dekker 1990; Argeloo & Dekker 1996; Baker & Butchart 2000; Gorog et al. 2005). All three projects feature moratoriums on collecting eggs and the use of guards at nesting fields; guards are either ex-harvesters (Summers 2007; van As 2007), other local people, or park rangers (John Tasirin in litt. 2007).

Two projects are based around the protection of eggs, where they are either left in situ (Summers 2007) or transferred to semi-natural hatcheries (Christy and Lentey 2002; Tasirin 2007); hatcheries have been used in maleo conservation efforts in the past (e.g., MacKinnon 1981; Dekker & Wattle 1987) and are widely recommended. That said, at some sites

95 hatcheries have been inadequately managed and their benefits are not well studied (Baker and Butchart 2000).

The improvement of nesting habitat by removing invasive and dense vegetation has been widely recommended (MacKinnon 1981; Dekker 1990; Baker & Butchart 2000; Gorog et al. 2005). This intervention is not well studied (Sinclair 2000b) and the removal of all dense vegetation may be counter-productive given chick survival is higher when there is escape cover (Göth and Vogel 2002; 2003).

The largest maleo conservation project in recent times is run by the Wildlife Conservation Society and Bogani Nani Wartabne National Park (Tasirin 2007). This site has hatcheries in three geothermal locations that are fenced against predators and poachers. Eggs are collected from fields, marked, measured and then transported to hatcheries. Emergent chicks are weighed and measured before being realised in nearby forest, while the adult population is monitored by counts of birds laying in fields (John Tasirin in litt. 2007).

All three projects are considered successful in that they have at least achieved high levels of community support and involvement. They have also resulted in more birds using enhanced nesting fields; one apparently abandoned nesting field has become active again (van As 2007) and cleared areas have significantly more activity and eggs laid that unclear areas (Christy & Lentey 2002). Voluntary moratoriums on collecting eggs have been put in place by local people to allow maleo populations to recover (Summers 2007), and many eggs have been protected and chicks hatched; at Taima over 1200 eggs were protected in the first 11 months of the project to July 2007, while at the Wildlife Conservation Society- Bogani Nani Wartabne National Park project the 4000th maleo chick was released in May 2007 (Tasirin 2007).

It appears that projects in areas where there is community support, using increased protection of nesting fields, hatcheries for eggs, and habitat improvement, are successful in reversing the deterioration of fields and in producing substantial numbers of megapode chicks. That said, there are several areas of research requiring attention to fully asses the validity of the methods used for the maleo: monitoring population changes independent of nesting fields is needed to determine if chick production results in more adult birds, and studies of population genetics are required to determine if there are high levels of natal philopatry (John Tasirin in litt. 2007). Comparisons of hatching success in situ versus in hatcheries are required to determine if hatcheries are the best approach, as is the study of adult movements from fields, and their home range and habitat use (Summers 2007).

Melanesian megapode in the Solomon Islands - sustainable harvests of eggs

The second case study is for the Melanesian megapode in the Solomon Islands where a research driven project was used to develop a sustainable harvest plan for eggs. (Unless other citations are given, the following is taken from Sinclair 1999b; 2001a; in prep.)

Melanesian megapodes (M. eremita) display considerable behavioural plasticity in their use of incubation sites and heat sources but are best known for the huge nesting fields from which hundreds of thousands eggs were harvested in the past (Bishop 1980; Broome et al. 1984). The species is sexually monomorphic, socially monogamous, territorial away from fields, and forage in both primary and secondary forest. The species occurs in the Bismarck Archipelago, Papua New Guinea, and Solomons Islands (Jones et al. 1995). Melanesian megapodes are considered to be widespread and abundant, and are classified as a species of Least Concern (IUCN 2006). Despite this classification, the over harvesting of eggs is widespread (King 1989; Kisokau 1991) and management is needed to protect many populations (Sinclair pers. obs.).

The rationale for this project is that there are no management models for nesting fields of megapodes (Jones 1999), yet over half of the species threatened with extinction use this system (IUCN 2006). This project is an opportunity to conduct management experiments with a common species that can be applied to problems with threatened species. Furthermore, eggs are an important source income and nutrition for people close to large fields (e.g.,

96 Dureau 1993) who therefore are supportive of these interventions and actively manage the megapode fields.

The objectives of the main research project were to conduct research to develop a community-based management plan on Simbo Island, Solomon Islands. At the start of the project local management involved a 2-month closed season and formalised local enforcement, but these were considered inadequate by the local people as the megapode population was believed to be declining. The research questions were broadly based around determining if eggs were escaping the harvest, and if not, whether closed seasons or hatcheries are the best management option. The risks faced by dispersing chicks were also studied.

On the island birds lay in two nesting areas where heat is generated by volcanic activity (Sibley 1946). Local people say they earn more than 60% of their cash-income from the sale of megapode eggs and their extensive Traditional Ecological Knowledge of the natural history of megapodes influences their interactions with them. In the 1970-1980s locals noticed declines in the megapode population and harvest of eggs, and in 1990 approved the Simbo Megapode Management Ordinance which featured 2-months closed to harvesting eggs, no killing of birds and other activities. Unfortunately, local people noticed continuing declines through the 1990s, which lead to the project detailed here.

The methods used in this project were participatory, involving the local people in all aspects of the project. Initially their Traditional Ecological Knowledge was surveyed to generate hypotheses to be tested. The size of the egg harvest was then determined by household surveys, the efficiency of harvesters estimated using a mark-recapture experiment with eggs, and the incubation period, hatching success and effectiveness of hatcheries investigated by marking and following eggs. The risks faced by dispersing chicks were estimated by radio- tracking.

In the 12-months of the study about 200,000 eggs were harvested, which equates to about 40 days of the average rural wage per family, confirming the harvest is economically important. The efficiency of harvesters was very high meaning that eggs had < 1% chance of escaping the harvest. The conclusion from these studies was that there was a need to manage harvests of megapode eggs.

1 18000 0.9 16000 0.8 14000

0.7 12000 0.6 10000 0.5 8000 0.4 6000 0.3 4000 0.2 produced Chicks

hatched Proportion 0.1 2000 0 0 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 Days of closed season Figure 3. Results of a preliminary model for closed seasons of 61 days (61d) and 91 days (91d). The left scale shows proportion of eggs laid on each day of the 61d (squares) and 91d (triangles) closed seasons that hatch before the end of the closed season. For example, of the eggs laid on the first day of the 91d closed season, 98% will hatch, whereas for the 61d closed season 38% hatch. The right scale is the cumulative number of chicks produced during the 61d (diamonds) and 91d (circles) closed seasons, with 95% Confidence Intervals. The Confidence Interval for the 61d closed season is too small to be seen on this scale. (Reproduced with permission from Sinclair (1999a)

97 The mean incubation period of eggs (about 65 days) was longer than the 2-month (61 days) closed season, indicating the closed season was too short to achieve its stated objective of allowing sufficient eggs to escape the harvest. A preliminary model developed to examine the productivity of closed seasons of different lengths estimated that a for 61-day closed season only 6% of eggs laid actually hatch producing between 1,200-2000 chicks, whereas for a 91-day closed season 31% of eggs hatch producing between 10,000-16,500 chicks (Figure 3).

In the hatchery experiment mortality was strongly dependent on treatment, with moving early- stage eggs resulting in significantly higher mortality than moving late-stage eggs, while the late-stage eggs did not have a significantly higher mortality than those left in situ. It was therefore concluded that hatcheries were a viable management option.

Hatching success was high for eggs of the Melanesian megapode (i.e., 65% hatched, 26% addled, 3% died in ground), but chick survival was very low (i.e., most chicks died < 48 hrs; all were dead or missing < 1 week). Megapode chicks face a significantly higher risk of death from cats and dogs than all other causes combined (Figure 2). The recommendation from this research was that natural predators should not be controlled, whereas introduced predators should be controlled if possible.

The most important outcomes of this project were a broad acceptance of the results of research leading to amendments to the management ordinance, including the closed season being extended to 3 months and a new community-based enforcement system. An extension of the close season to 3 months produces as many chicks as 60 hatcheries, so closed seasons are a more practical approach than hatcheries for community management.

DF2

WBT

Random NGM

BCT

DF1

Figure 4. Distribution of scores from a Discriminant Function Analysis showing separation of mound sites of Wattled Brush-turkeys (WBT), New Guinea Megapodes (NGM) and Brown- collared Talegallas (BCT) and random points along two discriminant axes (DF1 and DF2). Going from habitat at random points to that at mound sites along the DF1 axis is moving from more disturbed forest to a more mature closed-canopy forest. Ellipses are 95% contours with points being the locations of group means. (Reproduced with permission from Sinclair (2002)

Mound-building megapodes in Papua New Guinea - management prescriptions, local empowerment

The third case study is for mound-building megapodes in Papua New Guinea. (Unless other citations are given, the following is taken from (Sinclair 1997; 2000a; 2000b; 2001b; 2002).

98 The nine species in three genera of megapodes on the mainland of New Guinea are all mound builders (Jones et al. 1995) and are all considered species of Least Concern (IUCN 2006). This research involved the New Guinea megapode (a socially monogamous species that nests year-round with >1 pair sharing some mounds), the wattled brush-turkey (a polygynous and polyandrous species that has a short breeding season where the male alone builds and maintains the mound), and the brown-collared talegalla (T. jobiensis; a little known species that shares mounds with other species over a short breeding season).

The rationale for this project is that very little is known about New Guinea megapodes, yet all species are heavily harvested and local extinctions have occurred Sinclair pers. obs.). This project is designed to provide information for local people to manage megapodes because 97% of land in Papua New Guinea is in customary ownership (Marat 1991), so local people control the harvesting of megapodes and their eggs. Local people base resource management decisions on their astounding Traditional Ecological Knowledge. Traditional Ecological Knowledge is based on long-term observations (Baines & Hviding 1992) and often involves management prescriptions (Gadgil & Berkes 1991). If research will benefit local people it must provide recommendations in a form local people can use and that makes sense to them in the context of their Traditional Ecological Knowledge.

The main objective of the study was to learn the basic natural history of the species. The habitat variables associated with mounds were investigated by comparing them to random points so recommendations could be made about the habitat modification that occurs throughout New Guinea. Given harvesting of eggs from mounds is widespread, its effects were investigated by putting temperature loggers in mounds. The behaviour of megapodes was studied by erecting hides at mounds and tagging and radio-tracking them.

(b)

35 (a)

C

o 30

25 Temp./

20 (c) 15 0 1200 2400 1200 2400 1200 2400

Time of day

Figure 5. Temperatures in the core of megapode mounds when they are (a) excavated and covered by birds in the process of laying eggs, (b) excavated then left uncovered by harvesters in search of megapode eggs and, (c) excavated then covered by harvesters. The change in temperature at about 1200 hrs in (c) is when the megapodes returned to the mound and covered the hole left by the harvesters. (Sinclair, unpublished data)

Compared to random points, forest at mound sites of all species is in relatively undisturbed primary forest with a very closed canopy (moving along DF1 in Figure 4). It was therefore recommended that gardens are not established within 100 m of mounds and no trees at all are cut within 20 m of a mound. It was also recommended that areas be excluded from logging concessions in which harvests of megapode eggs are required in the future.

Because most megapode activity at mounds is in the morning, and that Brush-turkeys at least readily desert mounds if disturbed, it was recommended that egg collectors visit mounds only in the afternoon.

99 From temperature-loggers buried in mounds excavated by megapodes and egg harvesters, it appears humans and birds have a similar effect on the temperature of the mound provided the mound is closed after excavation. When the mound is left open after harvesting, as is the practice of many egg collectors, mounds cool rapidly and take a long time to recover to incubation temperatures (Figure 5). Leaving mounds uncovered could result in increased egg mortality when not all the eggs have been found and may lead to abandonment of some mounds. Given this, it was recommended that egg collectors close mounds after excavating them.

As with most species of megapode, the potential exists for over-harvesting of eggs, especially as the human population expands rapidly. In addition, wattled brush-turkeys may experience greater impacts than other species due to their shorter breeding season and concentration of eggs in some mounds. Accordingly, it was recommended that egg collectors visit mounds of wattled brush-turkeys only once per season and those of the other species not more than three times per year.

These recommendations were made to landowners at the study site through a series of meetings, and used in a curricula developed for environmental education in Papua New Guinea. Posters with these recommendations are now being trialled in villages in Papua New Guinea as a way to disseminate the information more widely .

Conclusion

The megapode family is at serious risk, and most if not all threatened species are in decline (IUCN 2006). With these levels of risk, if current decline continue, more species will be added to the list of extinct megapodes. The actions needed to address these declines have been identified and published in two Megapode Conservation Action Plans (Dekker et al. 1995b; Dekker et al. 2000). Those few actions that have been fully implemented have been largely successful. Most actions, however, have not been implemented sufficiently to conserve the species concerned (Table 1). We know what to do to conserve most megapodes; now we need to do it.

Acknowledgements

I would like to thank the organising committee of the Malleefowl Forum 2007, particularly Stephen Davies and Carl Danzi, and Rene Dekker, Heribert Hofer and Monica Hilker. I am grateful for the assistance from Lorima Tuke, Billy Sada and Jackie Fanning on the Melanesian megapode project, my many assistants on the mound-builders project, and the permission of the people of Simbo Island and Pawaia people of Crater Mountain for permission to work on these projects, respectively. Nick Brickle, John Tasirin and Martin Christy provided unpublished reports on the maleo project, Rene Dekker and Jez Bird provided useful information, while Zoe Coulson provided logistical support.

References

Anonymous (1995). Major nesting ground of the Moluccan Megapode Eulipoa wallacei rediscovered. Megapode Newsletter 9:1-2.

Argeloo, M. (1994). The Maleo Macrocephalon maleo: New information on the distribution and status of Sulawesi's endemic megapode. Bird Conservation International 4:383- 393.

Argeloo, M. & Dekker, R. W. R. J. (1996). Exploitation of megapode eggs in Indonesia: The role of traditional methods in the conservation of megapodes. Oryx 30(1): 59-64.

Baines, G. & Hviding, E. (1992). Traditional Environmental Knowledge from the Marovo Area of the Solomon Islands. LORE Capturing Traditional Environmental Knowledge. M.

100 Johnson. Ottawa, Dene Cultural Institute and the International Development Research Centre: 83-103.

Baker, G. C. & Butchart, S. H. M. (2000). Threats to the Maleo Macrocephalon maleo and recommendations for its conservation. Oryx 34:255-262.

Baker, G. C. & Dekker, R. (2000). Lunar synchrony in the reproduction of the Moluccan Megapode Megapodius wallacei. Ibis 142:382-388.

Baker, G. C. (2002). Conservation status of Maleo Macrocephalon maleo nesting grounds: an update. Megapode Newsletter 16:4-6.

Bennett, P. M. & Owens, I. P. F. (1997). Variation in extinction risk among birds: chance or evolutionary predisposition? Proceedings of the Royal Society B: Biological Sciences 264:401.

Benshemesh, J. (1992). The conservation ecology of Malleefowl, with particular regard to fire. Clayton, Monash University. PhD thesis.

Benshemesh, J. (1999). The National Malleefowl Recovery Plan: a framework for conserving the species across Australia. Zoologische Verhandelingen 327:101-124.

Benshemesh, J. (2000). National Recovery Plan for Malleefowl, Environment Australia.

BirdLife International and IUCN (in prep.). Threatened birds of the world 2008. BirdLife International, Cambridge.

Birks, S. M. (1997). Paternity in the Australian brush-turkey, Alectura lathami, a megapode bird with uniparental male care. Behavioral Ecology 8:560-568.

Bishop, D. (1980). Birds of the volcanoes. The Scrubfowl of West New Britain. World Pheasant Association Journal 5:80-90.

Bowen, J. (1996). Notes on the Vanuatu Megapode Megapodius layardi on Ambrym, Vanuatu. Bird Conservation International 6:401-408.

Brickhill, J. (1986). Breeding success of Malleefowl Leipoa ocellata in Central New South Wales. Emu 87:42-45.

Brom, T. G. & dekker, R. W. R. J. (1992). Current studies on megapode phylogeny. Zoologische Verhandelingen 278:7-17.

Broome, L. S., Bishop, K. D. & Anderson, D. R. (1984). Population density and habitat use of Megapodius freycinet eremita in West New Britain. Australian Wildlife Research 11:161-171.

Butchart, S. H. M. & Baker, G. C. (2000). Priority sites for conservation of maleos (Macrocephalon maleo) in central Sulawesi. Biological Conservation 94:79-91.

Christy, M. J. & Lentey, S. M. (2002). Maleo Project – Phase 3 North Sulawesi, Indonesia. Bogani Nani Wartabone National Park, Annual Summary, Aug 2001 – Jul 2002. The Wildlife Conservation Society, Sulawesi.

Collar, N. (1994). Red Data Books, Action Plans, and the need for site specific synthesis. Species 21/22:132-133.

Crome, F. H. J. & Brown, H. E. (1979). Notes on social organisation and breeding of the Orange Footed Scrubfowl Megapodius reinwardt. Emu 79:111-119.

101 Dekker, R. W. R. J. & Wattle, J. (1987). Egg and image: new and traditional uses for the Maleo (Macrocephalon maleo). In The value of birds. D.A.W. and F. L. Filion, International Council for Bird Preservation, Technical Publication No. 6:83-87.

Dekker, R. W. R. J. (1989). Predation and western limits of the megapode distribution. Journal of Biogeography 6:317-321.

Dekker, R. W. R. J. (1990). The distribution and status of nesting grounds of the Maleo Macrocephalon maleo in Sulawesi, Indonesia. Biological Conservation 51:139-150.

Dekker, R. W. R. J., Argeloo. M. & Jepson, P. (1995a). Notes on the Moluccan Megapode Eulipoa wallacei (G.R. Gray, 1860) following the rediscovery of two major nesting grounds. Zoologische Mededelingen Leiden 69:251-260.

Dekker, R. W. R. J., McGowan, P. J. K. & WPA/BirdLife/SSC Megapode Specialist Group. (1995b). Megapodes: an action plan for their conservation, 1995-1999. Gland, Switzerland, IUCN.

Dekker, R. W. R. J., Fuller, R. A. and Baker, G. C. Eds. (2000). Megapodes: Status Survey and Conservation Action Plan 2000-2004. WPA/BirdLife/SSC Megapode Specialist Group. Gland, Switzerland, IUCN and the World Pheasant Association.

Dekker, R. W. R. J. (2000). News about Bruijn's brush-turkey and Waigeo. Megapode Newsletter 14:1-2.

Dekker, R. (2007). Distribution and speciation of megapodes (Megapodiidae) and subsequent development of their breeding. In Biogeography, Time, and Place: Distributions, Barriers, and Islands. W. Renema (ed.): 93-102. del Hoyo, J., Elliott, A. & Sargatal, J. (1992). Handbook of the birds of the world. Barcelona, Lynx Edicions.

Diamond, J. M. (1983). The reproductive biology of mound-building birds. Nature 301: 288- 289.

Dureau, C. (1993). Nobody asked the mother: women and maternity on Simbo, western Solomon Islands. Oceania 64:18.

Dwyer, P. D. (1981). Two species of megapode laying in the same mound. Emu 81: 173-174.

Foster, T. (1999). Update on the Vanuatu Megapode Megapodius layardi on Ambrym, Vanuatu. Bird Conservation International 9:63-71.

Frith, H. J. (1956a). Breeding habits in the family of Megapodiidae. Ibis 98:620-640.

Frith, H. J. (1956b). Temperature regulation in the nesting mounds of the Malleefowl, Leipoa ocellata. Australian Wildlife Research 1:79-95.

Frith, H. J. (1959). Breeding of the mallee fowl, Leipoa ocellata Gould (Megapodiidae). CSIRO Wildlife Research 4:31-60.

Fuller, R. A., McGowan, P. J. K., Carroll, J. P.,Dekker, R. W. R. J. & Garson, P. J. (2003). What does IUCN species action planning contribute to the conservation process? Biological Conservation 112:343.

Gadgil, M. & Berkes, F. (1991). Traditional resource management systems. Resource Management and Optimization 8:127-141.

102 Gorog, A. J., Pamungkas, B. & Lee, R. J. (2005). Nesting ground abandonment by the maleo (Macrocephalon maleo) in North Sulawesi: Identifying conservation priorities for Indonesia's endemic megapode. Biological Conservation 126:548-555.

Göth, A. & Vogel, U. (1995). Status of the Polynesian Megapode Megapodius pritchardii on Niuafo'ou (Tonga). l Bird Conservation International 5:117-128.

Göth, A. & Vogel, U. (1999). Notes on breeding and conservation of birds on Niuafo'ou Island, Kingdom of Tonga. Pacific Conservation Biology 5:103-114.

Göth, A. & Vogel, U. (2002). Chick survival in the megapode Alectura lathami (Australian brush-turkey). Wildlife Research 29:503-511.

Göth, A. & Vogel, U. (2003). Juvenile dispersal and habitat selectivity in the megapode Alectura lathami (Australian brush-turkey). Wildlife Research 30:69-74.

Heij, C. J. (1997). The behaviour of the Moluccan megapode Eulipoa wallacei (Aves: Megapodiidae) in nesting grounds. Treubia 31:169-176.

Heij, C. J., Rompas, C. F. E. & Moeliker, C. W. (1997). The biology of the Moluccan megapode Eulipoa wallacei (Aves, Galliformes, Megapodiidae) on Haruku and other Moluccan islands; part 2: final report. Deinsea 3:1-124 . IUCN (2006). 2006 IUCN Red List of Threatened Species. www.iucnredlist.org. Downloaded on 30 May 2007.

Jones, D. N. (1987). Selection of incubation mound sites by the Australian Brush-turkey Alectura lathami. Ibis 130:251-260.

Jones, D. N. (1988a). Construction and maintenance of the incubation mounds of the Australian Brush-turkey Alectura lathami. Emu 88:210-218.

Jones, D. N. (1988b). Hatching success of the Australian Brush-turkey Alectura lathami in South-East Queensland. Emu 88:260-263.

Jones, D. N. (1989). Modern megapode research: a post-Frith review. Corella 13:145-154.

Jones, D. N. (1990a). Male mating tactics in a promiscuous megapode: patterns of incubation mound ownership. Behavioral Ecology 1:107-115.

Jones, D. N. (1990b). Social organization and sexual interactions in Australian Brush-turkeys Alectura lathami: Implications of promiscuity in a mound-building megapode. Ethology 84:89-104.

Jones, D. & S. Birks (1992). Megapodes - Recent ideas on origins, adaptations and reproduction. Trends in Ecology and Evolution 7:88-91.

Jones, D. N. (1992). An evolutionary approach to megapode mating systems. Zoologische Verhandelingen 278:33-42.

Jones, D. N. (1994). Reproduction without parenthood: male tactics and female choice in a promiscuous bird. Animal Societies. In J. P.J. and A. Rossitter (eds). Kyoto, Kyoto University Press.

Jones, D. N., Dekker, R. W. R. J. & Roselaar, C. S., Eds. (1995). The Megapodes. Bird Families of the World. Oxford, Oxford University Press.

Jones, D. N. (1999). What we don't know about megapodes. Zoologische Verhandelingen 327:159-168.

103 King, B. (1989). Does wildlife management by the people work? A case study of the Pokili Wildlife Management Area, Papua New Guinea. Science in New Guinea 15:111-118.

Kisokau, K. M. (1976). A study in the biology of the megapodes of West New Britain. Papua New Guinea Bird Society Newsletter 121:20-23.

Kisokau, K. (1991). Biology and ecology of the wildfowl, Megapodius freycinet Gaimard, at Garu egg grounds, West New Britain, Papua New Guinea. Biology. Port Moresby, University of Papua New Guinea. MSc thesis.

Kratter, A. W., Kirchman, J. J. & Steadman, D. W. (2006). Upland bird communities on Santo, Vanuatu, Southwest Pacific. Wilson Journal of Ornithology 118:295-308.

Mack, A. & West, P. (2005). Ten thousand tonnes of small animals: wildlife consumption in Papua New Guinea, a vital resource in need of management. Resource Management in Asia-Pacific Working Paper No. 61. Canberra, Australian National University.

MacKinnon, J. (1978). Sulawesi megapodes. Journal of the World Pheasant Association 3:96- 103.

MacKinnon, J. (1981). Methods for the conservation of Maleo birds, Macrocephalon maleo on the island of Sulawesi, Indonesia. Biological Conservation 20:183-193.

Marat, A. (1991). The Land tenure System of Papua New Guinea. Conservation and environment in Papua New Guinea: establishing research priorities. Waigani, Papua New Guinea, Department of Environment and Conservation, Papua New Guinea.

Mauro, I. (2005). Field discovery, mound characteristics, bare parts, vocalisations and behaviour of Bruijn's Brush-turkey (Aepypodius bruijnii). Emu 105:273-281.

Mauro, I. (2006). Habitat, microdistribution and conservation status of the enigmatic Bruijn's Brush-turkey Aepypodius bruijnii. Bird Conservation International 16:279-292.

O'Brien, M., Beaumont, D. J., Peacock, M. A., Hills, R. & Edwin, H. (2003). The Vanuatu Megapode Megapodius layardi, monitoring and conservation. Unpublished report. Sandy, Bedfordshire, RSPB.

O'Brien, T. G. & Kinnaird, M. F. (1996). Changing populations of birds and mammals in North Sulawesi. Oryx 30:150-156.

Olson, S. L. (1980). The significance of the distribution of the Megapodiidae. Emu 80:21-24.

Priddel, D. & Wheeler, R. (1990). Survival of Malleefowl Leipoa ocellata chicks in the absence of ground-dwelling predators. Emu 90:81-87.

Priddel, D. & Wheeler, R. (1994). Mortality of captive-raised Malleefowl, Leipoa ocellata, released into a mallee remnant within the wheat-belt of New South Wales. Wildlife Research 21:543-552.

Priddel, D. & Wheeler, R. (1996). Effect of age at release on the susceptibility of captive- reared Malleefowl Leipoa ocellata to predation by the introduced Fox Vulpes vulpes. Emu 96:32-41.

Rand, A. L. & Gilliard, E. T. (1967). Handbook of New Guinea Birds. London, Weindefeld and Necolson.

Rinke, D. (1993). Safe islands for the Malau. Ornithological Society of New Zealand News - Supplement to Notornis 40:1.

104 Rinke, D. R., Soakai, L. H. & Usback, A. (1993). Koe Malau. Life and future of the Malau. Bonn, Germany and Nuku'alofa, Kingdom of Tonga, Brehm Fund for International Bird Conservation.

Samson, M. (2007). Incubation sites of the Melanesian Megapode (Megapodius eremita) on Lihir Island, Papua New Guinea. Biology. Port Moresby, University of Papua New Guinea. BSc (Hons) Thesis.

Sankaran, R. (1995). The distribution, status and conservation of the Nicobar megapode Megapodius nicobariensis. Biological Conservation 72:17-25.

Sankaran, R. & Sivakumar, K. (1999). Preliminary results of an ongoing study of the Nicobar megapode Megapodius nicobariensis Blyth. Zoologische Verhandelingen 327:75-90.

Seymour, R. S. & Bradford, D. F. (1992). Temperature regulation in the incubation mounds of the Australian Brush-turkey. Condor 94:134-150.

Sibley, C. G. (1946). Breeding habits of megapodes on Simbo, central Solomon Islands. Condor 48:92-93.

Sinclair, J. R. (1997). Preliminary results from a study of the behaviour, ecology and conservation of three species of megapode in Papua New Guinea. Megapode Newsletter 11:2-3.

Sinclair, J. R. (1999a). Lape forever: research and monitoring of the Melanesian megapode Megapodius eremita on Simbo Island, Solomon Islands Sept 1997 to Mar 1999. Megapode Newsletter, Technical report to World Wildlife Fund Solomon Islands Community Resource Conservation and Development Project: 33.

Sinclair, J. R. (1999b). The Melanesian Megapode Megapodius eremita on Simbo Island, Solomon Islands: participatory research towards a management plan. Megapode Newsletter 13:6-8.

Sinclair, J. R., O'Brien, T. G. & Kinnaird, M. F. (1999). Observations on the breeding biology of the Philippine Megapode (Megapodius cumingii) in North Sulawesi, Indonesia. Tropical Biodiversity 6:87-97.

Sinclair, J. R. (2000a). The Behaviour, Ecology and Conservation of Three Species of Megapode in Papua New Guinea. Department of Zoology. Dunedin, University of Otago. MSc thesis.

Sinclair, J. R. (2000b). Management recommendations for three sympatric species of megapode in Papua New Guinea. Megapode Newsletter 14:9-14.

Sinclair, J. R. (2001a). Lape forever - research and monitoring of the Melanesian Megapode Megapodius eremita on Simbo Island, Solomon Islands 1997 to 1999. Megapode Newsletter 15:5-14.

Sinclair, J. R. (2001b). Temperature regulation in mounds of three sympatric species of megapode (Aves: Megapodiidae) in Papua New Guinea: testing the 'Seymour Model'. Australian Journal of Zoology 49:675-694.

Sinclair, J. R. (2002). Selection of incubation mound sites by three sympatric megapodes in Papua New Guinea. Condor 104:395-406.

Sinclair, J. R., O'Brien, T. G. & Kinnaird, M. F. (2002). The selection of incubation sites by the Philippine Megapode, Megapodius cumingii, in North Sulawesi, Indonesia. Emu 102:151-158.

105 Sinclair, J. R. (in prep.). High hatching success and low chick survival in Melanesian megapodes Megapodius eremita on Simbo Island, Solomon Islands.

Sivakumar, K. (2007). The Nicobar megapode: Status, ecology and conservation: Aftermath tsunami. Dehradun, Unpublished report, Wildlife Institute of India.

Steadman, D. W. (1993). Biogeography of Tongan birds before and after human impact. Proceedings of the Natural Academy of Science USA 90:818-822.

Steadman, D. W. (1999). The biogeography and extinction of megapodes in Oceania. Zoologische Verhandelingen 327:7-21.

Steadman, D. W. (2006). Extinction and Biogeography of Tropical Pacific Birds, University Of Chicago Press.

Stinton, D. W. & Glass, P. O. (1992). The Micronesian megapode Megapodius laperouse: Conservation and research needs. Zoologische Verhandelingen 278: 53-55.

Summers, M. (2007). Report of conservation activities at maleo nesting ground Libuun, Taima, Tompotika, Central Sulawesi, Indonesia August 2006-June 2007. Megapode Newsletter 20:4-5.

Tasirin, J. (2007). Release of 4000th maleo chick in Sulawesi. Megapode Newsletter 20:7-8.

USFWS (1998). Recovery plan for the Micronesian megapode (Megapodius laperouse laperouse). Portland, U.S. Fish and Wildlife Service. van As, J. (2007). Maleo nesting ground project in Tangkoko, North-Sulawesi, Indonesia. Megapode Newsletter 20:6-7.

Watling, D. (2003). Report on a visit to Late and Fonualei Islands, Vava'u Group, Kingdom of Tonga. Leiden, The Van Tienhoven Foundation for International Nature Protection.

Weathers, W. W., Weathers, D. L. & Seymour, R. S. (1990). Polygyny and reproductive effort in the Malleefowl Leipoa ocellata. Emu 90:1-6.

Weir, D. G. (1973). Status and habits of Megapodius pritchardii. Wilson Bulletin 85:79-82.

Whitmore, T. C. (1982). Wallace's Line: A result of plate tectonics. Annals of the Missouri Botanical Garden 69:668.

106

13. BRUIJN”S BRUSH-TURKEY Aepypodius bruijnI: FIELD DISCOVERY, MONITORING AND CONSERVATION OF AN ENIGMA

Iwein Mauro P.O. Box 289, 98401 Sorong, Indonesia, e-mail: [email protected]

Abstract

The endemic brush-turkey of Waigeo Island in eastern Indonesia, Aepypodius bruijnii, managed to remain entirely unknown in the living world during the more than 120 years that elapsed in between its formal description from trade skins in 1880 and its field discovery by the speaker in May 2002. Since then, Bruijn‘s Brush-turkey has been shown to nest only on the island‘s previously ornithologically unexplored highest summits, along an ecological gradient above 620 m elevation, where a structurally distinctive, wind-sheared and possibly locally edaphically controlled, stunted cloud-forest grows on infertile substrates. This breeding habitat comprises just 60 km2 or 1.9 % of the island‘s area and is contained within six locations sensu IUCN (2001), three of which are now confirmed to support breeding populations. The species‘ global population has been estimated at 349 mound-owning males or 977 mature individuals, with 98 % of the population restricted to just three locations in the eastern part of the island (Mauro 2004; 2006). Threats that impinge on the species are discussed, and it is concluded that its current precautionary treatment as Vulnerable warrants upgrading to Endangered in accordance with the revised IUCN Red List categories and criteria. Habitat destruction as a consequence of wild fires and a recent logging epidemic had previously been identified as the major factor threatening this unique brush-turkey‘s long-term survival (Mauro 2004; 2006). In addition, the species‘ second and third largest breeding populations could now rapidly come under extreme pressure if proposed mineral extraction projects in Waigeo‘s northern ultramafic belt materialize.

Introduction

Bruijn‘s Brush-turkey Aepypodius bruijnii, (Figure 1) is endemic to the Indonesian island of Waigeo (0°12‘S; 130°45‘E), situated just below the Equator, at the northern end of the Raja Ampat or ‗Four Kings‘ archipelago off New Guinea‘s northwestern tip. With a total area surface of c. 3,100 km2 Waigeo is the largest island in the group, which is further made up of the main islands of Misol, Salawati, and Batanta, plus a myriad of more than 600 satellites scattered around these, altogether comprising over 40,000 km2 of land and sea. Waigeo reaches a highest point at 970 m a.s.l. on the summit of Mt. Sau Lal, but the island‘s most notable topographic features are probably the cone of Mt. Nok or Buffelhorn, standing c. 880 m a.s.l., and the stunningly beautiful fjord-like Mayalibit Bay, surrounded by precipitous limestone karst country, dividing Waigeo in roughly two equal halves. Access to the island is still only through a 60 km long boat journey across the Dampier Strait from the gateway town of Sorong on the westernmost tip of the Bird‘s Head Peninsula.

Bruijn‘s Brush-turkey became known from 23 historical trade specimens that reached European museum cabinets between 1880 and 1904, through the services of the immortal Dutch merchant of Ternate, Anton August Bruijn, a dealer in almost every product that the Moluccas and Bird‘s Head region had to offer, including natural history specimens (Oustalet 1880; Jones et al. 1995; Voisin et al. 2000; Dekker 2000; Moeliker et al. 2003). Some of these specimens are actually lovely mounts (Voisin et al. 2000), testimony to the great skills of the native field collectors and preparators

107

Figure 1. Bruijn‘s Brush-turkey. Photo Iwein Mauro. that Bruijn was employing, and many were labelled ‗Waigeo‘. But subsequent authors (Rothschild et al. 1932; Peters 1934) queried whether this island was really the species‘ homeland. And this doubt appeared justified considering that 15 ornithological expeditions between 1793–1937 failed to record the species on Waigeo (Rothschild et al. 1932; Rand & Gilliard 1967; Frith & Beehler 1998), and Bruijn, after all, was notorious for his inexact labeling practices (De Schauensee 1940). A classic example is the duck Anas waigiuensis, which was originally described as a monotypic genus, Salvadorina, from a specimen supplied by Bruijn that allegedly originated from Waigeo and hence it‘s scientific binomial. But Salvadori‘s Teal has not subsequently been relocated on the island, and although Waigeo does not give up its secrets easily, most probably does not occur there.

However, Bruijn‘s Brush-turkey was rediscovered on Waigeo in 1938 when Jusup Kakiaij, a native Misolese field collector first trained during the Denison-Crockett Expedition to New Guinea, collected a female specimen from Jeimon in the Siam R. valley on the east side of Mayalibit Bay, at the behest of the Academy of Natural Sciences, Philadelphia (De Schauensee 1940). Another six expeditions between 1948 and 2000 failed to record the species in the wild, despite prevailing confidence that Bruijn‘s Brush-turkey still existed on the island (see e.g. Jones et al. 1995; Dekker et al. 2000; BirdLife International 2000). Then, in February 2001, local hunters presented the head and gnawed bones of a female they had hunted and consumed to a local contact person working for the operation ‗Waigeo Expeditions‘, which ran under the auspices of Natuurmuseum Rotterdam (Heij & Post 2001). However, subsequent searches at the location where the bird had been hunted, yielded no success, and again Bruijn‘s Brush-turkey could not be located in the wild.

The field discovery of Bruijn‟s Brush-turkey

In 1985 J. M. Diamond had pointed out that it remained unknown whether any ornithological collectors had reached Waigeo‘s summits (Diamond 1985). Moreover, the idea first introduced by Dekker & Argeloo (1993), that Bruijn‘s Brush-turkey, by analogy with the Wattled Brush-turkey of continental New Guinea and Misol, presumably inhabits the

108 mountains proper, became the dominant view by the turn of the century (see e.g. Jones et al. 1995; Dekker et al. 2000; BirdLife International 2000). Nevertheless, the long-awaited decisive assault on the island‘s peaks did not occur during three consecutive ‗Waigeo Expeditions‘ in 1999, 2000 and 2001, devoted primarily to rediscovering Bruijn‘s Brush-turkey (Heij & Post 2001).

So when I visited Waigeo in the spring of 2002 with the strong desire to observe this most enigmatic of all New Guinea bird species, I embraced these insights and headed for the summit of Mt. Nok, Waigeo‘s best known mountain. In so doing, I located an active incubation mound that I instantly recognized as pertaining to Aepypodius, just one hour after reaching cloud-forest at 715 m elevation on the mountain‘s eastern spine on 14 May 2002. I monitored this mound from a well-concealed hide for a total of 91 hours spread over 13 consecutive mornings from 15 to 27 May between first light and noon (Mauro 2002; 2004; 2005). This produced five excellent sightings of the mound-tending male down to point-blank range. On one occasion the male was watched during more than 20 minutes, digging a test hole in the centre of the flattened top of the mound. Another five sightings, all of solitary birds scurrying away, were made in forests in the vicinity. Novel insights into the species‘ broad habitat requirements and conservation needs (Mauro 2002; Anonymous 2002ab) served as the basis for follow-up field work at the request of the WPA/BirdLife/SSC Megapode Specialist Group in October–December 2002 and again in May–June 2003, bringing to light the existence of a viable population of the species (Mauro 2003; Dekker 2003). I eventually published the first comprehensive descriptions of the species‘ bare parts, which until my work remained essentially unknown (Mauro 2004; 2005). I described the species‘ appearance, vocalizations and behaviour in the field, and further unveiled various aspects of its life history and social organization, corroborating and adding considerable detail to previous assumptions founded on its generic affinities. In the first place, I confirmed that Bruijn‘s Brush-turkey indeed is a mound-builder, and provided the first descriptions of its mounds at different stages of development (Mauro 2004; 2005). I also provided evidence for a male resource-defense polygyny and polyandry mating system, which, uniquely among the megapodes is rooted within the closely allied New Guinean and Australian brush-turkey genera Aepypodius and Alectura (henceforth collectively referred to as the brush-turkeys).

In retrospect, the overall neglect of the island‘s summits, which can now be identified as the major flaw in the hunt for Bruijn‘s Brush-turkey, was probably due to a combination of three factors. First, and foremost perhaps, the possibility that the island could harbour strictly montane bird populations may have been deemed unrealistic in view of its modest topography not exceeding 1,000 m in elevation and effectively lacking significantly large areas above the 900 m contour where montane conditions normally set in on New Guinea (Paijmans 1976). What is more, the sole locality-specific historical specimen of Bruijn‘s Brush- turkey, definitely collected below 640 m elevation from Jeimon in the Siam R. valley in 1938 (De Schauensee 1940), gave rise to the common misconception, most recently perpetuated by Moeliker (2002abc), that the species is rare but widely distributed across Waigeo. Second, virtually all historical collecting and modern reconnaissance took place in coastal, rugged limestone karst country where higher elevations are very difficult to reach (cf. Dekker & McGowan 1995). On the other hand, the only ornithologist having searched the more accessible ultramafics, G. Stein, rather discouragingly tagged this zone ‗a generally strikingly animal-poor region‘ (Rothschild et al. 1932). Third, Waigeo‘s interior remained preponderantly uncharted until long after World War II, and its crucial montane sites were not appreciated until the recent advent of high-resolution digital elevation models.

Finally, transfer of local knowledge often proves crucial in tracking-down little-known species. However, in the instance of Bruijn‘s Brush-turkey, I found that information conveyed by local respondents rested almost entirely on tribal oral lore and narration, an increasingly fragmented, decaying ‗collective memory‘ so to speak, which albeit riveting by no means is necessarily true and accurate, and which in the overall absence of counterevidence from the field, unfortunately, was all too credulously embraced in the past. That the species appears to be known and goes by its own local name in certain villages, as various sources reported (Diamond 1986; Dekker & Argeloo 1993; Post 2001), cannot be considered proof that it occurs, not even incidentally, in the wider vicinity as suggested in the Megapode Action Plans (Dekker & McGowan 1995; Dekker et al. 2000). The fact that none of the respondents

109 recognized the species as a mound-builder in the first place, quite understandably rather perceiving it as an ordinary fowl, to me suggests that the inhabitants of Waigeo perhaps always have possessed merely casual knowledge of the species and never entered its secluded breeding habitat. In any case, the method of extensive village interviewing in order to gather presence/absence data recommended in the last Megapode Action Plan (Dekker et al. 2000), especially as executed by the ‗Waigeo Expeditions‘ without earnest ground-truthing, proved immensely ineffective and to generate merely secondary, ambiguous information, prone to considerable misinterpretation.

Monitoring

It is the males that site, construct, tend and own the incubation mounds in the brush-turkeys, and adult male territoriality is assumed to centre on the mound. In addition, vocal advertisement is restricted entirely to the mating season and may primarily be associated with sexual display on or in the immediate vicinity of the mound (Jones et al. 1995; Sinclair 2002; I.M. unpublished data). The number of reproductively mature males capable of permanently defending a mound is the single most relevant population unit to conservation assessment in a male resource-defending promiscuous megapode given that exactly these males secure virtually all copulations, and monopolize the chance of successful progeny (Jones 1990ab).

Waigeo receives an annual rainfall of 1,500–2,000 mm (van Royen 1960), most of which falls in a distinctive wet season from December through to May (local information). Elsewhere in seasonally humid environments the onset of breeding for mound-building megapodes has been found to be defined by local precipitation regimes (Jones et al. 1995). Incubation evidently was in progress on Mt Nok in mid-May 2002 following normal rainfall in the year 2001 and the first quarter of 2002. More recently, I witnessed territorial display on Mt. Danai in early April 2007, further corroborating that the species‘ habitual breeding season unfolds in the March-May second half of the normal wet season on the island (Mauro 2002; 2004; 2005; 2006; Papua Expeditions 2007). Severe ENSO-induced drought conditions, however, afflicted the entire Raja Ampat region during the follow-up field work, and reproductive and vocal activity in the Bruijn‘s Brush-turkey was not forthcoming. Hence I relied exclusively on qualitative mapping of its mounds to assess population status.

Assisted by a variable team of local hunters, I searched for incubation mounds during c. 4,500 person-hour, initially within every altitudinal belt, later on mainly confined to the species‘ increasingly tested and confirmed altitudinal breeding distribution (Mauro 2004; 2005; 2006). I spot-mapped each mound using GPS and barometric altimeter, and assigned it to one of four distinctive phases through which the mounds of this species normally pass: Construction phase, Active phase, Collapsed phase and Spread-out phase. By repeating this procedure over disjunct survey periods, I obtained static snapshots that provide insight into spatio- temporal distribution and utilization of incubation sites.

In retrospect, such laborious qualitative ‗limited resource‘ mapping would anyhow have been desirable since: (1) both the population of Bruijn‘s Brush-turkey and the extent of its breeding habitat on Mt. Sau Lal are comparatively small and manageable with a group of motivated searchers; (2) for reasons that we were unable to quantify within the given time-frame, the spatial distribution of incubation mounds was truncated in many seemingly suitable environments, indicating that highly specialized ecological and behavioural requirements determine selection of incubation sites; and (3) we now know that the species has very timid territorial vocalizations, consisting of Gallus-like wing-flapping as in the Wattled Brush-turkey, followed by a short nasal hiss (I.M. unpublished data). Such vocalizations are unlikely to be audible beyond 75 m in dense cloud-forest, while average Nearest Neighbour distance between concurrently tended mounds on Mt. Sau Lal was slightly more than 200 m (Mauro 2004, 2006).

The lowest altitude at which I recorded a mound of Bruijn‘s Brush-turkey was 620 m. Adopting this as an altitudinal cut-off, I delineated in a study area for Mts. Nok and Sau Lal the external boundary of breeding habitat realistically suitable to the species through exclusion of geomorphologically unsuited terrain. I then applied Minimum Spanning Tree, Nearest Neighbour and Delaunay/Dirichlet (Voronoy Polygon) Tesselation connection schemes

110 (Rosenberg 2001) to describe and analyse spatio-temporal distribution of incubation sites within the Mt. Sau Lal SA in GIS. Finally, I used the unedited NGA-NASA SRTM-3 three arc- second resolution digital elevation model to assess distribution and extent of land above 620 m elevation on Waigeo, and estimate the species‘ breeding range and global population (Mauro 2004; 2006).

Conservation

I found incubation sites of Bruijn‘s Brush-turkey only on Mts. Nok, Sau Lal and Danai along an ecological gradient above 620 m elevation, where a structurally distinctive, wind-sheared and possibly locally edaphically controlled, stunted cloud-forest grows on infertile substrates (Mauro 2004; 2006; Papua Expeditions 2007). Compression of the altitudinal vegetation belts on low and exposed mountains as on Waigeo, the so-called ‗Massenerhebung‘ effect, has long been documented for many Wallacean and Northern Melanesian islands (Mayr & Diamond 2001). Conversely, it remains possible that a shortage of one or more macronutrients (except Mg), very high soil Mg/Ca quotients or high soil Ni concentrations contribute to the dwarfed nature of forests growing on ultramafic outcrops like Mts. Nok and Sau Lal (Proctor 2003). In any case, these stands of cloud-forest on the island are sufficiently large to have biological meaning, and there is prospect for substantial future taxonomic discovery in the Waigeo cloud-forest zone. In addition to Bruijn‘s Brush-turkey I found the owlet-nightjar Aegotheles wallacii, the scrubwren Sericornis beccarii, and the honeyeaters Meliphaga orientalis facialis and Myzomela [nigrita] steini to be normally confined to the cloud-forest on the island (Mauro 2004; Papua Expeditions 2007). The exact taxonomic affinities of the novel populations of the owlet-nightjar and scrubwren, recorded for the first time on the island during my work, are under systematic investigation currently, whereas the morphologically highly distinctive steini-subspecies of the Black Myzomela certainly qualifies as an endemic allospecies (I.M. unpublished data). The restriction to cloud-forest of bird taxa distinct at the species and allospecies level, and the complete dissimilarity between the mix of montane bird populations on Waigeo and Batanta (Mauro 2004), are testimony that potential even minor ecological gradients can exercise to isolate taxa and sustain speciation across Pleistocene land-bridge islands.

Within its specialized breeding habitat I observed a density of 4 and 8 mound-territories/km2 respectively for the small relict Mt. Nok and the sizeable Mt. Sau Lal (Mauro 2004; 2006). While little comparable published information exists on population densities in megapodes derived from qualitative mapping of incubation mounds, Bruijn‘s Brush-turkey appears to rank as a low-density species. In comparison, Australian Brush-turkey in the wild occurs at densities up to 22.2 mound-territories/km2 (Marchant & Higgins 1993), an order of magnitude 2.4–5.5 times greater. Importantly, more than 20 years after the last fire event on Mt. Sau Lal, the species‘ population density in areas of post-fire succession there was still 4.5 times lower than in adjacent primary tracts (Mauro 2004; 2006).

The rarity of Bruijn‘s Brush-turkey has been linked to potential competitive exclusion by the only other species of megapode on Waigeo, the Dusky Megapode Megapodius freycinet (Dekker & Argeloo 1993; BirdLife International 2001; cf. Ripley 1960), considered a more recent arrival on the island. However, Dusky Megapode was found to occur chiefly in flat alluvial lowland forests along the broader river valleys between sea level and c. 100 m a.s.l., and occurs only patchily up to 285 m a.s.l., where extensive expanses of flat or only slightly undulating terrain are available. Hence the two species of megapode on the island breed in complete vertical parapatry, separated by an altitudinal gap exceeding 300 m, set against a maximum elevation of 970 m a.s.l. available for the island. Moreover, these mutually exclusive breeding ranges are mirrored by a comparable occupancy pattern of Dusky Megapode on neighbouring Batanta (I.M. unpublished data), where Aepypodius is absent till further notice (but see below). All this suggests that the brush-turkey‘s isolated montane breeding distribution may be better explained in terms of the unique evolutionary adaptation to upland forests of the genus Aepypodius. Recent molecular work brought to light that Dusky Megapode may be paraphyletic, M. f. quoyii and nominate freycinet each clustering with other species rather than as sisters in a ND2 mitochondrial gene tree (Birks & Edwards 2002). Such a relatively steep phylogenetic gap within Dusky Megapode suggests that Megapodius also

111 has an ancient history on Waigeo and that the disjunct breeding zones both species maintain are a result of a long-term and stable evolutionary status quo.

Bruijn‘s Brush-turkey presently is confirmed to breed at only three locations sensu IUCN (2001), and the known population totals 54 mound-owning males (Mauro 2004; 2006; I.M. unpublished data). The species‘ global population is estimated at 349 mound-owning males or 977 mature individuals, primarily confined to 60 km2 of inferred cloud-forest breeding habitat spread over six locations, with 94 % of breeding habitat and 98 % of the world population contained within three locations east of Mayalibit Bay (Mauro 2004; 2006). These figures pinpoint Bruijn‘s Brush-turkey among Indonesia‘s most restricted-range and rarer bird species (Stattersfield et al. 1998; BirdLife International 2000; 2001). Mt. Danai alone is believed to contain 60 % of breeding habitat and 65 % of the world population of Bruijn‘s Brush-turkey, and the paramount importance of this mountain for the species‘ long-term survival cannot be overstated. The only location in Waigeo‘s western sector to conceivably harbour a tiny population is restricted and isolated to such an extent that until a population is proven extant there indeed, I recommended querying the species‘ occurrence west of Mayalibit Bay (Mauro 2004; 2006).

Lowland occurrences of the Bruijn‘s Brush-turkey remain poorly understood. Six traceable positions of lowland records were on average only 6 km distant from the nearest known or inferred breeding habitat and lie east of Mayalibit Bay, and all refer to female type individuals (Mauro 2004; 2006). Some cases demonstrably coincided with prolonged periods of drought. Indeed, adult mound-tending males were found to endure extreme drought conditions on territory, whereas female type individuals appeared to have completely vacated optimum cloud-forest habitat. However, the species ranks as technically undetectable when dispersed in small numbers across vast lowlands hence the possibility looms that such apparent altitudinal shifting of female type individuals extends far beyond the casual disturbance- response and obscures juvenile dispersal and/or genetically strategic interchange of adults between isolated breeding habitats. As a matter of reasonable precaution, any in situ conservation initiative for the species should therefore soundly consider this potential temporal dependency upon lowlands of at least a subset of the population. Nonetheless, I recommended that these lowland occurrences be treated as cases of vagrancy with regard to applying criterion B of the IUCN Red List criteria (IUCN 2001).

While we now know more about Bruijn‘s Brush-turkey than ever before, the threats that impinge on the species too have never been greater. Waigeo consists predominantly of limestone and ultramafics, infertile substrates that are known to delay recovery of vegetation following environmental perturbation (overview in Proctor 2003). I identified habitat destruction as a result of wild fires as the most serious threatening process for Bruijn‘s Brush- turkey. Four traceable fire paths from the past 25 years averaged nearly 65 km2, together affecting c. 260 km2 or 8.4 % of the island‘s area (Mauro 2004; 2006). These known fire events acted over disproportionately large surfaces when compared to the size of individual locations of the species. Fires are directly responsible for severe internal degradation and partial obliteration of 18 % of cloud-forest on Mt. Sau Lal plus a presently unquantifiable yet significant portion of the Mnier Hills, and probably irretrievably degraded hundreds of km2 of low-lying forested areas, including corridors between locations. The impact of fire must also be assessed against the backdrop that the predominant scrub vegetation over ultramafic outcrops on Waigeo, long interpreted as an edaphic climax, now is considered to represent an early stage in a successional sequence induced by fire (Takeuchi 2003; see also Proctor 2003). Finally, three out of four wild fires occurred during the El Niño-induced drought event of 1982–83, but interviews indicated that most fires on the island are started deliberately, and the use of fire certainly increased significantly in the wake of logging operations.

In the accessible lowlands of Waigeo, forest was being lost at an extremely alarming pace due to illegal logging from 1999 to 2005. About 14 % of the West Waigeo Nature Reserve was already affected in 2000 (Anonymous 2000), and while commercial timber extraction was halted in mid-2005 following a large-scale crack-down operation throughout Indonesian Papua, evidence is mounting that most of the accessible tracts within the reserve have been severely degraded by the usual interplay of logging, road building, and the widespread subsequent use of fire as a cover up. While timber extraction likely will only affect low-lying

112 areas in the short term, the preservation of lowland forest corridors linking breeding habitats of Bruijn‘s Brush-turkey is imperative in view of a potential temporal dependency upon lowlands of at least a subset of the population.

The potential for nickel extraction is an old curse resting on Waigeo‘s mineral-rich northern ultramafic belt, and very recently loomed up again (Hermanto 2007). During the course of 2005 and 2006, the government of Raja Ampat regency has granted concession rights for nickel exploration in 11 contiguous blocks covering c. 900 km2 of land over ultramafic rocks, or 29 % of Waigeo‘s area. Included are the entire Mnier–Werar Hills and sizeable parts of the Mount Sau Lal–Waimila ridge system, respectively harbouring the world‘s second and third largest populations of Bruijn‘s Brush-turkey. Concessionaires have started to move ground during the course of 2006, and earlier this year the Raja Ampat government has approved actual exploitation in three blocks. Fortunately, this decision was promptly overruled at provincial level on account of these operations being in conflict with the protected area network on Waigeo. However, a reduction in size of the strict nature reserves on the island in order to accommodate mining has long been hanging and it appears that the government of Raja Ampat is now actively pursuing this option. It is feared that the materialization of these mineral extraction projects potentially could signify the total obliteration of breeding habitat of Bruijn‘s Brush-turkey within the ultramafic belt, and entail a reduction in the species‘ global population by as much as 34 % in the short- to medium-term future. Clearly, the development of the novel Raja Ampat regency, with Waisai on Waigeo‘s south coast as its capital, impels very attentive monitoring.

My work suggests that human predation, contrary to previous speculations (e.g. Dekker & McGowan 1995; Moeliker 2002ab), currently is a negligible threat to the population, and that introduced mammalian predators may be a more serious cause for concern. I listed the prehistorically introduced Wild Boar Sus scrofa as a potential nest predator of the Bruijn‘s Brush-turkey (Mauro 2004), and recently found mound raiding by boars to be at least locally rampant on Mt. Danai in April and June 2007 (Mauro et al., in prep.). Runaway dogs are also a problem locally in Waigeo‘s forests, and potentially imply a hazard to Bruijn‘s Brush-turkey considering its behavioural reluctance to flush (Mauro 2002; 2004; 2005; 2006). In addition, my work suggests that natural predation by the native monitor lizard Varanus indicus kalabeck, which unexpectedly manages to exist near permanent water sources up to at least 840 m elevation locally in the mountains of Waigeo (I.M. unpublished data), may pose a contributory threat, especially where subpopulations already are dwindling and territories occur at low densities such as on Mt. Nok.

I formally proposed that Bruijn‘s Brush-turkey, currently treated as Vulnerable (BirdLife International 2001), be upgraded to Endangered based on criteria EN B1+2ab(ii,iii,v) (IUCN 2001; Mauro 2004; 2006). The species is almost certainly confined to an area of 1,734 km2 in the eastern half of Waigeo and Extent of Occurrence as delineated by Minimum Convex Polygon (MCP) comprising all locations there amounts to 751 km2 [<5,000 km2; criterion B1]. The species‘ Area of Occupancy (AOO) now stands at 24 km2 and even if all locations within its Extent of Occurrence eventually should prove to be fully saturated at the IUCN 4 km 2- gridcell reference scale, still could not exceed 192 km2 [<500 km2; criterion B2]. Viable populations that are safe from genetic stochasticity exist at only three locations [<5 locations; qualifier a]. Fires severely degraded and partially obliterated 18 % of cloud-forest on Mt. Sau Lal plus a significant portion of the Mnier Hills, and quite possibly irretrievably degraded hundreds of km2 of low-lying forested areas including corridors between disjunct breeding zones [qualifier b(ii,iii)]. A continuing decline in number of mature individuals is inferred from the species‘ observed constrained population density in post-fire successional regrowth on Mt. Sau Lal [qualifier b(v)] (Mauro 2004; 2006).

Where from here?

I make the following recommendations:

[1] Urgently invest in preferably immediately effective cash-generating for-profit project interventions in order to alleviate increasing claims to deleterious resource-extraction.

113

Following endorsement of law No. 21/2001 on special autonomy for Indonesian Papua, customary land rights over forested land — previously considered state land — increasingly are being reinstated throughout the territory. While this, theoretically, tips the balance of power towards the communities, and while traditional societies may appear more receptive to long-term conservation goals than governments without adequate budget, I would at this point in time very cautiously say that it remains to be seen whether the conservation outlook for Bruijn‘s Brush-turkey can really be enhanced under this scenario.

Traditional societies on Waigeo are rapidly being overtaken by the 21st century. Adherence to customary law, ownership and tenure of land, reefs and natural resources slackens as integration into the market-oriented cash economy accelerates rapidly. Growing cash- dependency combined with dramatically increased prices of basic commodities and services deeply impact the islanders‘ livelihood. Hence to ease the financial burden, there are enormous incentives from the communities‘ perspective to sell off natural resources on customary-owned ancestral land. Internal conflicts emerge because young generations increasingly reject the limitations on resource exploitation imposed by customary so-called Sasi regimes that essentially evolved to ensure long-term sustainability through moderation and temporary abstinence. Waigeo‘s population is also structurally diverse (Masinambouw 1983; Silzer & Clouse 1991), further complicating the issue of customary land and resource ownership because original inhabitants, ancient migrants, and indeed any community inhabiting a given sector for a prolonged period of time, all readily claim access rights beyond subsistence use (Donnelly et al. 2003; I.M. unpublished data). Finally, another important element is the fact that the private sector as a whole essentially took over the central role of governments in providing basic services, and because the communities often are not principally opposed to resource extraction in itself but merely expect to reciprocally benefit from it, convincing them not to enter into novel, or terminate existing, unsustainable resource exploitation might prove exceedingly difficult, even if they do understand the long-term deleterious impact of such practices. To seriously rival destructive resource use, any conservation project intervention therefore likely will require immediately effective cash- generating alternatives, enabling the communities to earn sufficient hard cash in order to obtain, at the very least, those basic goods and services that cannot be fulfilled from their subsistence way of living. However, the development expectations of local communities are already known to extend beyond merely acquiring basic goods and services, and it can also be expected that companies responding by evincing greater corporate responsibility through more equitable profit-sharing with compliant communities, due to the remunerative nature of unsustainable exploitation, will prove unchallengeable. Let the ‗battle for the hearts and minds‘ of traditional societies living inside biodiversity hotspots begin!

[2] Through an elaborative process of participatory mapping and consultation with local communities, secure a consensus protected area east of Mayalibit Bay containing at least the c. 750 km2 convex hull that incorporates and interconnects all known and inferred breeding zones through overall well-forested corridors covering the species‘ complete altitudinal range.

The whole of the world population of Bruijn‘s Brush-turkey is believed to occur east of Mayalibit Bay and this has far-stretching implications for conservation planning and prioritization on Waigeo. Based on the principle of representation, both in terms of species diversity as habitats, the region east of Mayalibit Bay must now immediately be upgraded as the key area for terrestrial conservation on Waigeo and project intervention a priori should serve that sector. The East Waigeo Nature Reserve covers 1,195 km2 or 69 % of the island‘s eastern half (Ministry of Forestry Decree 251/Kpts-II/1996 d.d. 25 November 1996; Donnelly et al. 2003) and, quite fortunately, largely includes the c. 750 km2 convex hull delineating the species‘ Extent of Occurrence. While no significant management on the terrain has been forthcoming since its establishment in 1996, the reserve at least has clearly shown its value in legally halting proposed mining operations contravening with its boundaries while already approved by customary landowners. I am therefore highly convinced that the ‗consensus‘ protected area should be within the legal framework of the national conservation network. But in order to secure a viable lowland forest corridor between Mt. Sau Lal and Mt. Danai, a concerted effort must somehow be made to include into the reserve the Siam and Lapon R. valleys, presently classified as production forest. However, local communities and customary

114 landowners were not consulted at the time when the reserve was created in 1996, and reopening the discussion now, under the present circumstances, has the potential to trigger a complete rejection of the reserve.

[3] Declare the species‘ core locations as strictly no hunting areas and restrict access for scientific monitoring only.

This measure should not pose any foreseeable problems to local communities since these highest ridges remain entirely unexploited, if ever visited by man even. A complete hunting ban on wild boar is probably highly unrealistic and even undesirable in view of locally rampant nest predation (see above).

[4] Set up an island-wide awareness campaign to prevent future wild fires, the majority of which occur because of either imprudence or deliberate action of man.

[5] Restrict ‗parforce‘ pig-hunting and snaring to designated sectors near habitation and subsistence areas.

[6] Conduct additional field work in order to ground-truth current population estimates for Mt. Danai and the Mnier–Werar Hills complex, and the remaining unsurveyed ridges and peaks identified by Mauro (2004; 2006).

Mt. Danai is believed to contain in excess of 60 % of inferred breeding habitat and 65 % of the estimated global population of Bruijn‘s Brush-turkey. While I recently ascertained the presence of the species on Mt. Danai in itself, ground-truthing is urgently required to confirm these figures. All the more, given that the long-term survival of populations on Mt. Sau Lal and in the Mnier-Werar Hills, could now rapidly come under extreme pressure through the development of proposed mineral extraction projects (see above).

[7] Establish beyond reasonable doubt whether the species is indeed absent from Batanta.

There is a contemporary report of a ‗large brush-turkey‘ of undetermined generic identity, yet speculated to involve Bruijn‘s Brush-turkey, from 700 m elevation on Mt. Batanta in the centre of the eponymous island (Diamond 1986). Yet two lines of albeit circumstantial evidence suggest occurrence of the species there is not particularly likely. Firstly, an ornithological expedition in 1964 collected 11 field days within cloud-forest in the summit area of this mountain without encountering the species (Greenway 1966). Second, my work showed that the mix of strictly montane bird populations persisting on Waigeo and Batanta respectively is totally dissimilar. However, if Bruijn‘s Brush-turkey, after all, would be proven extant on the latter island, Mt. Batanta, supporting 47 km2 of optimum breeding habitat above the 620 m contour, in effect could support the species‘ most viable subpopulation. Negative evidence is hard to prove but a meticulous search of Mt. Batanta is clearly warranted in view of the species‘ sheer elusiveness.

[8] Investigate the impact of especially alien mammalian predators on Waigeo.

[9] Study nesting site philopatry and the extent of gene flow across locations, preferably using a non-invasive molecular technique.

[10] Study the impact of ENSO-induced drought events on the species‘ reproductive success.

The impact of ENSO-induced drought events on the breeding activity of mound-building megapodes has not been studied. However, Bruijn‘s Brush-turkey may be a particularly vulnerable species in this respect because drought has the capacity to directly reset the reproductive cycle to the following year‘s nesting period when coinciding with the species‘ habitual breeding season in a seasonally wet environment like Waigeo, and my work already provided circumstantial evidence for this (I.M. unpublished data). Moreover, because of their comparatively lower critical mass of near-exclusive organic composition Aepypodius mounds decompose and cool much faster when drought conditions start impinging upon the microbial activity that generates heat; Sinclair (2001) showed that they are less stable homeotherms.

115

Acknowledgments

The Australian National Malleefowl Recovery Team provided me with the opportunity to speak at the ‗National Malleefowl Forum and First Western Australian Megapode Symposium‘ held Katanning, September 7–11, 2007, and S. Davies kindly raised adequate travel funds. R. W. R. J. Dekker, in his capacity of chairman of the WPA/BirdLife/SSC Megapode Specialist Group (MSG), has been a constant source of advice, logistical support and encouragement. Field work was funded through two small grants of the Van Tienhoven Foundation for International Nature Protection and a donation received from Project Bird Watch/Indonesian Parrot Project. To all these people and institutions I am greatly indebted and extend my sincerest thanks. Above all, I am grateful to the people of Waigeo, the village authorities and landowners who granted permission to trespass traditionally owned land, and especially the more than 20 field assistants who contributed immensely to the results presented here, and made this project both a successful and enjoyable one.

References

Anonymous (2000). Perubahan status sebagian kawasan hutan Cagar Alam Waigeo Barat [Status change of a part of the forest area West Waigeo Nature Reserve.]. Technical Paper, BKSDA Sorong.

Anonymous, (2001). Long-lost bird raises its head. Science 291:2309.

Anonymous, (2002a). Bruijn‘s Brush-turkey seen. World Birdwatch 24:5.

Anonymous, (2002b). Observations of Bruijn‘s Brush-turkey Aepypodius bruijnii on Waigeo, Papua. OBC Bulletin 36:40–41, 49.

BirdLife International, (2000). Threatened birds of the world. Lynx Edicions and BirdLife International, Barcelona and Cambridge.

BirdLife International, (2001). Threatened Birds of Asia: The BirdLife International Red Data Book. BirdLife International. Cambridge.

Birks, S. M. & Edwards, S. V. (2002). A phylogeny of the megapodes (Aves: Megapodiidae) based on nuclear and mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 23:408–421.

Dekker, R. W. R. J. (2000). News about Bruijn‘s Brush-turkey and Waigeo. Megapode Newsletter 14:7–8.

Dekker, R. W. R. J. (2002). Boshoen. NRC Handelsblad 6 juli 2002.

Dekker, R. W. R. J. (2003). Megapode Specialist Group: Mounds “all over the place”. In: The World Pheasant Association Review 2002/3. A report on the activities of the World Pheasant Association. WPA, Fordingsbridge.

Dekker, R. W. R. J. & Argeloo, M. (1993). Bruijn‘s Brush-turkey Aepypodius bruijnii remains a mystery. Megapode Newsletter 7:15–17.

Dekker, R. W. R. J. & McGowan, J. K. (1995). Megapodes: An action plan for their conservation 1995–1999. WPA/BirdLife/SSC Megapode Specialist Group. Gland, Switzerland, Cambridge and Reading, U.K.: IUCN and World Pheasant Association.

Dekker, R. W. R. J., Fuller, R. A. & Baker, G. C., Eds. (2000). Megapodes: Status survey and conservation action plan 2000–2004. WPA/BirdLife/SSC Megapode Specialist Group.

116 Gland, Switzerland, Cambridge and Reading, U.K.: IUCN and World Pheasant Association.

De Schauensee, R. M. (1940). Rediscovery of the megapode, Aepypodius bruynii. Auk 57:83–84.

Diamond, J. M. (1985). New Distributional records and taxa from the outlying mountain ranges of New Guinea. Emu 85:65–91.

Diamond, J. (1986). Proposed studies of conservation problems on the western Papuan islands (Raja Ampat Islands), Irian Jaya. World Wide Fund for Nature Indonesia. Unpublished Report.

Donnelly, R., Neville, D. & Mous, P. J. (Eds.) (2003). Report on a rapid ecological assessment of the Raja Ampat Islands, Papua, Eastern Indonesia, held October 30 – November 22, 2002. Final Draft November 2003. The Nature Conservancy - Southeast Asia Center for Marine Protected Areas, Bali.

Frith, C. B. & Beehler, B. M. (1998). The Birds of Paradise Paradisaeidae. Oxford Univesity Press. Oxford.

Greenway, J. C. Jr. (1966). Birds Collected on Batanta, off Western New Guinea, by E. Thomas Gilliard in 1964. Amer. Mus. Novit. 2258:1–27.

Heij, C. J. & Post, J. N. J. (2001). Bruijn‘s Brush-turkey Aepypodius bruijnii rediscovered on Waigeo. Megapode Newsletter 15:2–5.

Hermanto, (2007). Bruijn's Brush-turkey Aepypodius bruijnii: uplist to Endangered? BirdLife International Globally Threatened Bird Forums, accessible from http://www.birdlifeforums.org/[email protected]@.2cba57d4

IUCN, (2001). IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, UK.

Jones, D. N. (1990a). Male mating tactics in a promiscuous megapode: patterns of incubation mound ownership. Behav. Ecol. 1:107–115.

Jones, D. N. (1990b). Social organisation and sexual interactions in Australian brush-turkeys (Alectura lathami): implications of promiscuity in a mound-building megapode. Ethology 84:89–104.

Jones, D. N., Dekker, R. W. R. J. & Roselaar, C. S. (1995). The Megapodes Megapodiidae. Oxford University Press. Oxford.

Marchant, S. & Higgins, P. J. (Eds.) (1993). Handbook of Australian, New Zealand & Antarctic Birds. Volume 2: Raptors to Lapwings. Oxford University Press, Melbourne.

Masinambouw, E. K. M. (Ed.) (1983). Halmahera dan Raja Ampat sebagai Kesatuan Majemuk [Halmahera and Raja Ampat as a Multiple-Unity]. LIPI-LEKNAS, Jakarta.

Mauro, I. (2002). Stop Press! 08/06/02 Bruijn‘s Brush-turkey Aepypodius bruijnii for the first time observed in the wild. Megapode Newsletter 16:2–3.

Mauro, I. (2003). Viable population of Bruijn‘s Brush-turkey Aepypodius bruijnii discovered. Megapode Newsletter 17:1.

Mauro, I. (2004). The field discovery, ecology, monitoring and conservation of an enigma: Bruijn’s Brush-turkey Aepypodius bruijnii Oustalet 1880. Final report to Van Tienhoven

117 Foundation for International Nature Protection and WPA/BirdLife/SSC Megapode Specialist Group.

Mauro, I. (2005). Field discovery, mound characteristics, bare parts, vocalisations and behaviour of Bruijn‘s Brush-turkey, Aepypodius bruijnii. Emu 105:273-281. doi: 10.1071/MU04052.

Mauro, I. (2006). Habitat, microdistribution and conservation status of the enigmatic Bruijn‘s Brush-turkey Aepypodius bruijnii. Bird Conservation International 16:279-292. doi: 10.1017/S0959270906000372.

Mauro, I., Wonggor, Z., & Wijaya, L. (in prep.). Nest predation by Wild Boar Sus scrofa in the Bruijn’s Brush-turkey Aepypodius bruijnii.

Mayr, E. & Diamond, J. M. (2001). The Birds of Northern Melanesia. Speciation, ecology and biogeography. Oxford University Press. New York.

Moeliker, C. W. (2002a). Gered uit de kookpot. NRC Handelsblad (Achterpagina) 9 September 2002.

Moeliker, C. W. (2002b). DB Actueel: Bruijn‘s Brush-turkey comes to life. Dutch Birding 24:323–324.

Moeliker, C. W. (2002c). Bruijn‘s Boshoen komt tot leven. Straatgras 14:26.

Moeliker, C. W., Heij, C. J., Post, J. N. J. & Kompanje, E. J. O. (2003). A new, a forgotten and a lost specimen of Bruijn‘s Brush-turkey Aepypodius bruijnii. Megapode Newsletter 17:2–8.

Oustalet, E. (1880). Monographie des oiseaux de la famille des Mégapodiidés. Bibliothèque de l‘Ecole des Hautes Etudes 22(5).

Paijmans, K. (Ed.) (1976). New Guinea vegetation. CSIRO and Australian National University Press, Canberra.

Peters, J. L. (1934). Check-list of birds of the world. Volume 2. Harvard University Press, Cambridge.

Post, J. N. J. (2001). De herontdekking van het Bruijn‘s Boshoen (Aepypodius bruijnii). Straatgras 13:2–7.

Proctor, J. (2003). Vegetation and soil and plant chemistry on ultramafic rocks in the tropical Far East. 6: Perspectives in Plant Ecology, Evolution and Systematics 6:105-124.

Rand, A. L. & Gilliard, E. T. (1967). The handbook of the New Guinea birds. Weidenfeld & Nicolson. London.

Ripley, S. D. (1960). Distribution and niche differentiation in species of megapodes in the Moluccas and Western Papuan area. Acta XII Congressus Internationalis Ornithologici 1:631–640.

Rosenberg, M. S. (2001). PASSAGE. Pattern Analysis, Spatial Statistics, and Geographic Exegesis. Version 1.1.1.3. Department of Biology, Arizona State University, Tempe, AZ.

Rothschild, W., Stresemann, E. & Paludan, K. (1932). Ornithologische Ergebnisse der Expedition Stein 1931-1932. Novit. Zool. 28:127–247.

Silzer, J. P. & Clouse, H. H. (1991). Index of Irian Jaya Languages. Uncen and SIL, Jayapura.

118 Sinclair, J. R. (2001). Temperature regulation in mounds of three sympatric species of megapode (Aves: Megapodiidae) in Papua New Guinea: testing the ‗Seymour Model. Australian Journal of Zoology 49:675–694.

Sinclair, J. R. (2002). Selection of incubation mound sites by three sympatric megapodes in Papua New Guinea. The Condor 104:396–406.

Stattersfield, A. J., Crosby, M. J., Long, A. J. & Wege, D. C. (1998). Endemic Bird Areas of the world: priorities for biodiversity conservation. BirdLife International. Cambridge.

Takeuchi, W. (2003). An Ecological Summary of the Raja Ampat Vegetation. In: Donnelly R., Neville, D. and Mous P. J. (Eds.). Report on a rapid ecological assessment of the Raja Ampat Islands, Papua, Eastern Indonesia, held October 30 – November 22, 2002. Final Draft November 2003. The Nature Conservancy - Southeast Asia Center for Marine Protected Areas, Bali. van Royen, P. (1960). Sertulum Papuanum 3. The vegetation of some parts of Waigeo Island. Nova Guinea, New Series, Bot. 10:25–62.

Voisin, C., Voisin J-F., Somadikarta, S. & Dekker, R. W. R. J. (2000). Six overlooked specimens of Bruijn‘s Brush-turkey Aepypodius bruijnii (Oustalet, 1880). Bull. B.O.C. 120:146–148.

119

14. MEGAPODES IN NORTHERN AUSTRALIA: A SUMMARY OF RECENT RESEARCH ON THE ORANGE-FOOTED SCRUBFOWL AND AUSTRALIAN BRUSH TURKEY

Darryl N Jones Centre for Innovative Conservation Strategies Griffith University Nathan Qld 4111 e-mail: [email protected]

Abstract

Malleefowl are unquestionably the aberrant megapode: all of the other 21 species live in subtropical and tropical regions, mainly in dense rainforests where the construction of incubation mounds is relatively straighforward. Malleefowl are of particular interest because they have somehow managed to adapt what would have been a life in the jungle to millennia of relentless drying. Coping with extreme climate change is possible but requires sophisticated adaptations. Comparisons with the two more northern megapodes that occur in Australia – the Australian Brush-turkey and the Orange-footed Scrubfowl - provides an assessment of the extent of this adaptation in Malleefowl as well as an opportunity to learn how the other species are faring. Although many megapodes remain seriously threatened, these two Australian species have learned how to live successfully among people, in some locations such as Brisbane, Darwin and Cairns, becoming nuisances in urban areas. Despite its abundance, however, the scrubfowl remains poorly studied. On the other hand, continuing work on the brush-turkey has greatly enhanced our understanding of the species and of many aspects that are common to the family as a whole. Of particular importance is the recent discovery of a specialised form of temperature sex determination in brush-turkeys in which different levels of mound heat result in highly skewed sex ratios. This presentation will review some of these recent findings, highlighting issues of particular relevance to Malleefowl conservation and ecology.

Introduction

Malleefowl Leipoa ocellata are unquestionably the aberrant megapode: all of the other 21 species live in subtropical or tropical regions, mainly in dense moist rainforests where the construction and maintenance of incubation mounds is relatively straightforward (Jones et al. 1995). Malleefowl are of particular interest because they are survivors of the Pre-Pleistocene era when much of Australia was covered in moist non-sclerophyllous forests. Like the extinct giant megapode Progura gallinacea, the ancestors of present-day Malleefowl would presumably have constructed large mounds in similar fashion to most of the extant mound- building megapode species. It is a measure of its resilience and adaptability that this bird has been able to maintain the typical megapode incubation process in the face of continental drying that transformed almost every element of mound construction and operation to be able to continue undeterred into the present day.

The primary focus of the 2007 Malleefowl Forum was on the current desperate conservation status of the Malleefowl in most parts of its range and on the task of recovery and securing the species‘ survival. The status of the other two Australian megapodes is, thankfully, quite different: both the Orange-footed Scrubfowl Megapodius reinwardt and the Australian Brush- turkey Alectura lathami are classified as of Least Concern, and both occur in most of their historical range (Dekker et al. 2000). Both are common and even abundant and increasing in many locations, especially in populations that have recently started to exploit urban environments (see below). There is, however, considerable contrast between the species in terms of the amount and type of research conducted on each. The Malleefowl, while only one of numerous megapode species that is critically threatened (Dekker et al. 2000), has been the subject of by the far the largest number of conservation biology studies. On the other hand, the Australian Brush-turkey, being the only abundant species to live in and near two of Australian largest cities (Sydney and Briisbane), is the best studied of all megapodes (Jones

120 1999). Despite occurring in Australia, the only developed country to support megapodes, the largely tropical and remote distribution of the Orange-footed Scrubfowl, as well as its secure conservation status, has resulted in relatively little research attention. The aim of this paper is to provide a concise summary of some recent studies on both of the two megapodes occurring to the north of the range of the Malleefowl. This resume concentrates on work completed since the 3rd International Megapode Symposium held in Nhill, Victoria in 1997 (Dekker & Jones 1999).

Orange-footed Scrubfowl

The Orange-footed Scrubfowl has by far the largest distribution of all megapodes, being found in a multitude of islands from the Lesser Sunda Islands near Bali in Indonesia, across northern Australia and most of southern New Guinea to the Trobrian Islands off the eastern tip of Papua (Jones et al. 1995). Nonetheless, since the first behavioural studies conducted near Cairns in the late 1970s (Crome & Brown 1979), the only direct research on the species has been conducted in coastal Northern Territory, primarily around Darwin. Palmer et al. (2000) investigated the site characteristics of incubation mounds in different forest types and found that ease of leaf litter collection was a major influence. While such organic matter was the primary source of incubation heat, solar radiation was also important in the sandy mounds constructed in coastal Vine thickets. Franklin and Baker (2005) investigated the increasing presence of the species in suburban Darwin and found that they lived and constructed their mounds mainly in areas of forest immediately adjacent to house-yards, into which they venture to forage opportunistically.

Very recently, important studies on the species have been carried out on the Indonesian Island of Komodo as part of continuing research into the Komodo dragon Varanus komodensis by Australian and Indonesian researchers (Jessop et al. in press). These workers conducted extensive surveys of mounds throughout the island to assess the features used by the birds in siting their mounds, and found that the birds preferred sandy or loamy soils in open monsoon forest with minimal shading. On Komodo, nesting activity began in the late dry season (October) and peaked during the late wet season in March. The mounds are crucial to reproduction by the Komodo dragons, 70% of which incubate their eggs along with those of the scrubfowl in the mounds. Nonetheless, there were relatively low levels of egg predation (<17%) and human disturbance was negligible.

There remain many excellent and fascinating opportunities for further studies of this widespread and still abundant species, especially in northern Queensland. Especially important would be conservation-related studies of this secure species that could be applied to many of the threatened Megapodius species for which manipulative or replicated research may be impossible or too risky.

Australian Brush-turkey

The Australian Brush-turkey is currently widespread and, in many places especially near certain cities, abundant. Recent surveys in Brisbane (Jones et al. 2004) and Gosford (Göth 2007a) indicate a continuing growth in suburban populations, accompanied by a concomitant increase in the number of complaints from residents. Earlier studies (Jones and Everding 1991) had intimated that the presence of brush-turkeys in urban environments was likely to be temporary, principally because of the almost complete mortality of the defenceless young due to, mainly, predation by cats. The pronounced growth in the number of sites within the Brisbane suburbs over a ten-year period (Jones et al. 2004) indicates that at least some chicks obviously do survive. However, two anthropogenic activities are also likely to enhance the success of urban brush-turkey populations: the on-going (and mainly illegal) translocations of nuisance birds – usually mound-building males (Jones et al. 2004), and the amount of feeding of the birds, whether intended or not, by householders (Göth 2000a).

Despite these extnsions of localised populations, the brush-turkey‘s overall pre-European distribution has contracted and fragmented significantly, especially in southern New South Wales (Göth et al. 2006). A detailed survey of all available historical records of the species (Göth et al. 2006) indicated that, during the early 1800s, the species occurred as far south as

121 Cape Howe 37.5 S) on the New South Wales-Victoria border and even in the Snowy Mountains near Jindabyne. To the west, historical sightings include Nyngan, Pilliga and Moree, that do not exist today. The western-most New South Wales extant population is in Mt. Kaputar National Park (150.1 E) (although the most inland population is probably that in Carnarvon Gorge National Park in central Queensland (147 E), 450 km from the coast)).

Although the demise of these southern and western populations has unquestionably decreased the distribution of the species, it is still found almost continuously from Cape York (10.4 S) in Queensland to its current southern-most outpost near Cambewarra (34.8 S) below Wollongong - a remarkable latitudinal range of over 30 (Jones et al. 1995, Goth et al. 2006). Evolutionary biologists are interested in species with extensive distributions for a range of reasons such as the process of and extent of gene flow and evidence of speciation within populations at the extremities of the species‘ distribution. One particularly useful technique associated with this field has been comparisons of ectoparasites living on the plumage of the birds from different parts of the range. This approach was applied by Proctor and Jones (2004) to the brush-turkeys sampled throughout the entire range of the species in Queensland. Six species of feather mites and two species of feather lice were identified. However, the extent of sharing of these species, even among populations over 1000 km distant, indicated unexpectedly similar parasite communities and no evidence of population differentiation.

Among the most significant implications of the megapode incubation system is the complete absence of post-hatching parental care for the offspring. Despite being extremely precocial at emergence, the young nonetheless start their lives without protection, guidance or even the possibility of social learning. While many commentators (Priddel & Wheeler 1996, Jones 1999, Göth 2001) have identified this situation as being of critical importance to understanding much subsequent development and behaviour in megapodes, generally – especially in relation to survival of individuals in threatened populations - the daunting logistical challenges have greatly limited most research on the topic. The existence of numerous high-density populations of the Australian Brush-turkey has provided an ideal opportunity for detailed studies, most of which have been conducted by Dr Ann Göth. Working initially in southern Queensland, Dr Göth completed a comprehensive series of studies on aspects of predator recognition (Göth 2001), foraging preferences (Göth & Proctor 2002), ‗underground ‗behaviour‘ following hatching (Göth 2002), chick survival and habitat preferences (Göth & Vogel 2002, 2003), and the ontogeny of social behaviour (Göth & Jones 2003).

This work continued and focussed with Dr Göth‘s move to Sydney, where fortuitously her arrival coincided with a marked expansion of populations in the vicinity of the northern suburbs of the city. Sophisticated studies of carefully reared captive chicks at the University of Macquarie elucidated the importance of visual ((Göth & Evans 2004a,b) and auditory (Barry & Göth 2005) cues, while increasingly detailed experiments have explored the importance of learning and testosterone in the development of mound building (Göth & Astheimer 2006) and further details of the fascinating area of female choice of mounds and mates (Göth 2007b). It is no exaggeration to suggest that this coherent and remarkable body of work is revolutionising our understanding of many aspects of megapode ecology, behaviour and even evolution.

However, it is a discovery during routine experiments on incubation in brush-turkey eggs that is possibly the most important of all these studies: the possibility of temperature-dependent sex ratios. While the phenomenon of temperature-dependent sex determination is well-known among a wide range of reptile species (Janzen & Paukstis 1991), the possibility of a similar influence in birds has always been impossible simply because sex in birds is genotypically determined at the point of fertilisation (Pike and Petrie 2002). Nonetheless, sex ratios in some birds is known to differ dramatically from parity, due mainly to pre-fertilisation hormonal influences of the female, or differential treatment of the hatchings (Hasselquist & Kampenaers 2002). Although incubation spans the period between these pre-fertilisation and post-hatching influences on sex ratios, this fundamental process has been deemed to be an unlikely source of variability because the temperature regime of almost all birds - being produced by the brooding bird‘s body heat will be identical for all eggs within the nest. Megapodes, however,

122 differ from almost all other birds by incubating their eggs by exploiting external sources of environmental heat, which, although typically manipulated to facilitate thermal stability nonetheless exhibit significant differences between mounds and within the same mound (Booth & Jones 2002). Could this variation in incubation temperature lead to changes in the sex ratio of the resulting chicks?

Characteristically, Göth acknowledges that it was a conversation with a North Queensland aboriginal elder, Warren Canendo, which triggered this line of research. Warren was adamant that pronounced differences in the seasons resulted in different numbers of males or females (Göth & Booth 2005). While our current level of understanding would suggest that this was unlikely, Ann was sufficiently intrigued to test the idea experimentally. In the wild the incubation temperatures of brush-turkey mounds may vary from 30 C to 38 C though the average is about 34 C (Göth 2007c). Ann Göth and her colleague David Booth placed eggs collected from natural mounds in artificial incubators set at three temperatures well within this range: 31 , 34 and 38 C (Göth & Booth 2005). The results were clear: at the lower temperature, more males hatched, more females hatched at the higher temperature while the sex ratio at ‗normal‘ temperatures was exactly equal. However, this result was only evident during the earliest stage of development, approximately the first 10 days. After this stage, the sex ratio remained equal, regardless of temperature.

As well as numerous theoretical implications, there are also some important practical possibilities of this remarkable finding. First, it raises the possibility that females could influence the sex of their offspring through the careful selection of the placement of the eggs at the time of laying. This is certainly possible, as it is already well known that the birds are capable of assessing the temperature of the incubation substrate, an activity observed in all megapode species. Such apparently adaptive manipulations of sex ratio of offspring prior to birth or hatching has been found in other animals but this is not in previously in birds (Pike & Petrie 2002).

A further finding from these experiments was that incubation temperature also strongly influenced the mass but not size of hatchlings (Göth & Booth 2005, Göth 2007c)); chick weight increased significantly with incubation temperature. Other investigations (Göth 2002, Göth & Evans 2004a) have demonstrated that this has important consequences for a range of behaviours including the speed of emergence from the mound, and even the probability of survival of chicks (Göth & Vogel 2003). Generally, larger chicks live longer.

This is the second pragmatic finding to emerge directly from these other potential theoretical studies: conservation plans that include artificial incubation may need to assess the potential impacts of incubation temperature on the sex ratio of the chicks being produced and of influences on survival after release.

References

Barry, K.L. & Göth, A. (2005). Call recognistion in chicks of the Australian brush-turkey (Alectura lathami). Animal Cognition 9: 47-54.

Booth, D.T. & Jones, D.N. (2002). Underground nesting in the megapodes. In Avian Incubation: Behaviour, Environment and Evolution. (Ed. D.C. Deeming). Oxford University Press, Oxford.

Crome, F.H.J. & Brown, H.E. (1979). Notes on social organization and breeding of the orange-footed scrubfowl Megapodius reinwardt. Emu 79: 111-119.

Dekker, W.R.J., Fuller, R.A. & Baker, G.C. (2000). Megapodes Status Survey and Conservation Action, Plan 2000-2004. IUCN, Gland.

Dekker, W.R.J., Jones, D.N. & Bensehmesh, J. (1999). Proceeding of 3rd International Megapode Symposium, Nhill, Australia. Zoologische Verhandelingen, Leiden..

123 Franklin, D.C. & Baker, B. (2005). The orange-footed scrubfowl Megapodius reinwardt as an urban bird in Darwin, Northern Territory. Australian Field Ornithology 22: 48-50.

Göth, A. (2001). Innate predator-recognition in Australian brush-turkey (Alectura lathami, Megapodiiae) hatchlings. Behaviour 138: 117-136.

Göth, A. (2002). Behaviour of Australian brush-turkey (Alectura lathami, Galliformes: Megapodiidae) hatchlings following underground hatching. Journal of Ornithology 143: 477-488.

Göth, A. (2007a). Australian Brush-turkeys, Corridors and Human-Wildlife Conflicts. Report for Gosford Shire Council. Pp. 19.

Göth, A. (2007b). Mound and mate choice in a polyandrous megapode: Females lay more and larger eggs in nesting mounds with the best incubation temperatures. Auk 124: 253-263.

Göth, A. (2007c). Incubation temperatures and sex ratios in Australian brush-turkey (Alectura lathami) mounds. Austral Ecology 32: 378-385.

Göth, A. & Astheimer, L. (2006). Development of mound-building in Australian brush-turkeys (Alectura lathami): the role of learning, testosterone and body mass. Australian Journal of Zoology 54: 71-78.

Göth, A. & Evans, C.S. (2004a). Egg size in Australian brush-turkey Alectura lathami hatchlings predicts motor performance and post-natal weight gain. Canadian Journal of Zoology 82: 972-979.

Göth, A. & Evans, C.S. (2004b). Social responses without early experience: Australian brush- turkey chicks use specific visual cues to aggregate with conspecifics. Journal of Experimental Biology 207: 2199-2208.

Göth, A. & Booth, D.T. (2005). Temperature-dependent sex ratio in a bird. Biology Letters 1: 31-33.

Göth, A., Nicol, K.P, Ross, G. & Shields, J.J. (2006). Present and past distribution of Australian Brush-turkeys Alectura lathami in New South Wales – implications for management. Pacific Conservation Biology 12: 22-30.

Göth, A. & Proctor, H. (2002). Pecking preferences in hatchlings of the Australian brush- turkeys, Alectura lathami (Megapodiidae): the role of food type and colour. Australian Journal of Zoology 50: 93-102.

Göth, A. & Vogel, U. (2002). Social behaviour of two free-ranging chicks of the Australian brush-turkey Alectura lathami. Corella 26: 1-3.

Göth, A. & Vogel, U. (2003). Juvenile dispersal and habitat selectivity in the megapode Alectura lathami (Australian brush-turkey). Wildlife Research 30: 69-74.

Hasselquist, D. & Kempenars, B. (2002). Parental care and adaptive brood sex ratio manipulation in birds. Philosophical Transactions of the Royal Society (London) B 357: 363-372.

Janzen, F.J. & Paukstris, G.L. (1991). Environmental sex determination in reptiles: ecology, evolution and experimental design. Quarterly Review of Biology 66: 149-179.

Jessop, T.S., Sumner, J., Imansyah, M.J., Purawandana, D., Ariefiandy, A. & Seno, A. (in press). Assessment of the distribution, seasonal use and predation of orange-footed scrubfowl on Komodo Island, Indonesia. Journal of Field Ornithology

124 Jones, D.N. (1999). What we don‘t know about megapodes. In Dekker, R.W.R.J., Jones, D.N. & Bensehmesh, J. (eds.) Proceeding of 3rd International Megapode Symposium, Nhill, Australia. Zoologische Verhandelingen 327: 159-168.

Jones, D.N., Dekker, R.W.R.J. & Roselaar, C.S. (1995). The Megapodes. Oxford University Press, Oxford.

Jones, D.N. & Everding, S.E. (1991). Australian brush-turkeys in a suburban environment: implications for conflict and conservation. Wildlife Research 18: 285-297.

Jones, D.N., Sonnenberg, R. & Sinden, K.E. (2004). Presence and distribution of Australian brush-turkeys in the Greater Brisbane region. Sunbird 34: 1-9.

Palmer, C. Christian, K.A. & Fisher, A. (2000). Mound characteristics and behaviour of the orange-footed scrubfowl in the seasonal tropics of Australia. Emu 100: 54-63.

Pike, T.W. & Petrie, M. (2002). Potential mechanisms of avian sex manipulation. Biological Reviews 78: 533-574.

Priddel, D. & Wheeler, R. (1996). Effect of age at release on the susceptibility of captive- reared Malleefowl Leipoa ocellata to predation by the introduced fox Vulpes vulpes. Wildlife Research 21: 543-552.

Proctor, H.C. & Jones, D.N. (2004). Geographical structuring of feather mite assemblages from the Australian Brush-turkey (Aves: Megapodiidae). Journal of Parasitology 90: 60-66.

Rich, P.V. & van Tets, G.F. (1985). Kadimakara: Extinct Vertebrates of Australia. Pioneer Design Studios, Lilydale.

125

15. MALLEEFOWL IN THE WESTERN AUSTRALIAN WHEATBELT; AN ECOLOGICAL STUDY INFORMED BY COMMUNITY KNOWLEDGE

Blair Parsons CSIRO Sustainable Ecosystems, Private Bag 5, Wembley WA, 6913, Australia. e-mail: [email protected]

Abstract

The Malleefowl (Leipoa ocellata) is listed as vulnerable across Australia and the National Recovery Plan suggests that its range has contracted by 45% in Western Australia. Estimates of decline are based on detailed studies of single populations or continent wide assessments based on presence only data. This study investigated whether presence-only data could reliably assess the status of Malleefowl by comparing presence with presence- absence data from Western Australia. The study also quantified changes in the range of Malleefowl in Western Australia and determined the relative influence that various threatening processes (e.g. land clearing, agricultural development), may have had on its range.

Presence-only data were vulnerable to false absence and bias in observer effort, but the study demonstrated a method that overcomes these weaknesses by verifying areas of apparent absence, resulting in the production of a presence/absence dataset. Malleefowl have suffered a range contraction within Western Australia but this contraction is less substantial than previously claimed. The contraction in range within the agricultural landscapes of south-west Western Australia is associated with extent of land clearing, the number of years since commencement of agricultural activity, and the number of sheep within a landscape. To conserve Malleefowl, landscapes developed for agriculture in recent decades must be protected to ensure they do not develop attributes found in landscapes that have been heavily cleared and occupied since the early 1900s.

In addition to the work detailed above, preliminary findings are presented from work aimed at understanding the role that fire plays in the ecology of Malleefowl in the Western Australian wheatbelt. This research utilises remote sensing to quantify the frequency and extent of wildfires in the Western Australian wheatbelt and vegetation survey to document how Malleefowl habitat recovers from such fire events.

Introduction

The Malleefowl (Leipoa ocellata) is one of many terrestrial fauna species that has experienced a decline in its distribution and abundance in Australia (Burbidge & McKenzie 1989; Recher & Lim 1990; Garnett & Crowley 2000). It is listed as endangered at the national level and the national recovery plan (Benshemesh 2000) suggests a contraction of 45% in the western portion of its range. Studies in south-eastern Australia (Frith 1962, Priddel & Wheeler 1999) have reported rapid and widespread declines of Malleefowl in response to agricultural development and the introduction of exotic predators. These threatening processes operate in Western Australia also and so there is concern that similar declines may have occurred. However, previous estimates of decline have been restricted to detailed studies of individual populations (e.g. Priddel & Wheeler 2003) or continent wide assessments based on presence-only data (Benshemesh 2000). Despite the wealth of research conducted on this species, there is uncertainty about the extent or even reality of decline in Western Australia with claims for both decline (Benshemesh, 2000) and increases in range/abundance (Serventy & Whittell, 1976).

There is a need to quantify more accurately any changes in distribution and abundance of Malleefowl in Western Australia and assess the relative contributions of various processes that threaten Malleefowl persistence. Currently, a wealth of community collected presence- only data exists for the species and despite the recognised shortcomings of this type of data (Austin 2002; Wintle et al. 2005) it has potential to answer key questions about the status of Malleefowl.

126

This study addressed the following questions: . Can we use presence-only data to assess changes in species distribution? . Has the Malleefowl suffered a range contraction within the Western Australian wheatbelt? . Is there a relationship between changes in the range of Malleefowl and landscape- scale environmental predictors?

Methods

The study sought to quantify changes in the range of Malleefowl in the Western Australian wheatbelt by (i) basing methods on those used in the national recovery plan and applying them to an expanded dataset of community collected sightings data dating from 1839 to 2006; and (ii) collecting presence-absence data from landholders and community members, allowing a rigorous examination of the decline. Areas where Malleefowl populations were apparently stable were compared with those with few recent (post-1990) sightings, and various landscape attributes were considered as possible causes of change. Issues relating to false absence were also explored.

Results

The results suggest that Malleefowl have suffered a range contraction within Western Australia but this contraction is smaller than previously claimed. In contrast to south-eastern Australia, Malleefowl appear to have persisted across much of their former range within the agricultural landscapes of south-west Western Australia, with declines largely confined to the western margin of the species‘ former range. The contraction in the range of Malleefowl was correlated with the number of years since commencement of agricultural activity and also with extent of land clearing. Analysis of sightings data also showed that false absences had the potential to exert substantial influence on estimates of species decline.

Conclusion

The Malleefowl has suffered less of a range contraction in Western Australia than previously claimed, particularly within the Western Australian wheatbelt. This conclusion is in part due to the compilation of a larger dataset than was previously available to researchers assessing this species. However, it is also likely that the species has persisted within large areas of the wheatbelt because agricultural development has been relatively recent (< 50 years ago) compared with areas in eastern Australia. As a result, these landscapes are of higher quality in that they are less fragmented and the remaining habitat has had less exposure to agents of degradation such as grazing, altered fire regimes and weed invasion. It is probable that the decline of Malleefowl will continue unless these processes are mitigated. The correlation between Malleefowl decline and extent of land clearing suggests that the future extinction of Malleefowl from highly fragmented areas within the wheatbelt is likely, particularly given the presence of exotic predators.

This study was limited to making broad estimates of Malleefowl decline using relatively coarse-scale spatial data but nevertheless represents a step forward in the use of presence- only data to assess species decline. Presence-only data can provide a useful starting point for understanding patterns of decline, provided an appreciation of bias within the data (e.g. false absence) is incorporated into the analysis. Collection of absence data via community survey is an effective way to account for such bias on a broad scale, particularly when studying a high-profile iconic species such as the Malleefowl. However, collection of spatially and temporally structured presence-absence data across a range of environmental gradients is necessary to provide rigorous assessments of population trends in decline over time.

Fire regimes of Malleefowl habitat in the Western Australian wheatbelt.

Wildfire is listed as a major threat to the persistence of Malleefowl (Benshemesh 2000) and it has been suggested that Malleefowl require habitat that is 30 to 60+ years post fire in order to maintain breeding populations (Benshemesh 1992). In eastern Australia, this is cause for

127 concern as Malleefowl habitat that meets such criteria is rare. However, in the Western Australian wheatbelt, the threat posed by fire has not previously been assessed. The frequency and spatial extent of fire events is not known, and the manner in which different habitats respond to fire has not been adequately studied.

This study seeks to understand the role that fire plays in the ecology of Malleefowl in the Western Australian wheatbelt. More specifically, it seeks to answer the following questions: 1) What are the frequency and extent of fire events in remnants in the Western Australian wheatbelt? Does this regime differ as remnant size changes? 2) What is the response of known Malleefowl habitats to fire? Are there differences in vulnerability/ recovery time to fire between different Malleefowl habitats (e.g. mallee, Acacia/Allocasuarina shrublands)?

Preliminary results

The frequency and extent of fire events in remnants in the Western Australian wheatbelt post 1988 were identified using a series of LANDSAT satellite images. Images existed for every second year between 1988 and 2004. Fire regimes were quantified for three categories of native vegetation: 1. Small remnants (100 - 500 ha); 2. Large remnants ( > 500 ha) 3. Continuous vegetation (within pastoral country adjacent to wheatbelt).

For the small remnant category, a sample consisted of the entire remnant. For the other two categories, 500 ha circular sample areas were taken from within the remnant. The position of these sample areas was determined at random. This method was used to maintain balanced sample effort between remnant categories.

Fire regimes for Malleefowl habitat in the wheatbelt are summarised in Table 1 below.

Table 1. The effect of fire on remnants of different size.

Sample remnants

Small Large Pastoral Number of samples 127 156 30 Burnt 1988-2004 (%) 5 16 33 Burnt prior to 1988 (%) 7 24 40 Avg proportion burnt (%) 26 48 69

The findings show that fire frequency and extent are low in small remnants within the Western Australian wheatbelt and suggest that fire does not represent a threat within these areas. Therefore, it may be more appropriate for conservation practitioners to focus on predator control or other threats that are active within small remnants. However, as remnant size increases (i.e. >500 ha), fire frequency and extent increases, suggesting that in large remnants or continuous vegetation, fire may represent a significant threat to the persistence of Malleefowl.

Future work

The response of known Malleefowl habitats to fire will be explored in a space for time study. The study will focus on the two habitat types most commonly occupied by Malleefowl within the Western Australian wheatbelt: (1) Mallee (most prominent in the southern wheatbelt); and (2) Acacia/Allocasuarina shrublands (most prominent in the northern wheatbelt).

Habitat attributes of direct relevance to Malleefowl (e.g. canopy height and cover, litter accumulation) will be measured in areas of varying post-fire age. These areas will range from

128 burnt pre-1968 to burnt in 2004. Measurements will be used to determine how long it takes for burnt habitat to resemble that of a long unburnt area.

By determining how long it takes for burnt habitat to structurally resemble that of long unburnt habitat, the study will help inform land managers about appropriate fire regimes for the conservation of Malleefowl.

References

Austin, M.P. (2002) Spatial prediction of species distribution: an interface between ecological theory and statistical modelling. Ecological Modelling 157: 101-118.

Benshemesh, J. (1992) The Conservation Ecology of Malleefowl With Particular Regard to Fire. Ph.D thesis. Monash University, Clayton, Vic.

Benshemesh, J. (2000) National Recovery Plan for Malleefowl, Department for Environment and Heritage, Adelaide.

Burbidge, A.A. & McKenzie, N.L. (1989) Patterns in modern decline of Western Australia's vertebrate fauna: causes and conservation implications. Biological Conservation 50: 143-198.

Frith, H.J. (1962) Conservation of the mallee fowl, Leipoa ocellata Gould (Megapodiidae). CSIRO Wildlife Research 7: 33-49.

Garnett, S. & Crowley, G. (2000) The Action Plan for Australian Birds 2000, Environment Australia, Canberra.

Priddel, D. & Wheeler, R. (1999) Malleefowl conservation in New South Wales: a review. . In Dekker, R.W.R.J., Jones, D.N. and Bensehmesh, J. (eds.) Proceeding of 3rd International Megapode Symposium, Nhill, Australia. Zoologische Verhandelingen 327:125-141.

Priddel, D. & Wheeler, R. (2003) Nesting activity and demography of an isolated population of Malleefowl (Leipoa ocellata). Wildlife Research 30: 451-464.

Recher, H.F. & Lim, L. (1990) A review of current ideas of the extinction, conservation and management of Australia's terrestrial vertebrate fauna. Proceedings of the Ecological Society of Australia 16: 287-301.

Serventy, D.L. & Whittell, H.M. (1976) Birds of Western Australia. University of Western Australia Press, Perth.

Wintle, B.A., Elith, J., & Potts, J.M. (2005) Fauna habitat modelling and mapping: a review and case study in the Lower Hunter Central Coast Region of NSW. Austral Ecology 30: 719-738.

129 16. HOW TO SUPPORT THE RECRUITMENT OF MALLEEFOWL Leipoa ocellata IN SMALL REMNANTS: AN OVERVIEW OF PROJECT FINDINGS

Jessica van der Waag Department of Animal Biology, University of Western Australia, Nedlands, 6009, Western Australia

Abstract

After emerging asynchronously from the nest mound, the chicks of the Malleefowl Leipoa ocellata (Megapodidae) disperse independently. Malleefowl chicks are thought to lead solitary lives until reaching maturity at three to four years of age. Very little is known of the dispersal, behaviour and habitat requirements of young Malleefowl during this period, particularly in a landscape of small habitat remnants. By radio tracking chicks and captive-raised young birds, it was possible to gather new information on young Malleefowl that may be used to inform management of small remnants, and have implications for large reserves. Findings include social interactions among young birds, dispersal and movements within the habitat, causes of mortality and recruitment events.

Introduction

After emerging asynchronously from the nest mound, the chicks of the Malleefowl Leipoa ocellata disperse independently. Malleefowl chicks are thought to lead solitary lives until reaching maturity at three to four years of age. Very little is known of the dispersal, behaviour and habitat requirements of young Malleefowl during this period, particularly in a landscape of small habitat remnants. In this paper, I will present some findings of particular interest from a study of Malleefowl chicks and juveniles including survival, social behaviour and recruitment events.

Methods

This study was conducted in the southern wheat belt shire of Gnowangerup in Western Australia. Two study sites were used. Foster is a 138 ha privately owned bush block, and Tieline is a 128 ha Department of Environment and Conservation Nature Reserve. Both remnants have a similar history being undisturbed since isolation by clearing for agriculture. Fox baiting is carried out once annually and is highly localised.

Eggs were collected from mounds at the study site and artificially incubated. The chicks were randomly assigned to two groups. The first were released at a few days of age once they had regained their hatch weight. The second group were captive raised in individual enclosures and released as juveniles. Radio-transmitters were attached using a simple backpack harness of two wing loops of 5mm black cotton tape (Benshemesh 1992; Priddel & Wheeler 1994). The birds were radio-tracked on foot. Releases occurred over three seasons; 2004/05, 2005/06 and 2006/07.

Results

These results are a selection of findings of particular interest from the study of Malleefowl chicks and juveniles. They include survival, social behaviour and recruitment events.

Survival

In examining the survival and causes of mortality in young Malleefowl, one of the aims of the study was to identify the critical age for chicks where survival is significantly increased. The fate of chicks and juveniles released at varying ages was recorded. As expected, there was a high rate of mortality in the first two weeks of life. However survival significantly increased for juveniles of 45 to 50 days of age (Figure 1).

130 300

275

250

225

200

175

150

125

Survival (days) 100

75

50

25

0 n = 37 0 10 20 30 40 50 60 70 80 90 100 110 120 Age at release (days)

Figure. 1. Survival of chicks released at differing ages. Black points indicate chicks survived a minimum of 300 days.

Introduced predators were the major cause of mortality among juveniles, followed by raptors. Of particular interest was the mortality of two chicks that, upon autopsy, appear to have died of head trauma. The birds‘ internal organs appeared normal and healthy, the body condition was good and the crop and gizzard full.

Malleefowl chicks have a larger suite of predators than juveniles, being taken by Varanids (goannas) and Artimidae (Currawongs/Butcher birds) in addition to introduced predators and raptors. Figure 2 presents the breakdown of the causes of mortality. An equal proportion of chicks were killed by introduced predators (38%) and by native aerial predators (38%). Native ground predators accounted for 10% of mortalities. Though only a low proportion of mortalities were directly attributed to starvation, it is well known that this can be a major factor for the survival of chicks (Priddel & Wheeler 1990; Benshemesh 1992). Where possible, remains after predation were examined to determine the physical condition of the chick. In most cases, however, there was insufficient remaining.

n = 22 Introduced pred Artamidae pred Varanidae pred Raptor pred Unknown pred Starved Other Survived

Figure 2. Fate of chicks released

The increased rate of survival at 45 to 50 days would be the result of the size of the chick, its level of physical development, and its increased skills for foraging and predator avoidance. Malleefowl chicks grow and develop quickly. From an average 112 grams at hatching, by 45 to 50 days, they are around 300 grams (Figure 3).

131 850

800

750

700

650

600

550

500

450

400

Weight(g) 350

300

250

200

150

100

50

0 0 10 20 30 40 50 60 70 80 90 100 110 120 130

Age (days)

Figure 3. Mean weight gain in the days following hatching.

A

B Figure 4. Chicks 1 to 14 days of age. A. 1 to 7 days of age. Nape to wingtip is approximately 10 cm. B. 14 to 24 days of age.

132 A

B

Figure 5. Malleefowl chick development from 23 to 40 days of age. A. 24 to 30 days of age. B 33 to 40 days of age. Nape to wingtip is approximately 18cm

They also develop quickly, with true tail feathers emerging by around 20 to 24 days (Figures 4 and 5).

Social behaviour

Social behaviour has been documented in young Brush-turkey chicks. Göth & Vogel (2002) found social interaction occurred among chicks between two days (in captivity) to four weeks of age. Mixed groups of three to six month old birds were also observed (Göth 2001).

Social interaction was not observed in Malleefowl younger than eight weeks of age in this study. Aggressive behaviour was observed in very young captive chicks of up to a week of age when two chicks encountered each other. One chick would display, lowering the head, raising the head feathers and wings (Figure 6). A loud growling vocalisation was also used. The second chick would generally flee without giving any display or vocalisation. Chick encounters were not observed in the field.

133

Figure 6. Aggressive display in a very young chick (three days old).

Social interaction was first observed in the field between two 10 week old juveniles in the 2004/05 season. These birds have been released two weeks apart and had moved and settled solitarily (Figure 7). After five weeks in the field, one bird (black) moved over a kilometre within the course of a morning to an area where the other bird (white) had settled in the north-eastern corner (after three weeks in the field). The birds remained together for two weeks; foraging within 20 meters and roosting within 50 meters. The interaction between the birds involved the use of visual displays, sparring and fighting, as well as vocalisations including an aerial predator alarm call and a contact call.

Figure 7. Movements of two juvenile birds in the 2004/05 season

For the next approximately seven months, the birds alternated between settling solitarily, and in close interaction. On two occasions, a third wild bird was seen with the pair.

In 2005/06, social groups of up to five members formed at both study sites. As with the pair in 2004/05, the groups were dynamic in membership, varying from two to five. Though no wild birds were observed with these groups, a surviving chick released at a few days of age joined a group at around 8 weeks of age. Table 1 presents the dynamics of social interaction

134 between birds on 13 separate visits (excluding visits where all the birds were solitary). For example, on visit 7, birds 041 and 042 were located together as a pair, 051, 052, 053 and 054 were together in a group, and bird 055 was within 50 meters of that group, but not interacting with the other birds.

Table 1. Group dynamics at Foster Road between April and August 2006

Group Dynamics at Foster Road Birds (rows) interacting at 13 visits (columns) between April and August 2006 1 2 3 4 5 6 7 8 9 10 11 12 13 041 ○ ● 042 ○ ● ○ ● ○ Wild ● bird 051 ● ● ● ● ● 052 ● ● ○ ● ● ○ ● ● ● ● ● ● 053 ● ● ○ ● ● ● ● ● ● ● 054 ● ● ● ● ● ● 055 ● ○ Shades of grey indicate separate groups within the site. Closed circles indicate interaction. Open circles indicate proximity (within 50 meters of each other), but not interaction.

Recruitment event of a young male

In late September 2005, the remains of an adult male bird were found 30 meters from one of the active mounds. The mound had been filled with litter and closed to allow the organic material to begin fermentation to generate the first stage of heat for incubation. As daily work had not yet commenced, it was not possible to observe the birds to determine if it was one of the pair.

In mid November, a radio-tracked eleven month old male was observed resting in scrub near the mound. As the mound was opened by the researcher, the young bird came onto the mound and began filling it back in. There was one egg in the mound (the mound of an established pair nearby had two eggs). Despite a poor season for breeding, at least four chicks were successfully hatched. In the following 2006/07 season, the male worked the mound again. There was a clutch of at least 12 eggs.

Acknowledgements

I am grateful for funding provided by the following; the Australian Academy of Science, the Malleefowl Preservation Group Community Conservation Trust, the WWF Threatened Species Network through the Yongergnow Australian Malleefowl Centre and Birds Australia. I would like to thank Stephen Davies for his continuing support and assistance.

The members of the Malleefowl Preservation Group have volunteered many hours to assist with fieldwork, my thanks to them, and also to L. and R. van der Waag.

135

References

Benshemesh, J. 1992, The conservation ecology of Malleefowl, with particular regard to fire. PhD Thesis, Monash University, Clayton, Australia

Göth, A. 2001, Survival, habitat selectivity and behavioural development of Australian Brush- turkey, Alectura lathami chicks, PhD thesis, Griffith University, Brisbane, Australia

Göth, A. & Vogel, U. 2002, Social behaviour of two free-ranging chicks of Australian Brush- turkey Alectura lathami, Corella, 26:1-3

Priddel, D. & Wheeler, R. 1990, Survival of Malleefowl Leipoa ocellata chicks in the absence of ground dwelling predators, Emu 90:81-87

Priddel, D. & Wheeler, R. 1994, Mortality of captive-raised Malleefowl Leipoa ocellata, released into a Mallee remnant within the wheat-belt of New South Wales, Wildlife Research, 21:543-52

Priddel, D. & Wheeler, R. 1996, Effect of age at release on the susceptibility of captive-reared Malleefowl Leipoa ocellata to predation by the introduced fox Vulpes vulpes, Emu 96:32-41

136

17. POST-RELEASE SURVIVAL OF CAPTIVE-REARED MALLEEFOWL IN WESTERN NEW SOUTH WALES

Coombes, C. J.1, Dehaan, R.2 and Wilson, A. L.3 1 Charles Sturt University, Albury, NSW. 2 Department of Environment and Heritage, GPO Box 1047, Adelaide, 5001 3 Charles Sturt University, LMB 588, Wagga Wagga, 2678, NSW.

Abstract

Malleefowl populations have undergone serious declines as a result of habitat loss and predation by introduced predators. Since 1990 a successful captive breeding program has been supplying chicks for release into Nature Reserves in western NSW. However, early studies indicated that post-release survival of the captive-bred chicks was low, mainly as a result of predation by foxes. In addition movement patterns of sub-adults following release is poorly known, making selection of optimal release sites difficult. This study investigated the survivorship of captive-bred Malleefowl following the introduction of aerial fox baiting. After behavioural and radio-transmitter attachments trials on captive birds, data on mortality and movement patterns of 14 sub-adult birds were collected using radio telemetry after release from November 2005 to April 2006. Survivorship in the first four months following release was relatively high with only one mortality confirmed. The majority of birds dispersed from the release site. However, variation in movement patterns between individuals was high, with some individuals moving large distances each day and others remaining in the vicinity of the release site. The results of this study suggest that fox baiting has been successful in improving survivorship of released Malleefowl.

Introduction

Malleefowl populations have undergone serious declines as a result of habitat loss and predation by introduced predators. Since 1990 a successful captive breeding program at Western Plains Zoo has been supplying chicks for release into nature reserves in western New South Wales. However, early studies (Priddel & Wheeler 1990) indicated that post- release survival of captive-bred chicks was low, mainly as a result of predation by foxes. In addition movement patterns of sub-adults following release is poorly known. Since that preliminary work with the release program no monitoring of these released birds has been done in the last ten years. This study aimed to see what was actually happening after the young birds were released and left to their own devices.

The study was done at Nombinnie Nature Reserve (Figure 1), which, with the adjoining Yathong and Round Hill Reserves, is the part of the largest continuous remnant of mallee in New South Wales, totalling 250,000 ha. There are mound counts in Yathong and Round Hill Nature Reserves, but none in Nombinnie, so releasing the birds there should not affect the mound counts at Yathong and Round Hill. The vegetation at Nombinnie is mainly mallee- spinifex with a shrub understorey of Acacia and Eremophila.

137 QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 1. The site of the study area in western New South Wales.

Work by Priddel & Wheeler (1994) indicated that there was significant predation by foxes on captive-bred Malleefowl when they were released; the effect of cats was unknown, but they do occur on Nombinnie as well as foxes. Foxes were considered the main mortality factor affecting Malleefowl in this study and to combat them an intensive fox baiting program is running. The park is baited from the air three times a year; 12,000 baits are dropped each time and three to five thousand additional baits are given to neighbours to distribute along roads and tracks. Park staff also distribute baits along tracks, so a large area is baited. Park staff report that the number of sightings of foxes on spot-lighting surveys have reduced over time, so the baiting appears to be having some effect. Many feral goats still occur in the park and graze the ground cover and shrubs, thereby competing for food with Malleefowl and destroying shelter. Goat traps have been installed at some water points and thousands of goats have been removed, but currently goat control is exercised by shooting them from a helicopter.

Methods

Of the 35 captive-reared birds released at Nombinnie in 2005, 14 were equipped with 12 gram transmitters. The transmitters had mortality sensors in them that doubled the rates of signal transmission if the transmitter did not move for ten hours. The transmitters were attached to anaesthetized birds in the scapular region between the shoulder blades following the methods of Goth & Jones (2001). Transmitters were only attached to birds that appeared to be less timid than others. Observations in the captive colonies indicated that the transmitters had a negligible effect on the behaviour of these birds. The birds weighed just over one kilogram (mean 1043g), at an average age of 286 days, when they were released.

All the birds with transmitters were released in November 2005 at one spot in Nombinnie; those without transmitters were released 6 km away. In 2006 we released another eight birds equipped with transmitters at Nombinnie. The released birds were tracked using an aerial mounted on a five metre pole to raise it above the mallee canopy. Compass readings were taken of the direction of the bird, almost simultaneously from two points a known distance apart and the position of the bird given by the intersection of these two directions, using the Arc Macro computer program. Another program ArcView enabled us to plot the movement path of each bird and its home range.

138 We were interested in the effect of vegetation density on the distribution and movement patterns of the birds. We used satellite imagery to map the density of vegetation in the reserves, classifying the density as low, medium and dense. Some ground truthing was undertaken to check the classification the satellite imagery presented to us.

Results

Patterns of movement

The transmitters lasted well. A few dropped off soon after release, but the average tracking time was 40 days and it was 137 days after release before we lost the last bird. There was only one confirmed death during the course of the study. The birds moved mostly east or west of the release point; only a few went south or north of it. Many moved a considerable distance at first, but then settled down and remained in one area for some time.

QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 2. The mean distances moved by individual birds released in 2005.

The distances moved in one day ranged from 187 m to 713 m, but most of the long moves were in the days shortly after release. Having moved away a few birds then returned to the release point (Figure 2). The pattern of movements resembled those shown by the Ongerup birds (van der Waag 2008), even to the point where some birds formed small groups and moved together. Figure 3 illustrates the observed movements of the birds in the 2005 release. The birds released in 2006 showed a similar pattern of movement but were not tracked for as long as the 2005 release. For the 2006 birds the first transmitter cut out after 17 days, the average tracking time was 21 days and the longest a bird was tracked was 22 days when tracking ceased.

areTIFF neededQuickTime™ (LZW) to decompressorsee and this a picture.

139

Figure 3. The movements of radio-tracked Malleefowl chicks released at Nombinnie in 2005.

Habitat selection

Figure 4 shows the distribution of the birds at the end of tracking, for both the 2005 and the 2006 release, overlain on the vegetation classification. It shows that the birds tended to favour medium to low density vegetation and avoid the densely vegetated areas. At the conclusion of tracking the birds had settled in three areas, encircled by the polygons in Figure 4, drawn by the ArcView program. One interesting aspect of the tracking was a time when a number of mortality sensors were set off together. This happened over 36 hours when a change-over of tracking teams was taking place. The second team went out in the morning to recover the transmitters from a number of dead birds, but when they located them the birds jumped up and ran off. Apparently a thunderstorm on the previous day had forced the birds to stay where they were at the time it broke and remain there for so long that the mortality sensors were activated.

Discussion

The behaviour of the released birds was very variable. Some stayed close to the release site, others moved considerable distances and then settled in an area, sometimes in a group with other birds. Some birds moved gradually away from the release site and continued moving for the duration of the study. Other birds remained as individuals, some returning to areas they had previously occupied. The patterns were similar to those recorded at Ongerup by van der Waag (2008). Survival was good over the period of tracking, with only one confirmed death.

We observed a lot more wildlife at Nombinnie than we expected and we attributed that to the success of the fox baiting program that has been running and to the removal of goats that would enable more resources to be available for Malleefowl.

140 QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 4. Final distribution of 2005 and 2006 releases at Nombinnie Nature Reserve in relation of vegetation density.

The study showed us that released Malleefowl do disperse, some considerable distances, showing that they are not going to remain in one spot, exhaust the resources there and die of starvation. One of the big questions is what effect do feral cats have once the foxes are removed and it would be worth doing more work to find that out. Are they serious predators of Malleefowl? The other big question is are these released birds, that appear to have survived well during the study, living to join the breeding population?

Acknowledgements

We are grateful to the Western Plains Zoo for breeding and providing the Malleefowl used in this study. Our thanks also to the staff of the New South Wales National Parks and Wildlife Service at Cobar, to the volunteers who assisted with the tracking and to Craig Poynter of SPAN.

References

Goth, A. & Jones, D. N. (2001). Transmitter attachment and effects on Australian brush-turkey chicks. Aust. Wildl. Res. 28:1-6.

Priddel, D. & Wheeler, R. (1994). Mortality of captive-reared malleefowl, Leipoa ocellata, released into a mallee remnant within the wheat-belt of New South Wales. Wildlife Research 21: 543-552.

Van der Waag, J. (2008). How to support the recruitment of Malleefowl Leipoa ocellata in small remnants :an overview of project findings. Curtin, Environmental Biology Bulletin No. 28.

141

18. MALLEEFOWL SEARCHES AT YEELIRRIE STATION IN ARID WESTERN AIUSTRALIA: A VALUABLE COLLABORATION BETWEEN INDUSTRY AND VOLUNTEERS

Joe Benshemesh1, Susanne Dennings2, Carl Danzi3 and Angela Saunders2 1 La Trobe University, School of Biological Sciences, Bundoora, Vic. 2 Malleefowl Preservation Group, Ongerup, WA 3 WA Malleefowl Network Facilitator, WWF-Australia, Perth

Abstract

Between 2000 and 2006 the Malleefowl Preservation Group undertook systematic surveys for Malleefowl at Yeelirrie Station, a 2750 km2 pastoral property managed by BHP Billiton Nickel West and located 400 km north-north-west of Kalgoorlie, Western Australia. The primary objective of this study was to describe the distribution of Malleefowl at Yeelirrie and to provide information on population trends. Volunteers walked over 800 km of systematic transects in search of Malleefowl footprints and other signs, and located at least 24 mounds. Six widely separated areas were inhabited by Malleefowl and breeding was confirmed in four of these areas. Malleefowl are highly localised at Yeelirrie, but we found no evidence of declines: footprints of Malleefowl were found in all five areas in which old mounds were found during our surveys, suggesting that Malleefowl range had not contracted over the past few decades, and in 2006 the birds still occupied areas in which their footprints were detected in 2000 and 2003. Recent management of Yeelirrie thus appears to have been beneficial for Malleefowl, although the frequency and extent of wildfire during the past five years (in particular) is a concern and may have removed some favourable habitat. We estimate that there are probably 10-20 breeding pairs on the property and conclude that the most important contribution that managers can make toward the conservation of Malleefowl at Yeelirrie is to continue to manage the property in an ecologically sensitive way, and to continue to monitor this important Malleefowl population in order to assess the success or otherwise of management actions.

Introduction

The Malleefowl (Leipoa ocellata) is gazetted in Western Australia as a species that is rare or threatened with extinction, and regarded as vulnerable nationally (Benshemesh 2000, Garnett & Crowley 2000). The species is best known from the semi-arid zone where clearing, overgrazing, and inappropriate fire regimes have caused substantial decline in its distribution and abundance. In the arid zone, few Malleefowl studies have been conducted and the species‘ distribution and abundance is poorly known. Scattered historical records from early this century suggest that Malleefowl were found in the south-west of the Northern Territory as far north as the Tanami Desert, throughout much of the western half of South Australia, and through much of southern and central Western Australia (Figure 1). Since then, Malleefowl appear to have declined substantially in the arid zone of Australia and they are probably extinct in the Northern Territory. These declines, and the increasingly uncertain future of Malleefowl, clearly indicate an urgent need to describe the distribution of the species, understand its habitat requirements, and to monitor its population trends in arid Australia. Such studies are regarded as being of utmost importance to the conservation of Malleefowl nationally (Benshemesh 2000, Garnett & Crowley 2000).

142

Figure 1. Estimated historical and current range of Malleefowl across Australia (from Benshemesh 2000)

While describing the distribution and trends of Malleefowl in the arid zone is of great importance, gathering the required information is a challenge. Conventional methods of survey and monitoring, such as sightings or trapping, are not efficient for detecting Malleefowl as the birds are shy and cryptic and often elude detection. Malleefowl nests, or mounds, are conspicuous and are used to monitor Malleefowl in southern areas where their densities are relatively high, but are widely dispersed in arid environments and difficult to find. Whereas in southern areas it is not unusual to find 10-20 old Malleefowl mounds per square kilometer, mound densities in arid areas are often tens or even hundreds of times more dispersed, and systematic searches for mounds are ineffective if not futile. However, there are some advantages as well. Where Malleefowl occur in arid areas, substrates are typically sandy, loose and exposed and these characteristics make tracking especially useful for detecting large ground-dwelling species such as Malleefowl. In fact, searching for the bird‘s footprints is often the best way of detecting Malleefowl in arid areas: Malleefowl prints are distinctive and conspicuous, and during mild weather these prints accumulate for days or even weeks to form a recent record of a bird‘s presence and movements. Yeelirrie station is a pastoral lease owned by BHP Billiton Nickel West (formerly WMC Resources) and is an example of the pastoral properties in central Western Australia that still retain much of their natural vegetation and provide important areas for the conservation of threatened species, such as Malleefowl. Yeelirrie has been sensitively managed for conservation in the past and this is likely to continue into the future. Increasing our knowledge of the distribution and abundance of Malleefowl at Yeelirrie will further encourage sensitive management and provide important information on the ecology and conservation of the species across the arid zone. Since 2000, the Malleefowl Preservation Group and WMC Resources/BHP Billiton Nickel West have been working together successfully on Malleefowl research at Yeelirrie. This partnership has enabled the Malleefowl Preservation Group to conduct searches for Malleefowl using community volunteers and provide information and management recommendations to enhance the conservation of Malleefowl at Yeelirrie.

The Yeelirrie Malleefowl project aimed to map and monitor the occurrence of Malleefowl within the station boundaries, and in nearby areas of particular interest. This aim was ambitious (Yeelirrie itself covers 2,750 km2) and of great importance to the conservation of Malleefowl throughout the arid zone. While methods of survey and monitoring Malleefowl in semi-arid areas are well established, these methods do not work well in arid habitats because of the scale of these areas and the characteristically low densities of Malleefowl. The Yeelirrie

143 project was the most thorough attempt to develop methods for use by community volunteers for survey and monitoring for Malleefowl in the arid zone. In this regard, the methods developed may be of interest to any group wishing to conduct similar work, while the results obtained in the study provide a basis for interpreting the ecology of Malleefowl in the arid zone.

Methods

Study area: Yeelirrie Station

Yeelirrie Station is an operational pastoral property leased by BHP Billiton Nickel West. The property spans some 275,000 ha and is located about 80 km south of Wiluna, 400 km north- north-west of Kalgoorlie, and 670 km north-east of Perth in the ‗mulga-belt‘ of central Western Australia (Figure 1). The climate is typically dry with an average rainfall of about 210 mm per annum with high yearly variation. Most rain (about 64%) falls in heavy summer storms, and occasionally the area also receives substantial downpours from dissipating tropical cyclones. Rain is received during the winter with the passage of southern cold fronts. The landform comprises a raised plateau (Old Plateau) that has eroded to form granitic breakaways or ‗jump-ups‘, and pediments and alluvial plains of the valley floor. Acacia woodland and shrubland, and spinifex grasslands (Triodia basedowii), dominate the vegetation. The property was developed for pastoral production in 1925, but between 1974 and 1986 Yeelirrie was largely destocked by Western Mining Corporation (now BPH Billiton, Nickel West). Subsequently the property ran a relatively small number of sheep until quite recently when it was once again destocked due to intractable predator (dingo/dog) problems.

Malleefowl are known to occur on Yeelirrie Station and have been seen there on several occasions by staff and other observers over the past few decades (see Benshemesh and Malleefowl Preservation Group 2001). In 2000 the Malleefowl Preservation Group began Malleefowl surveys at Yeelirrie and these have continued every few years to 2006. The specific aims of the Malleefowl project at Yeelirrie were to:

Describe the distribution of Malleefowl Assess the likelihood of recent declines Provide a foundation for monitoring

Training

Training of volunteers involved instruction on the recognition of animal tracks, the use of equipment (GPS, palm handheld computers, digital cameras, radios), and safety procedures. In particular, volunteers were provided with scaled photographs of the prints of various animals likely to be encountered at Yeelirrie, and a manual that detailed methods. Before searching for Malleefowl prints, volunteers were taken to areas in which Malleefowl prints could be seen on the ground so that each person had a clear search-image of the target of the surveys.

Tracking

Malleefowl are difficult animals to survey because they are rare, secretive and well camouflaged. However, their footprints are relatively large and distinctive and Malleefowl can be readily tracked in suitable habitat. To sample the habitats at Yeelirrie, we walked predetermined transects in search of the birds footprints. Transects were rectangles with long sides of about 4 km and short sides of 750-1000m, and orientated such that the long sides were orientated east-west to maximise lighting from the side and ahead.

Volunteers were instructed to record every Malleefowl track encountered. In 2006, volunteers photographed at least the first three Malleefowl prints encountered on each transect or part transect completed for later verification. Every 200m, volunteers stopped and recorded: 1) general habitat (open, shrubby, or woodland) and canopy height (less than 2m, 2-4m, and

144 over 4m); Tracking conditions (good, fair, poor, impossible); prints of other animals observed in the past 200m; and obtained a habitat photograph.

Selection of Transects

Transects were selected on the basis of a number of considerations using available satellite imagery and GIS mapping (kindly provided by David Gifford of BHP Billiton Nickel West, Breton Clifford, and Google Earth): Access: Track access is limited in Yeelirrie and priority was given to areas near tracks Previous knowledge: To measure the persistence of Malleefowl in the landscape, we revisited all areas in which they had previously been recorded. Historical records were included where possible, although the accuracy of the records, especially the associated locations, was often uncertain Habitat: Stony granitic areas and wash plains were avoided as tracking is generally not possible on these substrates. At Yeelirrie, these areas were usually very open and were less suitable for Malleefowl that tend to prefer sandy, well vegetated habitat (Frith 1962a, b, Benshemesh 2000). However, stony and open habitats were nonetheless sampled on many transects which unavoidably included a range of habitats Spread: We spread transects as widely as possible, other considerations notwithstanding Revisiting mounds

Malleefowl Preservation Group members routinely re-visited all previously known mounds at Yeelirrie in order to assess their recent or current use. Mounds were monitored using standard national monitoring protocols for Malleefowl.

Electronic data capture: Cybertrackertm

All data were recorded on Palm handheld computers connected to GPS units. We used an application named Cybertracker to program a series of screens on the Palm devises for data collection. Cybertracker also records GPS coordinates with every data record. Data were downloaded from palms every evening in the field and were later transferred to a Microsoft Access database.

Photos

Digital cameras were used to record prints for later verification, and in 2006 digital photos were also used to photograph habitat for reference. Volunteers photographed the habitat every 200m along transects, and the first three Malleefowl footprints they encountered on a transect. Digital photos were downloaded from cameras in the field and filed by date and transect. To provide ready access to these photos, we later renamed all photos in regard to the transect name, date, and a code that uniquely identified each photo with its associated record on the database. This provided a very quick, easy and objective means of recording habitat condition that we used during the analysis of data, and a valuable photographic reference to monitor habitat changes in the future.

2000-2006 surveys

2000 - The Yeelirrie project began in 2000 with a team of Malleefowl Preservation Group volunteers (Benshemesh & Malleefowl Preservation Group 2001). For the most part, this trip concentrated on developing methods that were suitable for volunteers and which could be easily taught, although there were a number of notable achievements and foundations were laid for subsequent surveys. Over 200km of transects were searched for Malleefowl signs, resulting in nearly 200 signs of Malleefowl (mostly prints) and over 750 site descriptions. Site descriptions involved information on the frequency of various large-animal prints, tracking conditions and habitat and were obtained every 200m along search transects. Seventeen

145 mounds were found during these surveys, and one of these was fully active (i.e. completed and likely to contain eggs).

2003 - The Malleefowl Preservation Group returned to Yeelirrie in 2003 and completed over 300km of transect searches (Sanders et al. 2003). However, heavy rain severely impaired tracking conditions and this resulted in few signs of Malleefowl (11 prints and 10 mounds) being recorded. Similarly, few site descriptions were also recorded. On the positive side, ten mounds were located at Yeelirrie in 2003, two of which were active, and Malleefowl prints were located in many areas where they were recorded in 2000, and in several new areas as well.

2004- George and Joan White monitored the 27 known mounds at Yeelirrie in November 2004 and recorded three of these as fully active.

2006 - The focus in 2006 was to consolidate the information collected to date and improve our knowledge of the distribution, abundance and conservation of Malleefowl at Yeelirrie (Benshemesh et al. 2007). Over 300km were searched for signs of Malleefowl in 31 transects that were selected to both monitor the persistence of Malleefowl in areas in which Malleefowl were previously recorded, and to sample in new areas or where sampling was judged to have previously been equivocal (usually due to rain before previous surveys). Nearly 1500 site descriptions were obtained for which we recorded habitat type, tracking conditions, and the occurrence of a range of animals every 200m along transects, and over 100 signs of Malleefowl were recorded. Habitat photos were obtained for virtually all 200m stops along transects and cross indexed with the data from these locations on the database.

Results and Discussion

While we have conducted surveys in search of Malleefowl footprints and mounds on three occasions since 2000, the data collected in 2006 represent the culmination and consolidation of previous surveys (especially as the 2003 surveys were hampered by rain). The following discussion focuses on the most salient results obtained in 2006, although mapping the distribution of Malleefowl was a progressive task.

General tracking

The average number of species detected in 200m sections of transects was not greatly influenced by habitat type and was fairly consistent at 2.1 to 2.6 species in different general habitat and canopy height habitat classes. Likewise, the average number of species recorded under most tracking conditions (‗good‘, ‗fair‘ and ‗poor‘) was similar, although the worst condition (‗impossible‘ category) was associated with much lower species counts than other categories and was omitted.

Kangaroos were the most frequently recorded species during transects, occurring in 79% of the 1423 sites (200m sections of transects). Echidna, emu and goanna were also commonly recorded and occurred in 49%, 46% and 42% of sites respectively, whereas rabbits occurred in 14% of sites. Malleefowl, dogs, foxes and cats were all recorded in only 4-5% of sites along transects. Sheep, camels and bustards were also recorded along transects, but rarely (less than 1% of sites).

Malleefowl distribution

The surveys by the Malleefowl Preservation Group have shown that there are at least six widely separated areas on Yeelirrie in which Malleefowl are resident. In 2006 we were lucky enough to find five mounds that were being prepared for breeding in four of these widely separated areas, but we expect that we also missed at least a similar number of active mounds. This is because our searches were not designed to find mounds (a much more laborious process), but rather to detect footprints and intercept Malleefowl ranging paths. While we did find some active mounds, it was clear from the patterns of footprints that there were also several clusters of Malleefowl prints that were distant from known active mounds. These clusters of prints probably represent resident pairs, and we suspect a more thorough

146 search of these areas would reveal at least another five active mounds in these areas in good rainfall years. It should also be noted that our transects did not sample all of the apparently suitable habitat as some potential sites were inaccessible, and that we may have missed some clusters of prints (due to windy or rainy weather, poor substrate, etc.). Given these considerations, we suggest the entire breeding population at Yeelirrie is probably around 10- 20 pairs.

The distribution of Malleefowl at Yeelirrie appears to be associated with the edges of the old plateau where there is relatively tall shrublands (over 2m) of bowgada, mulga and other acacias. These areas support a gradient of soil type and vegetation, ranging from the open spinifex grasslands on deep sand (typical Bullimore land system), often with shrubby sugar brother (Acacia coolgardiensis), through to bowgada (A. ramulosa var. linophyla) and eventually mulga (A. aneura) on the hardpan soils (typically Yanganoo and Kalli land systems) near breakaways (Sherwood land system). Malleefowl appear to occupy all of these habitats to some degree.

The major association of Malleefowl and habitat types, however, was not with land systems but with habitat structure and recent fire history. Malleefowl showed a strong preference for shrubby habitat with a 2-4m canopy, the most common association at Yeelirrie, and an avoidance of other habitats. Kangaroos, emus, echidnas, sheep rabbits and goannas also tended to prefer habitats recorded as shrubby and avoid those regarded as open, although emus showed a distinct preference for open and shrubby habitat with less than 2m canopy. Mammalian predators (dogs, foxes and cats) showed little preference with habitat-canopy classes. Interesting and plausible as these associations are, it should be noted that volunteers received only brief training on identifying animal prints other than Malleefowl and some people stated their confusion in distinguishing between dog, fox and cat prints, and occasionally between emus and bustards.

Malleefowl also showed a strong avoidance of areas that were burnt within the past decade or so. This result is not unexpected given the preference of the birds for shrubby and tall vegetation in this study, and Malleefowl responses to fire elsewhere (reviewed in Benshemesh 2000), and highlights the importance of protecting the known Malleefowl areas from further losses due to fire. Although we were unable to source satellite images showing the extent of the recent fires, the photographic record of habitats obtained during this study show that about 40% of sites had been burnt in the last five years and that these fires have encroacheda on Malleefowl b ) )

147

Figure 2. Transect search coverage (black rectangles, typically 4km long) of Yeelirrie over the three survey periods in 2000 (a), 2003 and 2006 (b) showing the minimum convex polygons enclosing a) Malleefowl prints (white polygons) where more than a single print has been recorded, and b) Malleefowl mounds (white crosshatched polygons). Based on 826 km of search transect on which 269 Malleefowl prints were recorded, and 35 mounds (24 confirmed). Property boundaries and tracks are also shown (thin black lines). Background satellite (LandSat) imagery obtained in 2001 before the widespread fires in the following years. Note that at one site at which prints were recorded along the transects there were no mounds encountered. habitat. Fortunately, Malleefowl still occurred in all the general areas in which they were detected in 2000 before the extensive fires, although habitat loss was considerable and some habitat was burnt that was previously utilized by the birds. It is likely that the proximity of all the Malleefowl sites to the granitic breakaways conferred some protection from fire. Nonetheless, given that most records of Malleefowl in 2006 were obtained from highly flammable habitats with spinifex ground-layer (Bullimore land system), it was lucky that more areas occupied by Malleefowl were not burnt.

Knowledge of where Malleefowl occur within the landscape is essential information for devising strategies for protecting such areas from fire, and documenting this information is a major achievement of Yeelirrie Malleefowl project. This information should now be used to devise fire management strategies. Mapping and digitising the extent and dates of all fires in the past 30 years would be an excellent way to start this process. Apart from providing a bias from which to develop fire management plans, such maps would provide useful information for analysis of the degree to which Malleefowl utilize habitat of different age since fire in the Yeelirrie landscape.

Overall, the survey and monitoring information obtained over the past six years suggests that the prognosis for Malleefowl at Yeelirrie is positive. Continued monitoring of these sites will provide the feedback that managers need to assess the success (or otherwise) of management actions.

Malleefowl persistence

Monitoring is generally recognised as being of critical importance for conserving Malleefowl, and this is especially true at Yeelirrie where the distribution of the species is now probably better known than anywhere else in the arid zone. The Malleefowl Preservation Group searches have provided a wealth of information on Malleefowl, as well as other species and habitats that occur at Yeelirrie, but arguably the greatest value of the project is in laying foundations and providing information on trends in Malleefowl populations.

In general, Malleefowl appear to be doing well at Yeelirrie. All of the three areas in which the birds were detected in 2000 were still occupied by Malleefowl in 2006. Similarly, all the areas surveyed in 2003 in which more than a solitary sign of Malleefowl was recorded were still occupied in 2006. There were two areas in which Malleefowl prints were recorded in 2003 but not in 2006, but in both cases these were solitary and unconfirmed records (i.e. not photographed) and most likely represented either wandering birds or errors.

Another way of looking for evidence of decline is to compare the distribution of mounds with that of prints (Figure 2). Mounds are persistent features of the landscape and suggest that Malleefowl have bred in an area within the last few decades, whereas prints indicate the current presence of the birds. While over 800 km of transects have been searched for signs

148 of Malleefowl since 2000, equivalent to about 2,000 km walked by individuals (as there were usually 2-3 people on a transect search), only 35 mounds have been recorded and of these 24 have been confirmed as Malleefowl mounds. These mounds occur in five clusters and Malleefowl prints were confirmed at each of these in 2006. In terms of search effort, more than half the distance walked since 2000 has been in areas where no nests have been found. There was however, one site at which clusters of prints were found but no nests were encountered. This was a site first visited in 2006 and the pattern of prints suggested that active and old nests would be found in this area.

These data indicate that Malleefowl are highly localised at Yeelirrie, that birds are still occupying areas in which they were detected in 2000, and that they are still occupying areas in which they built mounds over the past few decades.

However, it is also true that Malleefowl have not been detected at a number of sites for which historical records exist. These data are more difficult to interpret because in most cases the records were of single sightings and may have been of wandering birds, and because the precision of many of these locations from the 1970s-1990s is uncertain (before GPS). Nonetheless, in most cases where signs of Malleefowl were not found, where they were once reported to have been seen, habitat appeared unsuitable for Malleefowl breeding, often because it had been burnt in the meantime (many of these records are discussed in Benshemesh and Malleefowl Preservation Group 2001).

Mitigating threats to Malleefowl

Current management practises at Yeelirrie are clearly sensitive to, and are generally appropriate for, the conservation of Malleefowl. Low stocking rates in particular have maintained habitats in relatively pristine condition, and this is apparent in the healthy shrub layer (Figure 3). Assuming these benevolent practises continue, the major threat to Malleefowl would appear to be that of wildfire which has the potential of suddenly eliminating the species over large areas and rendering habitats unsuitable for the birds for several decades. This threat has been temporarily mitigated by the recent fires in some areas: fuel loads are currently low in recently burnt habitat and the continuity of fuel has been disrupted. However, the degree to which the six main areas in which Malleefowl occur are vulnerable to further fires is unclear, and whatever the degree of protection the recent fires may provide, this is likely to be short lived. A careful consideration of the vulnerabilities of the Malleefowl habitats, and the development of a fire management plan in regard to Malleefowl conservation, are urgent priorities.

Figure 3. Photographs of Malleefowl mounds on Yeelirrie station (left) and on a neighboring property that has been heavily grazed (right) showing the pronounced differences in shrub layers. With little shelter from the elements or predators, and few food sources, it is unlikely that Malleefowl would survive for long in the more open and degraded habitat.

Mammalian predators, particularly foxes, are often regarded as presenting a threat to Malleefowl (Priddel et al. 2007), although the evidence at the population level is not strong and the need for control is questionable. This uncertainty is exemplified at Yeelirrie: on one hand the continued presence of Malleefowl might be attributed to many years of predator control; on the other hand there is no evidence of a decline in Malleefowl that might be

149 expected due to the high dog/dingo numbers that over the last few years have necessitated the destocking of the property of sheep (Lucy Browlie, pers. comm.). While this uncertainty does not provide clear management directions in regard to the need for predator control, it does highlight the need for monitoring of the Malleefowl population. Whatever management actions are taken, the effects of these actions can at least be observed on the resident Malleefowl populations if they are appropriately monitored for change.

One possible downside of the control of predators, particularly dogs, is the benefit provided to large herbivores such as kangaroos and emus. While Yeelirrie has been destocked of sheep, kangaroos in particular are at high numbers and are seemingly ubiquitous, having been recorded in nearly 80% of over 1400 sites. Detrimental effects on Malleefowl have been documented due to high numbers of sheep (Frith 1962a, b), and similar effects might occur due to kangaroos where their numbers are unusually high. Kangaroos are culled at Yeelirrie and this practice should continue while their numbers are high, and while their natural predators are controlled and water is artificially provided.

Improvements: how we’d do things better next time

Much of the success of the Yeelirrie Malleefowl project can be attributed to the commitment and determination of the volunteers who applied themselves conscientiously and methodically to the work at hand. For the most part, we felt the field techniques made good use of this dedicated workforce, especially the use of electronic data-capture (Cybertracker), GPS, digital photography and the partitioned, rectangular transects which worked well at Yeelirrie.

Nonetheless, there were some important things we would recommend for future surveys at Yeelirrie or similar projects elsewhere. First, while the training that volunteers received was thorough and adequate in most cases, few volunteers probably needed additional instruction. In the future, we recommend that the ability of volunteers to identify Malleefowl prints be tested in the field as part of the training regime, perhaps by having individuals walk a few hundred meters along which a known number of Malleefowl prints occur. A simple test such as this would quickly identify the volunteers that require additional instruction or practice.

Secondly, more photographs of Malleefowl and other animal prints would be advantageous. Photographing prints is quick and easy in the field and provides a means of verifying the identifications made by volunteers, and would not add greatly to the workload in the field or in processing the information. Given that we were able to mesh the field data with the large number of digital photographs more easily than we expected, we would strongly recommend that a much larger sample of prints that are recorded be photographed in order to assess the accuracy of identifications, and correct misconceptions wherever possible.

The Future: Monitoring Malleefowl at Yeelirrie

Assuming that Yeelirrie continues to be managed in an ecologically sensitive way, the most important contribution that volunteers or managers can make toward the conservation of Malleefowl at Yeelirrie is to carefully monitor the existing population. This is not to say that other management activities are unimportant, but rather that whatever management is applied, it is essential to have some means of assessing its impact on Malleefowl. Without monitoring at some level, management is blind.

In light of our findings, monitoring could most efficiently be achieved at Yeelirrie by using a hierarchical approach, targeting the occurrence of the species in areas rather than the breeding density which is the metric of choice in southern areas where Malleefowl are more common (Natural Heritage Trust National Malleefowl Monitoring Project 2007). In this regard, the 35 mounds recorded at Yeelirrie provide an excellent basis for monitoring the occurrence of Malleefowl both because the presence of mounds is strong evidence that Malleefowl once resided and bred in the area, and because Malleefowl appear to habitually visit old mound sites in their home range, making these structures ideal places to find Malleefowl footprints. If recent signs of Malleefowl are detected near known mounds, it is likely that the birds are resident in the area and have intentionally visited the mound. However, if signs are not detected it would be necessary to search for signs more widely (several km2) in the

150 surrounding area before being able to conclude with any confidence that Malleefowl are not in the local area. Thus, while a fresh Malleefowl sign at a mound provides evidence of occupancy, systematic searches for prints and other signs is necessary to demonstrate that the birds are absent. Ideally, monitoring the Malleefowl occupancy in an area should be associated with survey for new sites. However, for the time being at least we feel that developing a systematic monitoring system for known sites is most urgent.

Although mounds provide a useful focus for the monitoring of persistence, it should be clear that knowing where mounds are is not a prerequisite of this type of monitoring. Areas in which Malleefowl are resident can be identified by their footprints and could be monitored even without knowing the location of mounds. At its simplest, this may simply involve searching for prints along a few selected transects where Malleefowl have been frequently encountered in the past.

Where there are concentrations of prints, active mounds are also likely during years of good winter rainfall, and it may be beneficial to focus on finding these mounds in the future. Techniques were described for finding active mounds in the 2006 project work plan, but in the field we gave priority to searching transects to obtain a better coverage of Yeelirrie. As this has now been achieved, the priorities should tip toward searching for active mounds during the next monitoring visit.

In conclusion, the Malleefowl population at Yeelirrie appears healthy and, given the size and location of the population, is of national significance. It is not known whether Malleefowl still occur in properties neighbouring Yeelirrie; occasional sightings are reported to wildlife authorities that may indicate breeding populations elsewhere, but the scarcity of sightings suggests that sizeable populations of Malleefowl are rare in central Western Australia. In any case, Yeelirrie has clearly been sensitively managed over the past few decades (especially in regard to low stocking rates), and is an excellent example of Malleefowl populations in the mulga lands of Western Australia. The number and persistence of Malleefowl on the property is a credit to its managers.

References

Benshemesh, J. (2000). National Recovery Plan for Malleefowl. Malleefowl Recovery Team and Department of Environment and Heritage, Adelaide.

Benshemesh, J., Dennings, S. & Danzi, C. (2007). Malleefowl Searches at Yeelirrie Station 2006. Unpublished report to the Malleefowl Preservation Group and BHP Billiton Nickel West, Ongerup and Perth, WA.

Benshemesh, J., & Malleefowl Preservation Group. (2001). Community searches for Malleefowl at Yeelirrie, August 2000. Unpublished report, Malleefowl Preservation Group, Ongerup, Western Australia.

Frith, H. J. (1962a). Conservation of the Mallee Fowl, Leipoa ocellata Gould (Megapodiidae). CSIRO Wildl. Res. 7:33-49.

Frith, H. J. (1962b). The Mallee Fowl. Angus and Robertson, Sydney.

Garnett, S. T., and Crowley, G. M.. (2000). The Action Plan for Australian Birds. Environment Australia, Canberra.

Natural Heritage Trust National Malleefowl Monitoring Project. (2007). National manual for the Malleefowl montoring system. Victorian Malleefowl Recovery Group and Mallee Catchment Management Authority, Melbourne.

Priddel, D., Wheeler, R. & Copley, P. (2007). Does the integrity or structure of mallee habitat influence the degree of Fox predation on Malleefowl (Leipoa ocellata)? Emu 107:100- 107.

151 Sanders, A., Benshemesh, J. & Malleefowl Preservation Group. (2003). WMC Operation Malleefowl Mulga to Mallee Research Project: A report on community searches for and monitoring of Malleefowl at Yeelirrie Station, August 2003. Malleefowl Preservation Group, Ongerup, Western Australia.

152 19. RE-SEARCHING MALLEEFOWL MONITORING SITES – A VICTORIAN MALLEEFOWL RECOVERY GROUP PERSPECTIVE

Peter Stokie Victorian Malleefowl Recovery Group

Abstract

The importance or searching monitoring sites periodically is essential for the continuation of scientifically valid data on Malleefowl breeding density. However it is probably the hardest thing to maintain due to the significant restraints of the time involved in conducting a systematic search and the difficulty of having sufficient personnel to conduct the search. This paper explores some of the methods the Victorian Malleefowl Recovery Group have employed in their searches, and an analysis of the relative successes and difficulties involved. The value of being well prepared, having well trained, motivated and supervised searchers, and using modern technologies to assiist with searches increases the viability to conduct successful searches. The history of past searches and the strategies used for recent, much improved searches will be presented as a model for other groups to consider.

Introduction

The importance of maintaining the scientific integrity of the Malleefowl monitoring sites cannot be stressed enough. The quality of data collected relies on the training and dedication of volunteers. The analysis of data gathered over the past twelve years by volunteers is of the highest standard. Volunteers visit every known Malleefowl mound within each site, the only variable being whether Malleefowl have started new mounds that are not known.

Why re-searching monitoring sites is important

• Malleefowl occasionally choose to build new mounds rather than renovate an old mound • It is important to search the monitoring sites every few years to record these • Mounds that are found during these repeat searches are added to the list of nests to check annually

In Victoria, there have been several attempts to maintain data integrity by re-searching existing sites using a variety of groups of volunteers, but due to the labor intense requirements to conduct a search it has been difficult to sustain the effort to meet the objective of re-searching every site every five years.

The initial approach to provide significant personnel for searches was to seek out organisations that provided assistance to environmental groups.

Pre 2004 search history

The Australian Conservation Volunteers and Greencorps were used in two major attempts to re-search the Victorian sites. Government contractors organised and supervised the searches.

– 1991/2 Australian Conservation Volunteers 4 sites – 1997 Greencorp 16 sites – In both instances the search teams were relatively young and inexperienced, especially in their knowledge of Malleefowl and working as a team in sometimes difficult terrain was a challenge.

Some of the key lessons learned from these efforts were: – Leadership – the choice of an appropriate group leader was crucial

153 – Organisation – being well prepared with directions and maps and local knowledge of the area to be searched – Keeping fun in the exercise – Developing technology to assist data collection – the use of GPS and hand held computers were not available in these searches and keeping records on paper was time consuming and inefficient

2004 – 2007 search history

The Victorian Malleefowl Recovery Group had taken on the responsibility of Malleefowl monitoring in 1999 to collect data from Mallefowl sites previously collected by government staff and contractors. Initially the Victorian Malleefowl Recovery Group gave no thought to the need to re-search sites, concentrating on data collection only.

The 2004 Annual Report on monitoring was a wake up reminder to the group that re- searching was an important aspect of the work of monitoring Malleefowl.

“Finally, I must reiterate that some of the declines we have witnessed may be more apparent than real because it has now been 8 years since most sites were searched for new nests.” J Benshemesh – Malleefowl monitoring in Victoria 2004/5

The Victorian Malleefowl Recovery Group rose to the challenge of this statement from the 2004 report and in the three years to 2007 have organised 15 existing sites re-searches and established 7 new sites.

A single strategy of using one organisation was no longer possible so different groups were used: – Greencorps – Technical and Further Education Students – Parks staff – Victorian Malleefowl Recovery Group members – Community

Group Re-search New

Greencorps 7 1 Technical and Further 5 Education Students Parks staff 1 Victorian Malleefowl 2 1 Recovery Group Community 1 4

• The Greencorps group was a partnership comprising the Mallee Catchment Management Authority, Greening Australia and the Victorian Malleefowl Recovery Group and operated similarly to the 1997 re-search, with the Victorian Malleefowl Recovery Group providing technical advice and helping with some searches • Technical and Further Education students and their teachers undertaking from Royal Melbourne Institute of Technology and Britafe Bendigo used the re-search exercise as a practical component of the course. Victorian Malleefowl Recovery Group provided an equipment and instructional material to each of these groups • Parks staff were self sufficient • Victorian Malleefowl Recovery Group members had not conducted re-searches previously, but their experience of tracking and familiarity with mallee country was highly beneficial • Community groups were generally starting from a very low knowledge and experience base and needed detailed assistance

154 • Search Efficiency

100%

90%

80%

70%

efficiency Search 60%

50%

19901992 1995 1997 2000 200520052006 2010

Figure 1. Search efficiency of individual re-searches.

Search efficiency has varied considerably in every attempt (Figure 1) and the essential elements to improve the effectiveness of a search are: • Leadership • Coordination • Equipment knowledge and availability – Maps – GPS – Cybertracker • Education • Enthusiasm

The Victorian Malleefowl Recovery Group has been the beneficiary of several small grants to enable groups to complete the tasks associated with re-searching without incurring much personal expense. Grants have been received from Parks Victoria, The Wilderness Society and The Natural Heritage Trust Malleefowl Multi regional project for community groups to re- search sites and establish new sites. The grants have been used to cover camping and food expenses and to offset travel costs through limited fuel reimbursement. The average expenditure for a site re-search has been $1000. This has been particularly helpful to Technical and Further Education students to keep the costs of a week out in the bush within their limited budgets, and a great help to attract community groups who otherwise may not have been able to afford the additional costs associated with this type of activity. In fact all community groups involved with the Victorian Malleefowl Recovery Group have used the opportunity to arrange the activity as a fundraiser for local organisations, and have donated their expenses in particular to local hospitals and kindergartens.

Community Groups that accepted the challenge of being involved in three or four days of site re-search activities include: • Hopetoun (Vic) Bike Riders • Hopetoun (Vic) Kindergarten parents • Natimuk Urban Landcare

155 • State Emergency Service from Nhill District • Nhill Lions Club • Mid Murray Field Naturalists • Wedderburn locals supporting the Wedderburn Conservation Management Network

Lessons Learned from 2004 – 2007 re-searches

The expected outcomes

• A significant number of sites, (60% of total sites) were successfully searched generally by people who had no previous experience of finding Malleefowl mounds or working in team situations • Search techniques were efficient with 85% of existing mounds found • Some ―new‖ mounds were found • New mound information has been incorporated into the database for future monitoring programs beginning in 2007 • Meeting an objective of the National Malleefowl Recovery Plan to re-search sites and maintain the scientific integrity of the Victorian Malleefowl Recovery Group monitoring program • Positive group experience through sharing information and friendship during searches and at barbecues and meal breaks. As a consequence enthusiasm was high and was maintained • Above all the activities were fun, but challenging

The unexpected outcomes

• New skills of using GPS, Palm Pilots and reading signs on mounds for Malleefowl activity, and other signs such as fox presence have been acquired by participants • The broadening of knowledge of the scientific program of monitoring Malleefowl was greatly enhanced • The broadening of knowledge of, and respect for, mallee vegetation and a broad scale understanding of mallee ecology was enhanced • Significant numbers of people and different groups involved • The take up rate from community groups was not as easily achieved as was indicated in surveys and discussions prior to funding applications being submitted • Very high percentage of people new to Malleefowl (80% and mainly young people) and the Mallee in general (50%) was a pleasing and unexpected outcome • Money allocated to local communities and educational institutions to offset costs was much appreciated • Requests from Technical and Further Education Colleges to continue their involvement with the activities of the Victorian Malleefowl Recovery Group provided positive feedback • Excellent local press coverage, articles in educational institution publications and word of mouth publicity generated much interest in the activities of the Victorian Malleefowl Recovery Group • High interest in new membership of the Victorian Malleefowl Recovery Group

The conclusions

• The task of re-searching sites is well received by community and students • Great reservoir of enthusiastic people willing to be involved and learn more about the bush and its conservation • Untapped resource of under-graduate students, and the value of enthusiastic and committed staff in tertiary institutions • Community welcome fund-raising opportunities, and are willing to be involved if they can see some benefit for their community • Inexperienced monitors will be successful with good training and quickly develop improved site searching techniques • Essential to be well organised, and ensure plenty of time for preparatory planning and preparation

156

Future Plans

• Develop an educational, instructional package for groups • Encourage community and educational groups to include re-searching as an annual activity for groups • Continue to support existing Technical and Further Education groups and investigate the involvement of other Tertiary Institutes and Community Groups • Seek the appointment of a paid co-ordinator in Victoria to arrange projects of this type, or have a volunteer co-ordinator‘s expenses paid to cover costs of administration • Seek more permanent sources of funding to expand the re-search activity

Concluding Remarks

One of the questions addressed in the data analysis during the Malleefowl Multi-regional Project was the issue of how often it is necessary to re-search existing sites. The current National Malleefowl Recovery Plan indicates that every five years was an ideal time frame. The data from the twenty re-search activities during the 1990‘s and the fifteen re-searches from 2004 to 2007 have been analysed to check on new mounds discovered and the activity associated with these mounds. The following conclusions can be drawn from this investigation: – Re-searching is important to locate new mounds – Re-searching sites burnt in the period of the past 20-30 years is most important when Malleefowl are moving back into the area: 3-5 years desirable – Re-searching sites where Malleefowl are increasing is important: 3-5 years desirable – At other sites 5-10 years is a reasonable re-search sequence

157

20. DOES THE INTEGRITY OR STRUCTURE OF MALLEE HABITAT INFLUENCE THE DEGREE OF FOX PREDATION ON MALLEEFOWL?

David Priddel1, Robert Wheeler1 and Peter Copley2 1Department of Environment and Climate Change (NSW), P.O. Box 1967, Hurstville, 2220, N.S.W. 2Department of Environment and Heritage, G.P.O. Box 1047, Adelaide, 5001, S.A.

Abstract

Mallleefowl are in decline across their range. Previous studies have found that the survival rate of young Malleefowl is low, the single greatest cause of mortality being predation by the introduced Red Fox (Vulpes vulpes). Many of these studies, particularly those in New South Wales, were conducted in habitats that were heavily modified by fire, exotic herbivores or plant harvesting. In this paper, we examine the survival of Malleefowl in relatively undisturbed mallee habitats within two conservation reserves in South Australia. Both reserves were long unburnt and free of large exotic herbivores, but differed greatly in understorey structure. Fifteen young captive-reared Malleefowl were released in each reserve. In all, 70% of these individuals were dead within 40 days. Fox predation was the prime cause of mortality, accounting for at least 35%, and perhaps as much as 96%, of all deaths. The rate and causes of mortality were similar in the two reserves. The overall rate of Malleefowl survival was (1) better than that recorded in more-disturbed habitat in New South Wales in the absence of fox control, but (2) substantially less than that in New South Wales after widespread fox control was implemented. This study indicates that Malleefowl in South Australia are subject to significant levels of fox predation, even in relatively undisturbed habitats. Available data indicate that during the past two decades Malleefowl populations in South Australia have declined at about the same rate as those in New South Wales. Current declines in South Australia are typically about one quarter of what they were 15 years ago. For the two mallee habitats examined, evidence suggests that understorey structure had no influence on the degree of predation. We conclude that habitat integrity and structure have very little effect on Malleefowl demography, and suggest that Malleefowl populations across Australia are threatened by foxes, placing the species at substantial risk of extinction.

Introduction

In 1990 we were able to test the possibility that captive–bred Malleefowl, released into the wild, would survive better in areas of undisturbed and long unburnt mallee compared with birds released into disturbed areas, where fox predation is known to cause high mortality in the released birds (Priddel et al. 2007). The study also examined differences in survival of captive-bred Malleefowl between dense and less dense stands of mallee. The project was a joint one between the New South Wales Department of Environment and Climate Change and the South Australian Department of Environment and Heritage. Historical data on number of active mounds, mortality rates and fox baiting protocols from both Departments were used in the analysis. The Monarto Zoological Park was able to provide captive-bred Malleefowl that they reared from eggs taken in the parks in which the birds were to be released.

The comparison was made between two mallee reserves in New South Wales, Yathong (107,241 ha, only half mallee) and Yalgogrin (558 ha) and two mallee reserves in South Australia, Bakara ( 1031 ha) and Ferries McDonald (843 ha). Yathong was badly disturbed. Until 1978 it was heavily grazed by sheep, rabbits and some cattle; when the sheep and cattle were removed, goats multiplied and were still numerous when the study was undertaken. Wildfires had burnt much of the reserve in 1957, 1975, 1984 and 1994. At Yalgogrin sheep graze the area and there is broombush harvesting and cutting eucalypts for oil as well. Bakara lies adjacent to 1300 ha of privately owned remnant mallee and has not been burnt for at least 80 years. The mallee is relatively thin with an understorey of spinifex. Ferries McDonald is dense mallee, reserved in 1956 and not burnt since.

158 Methods

Before the releases at Bakara and Ferries-McDonald a number of releases of captive-bred Malleefowl, raised at Western Plains Zoo, had been made at Yathong and Yalgogrin over the years. These previous releases were of chicks ranging in age from a week to 6 months of age, and all were fitted with radio transmitters. Fifteen Malleefowl chicks between 97 and 190 days of age were released at each of Bakara and Ferries McDonald on 1-2 June 1990. Each carried a 25g radio transmitter (AVM type, SB2). They were tracked and located daily for the first 11 days after release and intermittently thereafter for the life of the transmitter, about 4 months. The signal from the transmitter doubled in rate if the bird did not move for 10 hours and this enabled searches to be made for chicks that died and the remains inspected to ascertain the cause of death.

The density of the mallee on the two South Australian reserves was measured quantitatively using quadrats. Canopy and understorey cover was recorded and a horizontal structural measure also taken. The results were used to calculate an index of vegetation density. Standard statistical techniques were used to examine trends in the numbers of active mounds from historical and recent data.

Results

The results of the various releases at Yathong, Yalgogrin, Bakara and Feries McDonald are given in Table 1. References to the details of the New South Wales releases are given in Priddel et al. (2007).

Table 1. The results of releasing captive-bred Malleefowl of various ages into four mallee remnants. Note that all the fox baiting was from vehicles, except the widespread baiting at Yathong that was done both from the air and from vehicles. The figure in the last column is a minimum, confirmed count. Many more may have been taken by foxes.

Location of Fox control Area Baits/ No. Age of % alive % alive Known remnant with sq. km birds birds: after after 90 % lost baits sq let go months 30 days days to fox km S.A. Bakara None 15 3-6 33 27 27 Ferries None 15 3-6 33 33 50 McDonald N.S.W. Yalgogrin None 17 0-4 24 0 18 Yalgogrin None 15 3-6 7 0 73 Yathong None 23 3-5 4 0 52 Yathong Localised 0.4 7.5 12 6-8 50 42 86 Yathong Localised 6.4 7.5 12 8-11 33 17 70 Yathong Localised 19.2 1.5 24 4-5 42 25 45 Yathong Widespread 107.2 3.5 24 6-7 88 75 33

159 QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 1. The numbers of active mounds of Malleefowl per square kilometre at Yalgoggrin in New South Wales (open circles) compared with the number of active mounds per square kilometre in six reserves in South Australia (closed circles).

Figure 1 shows, as mounds per square kilometre, the numbers of active mounds at Yalgogrin in New South Wales between 1986 and 1999 compared with the numbers in South Australia. The South Australian figures are combined data from six sites that were monitored between 1989 and 2005. The regression lines have been drawn separately for the New South Wales and South Australian data.

Discussion

The survival rate of captive-bred Malleefowl released into New South Wales reserves has been shown to be dependent on the level of fox control implemented (Priddel & Wheeler 1999). Table 1 shows that at Yalgogrin and Yathong no birds survived to 3 months without fox control. When localised baiting was undertaken 33-50% of the released birds survived for at least one month and 17-42% survived at least 3 months. After widespread fox baiting, 75% of the birds survived for at least 3 months. The survival of captive-bred Malleefowl released at Bakara and Ferries McDonald in the absence of fox control was 30% after 3 months. Thus compared with a series of releases at Yathong and Yalgogrin where the vegetation was more disturbed than at the South Australian sites, Malleefowl released at Bakara and Ferries McDonald survived better than those released in the absence of fox control, about as well as those released when localised fox control was undertaken and less well than those released when widespread fox control was implemented. There was no discernable difference between the survival of birds released into Bakara and Ferries McDonald, although the vegetation was twice as dense at Ferries McDonald as at Bakara. Apparently the dense vegetation did not offer additional protection to the Malleefowl from predation by foxes. The numbers of active Malleefowl mounds is declining at the same rate in New South Wales and South Australia and the trend indicates that the bird may soon be extinct in the areas from which the data were derived.

It is difficult to avoid the conclusion from the experiments undertaken so far, that fox control is only effective if it is implemented intensely and simultaneously over a wide area.

160 Acknowledgements

We are grateful to the staff of the South Australian Department of Environment and Heritage who collected the eggs for the South Australian trials, and to the Monarto Zoological Park who incubated those eggs and reared the chicks. Many people assisted with the radio tracking and we are grateful for their dedicated help.

References

Priddel, D. & Wheeler, R. (1999) Malleefowl conservation in New South Wales: a review. . In Dekker, R.W.R.J., Jones, D.N. and Bensehmesh, J. (eds.) Proceeding of 3rd International Megapode Symposium, Nhill, Australia. Zoologische Verhandelingen 327:125-141.

Priddel, D., Wheeler, R. & Copley, P. (2007). Does the integrity or structure of mallee habitat influence the degree of fox predation on Malleefowl (Leipoa ocellata)? Emu 107:100- 107.

161 21. MALLEEFOWL AT MT. GIBSON MINESITE

Martine Scheltema1 and Peter Jones2 1Coffey Environmental, 2 Bulwer Street, Perth, 6000, Western Australia 2Extension Hill Pty., Ltd., Level 2, 7 Havelock Street, Perth

Introduction The Mt. Gibson Iron Ore project site lies about 280 km west of Geraldton in Western Australia. The mining lease contains a substantial population of Malleefowl that the company is committed to conserve, indeed it is compelled to do so by state and commonwealth legislation. The mine will extract two forms of iron ore, heamatite and magnetite. The magnetite will travel as a slurry along a pipeline direct to Geraldton and be exported from there. The corridor that contains the pipeline will also hold a road, a gas pipeline and a water pipeline to carry water from the Tathra borefield and return water to the Port of Geraldton. The project has an expected life of 20 years but has not yet had final environmental approval, although it has had a Public Environmental Review, the release of the Appeals Report and is expecting ministerial approval shortly. Once it has state approval it will then need to get commonwealth approval, but we are confident that both these will come through and are planning accordingly.

The lease area lies in the precinct of a number of private properties that are being actively managed for conservation. The Mt. Gibson Sanctuary of the Australian Wildlife Conservancy lies adjacent to the east of the proposed mine site, the Charles Darwin Reserve of Bush Heritage Australia lies to its west and Ninghan Station, owned by the Pindiddy Aboriginal Corporation and an indigenous protection are to its north.

The country is a banded ironstone formation, hilly, thickly vegetated, and difficult to traverse on foot. There is some flat country, but the Mt. Gibson Range dominates the lease area and is covered with Acacias, particularly mulga Acacia aneura, bowgida A. linophylla and sugar brother, A. collgardienesis. The mine site also lies at the intesection of two bioregions, the Yalgoo and the Avon Wheatbelt; the pipeline will also traverse the Geraldton Sandplain bioregion. The Malleefowl is one of several vulnerable or endangered organisms living in the area.

Methods

In order to assess the impact of the mine on the Malleefowl we consulted earlier records held by the North Central Malleefowl Preservation Group, Professor Harry Recher, Birds Australia, the Australian Wildlife Conservancy and the Western Australian Museum. We also undertook our own surveys, walking through the scrub, five metres apart and searching for mounds, active and inactive, on the minesite and surrounding areas, but omitting the top of the range where Malleefowl had not been seen. We scored mounds as active if they looked active; we did not dig them up. These surveys involved eight people for many weeks and were very labour intensive, but achieved results.

Because the vegetation is so thick and the ground surveys so expensive, time-consuming and restricted in area, we have been seeking methods that are more efficient than large parties searching on foot. The company is investigating the possibility of adapting Heli West‘s system that is used by Western Power to locate pole top fires, electromagnetic faults and fires that can‘t be observed visually. It is a three camera system with an ultra violet sensitive camera, an infra red long wave radiometric camera as well as a high resolution digital video to pick up temperature differentials between mounds and the surrounding soil. Heli West have ground truthed the system, comparing the camera technology with visual assessment and they think it will work well. If we can get it to work it will be potentially cost effective and will mean that much larger areas can be surveyed quickly and effectively so we will get realistic regional population numbers.

162 QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 1. Regional distribution of Mallefowl in the Mt Gibson area, based on various historical data sources.

QuickTime™ and a TIFF (LZW) decompressor are needed to see this picture.

Figure 2. The results of searching for Malleefowl mounds in the vicinity of Mt. Gibson minesite in 2005.

Results

163 A review of earlier observations showed that Malleefowl previously occurred in many parts of the area surrounding the mine site. Figure 1 shows the distribution of records of Malleefowl derived from various historical sources. The results of our recent searches for Malleefowl mounds in the area around the mine site are shown in Figure 2. During our search in 2005 we found 115 mounds, of which 15 were active. Most, but not all, of the mounds were in thicket country. Checking the active mounds again in 2007, we found that eight of them were not active in that year. In 2005 we also saw 4 live birds and one bird that appeared to have been killed and partly eaten by a fox.

The mine site itself will directly impact on three mounds active in 2005 and 25 mounds that were inactive in 2005.

Discussion

With regard to the Malleefowl the target will be no change in the number of active Malleefowl mounds in the general region. So it becomes very important to the company to have a valid way to measure active Malleefowl mounds because we are using that to monitor the Malleefowl population. Hence our interest in the Heli West system.

Mining at Mt. Gibson will have some direct negative effects on the Malleefowl. There will be a direct impact of clearing and development of the mine and its associated infrastructure, resulting in removal so some mounds and loss of habitat for nearby birds. There will also be disturbance from traffic on the many roads that will need to be built and disturbance at night from lighting around the mine and plant. In addition Malleefowl will suffer from fires and from predation by introduced predators.

The company is very conscious of its responsibility to mitigate these threats to the Malleefowl as far as possible. It has prepared a Malleefowl Conservation Plan that designates buffer zones around active and inactive mounds that are no go areas during the time breeding is occurring. Clearing impacting on Malleefowl mounds is only to be undertaken between April and June, that is, outside the breeding time. There is to be annual monitoring of active mounds and remedial action will be taken if a decline in the number of active mounds is observed. The company will support a regional predator control programme. There will be close collaboration with neighbouring properties, the North Central Malleefowl Preservation Group and the Department of Environment and Conservation to ensure a regional approach to Malleefowl management.

In addition to the specific conservation plan for Malleefowl the company plans to contribute to conservation efforts at a regional level. It wants to have an environmental benefit outside the footprint of the mine. It will take a regional approach to fire control, recognising that the area is prone to lighting strike, and that these often start fires in inaccessible area. It will have well- maintained fire fighting machinery on hand, and although it will not be able to prevent fires starting in the area, it should be able to prevent them becoming extensive and damaging large areas of bushland. The company also plans to help start and to support a Regional Conservation Association and provide funding of $100,000 a year to support on ground projects in that region, be they on public land or private land, so long as the projects are aimed at conservation. In addition the company has said the stations in the region that already focus on private conservation, will be provided with $150,000 a year for each station to do work on the ground.

In conclusion the company‘s aim is to continue to work with groups through direct on ground work, but also through knowledge, to maximise the knowledge and the conservation of the populations of Malleefowl in the region, both in the Mt. Gibson Ranges, but also in the general area. When the company says working with groups, that means with groups that are focused on conservation, Australian Bush Heritage, Australian Wildlife Conservancy, North Central Malleefowl Preservation Group, Ninghan Station, other mining companies in the area and both state and commonwealth agencies. If this comes to fruition it will be an opportunity to really maximise the benefit.

164

22. LITTLE DESERT NATURE LODGE MALLEEFOWL

Ray „Whimpey‟ Reichelt OAM Little Desert Nature Lodge, Nhill, Victoria

The Little Desert Nature Lodge is situated 16km from Nhill on the Nhill – Harrow Road. Our shire was named the ‗Shire of Lowan‘ in 1888, which is aboriginal word, which is a call the Malleefowl use when in danger. It was the largest shire in Victoria until broken up into smaller councils.

Whimpey started observing Malleefowl in 1967 with son Phillip. From their house they could easily get to the mound that was only 7 km from their home. Half of this block of land was cleared for farming in 1969 which stopped them watching these birds, as once cleared, the birds disappeared from this area.

In 1972 they started watching Malleefowl on the ‗Malleefowl Block‘ now known as the Malleefowl Sanctuary. This was to be cleared in 1973, they purchased this unburnt block of land to save it from being cleared. Maureen and Whimpey had to borrow half of the money to purchase the property, Maureen had put money away for kids education, which they had to use to pay for half of the block.

In 1979, Whimpey started seriously recording mound shapes, temperature and humidity and behaviour in the bush and at the mounds, paying a lot of attention to their diet. In 1988 they put a Conservation Covenant on the Sanctuary to protect it forever. In 1999 the blocks of land (including the Lodge Property) were feral proof fenced at a cost of $371,000 and finished in 2000, this was to keep out feral cats, foxes, rabbits and hares. This block of land was put into a tax deductable foundation in 2000.

The pair Whimpey has been watching, Charlie and Di, were together from 1979 – 1995. In that year Romeo lost Juliet and came and belted Charlie and took Di back with him to his mound known as Buckshot. Whimpey didn‘t see Charlie for three months, and at that time presumed he was dead. He returned after three months. Whimpey released three 12 month old chicks with Charlie and he eventually paired up with one in 1996. Di ended up with Romeo and they were together on Buckshot Mound from 1995 – 2000 when we lost old Romeo. Di called for a couple of months and a young male turned up, he has taken her to his mound called Phillips mound. Whimpey firmly believes that after watching these birds for so long that Malleefowl are not monogamous as Di is now with the third male.

Over the many years, Whimpey has hatched 147 chicks, these chicks are held between 8 -10 months until they are surgically sexed and get a new wing tag and leg band. The birds are then released in unrelated pairs into the National Park and on to private land.

Some 18 years ago, the female started with her high pitch call on the mound, the male moved to the mound and also started, it resembles the sound Lowan, very close to the mound was a fox coming in, it is a call they make when in danger.

After watching Malleefowl for some years, Whimpey heard the female calling one particular morning and went to see where she was. That morning he saw the birds copulate 4 times, shortly after she entered the mound and laid an egg. Not every time but most times before they cover the egg up, the male comes and scratches the pink off the egg. No one can tell me why, but I have caught on film on a number of occasions.

Whimpey is still uncertain what the birds use to check the temperature, My opinion is that the birds must have heat sensor in the roof of their mouth, why otherwise would the birds put their beak right up to their eyes into the hot sand of the mound?

Whimpey has been watching Malleefowl since 1979 making records of mound shapes, temperature, humidity and food sources. Until the 5/9/07 Whimpey has made 4,198 day visits to the mounds, filming and making many records. In autumn the birds gorge themselves with

165 Mushrooms, Toadstools and Fungus. Their droppings have been analysed by Dr Tom May of the Melbourne Herbarium and the same year, during the summer of 35 – 40 degrees the birds were digging in the grey clay and they were eating some small round tubers that resembled yams. Whimpey grew some of these and they turned out to be the Terestylus nana (Greenhood Orchid). There was nothing above the ground to see, so whether they can smell or remember where they were, we are still trying to work out.

Professor Darryl Jones and Whimpey have just finished the Food Chart that is now at the referees prior to being published this year. Our next two papers will be based on behaviour at the mound and mound shapes. They will most likely be published in 2008.

This year Whimpey recorded in Tower mound 6 Echidnas, and one in Phillips mound during the June/July period. The Echidnas lift the rotting vegetation and aerate the litter that assists with the composting of the vegetation.

In 1984 Joe Benshemesh brought 9 Malleefowl eggs from Wyperfeld National Park for Whimpey to hatch. Once hatched these chicks had transmitters attached to their backs. To Whimpey‘s knowledge they were dead in a matter of 6 weeks.

Maureen and Whimpey have recently purchased a 276 hectare property close by. This property is prime bush unburnt since 1958, used to protect more endangered species of flora and fauna including reptiles. This block is used by Swunburn University to further their environmental studies.

166