Integrating Conservation with Production:

The Ecology of Three Threatened Black Cockatoos within a

Mining Production Landscape in the Jarrah-Marri Forest of

Western Australia

Jessica Grace Huiyi Lee

Bachelor of Science in Biomedical Sciences and Molecular

Biology with Honours in Molecular Biology

Graduate Certificate in Zoology

This thesis presented for the Doctor of Philosophy in Biological

(Environmental and Conservation) Sciences at the School of Veterinary

and Life Sciences, Murdoch University

  i  

Declaration

I declare this thesis is my own account of my research and contains as its main content,

work, which has not been previously submitted for a degree at any tertiary educational

institution.

Jessica, Grace, Huiyi Lee

(28th February 2013)

  ii  

Abstract

Three threatened black cockatoos inhabit the Jarrah Eucalyptus marginata-Marri

Corymbia calophylla forest of southwestern Australia: Baudin's Cockatoo

Calyptorhynchus baudinii, Carnaby's Cockatoo C. latirostris, and the Forest Red-tailed

Black Cockatoo C. banksii naso (FRTBC). Their local ecology in relation to

anthropogenic disturbance is poorly known, hampering conservation management. This

study investigated their ecology at the Newmont Boddington Gold (NBG) mine, 130

km southeast of Perth, Western Australia, along the eastern margin of the Jarrah-Marri

forest. To improve the scientific basis for conserving black cockatoos and their habitat

at NBG, I aimed to: (1) describe the ecology of the three species at NBG, particularly

group size, site occupancy, habitat use, and food plant use (including seasonal and inter-

annual changes); (2) examine the effectiveness of ground-based hollow surveys, post-

felling inspections of hollows, and behavioural observations for assessing black

cockatoo breeding habitat; (3) assess the successional stage of the rehabilitated mine

pits and characterise variation in the structure and floristics of pits, in order to identify

features that might influence the availability of food resources for black cockatoos; (4)

document feeding activity by black cockatoos within rehabilitated mine pits and any

associations with structural or floristic features; (5) trial artificial nest hollows to

support breeding on-site and compensate for the loss of natural hollows; (6) review the

use of artificial nest hollows for black cockatoos to assess their value for mitigating

natural hollow loss; and (7) investigate black cockatoo use of natural and artificial water

sources at NBG and assess the potential for black cockatoo interactions with residue

disposal areas.

All three black cockatoos used remnant forest habitat as well as human-modified

habitats such as mine-site rehabilitation, water sumps, farm paddocks, and pine

plantations. Carnaby’s Cockatoos used the broadest range of habitats and fed on at least

ten plant species at NBG. FRTBC showed similar group sizes and occupancy across

seasons, suggesting year-round residency. In contrast, group size and occupancy

changed across seasons for Carnaby’s Cockatoos, indicating migrating flocks as well as

some birds present year-round. Few Baudin’s Cockatoos were observed in spring and

  iii  

summer, but they were more abundant during autumn and winter, which is when flocks

migrate northwards.

Three methods – ground-based surveys, post-felling inspections, and behavioural

observations of black cockatoos – were used to describe the availability of potentially

suitable nesting hollows across three tree species (Jarrah, Marri, Wandoo E. wandoo)

and to document probable nest sites. Eleven probable black cockatoo nest hollows were

identified at NBG and surrounds (Carnaby’s Cockatoos: n = 7; FRTBC: n = 3; and

unknown black cockatoo species: n = 1). Behavioural observations, using visual or

acoustic cues followed by physical ‘tree-knocking’, were the most effective approach to

identify probable nest hollows (n = 10 hollows). Ground-based surveys yielded only

one probable nest hollow, while post-felling inspections identified none, despite large

sample sizes and extensive field survey periods. Of the probable nest hollows, six were

in Marri, four in Wandoo, and one in Jarrah. Ground-based surveys identified 149

potential hollow-bearing trees, of which 119 (80%) survived felling intact enough for

inspection. Few potential hollows in Jarrah were large enough for black cockatoos (n =

28 of 89 trees inspected, 31.5%). Large hollows occurred more frequently in Marri (n =

14 of 22 trees inspected, 63.6%) and Wandoo (n = 8 of 12 trees inspected, 66.7%).

Thus ground-based surveys may significantly overestimate potentially suitable hollows,

but they may provide useful assessments of relative hollow abundance if biases are

identified and corrected. Post-felling inspections are ineffective at characterising hollow

occupancy, but provide other data. Targeted behavioural observations are most reliable

for identifying probable nest hollows, provided that surveys are undertaken at dawn and

dusk during known breeding seasons.

The rehabilitation pits at NBG were 7 - 10 years old and in an early successional stage,

with a wide range of proteaceous vegetation. The larger myrtaceous trees were

becoming prominent and will eventually shade out the proteaceous understorey.

Carnaby’s Cockatoo fed on seeds and flowers from proteaceous shrubs (Banksia and

Hakea spp.), while Baudin’s Cockatoo and FRTBC fed on Marri seeds. Examination of

food residues proved critical in demonstrating feeding activity by FRTBC within

rehabilitated mining pits, as FRTBC were not observed. No particular floristic or

structural features characterised feed plots relative to non-feed plots, suggesting little

support for substantially altering rehabilitation prescriptions to cater to the food

  iv  

requirements of black cockatoos. Rather, food availability within rehabilitated areas

reflects vegetation succession, with fast-maturing proteaceous species providing

abundant food in young rehabilitation sites, followed by regenerating eucalypts (e.g.

Marri). Therefore, as succession proceeds and the structure (and species composition) of

the vegetation becomes closer to that of the native Jarrah-Marri forest, the availability

of habitat resources is likely to change. Rehabilitation of mining pits can provide short-

term benefits for black cockatoos, emphasising the broader value of revegetating

landscapes to support black cockatoo conservation. However, tree hollows suitable for

breeding take over a century to form, so conserving old, hollow-bearing trees

complements restoring food plants.

Trials of plastic and wooden artificial nest hollows (ANHs) at NBG by black cockatoos

yielded no observations of inspection or nesting by black cockatoos. This likely related

to ANH position (i.e. too low in the tree) or location (i.e. far from an established nesting

site), and the presence of sufficient natural hollows locally. A state-wide survey of ANH

use by black cockatoos in Western Australia was also undertaken. Responses indicated

that ANHs have been widely used (at least 157 ANH installations) in Western Australia

from as early as 1996. Three major ANH designs have been used: ‘cockatubes’,

hollowed-out log sections, and wooden box-type ANHs. Black cockatoos nested and

reared young in all designs. Most observations of black cockatoos using ANHs involved

Carnaby’s Cockatoos. There were few records of ANHs by FRTBC and none by

Baudin’s Cockatoo. Thus ANHs may present a short-term mitigation option, especially

in areas deficient in natural hollows. However, over the long-term, they cannot

substitute for natural hollows.

Observations of black cockatoo drinking sites at NBG documented the use of both

natural and man-made water sources. The site, and the surrounding landscape, provides

water for black cockatoos across the year. Black cockatoos prefer water sources with

firm and gently inclined edges surrounded by vegetation. They rarely came into

proximity with potentially hazardous sections of the Residue Disposal Areas (RDAs).

RDAs may be less attractive because they occur in very open landscapes. Black

cockatoos used faunal drinking points around the RDAs, suggesting that they minimise

black cockatoo-RDA interactions.

  v  

Overall, the three black cockatoos exhibited different ecologies within the same

landscape, differing in their responses to disturbance and their capacity to use

anthropogenic resources. Significantly, the value of young rehabilitation for feeding

indicates that feeding habitat can be restored on a scale of a decade or two rather than

waiting for nearly a century as the forest matures. Breeding resources, however, are

only restored naturally on a scale of centuries, placing a premium on conserving prime

breeding habitat and artificial supplementation of the breeding hollow resource.

Given that much of the remaining Jarrah-Marri forest is within the tenure of State Forest

or mining companies such as NBG, there is considerable scope for adaptive

management initiatives to further inform conservation actions for black cockatoos.

These initiatives could investigate: (1) management of orchards; (2) captive breeding;

(3) general population ecology and baseline data collection; (4) issues related to nesting

hollows; (5) management issues specific to mine-sites; and (6) management issues

relevant to forested areas of the southwest, in particular State Forest.

  vi  

Publication

Some of the work presented in this thesis is published. While I took the major role in

the work, M. C. Calver and H. C. Finn assisted in the survey design and analysis. H. C.

Finn also assisted in fieldwork and in liaison with Newmont Boddington Gold.

Lee, J., Finn, H. and Calver, M. C., (2010) Mine-site revegetation monitoring detects

feeding by threatened black-cockatoos within 8 years. Ecological Management and

Restoration 11 (2): 141 - 143.

Lee, J.G.H., Finn, H.C. and Calver, M.C., (2013) Ecology of black cockatoos at a mine-

site in the eastern Jarrah-Marri forest, western Australia. Pacific Conservation Biology

19 (1): 76 - 90.

Lee, J., Finn, H. and Calver, M., (2013) Feeding activity of threatened black cockatoos

in mine-site rehabilitation in the jarrah forest of south-western Australia. Australian

Journal of Zoology 61 (2): 119 - 131.

  vii  

Acknowledgements

First and foremost, my utmost thanks and appreciation goes to my supervisors - Mike

Calver and Hugh Finn for their unceasing and relentless help, unwavering and persistent

patience, their constant encouragement, abounding wisdom, and most of all their

commitment to seeing me through this extensive intellectual experience. To Mike –

thanks for the long catch-ups, YouTubing sessions, humorous interruptions and for

giving me the lowdown on stats. To Hugh – thanks for showing me the ropes with the

fieldwork, the long dawn and dusk drives, and most of all for putting up with Echo, the

steamboat sessions, ‘El Wacko’ and ‘A Soft Touch of Asia’! Truly, I would not be

standing where I am today without your resolutions to this cause called a PhD, and this

thesis would not have been possible without you two. Surely, there is light out of rehab

and at the end of the hollow, and now truly, there’s no more going back to the PAST!

From the Newmont Boddington Gold Environmental team, my greatest thanks to

environmental managers - Tom Muth and Rynhard Kok, and to environmental officers -

Melanie Durack, Ross Polis, Sarah Bederak, Paul NcNeil, Devi Shanty, Mel Graham

and Jesse Steele for their continued help with on-site administration and assistance. To

the Security Team – thanks for all the black cocky tip-offs and bearing with my constant

radio check-ins. To the environmental technicians – Kerri and…especially Jo Batt! Gal,

remember those super-long rehab hours? Well, I would not have been able make it

through without you. Thanks for being amazing to work with, and for being such a great

gal-friend on-site and off. Thanks especially, for your companionship, hospitality and

cheer in brightening up the often monotonous mornings at the site.

(Newmont Boddington Gold provided support for this research)

I would also like to extend my thanks to Brad Stokes from Worsley Mine, and the

Alcoa Huntly Mine team – Paige Salmon, but especially Vicky Stokes -for her

friendship and help with organising the vegetation sampling at the Hedges site. It’s been

a pleasure!

To Tim Doherty and Briana Wingfield – it’s been a blast bashing through the rehab with

you two! Thanks for all your help, for the laughter and company! Wishing you two the

best life has to offer! And also to Erin Biggs and Rebecca Stutz who started out with

me!

  viii  

Thanks to Libby Mattiske and Mike ‘Spike’ Craig for your help in advising with the

rehab sampling.

Cheers to David Donato and team with providing some of the watering data.

My gratitude to Belinda Cale for drawing the figures (especially the horrendograms!) in

the thesis.

For black cocky help, big or small, thanks for the contributions of - Nicole White, Anna

LeSouef, Mike Bamford, Maree Weerheim, Jason VanWeenen, Peter (TasNature), Bill

(Dinosaur World), Neville Conners, Kim Whitford, Alan Burbidge, Tamara Chapman,

Hugh Possingham, David Paton, Martine Maron, Amy Koch, Teagan Johnston, Lynn

Pedlar, Leo Joseph, Mick Todd, Janelle Thomas, Mark Wapstra, Michael Barth,

Richard Hill, Matt Cameron and Stephen Garnett.

For other parrot help, thanks to the contributions of Pedro Martinez (Costa Rica),

Donald Brightsmith (Peru/USA), Neiva Guedes (Brazil) and André von Thuronyi

(Brazil).

My sincere thanks to the Artificial Nest Hollow Survey participants for their time, effort

and patience in contributing to the questionnaire - Dejan Stojanovic, Raana Scott, Peter

Mawson, Rick Dawson, Ron Johnstone, Tony Kirkby, Allan Elliot, Glen Byleveld,

Glenn Dewhurst, Caroline Minton and Stephen Davies. Thanks especially to Dejan for

your friendship and help in planning the questionnaire and contact list, as well as for all

the catch-up chats and calls. Cheers to Tony Kirkby for taking me out black cockatoo-

ing and teaching me the ID-ing skills. Thanks to Allan Elliot and Caroline Minton for

showing me the artificial hollows and Glenn Dewhurst for the aviary visit. Finally,

thanks to Christine Groom for your collaboration with the artificial hollow work.

To the three examiners (Stephen Davies, Dianne Brunton and Matt Cameron) – thank

you for your extensive feedback and comments. I have addressed them to the best of my

abilities, but may not have necessarily responded to or satisfied all of the comments

made in the reports.

To my family and friends, thanks for your steadfast patience, encouragement and

support.

Mom, Dad - thanks for continually pushing me forward, for your prayers,

understanding, your providence, concern and most importantly - for supporting the

pursuit of my dreams.

  ix  

Samy-sis…thanks for our late-night chats and your listening ear. Thanks for being one

who knows me and for the words of inspiration that only a sister can provide. I have and

am truly soaring on those ‘trusty pair of wings…’.

Bro – thanks for keeping me company while I work at home ;”>

‘Ah Kong’ – thanks for starting this whole thing off 25 years ago! I remembered the

first day you brought back a pair of budgies for me – I must have been 4 years old? I

recalled thinking they did not belong in a cage and set them free, you were quite

disappointed J But because of that, you planted that ‘egg’ in my heart that did not

hatch till I turned 12. That was when I knew that this was the flight path for me. I’d like

to dedicate this thesis to you, for helping me find my wings.

Andy and Priss – the other two amigos – thanks for being there for me through these

years. Thanks for the fixed but sporadic get-togethers, mini-road-trips and eating

sessions that has fueled me this far.

Pedro ‘Lapa’…I guess it can also be said ‘psittacines of a feather, flock together’.

Thanks for being part of that flock, for all the words of encouragement, for the fun and

unwinding chat sessions over bubble-tea sharing with my PhD rants, for the deep and

meaning conversations over coffee and the intellectually challenging scientific-religious

debates!

To my uni-mates – Heather, Elvina, Mel, Kat, Narelle, Wan Ho, Tina, Drew, and more

(to name a few). Thanks for the company, complaining and chillaxing sessions these

last years, and especially for giving me a place to crash at uni!

Ps Soks! Thanks for always checking up on me, for your care and your constant prayers,

but most importantly for being my gracious mentor!

To the CG – thanks for all your thoughts and prayers! Especially you ‘Chi-chi’ – thanks

for always keeping me in your thoughts!

But those who wait upon the Lord will renew their strength; they will mount up

and soar on wings like eagles, they will run and not grow weary, they will walk and not

be faint.

Isaiah 40:31.

Thanks to all the birds in my life that have inspired me (especially those special few –

you know who you are!) You bunch are the driving force fueling the pursuit of my

endeavors that has brought me thus far. Thanks to all the black cockies out there…y’all

are my motivation! This PhD would not have been possible without the lot of you!

  x  

Finally, my utmost thanks to God, for His grace in seeing me through this season of my

life; and for giving me the patience and the opportunity to be able to take off, pursue

and make a reality the passion that budded in my heart when I was a child.

Lend Me A Bird

"I will lend to you for a while, a bird", God said.

For you to love him while he lives and to mourn for him when he is dead.

Maybe for twelve or fourteen years, or maybe for two or three.

But will you, till I call him back, take care of him for me?

He'll bring his charms to gladden you and should his stay be brief,

you'll always have his memories as solace for your grief.

I cannot promise that he will stay, since all from earth return,

but there are lessons taught below I want this bird to learn.

I've looked the whole world over in search of teachers true.

And from the folks that crowd life's land, I have chosen you.

Now will you give him all your love; nor think the labor vain;

nor hate me when I come to take my lovely bird again?

I fancied that I heard them say, "Dear Lord, thy will be done,

for all the joys this bird will bring, the risk of grief we'll run."

Will you shelter him with tenderness?

Will you love him while you may?

And for the happiness you'll know forever-grateful stay?

But should I call him back much sooner than you've planned,

Please brave the bitter grief that comes and try to understand.

If, by your love, you've managed, my wishes to achieve,

In memory of him you've loved; be thankful; do not grieve.

Cherish every moment of your feathered charge.

He filled your home with songs of joy the time he was alive.

Let not his passing take from you those memories to enjoy.

"I will lend to you, a Bird", God said, and teach you all you have to do.

And when I call him back to heaven, you will know he loved you too.

(Author Unknown)

  xi  

Table of Contents

Declaration i Abstract ii Publication vi Acknowledgments vii Table of Contents xi List of Figures xvi List of Tables xx Acronyms and Abbreviations xxiv

Chapter 1 General Introduction 1 1.1 Overview of Chapter 1 1.2 Integrating Natural Resource Production and Nature Conservation 2 1.3 Human Activity and Biodiversity Conservation in the Jarrah-Marri Forest 9

1.3.1 The Jarrah Forest Bioregion 9 1.3.2 Timber Production 10 1.3.3 Mining 10 1.3.4 Reserve System 12

1.4 Problem Formulation 12 1.5 Study Aims and Thesis Structure 15

Chapter 2 Description of Study Species and Study Area 16 2.1 Study Species 16

2.1.1 Overview 16 2.1.2 Conservation Status 17 2.1.3 Species Description 18 2.1.4 Distribution, Range and Habitat 20 2.1.5 Seasonal Patterns and Movement 21 2.1.6 Foraging Ecology 22 2.1.7 Breeding Biology 23 2.1.8 Survival 24 2.1.9 Reasons behind Black Cockatoos being listed as threatened 24

2.2 The Study Area – Newmont Boddington Gold 25 2.2.1 Regional and Geographical Setting 25 2.2.2 Disturbance History 27 2.2.3 Biogeography, Topography, Hydrology and Climate 28 2.2.4 Vegetation Structure and Flora 28 2.2.5 Vertebrate Fauna 29 2.2.6 Black Cockatoos 29 2.2.7 Conservation Values, Concerns and Issues for Newmont Boddington Gold Black Cockatoos 29

Chapter 3 Ecology of Black Cockatoos at a Mine-site in the Eastern Jarrah-Marri Forest, Western Australia

31

  xii  

3.1 Introduction 32 3.2 Materials and Methods 34

3.2.1 Study Site and Species 34 3.2.2 General Distribution Survey (GDS) 36 3.2.3 Behavioural Methodology 40 3.2.4 Group Size 41 3.2.5 Habitat Type and Use 42 3.2.6 Food Plants 43 3.2.7 Site Occupancy 44

3.3 Results 44 3.3.1 Group Size 44 3.3.2 Habitat Use 47 3.3.3 Food Plants 47 3.3.4 Site Occupancy 52

3.4 Discussion 56 3.4.1 Baudin’s Cockatoos 56 3.4.2 Carnaby’s Cockatoos 57 3.4.3 Forest Red-Tailed Black Cockatoo 58 3.4.4 Management Implications 60

Chapter 4 Methods to Assess Breeding Habitat for Black Cockatoos in the Jarrah-Marri Forest of Southwestern Australia

63

4.1 Introduction 63 4.2 Methods 64

4.2.1 Study Area Characteristics 64 4.2.2 Ground-based Surveys 65 4.2.3 Post-felling Inspection 66 4.2.4 Behavioural Observations 67

4.3 Results 68 4.3.1 Ground-based Surveys 68 4.3.2 Post-felling Inspection 71 4.3.3. Behavioural Observations 71

4.4 Discussion 72 4.4.1 Methodological Implications 72 4.4.2 Management Implications 75

Chapter 5 Variation in the Structural and Floristic Features of Mine-Site Revegetation Pits of a Similar Age

77

5.1 Introduction 77 5.1.1 Problem Formulation 77 5.1.2 Objectives 80

5.2 Methodology 80 5.2.1 Study Area 80

  xiii  

5.2.2 Rehabilitation Protocols 81 5.2.3 Selection of Pits for Sampling 82 5.2.4 Vegetation Structure, Floristics and Phenology: Variables 83 5.2.5 Vegetation Structure and Floristics: Sampling Methods 84 5.2.6 Analysis 86

5.3 Results 89 5.3.1 Structural Variables 89 5.3.2 Floristics 96 5.3.3 Successional Stage 99

5.4 Discussion 100 5.4.1 Successional Stage 100 5.4.2 Differences between Exterior and Interior Plots 100 5.4.3 Differences between Pits 100 5.4.4 Implications 101

Chapter 6 Recolonisation of Mine-site Revegetation by Threatened Black Cockatoos in the Jarrah-Marri Forest of Southwestern Australia

103

6.1 Introduction 103 6.2 Methods 104

6.2.1 Study Area and Rehabilitation Protocols 104 6.2.2 Observations of Black Cockatoos in Rehabilitated Mine Pits 104 6.2.3 Identifying Feeding Residues in Rehabilitated Mine Pits 105 6.2.4 Marri Feeding 107 6.2.5 Assessment of Feeding Activity 108 6.2.6 Data Analysis 109

6.3 Results 111 6.3.1 Observations of Black Cockatoos in Rehabilitated Mine Pits 111 6.3.2 Patterns of Feeding Activity Revealed by Feeding Residues 112 6.3.3 Feeding on Marri 125 6.3.4 Assessments of Feeding Activity 125

6.4 Discussion 127 6.4.1 Observations of Black Cockatoos in Rehabilitated Mine Pits 127 6.4.2 Patterns of Feeding Activity Revealed by Feeding Residues 128 6.4.3 Assessments of Feeding Activity 129 6.4.4 Management Implications 129

Chapter 7 Suitability of Artificial Nest Hollows for Black Cockatoos at Newmont Boddington Gold

132

7.1 Introduction 132 7.2 Methodology 133

7.2.1 Artificial Nest Hollow Design 133 7.2.2 Artificial Nest Hollow Placement 135 7.2.3 Artificial Nest Hollow Monitoring 140

7.3 Results 140

  xiv  

7.4 Discussion 140 7.4.1 General Discussion 140 7.4.2 Management Implications 143 7.4.3 Concluding Remarks 146

Chapter 8 Are Artificial Nest Hollows a Suitable Method for Mitigating Impacts of Mining on Black Cockatoos in the Jarrah-Marri Forest? – A Survey of the Users of Artificial Nest Hollows in Western Australia

147

8.1 Introduction and Problem Formulation 147 8.2 Methodology 152

8.2.1 Design of Survey 152 8.2.2 Survey Procedure 152 8.2.3 Presentation of Findings 153

8.3 Results 153 8.3.1 Survey Details 153 8.3.2 Survey Participants 153 8.3.3 History of Implementation and Designs of Artificial Nest Hollows 156

8.3.4 Locations where Artificial Nest Hollows were Used 156 8.3.5 Characteristics of Successful Artificial Nest Hollows 163 8.3.6 Installation, Maintenance, Monitoring and Costs Associated with Successful Artificial Nest Hollows 163

8.3.7 Hazards to Black Cockatoos 168 8.3.8 Hazards to Artificial Nest Hollows 172 8.3.9 Respondents’ Priorities for Future Research on Artificial Nest Hollows 172

8.4 Discussion 172 8.4.1 What Factors Contribute to a Successful Artificial Nest Hollow Placement? 172

8.4.2 Risks to Black Cockatoos or Workers in Establishing and Maintaining Artificial Nest Hollows 173

8.4.3 The Place of Artificial Nest Hollows in Black Cockatoo Conservation and Research Priorities for Artificial Nest Hollows 174

8.4.4 Strengths and Limitations of the Study 176 8.4.5 Concluding remarks 177

Chapter 9 Water Use by Black Cockatoos at the Newmont Boddington Gold Mine: Natural and Artificial Water Sources and Risk of Interaction with Residue Disposal Areas

179

9.1 Introduction 179 9.1.1 Problem Formulation 179 9.1.2 Black Cockatoos and Water 179 9.1.3 Fauna-Residue Disposal Area Interaction at Newmont Boddington Gold 183

  xv  

9.1.4 Aims 190 9.2 Methodology 190

9.2.1 Water Sources: Types, Timing and Nature of Use 190 9.2.2 Behavioural Observations 193 9.2.3 Camera Trap Observations 194 9.2.4 Wildlife Monitoring by Newmont Boddington Gold Staff 197 9.2.5 Monitoring of Wildlife Mortality by Donato Environmental Services Personnel and Newmont Boddington Gold Staff 197

9.3 Results 198 9.3.1 Behavioural Observations 198 9.3.2 Camera Trap Observations 205 9.3.3 Donato Environmental Services Observations 205 9.3.4 Wildlife Monitoring by Newmont Boddington Gold Staff 206 9.3.5 Monitoring of Wildlife Mortality 207

9.4 Discussion 207 9.4.1 Behavioural Ecology of Black Cockatoo Water Use at Newmont Boddington Gold 207

9.4.2 Range and Characteristics of Water Sources 208 9.4.3 Is there Evidence that Black Cockatoos Use, or Occur in Proximity to, the Residue Disposal Areas? 210

9.4.4 Use of Fauna Drinking Points 212 9.4.5 Limitations and Future Improvements 212

Chapter 10 General Discussion and Synthesis 214 10.1 Overview of Principal Findings 214 10.2 Original Contributions of the Research 226 10.3 An Integrated Adaptive Management Approach to Conserving Black Cockatoos at the Newmont Boddington Gold Site and More Broadly in the Forested Southwest

227

10.3.1 Adaptive Management 227 10.3.2 Adaptive Management and the Newmont Boddington Gold Site 232

10.3.3 The Newmont Boddington Gold Site 232 10.3.4 The Wider Southwest 234

10.4 Implications for Psittacine Conservation Globally 237 10.4.1 Global Threats to Psittacines 237 10.4.2 What is being done about the threats? 240 10.4.3 What are the implications of my findings for conservation of psittacines elsewhere? 249

10.5 Concluding Remarks 251

References 253

Appendix I 298

Appendix II 300

  xvi  

Thesis List of Figures

Figure 1.1 The forested bioregions of southwestern Australia: Swan Coastal Plain, Jarrah Forest and Warren (adapted from Calver and Wardell-Johnson 2004). 9

Figure 2.1 (a) Carnaby’s Cockatoo male ♂ (left) and female ♀ (right); (b) Baudin’s Cockatoo male ♂ (left) and female ♀ (right) and (c) FRTBC male ♂ (left) and female ♀ (right). Images from Johnstone and Storr (1998).

19

Figure 2.2 Distribution and range of (a) Carnaby’s Cockatoos; (b) Baudin’s Cockatoos and (c) FRTBC, respectively (Images reproduced from Department of Sustainability, Environment, Water, Population and Communities 2011).

21

Figure 2.3 The location of the study site near Boddington, southwestern Australia. 26

Figure 3.1 Map of study area showing the NBG mining areas and residue disposal area, the Land-Exchange Area, and surrounds. The cleared land/non-native vegetation to the northeast of the residue disposal area is largely pine plantations and blue-gum (Eucalyptus globulus) plantations, while that to the east and southeast is mainly paddocks.

35

Figure 3.2 Location of the site survey points for the black cockatoo general distribution survey across the NBG study site. 37

Figure 3.3 Location of site survey points for the black cockatoo general distribution survey across the LEA study site. 38

Figure 3.4 Flock sizes of three species of black cockatoos seasonally at NBG and LEA, using data aggregated over the period 2007 – 2010. 46

Figure 4.1 Locations of black cockatoo nesting and hollows identified in the study site. 70

Figure 5.1 General site map and location of pits. Black dots indicate pits used in the vegetation sampling. 81

Figure 5.2 Dimensions and layout of the vegetation sampling design: (a) 10 m × 10 m interior plot with the 10 m transect and (b) 20 m × 5 m exterior plot with the 10 m transect, and (c) positions of sampling plots around the rehabilitation for sampling (not drawn to scale).

85

Figure 5.3 Bar graphs showing for interior and exterior plots at each of nine rehabilitation pits: mean canopy cover, mean canopy height, mean non-canopy height, mean stem density (live), mean stem density (dead) (a) – (e) and species diversity (live) (where a species diversity was calculated across all interior plots and all exterior plots for each rehabilitation pit) (f). Error bars are standard errors for all mean values and bootstrapped 95% confidence intervals for species diversity. The bootstrapped 95% confidence limits are not symmetrical, so upper and lower boundaries are shown. For all except species diversity, pits shown to differ in mean values at the 5% level using Tukey’s honest significant difference following ANOVA are shown with superscripts. For species diversity, similar superscripts indicate significant differences between species diversity values for interior plots (lower case) and exterior plots (upper case) as determined by t-tests after sequential Bonferroni correction.

90 - 91

  xvii  

Figure 5.4 (a) Non-metric MDS for the structural similarity of nine rehabilitation pits using the variables canopy cover, canopy height, non-canopy height, stem density (live) and stem density (dead). Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.1753. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units.

93

Figure 5.4 (b) Non-metric MDS for the structural similarity of interior and exterior plots across nine rehabilitation pits using the variables canopy cover, canopy height, non-canopy height, stem density (live), stem density (dead) and species diversity of live plants (here interior pits are indicated by the solid line and square points, while exterior pits are indicated by the broken lines and cross points). The similarity measure was Bray-Curtis and the stress was 0.1753. 95% confidence ellipses are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

94

Figure 5.5 (a) Non-metric MDS for the floristic similarity of nine rehabilitation pits, based on the number of live stems of 15 feed tree species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.202. 95% confidence ellipses are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

96

Figure 5.5 (b) Non-metric MDS for the floristic similarity of interior and exterior plots across nine rehabilitation pits, based on the number of live stems of 15 feed tree species (here interior pits are indicated by the solid line and square points, while exterior pits are indicated by the broken lines and cross points). The similarity measure was Bray-Curtis and the stress was 0.202. 95% confidence ellipses are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

97

Figure 6.1 Features of proteaceous feeding used to recognise black cockatoo feed residues: (a) branches bitten off displaying the characteristic 45 ° angled ‘snipped’ tip; (b) broken branches B. squarrosa; (c) broken branches of H. prostrata and (d) H. undulata, and (e) cracked seedpods of H. cyclocarpa.

106

Figure 6.2 Marri feed residues in the plot (a). Marri fruit husks displaying: (b) characteristic bite profile of Baudin’s Cockatoo and (c) shorn pattern of FRTBC. (d) – (f) illustrates how the condition of the residue, particularly its colour, can also indicate its age (i.e. the approximate time when the feeding event occurred) where - (d) 3 day old; (e) 3 month and (f) 7 month old Marri residue from Baudin’s Cockatoo ((d) – (f) are courtesy of Tony Kirkby, Western Australian Museum).

108

Figure 6.3 The number of plots containing residues of particular food plant species summed across all pits for each of three sampling sessions (a) – (c). Legend: Hu Hakea undulata, D = Banksia spp., M = Corymbia calophylla, DS = B. sessilis, HC = H. cyclocarpa, HT = H. trifurcata, HP = H. prostrata, HR = H. ruscifolia, HA = H. amplexicircus, DSQ = Banksia squarrosa, H = Hakea sp., BG = B. grandis, J = Eucalyptus marginata, HV = H. varia.

114

Figure 6.4 The number of plots containing residues of particular food plant species observed in each pit across all three sampling sessions. The legend is given in Figure 6.3.

115

  xviii  

Figure 6.5 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the first sampling occasion, based on the feeding residue counts for 15 feed tree species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.07241. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units.

116

Figure 6.5 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the first sampling occasion, based on the feeding residue counts for 15 feed tree species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.07421. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

117

Figure 6.6 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the second sampling occasion, based on the feeding residue counts for 15 feed plant species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.08648. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

118

Figure 6.6 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the second sampling occasion, based on the feeding residue counts for 15 feed plant species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.08648. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

119

Figure 6.7 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the third sampling occasion, based on the feeding residue counts for 15 feed plant species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.1181. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

120

  xix  

Figure 6.7 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the third sampling occasion, based on the feeding residue counts for 15 feed plant species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.1181. Minimum convex hulls, the smallest polygon incorporating all data points,are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

121

Figure 6.8 Non-metric MDS for the similarity in vegetation structure and floristics between feed and non-feed plots across nine rehabilitation pits on the first sampling occasion. The final solution is three-dimensional, shown with two separate two-dimensional plots. The unfilled square points indicate non-feed plots, while filled squares indicate feed plots. The similarity measure was Bray-Curtis and the stress was 0.1667. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for feed and non-feed plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another.

126

Figure 7.1 (a) PVC ‘cockatube’ ANH, and (b) wooden box-type ANH installed at the NBG and LEA sites. 134

Figure 7.2 ANH establishment on-site: (a) EWP used for installation; (b) cockatube on a Marri tree, and (c) wooden box-type ANH on a Jarrah tree. 139

Figure 7.3 ANH fell when a dead stag fell; ANH was subsequently relocated onto another tree. 145

Figure 9.1 Spatial distribution of mean annual and seasonal historical rainfall (1975 – 2007) in the southwest of Western Australia, including the Jarrah Forest Bioregion (taken from Charles et al. 2010, p. 83).

180

Figure 9.2 Spatial distribution of trends in mean annual and seasonal historical rainfall (1975 – 2007) in the southwest of Western Australia, including the Jarrah Forest Bioregion. The scale shows mean rate of change in mm/y/y (taken from Charles et al. 2010, p. 83).

181

Figure 9.3 Location of FDPs (CS01 - CS13) installed around the northern and western perimeters of the RDAs at NBG in 2009 and 2010. Camera trapping occurred at CS03 and CS07 - CS11.

185

Figure 9.4 Examples of FDPs located around the perimeter of the F1 RDA. 186

Figure 9.5 Location of PWSs where drinking by black cockatoos was observed or detected between December 2007 and May 2011. 191

Figure 9.6 Examples of PWSs at the NBG site: (a) a natural swamp, and (b) - (c) man-made sumps. 192

Figure 9.7 Black cockatoo drinking episodes: (a) flock of Carnaby’s Cockatoos at a man-made sump; (b) flock of FRTBC at a WSR; (c) Carnaby’s Cockatoo at a natural puddle; (d) motion-triggered camera image of a black cockatoo at a man-made sump; (e) motion-triggered camera images of a group of FRTBC, and (f) Carnaby’s Cockatoo drinking at two of the more recently installed FDPs ((e) – (f) are courtesy of NBG).

203 - 204

  xx  

Thesis List of Tables Table 1.1 Terms used to describe landscape fragmentation (following Samways 1995). 2

Table 3.1 Habitat types identified within study area. 40

Table 3.2 Results of ANOVA for the data in Figure 3.4 after logarithmic transformation. Species refers to black cockatoos and season refers to: summer, autumn, winter and spring. Tukey’s Honest Significant Difference (HSD) post hoc tests for unequal sample sizes to compare differences between flock sizes of black cockatoos. Significant differences are indicated with a *. Tukey’s HSD post hoc tests for unequal sample sizes to compare differences between flock sizes of black cockatoos across seasons. Significant differences are indicated with a *.

46

Table 3.3 The numbers of encounters with black cockatoos in different habitat types at NBG and LEA combined. Percentages are shown in parentheses. The Shannon diversity and evenness for the data are shown, with 95% confidence limits determined by bootstrapping.

47

Table 3.4 (a) Inventory of (a) canopy-forming tree species and (b) proteaceous shrubs that are food plant species used by black cockatoos at NBG and surrounds. The first column gives the common name and scientific name for the food plant species. The second column indicates whether a behavioural observation was made of black cockatoos feeding on this food plant within the study area. The habitat type(s) in which the behavioural observation(s) occurred is also indicated. The third column indicates whether feeding residues for the food plant were observed. The habitat type(s) in which the feeding residues occurred is also indicated, as if the form of residue observed. The middle three columns indicate which black cockatoo species we observed feeding on the food plant (during a behavioural observation); shading indicates that an observation occurred for that species. The last three columns indicate the habitat type(s) feeding on the food plant occurred, based on behavioural observations and feeding residues; shading indicates that an observation occurred for that species Codes: NF = Native Forest; RV = Mine-site Rehabilitation; PP = Pine Plantation; fc = fruit capsule; br = branch; fl = flower; sd = seed; fs = flower spike; sc = seed casing; co = cone; ? = black cockatoo species could not be determined from residue.

48

  xxi  

Table 3.4 (b) Inventory of proteaceous shrub species that are food plant species that are food plant species used by black cockatoos at NBG and surrounds. The first column gives the common name and scientific name for the food plant species. The second column indicates whether a behavioural observation was made of black cockatoos feeding on this food plant within the study area. The habitat type(s) in which the behavioural observation(s) occurred is also indicated. The third column indicates whether feeding residues for the food plant were observed. The habitat type(s) in which the feeding residues occurred is also indicated, as if the form of residue observed. The middle three columns indicate which black cockatoo species we observed feeding on the food plant (during a behavioural observation); shading indicates that an observation occurred for that species. The last three columns indicate the habitat type(s) feeding on the food plant occurred, based on behavioural observations and feeding residues; shading indicates that an observation occurred for that species Codes: NF = Native Forest; RV = Mine-site Rehabilitation; PP = Pine Plantation; fc = fruit capsule; br = branch; fl = flower; sd = seed; fs = flower spike; sc = seed casing; co = cone; ? = black cockatoo species could not be determined from residue.

49 – 50

Table 3.5 Plants where black cockatoos were observed feeding at NBG and LEA combined. Percentages are shown in parentheses. The foods were flowers and seeds in most cases, but lifting bark to search for grubs is entered separately.

51

Table 3.6 The number of detections of three species of black cockatoo at sites sampled for the GDS within NBG tenements and the LEA in each year over the duration of the study. The number of samples in each year is given in parentheses in the year row. The chi-squared tests the null hypothesis of equal numbers of detections in each year against the alternative that detections varied with year. In calculating the expected values for the chi-squared tests, equal numbers of detections were assumed in each year and those values then adjusted for the sampling intensity. The expected values are given next to the observed in parentheses. For example, the expected number of detections of Carnaby’s Cockatoos at NBG in 2007 was calculated as 58 (number of samples in 2007)/904 (sum of all samples) x 21 (sum of all Carnaby’s Cockatoo detections over the study period).

53

Table 3.7 The number of detections of three species of black cockatoo at sites sampled for the GDS within NBG tenements and the LEA in each season over the duration of the study. The number of samples in each season is given in parentheses in the top row. The chi-squared tests the null hypothesis of equal numbers of detections in each season against the alternative that detections varied with season. In calculating the expected values for the chi-squared tests, equal numbers of detections were assumed in each season and those values then adjusted for the sampling intensity. The expected values are given next to the observed in parentheses.

55

Table 4.1 Results of post-felling inspections of trees identified as having potentially-suitable hollows during ground-based surveys (GBS). Percentages in column 1 are relative to the overall number of trees identified (i.e. n = 149); percentages in columns 2 relative to column 1 (i.e. species-specific totals for trees identified); and percentages in columns 4 and 5 relative to column 3 (i.e. species-specific totals for trees with hollows intact after felling).

69

Table 5.1 Year of establishment of the rehabilitation pits studied. 82

Table 5.2 Structural, floristic and phenological variables measured on each plot. 84

Table 5.3 The two-way ANOVAs for the five structural variables plotted in Figure 5.3. The factors are Pit (1 – 9) and plot location (interior/exterior). The significance level for main effects and interactions was set at 0.01 after Bonferroni correction to

92

  xxii  

allow for the multiple tests. -Mean non-canopy height (the dependent variable was log-transformed before analysis) -Mean live stem density (the dependent variable was log-transformed before analysis) -Mean dead stem density (the dependent variable was log-transformed before analysis) -Mean canopy height -Mean canopy cover Table 5.4 Results of SIMPER for the percentage differences in structural variables between (a) pits and (b) interior and exterior plots. Note that the mean scores are based on variables range-standardised to a 0 – 1 scale. Cumulative % (column 2) is calculated by dividing the sum of the relevant contributions by the total contribution.

95

Table 5.5 Results of SIMPER for the percentage differences in floristic variables (live stem densities of indicated species) between pits. Note that the mean scores are based on the live stem densities of each plant species. This is the same scale for each species, so no standardisation has been done. A dummy variable with a score of 1 for each pit was included (see text). Cumulative % (column 2) is calculated by dividing the sum of the relevant contributions by the total contribution.

98

Table 5.6 Characteristics of desired states and transitions suggested for Alcoa sites – from Grant (2006), 99

Table 6.1 Number of occasions that one of the three black cockatoo species was detected, either visually or acoustically, during systematic surveys of 23 rehabilitated mine pits over 22 months of sampling.

111

Table 6.2 Results of SIMPER for the differences in feeding residues between pits on each of three sampling occasions: (a) winter 2009 (b) summer 2010, and (c) winter 2010. Note that the mean scores are based on variables range-standardised to a 0 – 1 scale.

122 – 124

Table 6.3 The total number of Marri trees where feeding residues were detected (i.e. ‘feed trees’) for: (a) different sampling occasions and (b) different rehabilitated mine pits. The number of trees where those residues could be attributed to Baudin’s Cockatoos or FRTBC is also shown. Some trees contained residues from both species, so the total number of trees with feeding residues present is less than the sum value of feed trees for the two species.

125

Table 7.1 Details of placement of all ANH used at the NBG and LEA sites and surroundings. 136 – 137

Table 8.1 Titles and organisational affiliations appear as they were at the time the survey was undertaken. 155

Table 8.2 History of ANH installation by participants between 1996 and 2009. *- ANHs were provided by LCJ-S. 158

Table 8.3 Design, placement and use by black cockatoos of ANHs placed by survey participants. 159 – 162

Table 8.4 Installation, maintenance, monitoring and costs associated with ANH implementation. 165 – 167

Table 8.5 Hazards to black cockatoos and hazards to longevity of ANHs. 169 – 171 Table 9.1 Protocol for artificial water points (FDPs) at NBG. 187 Table 9.2 Records of black cockatoo mortalities associated with the RDAs at NBG. 189 Table 9.3 Summary of DES wildlife monitoring survey effort between April 2008 and January 2010 for the F1 RDA, TDZ and the alternative water bodies, the F3 RDA, the R4 RDA and the D1 WSR (from Smith et al. 2010, p. 32).

197

Table 9.4 Observations of black cockatoos drinking at water sources at NBG and 199 - 201

  xxiii  

characteristics of water sources. Table 10.1 Summary table comparing all three black cockatoos for each of the key social or ecological parameters measured. 224 - 225

Table 10.2 Specific recommendations for black cockatoo conservation from Cale (2003) and Chapman (2008). 230 - 231

Table 10.3 Summarised list of global causes of decline and examples. 239 Table 10.4 List of natural and life history traits/parameters or strategies of psittacines that make them susceptible to threats. 240

Table 10.5 Summary of the list of conservation measures for psittacines globally. 242 - 248

  xxiv  

Acronyms and Abbreviations

ALCOA Aluminium Company of America ANH(s) Artificial Nest Hollow(s) ANOSIM Analysis of similarity ANOVA Analysis of Variance ‘CAR’ principle ‘Comprehensive, Adequate and Representative’ Principle CL Confidence Limits DEC Department of Environment and Conservation DES Donato Environmental Services DSEWPaC Department of Sustainability, Environment, Water, Population and Communities EPBC Environment Protection and Biodiversity Conservation Act 1999 (Commonwealth) EPA Environmental Protection Authority EWP Elevated Work Platform FDP(s) Faunal Drinking Point(s) FRTBC Forest Red-tailed Black Cockatoo GDS General Distribution Survey HSD Honest Significant Difference IUCN International Union for Conservation of Nature LEA Land-Exchange Area NBG Newmont Boddington Gold NGO(s) Non-Government Organisation(s) nMDS non-Metric Multi-Dimensional Scaling PAST PAlaeontological STatistics PVC Polyvinylchloride PWS(s) Permanent Water Source(s) RDA(s) Residue Disposal Area(s) RFA Regional Forest Agreement SEM Standard Error of the Mean SIMPER Similarity Percentage TDS Total Dissolved Solids TDZ Tailings Discharge Zone TSSC Threatened Species Scientific Committee WA Western Australia WCA Wildlife Conservation Act 1950 (Western Australia) WSR(s) Water Supply Reservoir(s)

  1  

Chapter 1 General Introduction

1.1 Overview of Chapter

As a result of extensive native vegetation loss, much of the biodiversity of southwestern

Australia is threatened or in decline. Key avifauna under threat include three endemic

black cockatoos that are listed as threatened under State and Commonwealth legislation:

Forest Red-tailed Black Cockatoos (Calyptorhynchus banksii naso) (a subspecies)

(FRTBC), Baudin’s Cockatoos (Calyptorhynchus baudinii), and Carnaby’s Cockatoos

(Calyptorhynchus latirostris). A conservation strategy based upon reserves alone is,

however, unlikely to protect these black cockatoos in perpetuity, making it necessary to

protect biodiversity within other landscapes, particularly those devoted to natural

resource production.

This thesis considers how we can conserve black cockatoos within landscapes devoted

to mining. It presents the findings of field research undertaken at the Newmont

Boddington Gold (NBG) mine, a mine-site situated approximately 130 km southeast of

Perth, Western Australia and located along eastern margin of the Jarrah-Marri forest,

the region’s largest contiguous forest habitat.

This chapter establishes the conceptual framework for the thesis. Firstly, the chapter

introduces the concept of the ‘production landscape’ as a means of describing a

landscape devoted to (and largely shaped by) the extraction of natural resources, and

then reviews the role that production landscapes can play in biological conservation.

Secondly, the chapter considers historical and current natural resource production in

Jarrah-Marri forest and its implication for biological conservation. Thirdly, the chapter

formulates the salient features for the conservation ‘problem’ this thesis addresses –

how to conserve black cockatoos within a landscape devoted to mineral resource

production. The chapter concludes with a description of the main aims of this study and

an outline of the structure of the thesis as a whole.

  2  

1.2 Integrating Natural Resource Production and Nature Conservation

Mankind has geometrized the landscape. Samways (1995, p. 104)

Land clearing and other habitat modifications have fragmented once continuous natural

landscapes into isolated patches of habitat, so that an estimated 39 – 50% of the land

area of the planet has been modified or degraded by human activity (Vitousek et al.

1997). These changes present a major threat to the world’s biodiversity (Gibbons and

Lindenmayer 2007). Landscape ecologists have developed several terms to describe the

consequences of landscape fragmentation (Table 1.1).

Table 1.1 Terms used to describe landscape fragmentation (following Samways 1995).

Landscape A grouping of interacting ecosystems that repeat across a heterogeneous land area

Matrix The most extensive and most connected ecosystem within a particular landscape

Patch A landscape element surrounded by the matrix but differing substantially in structure

Corridor Similar to a patch, but differs in being long and linear in shape; a corridor may connect otherwise isolated patches

Connectivity The extent to which particular landscape elements are connected to each other

The term ‘production landscape’ refers to a landscape in which human land uses

predominate and the matrix is therefore dedicated largely or wholly to natural resource

production (McIntyre and Hobbs 2000; Fischer et al. 2006; Hobbs et al. 2006; Chazdon

et al. 2009a,b). Land uses that may create production landscapes include agriculture,

pastoralism, mining, tree plantations, timber harvesting, urban development, recreation

and tourism, and water catchments (Calver and Wardell-Johnson 2004). In forest areas,

landscape-scale disturbances range from selective forestry practices, which retain some

of the original forest structure (Calver and Dell 1998; Loyn 2000; Lindenmayer and

Franklin 2002), to the complete conversion of native forest to agriculture, plantations,

or mining use (Burns et al. 2000; Hobbs and Saunders 2000). However, patches of

native vegetation may remain and may be connected to some degree. Linear

connections, or corridors, sometimes follow natural features such as watercourses or

human constructions such as road verges.

  3  

The fragmentation and modification of natural matrices for human land uses is a

significant driver of biodiversity decline in Australia (Recher 2002; Gibbons and

Lindenmayer 2007; Garnett et al. 2011). Effects on biodiversity may arise through the

loss of native vegetation and associated effects on the feeding, breeding, movement and

dispersal of native species (Calver et al. 1998; Marzluff and Ewing 2001; Lindenmayer

and Fischer 2006). Patches of remnant habitat within production landscapes may be

subjected to pressures such as recreational use, salinisation, altered hydrology,

introduced species, weeds, disease pathogens, altered fire regimes, disruption by roads

and other infrastructure, illegal culling or harvesting and climate change (Saunders et al.

1991; Bamford et al. 2004; Wardell-Johnson et al. 2004). These pressures may lead to

the continued loss of native plant and animal species long after the initial modification

of the landscape has occurred (Hobbs 1993).

Biodiversity conservation often focuses on establishing reserves where production

activities are restricted or prohibited (Lindenmayer and Franklin 2002; Fischer et al.

2006). Reserves can support the conservation of susceptible species and ecosystems

from potential or known impacts of disturbance (Pressey and Taffs 2001). They can also

serve as control zones to study and compare against the human exploitation impacts in

production areas (Chazdon et al. 2009b; Lindenmayer 2009), and also allow natural

evolutionary processes to continue (Lindenmayer and Franklin 2002).

Where a reserve system is designed to conserve the full range of biodiversity for a

region, the CAR principle – Comprehensive, Adequate and Representative – is often

adopted (Dickson et al. 1997; Bryan 2002; Fitzsimons and Robertson 2005). A

comprehensive reserve system encompasses all aspects of biodiversity from genes

through species to communities and ecosystems. To be adequate, reserve establishment

should set aside and maintain sufficiently large tracts of land to serve as feeding and

breeding habitat to maintain viable populations (Lindenmayer and Possingham 1998).

To be representative, species, communities and ecosystems should be included from

across their geographic ranges. Lindenmayer and Franklin (2002) propose adding a

second R for replication, arguing that multiple reserves are often needed to avert the

risk of a catastrophe eliminating a single reserve (e.g. Danks 1994, 1998).

However, many reserve systems are too small, too poorly connected, or too heavily-

impacted to sustain viable fauna and flora populations and natural ecological processes

  4  

over the long-term (Rundle 1996; Bryan 2002; Fitzsimons and Robertson 2005). Thus,

conservation biologists have increasingly attempted to use a mixed approach involving

both conservation within reserves and conservation within production landscapes

(Grigg et al. 1995; Hale and Lamb 1997; Lindenmayer and Franklin 2002; Chazdon et

al. 2009a,b). This combines a CAR-type reserve system with sympathetic management

of agricultural, forestry and mining production landscapes within forest and woodland

habitats.

As production landscapes dominate many bioregions, the long-term persistence of a

species may reflect in part its ability to utilise novel or modified habitat types and

resources (Saunders and Ingram 1995; Fahrig 2001; Hobbs et al. 2006). A combination

of natural and modified landscapes can provide species with a mosaic of habitat patches

differing in size, quality, and other factors (Saunders et al. 1991; Lindenmayer et al.

2002; Manning et al. 2004; Recher 2004). Thus, conservation efforts may need to

consider the range of natural and anthropogenic habitat types available within a given

landscape (Recher 2004), as well as information about natural ecosystem structure and

disturbance regimes. Three examples will illustrate these ideas in action.

Spotted Owls in North American Forests (Forestry):

Three subspecies of spotted owls - the California Spotted Owl (Strix occidentalis ssp.

occidentalis), Northern Spotted Owl (ssp. caurina) and the Mexican Spotted Owl (ssp.

lucida) have varying autecologies and occur in distinctively different biotic regions of

western forests of North America (Lindenmayer and Franklin 2002). Over the past

century, habitat loss and fragmentation as a result of timber harvesting operations (e.g.

coniferous species) has been a major contributing factor threatening the decline of all

three species. The conservation strategies adopted varied between the three species

(Lindenmayer and Franklin 2002). However, they all recognised, at both a stand and

regional level, the importance of managing the matrix (i.e. landscapes comprising

mosaics of different vegetation types) for maintaining populations of forest species,

along with the establishment and maintenance of ecological reserve systems (i.e. a

multi-scale landscape strategy). These matrix-based strategies looked at management of

the entire forest matrix that comprised reserves, old-growth forests, young regrowth

forests, national forests and harvested lands across multiple spatial scales (Lindenmayer

and Franklin 2002).

  5  

Briefly summarising here, the strategies included aspects such as improving owl habitat

conditions and managing existing owl habitat areas in the forest matrix across the entire

range of each species (i.e. meso-scale reserves), protecting and restoring habitat

potentially capable of supporting populations, thus increasing habitat over time (e.g.

large areas of old forest reserves), actively managing the general forest areas and

facilitating connectivity between habitat areas within the managed landscape. This

involved the retention of snags and large (diameter) trees on harvested lands in various

sized aggregates, implementing uneven-aged harvest systems, longer rotations and

monitoring programs (Lindenmayer and Franklin 2002). These studies on spotted owls

have shown that a flexible and adaptive approach to matrix management is required,

considering the different habitat requirements of individual species/subspecies. The

conservation strategies developed also led to the maintenance of populations of other

species, as well as other ecosystem processes (Lindenmayer and Franklin 2002).

Leadbeater's Possum in Southeastern Australia (Forestry and Water Catchment):

Leadbeater’s Possum (Gymnobelideus leadbeateri) is an arboreal marsupial that

inhabits the Mountain Ash forests of Victoria, southeastern Australia. Similar to the

spotted owl example; widespread logging by clearcutting is a major threatening process

for the species, along with land clearing for water catchments (Lindenmayer and

Franklin 2002). Mountain Ash is a valuable hardwood species for timber production

and pulpwood in eastern Australia, leading to the development of a large forestry

industry. Clearcutting removes large areas of forest, leaving only a few live standing

trees, and is usually followed by high-intensity slash fires to burn logging debris, reduce

fuel loads and create nutrient-rich ash ground for new stand regeneration (Lindenmayer

and Franklin 2002). Forestry impacts on the long-term conservation of the species have

been an issue for over four decades, however recent commitments to multiple forest use

and native wildlife conservation in timber production forests by the Victorian

government shed some hope on the plight of the possum (Lindenmayer and Franklin

2002).

Studies have demonstrated that a comprehensive and integrated matrix-based forest

management plan across multiple spatial scales and ecological processes is needed to

achieve conservation of the species. This requires consideration of the species' habitat

requirements and metapopulation dynamics, connectivity of habitat, and ecosystem

  6  

processes such as logging, fire and altered climatic conditions (Lindenmayer and

Franklin 2002).

To summarise, these are (i) large ecological reserves – the Yarra Ranges National Park

contains three large water catchments, but is the still most extensive area of multi-aged

forest (including mature old-growth forests) in the Central Highlands that spans a wide

range of elevational gradients. It contributes significantly to Leadbeater’s Possum

conservation, but is still restricted as it is threatened by potential impacts from

catastrophic wildfires and climate change. This emphasises the need for off-reserve

conservation within the matrix landscape dedicated to timber harvesting (Lindenmayer

and Franklin 2002). (ii) Midspatial-scale strategies for long-term conservation such as

the establishment of wildlife corridors, riparian strips and protected areas (i.e. old-

growth forest) within the wood production matrix lands to develop suitable matrix

habitat. The expansion of old-growth area through restoration and the withdrawing of

these regrowth forests from timber harvesting to be protected as multiple reserves and

allow them to develop into suitable possum habitat (especially those areas adjacent to

old-growth patches). Another strategy is a general forest management zoning approach

where high conservation value sites are designated special management zones for

conservation, and the rest of the forests where intensive forestry can be practiced are

general management zones (Lindenmayer and Franklin 2002). And (iii) fine-scale,

stand-level and matrix-based management strategies within the timber production

landscape that require modifications to existing silvicultural systems. This includes the

structural retention of snags and stands of large living cavity-trees in harvested areas,

protection of understorey vegetation and large logs, creation of multi-aged stands, and

altered cutting regimes on logged sites within the matrix. By resembling natural

disturbance regimes, logging and regeneration methods could promote structural

complexity in harvested stands to enhance wildlife habitat values (Lindenmayer and

Franklin 2002).

Finally, all these should be reinforced by the implementation of monitoring programs to

assess the effectiveness of the multiple management strategies in terms of conservation

outcomes for Leadbeater’s Possum. At the moment, all but point (iii) have been

implemented or addressed (Lindenmayer and Franklin 2002).

  7  

Superb Parrot (Agriculture):

The Superb Parrot (Polytelis swainsonii) is a vulnerable species restricted to

southeastern Australia. In the Murray-Murrumbidgee valleys, its breeding riparian river

red gum habitat is spatially differentiated from its foraging box-gum habitat and may be

joined by linear corridors of vegetation (i.e. along roads, fence-lines and water courses),

but in the scattered box-gum-Blakely’s Red Gum woodlands of the South-West Slopes

of New South Wales, its breeding and foraging habitats coincide (Manning et al. 2004;

Manning et al. 2006, 2007). A major threatening process to the survival of the species is

the ongoing loss of breeding habitat in the form of hollow-bearing trees as a result of

land clearing for agricultural landscapes and decline of the remaining standing trees that

provide nest sites (Manning et al. 2004). Other related threats include the clearing and

modification of feeding grounds, poor and/or cessation of regeneration of nest trees and

feeding resources, car-strikes while feeding on spilt grain along the roadside and nest

site competition. This is also associated with other degrading processes such as

grazing/pastoralism and rural dieback (Manning et al. 2004; Manning et al. 2007).

The imposition of agriculture on natural systems generally leads to habitats within the

agricultural matrix being modified but not completely destroyed. This leads to

variegated landscapes where human land uses occur together with pre-existing

vegetation in the form of isolated and scattered trees or patches. Protected areas (e.g.

nature reserves, remnant patches and corridors) will remain as important conservation

strategies important (Manning et al. 2004; Manning et al. 2006).

However, the Superb Parrot persists in the adjacent agricultural matrix of highly

productive, privately owned land in many parts of its range, and thus a reserve-only

strategy may not be sufficient for its survival. Conservation outcomes may be greatly

enhanced if remnant patches and corridors were intermixed into a matrix of self-

sustaining trees (Manning et al. 2004; Manning et al. 2006). This emphasises the

importance of integrating conservation and production in the same landscape, as these

modified ecosystems can continue to have considerable conservation value or benefits

through the long-term sustainability of these variegated landscapes (Manning et al.

2004; Manning et al. 2006). Studies have found an association with nesting and

increasing tree diameter, as well as a preference for nesting in dead trees and Blakely’s

Red Gum. Specific issues noted on the South-West Slopes (particularly) include the

scattered and isolated nature of hollow-bearing trees in the agricultural matrix, the

  8  

death/senescence of many existing trees and lack of or limited tree regeneration around

these trees at a landscape-scale (Manning et al. 2004). The large size and poor health of

nest trees require urgent habitat management to provide alterative nest sites and

regenerate trees at a landscape scale across the whole agricultural matrix. This includes

combined strategies such as tree protection (i.e. existing hollow-bearing trees, and

smaller mature trees with or without hollows as they have conservation value as

potential nest trees), as well as tree regeneration (i.e. established seedlings, tree

planting) (Manning et al. 2004). This is because all trees in these age classes can

contribute to Superb Parrot conservation through continuity of hollow resource, and for

the Superb Parrot – regenerating or retaining scattered trees across the whole landscape

of the agricultural matrix appear to have the greatest potential for restoration (Manning

et al. 2004).

To date, rural restoration focuses on tree planting. This has issues such as the

patch/linear planting configurations not being able to replace the current scattered

pattern of habitat used by the parrots in the landscape, and the uncertainty that planted

trees will have the same hollow-forming capabilities as existing native trees since

hollow formation varies with soil type and topography (Manning et al. 2004). However,

natural revegetation of trees (and understorey vegetation) may be a better (and cheaper)

alternative as it has greater potential over larger scales. It can be achieved post-grazing

through fencing and can be planned and managed in grazing regimes (e.g. cell or

rotational grazing) by using principles of restoration triage to determine appropriate

type of restoration action for particular sections of the landscape (e.g. temporary fencing

and rotational grazing for natural regeneration, and permanent fencing and planting in

cultivated paddocks) (Manning et al. 2004). Another conservation strategy that may be

viable as an alternative for nesting while replacement trees are growing (given the slow

process of hollow formation) is the implementation of nest-boxes. Overall, this requires

long-term planning and coordination on the scale of landscapes (Manning et al. 2004).

All the issues developed in these examples can be seen in the complex management of

the Jarrah-Marri forest of southwestern Australia.

  9  

1.3 Human Activity and Biodiversity Conservation in the Jarrah-Marri Forest

1.3.1 The Jarrah Forest Bioregion

Native forests in southwestern Western Australia cover c. 4.25 million ha and occur

across three biogeographic regions: Swan Coastal Plain, Jarrah Forest and Warren

(Department of Sustainability, Environment, Water, Population and Communities 2011)

(Figure 1.1). They are bounded to the east by the Wheatbelt (the bioregions Avon

Wheatbelt, Mallee and Esperance Sandplains), which has been cleared extensively for

agriculture (Jarvis 1981).

Figure 1.1 The forested bioregions of southwestern Australia: Swan Coastal Plain, Jarrah Forest and Warren (adapted from Calver and Wardell-Johnson 2004).

This thesis is concerned with the Jarrah Forest Bioregion, named for the commercially

important hardwood (Eucalyptus marginata). Marri (Corymbia calophylla) is another

widespread tree in the region and, although commercially less important than Jarrah, it

has a major ecological role (Paap 2006). The region is therefore aptly described as the

Jarrah-Marri forest. The whole bioregion is heterogeneous, with diverse vegetation

types shaped by extended geological and climatic histories (Havell 2000; Wardell-

  10  

Johnson et al. 2004). Although its terrestrial vertebrate fauna is less biodiverse than

forests in eastern Australia (Nichols and Muir 1989), the mixed Jarrah-Marri forest is

nevertheless vital for conserving threatened mammals and birds (see Maxwell et al.

1996 for mammals; Garnett et al. 2011 for birds). Its herpetofauna also differs notably

from that of the neighbouring Wheatbelt (Chapman and Dell 1985). For example, the

Geocrinia rosea complex of four allopatric frog species is endemic (Roberts et al. 1990;

Wardell-Johnson and Roberts 1991). Some amphibians are now largely restricted to the

Jarrah-Marri forest because of land clearing in neighbouring environments (Calver and

Wardell-Johnston 2004).

1.3.2 Timber Production

Human activities since first European settlement in 1826 have greatly altered the

landscape of the southwest, including the Jarrah-Marri forest. Much of the Jarrah-Marri

forest was allocated as State Forest, which led to its protection from clearing for

agricultural purposes (Williamson 2005; Rundle 2006). It has also meant that logging

and associated activities (e.g. road building, fuel reduction burns and other silvicultural

practices such as thinning) are historically the most extensive anthropogenic

disturbances in areas designated as State Forest (Heberle 1997; Calver and Wardell-

Johnson 2004). These disturbances have been substantial. For example, by 1920

‘…nearly one million acres of the jarrah forest were cut over for the removal of 750

million cubic feet of logs, causing a reduction of almost 50% in the forest canopy’

(Wallace 1965, p. 35). Under the current Forest Management Plan, approximately 800

000 ha of the Jarrah-Marri forest is available for logging (Conservation Commission of

Western Australia 2003).

1.3.3 Mining

In comparison with forestry, mining is a more recent land use in the Jarrah-Marri forest,

with large-scale mining operations commencing only within the last fifty years (Ward et

al. 1990). The Darling Plateau, on which the Jarrah-Marri forest lies, is valued for its

rich endowment of mineral resources (Dell et al. 1989; RFA 1998; Wali 1999). The first

mines established were for coal mining around Collie in the late 1890s (Bartle and

Slessar 1989). Alcoa World Alumina established the first bauxite mine in the Jarrah-

Marri forest at Jarrahdale in 1963. Alcoa has since expanded its operations within the

western Jarrah-Marri forest (Gardner and Bell 2007). In 1984 a second bauxite mine

operation operated by Worsley Alumina opened at Mt Saddleback (near Boddington) in

  11  

the eastern Jarrah-Marri forest (Bartle and Slessar 1989; Worsley Alumina Pty. Ltd.

1985).

Gold mining began in the Jarrah-Marri forest in the late 1980s, with mining operations

at Boddington Gold Mine and at the adjacent Hedges Gold Mine (Worsley Alumina

Pty. Ltd. 1999). These two mine-sites combined in 1998, but ceased operation in 2001.

The old Boddington Gold Mine site is now the site of Newmont Boddington Gold

(Newmont Mining Corporation 2007).

Individual mining operations have intense but localised impacts within the Jarrah-Marri

forest. Various State Agreement Acts, along with other State legislation, leave much of

the Jarrah-Marri forest subject to mining or exploration leases, so mining operations can

potentially expand over much larger areas (RFA 1998). To date, 45% of State Forest

and timber reserves is covered by State Agreements or approved mining leases and a

further 34% of these land tenures is covered by pending mining leases or

granted/pending exploration licences. Approximately 1 000 ha per year of native forest

are cleared for mining, principally bauxite, coal and gold (Conservation Commission of

Western Australia 2012).

The two main mining companies are Alcoa World Alumina and Worsley Alumina.

Between 1966 and 2010 Alcoa has cleared 18 200 ha of native forest, at a rate of

approximately 630 ha per year (Koch 2007; Legislative Council of Western Australia

2010). Worsley has cleared 2 712 ha between 1980 and 2009, at a rate of 240 ha per

year (Legislative Council of Western Australia 2010). The cumulative effects of mining

by these and other companies are estimated to ultimately lead to the clearing of 83 000

ha (7% of State Forest) and the fragmentation of 337 000 ha (28% of State Forest)

(Conservation Commission of Western Australia 2012).

Environmental impacts from mining operations include: clearing of native vegetation to

allow for extraction of ore and construction of infrastructure such as processing

facilities and haul roads (Schofield and Bartle 1984); potential spread of plant

pathogens such as ‘dieback’ Phytophthora cinnamomi (Bartle and Slessar 1989; Dell et

al. 1989); exposure of acid-sulphate soils that may contaminate drainage water and

degrade essential organic matter in soils (a risk in coal mining, Bartle and Riches 1978);

hydrologic changes such as soil erosion, stream disturbance and increases in turbidity

  12  

and salinity and increased surface-runoff into streams following clearing (Bartle and

Slessar 1989); and, in the case of gold mining, environmental contamination from

cyanide compounds (Finnie et al. 2009). In recognition of these problems, extensive

research has taken place on hygiene measures to prevent pathogen spread (Underwood

and Murch 1984; Garkaklis 2004) and on restoration techniques for mining areas (Ward

et al. 1990; Grant et al. 2007; Koch 2007; Koch and Hobbs 2007).

1.3.4 Reserve System

Until recently, the creation of a reserve system within the Jarrah-Marri forest and

neighbouring bioregions had been largely ad hoc, with management driven more by

lack of commercial interest in certain lands or by the presence of specific

tourism/recreation demands rather than by specific biodiversity protection goals

(Rundle 1996; Moore 1993). Just over 40% of the Jarrah-Marri forest is currently

dedicated to biodiversity conservation, with the remainder allocated to land uses

ranging from water catchment, State Forest and timber reserves, private native forest,

(non-native) tree plantation, mining (bauxite, gold, coal, and mineral sands), and human

settlement (Conservation Commission of Western Australia 2003). Detailed reservation

targets for specific forest ecosystems and old growth forest are listed in Appendices 9

and 10 of the Forest Management Plan 2004-2013 (Conservation Commission of

Western Australia 2003). There is general recognition that biodiversity conservation

goals will only be met if resource production is sympathetic to them.

1.4 Problem Formulation

Conservation biology is a problem-solving discipline (Meine et al. 2006). However, in

order to ‘solve’ a conservation problem, it is first necessary to ‘formulate’ the problem,

much as one would describe the features of a medical pathology before undertaking

tests to reach a clinical diagnosis (Caughley and Gunn 1996). Here, I present the key

features of the problem for this thesis – how to conserve black cockatoos within mining

landscapes. The features are presented in a sequence, as if comprising a series of logical

steps.

1: Three black cockatoos occur within the Jarrah-Marri forest: the Forest Red-tailed

Black Cockatoo Calyptorhynchus banksii naso (a subspecies) (FRTBC), Baudin’s

Cockatoo Calyptorhynchus baudinii and Carnaby’s Cockatoo Calyptorhynchus

  13  

latirostris (Johnstone and Storr 1998). All three are endemic to the southwest of

Western Australia. FRTBC and Baudin’s Cockatoos occur mainly in the forested areas

of the southwest, whereas Carnaby’s Cockatoos also range within the region’s

woodland and coastal heathland habitats.

2: All three black cockatoos have suffered substantial range contractions over the last

half-century and are considered to be in decline (Garnett et al. 2011). Using IUCN

(International Union for Conservation of Nature) criteria, Garnett et al. (2011) classified

FRTBC as vulnerable and Carnaby’s Cockatoo and Baudin’s Cockatoos as endangered.

All three species are listed as threatened under State and Commonwealth legislation.

3: Land clearing has reduced the total area of available habitat for all three black

cockatoos and recovery plans for the species have prioritised actions to protect breeding

and feeding habitat (Cale 2003; Chapman 2008). The Jarrah-Marri forest provides

breeding and feeding habitat for all three black cockatoos and, as the largest forest

habitat in southwestern Australia, therefore represents an important focus for

conservation efforts (Biggs et al. 2011).

4: Human activities differ in their effects on black cockatoo habitat. Fire management

may influence forest productivity and phenological patterns of Jarrah and Marri, which

may affect food availability for black cockatoos. Logging alters the structure and

composition of forests, and may change the mix of tree species and age classes present

(Recher 1996). Marri, in particular, has been selectively removed from Jarrah-Marri

forest stands to increase productivity of the more highly valued Jarrah (Calver and Dell

1998). Views differ on the influence of logging on hollow availability. Abbott (1998)

examined data from felled trees and a forest inventory undertaken in 1989 – 1991 and

concluded that there was a surplus of trees of sufficient size to have a suitable nest

hollow for FRTBC (and, by inference, for other black cockatoos as well). By contrast,

Johnstone et al. (2013a) suggest that Abbott’s calculations are overestimates because:

(i) they assume that Jarrah and Marri are equally suitable for nesting, whereas Johnstone

et al. (2013a) found that Marri was the predominant species used for nesting, and (ii)

inspection of putative hollows by climbing trees indicated that many ‘suitable’ hollows

identified from the ground were not useable.

  14  

5: Mining removes all native vegetation from an area. Mine-site rehabilitation can

replace a vegetation cover quickly and black cockatoos may begin feeding on

revegetation within eight years (Lee et al. 2010). However, little is known about

whether and how black cockatoos will use revegetation as a feeding habitat because its

structure and composition may be very different to mature native forest, at least during

early successional stages. The replacement of suitable breeding sites will only occur at a

scale of centuries, with the youngest suitable trees being 131 years old according to

Abbott and Whitford (2002) and 209 years old according to Johnstone et al. (2013a).

6: The conservation of fauna within mining landscapes requires research to characterise

the habitat needs of fauna species, information that can then guide efforts to protect

these habitat features from mining or to restore them after mining has ceased (Bartle

and Slessar 1989). This process is especially challenging at forested mine-sites because

mining removes all forest vegetation (Craig 2007; Webala et al. 2010, 2011). These

problems are not new, and this study draws on a rich theoretical and empirical

framework developed from previous applied research. Such research has been

undertaken in a diverse range of production landscapes including: mine-sites (e.g. Craig

et al. 2010); agricultural landscapes (e.g. Lambeck et al. 2005); tree plantations (e.g.

Foster et al. 2011; Pryke and Samways 2012); and native timber reserves (Gunarso and

Davie 2005).

7: Efforts to conserve black cockatoos within mining landscapes require information on:

(1) the habitat resources present (e.g. food, nest hollows, water); (2) the availability of

these resources within native forest areas and within habitats created by mining or

restored after mining ceases (e.g. mine-site rehabilitation, residue disposal areas); (3)

how black cockatoos use these resources (e.g. spatial and temporal patterns in feeding

activity); and (4) any risks presented by potential interactions between birds and mining

activities. This information can then be applied to identify appropriate actions for

protecting and restoring habitat and for minimising adverse interactions.

8: Black cockatoos present particular challenges as study subjects. For example, they

are highly mobile, with daily foraging ranges of several kilometres and (for some

populations) seasonal migratory patterns extending over hundreds of kilometres. Thus,

to obtain data on a scale relevant to these birds, it may be necessary to sample large

areas and to conduct surveys at different time periods. However, many of these

  15  

characteristics are shared with other terrestrial fauna, and thus the findings from this

study may be relevant to the conservation of other large mobile species within

production landscapes.

1.5 Study Aims and Thesis Structure

Chapter 2 describes the NBG study site and gives background information on the

ecology and distribution of the three black cockatoos.

Chapters 3 - 9 address the research aims of this research, which are to:

Describe the ecology of the three black cockatoos at NBG, particularly

patterns in group size, site occupancy, habitat use, and food plant use (Chapter

3);

Examine the effectiveness of ground-based hollow surveys, post-felling

inspections of hollows, and behavioural observations as approaches for

assessing black cockatoo breeding habitat in the Jarrah-Marri forest (Chapter

4);

Assess the successional stage of the rehabilitated mine pits and characterise

variation in the structure and floristics of pits, in order to identify features that

might influence the availability of food resources for black cockatoos

(Chapter 5)

Document feeding activity within rehabilitated mine pits and any associations

with structural or floristic features of the vegetation (Chapter 6);

Trial the implementation of artificial nest hollows to support breeding on-site

and compensate for the loss of natural hollows cleared by mining (Chapter 7);

Conduct a review of the use of artificial nest hollows for black cockatoos to

assess their value as a tool for mitigating natural hollow loss (Chapter 8);

Investigate the use of natural and artificial water sources at NBG and assess

the potential for interactions with residue disposal areas (Chapter 9).

The concluding chapter (Chapter 10) integrates the findings from earlier chapters and

synthesises management implications and conclusions.

16

Chapter 2 Description of Study Species and Study Area

This chapter presents background information on the ecology and conservation biology

of three black cockatoos occurring in the Jarrah-Marri forest of southwestern Australia:

the Forest Red-tailed Black Cockatoo (FRTBC) (Calyptorhynchus banksii naso),

Carnaby’s Cockatoo (Calyptorhynchus latirostris), and Baudin’s Cockatoo

(Calyptorhynchus baudinii). It also describes the regional and geographic setting for the

study, including the gold mining operations at the Newmont Boddington Gold (NBG)

mine-site.

2.1 Study Species

2.1.1 Overview

Black cockatoos belong to the Australian endemic genus Calyptorhynchus. Though

parts of the phylogeny and evolutionary history of black cockatoos have been

determined, the systematics of the Calyptorhynchus genus is unresolved, especially for

Carnaby’s Cockatoos and Baudin’s Cockatoos (White et al. 2011).

Three black cockatoos (FRTBC, Carnaby’s Cockatoos, and Baudin’s Cockatoos) are

restricted exclusively to southwestern Australia (Johnstone and Storr 1998). A fourth,

the Inland Forest Red-tailed Black Cockatoo (Calyptorhynchus banksii samueli) (a

subspecies), also occurs within the northern and eastern Wheatbelt region of the

southwest. To describe these birds, I used the nomenclature stipulated in the Wildlife

Conservation (Specially Protected Fauna) Notice 2012 (Western Australia), which

includes amendments to align nomenclature with names endorsed by the Western

Australian Museum.

Black cockatoos are among the largest Australian arboreal birds. They are highly

mobile and social species that display similar life history traits, although their ecologies

may differ significantly (Cameron 2007). The three black cockatoos occurring in the

southwest each have particular distributions, diets, and preferences for nesting trees

(though with moderate to substantial overlap) (Johnstone and Storr 1998). Breeding

17

pairs tend to maintain life-long, monogamous bonds and generation times are more than

ten years (Saunders 1982; Johnstone and Storr 1998; Garnett et al. 2011). Such

characteristics increase their vulnerability to anthropogenic threats, and over the last

century, the three ‘southwestern’ black cockatoos have experienced substantial declines

in population size and range (Saunders et al. 1985; Forshaw 2002; Cale 2003; Chapman

2008; Garnett et al. 2011).

The ecology of Carnaby’s Cockatoos has been studied intensively by Denis Saunders

for more than thirty years (e.g. Saunders 1974, 1980; Saunders and Ingram 1998;

Saunders and Dawson 2009). The ecologies of Baudin’s Cockatoos and FRTBC are less

well known, although long-term studies by the Western Australian Museum are

addressing gaps in knowledge (e.g. Johnstone and Kirkby 1999, 2008, 2009, 2010).

Early published records did not distinguish between Carnaby’s Cockatoos and Baudin’s

Cockatoos, which both have white bands on the tail (in contrast to the red bands on the

FRTBC) (Abbott 2008). The tall and densely forested habitats of the southwest also

make it difficult, expensive and time-consuming to obtain ecological data, especially in

relation to breeding (Chapman 2008; Johnstone and Kirkby 2008).

2.1.2 Conservation Status

As native fauna, all three black cockatoos are protected under the Western Australian

Wildlife Conservation Act 1950 (WCA). They are also listed as Schedule 1 fauna under

provisions of the act (species that are ‘rare or likely to become extinct and in need of

special protection’). The Western Australian Threatened Species Scientific Committee

(TSSC) classifies Carnaby’s Cockatoos and Baudin’s Cockatoos as Endangered

(‘considered to be facing a very high risk of extinction in the wild’) and FRTBC as

Vulnerable (‘considered to be facing a high risk of extinction in the wild’) (Western

Australian DEC 2012; Western Australian TSSC 2012).

The Commonwealth Environment Protection and Biodiversity Conservation Act 1999

(EPBC) lists Carnaby’s Cockatoos as endangered, and FRTBC and Baudin’s Cockatoos

as vulnerable. Recovery plans have been prepared for the three black cockatoos (Cale

2002; Chapman 2008). Listing as a threatened species under the EPBC makes the

conservation of these species a matter of ‘national environmental significance’ under

the act, and therefore any proposed actions that may have a ‘significant impact’ on these

species must be referred to the Commonwealth Department of Sustainability,

18

Environment, Water, Population and Communities (DSEWPaC 2011) for assessment.

Guidelines for the EPBC describe a ‘significant impact’ as an ‘impact which is

important, notable, or of consequence, having regard to its context or intensity. Whether

or not an action is likely to have a significant impact depends upon the sensitivity,

value, and quality of the environment, which is impacted, and upon the intensity,

duration, magnitude and geographic extent of the impacts’ (Commonwealth of Australia

2009a, p. 3).

Many Australian jurisdictions recognise IUCN (International Union for Conservation of

Nature) classification criteria as useful guidelines for conservation status. The most

recent IUCN classifications classify Carnaby’s Cockatoos as endangered (‘facing a very

high risk of extinction in the wild in the near future’) (IUCN 2011), based on evidence

of a rapid and continuing decline over the last half-century (BirdLife International

2010a). Baudin’s Cockatoos are also classified as endangered, based on evidence that

only 10% of the extant population is believed to be breeding and the population is

declining (BirdLife International 2010b). No classification is given for FRTBC, a

subspecies.

2.1.3 Species Description

Carnaby’s Cockatoos and Baudin’s Cockatoos possess dusky-black body plumages,

tails and crests (Johnstone and Storr 1998) (Figure 2.1). Adults are approximately 50 cm

long with a wingspan of approximately 1 m and broad tails that measure almost half

their body lengths. The tails are characterised by a broad sub-terminal white band

(Higgins 1999; Forshaw 2002; Connors and Connors 2005). Males have dark bills and

legs, irises encircled by pink eye-rings and faded-white ear coverts. Females are similar,

but have larger, more prominent white ear coverts, lighter-coloured bills and legs, and

grey eye-rings (Johnstone and Storr 1998; Forshaw 2002; Connors and Connors 2005).

Juveniles resemble adult females, but are often distinguishable from a distance by their

behaviour, e.g. submissive postures, making ‘begging’ calls. They moult into full adult

plumage in their fourth year (Johnstone and Storr 1998; Higgins 1999; Forshaw 2002).

Carnaby’s Cockatoos have a much finer and narrower curved upper mandible or

culmen, with a prolonged, tapered tip extending beyond the maxilla. Their contact calls

also differ (Johnstone et al. 2010). Both White-tailed species are strong fliers (Forshaw

2002).

19

FRTBC are similar in size to the two White-tailed species, but with a more marked

sexual dimorphism (Johnstone and Storr 1998) (Figure 2.1). Males possess dark bills,

glossy-black plumages, crests and tails with a broad sub-terminal band of crimson-red.

Females have lighter bills, brown-black plumages and crests heavily spotted with

numerous yellow speckles to their shoulders, yellow-orange barred feathers covering

the underparts and orange-red barring on their tail feathers (Johnstone and Storr 1998;

Higgins 1999; Forshaw 2002; Connors and Connors 2005). Juvenile FRTBC resemble

adult females, attaining adult plumage in their fourth year (Johnstone and Storr 1998;

Higgins 1999; Forshaw 2002). When in flight, FRTBC has a distinctive grating call.

Juvenile begging calls are softer, and similar to that of the two White-tailed species.

They are also strong fliers (Forshaw 2002).

(a) (b)

(c)

Figure 2.1 (a) Carnaby’s Cockatoo male ♂ (left) and female ♀ (right); (b) Baudin’s Cockatoo male ♂ (left) and female ♀ (right) and (c) FRTBC male ♂ (left) and female ♀ (right). Images from Johnstone and Storr (1998).

20

2.1.4 Distribution, Range and Habitat

Carnaby’s Cockatoo has the largest range. It is encountered throughout the Avon

Wheatbelt, the Mallee, the Esperance Sandplains, the Jarrah Forest, and the Swan

Coastal Plain Bioregions (Figure 2.2) (Saunders 1979; Saunders and Ingram 1995;

Johnstone and Storr 1998; Barrett et al. 2003; Johnstone et al. 2008). Thus, Carnaby’s

Cockatoo is sometimes referred to as the more ‘kwongan-heath’ species, given the

dominance of woodland and heathland habitat through much of its range (Johnstone et

al. 2008). Large-scale clearing has reduced its range in the Avon Wheatbelt and the

Mallee, and the centre of distribution of Carnaby’s Cockatoo has shifted significantly

westward and southward (Johnstone et al. 2008).

Baudin’s Cockatoo and FRTBC primarily inhabit the eucalypt forests of the Jarrah-

Marri forest or the Karri (Eucalyptus diversicolor) forest (in the Warren Bioregion).

FRTBC also occur on the Swan Coastal Plain (Figure 2.2) and occasionally Baudin’s

Cockatoos) as well (Johnstone and Storr 1998; Johnstone and Kirkby 2008). Baudin’s

Cockatoo and FRTBC are referred to as the ‘forest black cockatoos’ because their

distribution, feeding and breeding requirements are closely linked to forested habitats

(Chapman 2008).

Though Baudin’s Cockatoo can be found as far north as the northern Darling Range,

their breeding activity is centred in the tall, dense stands of Karri forest (Johnstone and

Kirkby 2008). Some breeding activity also occurs in the Jarrah-Marri forest where there

are recent records of ‘postnuptial visitors’ in the Wungong catchment (Johnstone and

Kirkby 2009; Johnstone et al. 2010). Baudin’s Cockatoo is not known to range beyond

the distribution of Marri, which is an important food resource (Saunders et al. 1985).

Breeding reports for FRTBC are mainly in the Jarrah-Marri forest (Saunders 1977;

Johnstone and Storr 1998; Chapman 2008).

21

Figure 2.2 Distribution and range of (a) Carnaby’s Cockatoos; (b) Baudin’s Cockatoos and (c) FRTBC, respectively (Images reproduced from Department of Sustainability, Environment, Water, Population and Communities 2011).

2.1.5 Seasonal Patterns and Movement

The seasonal movement patterns of Carnaby’s Cockatoos are variable. In higher rainfall

areas Carnaby’s Cockatoos may be resident year-round, while in woodland areas north

and east of the Jarrah-Marri and Karri forests, they are postnuptial migrants (Johnstone

and Storr 1998). In the non-breeding period (roughly January - June), Carnaby’s

Cockatoos form large flocks or hundreds of birds that migrate from inland districts to

higher rainfall sites in the Swan Coastal Plain and the Jarrah Forest Bioregions. Post-

breeding movements include regular visitations to Pine (Pinus spp.) plantations in

coastal areas (Saunders 1974, 1980; Shah 2006; Johnstone et al. 2008; Finn et al. 2009;

Saunders and Dawson 2009). In the breeding season (roughly July - December), flocks

return from the coast, breaking into smaller groups that travel to their traditional inland

breeding areas in the Wheatbelt (Johnstone et al. 2008).

Baudin’s Cockatoos also undertake a post-breeding migration (from roughly March to

September), leaving their nesting areas in the Karri forest and moving northwards into

the Jarrah-Marri forest to feed on Marri (Johnstone and Kirkby 2008). In October, they

return southwards for the breeding season (roughly October - February) (Johnstone and

Kirkby 2008). FRTBC are generally thought to be resident year-round within a defined

home range, though smaller-scale seasonal movements may occur because of changes in

(a) (b) (c)

22

water and food availability (Abbott 1998; Johnstone and Storr 1998; Johnstone and

Kirkby 2009, 2010; Johnstone et al. 2010; DSEWPaC 2012a).

2.1.6 Foraging Ecology

While Carnaby’s Cockatoos are sometimes characterised as preferring the seeds and

flowers of proteaceous plants and Baudin’s Cockatoos as being Marri specialists, the

reality is more complex. Both species feed on the seeds, nectar, fruits, buds and

flowers/blossoms of proteaceous species including Hakea and Banksia, as well as

myrtaceous species such as Jarrah and Marri (e.g. Saunders 1974a,b, 1979, 1980;

Johnstone and Storr 1998; Johnstone and Kirkby 2009; Johnstone et al. 2008; Valentine

and Stock 2009). Nonetheless, there are clear differences in the feeding ecology of the

two species, e.g. Baudin’s Cockatoos can extract seeds from mature Marri fruits, while

generally Carnaby’s Cockatoos can only feed on immature (and therefore softer) Marri

fruits. These differences in food preferences reflect different morphological adaptations

in bill shape and in feeding technique (Cooper 2000). Baudin’s Cockatoos pry seeds out

of the Marri fruit capsule with their elongated upper mandible, while Carnaby’s

Cockatoos breaks open and damages the rim of the fruit with its shorter, blunt mandible

(Cooper 2000). Carnaby’s Cockatoos may also chew off the side of the fruit capsule to

extract seeds (J Lee, Murdoch University, pers. obs.).

Both Baudin’s Cockatoos and Carnaby’s Cockatoos feed on introduced plant species

(Robinson 1960; Saunders 1974, 1977, 1980; Johnstone et al. 2008). Carnaby’s

Cockatoos regularly eat exotic plants, particularly Pine species (Pinus pinaster and P.

radiata) in plantations (Perry 1948; Saunders 1974; Finn et al. 2009). Baudin’s

Cockatoo feed on the seeds of cultivated nuts (e.g. almonds) and fruits (e.g. apples and

pears) (Johnstone and Kirkby 2008). Johnstone and Kirkby (2008) suggest these foods

are eaten when the seasonal production of their native food sources is low.

All three species eat insect larvae. Baudin’s Cockatoo use their longer mandibles to

extract insect larvae from under the bark of trees, a process called ‘grubbing’ (Robinson

1965). Carnaby’s Cockatoo dig out larvae hidden in the budding stems of Banksia

species (e.g. Seed-eating Weevil Alphitopis nivea in B. attenuata) (Scott and Black

1981). FRTBC strip the bark off trees to harvest insect larvae underneath roosting

(Chapman 2008).

23

FRTBC are essentially arboreal, feeding in the canopy of trees or tall shrubs and only

coming the ground to drink (Forshaw 2002). FRTBC feeds on the seeds, fruits, berries,

nectar and blossoms of eucalypts, Allocasuarina, Persoonia and other species

(Johnstone and Storr 1998; Johnstone et al. 2010). FRTBC extract Marri seeds by

shearing and chewing through the shell of the fruit with their large bills (Johnstone and

Kirkby 1999; Cooper 2000). Seeds of Jarrah and Marri make up almost 90% of their

food-intake, making these vitally important foods (Johnstone and Kirkby 1999; Cooper

et al. 2003). Johnstone and Kirkby (1999) suggested that the energy return from Marri

and Jarrah fruits, or volume of fruits produced by trees traditionally used for FRTBC

feeding, may limit the number of birds that can breed in the area.

2.1.7 Breeding Biology

While the breeding biology of Carnaby’s Cockatoo has been well-studied (e.g. Saunders

1982; Saunders and Ingram 1998; Saunders and Dawson 2009), breeding in Baudin’s

Cockatoo and FRTBC is less well-understood because of difficulties in locating nesting

trees in tall forest where these birds breed (Saunders 1977; Johnstone and Kirkby 1999,

2008). Nonetheless, as all three black cockatoos are similar in size, the three likely share

similar requirements in terms of hollow dimensions.

Carnaby’s Cockatoo hollows are on average 5 - 6 m high in smooth-barked Wandoo (E.

wandoo) and Salmon Gum (E. salmonophloia) (Saunders et al. 1982). Baudin’s

Cockatoo and FRTBC usually nest in taller Karri, Marri, Wandoo and Jarrah trees

(Johnstone and Storr 1998); FRTBC may prefer much larger hollows located near a

water source (Forshaw 2002). Some pairs reuse hollows from the previous breeding

season, although if the breeding attempt was unsuccessful, they rarely return to the same

hollow (Saunders 1982).

For Carnaby’s Cockatoo, breeding success depends on sufficient feeding sites adjacent

to nest sites (Saunders 1982, 1986). Sites with extensive areas of native vegetation have

greater breeding success than sites with little native vegetation following land clearing

(Saunders 1986). Although a surplus of trees with adequate hollows may occur at some

sites, the clearing of native vegetation has left little heathland with proteaceous food

plants adjacent to the remaining woodlands (Saunders 1980; Saunders 1986; Saunders

and Ingram 1987; Saunders 1990).

24

2.1.8 Survival

All three species are long-lived, although little is known about their lifespan or survival

rates in the wild. The Action Plan for Australian Birds 2010 reports generation times as

19.2 years for Baudin’s Cockatoos, Carnaby’s Cockatoos, and FRTBC (Garnett et al.

2011). The annual survival rate for adult Carnaby’s Cockatoo is 61 - 69%, compared to

15% for 12-month-old juveniles (Saunders 1982). For FRTBC, in some breeding areas

only 10% of the adult population breeds every year and the juvenile survival rate (one

year post-fledging) is 15% (Chapman 2008). For all three species, known primary

causes of death include Wedge-tailed Eagle (Aquila audax) attacks, road deaths/car

strikes and shooting (Cale 2003, Chapman 2008).

2.1.9 Reasons behind Black Cockatoos being listed as threatened

Conservation concerns for these black cockatoos are based on evidence of: (1) major

declines from historic abundances; (2) contraction of historic ranges following broad-

scale habitat clearing, and (3) continuing and/or mounting threatening processes (Cale

2003, Chapman 2008). All three black cockatoos display traits characteristic of K-

selected species, including: large body sizes, long life expectancies, high adult survival

and low recruitment rates (Begon et al. 2006). This makes them particularly sensitive to

increases in mortality and declines in recruitment.

Major causes of adult mortality include road strikes (especially where birds drink at

road side puddles) and, in the case of Baudin’s Cockatoo, illegal shooting or poisoning

by orchardists (Cale 2003; Chapman 2008; Johnstone and Kirkby 2008; DSEWPaC

2012a,b,c). Disease has also been postulated as an important factor regulating breeding

populations (Saunders et al. 2011), and pathology and epidemiology remain important –

yet little known – areas of research.

Clearing of native vegetation removes nesting trees as well as feeding habitat (Saunders

1977, 1982; Saunders et al. 1985; Mawson and Long 1994). Even if nesting trees are

retained in a cleared landscape, they may not be attractive to birds if no suitable food

occurs nearby (Saunders 1982, 1985). Agricultural clearing has been more limited

within the Jarrah-Marri forest and the Karri forest than in the wheatbelt, and has often

focused on fertile soils along waterways (see maps in Jarvis 1981). Timber production

in the Jarrah-Marri forest and the Karri forest regenerates forest stands after logging and

now retains a minimum density of habitat trees on logged plots (Conservation

25

Commission 2003; Stoneman 2007). However, early 20th century management involved

removal of over-mature, senescent trees and trees of non-commercial species, which

likely reduced hollow availability (Wardell-Johnson et al. 2004).

Almost all of the Jarrah-Marri forest has been logged sometime in the past, with

remnant unlogged forests occurring primarily along watercourses or on hillcrests.

Intense wildfires can also cause hollow-bearing trees to fall (Wardell-Johnson et al.

2004; Parnaby et al. 2010). The remaining hollows may be subject to intense

competition from invasive hollow-using native species such as Australian Ringnecks

(Barnardius zonarius), Western Corellas (Cacatua pastinator butleri), Galahs (Cacatua

roseicapilla) and even certain tree duck species (e.g. Grey Teal; Anas gracilis) (Cale

2003; Chapman 2008). Feral populations of the introduced European honeybee (Apis

mellifera) are major nest competitors in forested areas (Johnstone and Kirkby 2007).

Mining may also reduce hollow availability within forest landscapes. Although

abandoned mines are rehabilitated with native vegetation, old trees with hollows take

over a century to return (Mawson and Long 1994; Abbott and Whitford 2002; Whitford

2002b; Whitford and Williams 2002; Whitford and Stoneman 2004).

2.2 The Study Area – Newmont Boddington Gold

2.2.1 Regional and Geographical Setting

The Newmont Boddington Gold (NBG) mine is a large, open-cut gold and copper

mining project located approximately 130 km southeast of Perth and 13 km northwest

of Boddington on the eastern margin of the Jarrah-Marri forest (Figure 2.3). The region

around Boddington and the NBG mine-site is an important source of potable water,

copper, bauxite and gold (Worsley Alumina Pty. Ltd. 1999).

26

Figure 2.3 The location of the study site near Boddington, southwestern Australia.

The NBG study site is a mining production landscape within a mosaic of other

production landscapes including farming, State Forest and plantation timber. It is

bordered to north and east by a series of monadnocks including Mount Wells

immediately to the northwest. To the north and east, NBG encroaches upon a large

silviculture operation, which covers c. 11 800 ha and consists of native forest timber

and non-native pine (principally Radiata Pine; Pinus radiata) and Tasmanian Blue Gum

(E. globulus) plantations (Worsley Alumina Pty. Ltd. 1999). The operation was

managed by Wesfarmers as Sotico Pty. Ltd. during the course of this study (it was sold

privately in late 2011). Agricultural holdings lie between NBG and the Hotham River.

These include Hotham Downs, a pastoral operation that was managed by Alcoa

(Aluminium Company of America) during this study, along with several private

landholdings. The Hotham River is about 4 km south of the main mining area at NBG.

NBG is bordered to the west by State Forest. The State Forest stands are logged on a

27

40-year rotation and logging occurred in several forest areas adjacent to NBG during

this study. Between the mine and the town are several residences and the site

accommodation village for NBG (Worsley Alumina Pty. Ltd. 1999).

NBG is not the only mining operation in the vicinity. It is approximately 50 km

southeast of Alcoa’s Huntly mine and 25 km north of Worsley’s Boddington Bauxite

Mine. An expansion project for the Worsley mine will shift mining activities

northwards towards NBG tenements (Worsley Alumina Pty. Ltd. 1999).

2.2.2 Disturbance History

Compared to the Avon Wheatbelt, Mallee, Esperance Sandplains and Swan Coastal

Plain Bioregions, the Jarrah Forest has suffered less direct clearing and local extinctions

(Abbott and Armstrong 1995). Forest management (including logging, road

construction and fuel reduction burning) was the major anthropogenic disturbance in the

Jarrah Forest Bioregion since European settlement of Western Australia in 1826 until

the second half of the 20th century (Wardell-Johnson and Horwitz 1996; Wardell-

Johnson et al. 2004; Calver and Wardell-Johnson 2004; Stoneman 2007). The over-

storey in logged forest was opened with a reduction of up to 50% in the canopy

(Wallace 1965), while piles of logging-slash abandoned on the forest floor increased

fuel loads. Much of the machinery was steam-driven and this led to intense fires over

wide areas (Wallace 1965; Forests Department 1971; McCaw and Burrows 1989;

Burrows et al. 1995). The NBG site was first logged in the early 1900s and has been

logged at least twice since (Heberle 1997; Worsley Alumina Pty. Ltd. 1999). Changes

in fire regimes associated with changing forestry practices over the past decades have

also impacted the surrounding forest ecosystems (Calver and Dell 1998).

Mining operations first commenced at the NBG mine-site in 1987 (Rayner et al. 1996).

Prior to this, the site was managed as State Forest and as a private forest holding by

Sotico Pty. Ltd. Most of the existing disturbance footprint for the current NBG mine-

site was established during the initial establishment and operation of the mine, including

the clearing of native forest for infrastructure (e.g. processing plant, haul roads), two

major pit areas, waste rock dumps, several small ‘satellite’ pits, residue disposal areas,

and water supply reservoirs/dams (Rayner et al. 1996).

28

In 2001, mining ceased and the mine-site entered a care-and-maintenance phase until a

transfer of ownership, which led to the initiation of an expansion project for the mine in

2006 and recommencement of mining operations in 2009. The current mining operation

targets low-grade, hard rock ores beneath existing pits mined during past operations. To

date, the recent expansion project has required clearing of c. 1 200 ha of native forest,

bringing the total disturbance footprint for the mine-site to c. 3 500 ha (Worsley

Alumina Pty. Ltd. 1999).

2.2.3 Biogeography, Topography, Hydrology and Climate

Mean annual rainfall is between the 700 mm and 800 mm isohyets (Rayner et al. 1996).

The study site is located on the highly leached soils of the Darling Plateau and Scarp

where the highest elevations reach 400 m (Dell et al. 1989). Rivers and their tributaries

that originate east in agricultural lands bisect the Darling Plateau around Boddington.

Water quality ranges from fresh to saline and drainage is intermittent. Most of the NBG

mine-site is situated within the Thirty-Four Mile Brook catchment, a tributary of the

Hotham River that merges with the Murray River closer to the coast (Worsley Alumina

Pty. Ltd. 1999).

2.2.4 Vegetation Structure and Flora

NBG is located at an ecotone where the Jarrah-Marri forest blends into Wandoo

woodland further east. The vegetation communities are determined by topography, soil

pH and fertility, moisture availability and disturbance history (Dell et al. 1989). The

canopy-storey layer is dominated by Jarrah trees intermingled with varying admixtures

of Marri and Wandoo (E. wandoo), interspersed with Yarri (E. patens) (Worsley

Alumina Pty. Ltd. 1999; Biggs et al. 2011). The understorey comprises mainly Bull

Banksia (Banksia grandis), Forest Sheoak (Allocasuarina fraseriana), and Persoonia

species, with an even lower shrub layer consisting of proteaceous shrubs (namely

Banksia and Hakea species) and other plants species (e.g. Macrozamia and

Xanthorrhoea spp.) (Worsley Alumina Pty. Ltd. 1999). Sandy soils along upper slopes

are associated with the presence of Forest Sheoak. The moist, gravelly and more fertile

soils supporting Marri and Bull Banksia occur lower in the landscape (Dell et al. 1989;

Worsley Alumina Pty. Ltd. 1999).

29

2.2.5 Vertebrate Fauna

The forest at NBG is an important habitat for many vertebrates. Fauna surveys on-site

have recorded 67 bird species, 13 native and 5 introduced mammals, 20 reptile species

and 13 amphibian species (Worsley Alumina Pty. Ltd. 1999, Ninox Wildlife Consulting

2003). Threatened fauna include the Chuditch (Dasyurus geoffroii) and Carpet Python

(Morelia spilota variegata), as well as the three black cockatoos. Thirty-Four Mile

Brook and the heath along Old Soldiers’ Road and around Boomerang Swamp are

important as feeding and breeding resources for maintaining populations of

insectivorous and nectar-feeding birds. Several swamps (e.g. Boomerang, Eight, Pillow

and Round Swamps) and ridges are important denning sites for Chuditch (Worsley

Alumina Pty. Ltd. 1999).

2.2.6 Black Cockatoos

Carnaby’s Cockatoo, Baudin’s Cockatoo, and FRTBC were all recorded in the NBG

mining tenement in the 1990s (Worsley Alumina Pty. Ltd. 1999). Prior to the recent

expansion project, the area was surveyed and all three species were observed using the

mining tenement for roosting, feeding or nesting, in numbers ranging from a single bird,

to pairs, small family groups and larger flocks (Ninox Wildlife Consulting 2003).

2.2.7 Conservation Values, Concerns and Issues for Newmont Boddington Gold Black

Cockatoos

The loss of feeding and breeding habitat within the Jarrah-Marri forest through

agricultural clearing, logging and mining has been identified as a conservation concern

for black cockatoos (e.g. Mawson and Long 1994; Wardell-Johnson et al. 2004;

Chapman 2008), although there is scientific debate about the nature of the impact of

different activities (Abbott 1998; Abbott and Whitford 2002). All three species feed in

the Jarrah-Marri forest, either during seasonal migrations (Carnaby’s Cockatoo and

Baudin’s Cockatoo) or all year-round (FRTBC) (Saunders 1974a, 1980; Johnstone and

Kirkby 1999; Johnstone et al. 2010; Biggs et al. 2011), so the forest represents an

important feeding ground. While recent research has indicated that restored vegetation

in rehabilitated mine pits of the Jarrah-Marri forest provides some food for black

cockatoos (Lee et al. 2010), the value of rehabilitation vegetation relative to native

forest has not been determined.

30

Additionally, black cockatoos are the largest hollow-using bird in the region. The large

hollows these birds require occur only in tall, mature and senescent trees more than 150

years old (Mawson and Long 1994; Abbott and Whitford 2002; Whitford 2002;

Whitford and Williams 2002; Whitford and Stoneman 2004). Carnaby’s Cockatoo and

FRTBC breed throughout the Jarrah-Marri forest, while Baudin’s Cockatoo breeds in

southern regions of the Jarrah-Marri forest, but primarily in the Karri forest of the far

south (Saunders and Ingram 1995; Johnstone and Storr 1998; Johnstone et al. 2010).

Thus the Jarrah-Marri forest represents an important breeding habitat for these species

as well. Subsequent chapters address the ecology of these black cockatoos at the NBG

site.

 

  31  

Chapter 3 Ecology of Black Cockatoos at a Mine-site in the

Eastern Jarrah-Marri Forest, Western Australia

This chapter is published:

Lee, J. G. H., Finn, H. C. and Calver, M. C., 2013. Ecology of black cockatoos at a

mine-site in the eastern Jarrah-Marri forest, Western Australia. Pacific

Conservation Biology 19: 76 - 90.

To keep a consistent style with the rest of the thesis, the abstract,

acknowledgements and keywords were deleted and the references included in the

main reference list at the end of the thesis. Numbering of headings and sub-

headings follows the format of the rest of the thesis. Otherwise the text follows the

published paper (except for the addition of an appendix requested by an examiner,

extra detail in the legend to Figure 3.4 and one small clarification in the text), even

though this involves some repetition of details from earlier thesis chapters in the

introduction and methods.

The chapter has two co-authors. M. C. Calver and H. C. Finn assisted in research

design and analysis, while H. C. Finn also assisted in data collection in the field.

 

  32  

3.1 Introduction

Three threatened black cockatoos occur in the Jarrah Eucalyptus marginata-Marri

Corymbia callophyla forest of Western Australia: Forest Red-tailed Black Cockatoo

Calyptorhynchus banksii naso [FRTBC] (a sub-species), Carnaby's Cockatoo C.

latirostris, and Baudin's Cockatoo C. baudinii. They are listed as threatened under the

Commonwealth Environmental Protection and Biodiversity Conservation Act 1999 and

as Schedule 1 fauna [species that are ‘rare or likely to become extinct and in need of

special protection’] under the Western Australia Wildlife Conservation Act 1950 (see

also Cale 2003; Chapman 2008; Garnett et al. 2011).

The Jarrah-Marri forest also supports timber and mineral production, water catchments,

and other land uses (Conservation Commission of Western Australia 2003). Loss of

breeding and feeding habitat within the Jarrah-Marri forest through logging and mining

are a potential concern for black cockatoos (Mawson and Long 1994; Calver and Dell

1998; Wardell-Johnson et al. 2004; Chapman 2008; Garnett et al. 2011), although the

significance of these impacts is debated (Stoneman et al. 1997; Abbott 1998; Abbott

and Whitford 2002).

Black cockatoos are especially vulnerable because they nest in large hollows occurring

only in mature and senescent trees generally more than 150 years old (Saunders et al.

1982; Whitford 2002a; Whitford and Williams 2002; Whitford and Stoneman 2004).

FRTBC and Carnaby's Cockatoos breed throughout the Jarrah-Marri forest, while

Baudin's Cockatoos breed in southern sections (Saunders et al. 1985; Johnstone and

Storr 1998; Johnstone and Kirkby 2008).

All three black cockatoos feed in the Jarrah-Marri forest year-round, seasonally, or

during seasonal migrations (Saunders 1974a, 1980; Johnstone and Kirkby 1999, 2008;

Biggs et al. 2011). Studies of crop contents and observations of feeding behaviour

indicate that Jarrah and Marri are the main foods within the forest (Saunders 1974a,b,

1980; Johnstone and Kirkby 1999, 2008). This reflects the energetic characteristics of

Jarrah and (in particular) Marri seeds and the dominance of these two overstorey species

across the Jarrah-Marri forest (Pryor 1959; Abbott and Loneragan 1986; Whitford

2002b; Cooper et al. 2003; Koch and Samsa 2007).

 

  33  

Baudin’s Cockatoos are considered Marri specialists, although they also eat proteaceous

shrubs, insect larvae, orchard fruit, and other plants, including the buds and flowers of

Banksia spp. and Eucalyptus spp. (Saunders 1974b; Johnstone and Storr 1998; Cale

2003; Chapman 2007; Johnstone and Kirkby 2008). Marri and Jarrah seeds comprise

around 90% of the diet of FRTBC, although they also eat seeds of other eucalypts,

Forest Sheoak Allocasuarina fraseriana and Snottygobble Persoonia longifolia

(Robinson 1965; Johnstone and Storr 1998; Johnstone and Kirby 1999; Cooper et al.

2003). The foraging ecology of FRTBC has changed over the past 12 years, including

changes in the proportions of different food plants, probably in response to changing

food availability (R Johnstone and T Kirkby, Western Australian Museum, 2009, pers.

comm.). Carnaby’s Cockatoos also feed within forested areas, but have a more varied

diet, including seeds, flowers, and nectar of Jarrah, Marri, Banksia spp., Hakea spp.,

and Pinus spp. These food plants occur within proteaceous scrubs and heathland,

eucalypt forests and woodlands, and pine plantations (Saunders 1974a,b; Saunders

1980). They also consume insects (Saunders 1980; Scott and Black 1981). All three

black cockatoos forage on mine-site revegetation (Lee et al. 2010).

Despite this basic understanding of the ecology of black cockatoos in the southwest,

little is known of their response to anthropogenic disturbance. While Weerheim (2008)

and Lee et al. (2010) present preliminary data, their work covered only short periods.

Given that the feeding ecology of black cockatoos can vary markedly between years in

response to the irregular flowering of Jarrah and Marri (e.g., Johnstone and Storr 1998;

Johnstone and Kirby 1999), this is a significant limitation. Furthermore, because black

cockatoos are highly mobile and occur at low density, observational methods are

imperfect for species detection (e.g., Craig and Roberts 2005) and for assessing food

use. Searches for distinctive feeding residues beneath plants (Cooper 2000, Johnstone

and Kirkby 1999, Biggs et al. 2011) can overcome these difficulties.

Here we describe four aspects of the black cockatoos’ ecology at the Newmont

Boddington Gold Mine (NBG), located at the eastern margin of the Jarrah-Marri forest,

based on data collected over three years. We used behavioural observations and

assessments of feeding residues to investigate: (1) group sizes, including seasonal

changes; (2) habitat use, especially of mine-site rehabilitation; (3) food plants; and (4)

occupancy patterns, including seasonal and inter-annual changes. This provides a

baseline to assess the value of the NBG landscape to black cockatoos, the likelihood

 

  34  

and severity of human impacts associated with mining and other activities, and the

value of rehabilitated mine-sites as a food resource.

3.2 Materials and Methods

3.2.1 Study Site and Species

The NBG mine-site is an open-cut gold and copper mine located 130 km southeast of

Perth along the eastern margin of the Jarrah-Marri Forest (Figure 3.1). Mean annual

rainfall is between the 700 mm and 800 mm isohyets and the site occurs along the

ecotone where the landscape changes to the Wandoo E. wandoo woodland of the

neighbouring Avon Wheatbelt bioregion (Thackway and Cresswell 1995; Rayner et al.

1996).

 

  35  

Figure 3.1 Map of study area showing the NBG mining areas and residue disposal area, the Land-Exchange Area, and surrounds. The cleared land/non-native vegetation to the northeast of the residue disposal area is largely pine plantations and blue-gum (Eucalyptus globulus) plantations, while that to the east and southeast is mainly paddocks.

 

  36  

The study site includes NBG mining tenements and lands immediately adjacent to them.

NBG tenements contain active mining areas (mining pits, waste rock dumps), mine-site

infrastructure (buildings, roads, water reservoirs), residue disposal areas (RDAs), water

supply reservoirs (WSRs), rehabilitated mining pits, and remnant native forest.

Adjacent lands include native forest (within State Forest, Monadnocks Conservation

Park, and private lands), agricultural operations (livestock pastures), and plantations

(principally Radiata Pine Pinus radiata and Tasmanian Blue Gum E. globulus) (Figure

3.1).

Jarrah is the dominant canopy species within native forest, with admixtures of Marri

and Wandoo and occasional patches of Yarri E. patens and Flooded Gum E. rudis

(Worsley Alumina Pty Ltd. 1999; Biggs et al. 2011). Mid-storey species include Forest

Sheoak and Bull Banksia Banksia grandis. The shrub layers include several proteaceous

food species for black cockatoos, including Parrot Bush B. sessilis∗, Harsh Hakea Hakea

prostrata, and Wavy-leaved Hakea H. undulata (Dell et al. 1989; Worsley Alumina Pty

Ltd. 1999; Biggs et al. 2011).

Nest sites for Carnaby’s Cockatoos and FRTBC occur within NBG tenements and their

surrounds (J Lee and H Finn, Murdoch University, unpublished data). NBG is beyond

the breeding range of Baudin’s Cockatoos, which breed in the southern Jarrah-Marri

forest and even further south in the Karri Eucalyptus diversicolor forest from spring to

autumn (Johnstone and Kirkby 2008). However, their breeding range may be shifting,

with recent records of breeding in the Wungong catchment in the northern Jarrah-Marri

forest (Johnstone and Kirkby 2008).

3.2.2 General Distribution Survey (GDS)

We conducted vehicle-based surveys monthly from December 2007 to July 2010 within

the NBG mine-site (Figure 3.2), and similar native forest to the north known as the

Land-Exchange Area (LEA) (Figure 3.3). The eastern margin of the LEA adjoined an

extensive pine and eucalypt plantation. To distinguish this sampling method from

others, we call it a ‘general distribution survey’ (GDS). Surveys within the NBG mine-

site and within the LEA were separate and are called either the NBG GDS or the LEA

∗ Previously known as Dryandra sessilis, but placed under Banksia by Mast and Thiele (2007).

 

  37  

GDS. We considered the areas separately to contrast species occurrence and to gather

habitat use data within the LEA’s mixed forest-plantation landscape.

Figure 3.2 Location of the site survey points for the black cockatoo general distribution survey across the NBG study site.

WA

Map Area

´

NEWMONT BODDINGTON GOLD

9

8

7

6

54 3

2

1

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

23

22

21

2019

1817

16

15

14

13

12

11

10

435000

435000

440000

440000

445000

445000

63

70

00

0

63

70

00

0

63

75

00

0

63

75

00

0

63

80

00

0

63

80

00

0

63

85

00

0

63

85

00

0

NBG SITE

GENERAL DISTRIBUTION SURVEY

Author: S. Myles

Drawn: V. Preedy

File:BOD_Env_Fau_CockatooSurveySitesNBG.mxd\AUS\WA\Boddington\_Environmental

Date:July 2012

Scale: 1:75 000

Proj: MGA94 Zone 50

Boddington Project

EWMONTNASIA PACIFIC

0 1 2

Kilometers

 

  38  

Figure 3.3 Location of site survey points for the black cockatoo general distribution survey across the LEA study site.

A GDS involved driving a pre-determined route through the NBG mine-site or LEA,

following roads and tracks while looking and listening for cockatoos at defined survey

points. At NBG, several factors limited where survey points could be located or

WA

Map Area

´

NEWMONT BODDINGTON GOLD

98

7

6

5

4

3

2

1

33

32

31

30

29

28

27

26

25

24

23

22

21

20

19

18

17

16

15

14

13

12

11

10

437500

437500

440000

440000

442500

442500

445000

445000

63

87

50

0

63

87

50

0

63

90

00

0

63

90

00

0

63

92

50

0

63

92

50

0

63

95

00

0

63

95

00

0

LAND EXCHANGE AREA

GENERAL DISTRIBUTION SURVEY

Author: S. Myles

Drawn: V. Preedy

File:BOD_Env_Fau_CockatooSurveySitesLEA.mxd\AUS\WA\Boddington\_Environmental

Date:July 2012

Scale: 1:50 000

Proj: MGA94 Zone 50

Boddington Project

EWMONTNASIA PACIFIC

0 1 2

Kilometers

 

  39  

prevented sampling the complete route. They included: construction, mining, and

maintenance activities; restrictions on access; blocked or dead-end roads; safety and

practicality; and closure of forest after heavy rainfall (a protocol to limit the spread of

plant pathogens). Therefore, the NBG GDS varied monthly in length and the survey

points sampled (Figure 3.2). The extensive network of unsealed forest tracks in the LEA

and the lack of human activities allowed complete sampling of a defined route (Figure

3.3).

The survey route measured 24 – 33 km for an NBG GDS and 27 km for the LEA GDS.

After initial site inspections, we established 38 designated survey points for the NBG

GDS (Figure 3.2) and 33 points for the LEA GDS (Figure 3.3). At NBG, distances

between survey points ranged between 0.7 and 1.2 km. At LEA, points were roughly

equal distances (0.8 km) apart. During the survey period at the NBG site, survey points

17 - 27 were lost to the mine expansion (near the RDAs) and survey points 6 - 8 to a

large bushfire.

A GDS usually commenced at first light and continued into the early afternoon. This

captured the peak activity when birds leave roost sites and disperse to feeding sites. The

survey route for the LEA GDS was always driven in the same direction because the site

layout made it impractical to modify the direction. We alternated the direction of the

NBG GDS from month-to-month, avoiding potential biases associated with the

sampling time. The time to complete an individual survey ranged from two to seven

hours, depending on encounters and environmental conditions.

This protocol risks double-counting birds because: (1) black cockatoos may move

several kilometers a day as they shift among roosting, drinking, and feeding sites and

(2) it is also not possible to recognise individual black cockatoos (or flocks).

Nevertheless, we attempted to keep track of where birds were observed and heard and

determine the direction in which flying birds were moving. These methods may have

reduced double-counting for survey points sampled within a short time of each other

(i.e. within 15 - 20 minutes), but do not justify using the data to estimate population

sizes.

At each survey point, we stopped for two minutes to listen and watch for black

cockatoos. Auditory and visual observations were recorded separately. For analyses, we

 

  40  

only included detections in which we identified the species. When birds were detected

at a GDS survey point, we tried to locate them and collect data on species, group size,

habitat use and activity.

3.2.3 Behavioural Methodology

We performed a behavioural survey each time that we observed black cockatoos,

whether opportunistically or during a systematic search. During behavioural surveys we

approached as closely as possible without causing disturbance. Behavioural data were

collected for the first ten minutes of the observation period using a scan sampling

protocol (Altmann 1974). Efforts are made to observe all the individuals present and to

ascertain their activity state (Altmann 1974). We recorded predominant activity (the

activity state of ≥50% of individuals during the 10-minute scan sample); group size;

predominant habitat type (the habitat type used by ≥50% of individuals during the 10-

minute scan sample, all samples met this criterion); and any foods eaten. For more

details, refer to the Appendix I for the behavioural survey form. We classified habitat

type to one of nine categories (Table 3.1).

Table 3.1 Habitat types identified within study area. Habitat Type Description

Native Forest

Continuous to nearly continuous canopy, usually with Jarrah and Marri main canopy forming trees; midstorey layer including A. fraseriana, B. grandis, B. sessilis, Persoonia longifolia, and Hakea spp.; ground and shrub layer including Banksia and Hakea spp.

Native Woodland Open habitat with discontinuous canopy, usually with Wandoo as the main canopy forming tree; ground and shrub layer including Banksia and Hakea spp.

Modified Landscape Human-modified aspects of the landscape not characterised by other habitat types (e.g. mining infrastructure such as powerlines, haul roads)

RDA/WSR Residue Disposal Area (RDA) or Water Supply Reservoir (WSR). Large (>1 km2) man-made waterbodies.

Mine-site Rehabilitation

Young (<12 years post-establishment) revegetation with high stem densities; Canopy species include Jarrah, Marri, and Wandoo; diverse and well-developed shrub later comprising such as Banksia and Hakea spp.

Sump Small man-made waterbodies designed as drainage and seepage sumps

Pine plantation Stands of pine trees within a commercial forestry landscape

Paddock Agricultural pasture areas with large open areas of grass and occasional single paddock trees

Remnant Vegetation Patch of remnant native vegetation within a forestry or agricultural matrix

 

  41  

We refer to each behavioural survey as an ‘encounter’, to distinguish behavioural

surveys from other field surveys (e.g. GDS) and from visual or acoustic detections

during a GDS. While we attempted to conduct a behavioural survey for every detection

during a GDS, this was often not practical or feasible (e.g. the birds were too distant).

3.2.4 Group Size

We focused on group (or ‘flock’) size as a measure of local abundance. Density

estimation is problematic for black cockatoos because they occur in very low densities

within the Jarrah-Marri forest (Craig and Roberts 2005; Weerheim 2008). Temporal

factors are also important, because flocks may feed in an area for only hours to days,

then move elsewhere nearby (e.g. for flocks resident within a locality) or further away

(e.g. for flocks migrating between breeding and non-breeding areas) (Saunders 1980;

Finn et al. 2009).

Flock sizes were counted as accurately as possible. If many individuals were present or

it was difficult to observe every bird, the estimated number of birds present was

recorded. Results were noted as counts or estimates. Groups more than 50 m apart were

considered distinct flocks, while groups less than 50 m apart were considered one flock.

To examine group size, data from behavioural surveys were summarised in descriptive

tables and graphs. We combined observations from the NBG GDS and LEA GDS

because our interest was in general patterns in group size for the three species across the

landscape. We also included opportunistic observations and observations collected

during other research activities (e.g. dawn/dusk surveys).

We used two-way analysis of variance (ANOVA) to assess the influence of season and

black cockatoo species on flock size, combining data for all years. We did not use a

repeated measures design because we could not follow individual flocks across seasons.

We classified seasons as: winter (June, July, August), spring (September, October,

November), summer (December, January, February), and autumn (March, April, May).

Data were highly heteroscedastic, necessitating logarithmic transformation. Significant

effects were further investigated using post hoc Tukey’s Honest Significant Difference

for unequal sample sizes.

 

  42  

Group sizes are presented as means (± standard errors) and as medians (25 and 75%,

interquartile range) because of skewed distributions and because the sample size for

Baudin’s Cockatoos was small relative to the others.

3.2.5 Habitat Type and Use

We considered only predominant habitat type recorded during the behavioural survey

for each encounter. We did not test whether or not encounters with each cockatoo

species were distributed evenly across the habitat types at each site because we did not

know the proportional land area of each habitat type at each site and could not calculate

expected values. Construction at NBG and the need to change to GDS routes also

argued against determining proportional land areas at NBG. Instead, we used a chi-

squared contingency table to test for an association between encounters with each

cockatoo species and habitat type.

We combined observations from NBG and LEA because sample sizes for the two areas

separately were too small for statistical analysis and because our interest was in

examining differences in how the three species used the range of habitats available to

them within both landscapes. We also included opportunistic observations and

observations collected during other research targeted activities. Very few encounters

were made in Sump, RDA/WSR, and Remnant Vegetation, so these categories were

combined. A conservative rule of thumb for contingency tables is that no more than

20% of cells should have an expected value <5, but Zar (2010) argued that this is too

restrictive. We followed his advice that for testing at the 5% level, the mean of the

expected values should be ≥6.

In addition, we used the diversity menu of the PAST (PAlaeontological Statistics)

software (Hammer et al. 2001) to calculate a Shannon diversity index, with 95%

confidence limits determined by bootstrapping, for the habitat types in which each

cockatoo species was encountered. We then compared the Shannon diversity indices

between species using PAST’s diversity t-test option. Lastly, we used PAST to calculate

evenness (with 95% confidence limits determined by bootstrapping) for each species’

habitat use, defined as the Shannon diversity for that species divided by the maximum

value it could assume if encounters occurred equally across habitat types. Evenness lies

between 0 and 1, with 1 indicating observations distributed equally and low values

 

  43  

indicating a weighting toward a few habitat categories. It thus indicates habitat

generalists (high evenness values) and specialists (low evenness values).

3.2.6 Food Plants

Black cockatoos may feed on the nectar, seeds, buds or flowers of several plants and on

insects under tree bark or on flowers and fruits. We considered a plant species to be a

food plant if we observed black cockatoos feeding on it or if we observed black

cockatoo feeding residues for it (e.g. eaten seed casings, cut branches). Black cockatoos

may leave species-distinctive markings on residues from some plant species (e.g. Marri)

(Cooper 2000; Johnstone and Kirkby 1999; Biggs et al. 2011). However, it is generally

not possible to attribute residues confidently to just one black cockatoo species.

Flora surveys at NBG recorded at least 30 known food plants for black cockatoos

(Worsley Alumina Pty. Ltd. 1999; Mattiske Consulting Pty. Ltd. 2005), based on

published reports of food plants for black cockatoos (Saunders 1980; Johnstone and

Kirkby 1999; Johnstone and Storr 1998; Johnstone et al. 2010; Valentine and Stock

2008) and observations at NBG (H Finn, Murdoch University, unpublished data). These

include: four eucalypts (Jarrah, Marri, Yarri, and Wandoo), seven Banksia species and

ten Hakea species. Most of the potential food plants flower or seed briefly or are rare,

and therefore represent only short-term or low abundance foods. Of the eucalypts, black

cockatoos are not thought to feed on Wandoo seeds (but just on the flowers and nectar),

in contrast to the other three eucalypts (Johnstone and Storr 1998; Valentine and Stock

2008).

To examine species differences in food plant use, we used only behavioural

observations and only observations in which the black cockatoo species or subspecies

was known. We used observations collected during GDS and opportunistically during

other research activities.

We used a chi-squared contingency table to test for associations between cockatoo

species and food plants. We also calculated a Shannon diversity index and evenness,

with 95% confidence limits determined by bootstrapping, for each cockatoo’s

predominant food plants. We then compared Shannon diversity indices between species

using the diversity t-test option in PAST (Hammer et al. 2001).

 

  44  

3.2.7 Site Occupancy

‘Occupancy’ refers to temporal patterns in presence within a defined area, e.g.

differences across seasons. For black cockatoos, this ranges between year-round

residency and transience (e.g. migrating and present for a few hours or days) (Johnstone

and Storr 1998; Cameron 2007). Individuals may be resident for part of the year if, for

example, they nest within the study area or if they use the study area as a feeding habitat

during non-breeding seasons.

To examine seasonal and annual patterns in occupancy, we examined black cockatoo

detections from the NBG GDS and the LEA GDS. We did not consider observations

collected outside GDS sampling sessions. Data were summarised in tables

documenting, for each GDS location (NBG or LEA), the numbers of sightings for each

species within: (1) each year of study (2007, 2008, 2009, 2010) with all seasons

combined and (2) each season of study (winter, spring, summer, autumn) with all years

combined.

We then analysed these data with chi-squared goodness of fit tests to determine

whether, for each species separately, there were similar numbers of detections at each

site in each season or year. Expected values for goodness of fit tests were calculated on

the assumption of equal numbers of detections in each season or year, corrected for

sampling intensity. Zar (2010) states that for testing at the 5% level, the chi-squared

goodness of fit test is appropriate if k ≥ 3, n ≥ 10, and n2/k ≥ 10 (where k is the number

of categories and n is the total number of observations). We followed this advice.

3.3 Results

Where appropriate, means are reported ± standard errors.

3.3.1 Group Size

Group size for Baudin’s Cockatoos ranged from two to 107, with a mean of 14.5 ± 3.5

and a median of 9.5 (25% = 3.00; 75% = 18.00) (n = 32 encounters). Few very large

flocks of Baudin’s Cockatoos were observed. The six largest group sizes were of 25 (n

= 3 groups), 27, 45, and 107 birds; 78.1% (n = 25 of 32 encounters) of groups contained

less than 20 birds.

 

  45  

Group size for Carnaby’s Cockatoos varied from one to 90, with a mean of 10.1 ± 1.3

and a median of 5.0 (25% = 2.00; 75% = 11.00) (n = 116 encounters). Few very large

flocks of Carnaby’s Cockatoos were observed. The five largest group sizes were of 40,

45, 65, 72, and 90 birds; 85.3% (n = 99 of 116 encounters) of groups contained less than

20 birds.

Group size for FRTBC ranged from one to 45, with a mean of 7.4 ± 0.6 and a median of

5.0 (25% = 3.00; 75% = 9.00) (n = 127 encounters). Few very large flocks of FRTBC

were observed. The five largest group sizes were of 25, 27 (n = 2 groups), 35, and 45

birds; 92.1% (n = 117 of 127 encounters) of groups contained less than 20 birds.

Group sizes differed significantly between species and seasons (Figure 3.4, Table 3.2).

Carnaby’s Cockatoos had significantly larger groups than FRTBC, but Baudin’s

Cockatoos group sizes were similar to FRTBC and Carnaby’s Cockatoos (Figure 3.4,

Table 3.2). Group sizes for all cockatoos combined were significantly larger in spring

than in autumn and winter (Figure 3.4, Table 3.2). Group sizes for Baudin’s Cockatoos

were particularly small in summer, suggesting they occurred in pairs or very small

flocks. Carnaby’s Cockatoo flocks were also smaller in winter and autumn. By

inspection, group sizes for FRTBC were constant across seasons, with means of less

than ten birds in all four seasons. The interaction between species and season was not

significant (Figure 3.4, Table 3.2).

 

  46  

Figure 3.4 Flock sizes of three species of black cockatoos seasonally at NBG and LEA, using data aggregated over the period 2007 – 2010. Error bars show standard errors. Table 3.2 Results of ANOVA for the data in Figure 3.4 after logarithmic transformation. Species refers to black cockatoos and season refers to: summer, autumn, winter and spring.

Factor df Effect MS Effect df Error MS Error F p-level

Species 2 0.348 263 0.113 3.081 0.048 Season 3 0.605 263 0.113 5.350 0.001 Species x season 6 0.227 263 0.113 2.012 0.064 Table 3.2 (continued) Tukey’s Honest Significant Difference (HSD) post hoc tests for unequal sample sizes to compare differences between flock sizes of black cockatoos. Significant differences are indicated with a *. Species Carnaby’s Cockatoo Baudin’s Cockatoo FRTBC

Carnaby’s Cockatoo - 0.70 0.03* Baudin’s Cockatoo - 0.09 FRTBC - Table 3.2 (continued) Tukey’s HSD post hoc tests for unequal sample sizes to compare differences between flock sizes of black cockatoos across seasons. Significant differences are indicated with a *. Season Summer Autumn Winter Spring

Summer - 0.06 0.24 0.45 Autumn - 0.10 <0.01* Winter - 0.04* Spring -

 

  47  

3.3.2 Habitat Use

Encounters with the three species were associated significantly with habitat types (

= 59.25, p < 0.001). While all three species were encountered mainly in native forest,

Carnaby’s Cockatoos and Baudin’s Cockatoos were encountered next most frequently

in mine-site rehabilitation areas, whereas FRTBC were not sighted in rehabilitation

areas (Table 3.3). Carnaby’s Cockatoos were the only species recorded within pine

plantations.

Evenness values suggested that Baudin’s Cockatoo was more generalist in habitat use

than Carnaby’s Cockatoo, which in turn was more generalist than FRTBC (Table 3.3).

Carnaby’s Cockatoo was significantly more diverse in its habitat use than FRTBC (t234

= 4.62, p < 0.001) and Baudin’s Cockatoo (t56 = 2.42, p < 0.001), but Baudin’s

Cockatoo and FRTBC were not significantly different (t64 = 0.94, p = 0.35). These

conclusions are unchanged if the critical value for each test is adjusted to 0.017 to allow

for multiple tests.

Table 3.3 The numbers of encounters with black cockatoos in different habitat types at NBG and LEA combined. Percentages are shown in parentheses. The Shannon diversity and evenness for the data are shown, with 95% confidence limits determined by bootstrapping.

Habitat type Carnaby’s Cockatoo Baudin’s Cockatoo FRTBC

Native Forest 54 (46) 20 (59) 87 (72) Native Woodland 15 (13) 4 (12) 13 (11) Modified Landscape 6 (5) 2 (6) 8 (7) Mine-site Rehabilitation 23 (19) 7 (21) 0 (0) RDA/WSR 3 (3) 0 (0) 1 (1) Sump 1 (1) 0 (0) 0 (0) Pine Plantation 14 (12) 0 (0) 0 (0) Paddock 2 (2) 1 (3) 11 (9) Remnant Vegetation 0 (0) 0 (0) 1 (1) Shannon H (± 95% CL)

1.546 (1.355-1.668)

1.16 (0.769-1.342)

0.954 (0.732-1.107)

Evenness (± 95% CL)

0.586 (0.517-0.728)

0.638 (0.513-0.847)

0.433 (0.364-0.637)

3.3.3 Food Plants

Sixteen native plant species at NBG within native forest and mine-site rehabilitation

habitats were used as food plants (Table 3.4 (a) and (b)). Feeding residues from Pinus

radiata were also observed in the pine plantation and occasionally within native forest

or rehabilitation areas close to pine stands.

  48  

Table 3.4 (a) Inventory of canopy-forming tree species that are food plant species used by black cockatoos at NBG and surrounds. The first column gives the common name and scientific name for the food plant species. The second column indicates whether a behavioural observation was made of black cockatoos feeding on this food plant within the study area. The habitat type(s) in which the behavioural observation(s) occurred is also indicated. The third column indicates whether feeding residues for the food plant were observed. The habitat type(s) in which the feeding residues occurred is also indicated, as is the form of residue observed. The middle three columns indicate which black cockatoo species we observed feeding on the food plant (during a behavioural observation); shading indicates that an observation occurred for that species. The last three columns indicate the habitat type(s) feeding on the food plant occurred, based on behavioural observations and feeding residues; shading indicates that an observation occurred for that species Codes: NF = Native Forest; RV = Mine-site Rehabilitation; PP = Pine Plantation; fc = fruit capsule; br = branch; fl = flower; sd = seed; fs = flower spike; sc = seed casing; co = cone; ? = black cockatoo species could not be determined from residue.

Food Plant Behavioural Observation Feeding Residue Baudin’s

Cockatoo Carnaby’s Cockatoo FRTBC Native

Forest Mine-site Rehabilitation Pine Plantation

Jarrah (Eucalyptus marginata)

NF

NF fc

Marri (Corymbia calophylla)

NF/RV

NF/RV/PP fc

Sheoak (Allocasuarina fraseriana)

NF

NF fc

Yarri (E. patens)

NF

NF fc

Montery/Radiata Pine (Pinus radiata)

NF/PP co

NF/PP/RV co

 

  49  

Table 3.4 (b) Inventory of proteaceous shrub species that are food plant species that are food plant species used by black cockatoos at NBG and surrounds. The first column gives the common name and scientific name for the food plant species. The second column indicates whether a behavioural observation was made of black cockatoos feeding on this food plant within the study area. The habitat type(s) in which the behavioural observation(s) occurred is also indicated. The third column indicates whether feeding residues for the food plant were observed. The habitat type(s) in which the feeding residues occurred is also indicated, as if the form of residue observed. The middle three columns indicate which black cockatoo species we observed feeding on the food plant (during a behavioural observation); shading indicates that an observation occurred for that species. The last three columns indicate the habitat type(s) feeding on the food plant occurred, based on behavioural observations and feeding residues; shading indicates that an observation occurred for that species Codes: NF = Native Forest; RV = Mine-site Rehabilitation; PP = Pine Plantation; fc = fruit capsule; br = branch; fl = flower; sd = seed; fs = flower spike; sc = seed casing; co = cone; ? = black cockatoo species could not be determined from residue.

Food Plant Behavioural Observation Feeding Residue Baudin’s

Cockatoo Carnaby’s Cockatoo FRTBC Native

Forest Mine-site Rehabilitation Pine Plantation

Parrot Bush Banksia sessilis NF/RV NF/RV

br ? ? ?

Pingle B. squarrosa RV NF/RV

br ? ? ?

Couch Honeypot B. dallaneyi FT

Bull Banksia B. grandis NF/RV NF

fs ? ? ?

Prickly Hakea Hakea amplexicaulis RV

br ? ? ?

Ram’s Horn H. cyclocarpa RV

br ? ? ?

Marble Hakea H. incrassata RV

br ? ? ?

 

  50  

Honeybush H. lissocarpha RV

br ? ? ?

Harsh Hakea H. prostrata RV NF/RV

br sc ? ? ?

Wavy-leaved Hakea H. undulata RV NF/RV

br sc ? ? ?

Two-leaf Hakea H. trifurcata RV

br sc ? ? ?

Variable-leaved Hakea H. varia

NF/RV br sc ? ? ?

 

  51  

Carnaby’s Cockatoos fed much more extensively on proteaceous shrubs (Banksia,

Hakea spp.) than the two other species, which fed mainly on Marri ( = 112.92, p <

0.001) (Table 3.5). Carnaby’s Cockatoos and Baudin’s Cockatoos were observed

peeling back bark to search for grubs. Evenness values suggested that Carnaby’s

Cockatoo was more generalist than the other species, which were similar. The diet of

Carnaby’s Cockatoo was significantly more diverse than that of FRTBC (t136 = 9.34, p <

0.001) and Baudin’s Cockatoos (t28 = 4.39, p < 0.001), but Baudin’s Cockatoos and

FRTBC were not significantly different (t31 = 0.28, p > 0.5). These conclusions are

unchanged if the critical value for each test is adjusted to 0.017, allowing for multiple

tests.

Table 3.5 Plants where black cockatoos were observed feeding at NBG and LEA combined. Percentages are shown in parentheses. The foods were flowers and seeds in most cases, but lifting bark to search for grubs is entered separately. Food plant Carnaby’s Cockatoo Baudin’s Cockatoo FRTBC

Banksia spp. 25 (34) 1 (4) 0 (0)

Hakea spp. 10 (14) 0 (0) 0 (0)

Jarrah 15 (20) 1 (4) 21 (30)

Marri 2 (3) 16 (70) 46 (65)

Pine 14 (19) 0 (0) 0 (0)

Forest Sheoak 1 (1) 0 (0) 1 (1)

Yarri 0 (0) 0 (0) 3 (4)

Grub (Wandoo) 1 (1) 4 (17) 0 (0)

Grub (Marri) 0 (0) 1 (4) 0 (0)

Shannon H (±95% CL)

1.789 (1.583-1.87)

0.966 (0.462-1.241)

0.835 (0.621-0.985)

Evenness (±95% CL)

0.748 (0.709-0.920)

0.525 (0.476-0.844)

0.576 (0.516-0.812)

Statistical tests for differences in Shannon H: Carnaby’s Cockatoo versus Baudin’s Cockatoo t28 = 4.39, p = 0.000145 Carnaby’s Cockatoo versus FRTBC t136 = 9.34, p < 0.0001 Baudin’s Cockatoo versus FRTBC t31= 0.28, n.s.

 

  52  

3.3.4 Site Occupancy

After corrections for GDS sampling intensity, there were annual fluctuations in GDS

detections of Carnaby’s Cockatoos and FRTBC at NBG (Table 3.6). Despite the low

sampling intensity in 2007, more Carnaby’s Cockatoos were seen than expected. For

FRTBC, there were fewer detections in 2010 than expected given the sampling

intensity. Within the LEA, Carnaby’s Cockatoos detections fluctuated significantly

across the years. In 2009, more Carnaby’s Cockatoos were seen than expected. The

sample size for Baudin’s Cockatoos was too small for analysis.

 

  53  

Table 3.6 The number of detections of three species of black cockatoo at sites sampled for the GDS within NBG tenements and the LEA in each year over the duration of the study. The number of samples in each year is given in parentheses in the year row. The chi-squared tests the null hypothesis of equal numbers of detections in each year against the alternative that detections varied with year. In calculating the expected values for the chi-squared tests, equal numbers of detections were assumed in each year and those values then adjusted for the sampling intensity. The expected values are given next to the observed in parentheses. For example, the expected number of detections of Carnaby’s Cockatoos at NBG in 2007 was calculated as 58 (number of samples in 2007)/904 (sum of all samples) x 21 (sum of all Carnaby’s Cockatoo detections over the study period). Location Cockatoo species 2007 (58) 2008 (439) 2009 (267) 2010 (140) Chi-squared test

NBG Carnaby’s Cockatoo 7 (1.4) 9 (10.4) 5 (6.0) 0 (3.2) = 25.96, p < 0.0001

NBG Baudin’s Cockatoo 0 3 1 0 Not calculated – sample size too small

NBG FRTBC 7 (3.8) 36 (27.8) 12 (16) 1 (8.4) = 12.63, p = 0.006

Location Cockatoo species 2007 (0) 2008 (165) 2009 (396) 2010 (231) Chi-squared test

LEA Carnaby’s Cockatoo - 3 (8.75) 32 (21) 7 (12.25) = 11.79, p = 0.003

LEA Baudin’s Cockatoo - 1 1 0 Not calculated – sample size too small

LEA FRTBC - 3 3 1 Not calculated – sample size too small

 

  54  

There were seasonal fluctuations in detections for Carnaby’s Cockatoos but not for

FRTBC (Table 3.7). Detections of Baudin’s Cockatoos were too few to examine for

seasonal variation, but they were not detected during summer GDS samples, although

present in the study area (based on non-GDS observations) (Figure 3.4). Detections for

Carnaby’s Cockatoos were highest in spring and summer. At NBG, FRTBC were

detected much more frequently than the other species, whereas within the LEA

Carnaby’s Cockatoos were detected most frequently (Tables 3.6 and 3.7).

Encounters (including non-GDS sightings) with Carnaby’s Cockatoos occurred in all

months. There were no encounters with Baudin’s Cockatoos recorded in October or

November and no encounters with FRTBC in June.

 

  55  

Table 3.7 The number of detections of three species of black cockatoo at sites sampled for the GDS within NBG tenements and the LEA in each season over the duration of the study. The number of samples in each season is given in parentheses in the top row. The chi-squared tests the null hypothesis of equal numbers of detections in each season against the alternative that detections varied with season. In calculating the expected values for the chi-squared tests, equal numbers of detections were assumed in each season and those values then adjusted for the sampling intensity. The expected values are given next to the observed in parentheses. Location Cockatoo species Summer (302) Autumn (247) Winter (209) Spring (146) Chi-squared test

NBG Carnaby’s Cockatoo 13 (7.01) 0 (5.74) 3 (4.86) 5 (3.39) = 12.31, p = 0.006

NBG Baudin’s Cockatoo 0 1 2 1 Not calculated – sample size too small

NBG FRTBC 20 (18.70) 16 (15.3) 12 (12.95) 8 (9.04) = 0.31, p = 0.96

Location Cockatoo species Summer (198) Autumn (198) Winter (198) Spring (198) Chi-squared test

LEA Carnaby’s Cockatoo 2 18 7 15 = 15.33, p = 0.002

LEA Baudin’s Cockatoo 0 0 2 0 Not calculated – sample size too small

LEA FRTBC 1 3 0 3 Not calculated – sample size too small

 

  56  

3.4 Discussion

3.4.1 Baudin’s Cockatoos

Group sizes for Baudin’s Cockatoos are variable and seasonally dynamic, with marked

differences between breeding and non-breeding periods (Johnstone and Kirkby 2008).

Johnstone and Kirkby (2008, 2009) report that the species, which is typically confined

to forest habitats, generally occurs in small flocks of up to 30 birds, and occasionally in

larger flocks of up to 50 or aggregations of c. 1 200 birds. Our finding of a mean group

size of 14.5 birds is similar to the 13 ± 2 Baudin’s Cockatoos reported by Weerheim

(2008), who reported group sizes ranging from one to 100 birds.

Johnstone and Kirkby (2008) report that, depending on their provenance, Baudin’s

Cockatoos may be year-round residents or postnuptial nomads and migrants, with the

majority of the population shifting northwards in autumn to the central and northern

Jarrah-Marri forest and adjacent areas of the Swan Coastal Plain, before migrating

southwards in spring to breeding areas in the Karri forest and southern Jarrah-Marri

forest. Johnstone and Kirkby (2008) suggest that the main wintering area for Baudin’s

Cockatoos is concentrated in the western portion of the northern Jarrah-Marri forest.

Weerheim (2008) reported Baudin’s Cockatoos to be absent or scarce in the western

Jarrah-Marri forest in October, December, and February. Johnstone and Kirkby (2008)

also report Baudin’s Cockatoos to be absent or scarce at roost sites near Perth between

October and January.

Though we detected few Baudin’s Cockatoos, they were less common in summer than

during autumn, winter, and spring. The birds seen in summer could have been non-

breeding juveniles or adults or breeding pairs. Although Boddington lies outside the

main breeding areas for the species, reports of breeding in the Wungong catchment 140

km northwest of Boddington (Johnstone and Kirkby 2009) suggest that breeding may

occur within the study area.

Baudin’s Cockatoos mainly occur in native forest and woodland at NBG, but also use

mine-site rehabilitation. Lee et al. (2010) found that Baudin’s Cockatoos feed on Marri

in rehabilitation at NBG within eight years of establishment. Though Marri appears the

main food source for Baudin’s Cockatoos at NBG, our list of food plants at NBG

probably excludes some food plants used by the species. For example, Baudin’s

 

  57  

Cockatoos feed on the flower spikes of Bull Banksia (Johnstone and Kirkby 2008), and

thus some Bull Banksia feeding residues observed at NBG may be from this species.

However, Bull Banksia flowers between September and November at NBG (H Finn,

Murdoch University, pers. obs.), when Baudin’s Cockatoos are absent or scarce.

3.4.2 Carnaby’s Cockatoos

We found groups of up to 90 Carnaby’s Cockatoos, averaging 10.1 birds, similar to

Weerheim (2008) who reported an average group size of 11 ± 2 birds (n = 34 sightings)

in the western Jarrah-Marri forest (range two to 60 birds). Finn et al. (2009) reported

larger foraging groups of 117.3 ± 28.1 (range 3 to 1 785) in Gnangara pine plantations

north of Perth on the Swan Coastal Plain. Similarly, Saunders (1980) reported feeding

groups between Perth and Moore River varying from 2 to 1 200 birds with a mean

group size of 129.3. Johnstone and Storr (1998) noted that, across their range,

Carnaby’s Cockatoos typically occur in pairs or small flocks, though sometimes in

flocks of up to 2 000 birds during the non-breeding season and in pine plantations.

Carnaby’s Cockatoos fed on the broadest range of plants (at least ten species), agreeing

with previous studies (Saunders 1974a,b; Saunders 1980; Shah 2006; Valentine and

Stock 2008). Saunders (1974b) observed that the crops of Baudin’s Cockatoos

contained mostly Marri seeds (in 89% of crops examined) but no pine seeds (0% of

crops), whereas crops of Carnaby’s Cockatoos contained mostly pine seeds (81% of

crops), followed by seeds from ‘Banksia’ spp. (20% of crops) and Hakea spp. (19% of

crops), but rarely Marri seeds (8% of crops). The breadth of the diet of Carnaby’s

Cockatoos at NBG is reflected in the finding that no food plant accounted for more than

20% of feeding observations. Nevertheless, Jarrah is an important food source for

Carnaby’s Cockatoos at NBG.

Carnaby’s Cockatoos used the broadest range of habitats (i.e. eight of the nine habitat

types), including two anthropogenic habitats – mine-site rehabilitation and pine

plantation – where they fed on proteaceous shrubs and pine (respectively). This agrees

with previous studies documenting the use of novel habitats and food sources by this

species (e.g. Saunders 1974a, b; Saunders 1980; Finn et al. 2009; Lee et al. 2010).

Carnaby’s Cockatoos were most common at NBG in summer and in the LEA in autumn

and spring. The peak during summer at NBG may reflect flocks migrating towards the

 

  58  

coast at the end of the breeding season (Saunders 1980; Johnstone and Storr 1998;

Johnstone and Kirkby 2009; Johnstone et al. 2010). The availability of reliable water

and food sources at NBG (e.g. proteaceous shrubs in mine-site rehabilitation areas) may

make NBG attractive to migrating flocks. The peak at LEA during autumn and spring

may also reflect seasonal movement, but abundance may be closely linked to

availability of mature pinecones in the plantation along the eastern edge. The strong

association of this species with pine probably accounts for it being observed within the

LEA more frequently than the other black cockatoos. The factors affecting occupancy

are also likely to explain seasonal variation in group size.

Like Weerheim (2008), we observed Carnaby’s Cockatoos throughout the year,

suggesting year-round residency for some individuals. However, Weerheim (2008)

found that Carnaby’s Cockatoos were the least abundant black cockatoo in the western

Jarrah-Marri forest. This contrasts with our findings that Carnaby’s Cockatoos were the

most common black cockatoo in the LEA and the second-most common species at NBG

(after FRTBC). This suggests potential regional differences in the distribution of

Carnaby’s Cockatoos, and may also reflect differences in movements of migratory

flocks and the availability of non-native foods. Canola, an important Wheatbelt food

source, occurs in farms along the eastern margin of the Jarrah-Marri forest and birds

have been observed feeding on canola 20 km to the southeast of NBG (H Finn,

Murdoch University, unpublished data). Johnstone and Kirkby (2009) reported small

numbers of Carnaby’s Cockatoos year-round in the Wungong catchment, as well as

several breeding pairs. This report, along with our observation of breeding pairs at

NBG, suggests that Carnaby’s Cockatoos may breed throughout the Jarrah-Marri forest,

at least in low numbers. If confirmed, this would emphasise the significance of the

Jarrah-Marri forest as breeding habitat.

3.4.3 Forest Red-Tailed Black Cockatoo

We found FRTBC mostly in small groups (mean of 7.4 birds), similar to Johnstone and

Kirkby (1999), who reported that FRTBC generally occurred as breeding pairs or in

small family units of 3 to 5 birds, though groups of up to 200 were occasionally

observed. Saunders (1977) reported that, in forested areas of the southwest, FRTBC

were most commonly in groups of less than ten, rarely larger than twenty. The largest

FRTBC flocks observed by Ford (1980) were 20 to 30 birds. Abbott (1998) reviewed 66

published records of FRTBC from 1900 - 1995 and found a mean number of 9.6 ± 1.5

 

  59  

cockatoos. He also reported a mean of 6.6 ± 1.5 for FRTBC sighted during two

extensive regional surveys in southwestern Australian (in 1995 - 1996 and 1996 - 1997).

Weerheim (2008) reported an average of 4 ± 0.2 FRTBC (n = 585 sightings), with

groups ranging from one to 50 birds.

FRTBC mostly use native forest and woodland at NBG. Paddocks are also important, as

noted elsewhere (Abbott 1998). Paddocks adjacent to NBG are used for feeding and

drinking, with the largest group at NBG being 45 birds at drinking site in a paddock (H

Finn, Murdoch University, pers. obs.).

Water in farm dams and near mine-sites may be important for FRTBC at NBG. FRTBC

have the highest basal metabolic rate and evaporative water loss of the four

Calyptorhynchus species occurring in southwestern Australian (including the Inland

Red-tailed Black Cockatoo, C. b. samueli), so water may determine their distribution

more than for Baudin’s or Carnaby’s Cockatoos (Cooper et al. 2002).

Although we did not observe FRTBC in mine-site rehabilitation, feeding residues from

this species are present in rehabilitation areas at NBG, indicating that some feeding

activity occurs (Lee et al. 2010). Weerheim (2008) also found that FRTBC were

infrequent in mine-site rehabilitation less than 20 years old at sites in the western

Jarrah-Marri forest. Weerheim (2008) and Lee et al. (2010) suggested that FRTBC may

avoid mine-site rehabilitation at an early successional stage because of perceived

predation risk from raptors. Observations of Baudin’s Cockatoos feeding on Marri

indicate that suitable food sources for FRTBC are available in rehabilitation (Lee et al.

2010).

FRTBC at NBG fed mostly from Marri and Jarrah, and less often on Forest Sheoak and

Yarri. This agrees with Johnstone and Kirkby (1999) and Weerheim (2008), who

reported that Marri and Jarrah are the main foods for FRTBC, although they may feed

on other plants with suitable energy content or nutritive value, particularly if Marri and

Jarrah are limited (Johnstone and Kirkby 1999; Cooper et al. 2002).

FRTBC are more sedentary than the other species (Abbott 1998; Johnstone and Storr

1998; Johnstone and Kirkby 1999; Johnstone et al. 2010). Nevertheless, movements

away from areas burned or lacking water are reported (Abbott 2001; Johnstone and

 

  60  

Kirkby 2009; DSEWPaC 2011), and birds range onto the Swan Coastal Plain to use

native and exotic food sources (e.g. Cape Lilac Melia azedarach) (Johnstone et al.

2010). Johnstone and Kirkby (1999; 2009) and Weerheim (2008) reported FRTBC as

present year-round in the Jarrah-Marri Forest.

Inter-annual changes of FRTBC in site occupancy at NBG occurred, but not seasonal

changes in occupancy, or seasonal changes in group size. FRTBC appeared to be more

locally abundant in 2007 and 2008 than in subsequent years. This may reflect

landscape-level patterns in food availability for Marri, as Biggs et al. (2011) reported

intensive feeding on Marri by FRTBC at NBG based on residues observed in 2008.

FRTBC are resident year-round at NBG, but their distribution shifts periodically,

probably in response to the availability and depletion of food and, in summer, to water.

Landscape-scale differences in the flowering and fruiting of Marri or Jarrah could

change distributions (Johnstone and Kirkby 1999; Weerheim 2008). We did not see

large transient or migratory flocks of FRTBC. The larger flocks we observed were

usually temporary aggregations at drinking sites, often as dusk. FRTBC may migrate

into and out of the study area, but as small flocks and without a marked seasonal

component.

3.4.4 Management Implications

This study illustrates how, within a single landscape, these three black cockatoos exhibit

distinct (though often overlapping) ecologies, including differences in group size,

habitats and food plants used, and occupancy (seasonal or year-round). These

differences suggest they will differ in their response to disturbance and use of

anthropogenic food sources such as pine, mine-site rehabilitation, and paddock trees.

Of the three species, FRTBC appear the most vulnerable to disturbance from intensive

activities such as mining and forestry because they tend to reside year-round within

defined locales in the Jarrah-Marri forest. FRTBC shift their distribution locally in

response to the availability of food and water, and are likely to do so in response to

local disturbance. However, populations of this species appear to be small (i.e. <150

individuals) and resident (Johnstone and Kirkby 1998), suggesting the potential for

local extinctions if local carrying capacities are greatly reduced (Wardell-Johnson et al.

2004). In contrast, Carnaby’s Cockatoos use more foods and habitats, and therefore are

 

  61  

better able to adapt to local disturbances, either by migrating elsewhere or by shifting to

other food plants. The abundance and distribution of Baudin’s Cockatoos is linked to

the availability of Marri, so retaining or restoring Marri is critical for this species.

Breeding success for black cockatoos depends on adequate supplies of suitable hollows

and sufficient food and water within a few kilometers from the nest (Saunders et al.

1985; Johnstone et al. 2010). Leaving aside hollow availability, an issue debated

elsewhere (e.g. Mawson and Long 1994; Stoneman et al. 1997; Abbott 1998; Calver

and Dell 1998; Abbott and Whitford 2002; Wardell-Johnson et al. 2004), this study

suggests that the availability of anthropogenic foods (e.g. pine, proteaceous shrubs in

rehabilitation areas) may support breeding by Carnaby’s Cockatoos within the Jarrah-

Marri forest. However, as FRTBC do not feed on pine and infrequently within young

rehabilitation, efforts to support FRTBC breeding are better focused on in-situ

conservation of native vegetation. However, the association between FRTBC and

paddock trees suggests that they may benefit from restoration that allows Marri stems to

obtain large canopy volumes.

Climate change has important implications for the conservation of black cockatoo

habitat in the Jarrah-Marri forest, because the lower rainfall predicted will affect the

productivity and species composition of vegetation (Charles et al. 2010). Climate-

related changes may be most pronounced in areas such as NBG, which lie along

existing ecotones where forest transitions to woodland.

We suggest two strategies to ensure that suitable feeding habitat is maintained for all

three black cockatoos. Firstly, restoration and rehabilitation efforts should continue to

establish a range of native food plants, including both proteaceous and myrtaceous

species. A diversity of food plants adds resilience to restored sites and also helps to

maintain food availability across seasons and years. The current restoration aims of

restoring a functioning Jarrah-Marri forest environment are compatible with this aim

and appear to be achieving it. Secondly, consideration may need to be given to areas

where restoration of Jarrah-Marri forest may be impractical in the face of climate

change. In such areas, and where this does not conflict with other restoration aims, non-

native (e.g. pine) and non-endemic (e.g. species from lower rainfall regions) could be

included in restoration. Several non-endemic Australian and exotic plants are used

intensively by black cockatoos where they occur in Western Australia (e.g. Cape Lilac

 

  62  

Melia azedarach, Lemon-scented Gum Corymbia citriodora, Liquid Amber

Liquidamber styraciflua) (Saunders 1980; Johnstone and Kirkby 1999; Kenneally

2002). Concerns over the use of non-endemic species for restoration of disturbed sites

may need to be reconsidered given the implications of changing rainfall for the region.

The inclusion of plant species adapted to lower rainfall conditions may be particularly

important in ensuring long-term food availability and in mitigating the short-term

impacts of drought. Such an approach may be appropriate when revegetating farmland,

as opposed to native forest restoration.

63

Chapter 4 Methods to Assess Breeding Habitat for Black

Cockatoos in the Jarrah-Marri Forest of Southwestern Australia

4.1 Introduction

Availability of suitable breeding habitat, especially tree hollows, is important for the

conservation and recovery of Carnaby’s Cockatoo, Baudin’s Cockatoo and FRTBC

(Saunders and Ingram 1987; Abbott 1998; Cale 2003; Chapman 2008). This emphasises

the need for appropriate methodologies to assess breeding habitat quality and nesting

activity. Here I examine the efficacy of different field survey techniques to identify (a)

potential breeding hollows and (b) probable nest sites, using data associated with the

clearing of 1 203 ha of Jarrah-Marri forest in southwestern Australia.

Similar to other arboreal fauna, researchers can collect data on the presence of potential

breeding hollows and of probable nest sites for black cockatoos using destructive (e.g.

post-felling inspection of hollows) or non-destructive approaches (e.g. McComb et al.

1994; Nelson and Morris 1994; Whitford 2001, 2002b; Whitford and Williams 2002;

Fierke et al. 2005; Cameron 2006). Non-destructive approaches may involve: ground-

based surveys to locate potential hollows; inspection of hollows in situ (e.g. using

elevated work platforms); behavioural observations of animals near potential nesting

areas; capturing, tagging, and tracking of animals; remote observational techniques (e.g.

camera traps, pole-cameras, temperature sensors); or ‘tree-knocking’ (i.e. an approach

in which an observer knocks at the base of a tree and watches to see if an animal

emerges from the hollow to investigate the disturbance) (e.g. Lindenmayer et al. 2000;

Gibbons and Lindenmayer 2002; Harper et al. 2004; Koch 2008; Cawthen et al. 2009).

Ethical, logistical, and practical limitations will influence what methodologies are

suitable or cost-effective.

There has been some discussion in the scientific literature about the effectiveness of

different techniques for assessing breeding habitat quality and nesting activity for black

cockatoos in southwestern Australia (e.g. Wardell-Johnson et al. 2004). The recovery

plans for these species also recognise the need for regulatory assessment of the potential

impact of on-ground works upon black cockatoo habitat (e.g. land clearing, burning)

(Cale 2003; Chapman 2008). However, there remains no specific regulatory guidance

64

for appropriate habitat assessment methodologies for black cockatoos, as exists for

other nationally threatened fauna in this region (e.g. Western Ringtail Possum

Pseudocheirus occidentalis (Commonwealth of Australia 2009b)).

In this chapter I examine ground-based hollow surveys, post-felling inspections of

hollows, and behavioural observations as approaches for assessing black cockatoo

breeding habitat in the Jarrah-Marri forest and other areas of southwestern Australia. In

particular, I evaluate the effectiveness of: (a) ground-based surveys and post-felling

inspections as methods to identify potential nest hollows, and (b) all three methods as

techniques to identify probable nest hollows. The data were collected by a range of

observers at a mine-site in the Jarrah-Marri forest over a six-year period (2005 - 2010),

and have certain limitations reflecting the opportunistic way in which they were

collected and the lack of an overall coordinator implementing a systematic plan. I use

these limitations to illustrate several methodological considerations that are instructive

for further work, and also to suggest protocols for assessing breeding habitat for black

cockatoos in Western Australia. The work complements the detailed work assessing the

distribution and occurrence of hollows across different ages and management tenures in

the Jarrah-Marri forest (Whitford 2001, 2002b; Whitford and Williams 2001, 2002;

Johnstone et al. 2013a, 2013b) with site-specific data. It also examines data from

procedures normally followed by environmental consultants conducting surveys for the

presence of habitat trees and breeding hollows as part of environmental impact

assessments.

4.2 Methods

4.2.1 Study Area Characteristics

An expansion project at the NBG mine required clearing 1 203 ha of Jarrah-Marri forest

over the period 2005 - 2010. The expansion area contained 955 ha of open forest or

woodland in which Jarrah and Marri were the dominant canopy species, and 224 ha

open woodland in which Wandoo (E. wandoo), Yarri (E. patens – Blackbutt), and/or

Flooded Gum (E. rudis) were the dominant tree species (an additional 24 ha were

heath/shrubland). Jarrah and Marri occur at a ratio of 3.2 : 1 at the NBG site (Biggs et

al. 2011), which is similar to other published stem ratios for the Jarrah-Marri forest. For

example, Whitford (2002b) reported the frequency ratio of Jarrah to Marri as

65

approximately 2 : 1 through the Jarrah-Marri forest, and Abbott (1998) cited

unpublished forest inventory data indicating that Marri accounts for between 16% of

stand basal area in the northern Jarrah-Marri forest and 33% in the southern Jarrah-

Marri forest. Site-specific logging history is given in Chapter 2, while site-specific data

on tree species’ abundance and distribution are given in Biggs et al. (2011).

Environmental impact assessment for the expansion project identified impacts on black

cockatoos as an important focus for environmental management. Carnaby’s Cockatoos,

Baudin’s Cockatoos, and FRTBC were all observed year-round in the study area (Lee et

al. 2010), but only Carnaby’s Cockatoos and FRTBC nest locally (Johnstone et al.

2010; H Finn, Murdoch University, unpublished data).

4.2.2 Ground-based Surveys

From 2005 – 2010, ground-based surveys were conducted to identify potential nest

hollows within 1 203 ha designated for clearing. Although precise records are not

available, I estimate that between 400 and 500 man-hours were expended on ground-

based surveys over this period, suggesting a search time of 2.4 - 3.0 ha of forest

searched per man-hour. No systematic data were collected on factors influencing search

times, but terrain, stand structure, number of potential hollows observed, and the detail

of the data collected were likely to have been important.

Ground-based surveys were conducted throughout the year (i.e. without regard to

potential breeding seasons) and generally between 0800 – 1600 hours. The overall study

area was surveyed incrementally over time, with the size of the areas designated for a

particular ground-based survey varying from <1 ha to >100 ha. Ground-based surveys

were conducted by applying a grid-based search pattern to the survey area, with the

lines of the grid spaced 50 - 100 m apart depending on stand density. Four individuals

conducted the majority of the ground-based surveys over the six years of the study.

Based on descriptions of black cockatoo nest hollows in southwestern Australia

(Saunders 1979; Saunders et al. 1982; Mawson and Long 1994; Johnstone and Storr

1998), potential nest hollows were selected by: (a) an opening at least horizontal relative

to the ground surface (i.e. not opening downwards); (b) an entrance appearing ≥12 cm

across; and (c) space for a hollow interior of ≥25 cm. Personnel also considered

indicators of senescence, decay, fire, and storm damage, such as decaying wood

66

structure, dead branches, and dead wood below the assumed hollow entrance, and the

presence of chew marks (because used nest hollows are generally chewed around the

entrance).

Given the lack of one overall coordinator for the entire study period, no systematic

criteria were adopted to reduce bias in hollow identification across observers (e.g.

Harper et al. 2004; Munks et al. 2007). This is a significant methodological limitation.

An additional limitation is the likelihood that ground-based surveys may not detect

some potential hollows (e.g. because hollows were obscured by vegetation or occurred

at angles making them difficult to observe from the ground). While I did not collect any

data relating to the prevalence of undetected hollows (e.g. by examining felled trees in

which no hollows were observed from the ground), I suggest that the number of

undetected hollows is likely to have been small because of the following characteristics

of vegetation at the study area: (a) most Jarrah, Marri, and Wandoo trees are less than

20 m high (Biggs et al. 2011); (b) individual tree canopies tend to be diffuse, allowing

clear observations of branches and trunks; and (c) stand densities are low enough to

support a discontinuous upper canopy layer.

I emphasise that the objective of ground-based surveys was to identify potentially

suitable hollows for black cockatoos, not to assess nesting activity. As such, protocols

for behavioural observations of black cockatoos were not included in methodologies for

ground-based surveys, and no consideration was given to whether these surveys were

conducted at a time of day (e.g. dawn/dusk) or year (e.g. breeding season) potentially

allowing for observations of nesting activity. Tree-knocking to assess hollow occupancy

was not included as an element of the ground-based survey methodology until 2009 (see

below). In retrospect, these factors effectively precluded many opportunities for

undertaking behavioural observations and/or assessing hollow occupancy during

ground-based surveys, and represent an important methodological limitation. As a

result, inferences drawn from the data are cautious. I have not attempted to estimate

hollow densities, for example, because these may be biased by the factors described

above.

4.2.3 Post-felling Inspection

Trees identified during ground-based surveys as having potential nest hollows were

marked to allow for inspection of hollows after trees were felled. During felling

67

operations, an observer accompanied the harvesting operator. This observer examined

each marked tree and identified the location of the potential hollow(s) prior to the tree

being felled. This generally allowed the harvesting operator to fell trees so that hollows

were accessible for inspection, or to manipulate felled trees when hollows were

obscured.

Hollows that survived felling were then inspected to determine if a large hollow was

present, and to look for evidence of hollow use by black cockatoos or other species (e.g.

feathers, fur, eggshells, carcasses, faeces). Another methodological limitation of this

study was that no systematic criteria were applied for measuring the dimensions of

hollows or for characterising their structural features, as has been implemented in other

studies (Whitford 2001, 2002b). Instead, observers classified a hollow as ‘large’ if it

could contain an unfolded A4 sheet of paper (210 mm × 297 mm). Some hollows were

damaged during felling, precluding inspection of contents and measurements. Thus,

another potential methodological limitation for this study is the possibility for some bias

to occur between tree species in their propensity to be damaged during felling.

4.2.4 Behavioural Observations

From 2008 - 2010, I undertook a long-term study of black cockatoo ecology at the NBG

mine (Chapter 3 of this thesis). This study provided opportunities to conduct

behavioural observations that led to the location of several probable black cockatoo nest

hollows at the site. As these observations were conducted on both an opportunistic (e.g.

during the course of research activities) and systematic (i.e. with the intention of

locating nest hollows) basis, it is difficult to determine how many man-hours were

expended in the effort of locating a probable nest hollow. To obtain a broad estimate, I

allocated an entire day of survey effort to each instance of locating a probable nest

hollow (n = 10 probable nest hollows × 12 man-hour survey day = 120 man-hours). I

urge caution with this estimate as, in my experience, the amount of effort required to

locate a nest varies with the experience and local knowledge of the observer, and with

other factors such as time of day and time of year. In the most opportune cases, a

probable nest hollow may be located within a matter of minutes; in others, it may take

repeated observations over several days.

Behavioural observations leading to the location of a probable nest hollow involved

three basic approaches: (a) ground-based observers knocking at the base of a tree

68

identified as having a potentially-suitable hollow and observing to see if a nesting black

cockatoo emerged to investigate the disturbance (‘tree-knocking’) (T Kirkby, Western

Australian Museum, 2009, pers. comm.); (b) hearing or observing black cockatoos

roosting in a nest tree (or in nearby tree); or (c) observing black cockatoos (e.g. at a

drink site) and visually tracking them back to the nest tree. Observations of birds

entering or leaving a hollow indicated whether a tree was likely to be used for nesting.

Inference was based on birds remaining in the hollow for some time (generally over

multiple observations), rather than briefly occupying a hollow, as pairs sometimes do

when inspecting hollows. I did not climb trees to confirm nesting because of

occupational health and safety concerns; this represents an important methodological

limitation.

4.3 Results

4.3.1 Ground-based Surveys

Ground-based surveys identified 149 trees as having potentially suitable hollows. Most

occurred in Jarrah and lesser numbers in Marri and Wandoo (Table 4.1). Other tree

species were encountered infrequently and were not observed with potential hollows.

Observers during ground-based surveys noted obvious indications of probable hollow

use by nesting by black cockatoos only once, when Carnaby’s Cockatoos were seen

around a hollow. Observers also observed a pair of Carnaby’s Cockatoos inspect a

hollow on one other occasion (Figure 4.1).

69

Table 4.1 Results of post-felling inspections of trees identified as having potentially-suitable hollows during ground-based surveys (GBS). Percentages in column 1 are relative to the overall number of trees identified (i.e. n = 149); percentages in columns 2 relative to column 1 (i.e. species-specific totals for trees identified); and percentages in columns 4 and 5 relative to column 3 (i.e. species-specific totals for trees with hollows intact after felling).

Tree species

1 Identified during GBS

2 Damaged in felling

3 Intact after felling

4 Large hollow present

5 Large hollow absent

Jarrah 110 (73.8%) 21 (19.1%) 89 28 (31.5%) 61 (68.6%)

Marri 26 (17.4%) 8 (30.8%) 18 10 (55.6%) 8 (44.4%)

Wandoo 13 (8.7%) 1 (7.7%) 12 8 (66.7%) 4 (33.3%)

Total 149 30 119 46 (38.7%) 73 (61.3%)

70

´

1

2

3

4

6

7

8

9

10

11

1

2

3 4

56

7

8

9

10

1112

13

14

15

16

18

19

21

22

23

24

435000

435000

440000

440000

445000

445000

450000

450000

455000

455000

460000

460000

63

75

00

0

63

75

00

0

63

80

00

0

63

80

00

0

63

85

00

0

63

85

00

0

63

90

00

0

63

90

00

0

COCKATOO NESTING

NEWMONT BODDINGTON GOLDEWMONTNASIA PACIFIC

0 1 2

Kilometers

Cockatoo Nesting

Cockatoo Hollows

Scale: 1:70 000 Projection: MGA94 Zone 50

Figure 4.1 Locations of black cockatoo nesting and hollows identified in the study site.

71

4.3.2 Post-felling Inspection

20.1% (n = 30) of trees were too damaged during felling to inspect. Post-felling

inspection of the remaining 119 found that most potential hollows identified during

ground-based surveys were not ‘large’ hollows (Table 4.1). Whether a large hollow was

present or absent was significantly associated with tree species (

χ22 = 8.08, p = 0.018),

with large hollows present in 31.5% of Jarrah trees (n = 28 of 89 felled trees), 63.6% of

Marri trees (n = 14 of 22 felled trees) and 69.2% of Wandoo trees (n = 9 of 13 felled

trees).

No hollow inspected showed signs of occupancy by black cockatoos. Two hollows

classified as ‘large’ showed evidence of use by owls, one contained a dead bat, and one

contained a beehive. Of hollows classified as ‘not large’, one contained a dead parrot

and a second contained parrot feathers (species not determined).

4.3.3. Behavioural Observations

Observations of black cockatoos using behavioural cues for nesting activity identified

ten potential nest hollows, eight for Carnaby’s Cockatoos (n = 6 Marri, n = 2 Wandoo)

and three for FRTBC (n = 3 Wandoo) (Figure 4.1). These observations occurred over

eight study days (n = 3 hollows over 3 days in 2008; n = 6 hollows over 4 days in 2009;

n = 1 hollow over 1 day in 2010). Nine observations involved some kind of visual or

acoustic cue indicating the potential presence of a nest site; in all cases hollow

occupancy was checked by tree-knocking. In one instance, hollow occupancy was

determined by the opportunistic tree-knocking of a tree considered to have a potentially

suitable hollow. Of the ten potential nest hollows identified through behavioural

observations, seven (70%) were identified while I was undertaking field research that

included nest site identification as an objective. The remaining three potential hollows

(30%) were identified opportunistically based on observations made while undertaking

other activities, such as driving along the mine-site access road or sampling vegetation.

All potential hollows were identified on the basis of an initial set of observations

conducted over a few minutes (e.g. seeing a potentially suitable tree and knocking at the

base of that tree) to a few hours (e.g. hearing and tracking birds as they moved to/from a

nest site). Subsequent observations were, however, necessary to characterise use of the

hollow (i.e. were birds only investigating the hollow or staying in the hollow over time).

72

4.4 Discussion

The findings of this study are in two areas: the availability of potential nest hollows and

the probable use of hollows by nesting black cockatoos. The study’s methodological

limitations make comparisons of techniques problematic, and preclude some

quantitative assessments such as estimates of hollow density across the landscape.

However, the data do provide an opportunity to review suitable study designs and

suggest best practice approaches in relation to regulatory requirements and further

research.

4.4.1 Methodological Implications

Ground-based surveys overestimate the incidence of potential nest hollows

Ground-based surveys avoid many costs, logistical requirements, and safety concerns

(Gibbons et al. 2000; Lindenmayer et al. 2000; Gibbons and Lindenmayer 2002), but

hollows may be difficult to locate and their interiors cannot be observed or measured

(Lindenmayer et al. 2000; Whitford 2002b; Harper et al. 2004; Munks et al. 2007; Koch

2008). This study, like Whitford (2002b), indicates that ground-based hollow surveys

overestimate hollow abundance, especially in Jarrah. Whitford (2002b) used post-felling

inspection to assess the accuracy of ground-based surveys for hollows, and found that

ground-based surveys greatly overestimated the actual abundance of hollows, largely

because of high prevalence rates for ‘false’ hollows in Jarrah.

More rigorous selection criteria (Lindenmayer et al. 2000; Harper et al. 2004; Munks et

al. 2007; Koch 2008; Rayner et al. 2010) may reduce, but not eliminate, problems

caused by difficulty in locating hollows and the impossibility of confirming internal

dimensions from the ground. However, ground-based surveys can provide useful

assessments of relative hollow occurrence and abundance, as long as the biases are

identified and accounted for across areas surveyed and observers used (e.g. Harper et al.

2004).

Ground-based surveys are not effective at identifying probable black cockatoo nest

hollows unless they include targeted behavioural observations and are conducted

during likely breeding periods.

Ground-based surveys identified only one nest site despite substantial survey effort and

the large area (1 203 ha) that was surveyed. Given that I located probable nest sites

73

elsewhere in the study area, nests for Carnaby’s Cockatoo and FRTBC were probably

present in the area searched by ground-survey, at least in low densities. The failure of

ground-based surveys to identify probable nesting activity likely reflects: (a) the failure

to include behavioural observations and assessments of hollow occupancy (tree-

knocking) as objectives for ground-based surveys, and (b) the fact that surveys were not

conducted at the most appropriate times of day (e.g. at dawn and dusk) or times of the

year (i.e. during breeding seasons). I suggest that the value of ground-based surveys as

an assessment approach is significantly enhanced if attention is given to these

considerations.

Post-felling inspections assess hollow availability but not actual use

Post-felling inspections are valuable in quantifying misidentification rates for ground-

based estimates and the factors associated with the scoring of real versus false hollows

(Harper et al. 2004; Koch 2008). Hollows can also be measured to confirm suitability

(Abbott and Whitford 2002; Whitford and Williams 2002; Wardell-Johnson et al.

2004), although damage or total destruction during felling may lead to underestimation

of the number of hollows present or to biased or inaccurate measurements of hollow

size and characteristics. I did not address the prevalence of undetected hollows in this

study, but I do consider it likely to be small, given that characteristics of canopy

vegetation at this site allow for clear visibility of branches and trunks and the height of

most trees is less than 20 m (see Methods).

Disturbance during felling may also displace or obscure residues from breeding birds.

The contents of most Jarrah and Marri hollows shifted during felling. This likely

explains why observers found no evidence of occupancy within any felled hollow, with

the caveat that nesting activity may not always leave obvious physical signs and

therefore other methods (e.g. genetic sampling of fecal source material) might be

appropriate (Long et al. 2008).

Methodological limitations may limit the value of opportunistic studies

The opportunistic study involved several observers operating with some independence,

raising limitations in terms of: inter-observer bias; few measurements of hollows in

felled trees; possible differential damage to hollows in different tree species during

felling; no climbing of trees to confirm nesting; the use of ad hoc bird observations; and

ground-based surveys outside the breeding season missing opportunities to observe

74

birds using hollows. Safety considerations are an important restriction on climbing trees

and there is probably little that can be done regarding differential damage to hollows in

different species of trees. However, the other biases can be overcome with advanced

planning and valuable data on hollow abundance in different areas of the Jarrah-Marri

forest could be obtained with collaboration between organisations involved with the

harvesting or clearing of forest areas.

Behavioural observations are the most effective method for locating nest sites

Behavioural observations located ten potential nest hollows at the study area. While I

have estimated search effort at 120 man-hours, this includes a combination of

opportunistic and systematic study. Greater efficiency could be gained by ensuring that

field observations are conducted at appropriate times. A key point in this argument is

that suitable nest hollows may be limited. If they are not and cockatoo populations are

low, then behavioural observations would be a measure of population size more than of

hollow availability. A conclusion of potentially limited hollows seems precautionary

from an environmental impact perspective (Calver et al. 1999), and accords with the

concern of Johnstone et al. (2013a, 2013b) about hollow availability for black

cockatoos.

I suggest that the most effective technique to identify black cockatoo nest sites in the

Jarrah-Marri forest is behavioural observations conducted: (a) at dawn and dusk; (b)

during known breeding periods, which should be defined broadly to allow for annual

variation in response to climatic variation; and (c) with the aim of identifying birds

travelling to and from a nest. Flight paths can then be re-traced to the nest, with contact

calls between birds providing a directional cue. Once a general area is established, ‘tree-

knocking’ can be used to test whether potential hollows are occupied. While I

acknowledge that this approach can be time-intensive and the effectiveness of (and

potential biases associated with) ‘tree-knocking’ are not known, it has proved effective

identifying probable nest sites at the NBG site (and elsewhere, e.g. R Johnstone and T

Kirkby, Western Australian Museum, 2009, pers. comm.). Consideration should

therefore be given to the application of this approach whenever information is needed

on black cockatoo nesting activity (e.g. for environmental impact assessments). This

necessarily implies certain planning requirements, particularly that observations be

conducted during breeding seasons.

75

4.4.2 Management Implications

Conservation of species other than Jarrah is important

The findings suggest that, in some Jarrah-Marri forest landscapes, mature and senescent

Marri and Wandoo may be disproportionately valuable as trees bearing large hollows.

Marri and Wandoo accounted for a disproportionate number of the large hollows

identified in this study, even though Jarrah predominated at the site. This finding might

apply elsewhere because, while this study was limited to one location, it did include

vegetation types and landforms that are ubiquitous within the northern Jarrah-Marri

forest. The disproportionate value of Marri and Wandoo as breeding habitat for black

cockatoos is also supported by observations that only three of the 90 nest hollows for

FRTBC identified by personnel from the Western Australian Museum have been in

Jarrah (Johnstone et al. 2013a, 2013b).

Therefore, efforts to retain areas abundant in Marri, Wandoo, and other similarly

hollow-bearing species may protect disproportionate numbers of large hollows,

particularly if large and senescent trees are present. In the northern Jarrah-Marri forest,

Marri is generally most abundant lower in the landscape (Havel 1975a,b; Havel 2000),

while Wandoo trees (and other species such as Yarri and Flooded Gum) typically occur

on lower slopes and along drainage lines. Marri and Wandoo may also be important

hollow-bearers within landscapes along the ecotone with Wandoo woodland at the

eastern margin of the Jarrah-Marri forest. Bauxite deposits in the Jarrah-Marri forest

occur in mid- to upper-slope areas, so bauxite mining operations could facilitate hollow

conservation in a cost-effective way by avoiding or minimising impacts (e.g. the

construction of infrastructure such as haul roads) on lower slopes, valley floors, and the

edges of swamps.

Environmental impact assessments in potential breeding habitats should include

targeted behavioural observations and, where practical, destructive sampling.

Silviculture and mining activities in the Jarrah-Marri forest provide important

opportunities for the use of destructive sampling (e.g. Whitford 2002b; Fierke et al.

2005). The collection of data during harvesting and clearing operations could provide

information on, for example: hollow abundance; hollow dimensions and occupancy;

characteristics of hollow-bearing trees; stand and landscape-scale factors associated

with hollow incidence; tree health; and other measures (e.g. McComb et al. 1994;

76

Gibbons et al. 2002; Gibbons and Lindenmayer 2002; Whitford 2002b; Whitford and

Williams 2002; Whitford and Stoneman 2004; Koch 2008). Destructive sampling also

offers opportunities to verify or refine predictions made about hollow occurrence by

ground-based observations or non-destructive sampling (Lindenmayer et al. 2001;

Whitford 2002b; Harper et al. 2004; Munks et al. 2007; Koch 2008). Together, targeted

behavioural observations and destructive sampling provide key information on the

occupancy, size and distribution of nest hollows for black cockatoos at the local scale,

which is an important complement to the generalisations Whitford (2001, 2002b) and

Whitford and Williams (2001, 2002) made for the Jarrah-Marri forest as a whole. These

recommendations should also encourage greater coordination and systematic planning

in data collection on hollow availability assessment.

77

Chapter 5 Variation in the Structural and Floristic Features

of Mine-Site Revegetation Pits of a Similar Age

5.1 Introduction

5.1.1 Problem Formulation

This chapter describes the structural and floristic features of mine-site rehabilitation

areas at Newmont Boddington Gold (NBG). The next chapter (Chapter 6) examines

associations between these features and the feeding activity of black cockatoos.

Documenting the successful recolonisation of rehabilitation at NBG by black cockatoos

use is an important research question because of Newmont’s existing regulatory

commitments under the Commonwealth’s Environment Protection and Biodiversity

Conservation Act (EPBC) 1999. The particular decision for EPBC 2006/2591 requires

Newmont to revegetate ‘areas disturbed by mining related activities … with the aim of

providing foraging, and longer-term potential nesting habitat for black cockatoos.’ The

return of black cockatoos to rehabilitation is also important to support other regulatory

commitments, including approvals held under the Environmental Protection Act 1986

(Western Australia).

The use of rehabilitation by black cockatoos also raises practical questions about the

return of the avifauna to landscapes disturbed by mining (Armstrong and Nichols 2000;

Nichols and Nichols 2003; Nichols and Grant 2007; Weerheim 2008). Mining activities

in the Jarrah-Marri forest require the clearing of native vegetation in order to extract the

ore present in the soil profile or underlying bedrock and to construct mining

infrastructure (Rayner et al. 1996; Koch 2007). Native vegetation is then re-established

after mining operations are complete. Since the early 1980s, the long-term objective of

mine-site revegetation has been to return a functional Jarrah-Marri forest ecosystem to

landscapes disturbed by mining (Koch 2007), with the exception of certain sites, such as

residue disposal areas, where the characteristics of the post-mining materials preclude

the replacement of the original vegetation. While the focus is on restoring a functional

ecosystem, there is no dedicated focus on particular species such as black cockatoos.

78

The objective of restoring a self-sustaining forest ecosystem constrains species-level

conservation because actions beneficial to a species may not necessarily be in the

interests of restoring the full ecosystem (e.g. the ideal species mix in revegetation for

black cockatoos may not be the ideal mix for Jarrah-Marri forest restoration). Below, I

discuss three important considerations for my study of black cockatoos and

rehabilitation at NBG.

(1) Rehabilitation is a process

The first consideration is that rehabilitation is a process of habitat creation (George and

Zack 2001). This process relies on vegetation undergoing a predictable series of

changes (i.e. a successional pattern) as it develops towards the goal state (Hobbs and

Harris 2001). This process begins when vegetation is re-established and continues until

the vegetation reaches maturity, and is functionally and structurally similar to the goal

state vegetation. This process can vary considerably in terms of its timing (e.g. when

certain features of the vegetation appear) and its trajectory (e.g. the sequence of

vegetation types it follows) (Grant 2006). Use of rehabilitated mining pits at NBG by

black cockatoos will therefore reflect the particular successional pattern(s) present

within rehabilitation areas at NBG.

Ideally, rehabilitation follows a consistent successional pattern towards the goal state

(Grant 2006). However, even when uniform rehabilitation prescriptions are applied,

rehabilitation can assume varying trajectories (Hobbs and Harris 2001). If these

trajectories culminate in undesirable states, this can add to the cost of rehabilitation if,

for example, remedial management actions such as planting or thinning are required

(Grant 2006). It can also reduce the benefits rehabilitation provides, because these areas

will not support the desired vegetation structure, biodiversity, or ecosystem processes

(Koch and Hobbs 2007).

The issue of undesirable trajectories is of particular concern for mining operations,

because mine-site rehabilitation typically must achieve specific completion criteria to

meet regulatory and environmental management objectives. Successful fauna

recolonisation of mine-site rehabilitation is often a criterion, and may be particularly

important to demonstrate when the fauna in question are threatened. These completion

criteria will not be met if rehabilitation does not provide the habitat necessary for fauna

recolonisation. It is therefore important to characterise the vegetation trajectories

79

present within the rehabilitated mining pits at NBG and how the variable nature of these

trajectories may influence the goal of providing habitat resources for black cockatoos.

(2) Habitat restoration was not an explicit objective at establishment

A second consideration is that the restoration of habitat for black cockatoos was not an

explicit objective for the rehabilitation prescriptions applied when vegetation was re-

established in these pits between 1998 and 2002. The return of black cockatoos to these

sites is therefore an incidental outcome of applying those prescriptions. To the extent

that rehabilitated pits at NBG provide food resources, this is an outcome of general

prescriptions intending to return these areas to a Jarrah-Marri forest ecosystem.

This leaves open the question of whether rehabilitation at NBG could – or should – be

modified to support the particular ecological requirements of certain species. Generally,

efforts to re-establish the habitat resources of particular fauna species must be consistent

with (or conflict directly with) the overall aims of rehabilitation. There is usually little

latitude for substantially altering prescriptions relating to the structural and floristic

composition of rehabilitated sites, such as the choice of plant species or desired stem

densities. However, in certain situations it may be possible to modify rehabilitation

prescriptions so that rehabilitation provides certain habitat resources (e.g. use of waste

timber to provide shelter logs for herpetofauna) (Brennan et al. 2005; Koch 2007).

(3) Rehabilitation is a long-term objective

A third consideration is that the overall goal of rehabilitation at NBG – the

establishment of a Jarrah-Marri forest ecosystem – is a long-term objective. This applies

both to: (1) food resources (i.e. food plants), which might be expected to return with

years to decades and to (2) breeding resources (i.e. hollows in mature eucalypts), which

may require upwards of 150 years to return (Whitford and Williams 2002). This

suggests two further points: (1) a two-year study will only obtain a snapshot of the state

of the rehabilitation and its use by black cockatoos at a particular successional stage. (2)

Investigations of how black cockatoos use the resources available at particular

successional stages must be seen as distinct from the question of habitat value these

sites might have when vegetation matures.

These points suggests that, while it is important to document various aspects of the

process of recolonisation (e.g. the time lag in return, the sequence by which species

80

return, the food plants used at various stages), there may be little value in assessing the

equivalency of rehabilitation to native forest vegetation until it obtains a structure and

floristic composition comparable to mature forest. Longer-term monitoring is necessary

to determine whether revegetation successfully returns habitat resources and supports

faunal recolonisation.

5.1.2 Objectives

Here I identify a set of floristic and structural features of the vegetation in rehabilitated

mine pits that could influence its suitability as feeding habitat for black cockatoos. I

investigated these features at two spatial scales: (1) between interior (near the core of

pits) and exterior (near the edge of pits) positions within a rehabilitated mine pit and (2)

between rehabilitated mine pits of the same age. My aim was to assess the successional

stage of the rehabilitated mine pits and to characterise variation in the structure and

floristics of pits, in order to identify features that might influence the availability of

food resources for black cockatoos.

5.2 Methodology

5.2.1 Study Area

At the time of the study, the NBG site contained 50 rehabilitated mining pits with a total

rehabilitation area of approximately 190 ha (Figure 5.1). These pits are called ‘satellite’

pits because they are peripheral to the main mining area, which consists of two large

(i.e. several km wide) and deep (<1 km) pits (Rayner et al. 1996). The area mined is

likely to expand in future. Furthermore, mining as a whole in the Jarrah-Marri forest

will ultimately lead to the clearing of 83 000 ha (7% of State Forest) and the

fragmentation of 337 000 ha (28% of State Forest) (Conservation Commission of

Western Australia 2012). The suitability of rehabilitated sites for black cockatoos is

therefore an important issue.

81

Figure 5.1 General site map and location of pits. Black dots indicate pits used in the vegetation sampling.

5.2.2 Rehabilitation Protocols

The pits were rehabilitated according to prescriptions described in Rayner et al. (1996)

and similar to other mining operations within the Jarrah-Marri forest in terms of

landscaping, soil management, seeding, and planting (Koch 2007). While some pits

were rehabilitated as early as 1992 - 1993, most were rehabilitated between 1998 and

2002 (Table 5.1). Prescriptions for establishing vegetation relied on the natural seed

bank present in soils used in rehabilitation, direct (broadcast) seeding from commercial

seed mixes, and planting of nursery-reared seedlings. The seed mixes contained only

82

plant species endemic to the native forest at the site and included Jarrah, Marri, and

Wandoo as canopy-forming tree species, and Banksia, Hakea, and Allocasuarina spp.

as midstorey and shrub species, along with a wide range of other shrub, herb, and forb

species. Rates and methodology of seeding and fertiliser application were standardised

across all pits and seeding covered the entire surface area of each pit (M Durack,

Newmont Boddington Gold, 2010, pers. comm.).

Table 5.1 Year of establishment of the rehabilitation pits studied. Rehabilitation Pit Year of Revegetation Used in Vegetation Sampling

D4 (SMG) 1998 ✔ (Pit 1) K1 2000 K2 2000 ✔ (Pit 2) K3 2000 K4 2000 K5 2000 K6 2000 ✔ (Pit 3) K7 2000 K8 2002 K9 2000 L1 1999/2001 L2 2001 ✔ (Pit 4) L3 2001 L4 2001 M1 2002 M2 2001 ✔ (Pit 5) M3 2001 M4 2001 ✔ (Pit 6) M5 2001 Q1 2002 ✔ (Pit 7) Q2 2002 Q3N 2002 Q3S 2002 ✔ (Pit 8) WTR 1996 ✔ (Pit 9)

5.2.3 Selection of Pits for Sampling

Twenty-three of the 50 pits were available for vegetation study based on size,

accessibility, and distance from active mining areas. I excluded all pits that were: less

than 1 ha in area, inaccessible to vehicles, and near mining operations (e.g. active

mining pits). These included six designated for re-clearing for mining activities, three

that were not yet revegetated, and 17 that were too small or were located in areas that

made them inaccessible or unsuitable because of safety concerns. I selected nine of

these 23 pits to use for vegetation sampling: D4 (pit 1), K2 (pit 2), K6 (pit 3), L2 (pit 4),

M2 (pit 5), M4 (pit 6), Q1 (pit 7), Q3S (pit 8) and WTR (pit 9) (Figure 5.1, Table 5.1). I

83

chose pits that had a similar (gentling sloping) topography; and were revegetated within

a short span of time. The selected pits were spread across the mine site (Figure 5.1,

Table 5.1).

5.2.4 Vegetation Structure, Floristics and Phenology: Variables

I measured 11 structural, floristic or phenological variables in each of the sampling

plots in each rehabilitated pit (Table 5.2). The structural variables canopy cover, canopy

height, non-canopy and stem density (living and dead) and the floristic variables 'stem

density of all food plants' and 'stem density of major feed plant species' provided

reliable and efficient measures of the structure and composition of re-established

vegetation and are commonly used in studies of revegetation (e.g. Vesk et al. 2008;

Munro et al. 2009, 2011; Gould 2011). I used them to compare the vegetation between

pits and between interior and exterior plots within pits. A different subset of variables

from Table 5.2 was used to compare characteristics of plots where black cockatoos fed

and where they did not feed (see Chapter 6).

84

Table 5.2 Structural, floristic and phenological variables measured on each plot.

Structural variables

Variable Definition

Canopy cover

Measured at 1m intervals along point-intercept transect using an optical densitometer (n = 10 measurements per plot). Results were recorded as 1 (fully covered), 0.5 (partially covered) and 0 (no cover). The ten measurements were summed to give a single canopy cover value between 0 and 10 for the plot.

Canopy height Mean height of stems from canopy-forming species (>0.5 m height). We regarded plants of the genera Eucalyptus, Corymbia, Allocasuarina, and Acacia as canopy-forming species.

Non-canopy height Mean height of plants from genera other than those regarded as canopy-forming (>0.5 m height).

Vegetation density: living and dead levy pole touches at 0.5m, 1.0m, 1.5m, 2.0m, 2.5m, 3.0m and 3.5m

Measured at 1 m intervals along point-intercept transect using a levy pole graduated into 0.5 m intervals to a height of 3.5 m. We recorded the number of touches by each plant species, living or dead, at each 0.5 m gradation. The ten measurements for each height point were summed to give a single value for that height for the plot.

Total stem count for living and dead stems

The total number of living and dead stems of any species (>0.5 m height) in the plot. If a plant was multi-stemmed (e.g., coppice regrowth) I recorded this as one stem only. Any plant that was mostly or fully brown or grey, including leaves, was classified as dead.

Floristic variables

Variable Definition Stem density of all food plants

The number of stems of living plants (>0.5 m height) within a plot for any of the 15 plant species known to be potential foods.

Stem density of major feed plant species

The number of stems of living jarrah, marri, Hakea undulata, H. prostrata and Banksia squarrosa (>0.5 m height) within a plot

Individual plant structural variables

Variable Definition

Canopy width

For every stem in the plot (both canopy and non-canopy species), I estimated the width of the canopy by holding the levy pole horizontally at the greatest width of the canopy. These values were averaged for each species to give a mean canopy-width for each species in the plot. Recorded for jarrah, marri, Hakea prostrata, Hakea undulata and Banksia squarrosa

Height of five most abundant food plant species

Mean height of stems (>0.5 m height) of jarrah, marri, Hakea undulata, H. prostrata and Banksia squarrosa. If the tree was multi-stemmed, height was taken for the tallest stem

Phenological variables

Variable Definition

Flowering or post-flowering

If buds were present or had opened as flowers or if fallen flowers were present beneath the plant. Recorded for jarrah, marri, Hakea undulata, H. prostrata and Banksia squarrosa

Seeding Open or closed seedpods were present. Recorded for jarrah, marri, Hakea undulata, H. prostrata and Banksia squarrosa

5.2.5 Vegetation Structure and Floristics: Sampling Methods

I sampled 10 replicate plots (each 100 m2 in area) within each of the nine rehabilitated

mining pits, thus providing a total of n = 90 plots (total area – 9 000 m2) for the study.

Each pit contained five interior plots (>25 m from any edge), which were located at

85

equal intervals along a transect running at a diagonal across the entire pit. Each pit also

contained five exterior plots, which were positioned at equal intervals around the

perimeter of the pit (Figure 5.2).

(a)

(b)

(c) Figure 5.2 Dimensions and layout of the vegetation sampling design: (a) 10 m × 10 m interior plot with the 10 m transect and (b) 20 m × 5 m exterior plot with the 10 m transect, and (c) positions of sampling plots around the rehabilitation for sampling (not drawn to scale).

Plot dimensions were 10 m × 10 m for interior plots and 20 m × 5 m for exterior plots,

and all plots were separated by at least 50 m (Figure 5.2). The longer (20 m) edge of the

86

exterior plot was positioned along at the outer edge of the vegetation at each site (i.e. so

that the plots measured only 5 m inwards from outermost stems) (Figure 5.2). I used the

combination of interior and exterior plots in order to examine potential edge effects and,

more specifically, to determine whether plot-positioning could detect differences in

vegetation characteristics within the interior of the rehabilitated area and around its

margins.

Table 5.2 provides a full description of the sampling procedure for each variable.

Measurements of canopy cover were taken using a 10 m point-intercept transect situated

within each sampling plot. For the 10 m × 10 m interior plots, I positioned the point-

intercept transect through the middle of the plot along the same axis as the transect

running across the entire pit (Figure 5.2). For the 20 m × 5 m exterior plots, I positioned

the point-intercept transect parallel to the long axis of the plot and in the middle of the

plot, so that the transect was 2.5 m from the long (20 m) side of the plot and 5 m from

short (5 m) of the plot (Figure 5.2).

Measurements of the other variables were taken using each plot as a 100 m2 sampling

area. For measurements of canopy height, non-canopy height, stem density (live stems),

and stem density (dead stems), an individual plant was considered as a ‘stem’ if it

measured at least 0.5 m in height. Stems smaller than 0.5 m in height were not

measured. I measured each stem in a plot that met the criteria for a particular variable.

Heights were measured using a levy pole as a guide to measure the highest point on the

plant. A stem was defined as within a plot if at least 50% of its bole was within the plot

boundaries.

5.2.6 Analysis

Analysis and presentation of the data on vegetation structure and floristics followed the

approach used by Catterall et al. (2004). This consisted of (1) an initial graphical

presentation of the data, accompanied by univariate statistics, and then (2) a more

extensive multivariate analysis to further explore potential differences in vegetation

structure and floristics between pits and between interior and exterior plots.

As a first step, the means of each of the five structural variables were used as dependent

variables in two-way analysis of variance (ANOVA) (after checking that the data

conformed to basic ANOVA assumptions such as homoscedascity and normality) with

87

factors of pit (the nine pits) and location (interior or exterior plots). Logarithmic

transformation was used to correct any problems found. The significance level for main

effects and interactions was set at 0.01 after Bonferroni correction to allow for the

multiple tests. When significant, ANOVA was followed with Tukey’s Honest

Significant Difference (HSD) tests to determine which pits differed from others. All

these analyses used the Statistica 7.1 software package (Statsoft 2006).

Structural differences across pits and between interior and exterior plots were then

explored further using non-metric multidimensional scaling (nMDS) followed by non-

parametric multivariate analysis of variance (PERMANOVA or NPMANOVA) and,

where PERMANOVA was significant, similarity percentage (SIMPER).

nMDS is a non-parametric, ordination-based statistical technique used for visual

exploration of the similarities or dissimilarities within or between data sets. It

determines relationships in space between subjects (in this case- rehabilitation pits) on

multiple axes, one for each variable (here, the structural variables measured in each

plot). For ease of comprehension, the relationships are graphed on two or three axes that

preserve as closely as possible the relative ranked positions determined when every

variable is used. Reducing the axes in this way loses information and ‘stress’ statistics

measure the extent of the loss (Clarke and Warwick 2001). If stress is low, the two- or

three-dimensional representation is a reasonable representation of the relative positions

of the subjects. Clarke and Warwick (2001, p. 5.5) state that a stress of <0.1 gives ‘a

good ordination with no real prospect of a misleading interpretation’, while a stress of

<0.2 ‘still gives a potentially useful … picture’.

nMDS begins with the calculation of a distance matrix to represent the relative

difference between points in the study. I used a distance matrix based on the Bray-

Curtis measure (Bray and Curtis 1957), which is widely used in ecology because it has

been successful in a range of tests and simulation studies (Faith et al. 1987; Clarke and

Warwick 2001). I range-standardised each of the structural variables to a scale between

0 and 1 by subtracting the smallest score from each value and dividing the result by the

difference between the largest and the smallest score for that variable, thereby allowing

an equal impact of each variable. I fitted each nMDS using two dimensions first, using

the stress statistic to determine the adequacy of the fit. If stress was ≤0.2, I accepted the

88

two-dimensional solution as adequate. If it was not, I fitted a three-dimensional solution

and checked the stress values again.

I wished to examine possible differences in vegetation structure between interior and

exterior plots and also between different rehabilitation pits. Therefore I ran one nMDS

using structural variables to compare interior and exterior plots across all sites

combined, and a second nMDS to compare vegetation structure across all nine

rehabilitation pits. nMDS gives a visual representation of the similarity or dissimilarity

of samples, but incorporates no test for statistical significance. This is provided by

PERMANOVA.

PERMANOVA is a non-parametric analogue to traditional MANOVA, providing an F

statistic with an associated p-value for testing the significance of each factor in the

design and the interaction between them (Anderson 2001). It can be based on any

distance measure, so I chose the Bray-Curtis measure to be consistent with nMDS.

While post hoc tests can be used for paired comparisons of levels within a significant

factor (for example, to test for differences between specific pairs of pits), I did not use

them because the pits did not represent different experimental treatments, but showed

the range of variation arising when different pits were rehabilitated with the same

approach. Accordingly, I was most interested in whether or not the pits overall were

similar rather than in the detail of any differences that might arise. However, where

PERMANOVA was significant, I used similarity percentage (SIMPER) to determine

the contribution of individual plant species to the difference. All analyses used the

relevant routines in the PAST software package (Hammer et al. 2001).

To analyse floristic data, I first used species-presence data from each sampling plot to

calculate a Shannon species diversity for each pit, with a 95% confidence limit for the

interior plots (based on all five plots combined) and for the exterior plots (based on all

five plots combined). Calculations were performed using the PAST software (Hammer

2001). The Shannon species diversity incorporates data on the number of taxa and the

number of individuals in each taxon in the study area, using the formula H=-

sum((ni/n)ln(ni/n)), where H is the Shannon species diversity, n is the total number of

individuals and ni is the number of individuals in category i. While any base of

logarithms can be used (Zar 2010), PAST uses base e. I first completed a univariate

analysis of species diversity across the nine pits for interior plots and then for exterior

89

plots. I used the relevant routines in PAST (Hammer et al. 2001), with the sequential

Bonferroni correction (Quinn and Keogh 2002) used to determine significance given the

multiple tests.

I also compared floristics between interior and exterior plots and across the nine

rehabilitation pits using nMDS, and two-way PERMANOVA, followed by SIMPER if

either of the variables in the PERMNOVA was significant. The dependent variables in

this case were the number of live stems of 15 plant species counted in each of the plots.

Range-standardisation was not necessary for the floristic data, because each species was

assessed with a common variable of stem number. However, there was an issue with

several scarce species across most plots that led to unacceptably high stress on nMDS

plots. To overcome this problem, I included a ‘dummy species’ with a value of 1 in all

plots in the analysis. This modification is recommended by Clarke et al. (2006) for

situations in which some samples contain few records and this is not an artifact of

unequal sampling intensity. Clarke et al. (2006) describe this correction as analogous to

the well-known log(x+1) data transformation in univariate statistics to deal with the

problem of small or zero observations. Apart from these differences, I used the same

procedures as described for the vegetation structure data.

Lastly, I made qualitative comparisons of the structural and floristic data against

descriptions of characteristics of desired states and transitions for Jarrah forest

rehabilitation (Grant 2006). These indicated the successional stage of the rehabilitation

plots.

5.3 Results

5.3.1 Structural Variables

Canopy cover and stem density (dead) were the only structural variables to differ

between interior and exterior plots. Both were higher in interior plots (Figure 5.3 and

Table 5.3). More substantial differences occurred across pits. With the exception of

average canopy height, all structural variables differed significantly across pits. There

were no significant interactions between pit (1 - 9) and location (interior/exterior)

(Figure 5.3 and Table 5.3).

90

91

Figure 5.3 Bar graphs showing for interior and exterior plots at each of nine rehabilitation pits: mean canopy cover, mean canopy height, mean non-canopy height, mean stem density (live), mean stem density (dead) (a) – (e) and species diversity (live) (where a species diversity was calculated across all interior plots and all exterior plots for each rehabilitation pit) (f). Error bars are standard errors for all mean values and bootstrapped 95% confidence intervals for species diversity. The bootstrapped 95% confidence limits are not symmetrical, so upper and lower boundaries are shown. For all except species diversity, pits shown to differ in mean values at the 5% level using Tukey’s honest significant difference following ANOVA are shown with superscripts. For species diversity, similar superscripts indicate significant differences between species diversity values for interior plots (lower case) and exterior plots (upper case) as determined by t-tests after sequential Bonferroni correction.

92

Table 5.3 The two-way ANOVAs for the five structural variables plotted in Figure 5.3. The factors are Pit (1 – 9) and plot location (interior/exterior). The significance level for main effects and interactions was set at 0.01 after Bonferroni correction to allow for the multiple tests. Mean non-canopy height (the dependent variable was log-transformed before analysis) SS df MS F p Intercept 3.05 1.00 3.05 899.65 <0.01 Pit 0.55 8.00 0.07 20.28 <0.01 Plot location 0.00 1.00 0.00 0.14 0.71 Pit*Plot 0.03 8.00 0.00 1.08 0.39 Error 0.24 72.00 0.00 - - Mean live stem density (the dependent variable was log-transformed before analysis) SS df MS F p Intercept 233.20 1.00 233.20 10307.72 <0.01 Pit 2.71 8.00 0.34 15.00 <0.01 Plot location 0.02 1.00 0.02 1.04 0.31 Pit*Plot 0.37 8.00 0.05 2.02 0.06 Error 1.63 72.00 0.02 - - Mean dead stem density (the dependent variable was log-transformed before analysis) SS df MS F p Intercept 62.47 1.00 62.47 817.34 <0.01 Pit 3.37 8.00 0.42 5.51 <0.01 Plot location 1.40 1.00 1.40 18.27 <0.01 Pit*Plot 0.76 8.00 0.09 1.24 0.29 Error 5.50 72.00 0.08 - - Mean canopy height SS df MS F p Intercept 1435.20 1.00 1435.20 459.80 <0.01 Pit 44.24 8.00 5.53 1.77 0.10 Plot location 0.54 1.00 0.54 0.17 0.68 Pit*Plot 17.33 8.00 2.17 0.69 0.70 Error 224.74 72.00 3.12 - - Mean canopy cover SS df MS F p Intercept 843.34 1.00 843.34 153.57 <0.01 Pit 263.49 8.00 32.94 6.00 <0.01 Plot location 56.80 1.00 56.80 10.34 <0.01 Pit*Plot 51.72 8.00 6.47 1.18 0.32 Error 395.40 72.00 5.49 - -

PERMANOVA found significant structural differences in the vegetation across pits

(F8,89 = 8.91, p < 0.01), and between exterior and interior plots (F1,89 = 5.40, p = 0.01).

The interaction between pits and interior and exterior plots was not significant (F8,89 =

1.13, p = 0.30). nMDS indicated that the significant difference between interior and

exterior plots was slight, as indicated by the high overlap in their minimum convex

hulls. Pits appeared more dissimilar, with pits L2 and Q1 overlapping more with each

other than with other pits (Figure 5.4 (a) and (b)).

93

Figure 5.4 (a) Non-metric MDS for the structural similarity of nine rehabilitation pits using the variables canopy cover, canopy height, non-canopy height, stem density (live) and stem density (dead). Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.1753. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units.

94

Figure 5.4 (b) Non-metric MDS for the structural similarity of interior and exterior plots across nine rehabilitation pits using the variables canopy cover, canopy height, non-canopy height, stem density (live), stem density (dead) and species diversity of live plants (here interior pits are indicated by the solid line and square points, while exterior pits are indicated by the broken lines and cross points). The similarity measure was Bray-Curtis and the stress was 0.1753. 95% confidence ellipses are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

SIMPER indicated that average non-canopy height (23.82%) and canopy cover

(23.53%) were responsible for almost half the differences across pits, and that canopy

cover (23.72%) and average non-canopy height (22.99%) were responsible for almost

half of the differences between interior and exterior plots (Table 5.4 (a) and (b)).

95

Table 5.4 Results of SIMPER for the percentage differences in structural variables between (a) pits and (b) interior and exterior plots. Note that the mean scores are based on variables range-standardised to a 0 – 1 scale. Cumulative % (column 2) is calculated by dividing the sum of the relevant contributions by the total contribution.

(a) Structure across pits

Measure (mean) Contribution Cumulative %

Mean score D4 Pit 1

Mean score K2 Pit 2

Mean score K6 Pit 3

Mean score L2 Pit 4

Mean score M2 Pit 5

Mean score M4 Pit 6

Mean score Q1 Pit 7

Mean score Q3S Pit 8

Mean score WTR Pit 9

Non-canopy height 9.134 23.82 0.445 0.0906 0.536 0.616 0.647 0.606 0.518 0.608 0.0805

Canopy cover 9.022 47.35 0.41 0.175 0.4 0.085 0.285 0.2 0.185 0.32 0.695 Stem density (alive) 6.877 65.29 0.497 0.146 0.489 0.309 0.475 0.365 0.407 0.612 0.154 Canopy height 6.823 83.09 0.473 0.403 0.525 0.464 0.66 0.564 0.478 0.435 0.356 Stem density (dead) 6.485 100 0.593 0.131 0.438 0.276 0.224 0.176 0.193 0.217 0.245

Total 38.341

(b) Structure interior/exterior Measure (mean) Contribution Cumulative % Mean score

interior plots Mean score exterior plots

Canopy cover 8.963 23.72 0.271 0.341 Non-canopy height 8.684 46.71 0.46 0.461 Canopy height 6.792 64.68 0.499 0.469 Stem density (dead) 6.711 82.45 0.357 0.197 Stem density (alive) 6.633 100 0.371 0.396 Total 37.783

96

5.3.2 Floristics

Floristics followed a similar pattern. Species diversity between interior and exterior

plots at the same pit was very similar, but varied across pits for both interior and

exterior plots (Figure 5.5 (a) and (b)). Pits 2 and 9 were markedly lower in species

diversity.

Figure 5.5 (a) Non-metric MDS for the floristic similarity of nine rehabilitation pits, based on the number of live stems of 15 feed tree species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.202. 95% confidence ellipses are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

97

Figure 5.5 (b) Non-metric MDS for the floristic similarity of interior and exterior plots across nine rehabilitation pits, based on the number of live stems of 15 feed tree species (here interior pits are indicated by the solid line and square points, while exterior pits are indicated by the broken lines and cross points). The similarity measure was Bray-Curtis and the stress was 0.202. 95% confidence ellipses are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the plots are relative distances from one another rather than absolute differences read in units.

PERMANOVA found significant differences in floristics across pits (F8,89 = 6.69, p <

0.01), but not between exterior and interior plots (F1,89 = 1.26, p = 0.26). The interaction

between pits and interior and exterior plots was not significant (F8,89 = 1.26, p = 0.26).

Interior and exterior plots also appeared similar in nMDS with high overlap in their

minimum convex hulls, whereas pits overlapped less in their minimum convex hulls

(Figure 5.5 (a) and (b)). SIMPER indicated that the number of live stems of Marri

(28.77%), number of live stems of Hakea undulata (15.15%) and number of live stems

of Jarrah (15.02%) were responsible for nearly 60% of the differences across pits (Table

5.5).

98

Table 5.5 Results of SIMPER for the percentage differences in floristic variables (live stem densities of indicated species) between pits. Note that the mean scores are based on the live stem densities of each plant species. This is the same scale for each species, so no standardisation has been done. A dummy variable with a score of 1 for each pit was included (see text). Cumulative % (column 2) is calculated by dividing the sum of the relevant contributions by the total contribution.

Taxon Contribution Cumulative %

Mean score D4 Pit 1

Mean score K2 Pit 2

Mean score K6 Pit 3

Mean score L2 Pit 4

Mean score M2 Pit 5

Mean score M4 Pit 6

Mean score Q1 Pit 7

Mean score Q3S Pit 8

Mean score WTR Pit 9

Marri 16.23 28.77 6.1 4 7.8 2.8 11.9 9 10.5 12 0.8

Hakea undulata 8.549 43.92 8 0 4.7 4.4 1.8 4.9 1.1 1.4 0.1

Jarrah 8.476 58.94 3.1 3 2.1 5.4 5.1 3.1 4.6 6.1 2.7

Banksia grandis 7.62 72.45 1.1 0 2.4 5 3.5 3 3.7 5.4 0

B. squarrosa 6.199 83.44 1.1 0.5 3.1 0 2.4 4 2.8 5 0.3

H. prostrata 4.427 91.28 0.6 0.4 0.7 1.2 2.6 4.1 1.5 1.4 0.2

H. trifurcata 1.34 93.66 2.4 0 0.2 0.2 0 0 0 0 0

B. sessilis 1.32 96 0.2 0 1.1 0 0.1 0.1 0.1 0 0.8

H. amplexicircus 0.9379 97.66 0.1 0 0.1 0.3 0.7 0.3 0.4 0.5 0

H. lissocarpha 0.4023 98.38 0.1 0.1 0 0.3 0 0 0.1 0.2 0

Hakea sp. 0.3436 98.98 0 0 0 0.4 0 0 0.1 0.2 0

H. cyclocarpa 0.3321 99.57 0 0 0 0.4 0 0 0.1 0.2 0

H. incrassata 0.1992 99.93 0 0 0 0.1 0 0.1 0 0.1 0

H. ruscifolia 0.04176 100 0 0 0 0.1 0 0 0 0 0

H. varia 0 100 0 0 0 0 0 0 0 0 0

Dummy 0 100 1 1 1 1 1 1 1 1 1

99

5.3.3 Successional Stage

Structural and floristic characteristics indicate that the pits are at an intermediate

successional change. Firstly, the stem density for live plants was higher than the stem

density for dead stems, indicating that revegetation has become established at these sites

rather than that the revegetation has failed. Secondly, the height differences between the

canopy and non-canopy species indicate a demarcation between the canopy and non-

canopy layers (i.e. shrub- or ground-level vegetation). The mean heights of the canopy

species ranged between 2 and 6 m, while the mean heights of the non-canopy species

were all less than 1 m. These overall values for heights were similar to post-bauxite

mining rehabilitation where canopy species achieved heights of more than 1.5 m a

minimum of five years post-rehabilitating (Koch and Samsa 2007). Thirdly, both forest

tree species and understorey vegetation were successfully established. These

characteristics indicate that these sites are similar to state S3 sites within Alcoa

(Aluminium Company of America) mine-site rehabilitation areas in the Jarrah-Marri

forest (Grant 2006) (Table 5.6). Table 5.6 Characteristics of desired states and transitions suggested for Alcoa sites – from Grant (2006).

S0: Mining Mined out pit prior to rehabilitation operations commencing

S1: 0 years Landscaped pit ripped on the contour to greater than 1.2 m with no unblasted caprock and greater than 90% topsoil cover

S2: 0 – 5 years

Rehabilitated pit with scrub vegetation dominated by eucalypts and acacias. Minimal weed invasion, moderate to high species richness (>50% of forest controls), 1,300 stems/ha of eucalypts (500 – 2 500 stems/ha), and 1 legume/m2 (0.5 – 2.0 legume/m2)

S3: 5 – 15 years Sapling to pole-sized eucalypts with dense Acacia understorey, high fuel loads, three-tiered vegetation structure, and moderate species richness (50 – 80% of forest controls)

S4: >15 years Pole-sized eucalypts with senescent Acacia understorey, high fuel loads, two-tiered vegetation structure, moderate species richness, minimum of 300 potential sawlogs/ha, including 100 stems/ha of Marri

S5: 0 – 5 years post-burn

Pole-sized eucalypts with regenerating diverse understorey, low fuel loads, two-tiered vegetation structure, moderate to high species richness (>50% of forest controls), and less than 5% tree deaths following burning

Sx: Rehabilitation objective

Multiple land use Jarrah forest with 500 to 1 500 stems/ha of Jarrah in the overstorey, species rich understorey including resprouters, two-tiered fuel structure, and ability to be incorporated into fire management with the surrounding unmined forest

100

5.4 Discussion

5.4.1 Successional Stage

Most sites had advanced beyond the initial ‘scrub’ successional stage and had reached

an interim stage in which canopy-forming trees were beginning to predominate (and to

form the most significant structural feature) but high densities of proteaceous shrubs

remained. It is likely that the abundance, and perhaps also the diversity, of these shrub

species will decline as the canopies of Jarrah and Marri increase in volume and shade

mid- and understorey vegetation. At these sites, it takes at least 15 years for

rehabilitation pits to closely resemble mature forest in species diversity and longer for

structure (Koch 2007; Koch and Hobbs 2007; Koch and Samsa 2007).

5.4.2 Differences between Exterior and Interior Plots

Structurally, interior and exterior plots differed only in the frequency of dead stems and

in canopy cover, both of which were higher in interior plots. This is consistent with the

greater canopy cover shading and killing understorey species and late developing

individuals of canopy species. Although PERMANOVA for comparison of vegetation

structure between interior and exterior plots was statistically significant, the nMDS

indicated that the magnitude of the difference was small. No significant differences

were found for floristics. Overall, the findings suggest few edge effects at these pits.

It is possible that the similarity of the exterior and interior vegetation relates to the size

of the rehabilitation pits (which ranged 0.7 ha to 14.9 ha) and that more substantial

differences might have arisen in larger pits with a greater interior area (Davis 2004).

However, more relevant explanations include: (1) the proximity (<50 m in most cases)

of the pits to remnant native forest vegetation, which may limit the influence of edge-

related effects such as wind shear, sedimentation, and scouring by airborne particles

(Goosem and Jago 2006); and (2) uniformity in the rehabilitation prescriptions across

pits (e.g. ripping, seeding).

5.4.3 Differences between Pits

Vegetation varied structurally and floristically across pits despite the use of similar seed

mixes and a similar age range (established between 1996 - 2002). This suggests that the

trajectory of each pit was influenced by pit-scale factors relating to the rehabilitation

landscape (e.g. aspect, topography), environment (e.g. precipitation, ambient

101

temperature), soil properties (e.g. nutrient availability, salinity, moisture content), or

some artifact of the rehabilitation process (e.g. differences in fertiliser application,

seeding rates) (Vesk and Mac Nally 2006; Koch and Hobbs 2007).

Pit-scale differences in structure are common in mine-site rehabilitation within the

Jarrah-Marri forest. For example, Koch and Samsa (2007) reported highly variable stem

densities for saplings of Jarrah (mean = 1 375 stems per ha; range = 550 to 3 175) and

Marri (mean = 333 stems per ha; range = 50 to 1 125 stems per ha) within 10 - 15 year-

old restored Alcoa sites. Species richness is also variable, though to a lesser degree

(Koch 2007).

Some of the differences between pits are readily observable. For example, the

rehabilitation at WTR was unique in lacking both Jarrah and Marri as canopy-forming

eucalypts; stem densities for proteaceous food plants were also low. The vegetation in

some of the ‘K’ pits (pit K2) also differed in having: (a) low overall stem densities and

(b) more ‘robust’ stems of Marri that had larger DBHs (diameter at breast heights) and

canopy volumes (and thus, likely, larger quantities of fruit). These K pit sites presented

a woodland-like appearance in comparison with the much thicker vegetation densities at

most other sites.

5.4.4 Implications

The availability of habitat resources is likely to change as succession proceeds and the

structure (and species composition) of the vegetation becomes closer to that of the

native forest, with lower stem densities, a well-developed overstorey, and moderately-

to poorly-developed understoreys. In general, these changes are likely to decrease the

availability of proteaceous food sources, as midstorey layers become thinner. While a

wide range of ages of rehabilitated pits was not available for study at NBG, extending

the approaches described here to a range of older rehabilitation at other mine-sites could

cover this deficiency.

Evidence of structural and floristic differences between pits likely means that

differences in food availability between pits also occurs, such as variation in the

abundance and quality of Marri fruits. Such differences in availability of food resources

are likely to result in variation in feeding activity across pits. This is the subject of the

next chapter. Without detailed studies of microclimate and soil conditions, reasons for

102

the differences in vegetation between pits cannot be assessed. Differences between

interior and exterior vegetation in individual pits may be important to black cockatoos if

predation risk varies between the two zones.

103

Chapter 6 Recolonisation of Mine-site Revegetation by

Threatened Black Cockatoos in the Jarrah-Marri Forest of

Southwestern Australia

6.1 Introduction

There is broad agreement that forest management practices and mine-site rehabilitation

should maintain or restore habitat resources required by endemic fauna (Conservation

Commission of Western Australia 2003; Koch and Hobbs 2007). Thus it is important to

determine when black cockatoos begin using mine-site rehabilitation and the structural

and floristic features associated with their return (Lee et al. 2010).

Confirming the presence of recolonising species can be difficult. For example, black

cockatoos are detected infrequently during fauna surveys within bauxite mine-site

restoration (Armstrong and Nichols 2000; Nichols and Nichols 2003; Nichols and Grant

2007). This likely reflects the general difficulty of detecting mobile species occurring at

low densities in a forest landscape (Craig and Roberts 2005). Feeding residues (or

‘trace’) provide an alternative means of species detection. Feeding black cockatoos

often generate distinctive residues such as fruit husks, seed pods, twigs, branches,

flowers, and buds) and, in the case of Marri, individual species can be distinguished

(Saunders 1974a,b; Johnstone and Kirkby 1999; Cooper 2000; Cooper et al. 2003;

Weerheim 2008; Biggs et al. 2011). The amount of feeding residues can also be

estimated, providing an index of feeding intensity. In a preliminary study, Lee et al.

(2010) and others (Clout 1989; Cameron and Cunningham 2006) demonstrated that

examination of feeding residues was a valuable complement to observational work in

assessing black cockatoo activity in rehabilitated mine-sites in the Jarrah-Marri forest.

This chapter examines the feeding ecology of black cockatoos within 7 - 14 year old

rehabilitated mine pits at Newmont Boddington Gold (NBG), located at the eastern

margin of the Jarrah-Marri forest. My study design had three components. Firstly, I

used behavioural observations of black cockatoos to assess how frequently black

cockatoos used rehabilitated mine pits and what activities were undertaken in these

areas, and to characterise differences among the three species. Secondly, I used

systematic plot-based sampling of feeding residues within rehabilitated mine pits to

104

describe: (1) the food plants eaten, (2) the distribution of feeding activity across pits and

between interior and exterior vegetation in each pit, and (3) temporal patterns in feeding

activity. Lastly, I used systematic plot-based sampling within rehabilitated mine pits to

obtain floristic and structural vegetation data in order to assess the characteristics of

plots favoured for feeding by black cockatoos (refer to Chapter 5). The results confirm

that black cockatoos feed in young rehabilitated mine pits at this site, and suggest the

value of approaches combining observational work with systematic assessments of

feeding residues and vegetation structure and phenology.

6.2 Methods

6.2.1 Study Area and Rehabilitation Protocols

In Chapter 5, I explained how 23 of the 50 rehabilitated pits on site were identified as

potential study pits for vegetation structure and floristics, and how nine of these pits

were chosen for detailed analysis. I used an observational protocol to assess black

cockatoo use of the larger set of 23 pits, and detailed analysis of feeding residues to

confirm their feeding in the subset of nine pits also studied for structural and floristic

composition (refer to Chapter 5 - Table 5.1, Figure 5.1). The nine pits chosen for

residue analysis were spread across the mine-site and all had similar gently sloping

topography. Eight of the sites were established with similar seed mixes. One differed

from the other eight sites in being dominated by Wandoo, with few stems of Jarrah and

Marri and only a weak proteaceous under-storey.

6.2.2 Observations of Black Cockatoos in Rehabilitated Mine Pits

I surveyed 23 pits once every two weeks in October - November 2008 and once a month

from December 2008 to July 2010. During these surveys, I occupied a vantage point

near each pit either within four hours of dawn or four hours of dusk and observed the pit

and its immediate surrounds for five minutes. These times coincided with peak feeding

activity of black cockatoos (T Kirkby, Western Australian Museum, 2009, pers.

comm.). I defined the ‘presence’ of black cockatoos at a pit as the detection of black

cockatoos: (1) visually, either within rehabilitation vegetation or flying directly above

the pit or (2) acoustically, based on contact calls appearing to originate from within the

rehabilitation vegetation or immediately adjacent to it. Black cockatoos were identified

to species.

105

Behavioural observations of black cockatoos were recorded for the first ten minutes

whenever birds were observed feeding or roosting within rehabilitation vegetation,

using a scan sampling protocol (Altmann 1974). I attempted to observe all the

individuals present and record their predominant activity (that of ≥50% of individuals

during the 10-minute scan sample), the flock size and any foods eaten.

Additionally, I conducted behavioural observations whenever black cockatoos were

encountered opportunistically in rehabilitated mine pits during other studies of black

cockatoo ecology at the NBG site (Lee et al. 2013b, Chapter 3). Data from systematic

and opportunistic encounters are distinguished in the results.

6.2.3 Identifying Feeding Residues in Rehabilitated Mine Pits

I used feeding residues to investigate patterns in feeding activity within and across pits.

To record residues, I established ten replicate sampling plots (each 100 m2) within nine

pits, providing a total of n = 90 plots (total area – 9 000 m2). Each pit contained five

interior plots (≥25 m from any edge), located at equal intervals along a transect running

diagonally across the entire pit. Each pit also contained five exterior plots positioned at

equal intervals around the perimeter (refer to Chapter 5 - Figure 5.2). Plot dimensions

were 10 m × 10 m for interior plots and 20 m × 5 m for exterior plots, and all plots were

separated by at least 50 m. The longer (20 m) edge of the exterior plot was positioned

along the outer edge of the vegetation at each pit so that the plots measured only 5 m

inwards from the outermost stems (Figure 5.2 in Chapter 5). I used the interior and

exterior plots to examine potential edge effects and, more specifically, to determine

whether plot-positioning could detect differences in feeding activity within the interior

of rehabilitated areas and around their margins.

I searched each 100 m2 plot for feeding residues by examining the ground beneath each

stem of a potential food plant (0.5 m high or taller) within the area covered by the

plant’s canopy. I removed leaf litter to ensure that feeding residues covered by leaf fall

were also identified. Feeding residues were identified using characteristics shown in

Figure 6.1. If the canopies of two or more plants of the same species overlapped, I

allocated equal numbers of feeding residues to each plant. If the canopy of a plant in the

plot overlapped with the canopy of a plant outside the plot, I also allocated equal

numbers of feeding residues to each plant.

106

Figure 6.1 Features of proteaceous feeding used to recognise black cockatoo feed residues: (a) branches bitten off displaying the characteristic 45 ° angled ‘snipped’ tip; (b) broken branches B. squarrosa; (c) broken branches of H. prostrata and (d) H. undulata, and (e) cracked seedpods of H. cyclocarpa. To my knowledge, given the unique characteristics, type and location of the residue, no other species leave similar residue.

Fruit husks, the feeding residues from myrtaceous plants, were counted within the

canopy diameter of each stem. These counts were then summed for each 100 m2 plot. In

two plots there were too many fruit husks to count, so the total number in each case was

estimated based on the area covered and the depth. Feeding residues for proteaceous

plants consisted mainly of broken branches and seedpods within the canopy diameter of

the plant. There were also occasional buds and flowers. Each of these residue types was

counted for each stem of each food plant within a plot and summed for each 100 m2

107

plot. Thus each plot had a summed residue score for each of the potential food plant

species.

Plots were checked for feeding residues twice in winter (July 2009 and July 2010) and

once in summer (January 2010). I removed feeding residues from plots at the end of the

first and second sampling sessions, so the residues observed during the second and third

sampling sessions included only those feeding residues produced within the six months

prior to the sample. Residues on the first occasion could have accumulated over longer

periods. For Marri trees in two of the 90 plots, the depth and extent of feeding residues

precluded their complete removal. These trees were noted and photographs taken of the

residues. Later samples recorded only recent feeding residues identified by comparison

with the photographs.

6.2.4 Marri Feeding

Marri feeding residues are a special case, because distinctive marks on the fruit husk

differentiate between feeding by Baudin’s Cockatoos and FRTBC (Figure 6.2).

Behavioural observations indicate that Carnaby’s Cockatoos only rarely feed on Marri

fruits at NBG (Lee et al. 2013b, Chapter 3). Therefore I chose to attribute all Marri

feeding residues to either Baudin’s Cockatoos or FRTBC, with the caveat that some

residues may have come from Carnaby’s Cockatoos.

108

Figure 6.2 Marri feed residues in the plot (a). Marri fruit husks displaying: (b) characteristic bite profile of Baudin’s Cockatoo and (c) shorn pattern of FRTBC. (d) – (f) illustrates how the condition of the residue, particularly its colour, can also indicate its age (i.e. the approximate time when the feeding event occurred) where - (d) 3 day old; (e) 3 month and (f) 7 month old Marri residue from Baudin’s Cockatoo ((d) – (f) are courtesy of Tony Kirkby, Western Australian Museum). To my knowledge, given the unique characteristics, type and location of the residue, no other species leave similar residue.

6.2.5 Assessment of Feeding Activity

I also used the 90 sampling plots sampled for feeding residues to characterise the

vegetation present in the rehabilitated mine pits. Within the pits, I measured a range of

community structural, floristic, individual plant structural and phenological variables

(refer to Chapter 5 - Table 5.2). I identified a subset of these variables as possible

indicators of black cockatoo feeding based on a literature review for birds feeding in

109

rehabilitated habitats: levy pole touches at each half meter from 0.5m to 3.5m, number

of live stems, crown width, height and phenological status (flowering and seeding) of

the five most abundant food plant species across the site

I measured the structural and floristic variables on the first sampling occasion only and

did not attempt to re-measure these variables during subsequent visitations to sampling

plots to check plots for feeding residues. My reasoning was that measurements of these

structural and floristic variables would not change significantly over the short time

intervals (i.e. 6 - 12 months) between the sampling sessions or over the overall study

duration (18 months) (L Mattiske, Mattiske Consulting Pty. Ltd., 2008, pers. comm.).

The study therefore represents a ‘snapshot’ assessment of vegetation structure and

floristics at a particular successional stage rather than a longitudinal assessment of

changes over time.

6.2.6 Data Analysis

Observational data for black cockatoos at rehabilitated mine pits were tabulated. These

document the numbers of black cockatoos seen, when and where they were seen, and

whether or not they were feeding or roosting.

Use of individual feed tree species by black cockatoos, as determined by feeding

residues, was represented with bar graphs. These indicated the number of 100 m2 plots

in which black cockatoo feeding residues for that feed tree species were observed for all

pits summed on each sampling occasion, and the number of plots in which black

cockatoo feeding residues for that feed tree species were observed for each pit summed

across all three sampling occasions.

I investigated the similarity in black cockatoo feeding between rehabilitation pits and

between interior and exterior plots, using non-metric multidimensional scaling (nMDS)

based on the Bray-Curtis similarity measure. The raw data for the feeding residues for

each feed plant species in the plots in each pit were first range-standardised to a range

of 0 – 1 to correct for the different scales of measurement used on individual feed plant

species (see Chapter 5 for details of range standardisation). I ran one nMDS to compare

feeding in interior and exterior plots across all sites combined, and a second nMDS to

compare feeding across all nine rehabilitation pits. I used stress statistics as indicators of

the adequacy of the final representations (Clarke and Warwick 2001), accepting stress

110

values less than 0.2 as adequate. These revealed an issue with several scarce feed plant

species across most plots that led to unacceptably high stress on nMDS plots. To

overcome this, I included a ‘dummy species’ with a value of 1 in all plots in the analysis

(Clarke et al. 2006).

Statistical significance of differences in feeding between pits and between interior and

exterior plots across pits on each sampling occasion was determined using a two-way

non-parametric multivariate analysis of variance (PERMANOVA or NPMANOVA)

with factors of Pit (the nine rehabilitation pits) and Location (the interior and exterior

plots) and dependent variables of the standardised feeding residues for each plant

species. This non-parametric analogue to traditional MANOVA provides an F statistic

with an associated p-value for testing the significance of each factor in the design and

the interaction between them (Anderson 2001). It can be based on any distance measure,

so I chose the Bray-Curtis measure to be consistent with nMDS. While post hoc tests

can be used for paired comparisons of levels within a significant factor (for example, to

test for differences between specific pairs of pits) I did not use them. My reasoning was

that the pits did not represent different experimental treatments, but showed the range of

variation arising when different pits were rehabilitated with the same approach.

Accordingly, I was most interested in whether or not the pits overall were similar rather

than in the detail of any differences that might arise. However, where PERMANOVA

was significant, I used the similarity percentage routine (SIMPER) to determine the

contribution of individual plant species to the difference. All analyses used the relevant

routines in the PAST software package (Hammer et al. 2001).

Lastly, I used nMDS and PERMANOVA to test if sampling plots where black

cockatoos fed differed from those in which they did not feed on the basis of the

variables from Table 5.2 (refer to Chapter 5). I used data from the first sampling

occasion only, on the basis that only phenological variables might differ markedly

across the three sampling occasions.

111

6.3 Results

6.3.1 Observations of Black Cockatoos in Rehabilitated Mine Pits

I recorded 52 visual or acoustic detections of black cockatoos during systematic surveys

of the 23 rehabilitated mine pits between October 2008 and July 2010. Visual detections

were of birds associated with vegetation in the pits (using the vegetation or flying

directly above it). Acoustic detections were of birds in the immediate vicinity of

rehabilitation vegetation (50 m was the distance). Carnaby’s Cockatoo was the most

frequently detected species (n = 28 of 52 detections, 53.8%), followed by Baudin’s

Cockatoo (n = 13 of 52 detections, 25%) and FRTBC (n = 11 of 52 detections, 21.1%)

(Table 6.1). Table 6.1 Number of occasions that one of the three black cockatoo species was detected, either visually or acoustically, during systematic surveys of 23 rehabilitated mine pits over 22 months of sampling.

Pit Carnaby’s Cockatoos Baudin’s Cockatoos FRTBC

1 3 0 1

2 1 0 0

3 3 0 0

4 4 0 0

5 1 0 1

6 1 1 0

7 3 0 0

8 2 0 0

9 0 0 1

10 0 0 0

11 2 0 0

12 0 2 0

13 0 1 0

14 0 1 0

15 0 2 1

16 0 1 0

17 1 0 0

18 1 2 0

19 1 0 1

20 0 0 1

21 1 0 1

22 0 1 2

23 4 2 2

Total 28 13 11

112

I saw Carnaby’s Cockatoos (n = 25 sightings, 71.4%) and Baudin’s Cockatoos (n = 10

sightings, 28.6%) feeding on or roosting upon rehabilitation vegetation between

October 2008 and July 2010. These sightings occurred during systematic pit surveys (n

= 10) or during opportunistic observations (n = 25). Birds were mainly feeding (n = 21

sightings, 60%) or short-term roosting (n = 12 sightings, 34.3%). Carnaby’s Cockatoos

fed on proteaceous shrubs (Hakea or Banksia spp.) (n = 11 sightings), Forest Sheoak

(Allocasuarina fraseriana) (n = 1 sighting), Radiata Pine (Pinus radiata) (n = 1

sighting), and Jarrah (n = 1 sighting). I also observed Carnaby’s Cockatoos feeding on

the ground among the rehabilitation vegetation on several occasions, though I could not

determine the food source. I only observed Baudin’s Cockatoos to feed on Marri (n = 8

sightings). I did not see FRTBC using rehabilitation vegetation, although I did see them

feeding in native vegetation directly adjacent to pits but not within the pits.

Group sizes for Carnaby’s Cockatoos ranged from one to 72 individuals, with a mean of

13.2 ± 3.2 birds and a median of 8 (25% = 3.3; 75% = 19). Group sizes for Baudin’s

Cockatoos ranged from three to 107 individuals, with a mean of 20.6 ± 9.8 birds and a

median of 13 (25% = 7.5; 75% = 18.5) (n = 10 sightings).

I observed raptors, the main predators of adult black cockatoos, soaring in the vicinity

of the rehabilitated mine pits six times during the systematic surveys (Wedge-tailed

Eagles; Aquila audax: n = 2 sightings; Square-tailed Kite; Lophoictinia isura: n = 1

sighting; unidentified raptors: n = 3 sightings). Raptors were observed chasing black

cockatoo flocks within rehabilitated mine pits on two occasions (Wedge-tailed Eagle: n

= 1 sighting; Square-tailed Kite: n = 1 sighting).

6.3.2 Patterns of Feeding Activity Revealed by Feeding Residues

Feeding residues occurred in most of the plots sampled. In the first sampling session,

Banksia squarrosa, Hakea undulata, and Marri had the highest standardised feeding

residues summed across all pits (Figure 6.3). In the second sampling session, the species

with the highest feeding residues summed across all pits were Marri, and the

proteaceous shrubs B. squarrosa, H. undulata, and H. prostrata. In the third sampling

session, Marri, H. undulata and B. squarrosa had the highest standardised feeding

residues summed across all pits (Figure 6.3). Considering the number of plots in which

residues were found across all sampling sessions, the number of species with feeding

residues present in a pit ranged from two (Pit 3) to eight (Pits 4 and 6) (Figure 6.4). H.

113

undulata and Marri were eaten in the most pits, with feeding residues at eight pits,

followed by B. squarrosa at seven pits (Figure 6.4).

114

Figure 6.3 The number of plots containing residues of particular food plant species summed across all pits for each of three sampling sessions: (a) July 2009, (b) January 2010 and (c) July 2010. Legend: Hu Hakea undulata, D = Banksia spp., M = Corymbia calophylla, BS = B. sessilis, HC = H. cyclocarpa, HT = H. trifurcata, HP = H. prostrata, HR = H. ruscifolia, HA = H. amplexicircus, BSQ = Banksia squarrosa, H = Hakea sp., BG = B. grandis, J = Eucalyptus marginata, HV = H. varia.

Figure 6.4 The number of plots containing residues of particular food plant species observed in each pit across all three sampling sessions. The legend is given in Figure 6.3.

115

116

On the first sampling occasion, PERMANOVA found significant differences in black

cockatoo feeding activity (as measured by the feeding residues from the 15 food plant

species) across pits (F8,89 = 4.04, p < 0.01), but not between exterior and interior plots

(F1,89 = 1.00, p = 0.39). The interaction between pits and interior and exterior plots was

not significant (F8,89 = 0.75, p = 0.87). The minimum convex hulls in two-dimensional

nMDS overlapped considerably across pits and also between interior and exterior plots

(Figure 6.5 (a) and (b)). This suggests that the differences between pits, while

significant, were slight.

Figure 6.5 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the first sampling occasion, based on the feeding residue counts for 15 feed tree species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.07241. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units.

117

Figure 6.5 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the first sampling occasion, based on the feeding residue counts for 15 feed tree species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.07421. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another rather than absolute differences read in units. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

Although there was much less time for residues to accumulate, there were also

significant differences across pits in the second (F8,89 = 3.53, p < 0.01) and third

sampling sessions (F8,89 = 3.28, p < 0.01). Interior and exterior plots did not differ in the

second (F1,89 = 1.26, p = 0.24) or the third sampling sessions (F1,89 = 1.66, p = 0.24).

The two-dimensional nMDS plots for pits and for interior an exterior plots indicate the

strong overlap between interior and exterior plots, but also suggest that the differences

between pits were small (Figures 6.6 (a) and (b), and 6.7 (a) and (b)).

118

Figure 6.6 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the second sampling occasion, based on the feeding residue counts for 15 feed plant species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.08648. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

119

Figure 6.6 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the second sampling occasion, based on the feeding residue counts for 15 feed plant species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.08648. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

120

Figure 6.7 (a) Non-metric MDS for the similarity in black cockatoo feeding across nine rehabilitation pits on the third sampling occasion, based on the feeding residue counts for 15 feed plant species. Here D4 = pit 1, K2 = pit 2, K6 = pit 3, L2 = pit 4, M2 = pit 5, M4 = pit 6, Q1 = pit 7, Q3S = pit 8 and WTR = pit 9. The similarity measure was Bray-Curtis and the stress was 0.1181. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for each pit. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

121

Figure 6.7 (b) Non-metric MDS for the similarity in black cockatoo feeding in interior and exterior plots across nine rehabilitation pits on the third sampling occasion, based on the feeding residue counts for 15 feed plant species (here interior pits are indicated by the solid line and filled square points, while exterior pits are indicated by the broken lines and unfilled square points). The similarity measure was Bray-Curtis and the stress was 0.1181. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for interior and exterior plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another. The dependent variables were the feeding residues from the 15 food plant species counted in each of the plots.

SIMPER showed that most of the variability across pits was caused by: (1) B. squarrosa

(28.95%), H. undulata (20.12%) and Marri (11.55%) in the first sampling session; (2)

B. squarrosa (22.51%), H. prostrata (22.39%) and H. undulata (18.31%) in the second

sampling session; and (3) Marri (50.86%) and H. undulata (18.75%) in the third

sampling session (Table 6.2).

122

Table 6.2 Results of SIMPER for the differences in feeding residues between pits on each of three sampling occasions: (a) winter 2009 (b) summer 2010, and (c) winter 2010. Note that the mean scores are based on variables range-standardised to a 0 – 1 scale. (a)

Taxon % Contribution

Cumulative %

Mean score D4 Pit 1

Mean score K2 Pit 2

Mean score K6 Pit 3

Mean score L2 Pit 4

Mean score M2 Pit 5

Mean score M4 Pit 6

Mean score Q1 Pit 7

Mean score Q3S Pit 8

Mean score WTR Pit 9

B. squarrosa 28.95 28.95 0.00 0.00 0.15 0.10 0.07 0.00 0.22 0.25 0.14

H. undulata 20.12 49.07 0.06 0.12 0.02 0.05 0.14 0.00 0.06 0.08 0.10

Marri 11.56 60.62 0.01 0.00 0.02 0.02 0.02 0.08 0.02 0.16 0.01

B. sessilis 8.16 68.79 0.00 0.13 0.08 0.00 0.05 0.00 0.00 0.00 0.00

H. prostrata 4.59 73.38 0.00 0.00 0.07 0.00 0.00 0.00 0.02 0.01 0.03

H. amplexicircus 4.18 77.56 0.00 0.00 0.04 0.00 0.00 0.00 0.10 0.00 0.00

H. incrassata 3.35 80.91 0.00 0.00 0.00 0.10 0.00 0.00 0.00 0.00 0.00

Jarrah 3.28 84.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10

Banksia sp. 2.93 87.13 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

B. grandis 2.78 89.91 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00

Hakea sp. 2.70 92.62 0.00 0.00 0.10 0.00 0.00 0.00 0.00 0.00 0.00

H. ruscifolia 2.56 95.18 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00

H. trifurcata 2.41 97.59 0.00 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. cyclocarpa 2.41 100.00 0.00 0.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. varia 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Total 20.04

123

(b)

Food Plant % Contribution

Cumulative %

Mean score D4 Pit 1

Mean score K2 Pit 2

Mean score K6 Pit 3

Mean score L2 Pit 4

Mean score M2 Pit 5

Mean score M4 Pit 6

Mean score Q1 Pit 7

Mean score Q3S Pit 8

Mean score WTR Pit 9

B. squarrosa 22.51 22.51 0.01 0.00 0.15 0.05 0.01 0.01 0.10 0.12 0.02

H. prostrata 22.39 44.90 0.00 0.17 0.09 0.08 0.00 0.00 0.06 0.00 0.05

H. undulata 18.31 63.21 0.00 0.04 0.03 0.03 0.13 0.00 0.02 0.10 0.02

Marri 9.78 72.99 0.01 0.00 0.00 0.05 0.01 0.01 0.01 0.04 0.05

H. varia 8.67 81.66 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.15 0.00

H. amplexicircus 4.85 86.52 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00

Jarrah 4.68 91.20 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00

B. grandis 4.57 95.77 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.00 0.00

B. sessilis 4.23 100.00 0.00 0.00 0.00 0.00 0.09 0.00 0.00 0.00 0.00

H. ruscifolia 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. trifurcata 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Hakea sp. 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. cyclocarpa 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. incrassata 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Banksia sp. 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Total 14.447

124

(c)

Food Plant % Contribution

Cumulative %

Mean score D4 Pit 1

Mean score K2 Pit 2

Mean score K6 Pit 3

Mean score L2 Pit 4

Mean score M2 Pit 5

Mean score M4 Pit 6

Mean score Q1 Pit 7

Mean score Q3S Pit 8

Mean score WTR Pit 9

Marri 50.86 50.86 0.12 0.01 0.03 0.01 0.03 0.06 0.08 0.04 0.01

H. undulata 18.75 69.61 0.00 0.02 0.00 0.00 0.00 0.00 0.09 0.02 0.01

B. squarrosa 10.07 79.68 0.00 0.00 0.01 0.02 0.00 0.00 0.02 0.01 0.01

H. cyclocarpa 8.21 87.89 0.00 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. amplexicircus 5.62 93.51 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00

B. sessilis 5.39 98.90 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. prostrata 1.10 100.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Jarrah 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

B. grandis 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. trifurcata 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Hakea sp. 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. ruscifolia 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. varia 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

H. incrassata 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Banksia sp. 0.00 100.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Total 6.31966

125

6.3.3 Feeding on Marri

Almost every Marri tree with feeding residue had been fed on by Baudin’s Cockatoos

(Table 6.3 (a) and (b)). In contrast, FRTBC fed on a smaller number of Marri trees, both

across the three sampling sessions (Table 6.3 (a)) and across the different pits (Table 6.3

(b)).

Table 6.3 The total number of Marri trees where feeding residues were detected (i.e. ‘feed trees’) for: (a) different sampling occasions and (b) different rehabilitated mine pits. The number of trees where those residues could be attributed to Baudin’s Cockatoos or FRTBC is also shown. Some trees contained residues from both species, so the total number of trees with feeding residues present is less than the sum value of feed trees for the two species. (a) Sampling occasions

Species Sampling 1 Winter 2009

Sampling 2 Summer 2009 - 2010

Sampling 3 Winter 2010 Total

Baudin’s Cockatoo 54 (98.2% 34 (97.1%) 66 (100%) 154 (98.7%)

FRTBC 8 (14.5%) 5 (14.3%) 3 (4.5%) 16 (10.4%)

Total 55 35 66 156

(b) Rehabilitation pits

Species Pit 1 D4

Pit 2 K2

Pit 3 K6

Pit 4 L2

Pit 5 M2

Pit 6 M4

Pit 7 Q1

Pit 8 Q3S

Pit 9 WTR Total

Baudin’s Cockatoo

9 (90.0%)

25 (96.2%) 0 13

(100%) 19 (100%)

13 (100%)

16 (100%)

40 (100%)

19 (100%) 154

FRTBC 1 (10.0%)

3 (11.5%) 0 2

(15.4%) 2 (10.5%)

4 (30.8%)

1 (6.3%) 0 3

(15.8%) 16

Total 10 26 0 13 19 13 16 40 19 156

6.3.4 Assessments of Feeding Activity

There was no difference between feed and non-feed plots in a range of structural and

floristic variables (F34,55 = 1.58, p =0.12). Three-dimensional nMDS plots (necessary to

reduce stress to an acceptable 0.1667) comparing feed and non-feed plots indicate a

high degree of overlap (Figure 6.8).

126

Figure 6.8 Non-metric MDS for the similarity in vegetation structure and floristics between feed and non-feed plots across nine rehabilitation pits on the first sampling occasion. The final solution is three-dimensional, shown with two separate two-dimensional plots. The unfilled square points indicate non-feed plots, while filled squares indicate feed plots. The similarity measure was Bray-Curtis and the stress was 0.1667. Minimum convex hulls, the smallest polygon incorporating all data points, are shown for feed and non-feed plots. Note that the distances along the axes are unit-less, and thus the positions of the points are relative distances from one another.

127

6.4 Discussion

My principal findings were that: (1) black cockatoos begin feeding in rehabilitated mine

pits within eight years of establishment and while vegetation is at an early successional

stage; (2) the three black cockatoos differed in their use of mine-site rehabilitation; and

(3) a combined approach of behavioural observation, assessment of feeding residues,

and vegetation sampling is useful for investigating the feeding activity of a large mobile

species occurring at low densities.

6.4.1 Observations of Black Cockatoos in Rehabilitated Mine Pits

Observational approaches confirmed the use of rehabilitation vegetation by Baudin’s

Cockatoos and Carnaby’s Cockatoos. However, the failure to detect FRTBC visually,

and the fact that all three black cockatoos were infrequently detected during the

systematic pit surveys, suggests that low-level (e.g. once monthly) observational

monitoring may be insufficient to confirm feeding. There were no conspicuous reasons

why detections at some pits were higher than others.

Birds used rehabilitation vegetation mainly for feeding, with Carnaby’s Cockatoos

eating seeds from proteaceous shrubs and Baudin’s Cockatoos taking Marri seeds.

Roosting behaviour was associated with feeding or with stop-over movements as birds

transited between habitats. As the canopy-forming trees in the pits occurred at high stem

densities and with low heights (<10 m) and narrow canopies (<5 m across), birds may

prefer to roost in mature trees in adjacent unmined forest, which provide better

concealment and shade.

The group sizes of Baudin’s Cockatoos and Carnaby’s Cockatoos within rehabilitated

mine pits were similar to those reported by Lee et al. (2013b) (Chapter 3) for nearby

remnant forest areas at NBG. Lee et al. (2013b) (Chapter 3) also reported that about

20% of sightings for Baudin’s Cockatoos and Carnaby’s Cockatoos at NBG occurred

within rehabilitation vegetation; most sightings for both species (71% and 59%,

respectively) were within native forest and woodland. This indicates that, for these two

species, the rehabilitated mine pits are an additional feeding habitat to those already

available at NBG and its surrounds.

128

6.4.2 Patterns of Feeding Activity Revealed by Feeding Residues

I found feeding residues in most sampling plots, indicating that black cockatoos feed

intensively within the rehabilitated mine pits at NBG. Residues were observed from

eight proteaceous species and two myrtaceous species, demonstrating the diversity of

food plants available in the pits. The discovery of new feeding residues during the

second and third sampling occasions also indicates that feeding activity occurs

throughout the year.

The presence of Marri feeding residues indicative of feeding by FRTBC confirmed that

FRTBC do feed within the rehabilitated mine pits, although the lack of sightings and

lower abundance of residues for FRTBC implies that FRTBC feed less in the pits than

the other two species. The lower feeding activity cannot be attributed to lower

abundance, as FRTBC were detected more frequently than Baudin’s Cockatoos and

Carnaby’s Cockatoos during site-wide surveys at NBG (Lee et al. 2013b, Chapter 3).

FRTBC may be more sensitive to predation risk from raptors than the other two species

(Weerheim 2008), and therefore may avoid open habitats, as is known for some black

cockatoos elsewhere in Australia (Chapman and Paton 2005; Cameron and Cunningham

2006). The incidental observations of raptors near rehabilitation plots demonstrate that

the risk of predation is apparent to birds. It remains to be seen if FRTBC will begin

using pits when they reach a later successional stage in which vegetation provides better

concealment and vantage points as well as Marri and other food plants.

Four species – Marri, H. undulata, H. prostrata and B. squarrosa – accounted for most

feeding residues, indicating that they represent the main food plants for black cockatoos

in rehabilitation pits. The predominance of Marri and three proteaceous shrubs is

consistent with the feeding habits of black cockatoos in southwestern Australia

(Saunders 1980; Johnstone and Storr 1998; Johnstone and Kirkby 1999; Cooper et al.

2002; Cale 2003; Chapman 2008; Johnstone et al. 2010; Biggs et al. 2011).

Feeding residues were not recorded for Acacia spp., Gastrolobium spp., Forest Sheoak

and Wandoo despite their abundance within pits. As black cockatoos rarely feed on

Wandoo (except for flowers), Acacia spp., or Gastrolobium spp., the lack of feeding

residues from these species/genera is unsurprising. However, FRTBC do feed on

Allocausurina spp. (Johnstone and Kirkby 1999; Pepper et al. 2000; Chapman and

Paton 2005, 2006) and it is unclear why they do not feed on Forest Sheoak in the

129

rehabilitation pits, where cones were observed on some trees. Similarly, there were far

fewer feeding residues of Jarrah than of Marri, despite other studies finding more even

use of the two food plants (e.g. Johnstone and Kirkby 1999; Biggs et al. 2011). These

differences could relate to the seed quality of regenerating Jarrah and Forest Sheoak.

There was some support for differences in feeding activity across pits, although the high

overlaps shown in the two-dimensional nMDS advise caution. Differences may relate to

variation in the occurrence and abundance of food plants. Some pits, for example, had

feeding residues from eight plant species while others had only feeding residues from

two or three plant species. Similarly, the number of Marri stems with feeding residues

varied substantially across pits. It is likely that variations in plant survival caused

differences in the relative proportions of species at different pits despite using the same

seed mix and rehabilitation approach. While edge effects may lead to differences in the

distribution and/or quality of food plants between interior and exterior plots, there was

little evidence of differences in feeding activity between interior and exterior plots.

6.4.3 Assessments of Feeding Activity

Feed and non-feed plots did not differ in a range of structural and floristic variables, so

no preferences are apparent at this scale. Other factors, such as pit location and pit size,

may be important. Black cockatoos may feed in particular pits more than others because

of their proximity to nesting or roosting sites, thereby reducing the energy spent in

travelling. The pits also varied in area (0.7 to 4.9 ha) and in shape, indicating

differences in, for example, the ratio of interior to exterior habitat within a pit and the

extent of the edge relative to the pit area. However, feeding activity in exterior and

interior plots was similar across pits, suggesting that any edge effects – if present – are

not substantial. However, the rehabilitated mine pits are small (i.e. no more than a few

ha) and more robust differences may emerge if pits are larger.

6.4.4 Management Implications

Revegetating disturbed sites can ameliorate the impact of native vegetation loss on

threatened fauna (Bennett et al. 2000; Munro et al. 2007). This study demonstrates that

establishing proteaceous food plants and Marri can provide feeding habitat for three

threatened black cockatoos within a few years. Thus, revegetation can provide tangible

benefits for black cockatoos quickly, emphasising the potential benefits of planting

130

black cockatoo food plants across the southwestern region (Hobbs and Saunders 1993;

Whisenant 1999; Hobbs and Lambeck 2002).

More generally, efforts to return fauna to disturbed sites are part of broader habitat

restoration strategies. At NBG, the aim of rehabilitating mine pits was to establish a

vegetation community similar to that in the surrounding native forest, a goal that can

only be achieved at a time-scale of decades (Koch and Hobbs 2007). Feeding activity

within early successional stage vegetation therefore represents a fortuitous outcome

allowing recolonisation of pits to occur within a decade post-establishment. This study

does not address whether rehabilitation vegetation adequately replaces undisturbed

native vegetation (Gould 2011). Equivalency must be assessed at appropriate time-

scales and requires comparative information on the quality and availability of food

plants in the two areas.

Restoration strategies must consider the patterns of vegetation succession likely to

prevail given site conditions and the plant species used (Norman et al. 2006; Walker et

al. 2007). As proteaceous species mature faster than the myrtaceous Jarrah and Marri,

Banksia and Hakea species soon established a thick shrub layer in the NBG pits,

providing a diverse mix of proteaceous seed and flower for black cockatoos. The

availability of proteaceous food plants is particularly important because of the apparent

time lag in the availability of suitable Jarrah fruit husks, which is a common food source

for all three black cockatoos at NBG (Biggs et al. 2011, Lee et al. 2013b, Chapter 3)

and elsewhere (Saunders 1980; Johnstone and Kirkby 1999; Johnstone et al. 2010).

Stems of regenerating Marri, though fed upon, do not maintain the large canopy

volumes (and therefore abundance of food) of mature trees. In eastern Australia, the

black cockatoo species Calyptorhynchus lathami halmaturinus and C. banksii

graptogyne prefer feeding in larger trees because they provide more food in a small

area, minimising energy-expensive movement between food plants in agreement with

optimal foraging theory (Reynolds 2012).

The phenology of food plants is also important to consider. Proteaceous species, which

are shorter-lived than myrtaceous species, tend to produce food (i.e. seeds, flowers) in

predictable seasons within the Jarrah-Marri forest (Koch 2007). In contrast, the

reproductive cycle of Marri and Jarrah depends on environmental conditions and

individual trees may only flower every few years (Nichols and Watkins 1984).

131

However, rehabilitation prescriptions, such as the use of water or fertiliser, may

encourage stronger growth and productivity, as well as atypical phenologies. The

intense feeding on Marri may reflect these factors.

As the rehabilitation at NBG ages proteaceous food plants are likely to decline as

myrtaceous trees mature and form a stronger canopy layer (Abbott et al. 1989; Williams

and Woinarski 1997). The foraging behaviour of black cockatoos is also likely to

change, with feeding on myrtaceous species by Baudin’s Cockatoos (and possibly also

FRTBC) becoming more frequent and feeding on proteaceous species by Carnaby’s

Cockatoos less common. This suggests two possible strategies for the restoration of

feeding habitat for black cockatoos: (1) the use of either proteaceous or myrtaceaous

species with the aim of maintaining a steady supply of particular food plants or (2) a

mixed-strategy employing both proteaceous or myrtaceaous species, with the

recognition that successional processes will change the mix of food plants available

over time. Climate change models predict reduced rainfall in much of the southwest, so

choice of plants for rehabilitation in future may need to consider drought tolerance to

ensure that rehabilitation can continue to provide food for black cockatoos within a

decade (Lee et al. 2013a).

There are four ways in which these results could be broadened. One would be to

compare the feeding of black cockatoos in rehabilitation with that in neighbouring

native vegetation. This would permit assessment of the significance of rehabilitation in

terms of food density in relation to native vegetation. A second area would involve

examining rehabilitation of older ages at other mines in the Jarrah-Marri forest. Such

comparisons would reveal changes in vegetation structure and feeding patterns related

to the maturing of rehabilitation vegetation. Thirdly, no account has been taken in this

study of the nutritional value of seeds or flowers from different plants, or of annual

differences in seed set. These could be part of a more detailed study of feeding in the

rehabilitation. Lastly, the use of rehabilitation by other bird species could be considered.

However, feeding is but one component of habitat quality for black cockatoos.

Although food resources return quickly, tree hollows suitable for breeding may take

over a century to form, so conservation of old, hollow-bearing trees is an essential

complement to restoring food plants. The significance of hollows is explored in

subsequent chapters.

132

Chapter 7 Suitability of Artificial Nest Hollows for Black

Cockatoos at Newmont Boddington Gold

7.1 Introduction

Removal of native vegetation is generally necessary if mining operations are to occur

(Hobbs 2000). This clearing may remove trees used for shelter and breeding by threatened

arboreal fauna such as black cockatoos (Lindenmayer et al. 2009). Black cockatoos are

particularly vulnerable to such impacts because they require large hollows in mature

eucalypt trees to breed (Saunders 1979; Saunders et al. 1982; Abbott 1998). Suitable

breeding hollows may take more than 150 years to form, raising concerns about their

availability in a forest landscape altered by logging and other forest management processes

(Wardell-Johnson et al. 2004).

Artificial nest hollows (ANHs) are one potential means of mitigating the impact of mining

on black cockatoo breeding habitat. ANHs have been used in eastern and northern Australia

to provide breeding habitat for glossy black cockatoos (Calyptorhynchus lathami) and other

large parrot species (Pell and Tidemann 1998; Garnett et al. 1999; Cameron 2006). In

Western Australia, there is little empirical information on the use of ANHs by black

cockatoos in the Jarrah-Marri forest. A recent Department of Environment survey of the use

of ANHs in southwestern Australia found that “(no) artificial hollows built for Carnaby’s

Cockatoos and placed in Jarrah-Marri forest areas were identified as being used by black

cockatoos during the study” (Groom 2010).

In Western Australia, the two main materials for constructing ANHs are polyvinylchloride

(PVC) piping and wood (Groom 2010; Chapter 8). Various ANH designs have been trialled

with black cockatoos in the wild and in cage settings, but with varying success rates (G

Byleveld, SERCUL1, 2009, pers. comm.; A Elliot, Serpentine-Jarrahdale Landcare Centre,

2009, pers. comm.; T Kirkby and R Johnstone, Western Australian Museum 2009, pers.

comm.; see Chapter 8). Little is known about what influences the use of ANH by black

1 South East Regional Centre for Urban Landcare

133

cockatoos, although the landscape-scale availability of natural hollows may be critical

(Groom 2010).

Chapters 3 and 4 presented evidence that the Newmont Boddington Gold (NBG) study area

supports breeding populations of Carnaby’s Cockatoos and FRTBC (and possibly Baudin’s

Cockatoos). As land clearing for mining is likely to impact on these populations it is

important to assess whether ANHs could provide at least short-term mitigation for the loss

of hollow-bearing trees by providing an artificial source of hollows. While ANHs cannot

provide long-term mitigation (unless they are maintained for the >150 years that it takes for

replacement trees to regrow), they may be valuable in the short-term, particularly if close to

suitable habitat features, such as feeding sites in rehabilitation areas and drinking sites. If

ANHs support the continued breeding of black cockatoos in the area, this may sustain local

populations and mitigate the overall impact of the breeding (and nearby feeding) habitat,

which was cleared for the expansion of the NBG mine.

To assess whether ANHs are a potential mitigation option, we need to determine whether

black cockatoos will actually use ANHs at NBG and what factors might influence their use.

To support this objective, I trialled two different ANH designs (PVC pipe and wooden box)

within the NBG study area. I monitored the use of these ANHs over the course of this study

to determine their use by black cockatoos. The two different designs were chosen to allow

for comparison of the use of the ANHs made from PVC piping versus those made from

wood. This chapter reports the results of this assessment and examines their implications.

7.2 Methodology

7.2.1 Artificial Nest Hollow Design

Two ANH designs were deployed: PVC tube piping (‘cockatubes’) and wooden box-type

design (Figure 7.1). The PVC ‘cockatube’ design was built by the Serpentine-Jarrahdale

Landcare Centre. These ANHs consisted of cylindrical black PVC tube with a diameter of

approximately 450 mm and a thickness of 2.4 - 2.8 cm. They were straight, 45° slanting

top-entry ANHs, measuring 1.2 m high at the back, tapering to 0.85 m high at the front. A

metal grill was attached to the inside of the front to act as a ladder for access by black

134

cockatoos. The ANHs were attached to trees through chains along either side and a single

chain at top of the ANH. The chains were secured with ‘screw-eyes’, shackles, and

stainless steel fence wire. These ‘cockatubes’ (as they are known colloquially) were filled

with a sacrificial wooden block visible from the ground (for birds to chew on) and leaf litter

from the forest floor nearby to provide nesting material.

Figure 7.1 (a) PVC ‘cockatube’ ANH, and (b) wooden box-type ANH installed at the NBG and LEA sites.

The wooden box-type ANHs were constructed on-site. These comprised a box design made

from wooden boards joined with non-toxic glue, reinforced with screws and two black

cable-ties/tie-wraps (Figure 7.1). They also had a protruding base and roof made from

wooden boards. These were top-entry ANHs. They measured 65 cm high in the front and 1

m high on the other three sides The internal dimensions of the ANH were roughly cubical

with a length and width of between 20 - 30 cm. Small wooden pieces were attached to the

inside of the front wall like rungs to allow birds to scale the wall upon entering and leaving

the ANH. The roof came up to two-thirds of the way to the top-entrance to allow for ease of

(a) (b)

135

access by the birds. The ANH was secured to a tree using stainless steel fence wire attached

to two pairs of ‘screw-eyes’ near the top and the bottom of both sides of the ANH.

7.2.2 Artificial Nest Hollow Placement

Twenty-four ANHs were set up within remnant forest areas across the NBG study area. Ten

‘cockatube’ ANHs were installed (n = 6 within the NBG site and n = 4 within the Land-

Exchange Area; LEA). They were installed in October and November 2008. Fourteen

wooden box-type ANHs were established within the eastern acquired lands in May 2010.

Details of the structure and placement of each box are given in Table 7.1.

136

Table 7.1 Details of placement of all ANH used at the NBG and LEA sites and surroundings.

Nest No. Construction Design Body material Top material Base material Other material Entry Front

length Back length

1 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 2 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 3 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 4 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 5 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 6 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 7 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 8 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 9 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 10 Cockatube Black plastic None Black plastic Chew post Top 0.85 1.2 11 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side - 12 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side - 13 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side - 14 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side - 15 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.64 1 16 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.64 1 17 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.64 1 18 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.65 1 19 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.65 1 20 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.66 1 21 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.67 1 22 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.67 1 23 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.66 1 24 Wooden box-type Wooden boards Wooden boards Wooden boards Chew post Side 0.69 1

137

Table 7.1 (continued).

Nest No. Inspection Use Proximity to

breeding Location Placement year Tree Host Live/Dead Entrance height

1 Monthly Not used 50 LEA site 2008 Marri L 10 2 Monthly Not used 50 LEA site 2008 Marri L 8 3 Monthly Not used 50 LEA site 2008 Jarrah L 6 4 Monthly Not used 50 LEA site 2008 Jarrah L - D 10 5 Monthly Not used 0.5 NBG site 2008 Marri L 10 6 Monthly Not used 0.5 NBG site 2008 Marri L 7 7 Monthly Not used 0.5 NBG site 2008 Marri L 9 8 Monthly Not used 0.5 NBG site 2008 Jarrah L 7 9 Monthly Not used 0.1 NBG site 2008 Jarrah L 10 10 Monthly Not used 0.5 NBG site 2008 Marri L 8.5 11 Monthly Not used 5 Eastern Remnant Forest 2010 - - 6.5 12 Monthly Not used 5 Eastern Remnant Forest 2010 - - 6 13 Monthly Not used 5 Eastern Remnant Forest 2010 - - 7 14 Monthly Not used 5 Eastern Remnant Forest 2010 - - 6 15 Monthly Not used 1 Eastern Remnant Forest 2010 Jarrah L 8 16 Monthly Not used 5 Eastern Remnant Forest 2010 Jarrah L 8.5 17 Monthly Not used 1 Eastern Remnant Forest 2010 Jarrah L 10 18 Monthly Not used 1 Eastern Remnant Forest 2010 Jarrah L 8.7 19 Monthly Not used 0.5 Eastern Remnant Forest 2010 Wandoo L 8.3 20 Monthly Not used 5 Eastern Remnant Forest 2010 Marri L 9 21 Monthly Not used 5 Eastern Remnant Forest 2010 Jarrah L 8.5 22 Monthly Not used 5 Eastern Remnant Forest 2010 Wandoo L 9 23 Monthly Not used 5 Eastern Remnant Forest 2010 Jarrah L 8 24 Monthly Not used 5 Eastern Remnant Forest 2010 Jarrah L 9

138

Placement was on living trees, with trees generally between 18 - 25 m in height and

diameters at breast height (DBHs) of >75 cm. ANHs were installed on three tree species:

Marri, Jarrah and Wandoo. These species were selected because they are the three most

common hollow-bearing trees within the study area and are known to be used for nesting

by black cockatoos. Where possible, ANHs were placed in forest areas within 3 - 5 km of

known or suspected drinking sites.

ANHs were set up using a trailer-borne elevated work platform (EWP) (Figure 7.2). This

EWP had a maximum working height of 10 m, thus restricting the installation height for

ANHs to 10 m or less. Trees were selected based on the presence of a large limb or major

fork in the trunk (at around 10 m in height) to provide a suitable platform for mounting the

ANHs. As natural limbs of trees did not always occur at 10 m of height, ANHs were set up

on limbs at heights between 7 – 10 m. Once in position on the tree, the ANH was attached

using chains, ‘screw-eyes’, and wire (depending on the design) (Figure 7.2). Aspect was

not considered.

139

Figure 7.2 ANH establishment on-site: (a) EWP used for installation; (b) cockatube on a Marri tree, and (c) wooden box-type ANH on a Jarrah tree.

(b) (a)

(c)

140

7.2.3 Artificial Nest Hollow Monitoring

Post-establishment, all the ANHs were monitored for use. From October 2008 to July 2010,

I visited the ANHs once monthly to inspect the hollows for occupancy or signs of use. This

involved scanning the rims or edges of the entry holes of each ANH (and wood on the tree)

with binoculars to note the presence of any claw or chew marks that may have been made

when black cockatoos were inspecting or using the ANHs. For the cockatubes, the ‘chew

blocks’ or sacrificial posts were also inspected for chewing that might suggest use of the

ANH. I also knocked the base of the tree on which an ANH was hung with a wooden stick

as this technique can cause nesting birds to emerge to observe the source of the noise. After

July 2010, irregular monitoring of the ANHs was continued by NBG personnel (note: as

January 2013, this monitoring had not identified any use of ANHs).

7.3 Results

I did not observe black cockatoos inspecting or using any ANHs. I did observe black

cockatoos in the vicinity of ANHs, including one occasion where black cockatoos were

feeding on Marri fruits on a tree bearing an ANH. Therefore the lack of use was not

because black cockatoos were avoiding the vicinity of ANHs.

7.4 Discussion

7.4.1 General Discussion

Chapters 3 and 4 presented evidence that at least low numbers of black cockatoos nest

within and around NBG tenements. The failure to observe use of ANHs at the site is

therefore curious. Studies of natural nest hollows, and of hollows and hollow-bearing trees

in the Jarrah-Marri forest, indicate that various factors could influence ANH use (Saunders

1982; Whitford 2002a; Whitford 2002b). Here I discuss potential explanations for why

birds may not have used the ANH installed at NBG. The lack of use of ANHs means that it

was not possible to make comparisons between the two ANH designs and I do not discuss

this issue further here.

141

Inadequate monitoring effort: It is possible that birds used hollows but that this use was not

documented. I consider this unlikely because: (a) I used the same observational methods

(visual inspection of hollows, ‘tree-knocking’) that were used to document the occupancy

of natural hollows at NBG, and (b) the survey effort was frequent enough (i.e. monthly) to

have provided at least two (and generally three) checks on hollows during a breeding

attempt (12 weeks from egg laying to fledging) (Saunders 1977; Bohner 1984; Smith and

Saunders 1986; Johnstone 1997). Nonetheless, I cannot exclude the possibility that some

hollows were used. Indeed, other users of ANHs have sometimes reported instances of

‘cryptic’ use of ANHs by black cockatoos (A Elliot, Serpentine-Jarrahdale Landcare

Centre, 2009, pers. comm.). Such concerns could be addressed by increasing the frequency

and duration of ANH monitoring surveys, particularly during breeding seasons (at least for

Carnaby’s Cockatoos) or by employing technology allowing for direct visual inspection of

hollows (e.g. a pole-camera device) (Nathan et al. 2001).

Inadequate local supplies of food and water: I also consider it unlikely that black cockatoos

did not use ANHs because food and water were unavailable. All six cockatubes used within

NBG tenements were located within 500 m of a known natural nest hollow and more than

half of the wooden box-type ANHs deployed in the eastern acquired lands were located

within 1 km of a known natural nest hollow. Moreover, ANHs were installed within a 3 - 5

km radius of known or suspected watering sites.

Disturbance: It is unlikely that the lack of ANH use could relate to the noise and

disturbance from vehicles because all ANHs were located on forest tracks that were very

infrequently used. Known natural nest hollows were also located near several of these

tracks.

Familiarity: It is possible that black cockatoos may not have used the ANHs because they

seemed foreign or unfamiliar and the birds were wary of coming close or using them (D

Stojanovic, Birds Australia, 2009, pers. comm.). Black cockatoos have been reported to

quickly locate and begin using ANHs, particularly if the ANHs are placed in a previously-

established nesting area that has been cleared of hollow-bearing trees (Groom 2010; R

Dawson, Department of Environment and Conservation, 2009, pers. comm.). However, in

other contexts, it may take many months and even years for ANHs to become occupied and

142

rates of ANH occupancy may therefore be quite low in the first years post-installation

(Brennan et al. 2005). This may be especially true for birds such as black cockatoos, as

ANHs can be only used once per nesting season and thus (depending on when the ANH

was installed) it may be at least a year or more before ANHs are found and used (Gibbons

and Lindenmayer 2002).

ANH structural characteristics: Certain characteristics of the ANH designs may have

deterred birds from using them. Features such as the ANH shape, nature of opening,

internal size/volume, dimensions, material or thickness may be unsuitable (Goldingay

2009). Other studies have reported that birds may show selective preferences for certain

ANH characteristics such as size of hollow opening (Brawn 1988; Gibbons and

Lindenmayer 2002). Aspect has not been shown to affect black cockatoo use of natural

hollows and is also deemed unlikely to affect artificial use (Saunders 1979; Nelson and

Morris 1994; Garnett et al. 1999). The thick walls of the PVC cockatubes may provide

good insulation in cooler periods (e.g. winter to spring), but create too high an internal

temperature during warmer periods (e.g. spring to summer) (A Elliot, Serpentine-Jarrahdale

Landcare Centre, 2009, pers. comm.).

These considerations must be weighed against the fact that the cockatube design has

attracted nesting birds in multiple locations (G Dewhurst, Black Cockatoo Preservation

Society, 2009, pers. comm.; A Elliot, Serpentine Jarrahdale Landcare Centre, 2009, pers.

comm.; P Mawson, Department of Environment and Conservation, 2009, pers. comm.; C

Minton, Murdoch University, 2010, pers. comm.; R Scott, BirdLife Australia, 2009, pers.

comm.). Although the wooden box-type ANH design used in this study differs slightly in

design from others used in Western Australia, black cockatoos do use wooden nest-boxes

(R Dawson, Department of Environment and Conservation, 2009, pers. comm.; R

Johnstone and T Kirkby, Western Australian Museum, 2010, pers. comm.).

ANH position and location: The lack of ANH use could be because the ANHs were placed

too low on the trees. Birds may prefer hollows situated higher than 10 m, particularly if

natural hollows tend to occur above this height in the surrounding landscape (T Kirkby,

Western Australian Museum, 2009, pers. comm.). Furthermore, low ANH heights may be

unappealing to black cockatoos because of a greater perceived risk of predation by ground-

143

based predators (Menkhorst 1984). Against this is the fact that, at NBG, suspected natural

nest hollows in Wandoo were often less than 7 m above the ground.

Locale may be an important factor in determining whether ANHs will be used, as black

cockatoos are known to display nest site-fidelity and to return to the same breeding area

(and often to the same hollow) (Saunders 1982, 1986; Johnstone and Storr 1998). I would

therefore expect that the pairs who might use the ANHs at NBG are likely to have bred in

the area previously and may, for example, have lost a previously used hollow, either to

natural deterioration or because of clearing at the site.

Sufficient availability of natural hollows: The lack of ANH use suggests that the NBG

landscape may contain adequate numbers of natural hollows to support breeding pairs

wishing to nest in the area. Groom (2010) found that use of ANH tended to be highest in

areas where the availability of natural nesting hollows was limiting. Johnstone and Kirkby

(2009) noted that not all breeding pairs in a local black cockatoo population may nest in a

given year and therefore only a small proportion may require hollows. Thus, the apparently

small breeding populations in the NBG area may not require ANHs to support breeding. To

test this hypothesis at NBG, better information is needed on hollow availability and

breeding populations in the area. This would require improved techniques to assess hollow

suitability.

7.4.2 Management Implications

Advocates of ANHs argue that they are appropriate if natural hollows are being lost faster

than they are replaced (Cameron 2006; Durant et al. 2009) In contrast, others such as

Lindenmayer et al. (2009) and Harley (2006), caution that adopting ANHs should be based

on an ecological and economic cost-benefit assessment, comparing the costs of establishing

and maintaining ANHs in the long term against the costs of improved landscape

management to protect and replenish natural hollows. Another view is that ANHs are costly

and temporary with limited long-term benefit (Harper et al. 2005). There is little doubt that

in-situ conservation of hollow-bearing trees is preferable to using ANHs as a mitigation

method for land clearing. However, given the reality of land clearing to support human

activities, it is important to determine the circumstances in which ANHs are used for

nesting and therefore provide a tangible conservation benefit.

144

Natural hollow availability at NBG is likely to decrease in the coming decades through: (1)

logging of State Forest to the west and north of NBG; (2) bauxite mining by Worsley

Alumina in 10 - 20 years; (3) possible logging of private forest to the south and to the east;

and (4) possible changes in the abundance of hollow competitors such as feral European

honey bees (Apis mellifera) and Galahs (Cacatua roseicapilla). Thus, future generations of

black cockatoos are likely to encounter a more fragmented landscape in which the number

of suitable nesting hollows is reduced, perhaps substantially so. ANHs, if they can be

attractive to black cockatoos and are economically viable to establish and maintain, may be

important for sustaining current breeding populations at the site in the face of these

changes. In the longer term, rehabilitation efforts and mine closure planning should

consider how best to conserve and, if possible, replace hollow resources.

In the next chapter, I present the findings of a survey of ANH-users in southwestern

Western Australia and review the key considerations for the installation of ANHs and the

factors influencing their use by black cockatoos. Here I briefly consider four specific issues

relevant to the deployment of ANHs at NBG: (1) durability and maintenance, (2)

placement, (3) control of nest hollow competitors, and (4) monitoring.

Durability and Maintenance: The cockatube and wooden box-type ANH designs used at

NBG can be difficult to install because of their size and weight (25 - 30 kg). The systems

needed to attach or replace ANHs are time-consuming and logistically taxing. The synthetic

cockatube is more durable, fire-resistant (J Lee, Murdoch University, pers. obs.), and can

last almost indefinitely. However, black cockatoos may chew through the sacrificial posts,

so these need replacement (A Elliot, Serpentine Jarrahdale Landcare Centre, 2009, pers.

comm.; T Kirkby, Western Australian Museum, 2009, pers. comm.). Similarly, wood-based

ANHs may require repair if the wood is chewed by the birds. In general, wood-based

ANHs do not last as long as the cockatubes (R Johnstone, Western Australian Museum,

2009, pers. comm.), as they may be damaged by rain, wind, temperature changes, abrasion

against the tree, insect or fungal attack or rusting of the screws and bolts with time (Smith

2004). Additionally, accretion of waste and nesting material will fill both types of ANH

and parasites or ants may accumulate, necessitating clearing (Rhind 1998; Gibbons and

Lindenmayer 2002; Smith 2004).

145

Another consideration is the need to maintain ANHs over the life cycle of the mine and

beyond mine closure. The longevity of these ANH designs relates both to the ANH

materials and to the circumstance of their being attached to the surface layer of bark and

heartwood of living (or dead) trees. One of the cockatubes installed at NBG fell when the

branch it was attached to broke off from the main stem of the tree (Figure 7.3).

Figure 7.3 ANH fell when a dead stag fell; ANH was subsequently relocated onto another tree.

Placement: Increasing the height of the attachment may improve the use of ANHs at NBG,

as this would (presumably) ensure that ANHs are installed at a height similar to that of

natural hollows in the surrounding forest (at least for ANH installed in Jarrah and Marri

trees). However, increasing the heights creates logistical challenges, both in terms of the

equipment needed to install ANHs and occupational health and safety regulations, given the

weight of ANHs and the risk inherent to attaching a heavy object to a tree at height (Smith

2004). One alternative might be to use ANH installed at the end of a metal pole. The pole

could then be lowered for maintenance.

146

Nest Competitors: Hollow competition from exotic species such as feral bees (Apis

mellifera) and native Galahs (Cacatua roseicapilla) (Gibbons and Lindenmayer 2002)

maybe important in the future, particularly as climate change advances. Currently, these

species are not present in sufficient abundance to cause concern, as there have been very

few observations of these species at the NBG study area. The relatively open-air style of

ANH design and material (such as plastic) is also thought to thwart the use of ANHs by

pests such as bees. However, bees have become established in cockatubes recently installed

in the Perth metropolitan area (C Minton, Murdoch University, 2009, pers. comm.).

Monitoring: Improved monitoring of ANHs would allow for better documentation of use

(or non-use). This monitoring would need to be maintained over multiple years because

ANH use may be delayed for one or more years until birds become familiar to them.

Another consideration is the need to maintain ANHs over the life cycle of the mine and, if

their use is continued, in the period beyond mine closure.

7.4.3 Concluding Remarks

If ongoing monitoring indicates that ANHs are used for nesting, their implementation

should become part of the rehabilitation and mitigation protocol at NBG. However, they

have a limited lifespan and cannot replace all the attributes provided by natural hollows.

Therefore ANHs are best seen as a short- to medium-term mitigation measure until the

restored habitat provides natural hollows.

If nest-boxes are not used by black cockatoos, then it may be appropriate to consider

alternatives such as translocating hollow-bearing sections of felled trees or creating hollows

via human manipulation (Brennan et al. 2005).

147

Chapter 8 Are Artificial Nest Hollows a Suitable Method for

Mitigating Impacts of Mining on Black Cockatoos in the

Jarrah-Marri Forest? – A Survey of the Users of Artificial Nest

Hollows in Western Australia

8.1 Introduction and Problem Formulation

Land clearing, modified fire regimes, introduced plant pathogens, selective removal of

some tree species during timber harvesting, range expansions of invasive or feral

species, and cutting of mature or dead trees for firewood or woodchips have all reduced

the availability of tree hollows for fauna in southwestern Australia, either by altering

processes of tree senescence or by removing hollow-bearing trees (Calver and Dell

1998; Wardell-Johnson et al. 2004). This affects black cockatoos because they require

large hollows in mature trees for breeding (Saunders 1977, 1979; Saunders et al. 1982;

Cameron 2007).

Black cockatoos require the largest hollows of all arboreal fauna in southwestern

Australia because of their large body size and the need to prevent their feathers from

being damaged or lost while occupying a hollow (Whitford 2001; Abbott and Whitford

2002; Cameron 2007). Large hollows take longer to form (Abbott and Whitford 2002)

and tend to form in the trunk or primary branches of trees, rather than in secondary

branches (Whitford 2002a). These factors make hollows suited to black cockatoos less

common than those for other arboreal fauna (Abbott and Whitford 2002). Furthermore,

black cockatoos display strong nest site-fidelity and return to the same nesting sites in

subsequent breeding seasons (Saunders 1986; Cameron 2007). Thus, loss of traditional

nesting sites may affect breeding success for several years, particularly if pairs are not

able to locate new breeding hollows in the surrounding landscape.

Reductions in hollow abundances through land clearing and other practices have

increased competition for available nest hollows amongst breeding black cockatoo

pairs, particularly given increasing numbers of feral species (e.g. feral bees; Apis

mellifera) (Johnstone and Kirkby 2007) and superabundant native pest species (e.g.

Galahs; Cacatua roseicapilla) in forest and woodland habitats (Barrett et al. 2003;

Johnstone and Kirkby 2004; Johnstone et al. 2010). This limits black cockatoo

148

abundance at sites with little native vegetation remaining, and may cause some to use

suboptimal nest sites, with a concomitant reduction in reproductive success (Saunders

and Ingram 1987). This lack of nesting hollows is suggested as the greatest factor

contributing to the decline in black cockatoo numbers (Groom 2010; Garnett et al.

2011).

Unfortunately, re-planting of tree species known to bear hollows can only offer long-

term mitigation for the reduction in hollow abundance. This is because of the time lag

(>150 years) between when trees are planted and when they might be mature enough to

bear a hollow suitably large for black cockatoos to use (Whitford and Williams 2002;

Goldingay 2009). Thus, it is necessary to consider strategies that would satisfy the

current needs of black cockatoos for hollows in which to breed.

One option is in-situ retention, principally by retaining mature, senescent, or known

hollow-bearing trees (i.e. ‘habitat trees’) at specific densities in production areas

(Conservation Commission of Western Australia 2003). Given the long time lag before

the regrowth of harvested coupes will produce trees with potentially suitable hollows

(Manning et al. 2013), it is questionable whether current prescriptions for the retention

of habitat trees are adequate to keep birds breeding while the regrowth matures.

Perpetuation of hollow-bearing trees includes retaining existing hollow-bearing trees

(e.g. residual) and/or recruitment trees (i.e. trees presently without hollows but are

likely to develop them in the future) (Gibbons and Lindenmayer 1996, 1997b). Tree

retention/recruitment prescriptions should be guided by sound research (Hunter and

Bond 2001), and need to take into consideration:

(i) Selection or choice of trees - the tree type/species known or most likely to develop

hollows suitable for fauna in a given area, the size/age (e.g. larger-diameter trees of

various ages), number (minimum required to support functionally viable

populations of the full range of hollow-dependent taxa given their ecologies) and

location or spatial arrangement of trees chosen to be retained (e.g. even and random

distribution), as well as the retention level appropriate for a particular landscape

(Gibbons and Lindenmayer 2002);

(ii) Protection of retained/recruited trees – trees selected for retention need to have

good prospects of survival (e.g. remain standing) and thus must be in relatively

good and healthy physiological condition. Henceforth, they have to be protected

149

from pathological disease conditions (e.g. dieback) and fire (e.g. buffer zones)

(Gibbons and Lindenmayer 1997a, 2002).

Current retention prescriptions for hollows are often lack adequate and detailed

information, and generally standardised over large areas of heterogeneous forest. This

makes them too generic and inappropriate for the site and its expected, often diverse,

assemblage of hollow-dependent fauna and their habitat requirements. This paucity may

stem from an under-utilisation of the literature or a genuine deficit of available

information (Gibbons and Lindenmayer 1996, 1997a). For a more in-depth review and

summary on the present prescriptions employed in Australia, refer to Gibbons and

Lindenmayer (1997a, 1997b, 2002).

Promoting or creating natural hollows in standing trees is a second option. I am

unaware of any published study for Western Australia that has attempted to

experimentally promote or create natural hollows in standing trees, although some work

has been done elsewhere (Brennan et al. 2005). The technique is practically and

logistically onerous and is unlikely to be appropriate for anything more than small-scale

remediation efforts.

However, there are a several ways in which hollow development may be promoted in

live, standing trees and, while many of these techniques have been trialled or assessed

overseas (e.g. North America) and there is a lack of such information and studies in

Australia (Adkins 2006). They include: (i) pheromone application to bait and attract

decay-promoting invertebrate infestations (e.g. termites) (Bull and Partridge 1986; Shea

et al. 2002). (ii) Inoculation of fungi or growth hormones (e.g. Connor et al. 1981;

Parks et al. 1995; Shea et al. 2002; Cockle et al. 2012; Bednarz et al. 2013). (iii)

Deliberately producing holes in trees through drilling or cutting wood (e.g. ringbarking,

stub-pruning, limbing, boring or routing) using chainsaws or explosives (Shepherd

1957; Bull et al. 1981; Carey and Sanderson 1981; Bull and Partridge 1986; Lewis

1998; Adkins 2006). (iv) Mechanically injuring or killing the trees through sub-lethal

poisoning (e.g. herbicides) (Connor et al. 1983; Bull and Partridge 1986; Whitford et al.

1995; Ryan 1996; Adkins 2006), as well as girdling or topping that will attract infection

by microorganisms such as fungi or bacteria and accelerate hollow formation (Hennon

and Loopstra 1991; Connor et al. 1981; Carey 1993; Lewis 1998; Hallett et al. 2001;

Shea et al. 2002). For example, Carey (1993) described the deliberate topping and even

150

killing of trees to provide habitat trees for Spotted Owls. And (v) the use of fire to

damage trees, therefore catalysing hollow formation (Hutto 1995; Adkins 2006).

However, research into the applicability and effectiveness of these techniques is still

required and this requires monitoring of use of these artificially engineered cavities

(Gibbons and Lindenmayer 2002). For eucalypt species in Australia that have a shorter

standing life when dead, techniques that injure rather than kill trees may be more

applicable as they will increase the standing life of trees. Also, given that hollow

formation in eucalypts is a slow process because of the lack of primary excavating

fauna (Gibbons and Lindenmayer 1996). Furthermore they should be concentrated on

larger trees that are more able to withstand the damages and are also more likely to

produce hollows for a greater range of faunal species (Gibbons and Lindenmayer 2002;

Adkins 2006). Strategies that may be applied at a landscape level may be more cost-

effective and appropriate economically and logistically for a country like Australia that

also regularly applies prescribing burning (Gibbons 1999; Adkins 2006).

Artificial nest hollows (ANHs) represent a third option to supplement the availability of

natural hollows in the short- to medium-term. Until recently, the factors affecting the

use of ANHs by the Western Australian black cockatoos were not researched in detail,

although use of some designs was reported by Davies (2005) and Scott (2009).

However, important features may be inferred (to some extent) from the factors known

to influence the use of the natural hollows by the species and their general nesting

requirements (Saunders et al. 1982; Abbott and Whitford 2002). Studies of black

cockatoos in eastern Australia (Emison 1996; Pedlar 1996; Garnett et al. 1999; Jarmyn

2000; Conners and Conners 2005; Cameron 2007) offer further insights. There are also

a few reports of ANH use in the ‘grey’ literature and several anecdotal accounts

(Conners and Conners 2005). Most of the ANHs reported in the ‘grey’ literature were

constructed on an ad hoc basis (Davies 2005), and such ‘trial and error’ approaches

usually lead to varying results. More extensive and detailed information is important for

assessing the potential application of ANHs in assisting black cockatoo conservation.

To collate existing anecdotal information and to establish new records where nothing

has been published, Groom (2010) undertook a substantial survey of ANHs for

Carnaby’s Cockatoos. She established accurate locations for 239 ANHs for Carnaby’s

Cockatoos, while a further 76 ANHs lacked locational information. She used site visits

and interviews with people using ANHs to collate detailed data on the design,

151

placement and construction of ANHs. Not all ANHs had been monitored to determine

use by Carnaby’s Cockatoos and imperfect data records restricted some descriptions

and analyses. Successful ANHs were concentrated in the northern agricultural areas and

occurred on a range of land tenures. There were problems with occupation of ANHs by

feral species such as honeybees (Apis mellifera) and abundant native species such as

Galahs (Cacatua roseicapilla).

Groom (2010) reported three main designs of ANHs: (1) ‘cockatube’, (2) Pedler (see

Pedler 1996) and (3) Kerkoff (see Davies 2003), as well as other less common custom

designs. While the Kerkoff appeared the most effective, this may be an artefact of their

location relative to known nesting sites. Groom (2010) found that successful design

features included:

Durable walls and bases offering good thermal insulation;

Top-entry entrances to reduce use by non-target species;

Ladders for access;

Sacrificial chewing posts (which the birds use to provide shavings for

nesting); and

Stable mountings.

To improve the information available on ANH use by all three black cockatoos in

southwestern Australia, I extended the information presented in Groom (2010) by

interviewing and surveying (via questionnaire) all known users (both lay and

professional) of ANHs for black cockatoos in Western Australia to collect information

on the use of ANHs, as well as their success in getting birds to breed. On the basis of

this information, I sought to:

Describe the design, installation, maintenance and location features of those

ANHs that were successful in encouraging birds to breed;

Highlight any risks to black cockatoos or to workers in establishing and

maintaining ANHs;

Identify current gaps in knowledge, so as to stimulate further research and

experimentation;

Assess whether ANHs provide a suitable method for mitigating the loss of

natural hollows associated with mining in the Jarrah-Marri forest.

152

8.2 Methodology

All survey work was conducted under Murdoch University Human Research Ethics

Committee Permit 2009/058.

8.2.1 Design of Survey

I devised 26 questions relating to the use of ANHs by black cockatoos in Western

Australia, organised around six major themes:

I. Installation, inspection, and monitoring

II. ANH design and materials

III. Maintenance and costs

IV. Location and tree characteristics

V. Use of ANHs by black cockatoos and other species

VI. Other outcomes

I then constructed a questionnaire (survey form) using these themes and questions (see

Appendix II). Black cockatoo researcher (Dejan Stojanovic) from Birds Australia

Western Australia collaborated in discussions to finalise the questions forming the core

of the questionnaire. In collaboration with Dejan Stojanovic, I established a list of

individuals working with black cockatoos who might be involved in the use of ANHs,

including the use of ANHs in captive situations and in natural environments. I then

contacted potential respondents by telephone or email to ask if they were willing to

participate in the survey. Thus, the survey group was self-selected (i.e. it included only

those individuals who agreed to be involved and returned a questionnaire).

8.2.2 Survey Procedure

Participants were interviewed by email, telephone, or in-person visits. The outline of the

interview schedule was:

1. Pre-interview: All prospective participants were contacted via email, letter or

telephone to introduce the project and invite their participation. No attempts were made

to persuade individuals who declined. Also, I explained to all participants that they

would not be personally identified in the project if they did not wish to be, and that they

should not feel compelled to answer all the questions if they did not wish to.

Furthermore, prior approval would be sought before acknowledging them by name in

any written publication.

153

2. Interview: When convenient to all parties, I visited participants in-person. Otherwise,

interviews were done by phone or via email. Only written notes were taken. No

structured analysis of the data was performed.

3. Post-interview: Summaries of responses were sent to participants by email or letter

for checking. Corrections were returned for further checking and approval.

8.2.3 Presentation of Findings

Several factors precluded the use of quantitative analyses, including the small sample

size of participants, the ‘pre-selected’ (i.e. involving only people willing to participate

in the survey) nature of the survey group, and the fact that many respondents did not

answer all questions. Interpretations were therefore based on qualitative analyses of

responses. I organised the information from the responses around eight major themes:

i. Survey participants

ii. History of implementation and designs

iii. Locations where ANHs were used

iv. Characteristics of successful ANHs

v. Installation, maintenance and cost of successful ANHs

vi. Risks to black cockatoos or workers from ANH designs

vii. Risks to ANHs

viii. Respondents’ priorities for future research

8.3 Results

8.3.1 Survey Details

Twelve of the 14 people approached agreed to be interviewed; the other two individuals

declined because they were too busy. Interviews and follow-up discussions were

conducted between March 2009 and December 2010.

8.3.2 Survey Participants

The list of survey participants indicates the range of individuals and organisations

involved in the use of ANHs for black cockatoos in Western Australia (Table 8.1). The

list includes representatives from what are (presumably) all of the key stakeholder

categories for black cockatoo conservation in Western Australia: State management

154

agencies, research institutions, environmental organisations, catchment management

and Landcare groups, other Non-Government Organisations (NGOs) and industry.

Independent lay-people and professionals were also represented. The range of interested

parties suggests there is considerable interest in the issue of whether ANHs can be used

to support black cockatoo breeding efforts.

The participants from Landcare Jarrahdale-Serpentine kindly provided details on ANHs

established by a range of other community groups. Those data are included in the

discussion of ANH installation and longevity/removal, but no other information was

available.

155

Table 8.1 Titles and organisational affiliations appear as they were at the time the survey was undertaken. *The superscript number next to the name is for future references to personal communication (pers. comm.) in the text of this chapter.

Participant Title, Organisation and Contact Email

Dejan Stojanovic1 Cross Regional Project Officer Birds Australia Western Australia (BAWA) dejan.stojanovic@anu.edu.au

Raana Scott2 Carnaby’s Black Cockatoo Recovery Project Manager (South Coast) Birds Australia Western Australia (BAWA) r.scott@birdsaustralia.com.au

Peter Mawson3

Wildlife Conservation Principal Zoologist Fauna Species and Communities Branch, Department of Environment and Conservation (DEC) Peter.Mawson@perthzoo.wa.gov.au

Rick Dawson4 Nature Protection Branch Senior Investigator Department of Environment and Conservation (DEC) Rick.Dawson@dec.wa.gov.au

Ron Johnstone5 Curator - Ornithology Western Australia Museum (WAM) Ron.Johnstone@museum.wa.gov.au

Tony Kirkby6 Biologist Western Australia Museum (WAM); Water Corporation (WC) akirkby@vtown.com.au

Allan Elliot7 Landcare Jarrahdale-Serpentine (LCJ-S) (with Shire of Serpentine-Jarrahdale and LotteryWest) kingia@reachnet.com.au

Glen Byleveld8 Landcare Jarrahdale-Serpentine (LCJ-S) (with Shire of Serpentine-Jarrahdale and LotteryWest) GlenByleveld@sercul.org.au

Glenn Dewhurst9 Black Cockatoo Preservation Society (BCPS) dewy3@bigpond.com

Caroline Minton10 Environmental Programme Manager Office of Commercial Services; Murdoch University (MU) c.minton@murdoch.edu.au

Steven Davies11

Adjunct Professor Murdoch University and Curtin University (CU) Moore River Catchment Group; World Wildlife Fund; Men Of Trees S.Davies@exchange.curtin.edu.au

Hugh Finn12 Postdoctoral Fellow Murdoch University (MU); Newmont Boddington Gold (NBG) h.finn@murdoch.edu.au

156

8.3.3 History of Implementation and Designs of Artificial Nest Hollows

The earliest ANHs recorded in this survey were established from 1996 and the most

recent in 2010. Although exact numbers were unknown for some installations, 79% of n

= 414 ANHs were installed from 2000 onwards (Table 8.2).

Respondents reported using three general types of ANHs, although minor modifications

of each type sometimes occurred (Table 8.3). These types were:

(1) ‘cockatubes’ (industrial polyvinyl chloride piping chained to trees);

(2) hollowed-out log sections which are raised on metal poles; and

(3) wooden box-type ANHs

Only a single wood-based ANH was treated with insecticide (WAM). All the

cockatubes had holes drilled into the bases for drainage. The hollowed-out log sections

on poles were either bolted or cemented into the ground. Some had lids on the top while

others did not. The wooden box-type ANHs were made of plywood (a softwood) or

Jarrah (a hardwood) and bases and lids were made of tin or wood. Some wooden box-

type ANHs had a natural hollow entrance included as the opening.

8.3.4 Locations where Artificial Nest Hollows were Used

BCPS was the only organisation to use ANHs within a captive setting (n = 14 ANHs).

ANHs were placed between 1 - 1.5 m in height within aviaries. All other participants

used ANHs in natural settings.

ANHs in natural settings were placed within a range of land tenures. Most were situated

on private property, either on remnant trees within cleared land or in remnant woodland

or forest plots. Few ANHs were placed on public (Crown) land; these were situated in

State Forest/Crown Reserve/National Parks (DEC; LCJ-S) or in natural remnant

vegetation or along roadside verges (WAM).

The choice of tree species for ANH placement usually reflected the dominant tree

species present in the area (e.g. Jarrah or Marri trees in the Jarrah-Marri forest). ANHs

on poles were placed adjacent to trees (CU, BAWA). Within Swan Coastal Plain and

the Jarrah-Marri Forest, ANHs were placed mainly on mature Jarrah, Marri, Tuart, and

Wandoo trees, as well as non-native Pine trees, while Salmon Gum and Wandoo trees

157

were utilised in woodlands within the Wheatbelt. Most of the ANHs were placed in

living trees, with only one ANH placed in a dead tree.

ANHs were placed on the main trunk or the largest fork, except for one ANH placed on

a stump (WAM). The height of ANHs typically ranged from 5 - 10 m, with the lowest

set at 2 m (BAWA) and the highest at 17 m (LCJ-S). While placement height in trees

depended on the general height of the local canopy, it was restricted by accessibility,

safety, and the nature of equipment utilised (e.g. cherry-picker or Elevated Work

Platform; EWP) The average natural hollow height in the surround vegetation was

generally used as a guide for placement. Only two groups faced ANHs northeast

direction to reduce effects from the elements (i.e. prevailing weather).

158

Table 8.2 History of ANH installation by participants between 1996 and 2009. *- ANHs were provided by LCJ-S.

No. Participant and Organisation

No. of ANHs installed Year installed Duration since set up (as

at time of survey in 2009)

1 Dejan Stojanovic BAWA (Regional) 22 From 2002 From 9 years

2 Raana Scott BAWA (South Coast)*

35 3 in 2006 25 in 2008* 7 in 2009

From 5 years From 3 years From 2 year

3 Peter Mawson DEC >100 From 1996 From 15 years

4 Rick Dawson DEC <50 From 2003 - 04 From 7 - 8 years

5 Ron Johnstone WAM

1 20 5 2

From 2002 From 2007 and 2009 From 2006 and 2007 From 2005 - 06

From 9 years From 4 and 2 years From 5 and 4 years From 5 - 6 years

6 Tony Kirkby WAM and WC

7 21 6 1

2000 2002 2003 2004

From 11 years From 9 years From 8 years From 7 years

7 Glenn Dewhurst BCPS 14 2007 From 4 years

8

Glen Byleveld & Alan Elliot LC S-J (with Shire of Serpentine-Jarrahdale and LotteryWest)*

34 30

From 2005 - 06 From 2007 - 08

From 4 - 5 years From 3 – 4 years

9 Caroline Minton Murdoch University* 6 From 2009 From 2 year

10 Steven Davies Curtin University 40 From 2003 From 8 years

11

Hugh Finn Murdoch University Newmont Boddington Gold*

10* 10

From 2008 From 2010

From 3 years From 1 years

159

Table 8.3 Design, placement and use by black cockatoos of ANHs placed by survey participants.

Worker ANH Design Materials and Dimensions Location Placement and Height

Arrangement and distance from other hollow (natural or artefact)

BAWA Hollowed-out log sections erected on metal poles

Cut-off sections of natural tree hollows attached to a metal lid and base atop a metal pole. On average- Internal diameter: ~300 mm and ~500 mm tall. Poles: 3 - 5 m long

Woodland remnants in private landholds

Erected on 3 – 5m long poles

Arranged in clusters across the site ≥10 m

BAWA* Cockatube

Industrial (high density) PVC piping. Slanting open top with sealed base chained to trees. Internal wire mesh for climbing and sacrificial chewing posts with woodchip bedding. On average- Internal diameter: 300 - 400 mm; 1 m tall with wall thickness: 20 - 25 mm

Varying sites (highly degraded, post-burn and remnant vegetation) in private landholds Also remnant vegetation on NGO property. (e.g. Bush Heritage)

Forks or trunks of living wandoo and marri 2 - 9 m

Arranged in clusters across the site 3 - 200 m

DEC

(1) Cockatube (2) Hollowed-out log sections erected on metal poles

(1) Industrial PVC piping (as above) (2) Hollowed-out (Wandoo) tree sections with metal lids and bases attached to metal pole on a concrete slab. On average- Internal width: ~300 mm and <1 m tall. Pole: 5-6 m and concrete slab: 1 m x 1m

Native open Salmon Gum woodland remnants and in private landholds with a mix of Red Morel, Salmon and York Gum.

Forks or trunks of living salmon gum or wandoo. 5 - 6 m

Arranged in clusters across the site ≤50 m

DEC

(1) Cockatube (2) Hollowed-out log sections erected on metal poles (3) Wooden box-type ANH

(1) Industrial PVC piping (as above) (2) Lidless hollowed-out tree sections with metal bases attached, erected on metal poles. On average- Internal width: 300 mm and 1 m tall. Pole: 100 - 150 mm diameter and <10 mm thick and 5 - 6 m long (3) box made from wood sections with lid and tin base *woodchips bedding for all ANHs

Private property in the Wheatbelt region

Facing northeast on living Salmon Gum and Wandoo 5 - 7 m

Not noted.

160

Worker ANH Design Materials and Dimensions Location Placement and Height

Arrangement and distance from other hollow (natural or artefact)

WAM

(1) Cockatube (2) Wooden box-type ANH with natural hollow entrance

(1) Slanting open-top black PVC pipe with metal or plastic base with ‘weldmesh’ internal ladder and fixed sacrificial chewing posts. Charcoal and woodchip bedding. On average- Internal and external diameters: 250 - 300 mm and 300 - 350 mm, respectively. 0.7 - 1.2m tall with internal ladder of 100 mm wide and hardwood sacrificial posts of 70 - 50 mm (2) Box made with wood sections attached to natural hollow top entrances. On average- Natural entrance diameter: 200 - 300 mm (some boxes treated with insecticide)

Remnant woodland, roadside verges and edges of paddocks

Stumps, forks or trunks of living and dead Marri and Wandoo Natural nest height

Not noted

WAM and WC

(1) Cockatube (2) Wooden box-type ANH

(1) Side-entry plastic tubing with steel lid (2) Side-entry box made from plywood sections. Ranging from small, medium to large in size

In forested areas near man-made reservoirs on WC property

Forks of living Jarrah or Marri Natural nest height

Not noted

BCPS Cockatube

White/black PVC or polyethylene piping sealed with plastic or metal at the top and base. Base had holes drilled for drainage with woodchip bedding and fixed jarrah sacrificial post of ~1.4 m. Top, side entry with internal ladder that came down to 75 - 100 mm from the floor. On average- Internal diameters: 280 - 360 mm and 1 m tall. Entrance holes: 150 mm x 190 mm

ANHs placed in aviary of rehabilitation centre

N/A 1 - 1.5m ~1.5 m

LCJ-S* Cockatube

Durable black high-density polyethylene or PVC with galvanised fittings for chaining to trees and sacrificial posts made from Marri, Jarrah or Tuart. On average- Internal diameter: 300 - 400 mm; wall thickness: 20 - 25 mm and 1 m tall

Native vegetation of Shire/public reserves and State Government land, as well as in private landhold- paddocks

Facing northeast on forks or trunks of living and dead Jarrah, Marri, Tuart, or Wandoo 8 - 17 m

Mostly scattered across the site, but clustered in certain areas due to space restrictions (i.e. private landholds) ~20 m

161

Worker ANH Design Materials and Dimensions Location Placement and Height

Arrangement and distance from other hollow (natural or artefact)

MU* Cockatube As above Remnant vegetation reserve of a public university

Trunks of living Pine and White Gum >5 m

≤50 m

CU Hollowed-out log sections erected on metal poles

Pieces of cut sections of hollowed-out Wandoo trees attached to metal bases and erected on metal poles that are hinged at the bottom

Cleared land on private landholds

Adjacent to mostly dead Wandoo Height unsure

Not noted

NBG (1) Cockatube (2) Wooden box-type ANH

(1) As above. On average- Internal diameter: 280 - 300 mm and 1 m tall (2) Box made with fresh new Jarrah wood sections with side-entry and partial roofing. Small wood blocks were attached to the inner wall to serve as rungs for climbing. On average- Wooden boxes were 1 m tall with a depth from side opening of 600 mm.

In remnant Jarrah-Marri forest near a Pine plantation and mining tenement

Forks or trunks of living and dead Jarrah, Marri and Wandoo trees 6 - 10 m

Arranged in clusters across the site ≥50 m

162

Table 8.3 (continued).

Worker Black cockatoo species Evidence of Use by Black Cockatoos Inspected or Used Subsequent Use

BAWA Carnaby’s Cockatoo FRTBC Yes; evidence of eggs and chicks Use– the first breeding season post-establishment Unconfirmed, but likely.

BAWA* Carnaby’s Cockatoo Yes; evidence of eggs and chicks Inspection- in a couple of weeks.

Use- in a month None

DEC Carnaby’s Cockatoo Yes; evidence of eggs and chicks Most were used and can be within days, months or

even years. Yes; for some ANHs

DEC Carnaby’s Cockatoo C. b. Samueli Yes; evidence of eggs and chicks Use- in that first year ~1 month before breeding

season Unconfirmed, but likely.

WAM Carnaby’s Cockatoo Yes; evidence of eggs and chicks Use- within a week Yes; for some ANHs

WAM and WC

Carnaby’s Cockatoo FRTBC

Yes; evidence of eggs and chicks (in later boxes without plywood sections) Unsure Unconfirmed

BCPS Carnaby’s Cockatoo FRTBC Baudin’s Cockatoo

Yes; evidence of eggs and chicks Inspection- almost immediately Use- from two weeks up to 12 months post-installation Unconfirmed, but likely.

LCJ-S* Carnaby’s Cockatoo FRTBC - Inspection- almost immediately

Use- unsure Unconfirmed, but likely.

MU* FRTBC Yes; evidence of egg and chick Inspection- in a few weeks to a few months Use- a year post-installation Unconfirmed

CU Carnaby’s Cockatoo Yes; evidence of eggs and chicks Inspection- in less than a week

Use- unsure Yes; as some birds were tagged

NBG None yet - - -

163

8.3.5 Characteristics of Successful Artificial Nest Hollows

Black cockatoos nested and reared young in all designs of ANHs. While wooden box-

type ANHs were generally successful, those made from plywood deteriorated so rapidly

that they had to be replaced before birds had an opportunity to use them. The only site

where ANHs were not used at all was at NBG (Table 8.3). Given that the cockatube and

wooden box-type ANH designs used at NBG were deployed successfully elsewhere,

factors other than design must account for their lack of use.

Most observations of black cockatoos using ANHs involved Carnaby’s Cockatoos, with

FRTBC accounting for a smaller number of records. One of the records of FRTBC

using an ANH represented the first time this subspecies has been documented using an

ANH in the metropolitan area (C Minton10, 2010, pers. comm.) To date, there are no

records of Baudin’s Cockatoos using ANHs in the wild. There was one record of an

Inland Red-tailed Black Cockatoo (C. banksii samueli) using an ANH. In an aviary

setting, Baudin’s Cockatoos, Carnaby’s Cockatoos, and FRTBC all used ANHs

(BCPS). This suggests that Baudin’s Cockatoos and FRTBC could use ANHs if the

were placed in suitable locations.

8.3.6 Installation, Maintenance, Monitoring and Costs Associated with Successful

Artificial Nest Hollows

BCPS were the only organisation to report on ANHs in a captive setting, with ANHs

fastened to the sides of the aviary cage mesh with metal chains. The aviary situation

facilitated monitoring, so ANHs were monitored at least monthly and usually weekly

when breeding was suspected. ANHs were removed only during checks, when two

needed maintenance before replacement. None of the 14 fell from the cage bars. The

construction and set-up costs of $400 - $500 for each ANH were similar to costs for

ANHs in natural settings (Table 8.4).

Installation methods for ANHs in natural settings included EWPs, rope-pulley systems,

and cherry-pickers. Installation involved mechanical assistance from vehicles (e.g. utes,

tractors, front-end loaders) or from personnel (harnessed with climbing ropes and

ascenders on extension or wire-cave ladder systems). ANHs were attached to a tree or a

metal pole cemented or bolted into the ground. The metal pole design may deter

climbing predators and cannot burn (R Dawson4, 2009, pers. comm.; P Mawson3, 2009,

pers. comm.) (Table 8.4).

164

ANHs in natural settings were monitored to determine usage by black cockatoos (or

other species) and requirements for maintenance, and to assess potential improvements

in design and installation. In most cases, monitoring was conducted opportunistically,

with the timing and frequency of inspections influenced by factors such as time,

travelling distance, cost, logistics and the expected time of breeding. Only two groups

checked their ANHs regularly (BCPS and NBG) and none checked more frequently

(Table 8.4).

Participants reported that it was difficult to determine whether hollows were used,

particularly where the placement of ANHs precluded easy observation into the hollow.

While the presence of black cockatoos and, in particular, a chick or nestling indicated

use for breeding, other evidence was circumstantial, such as chewed areas around the

hollow or sacrificial posts (Table 8.4).

The two most common inspection methods were: (1) climbing with a harness and ladder

to check inside of the ANHs and (2) looking for signs of use by checking for ‘chew-

marks’ on the exterior sacrificial posts in the ANHs or around the edges of the entry-

holes to ANHs. Another common technique to check use of ANHs was tapping on the

base of the tree to either flush the sitting parent or bring it to the ANH entrance.

Respondents also reported using black cockatoo courtship behaviour and

nestling/fledgling calls to infer nesting in the area around the ANH. Another, less

common, method was using telescopic camera poles or mirrors to check the interior

(Table 8.4).

Maintenance requirements involved replacing sacrificial posts and/or re-filling the

woodchip nesting material for the cockatubes. The intensity and frequency of the

chewing habit varied with individual birds. Not all sacrificial posts were replaced every

breeding season. In only one case were ANHs subject to monthly maintenance (BCPS).

Gaps in the wood of the walls or bases of wood-based ANHs and hollowed-log section

ANHs needed to be filled when they cracked from natural decomposition or were

chewed by the nesting birds.

The total cost of making, transporting and setting up single ANH did not vary much

between groups and design. Reported costs ranged from as low as $400 to $700 overall.

Costs do not include monitoring, as this is an ongoing activity.

165

Table 8.4 Installation, maintenance, monitoring and costs associated with ANH implementation.

Worker ANH Design Installation Monitoring Maintenance Total costs per ANH (construction and installation)

BAWA Hollowed-out log sections erected on metal poles

The ANH was screwed into a metal base attached to a pole. The pole is erected using a rope attached to a moving car

ANHs were monitored annually via climbing

Annual checks and repairs as necessary

$540

BAWA* Cockatube Personnel on ladder attached ANHs to trees with the aid of a front-loader or EWP

ANHs were monitored every nesting season at least once to a few times. Via tapping the base of ANH tree to flush bird or looking for sacrificial post use

Every nesting season. Sacrificial posts and woodchip bedding replace when necessary

$500 - 700

DEC

(1) Cockatube (2) Hollowed-out log sections erected on metal poles

Cockatubes were attached to trees using a rope-pulley system guided by a harnessed worker on a ladder. ANHs attached to poles were erected using a tractor or front-loader and cemented to the ground

Monitored using a ladder and harness system at least once annually and twice of breeding was observed over the nesting season to observe chicks

Sacrificial posts replaced every 3 - 5 years for cockatubes. Gaps in hollow log-section ANHs are filled when required

$400 - 500

DEC

(1) Cockatube (2) Hollowed-out log sections erected on metal poles (3) wooden box-type ANH

Cockatubes were attached to trees with assistance of a front-loader. While ANHs on poles were erected using a tractor and bolted to the ground

Monitored twice a breeding season using a ~6 m long telescopic camera or climbing with the use of a harness and ladder

Wooden box-type and hollow log-section ANHs are repaired with tin sheets when required. Usually every 4 - 5 years

$500 - 520

166

Worker ANH Design Installation Monitoring Maintenance Total costs per ANH (construction and installation)

WAM

(1) Cockatube (2) Wooden box-type ANH with natural hollow entrance

Cockatubes and wooden box-type ANHs were fastened to trees by workers on extension and wire-cave ladders and harnessed to climbing ropes with ascenders with the aid of elevated work platforms

Monitored regularly every breeding season. ANHs were monitored via climbing, raking the tree with a stick or inspecting the ground for fresh chewings

Replacement of sacrificial posts for cockatubes every 2 years

$500

WAM and WC

(1) Plastic tube (2) Wooden box-type ANH

ANHs were fastened to trees using a ladder and harness system assisted with an EWP

Monitored occasionally and at least once annually. Performed by tapping the tree base, checking from the ground, ladders or elevated work platforms

No particular maintenance requirements as not regularly maintained

Unsure

BCPS Cockatube Workers on the ground fastened ANHs to the cage mesh via metal chains

Monitored every week or month using mirrors to check the inside of the ANHs

Externally checked weekly. Internally checked every 3 - 12 months (i.e. replacement of sacrificial posts)

$400 - 550

LCJ-S* Cockatube ANHs lifted by a pulley system were screwed and fastened to trees by workers on a cherry-picker

ANHs in reserves were monitored annually, and opportunistically for those on private landholds Base of trees were knocked and sacrificial posts checked for chewings

Maintenance of sacrificial posts Every two breeding seasons or 4 - 5 years

$500 - 700

167

Worker ANH Design Installation Monitoring Maintenance Total costs per ANH (construction and installation)

MU* Cockatube ANHs on a rope were winched up using a car and chained and bolted to trees by workers on a cherry-picker.

ANHs were monitored on a random and ad hoc basis but usually every couple of months via knocking on trees or visual observations of ANHs

Inspected every few weeks or opportunistically. Will replace sacrificial posts when required

~$600

CU Hollowed-out log sections erected on metal poles

ANHs on poles were erected using a front-loader and the poles were bolted to the ground

Monitored once monthly by checking sacrificial posts

Sealing of holes in wood when necessary

~$500

NBG (1) Cockatube (2) Wooden box-type ANH

EWP were used to assist workers. Cockatubes were chained to trees, while the wooden box-type ANH here attached using wires

Monitored monthly via knocking the base of trees or checking for chewings on sacrificial posts

Externally checked monthly. Including checking of sacrificial posts and ANHs for use and tapping of base of tree

~$500

168

8.3.7 Hazards to Black Cockatoos

In the artificial cage setting one Carnaby’s Cockatoo chick left the nest approximately

three weeks earlier than it would in the wild. The chick was fed by its parents on the

ground and survived.

Some ANHs were occupied by non-target species including feral bees (Apis mellifera),

Galahs (Cacatua roseicapilla), Western Corellas (C. pastinator), Ringneck (Twenty-

Eight) Parrots (Barnardius zonarius), mountain/wood ducks (Chenonetta jubata and

Tadorna tadornoides) and even Boobook Owls (Ninox novaeseelandiae) (Table 8.5).

The small number of injuries caused by the ANHs were all associated with the ladder

and occurred in the captive setting. One chick caught its leg in the ladder, one female’s

wing caught on the ladder and bled, and one bird was snagged on the outside ANH wall,

metal chain, and mesh (Table 8.5).

Predation in or near ANH was not commonly observed. One chick was suspected to

have been taken by a feral cat and two were taken by monitor lizards on separate

occasions (R Dawson4, 2009, pers. comm.; R Johnstone5, 2009, pers. comm.; P

Mawson3, 2009, pers. comm.; R Scott2, 2009, pers. comm.) It is unknown as to whether

the predation was ‘enhanced’ as a result of ANH use or was simply a natural incident.

169

Table 8.5 Hazards to black cockatoos and hazards to longevity of ANHs.

Worker ANH-related Injury or Fatality

Damage/Manipulation to ANH ANH-induced Behaviour Competition Predation ANH taken down/fallen

BAWA None Fairly benign chewing to inner ANH wall None By Galahs None None

BAWA* None Some chewing to inner wall None None

Unlikely. Unsure- suspected monitor lizard

None

DEC

One occasion a chick’s leg was caught in the ladder. Mesh, but was freed

Chewing to inner wall and ANH entrances None

By Galahs, Western Long-billed Corellas and mountain ducks

Once; goanna

None. One ANH on a pole that was bent as a result of human activity but was straightened and left in place

DEC One death; unsure if ANH-related. Chewing to ANH

Birds seemed quicker and more physically able to climb out of ANHs than natural hollows

By Galahs Low levels; by feral cats

3 - 6 taken done due to lack of breeding activity

WAM None

Extensive chewing to the walls and damages to floors of wooden box-type ANHs that needed repairs. Extensive chewing of the sacrificial posts, but none of PVC tube

None

Wooden box-type ANHs used by Australian Shelduck, Boobook Owl and Galahs and invaded by feral bees

Once; by feral cats

One taken down as a result of use by non-target species modified and replace at another site. Some decayed, but did not fall

170

Worker ANH-related Injury or Fatality

Damage/Manipulation to ANH ANH-induced Behaviour Competition Predation ANH taken down/fallen

WAM and WC Unsure Unsure Unsure

Galahs, feral bees, wood ducks and possums

Unsure One removed due to invasion by feral bees

BCPS

One injury; mother bird caught her wing on the ladder mesh and bled. On two other occasions, parent birds were stuck between the ladder mesh and ANH wall

Chewing to inner edges of PVC ANH entrances and ladder mesh Birds observed undoing bolts and nuts, dislodging an ANH from the cage mesh

One chick left ANH three weeks earlier and continued to be fed from the ground

Attempts by Galahs and feral bees None One nearly fell due to

manipulation by bird

LCJ-S* None Chewing on sacrificial posts and section of tree trunk near ANH entrance

Only females investigated ANHs, and appear to play with ladder mesh

By Galahs, wood ducks and an unknown parrot species

None

Two ANHs fell when trees fell. One was removed due to tree felling and returned, and one was exposed to but unaffected by a fuel reduction burn

MU* None Unknown

Birds seemed to hang around ANHs all day for many days and rarely left the tree

By Galahs and invasion by feral bees

None. Birds were observed scared away by ravens in the vicinity

None

CU None Chewing to the wood of natural hollowed-out log ANHs

None Galahs and Ringneck 28 Parrots Unsure; possible Unknown

171

Worker ANH-related Injury or Fatality

Damage/Manipulation to ANH ANH-induced Behaviour Competition Predation ANH taken down/fallen

NBG* N/A N/A N/A None None

One fell when an old dead tree fell and was replaced. One was exposed to a fuel reduction burn but was unaffected

172

8.3.8 Hazards to Artificial Nest Hollows

Damages to ANHs by the birds were fairly benign. Birds chewed the inside walls, top

edges/entrances or the section of tree close to the ANH opening, while the harder metal

surfaces were rarely damaged. The sacrificial wooden posts placed in the cockatubes

were the most extensively chewed. There was only one situation – in the artificial

setting - where the birds undid the nuts/bolts that held the ANH to the cage mesh. This

resulted in the ANH dangling from the cage mesh (Table 8.5).

In natural settings, the fate of only 61% (n = 96 of 157) ANHs was known. Few ANHs

(9%) were removed permanently. One respondent removed six ANHs because there

was no sign of breeding within them for at least three years post-establishment, two

were destroyed when their trees fell and one removed and not replaced following

colonisation by non-target species. A further two ANHs were replaced after the tree fell

and one (after modification) following colonisation by feral bees. Two cockatubes

survived fuel reduction burns (LCJ-S and NBG) (Table 8.5).

8.3.9 Respondents’ Priorities for Future Research on Artificial Nest Hollows

Overall, the major concerns or issues ANH workers were with costs of construction,

maintenance and monitoring, and safety issues to workers and the general public.

8.4 Discussion

8.4.1 What Factors Contribute to a Successful Artificial Nest Hollow Placement?

All the ANH designs examined here were used for breeding in one or more cases. The

limited sample sizes make it difficult to conclude whether some designs are more suited

to some species than others, or whether particular designs might be preferred if multiple

choices were available in the same locale. Nevertheless, a commonly reported belief

was that positioning ANHs in known breeding areas where natural hollow abundance

has been reduced was most important in determining success. The logic behind this

observation is that black cockatoos show fidelity to breeding sites and hollows

(Saunders 1982) and that breeding hollows must be in proximity to suitable food and

water. Groom (2010) also noted a correlation between the utilisation of ANHs by

Carnaby’s Cockatoos and the close proximity of natural hollows in both the southern

forests (Jarrah-Marri forest) and the northern Wheatbelt.

173

In considering characteristics of successful placements, it is instructive to examine

situations where ANHs were not used for breeding and monitoring was adequate to

detect breeding if it had occurred. The trials at NBG (reported in more detail in Chapter

7) are a case in point. At NBG, breeding was not detected in either cockatubes or

wooden box-type ANHs despite monthly monitoring over 20 months from 2008 – 2010.

Breeding was noted in 11 natural hollows in the study area. The problem may have been

that the main canopy height in the area was 18 – 25 m, whereas the ANHs were

positioned at 6 – 10 m heights because of EWP constraints restricting placement height.

Overall, it is likely that successful ANHs must be positioned near known breeding

grounds with access to food and water and at a similar height to natural hollows in the

area.

Most of the successful ANH deployments occurred in areas where Saunders (1979) had

reported that land clearing had caused substantial reductions in the numbers of natural

hollows present. In contrast, the abundance of natural hollows may not be limiting (or

as limiting) in forested regions. Despite logging and other land uses that remove the

larger and older (>150 years) hollow-bearing trees (Whitford and Williams 2002), some

habitat trees are retained and natural recruitment may occur. Nevertheless, there is a risk

that retained trees will be lost to logging or natural tree-fall resulting in a lack of the

larger and older hollow-bearing trees that these black cockatoos need to breed in

(Whitford and Williams 2001; Johnstone et al. 2005).

8.4.2 Risks to Black Cockatoos or Workers in Establishing and Maintaining Artificial

Nest Hollows

Risks to black cockatoos appear small based on the experience of participants in this

study. Nevertheless, the materials used, particularly synthetic ones, should be tested

chemically to ensure they are non-toxic, especially given the chewing habits of black

cockatoos (Conners and Conners 2005). Even the wood-based materials may have

residual chemicals on the surface from past use (such as to preserve the wood). Other

factors that may affect use and nesting success include the presence of competitor

species such as feral bees, Galahs and corellas (Saunders 1979; Garnett et al. 1999;

Barrett et al. 2003; Johnstone and Kirkby 2007). While predation was not commonly

observed by respondents, it should be monitored.

174

Respondents were concerned about risks to workers. ANHs are heavy and bulky; each

weighing anywhere from 20 - 30 kg (A Elliot7, 2009, pers. comm.). They are installed at

heights of up to 17 m, so there is a risk of serious falls that may be aggravated during

bad weather. Furthermore, if the host tree selected is a dead tree, there is an increased

risk that limbs may fall off during instalment.

At NBG, specific safety procedures were required to install ANHs. Wearing gloves

during checks (or installation) may also protect workers, as well as avoid getting human

scent on the hollows. Finally, signs may be needed to deter people from approaching

beneath ANHs in case they fall. These signs should caution against tree fall to prevent

giving clues of ANH locations to potential poachers.

8.4.3 The Place of Artificial Nest Hollows in Black Cockatoo Conservation and

Research Priorities for Artificial Nest Hollows

Western Australian black cockatoos require the largest hollows of all arboreal fauna in

the state (Abbott and Whitford 2002). The lack of natural hollows, as a result of land

clearing in the Wheatbelt and other areas of the southwest, is now a major factor

limiting their reproductive success (Cale 2003; Chapman 2008; Groom 2010; Garnett et

al. 2011). It is not clear whether the current and future needs of these hollow-using

cockatoos will be satisfied. Some authors believe that there is a deficit of suitable

hollows (e.g. Johnstone et al. 2013a), which is exacerbated by the time lag of many

decades before regenerating trees (e.g. planted, regenerating in harvested coupes) can

produce suitable size hollows (Vesk et al. 2008). It is also important to note that ANHs

are suitable in some contexts. For example, in the tall Karri forests of the far southwest

it is impractical to position ANHs at the heights where natural hollows would occur.

This study supports several recommendations proposed by Groom (2010) regarding the

use of ANHs in southwestern Australia. I have grouped them into three broad topics:

Improved monitoring of ANHs, including the establishment of an ANH registry

Both Groom (2010) and the respondents in my study indicated that monitoring was

often ad hoc or absent. Therefore maintenance could not be made systematically.

Groom (2010) recommended requiring groups or individuals seeking funding from both

State and Commonwealth bodies to establish ANH to register their activities to facilitate

monitoring. I concur, as this will address the significant issue of inadequate monitoring.

175

Frequent and long-term monitoring of ANHs increases certainty in assessing ANH use

(Saunders 1986) and allows rapid and effective response to colonisation by non-target

species, although with the risk that too frequent monitoring may repel birds. It may also

suggest improvements to existing ANH designs. Observations could also be made of

insect activity around the ANHs (e.g. flies or ants attracted to droppings) or presence of

cobwebs (ANH unused) or chewings around the ANHs (ANH used).

Practical and financial constraints often constrained monitoring to annual or ad hoc

inspections. At the least, monitoring prior to the breeding season and monthly

throughout will give a good indication of use. Infrequent or inadequate monitoring is a

widespread problem in many community-based conservation or restoration endeavours

and not unique to use of ANHs (e.g. Brooks and Lake 2007).

Design principles for successful ANHs

Existing designs all consider the important principles of thermal insulation, durability,

adequate internal dimensions, drainage, accessibility and provision of sacrificial

chewing material. General principles include: positioning ANHs on trunks or vertical

poles of long-lasting materials, ensuring dimensions allow a nesting bird and chicks to

nest comfortably without damaging tail feathers, using durable materials that simulate

the microclimate (e.g. thermal properties) of natural hollows, providing nesting

materials, and choosing a base that allows drainage while supporting the bird and chick.

Entrance holes deserve special attention. They should be just large enough for a black

cockatoo to access without incurring injury while deterring nest competitors and

predators. They must also allow viewing for monitoring and access for maintenance.

Some respondents believed that entrances should be placed vertically, as opposed to

horizontally, to deter bees. Some respondents believed that a northeasterly aspect was

important, but this requires experimental verification.

Sacrificial posts should be installed, especially inside the synthetic ANHs, for natural

chewing activities and for the birds to prepare a dry base for nesting eggs and chicks.

Ladders or steps in the wood are needed for the birds’ easy and safe access.

176

Research into use and maintenance of ANHs

Although inadequate monitoring complicates assessments of ANH use, it is clear that

some ANHs are unused. This may be because of height, placement in relation to food

sources, water or existing breeding sites, differences in preferences across species, or

conditions of the ANH itself (e.g. chemical residues, inappropriate microclimate).

Maintenance schedules, including issues such as replacement of sacrificial chewing

posts, should also be considered. Longevity is a major consideration for ANHs; both for

the desirability of installing a long-lasting structure and for the costs associated with

maintenance, particularly with wood-based ANHs that degrade naturally. Synthetic

ANHs are more durable and cost-effective over the long-term, but still need monitoring

and maintenance. Lastly, installation of ANHs on living host trees minimises the loss of

the ANH when a dead tree falls.

In addition to the recommendations in Groom (2010), I also suggest that research could

consider methods of encouraging hollow development in standing trees. These include

developing natural hollows more rapidly through, for example, inoculation with fungi

or pheromones of wood-decaying beetles, physically making holes in trees and trunks,

and burning (Brennan et al. 2005). It may also be possible to offer multiple designs of

ANHs in the same site, thereby testing whether there is a preference for one design over

another. The hypothesis that breeding could be increased at established breeding sites if

ANHs were provided should also be tested. Future work could also consider feeding

supplementation to encourage use of ANHs given that food availability influences

breeding (Saunders 1977, 1986) and applications of ANHs in forested areas with high

canopies.

8.4.4 Strengths and Limitations of the Study

Prior to 2010, most of the information available regarding ANH use for black cockatoos

in southwestern Australia is unpublished, so the major value of Groom (2010) and this

study is in collecting anecdotal records for the use of researchers, bird conservationists,

and other interested parties. By adopting a planned survey approach, I ensured that

similar questions were asked of all participants in the hope of gathering the broadest

range of information suitable for comparisons of outcomes between different designs of

ANHs.

177

However, being questionnaire-based, most of the data provided depended on the ability

of the interviewees to answer the questions and the ease at which they were able to

access their own data. I also depended on the willingness and availability of the

interviewees to provide the time and information required to complete the

questionnaire. In some cases, responses to questions were not optimal, e.g. some

questions were not answered at all or were incomplete or were ambiguous (with no

further details were provided). As a result, the overall set of survey responses is patchy

with information not available for all the variables that I would have liked to consider.

An additional problem related to the lack of proper and/or consistent monitoring and

uncertainty in determining use by black cockatoos. These issues, along with the small

sample size of participants, precluded quantitative analyses.

Nevertheless, I believe that my main recommendations improved monitoring and the

need for research into ANH use and maintenance are robust because:

Successful ANHs were reported by multiple, independent groups. Therefore

success was not attributable to unique features of one group at one site.

Participants appeared frank about failures and also about the extent of

monitoring, which is important in deciding whether a failure is real or simply

an artefact of inadequate monitoring.

ANH designs that were successful at some sites were unsuccessful at others,

suggesting features of location and establishment that should be investigated

in controlled experimental studies.

8.4.5 Concluding remarks

It is likely that ANHs will play a more important role in the near future as an ‘interim’

mitigation measure to the problem of hollow deficiency in many landscapes. The

general ‘gist’ drawn from the opinions of interviewees from this survey (that is

Question 26) is that ANHs are a good short-term strategy, especially in areas deficient

in natural hollows. However, to sustain current populations of black cockatoos, they

cannot substitute for natural hollows, so the major focus of conservation should be the

retention of breeding and feeding habitats.

Information exchange among different researchers from various organisations or public

interest groups is also critical. Although it was not a specific question in my study,

respondents often asked about any other work undertaken and were keen for

178

information. Not only does this allow groups to follow best practice, but also it prevents

wasting funds and other resources in repeating experiments.

179

Chapter 9 Water Use by Black Cockatoos at the Newmont

Boddington Gold Mine: Natural and Artificial Water Sources and

Risk of Interaction with Residue Disposal Areas

9.1 Introduction

9.1.1 Problem Formulation

Black cockatoos drink from natural and man-made water sources within the Newmont

Boddington Gold (NBG) mine-site. Standing water within the Residue Disposal Areas

(RDAs) presents the risk that black cockatoos could drink water containing harmful

cyanide concentrations (Smith et al. 2010). This chapter examines how black cockatoos

use the water sources at NBG and characterises, in a general way, the risk of

interactions between black cockatoos and RDAs. It does not represent a formal risk

assessment. The chapter also addresses the value of potential mitigation measures,

particularly providing alternative water sources near the RDAs.

9.1.2 Black Cockatoos and Water

Water is scarce within the Jarrah-Marri forest. In general, surface water is limited,

stream flow is seasonal and irregular with extended dry periods, and runoff is minimal

(Gentilli 1989; Schofield et al. 1989). Rainfall for the region varies annually, across

seasons, and across east-west and north-south gradients and averages 600 – 1000 mm

per annum (Figure 9.1; Silberstein et al. 2010). Most rain falls between April and

September and the southern and western regions receive more rain than the northern and

eastern ones. In summer, high temperatures lead to high evaporation rates and drought

is common (Gentilli 1989; Schofield et al. 1989). Since the 1970s there has been a

marked trend to reduced rainfall across the Jarrah-Marri forest and more widely in the

southwest (Figure 9.2; Silberstein et al. 2010). Thus, similar to other Mediterranean-

type climates, the Jarrah-Marri forest is a seasonally stressful environment with limited

water availability, especially during summer (Croton and Reed 2007).

180

Figure 9.1 Spatial distribution of mean annual and seasonal historical rainfall (1975 – 2007) in the southwest of Western Australia, including the Jarrah Forest Bioregion (taken from Charles et al. 2010, p. 83).

181

Figure 9.2 Spatial distribution of trends in mean annual and seasonal historical rainfall (1975 – 2007) in the southwest of Western Australia, including the Jarrah Forest Bioregion. The scale shows mean rate of change in mm/y/y (taken from Charles et al. 2010, p. 83).

The availability and location of water sources influences the movement and activity

patterns of many parrots, particularly those inhabiting desert or seasonally arid

environments (Fisher et al. 1972; Davies 1982; Warburton and Perrin 2005), including

black cockatoos (Saunders 1977, 1980; Davies 1982; Finn et al. 2009). Black cockatoos

are typically – though not exclusively – granivores and thus do not usually obtain

adequate hydration from water present in food or obtained through metabolic processes

(Williams et al. 1991; MacMillen and Baudinette 1993; Bentley 2002; McNab 2002).

As such, black cockatoos generally must drink standing water sources at least once

182

daily. High ambient temperatures increase moisture loss through evaporative cooling, so

birds drink more frequently (Fisher et al. 1972; Davies 1982; Cameron 2007). Other

factors, such as exertion, can also increase water requirements. The consequence of

water stress was demonstrated recently in the Esperance-Hopetoun region southeast of

the forest block, where about 150 Carnaby’s Cockatoos died when temperatures rose

rapidly above 47 ºC and birds could not locate adequate water (Saunders et al. 2011).

Such events may become common following climate change (McKechnie and Wolf

2010).

The need for daily water intake means that black cockatoos are rarely found more than

10 km from a water source and their distribution patterns, both locally and regionally,

may reflect drainage patterns associated with rivers, streams, and other water bodies

(Johnstone and Kirkby 1999; Cameron 2007). Black cockatoos often drink in the

evening, but may also drink in the morning (Cale 2003; Chapman 2008; Cameron 2007;

Finn et al. 2009).

Sources of drinking water for black cockatoos in the wild include freshwater lakes,

pools, rivers, streams, and swamps, as well as other landscape features that concentrate

runoff or store water such as rocky outcrops, tree hollows, and puddles on hard ground

or beside paths and roads (Abbott 2001; Cameron 2007; T Kirkby, Western Australia

Museum, 2009, pers. comm.). Artificial water sources such as cattle troughs and farm

ponds can be vital habitat features within landscapes lacking natural water sources and

may be used daily, particularly during summer. With forest landscapes, many water

sources are small, ephemeral, and obscured by vegetation. In urban areas, black

cockatoos are known to drink from vases and urns in cemeteries (Berry 2008).

Various components of the mine infrastructure at NBG provide water sources for black

cockatoos year-round. These artificial water sources include drainage sumps for roads,

seepage sumps for water bodies, Water Supply Reservoirs (WSRs), and ponds

associated with wetland areas. The adjacent agricultural and plantation areas also

contain numerous dams within paddocks or near timber stands. Natural water sources

include brooks, streams, swamps, and the Hotham River. In winter, water is more

widely available through ephemeral pools and puddles along roads and cleared areas,

and on top of hard-capped soils within areas of remnant forest.

183

Black cockatoos were observed drinking from many of the artificial water sources at

NBG and the availability of permanent water sources might influence their daily

movement patterns and distribution. It is also possible that the dependability and

relative abundance of water sources enhances the suitability of the NBG landscape for

feeding and breeding. These factors and the natural mobility of black cockatoos suggest

that they may attempt to drink from the RDAs in the northern half of the site.

9.1.3 Fauna-Residue Disposal Area Interaction at Newmont Boddington Gold

At NBG, cyanide is used to extract gold from ore, producing leachates or cyanide-

bearing tailings. These can be toxic to wildlife if consumed (Eisler and Wiemeyer 2004;

Brasel et al. 2006; Griffiths et al. 2009). Cyanide-related hazards are more likely to

occur at mines processing copper-gold ore, which generates copper-cyanide complexes

as well as gold-cyanide complexes, both of which may be toxic to birds (Eisler and

Wiemeyer 2004; Donato et al. 2007).

Cyanide degrades rapidly when exposed to air and sunlight and does not biomagnify or

cycle in natural food webs or ecosystems (Eisler and Wiemeyer 2004). Thus, the main

risk of cyanide toxicity occurs near waste discharge locations containing high residual

concentrations of cyanide. These waste solutions are usually stored in large open water

storage facilities or impoundments (i.e. RDAs) that may cover several hundred ha

(Donato et al. 2007).

Free-ranging fauna may be attracted to these areas to drink or, for waterbirds, because

they contain potential aquatic habitat (Clark 1993; Eisler and Wiemeyer 2004). Birds

are especially vulnerable because they fly long distances to drink at these contaminated

ponds, and they constitute most of the wildlife mortalities from cyanide toxicity (Eisler

and Wiemeyer 2004; Donato et al. 2007). These mortalities may be under-reported if

monitoring is inadequate, cursory, or based on voluntary reporting (Brasel et al. 2006;

Donato et al. 2007). Estimates of cyanide-related mortality are further complicated

because death may occur away from the tailings area and some time after drinking.

Even prompt post-mortem examination may fail to detect the clinical signs of cyanide-

related mortality (Eisler and Wiemeyer 2004).

There is little detailed peer-reviewed information on the risks of cyanide-associated

mine waste for wildlife, and few guidelines for managing the risks to wildlife or for

184

monitoring the use of tailings facilities by wildlife (Smith et al. 2007; Donato et al.

2007, 2008; Griffiths et al. 2009). Mitigation methods include: managing waste

solutions and discharge procedures so that cyanide-containing solutions are kept at

concentrations below potential thresholds of toxicity; establishing decoy water bodies;

positioning flagging devices or floating polyethylene balls in ponds to dissuade landing

birds; or setting up physical barriers such as drift fences or screens (Eisler and

Wiemeyer 2004; Johnson and Donato 2005; Smith et al. 2010). Chemical repellents and

hazing techniques (e.g. acoustic, visual, or physical deterrents such recordings of raptor

calls, watercraft or personnel to deter birds) are costly, logistically intensive and often

ineffective (Clark 1993; Eisler and Wiemeyer 2004; Donato et al. 2007).

As part of an overall strategy to mitigate interactions between wildlife and the RDAs,

13 artificial fauna drinking points (FDPs) were established around the northern and

western perimeters of the F1 RDA in late 2009 and early 2010 (Figures 9.3 and 9.4;

Table 9.1). The FDPs provide black cockatoos and other fauna with alternative water

sources (i.e. other than the RDAs themselves). The design specifications for the FDPs

were developed based on the characteristics identified for artificial water sources used

by fauna at NBG and at other locations (D Donato, Donato Environmental Services,

2011, pers. comm.; H Finn, Murdoch University, 2009, pers. comm.). These

characteristics include features allowing black cockatoos to drink with ease, e.g. firm

(i.e. non-muddy) substrates, gently sloping banks, and the presence of vegetation nearby

to provide cover or protection from predators.

185

Figure 9.3 Location of FDPs (CS01 - CS13) installed around the northern and western perimeters of the RDAs at NBG in 2009 and 2010. Camera trapping occurred at CS03 and CS07 - CS11.

186

Figure 9.4 Examples of FDPs located around the perimeter of the F1 RDA.

(a)

(b)

187

Table 9.1 Protocol for artificial water points (FDPs) at NBG.

Appendix III NBG Protocol for Artificial Water Points

To Murray Allan From Ross Polis, Environmental Officer Department Construction Department Environmental

Copies To Steve Higgins, Hugh Finn Date 25 February 2009

1. Requirement

Artificial drinking points for wildlife around the perimeter of the RDAs to deter wildlife interaction with the F1/3 RDA discharge and supernatant is a requirement of:

Environmental Protection and Biodiversity Conservation (EPBC) Act (Commonwealth) approval for Black Cockatoos.

International Cyanide Management Code. These points require installation prior to the first summer following start-up.

2. Design & Examples

Size: No specific size requirement; however smaller would require less water and maximise surface area. Suggest width of a dozer blade and few meters long would be acceptable.

Lined with clay so that they can retain water. At least one side must be angled and compacted (or lined with gravel) so that birds and

terrestrial animals can walk down to the edge to drink. Examples are below: the first is a seepage sump to the SE of R4, the others are a

drainage sump off the L & M pit haul road (note: do not need to be this large, they can be much smaller as long as they can hold water).

3. Suggested Locations

At ~1km intervals around the F1/3 RDA between the stormwater sump to SE of Booster Station through to Gate 12 (i.e. stockpile to north of F3). The specific locations can be marked on map and field fitted to align with contour.

Require vegetation around the drinking point, so positioning near forest edge is preferable with trees for Black Cockatoo roosting.

Positioned so that they gather water through natural landscape drainage, positioning along culverts or using road sumps is acceptable.

Positioned so they can be filled by a water truck or reticulation.

The FDPs therefore provide more attractive watering sites than the RDAs which are

large, muddy, and open areas containing turbid and saline water (though still palatable,

being <10 000 mg/L TDS (Total Dissolved Solids) (Smith et al. 2010). As additional

control measures, NBG also maintains: (1) a ‘decoy’ wetland and WSR (the R4 RDA)

as an alternative water body to the F1 RDA and (2) adequate (i.e. for wildlife to drink)

water quality within the other water bodies that occur on-site (e.g. D1 and D4 WSRs).

There are two recorded instances of black cockatoo mortalities associated with the

RDAs at NBG (Table 9.2). In 2000, during the previous operation of the mine, two

black cockatoo carcasses were observed at the F3 RDA and at the R4 RDA (which was

used as an active RDA at that time), one in January 2000 and the other in February

2000, respectively. On both occasions the mortality was associated with birds using

188

recently deposited residue that was still unconsolidated (i.e. was still ‘mud-like’),

indicating that mortality could have related to ingestion of cyanide-containing water or

to entrapment within the material. It is difficult to infer exposure based on the WAD CN

(Weak Acid Dissociable-Cyanide) concentrations reported, which has been interpreted

to mean decant concentration.

189

Table 9.2 Records of black cockatoo mortalities associated with the RDAs at NBG. Incident 1 (recorded as Incident 5690)

Date: 10 January 2000 Time: 1400 Fauna Type: white-tailed black cockatoo Description: A white-tailed black cockatoo was located at the edge of a small pond created by residue deposition on the periphery of the F3 RDA (near dropper #9A). The cockatoo had left imprints in the tailings indicating it had become stuck, whilst accessing the pond to drink. This individual was recovered for post mortem analysis. The results of the post-mortem found cyanide was beneath detection limits. Decant WAD CN Concentration: 20 mg/L* on 29 December 1999; 25 mg/L* on 12 January 2000 * tailings concentration should be interpreted as decant concentration Suggested Cause of Demise: Landing on unconsolidated residue.

Incident 2 (recorded as Incident 5718)

Date: 15 February 2000 Time: 1030 Fauna Type: black cockatoo Description: A deceased black cockatoo was observed on recently deposited residue in the R4 RDA on the northwest embankment. The cockatoo had left significant imprints in the residue indicating it had landed on unconsolidated residue. The body could not be recovered; hence species identification was not ascertained. Decant WAD CN Concentration: 1 mg/L* on 9 February 2000 * tailings concentration should be interpreted as decant concentration Suggested Cause of Demise: Landing on unconsolidated residue.

190

9.1.4 Aims

The aim of the chapter is to examine how black cockatoos use the range of water

sources of available to them at NBG and to characterise the risk that black cockatoos

may use the RDAs as a water source. I use data from three sources: (a) behavioural

observations of black cockatoos drinking at water sources; (b) camera trap detections of

black cockatoos at water sources; and (c) monitoring data collected for the RDAs by

Donato Environmental Services. I use the information from these data sources to

address four issues necessary to adequately characterise the risk of harmful interactions

between black cockatoos and the RDAs:

Describe behavioural ecology of water use by black cockatoos at NBG,

including behaviours used, flock size during drinking bouts, and associations

with time of day and season;

Identify the range of water sources used by black cockatoos at NBG and

describe their salient features;

Examine evidence of black cockatoo interaction with, or proximity to, the

RDAs at NBG; and

Determine whether black cockatoos use the alternative water sources

established around the perimeter of the RDAs.

9.2 Methodology

9.2.1 Water Sources: Types, Timing and Nature of Use

Black cockatoos use several water sources at NBG, including: (1) man-made features

such as drainage or seepage sumps, WSRs, dams, and paddock ponds; (2) natural

drainage features such as streams and swamps (Figures 9.5 and 9.6); and (3) more

ephemeral water sources, such as puddles, which sometimes occurred in the landscape

following rainfall events. I refer to (1) and (2) as ‘permanent water sources’ (PWSs) to

emphasise their relationship with persistent man-made or natural features and their

tendency to retain water for long periods (sometimes even year-round). The FDPs

represent a form of PWS, as do the D1, D4, and R4 water supply reservoirs and the F1

RDA.

191

Figure 9.5 Location of PWSs where drinking by black cockatoos was observed or detected between December 2007 and May 2011.

192

Figure 9.6 Examples of PWSs at the NBG site: (a) a natural swamp, and (b) - (c) man-made sumps.

(a)

(b)

(c)

193

Observational study of black cockatoo drinking behaviour at NBG began in December

2007, with more targeted study of drinking sites beginning in October 2008 and

observations continuing in some form until autumn 2011. The methods used and the

timeframe for their use is discussed further below.

Discharge of tailings into the F1 RDA began on 18 September 2009. The F1 RDA was

the only RDA used during the study. Tailings residues were discharged into a confined

area of F1 known as the tailings discharge zone (TDZ). The FDPs were established

between late 2009 and mid-2010. The R4, D1, and D4 water bodies held water

throughout all or most of this study. The D1 and D4 water reservoirs have never been

used to contain tailing residues. R4 has not been used for contain tailing residues since

2001 (i.e. when the mine was in a care and maintenance phase).

9.2.2 Behavioural Observations

Behavioural observations were collected using opportunistic and systematic approaches.

Most observations of black cockatoos drinking occurred opportunistically, usually

during the course of behavioural surveys conducted for other research purposes. Other

opportunistic observations also occurred while travelling around the site and stopping to

observe birds sighted from the vehicle.

If birds were observed drinking I recorded the episode as ‘definite drinking’. I recorded

an episode as ‘probable drinking’ if I did not observe birds drinking (e.g. because

vegetation obscured the water body), but the circumstances (e.g. proximity to a water

source) and associated behaviours strongly indicated drinking activity (e.g. birds

roosting in the vegetation alongside the water source, then leaving their roosting

position and descending towards the ground).

To complement the opportunistic observations, and in an effort to characterise the

frequency and intensity with which birds used PWSs at NBG, I also conducted

systematic surveys at dawn and dusk. Each survey included eight (four dawn and four

dusk) surveys across the five days of the week. I conducted two week-long surveys

during each season, starting in October 2008 (spring) and ending in December 2009

(summer), making for n = 12 week-long survey periods and for n = 96 dawn/dusk

visitations to PWSs.

194

The surveys commenced 30 minutes to an hour before dawn and before dusk on the

same day, and lasted for three to four hours. I selected 10 PWSs to visit for these

surveys. These PWS locations were selected on the basis of accessibility, demonstrated

ability to retain standing water for at least a portion of the summer dry period, and

previous observations of birds drinking from them. The PWSs visited in systematic

surveys were distributed across the site (Figure 9.5).

All 10 PWS locations were visited at both dawn and dusk on the same day. If birds were

encountered, I observed the flock for 10 minutes with binoculars, approaching as

closely as possible without causing disturbance and noting the species. I sometimes

encountered black cockatoos drinking at PWSs outside dawn and dusk surveys, or at

ephemeral water sources such as roadside puddles. In such cases, I recorded

observations as described for the systematic surveys.

9.2.3 Camera Trap Observations

I also used motion-sensitive ‘camera traps’ to collect observations of black cockatoos

using water sources at NBG (Dixon et al. 2009; Treves et al. 2010). Originally I had

planned to deploy the camera traps in a systematic manner, in order to establish a

reliable and ongoing method for monitoring black cockatoo drinking activity at selected

water sources. However, two logistical and methodological difficulties prevented this.

Firstly, several factors caused damage or loss of camera traps, including theft, bushfires,

rain, and vehicle movement. Secondly, several factors made it difficult to quantify, and

thus to control for, sampling effort. These factors included: ineffective camera

placement because of glare (at certain periods of the day) or other causes; varying

battery duration; and varying fields of view because of vegetation and the physical

layout of the site. The latter factor was particularly problematic as sites varied

considerably in the ease with which birds could be detected. For these reasons, I also

did not attempt to calibrate probabilities of detection.

These considerations suggested, in the circumstances, the camera traps were best used

as an additional source of opportunistic observations and, in particular, to assess the use

of water sources for which we had no behavioural observations. I reasoned that while

the absence of detection records could not be interpreted reliably as meaning a lack of

black cockatoo visitation to a PWS site, recorded images of black cockatoos did provide

a reliable ‘positive’ detection of birds using a watering point.

195

Camera traps were deployed at 11 PWSs (n = 6 FDPs located around the F1 RDA

(Figure 9.3), and n = 5 seepage or drainage sumps (Figure 9.5)). A single camera was

set up to record continuously for 20 – 30 days at each of the sites from January 2010 to

May 2011. Camera traps were also deployed at two sites for a one-month period in

November - December 2008. At all locations, cameras were tied to a tree trunk or set up

on a pole, facing the bank near the water surface. I selected these five seepage or

drainage sumps because black cockatoos had previously been observed drinking from

these sites. This offered the opportunity to check whether black cockatoos would trigger

the camera trap sensors and to collect additional data on the use of these water sources.

There was no a priori rationale for the selection of FDPs for camera deployments, other

than to sample as many of the FDPs as possible. However, at one site, I deployed a

camera trap at one FDP on multiple occasions because of the apparent suitability of this

site as water source for various fauna.

Donato Environmental Services (DES) investigated the presence and activity of birds

(and other wildlife) at the F1 RDA and at other water bodies at NBG during a series of

site visits between April 2008 and January 2010 (Smith et al. 2010). The aim of these

investigations was to “document wildlife ecology, wildlife mortalities and exposure

pathways to mine waste solutions in order to explore and assess potential protective

mechanisms” (Smith et al. 2010, p. 262).

These investigations involved: (i) a baseline survey conducted in April 2008, July 2008,

and September 2009 (n = 18 days) and (ii) a monitoring regime during the initial

operational phase of the F1 RDA conducted in November 2009 and January 2010 (n =

13 days). DES and Newmont have kindly made their data available for presentation

here.

DES used three observational methodologies that provide information relevant to

assessing interactions between black cockatoos and the F1 RDA: (1) diurnal visual

observation; (2) camera trap monitoring and (3) acoustic monitoring. Sampling effort

for the different observational methods across the water bodies at NBG is shown in

Table 9.3. The methodologies are described in detail in Smith et al. (2010). The relevant

aspects of these methodologies are presented below.

196

(1) Diurnal visual observations: DES observers used binoculars with 8 × magnification

and a telescope with 20 – 60 × magnification to conduct visual observations, usually

within three hours of sunrise. These observations included: (a) 20-minute intensive

surveys in which all wildlife present, entering, and leaving the water body were

recorded; (b) wildlife behaviour and habitat use surveys in which wildlife were

observed at intervals and individuals present classified according to the behaviours

exhibited and habitats used; and (c) interaction surveys of wildlife using cyanide-

bearing habitats. Within the F1 RDA, separate surveys were undertaken for the TDZ (a

zone of high hazard) and the broader RDA area.

(2) Camera trap observations: Two Reconyx RC55 camera traps were deployed at

beaches of the R4 RDA (n = 40 days) and the F1 RDA (n = 100 days) between 15

December 2009 and 28 February 2010 to monitor wildlife activity. The camera traps

were set to be active continuously and to be triggered by motion or heat. Cameras were

positioned along the interface of bare ground and wet tailings, clear pools, or tailings

streams. As a check for cases where birds did do not trigger the cameras, some cameras

were set up to take photographs every 15 minutes. Smith et al. (2010) suggested that

wildlife of duck size or larger may trigger cameras at a distance of 50 – 100 m

(depending on the environment and prevailing conditions).

(3) Diurnal and nocturnal acoustic recordings: Voice-recorders or ‘songmeters’

(Wildlife Acoustics SM2) were used to detect wildlife vocalisations (e.g. calls) at the F1

RDA and R4 RDA. These devices ran for 20 minutes every hour on the hour each day

from 25 June to 20 July 2010 to complement the surveys at the TDZ, and were

primarily used for detecting nocturnal wildlife activity. The songmeters record sounds

between 20 - 20 000 Hz, a range that is audible to humans. Recordings were analysed

by personnel trained in identifying birdcalls to species.

197

Table 9.3 Summary of DES wildlife monitoring survey effort between April 2008 and January 2010 for the F1 RDA, TDZ and the alternative water bodies, the F3 RDA, the R4 RDA and the D1 WSR (from Smith et al. 2010, p. 32).

DES survey effort F1 RDA

TDZ F3 RDA

R4 RDA D1 WSR

Baseline Survey (April 2008 to September 2009)

Days survey conducted 18 - 12 17 8

20-minute surveys 58 - 22 33 9

1-minute foraging surveys 65 - 239 322 0

Operational Phase (November 2009 and January 2010)

Days survey conducted 13 13 1 3 2

20-minute surveys 22 46 - - -

Interaction surveys 9 20 - - -

Habitat and behaviour surveys 71 100 - - -

9.2.4 Wildlife Monitoring by Newmont Boddington Gold Staff

In addition to the DES investigations, NBG staff (or subcontractors) conducted wildlife

monitoring surveys at the F1 RDA and three other water bodies (R4 RDA, F3 RDA,

and D1 WSR) between 9 November 2007 and 28 February 2010. These surveys were

usually conducted within three hours of sunrise and from set observation (vantage)

points. Observers were trained to classify birds present at the water bodies to guilds

(e.g. ‘Waders’, ‘Grebe spp.’) and, in some cases, to species. The survey protocols

included a category ‘Black Cockatoo’.

9.2.5 Monitoring of Wildlife Mortality by Donato Environmental Services Personnel

and Newmont Boddington Gold Staff

DES personnel monitored the F1 RDA for carcasses once tailings disposal had begun.

To obtain background information on carcass detection rates, NBG staff monitored the

R4 RDA for carcasses between November 2007 and February 2010 as part of the

wildlife monitoring surveys.

198

9.3 Results

9.3.1 Behavioural Observations

Twenty-five encounters of one or more black cockatoo species were recorded near

water sources between December 2007 and January 2010 (FRTBC: n = 13; Carnaby’s

Cockatoos: n = 9; Baudin’s Cockatoos: n = 4) (Table 9.4). Four (16%) of these

encounters were classified as probable drinking events (based on behaviours and

environmental contexts indicative of drinking). Three (12%) of these encounters

occurred during systematic dawn/dusk surveys between October 2008 and December

2009, all of which involved FRTBC; the other 22 (88%) encounters involved

opportunistic observations.

Encounters occurred at 11 PWSs (n = 7 sumps; n = 3 WSR; n = 1 dam) and three

natural water sources (n = 1 stream; n = 1 swamp; n = 1 puddle) (Figure 9.5).

Observations of drinking events occurred in seven months (January: n = 8; March: n =

2; April: n = 5; September: n = 2; October: n = 3; November: n = 2; December: n = 3).

The overall mean time difference between time of encounter and time of sunset or

sunrise was 1:26 (range: 0:04 – 03:59; n = 25). The mean time difference between time

of encounter and time of sunrise was 2:00 (range: 0:11 – 3:59; n = 11). The mean time

difference between time of encounter and time of sunset was 1:00 (range: 0:04 – 2:18; n

= 14). Two of the encounters around dusk occurred after sunset (0:04 and 0:25 minutes

after sunset). Eighteen (72%) of the encounters were within two hours of sunset or

sunrise.

The number of birds present during drinking events varied from one to 45 (mean: 11.7 ±

2.16 SE; n = 25) (Table 9.4). Though the sample sizes are small, the mean group sizes

for FRTBC and Baudin’s Cockatoos were more than 10 individuals (FRTBC: 13.2 ±

3.2, range 2 - 45, n = 13 encounters; Baudin’s Cockatoos: 11.5 ± 1.2, range 10 - 15, n =

4 encounters). Mean group size for Carnaby’s Cockatoos was smaller (4.6 ± 1.1, range

1 - 9, n = 8 encounters).

199

Table 9.4 Observations of black cockatoos drinking at water sources at NBG and characteristics of water sources. Sunrise – Sunset: time difference between the time of sighting and the time of sunrise (i.e. after sunrise) or sunset (i.e. before sunset) (a ‘+’ means after sunset). * Small sump between Narrow Pipeline Road and Pipeline Corridor Road: 0439959 / 63782414 # Occurred during dawn/dusk survey FDP: Fauna Drinking Point LEA: Land-Exchange Area (near Sotico Plantation) WT: White-Tailed black cockatoo (when not able to confirm species) Description Detection History

No. Name and General Location Type Method of Detection Species No. Obs. Date Time (h) Sunrise -

Sunset Notes

Permanent Water Sources at NBG

1 CS11 – east of Gate 12 FDP

Camera Trap FRTBC 6 30 Mar 2010 0820 2:04

Camera Trap FRTBC 1 10 Apr 2011 1217 5:45

Camera Trap FRTBC 6 1 May 2011 1700 0:39

Camera Trap FRTBC 7 2 May 2011 1730 0:08

2 CS09 – northwest of site FDP Camera Trap WT 1 22 Jan 2010 1906 0:18

3 Southeast corner R4 Sump R4EESU1

Sump

Camera Trap WT 1 27 Jan 2010 0835 2:58

Camera Trap WT 1 28 Jan 2010 1010 4:32

Behavioural Carnaby 3 14 Dec 2007 0757 1:52

4 Hollow Brook (along Siding Road)

Natural

Behavioural FRTBC 45+ 14 Dec 2007 1800 2:18

Behavioural FRTBC 6 15 Oct 2008 1745 0:43

Behavioural FRTBC 3 11 Nov 2008 1751 1:59

200

5 North of Security Main Gate Sump

Behavioural Carnaby 5+ 22 Jan 2008 1030 3:59 Probable drink

Behavioural FRTBC 20+ 23 Jan 2008 1930 0:53

Behavioural FRTBC 21 17 Sep 2008 1735 0:35 Probable drink

Behavioural FRTBC Carnaby 7 and 4 14 Oct 2008 1727 1:00

Behavioural Carnaby 9 11 Nov 2008 1926 0:25 Probable drink

6 South of L and M Haul Road (eastern) Sump

Camera Trap WT 1 9 Jan 2008 1915 1:11

Camera Trap FRTBC 1 12 Jan 2008 0953 3:31

Camera Trap FRTBC 2 13 Jan 2008 1730 2:56

Camera Trap WT 2 13 Jan 2008 1900 1:26

Behavioural Carnaby 1 2 Jan 2008 1940 0:46

Behavioural FRTBC 10 9 Jan 2008 0712 0:52

Behavioural Carnaby 8 9 Jan 2008 1000 3:40

7 Southeast corner R4 (no. 2) Sump Behavioural FRTBC 4 18 Dec 2007 1815 2:05

8 South of L and M Haul Road (western) Sump Behavioural Carnaby 1 2 Jan 2008 1915 1:11

9 D4 WSR

Behavioural FRTBC 2 18 Jan 2008 0745 1:18 Probable drink

Behavioural FRTBC 21 2 Apr 2009 1700 1:12 #

Behavioural FRTBC 12 17 Apr 2009 0738 1:00 #

10 Eastern perimeter of R4 small pool along edge at northeast corner

WSR Behavioural FRTBC 8 7 Mar 2008 0721 0:11

11 Swamp near to D4 wash-down Natural Behavioural Baudin 10 9 Sep 2008 0855 2:33

12 North of L and M Haul Road Sump Behavioural Baudin 15 26 Mar 2009 0820 0:57

Behavioural Baudin 11 9 Apr 2009 0855 2:23

201

13 D1 WSR Behavioural FRTBC 12 15 Apr 2009 1730 0:26 #

14 Between PCR and NPR* Sump Behavioural Baudin 10 15 Apr 2009 1800 +0:04

15 North of Site Village Dam Behavioural Carnaby 6 19 Jan 2010 1950 +0:25 Acquired Lands

Other Water Sources

16 Ephemeral Source (e.g. puddles)

n/a Behavioural Carnaby 39+ 14 Oct 2009 0857 3:00 LEA

202

During drinking bouts, individuals often remained at the water’s edge for more than a

minute (Figure 9.7). However, while birds sometimes engaged in antagonistic

behaviour while on the ground (e.g. briefly ‘squabbling’), they were not observed to

engage in social behaviour or to preen themselves, even though they exhibited these

behaviours while roosting in trees nearby. In addition, they appeared quite sensitive to

disturbances (e.g. unusual noises, presence of people) while drinking.

203

(a)

(b)

(c)

204

(d)

(e)

(f)

Figure 9.7 Black cockatoo drinking episodes: (a) flock of Carnaby’s Cockatoos at a man-made sump; (b) flock of FRTBC at a WSR; (c) Carnaby’s Cockatoo at a natural puddle; (d) motion-triggered camera image of a black cockatoo at a man-made sump; (e) motion-triggered camera images of a group of FRTBC, and (f) Carnaby’s Cockatoo drinking at two of the more recently installed FDPs ((e) – (f) are courtesy of NBG).

205

9.3.2 Camera Trap Observations

Camera trap deployments yielded 11 detections of drinking events at four (36.4%) of

the 11 PWSs where deployments occurred (Table 9.4). Black cockatoos were detected

at two (33.3%) of the six FDPs and two (40%) of the five drainage and seepage sumps

monitored (Figure 9.7).

Six (54.5%) detections were of FRTBC and five (45.5%) were of White-tailed black

cockatoos (i.e. either Baudin’s Cockatoos or Carnaby’s Cockatoos). Observations of

drinking events were detected in four months (January: n = 6; March: n = 1; April: n =

1; May: n = 2). Black cockatoos were recorded in every month where cameras were

active except February. The number of birds detected ranged from one to seven, with

most detections (n = 6, 54.5%) being of a single bird.

The overall mean time difference between time of sighting and time of sunset or sunrise

for camera trap detections was 2:18 (range: 0:08 – 5:45 hours; n = 11). The mean time

difference between time of sighting and time of sunrise was 3:16 (range: 2:04 – 4:32; n

= 4). The mean time difference between time of sighting and time of sunset was 1:46

(range: 0:08 – 5:45; n = 7). In slight contrast to the behavioural observations, only four

of the 11 detections (36.4%) of the detections were within two hours of sunset or

sunrise.

9.3.3 Donato Environmental Services Observations

(1) Diurnal Visual Observations

DES personnel observed 25 bird species and recorded 8 701 wildlife visitations to the

F1 RDA during diurnal observations conducted in the baseline survey period (2008 -

2009) and in the monitoring regime for the initial operational phase of the F1 RDA

(2009 - 2010) (including observations conducted during 20-minute intensive surveys

and opportunistically). These visitations included one sighting of two FRTBC and one

sighting of 23 Baudin’s Cockatoos. Both sightings were during the baseline survey

period. No black cockatoos were observed during monitoring in the operational phase,

which included 68 intensive 20-minute surveys of the supernatant and TDZ in

November 2009 (n = 39 surveys) and January 2010 (n = 29 surveys). Smith et al. (2010)

state that the two sightings were of flocks: ‘flying overhead between forest blocks and

the records and absence in the causation data [i.e. monitoring during the operational

phase of the F1 RDA] is unrelated to tailings disposal’ (p. 278) and more specifically

206

‘flying over the northern tongue from the one patch of forest to another on the other side

of the RDA system’ (p. 48).

DES personnel observed 13 bird species and recorded 789 during observations of

wildlife habitat use at the F1 TDZ between November 2009 and January 2010 (n = 100

counts of bird flocks). These observations included one sighting of three FRTBC at the

TDZ (0.38% of the individuals recorded). The activity recorded for these individuals

was ‘Locomotion/Patrolling - Aerial’.

DES personnel observed 15 bird species and recorded 789 during observations of

wildlife habitat use at non-TDZ habitats at the F1 RDA between November 2009 and

January 2010 (n = 111 counts of bird flocks). Black cockatoos were not recorded during

these surveys.

Black cockatoos were not recorded during observations conducted at the R4 RDA

during the baseline survey period (April 2008, July 2008, September 2009). Black

cockatoos were observed feeding on weeds present along the beach at an RDA on one

occasion (D Donato, Donato Environmental Services, 2011, pers. comm.).

(2) Camera Trap Observations

The DES camera traps did not capture the presence of black cockatoos at either the F1

TDZ or the R4 beach. Two wildlife species were recorded at the F1 TDZ – Emus and

Western Grey Kangaroos. Several larger bird species (e.g. ducks, herons) were detected

at the R4 beach.

(3) Diurnal and Nocturnal Acoustic Recordings

Voice recognition detected no black cockatoos during 53 hours of recording at the F1

TDZ and 36 hours of recording at the R4 RDA.

9.3.4 Wildlife Monitoring by Newmont Boddington Gold Staff

NBG staff recorded 13 611 wildlife observations from at least 28 species during 337

surveys of the F1 RDA between November 2007 and February 2010 (n = 148 prior to

tailings discharge, n = 189 post tailings discharge). No black cockatoos were observed

during these surveys.

207

NBG staff also conducted 358 surveys at the R4 RDA (n = 124 surveys), F3 RDA (n =

118), and D1 WSR (n = 116) water bodies between November 2007 and February 2010.

No black cockatoos were detected at the R4 and F3 RDAs. Seventeen black cockatoos

were detected at the D1 WSR.

9.3.5 Monitoring of Wildlife Mortality

No carcasses of black cockatoos were recovered by DES personnel or NBG staff. DES

detected six wildlife carcasses at the F1 RDA in January - February 2010. NBG staff

detected 11 wildlife carcasses at the R4 RDA during the 27 months of monitoring.

9.4 Discussion

This chapter addressed the key questions of: (i) the ecology of black cockatoos and

water at NBG; (ii) the characteristics of water sources at NBG; (iii) evidence relating to

interactions between black cockatoos and RDAs; (iv) use of FPDs; and (v) limitations

of the study and future improvements.

9.4.1 Behavioural Ecology of Black Cockatoo Water Use at Newmont Boddington Gold

Behaviours while drinking: The behaviour of birds during drinking bouts suggest that

birds are quire wary of predation while on the ground and drinking, e.g. drinking bouts

were brief, birds immediately returned to roosting in nearby trees after drinking, birds

appeared sensitive to disturbance while drinking.

Activity patterns: Most observations of drinking events were within a few hours of

sunrise or sunset. At other locations, black cockatoos often over-night roost near water

sources and drink when leaving or occupying these roosts, though they may drink at

other periods (e.g. after morning feeding bouts) (Saunders 1977, 1980; Davies 1982;

Shah 2006; Cameron 2007; Finn et al. 2009). Over-night roosting near water sources

also occurs at NBG (J Lee, Murdoch University, pers. obs.). The camera trap detections

indicate, however, that birds may drink at any time during the day. The camera trap

detections are an important supplement to the behavioural observations as observational

effort was concentrated in the early mornings and late afternoons.

Interspecies variability: All three species were observed using PWSs at NBG. The

drinking behaviour of the three species is likely to differ, as the three species vary in the

208

frequency with which they will come to ground. Carnaby’s Cockatoos commonly

forage on the ground, either because to feed on ground-based foods (e.g. Banksia

dallaneyi) or on the seeds in fallen fruits and cones (Chapter 3; Saunders 1974, 1980;

Finn et al. 2009). Baudin’s Cockatoos do not generally feed on ground-based foods, but

may occasionally do so (Johnstone and Kirkby 2010). FRTBC rarely come to the

ground apart from when they have to drink and are cautious when doing so (T Kirkby,

Western Australia Museum, 2009, pers. comm.).

Though the sample sizes are small, the findings suggest that FRTBC and Baudin’s

Cockatoos may drink in larger groups than do Carnaby’s Cockatoos. This differs from

the patterns in group sizes observed for the three species generally (see Chapter 3). If

these patterns are real, then they may reflect differences in aggregative behaviour

associated with drinking events and/or the dawn/dusk periods. Anecdotal observations

of FRTBC suggest that smaller groupings (e.g. pairs, family units) may aggregate

around water sources around dusk (J Lee, Murdoch University, pers. obs.). This would

be consistent with an ecology in which FRTBC are dispersed in small groups across the

landscape during the day, but may form larger groups to drink and possibly also to over-

night roost. FRTBC have a metabolic rate higher than the White-tailed species and

therefore drink more frequently (Cooper et al. 2002). Their distribution patterns, and

tendency to be resident within defined areas, may therefore reflect, the need to ensure

reliable access to known water sources. Observations of FRTBC in the eastern acquired

lands suggest that, in summer, FRTBC are concentrated near paddock areas where farm

dams maintain water through the summer period (Finn 2011).

The observations of Baudin’s Cockatoos group size appear consistent with their general

ecology in the NBG area (i.e. transient groupings of 10 - 25 birds). The observations of

Carnaby’s Cockatoos group size likely reflect the more variable ecology of this species

at NBG. For example, single Carnaby’s Cockatoos were sometimes observed during the

breeding season, which suggests one of breeding pair coming away from the nest to

drink.

9.4.2 Range and Characteristics of Water Sources

Behavioural observations and camera trap detections demonstrated use of 15 different

water sources at NBG (not including ephemeral water sources). It is likely that black

cockatoos also used other water sources, both natural and man-made, but that survey

209

effort failed to document this use. The findings suggest that black cockatoos clearly do

make use of existing artificial water sources on-site and thus may be expected to make

use of new sources when/if these are created.

The range of water sources observed suggests four general categories: (1) sumps and

dams; (2) water reservoirs; (3) streams and wetlands; and (4) ephemeral pools and

puddles. While categories (1) and (2) obviously refer to man-made structures, the water

sources falling within categories (3) and (4) may be ‘natural’ (e.g. streams in native

forest habitats) or have some degree of anthropogenic influence (e.g. streams and

swamps within paddocks or near roads). An additional factor for classification is the

persistence of standing water and whether the water sources are ephemeral (i.e.

available for hours to days), longer-term to semi-permanent (i.e. dry during some

portion of summer), or permanent (i.e. having water year-round regardless of rainfall).

Most of the water sources were sumps or dams. These structures were typically

designed to assist in drainage, and therefore were built either to retain surface water

runoff from roads or to contain sub-surface seepage from R4 (which was originally

designed an RDA). Only three water sources – the Siding Road creek, the D4 wash-

down swamp, and the various ‘puddles’ were not man-made structures. However, even

these water sources arguably had a significant amount of anthropogenic influence

through the imposition of physical barriers (a dam wall), the removal of surrounding

vegetation, or the physical alteration of the landscape.

Known drinking sites at NBG are generally surrounded by vegetation, but usually have

areas of open ground immediately around the water. In addition, there are often large

trees nearby in which birds can roost before, during and after drinking. Commonly,

some birds will roost in the tree(s) while other birds drink. Most water sources had trees

greater than 10 m in height within a 10 – 20 m distance of the water edge and were

bordered by vegetation on one or more sides. The exceptions were the D1 and D4 water

supply reservoirs and the use of a pool of water along the outskirts of the R4 reservoir.

As described above, birds used the trees adjacent to the water sources to roost in before

and after drinking. This shortened the distance they needed to fly from cover and down

to the water source and thus minimised the time in which they were exposed on the

ground.

210

Another salient feature of the water sources appears to have been the presence of a firm

and gently inclined approach edge, in order to allow the birds to access the water source

without slipping or becoming mired in mud.

9.4.3 Is there Evidence that Black Cockatoos Use, or Occur in Proximity to, the Residue

Disposal Areas?

The findings suggest that black cockatoos only infrequently come into proximity with

potentially hazardous sections of RDAs. Although it is problematic to attribute a

behavioural preference without a more rigorous experimental design, the combination

of the DES observations with my observations suggests that black cockatoos prefer to

use water sources other the RDAs. With one exception, black cockatoos were only

observed flying overhead of the RDAs. There were no records of them drinking from

the F1 RDA, nor were any carcasses recovered, despite a reasonably intensive

monitoring regime for the F1 RDA.

Nonetheless, certain observations do suggest situations in which interactions could have

occurred. Firstly, DES personnel observed FRTBC feeding on weeds along the beach of

the RDA. Secondly, birds were observed drinking along the perimeters of the three

water supply reservoirs at NBG – R4, D1, and D4. These observations, along with the

detection of two carcasses at F3 and R4, suggest that black cockatoos may occasionally

encounter hazardous portions of the RDAs and that the risk of mortality may relate to

ingestion of water containing toxic concentrations of cyanide and/or entrapment within

‘muddy’ unconsolidated materials.

If this conclusion is correct and black cockatoos rarely, if ever, drink from the active

portions of the RDAs, why might this occur? While such comments are necessarily

speculative, several factors could make, either by themselves or in combination with

each other, the RDAs undesirable as water sources.

One factor could the openness of the landscape around the RDAs. The RDA sites are

large, open bodies of water located some distance from a forest edge, forcing birds to

travel some distance before they can land to drink. Drinking birds thus may perceive a

greater risk of predation (Cameron 2006). While observations of birds drinking from

R4, D1, and D4 clearly indicate that openness does not necessarily preclude drinking

activity, the balance of the behavioural observations does suggest that birds prefer water

211

sources where vegetation is close by, allowing birds to shelter when not on the ground.

In addition, anecdotal observations of birds feeding and drinking do suggest that some

birds maintain a ‘look-out’ for predators while other birds feed or drink (J Lee,

Murdoch University, pers. obs.).

The second factor could be muddiness in areas of active discharge (i.e. the areas where

birds might be exposed to solutions containing potentially harmful concentrations of

cyanide). The behavioural observations indicate that black cockatoos prefer a firm

substrate on which to approach a water source, based on the combination of gravel and

hard clay that surround the sumps and dams used as drink sites.

A third factor could relate to the current topography of the F1 RDA. Currently the

active discharge areas for F1 is ‘low’ relative the surrounding landscape, meaning that

birds must fly down towards it (D Donato, Donato Environmental Services, 2011, pers.

comm.). This would be analogous to flying down towards a stream at the base of a

valley. As with the openness of the landscape, the current ‘lowland’ topography of the

F1 RDA may increase the perceived risk of predation.

A fourth factor, though not characteristic of the RDAs themselves, may be that the

surrounding landscape maintains alternative, and more suitable, water sources. There

are several water sources at NBG that hold water year-round and birds may not need to

fly out to the RDAs to drink, even during drier months. This hypothesis imputes a

certain degree of ‘local knowledge’ to the flocks occurring at NBG and thus it is logical

to conclude that flocks unfamiliar with the site may be at the greatest risk of using the

RDAs (T Kirkby, Western Australian Museum, 2011, pers. comm.). The availability of

year-round water sources may also mean that birds are present at NBG when other areas

of the broader forest landscape have inadequate water sources. Other factors that could

deter usage of the RDAs include: the ‘saltiness’ of the discharge solutions; the frequent

traffic and human activity around the RDAs; and the presence of raptors.

It is likely that some or all of these factors make the RDAs less preferable than the other

waters sources. If these ‘undesirable’ characteristics can be maintained or enhanced, this

would minimise the risk of interaction. Similarly, the observation of birds feeding

around the beach area suggests the need to remove potential food sources near the

RDAs.

212

9.4.4 Use of Fauna Drinking Points

The camera trap deployments confirmed that black cockatoos drink at some of the

newly established FDPs located around the RDAs, but along the edge of the forest-line.

This suggests that these artificial drinking points may be successful and valuable in

keeping black cockatoos away from the RDAs.

9.4.5 Limitations and Future Improvements

The use of, and probable reliance on, man-made water sources is not unique to NBG.

Such usage commonly occurs in agricultural and urban areas where areas of standing

water are artificially maintained in dams, ponds, fountains, and similar structures or by

watering gardens and lawns (Berry 2008, Finn et al. 2009). However, the availability at

NBG of such a ‘network’ of water sources increases the overall suitability of NBG for

breeding and for feeding. Further, the value of these water sources may be particularly

important during summer and autumn when other natural and man-made water sources

may lack water. Thus the NBG landscape may be better able to sustain flocks during

dry periods than some other habitats within the Jarrah-Marri forest landscape.

Ideally, future work would systematically monitor water sources and quantify the

frequency with which particular water sources are used. The DES monitoring followed

a set sampling methodology and was site-focused (i.e. focused detecting wildlife around

the RDAs and the FDZ in particular). In contrast, observational effort in this study was

more opportunistic. For example, the study only involved targeted dawn and dusk

surveys for limited periods each season, which may be insufficient in capturing large

amounts or consistent drinking data, especially considering the migratory habits of the

Carnaby’s Cockatoos and Baudin’s Cockatoos and the nomadic nature of FRTBC (the

seasonal movements of FRTBC are irregular) (Sedgwick 1949; Storr and Johnstone

1988). The importance of ephemeral water bodies was not assessed. One suggestion to

provide long-term water use data would be to increase the number of surveys performed

each season.

Camera traps were useful for detecting the use of water sources, recording instances of

birds drinking at mid-morning and noon, when behavioural observations were likely to

underestimate. Likewise, without intensive behavioural sampling, it may not have been

possible to document the (apparently) infrequent use of certain water sources. These are

213

very clear examples of the value of camera traps for increasing sampling effort and in

obtaining rare or atypical observations.

If the practical and methodological issues involving factors as placement of traps,

sensitivity settings, and battery duration can be resolved, it may be possible to estimate

sampling effort and probabilities of detection obtained. Calibration is essential to ensure

that cameras detect birds if they are present, so negative results can be interpreted

confidently. The deployment of multiple camera traps at a site may allow for this. In

addition, setting cameras to take photographs at specified intervals identifies situations

when birds are not triggering motion sensors. Furthermore, cameras may underestimate

drinking flock sizes, because they focus on one area and birds may be drinking

elsewhere. Multiple cameras would also help correct this. Finally, the use of higher

resolution images may help in identifying species of White-tailed black cockatoo.

It is difficult to determine to if there is a clear association between the presence of the

FDPs and the apparent absence of black cockatoos at the RDAs. The findings indicate

that black cockatoos used FDPs and that birds were not observed to use the RDAs when

the FDPs were in place. However, we lack baseline observations for use of the RDAs

before the FDPs were constructed. Nonetheless, given the potential hazard of the RDAs

to black cockatoos, using a range of mitigation measures increases confidence in

keeping exposure low.

Consideration could also be given to revegetation near watering points to provide

shelter from predators for birds gathering to drink. Raptors hunt in more ‘open areas’

(such as around the areas of grass-like vegetation around the RDAs), than in the more

forested habitats, where the overarching canopy cover makes these areas intrinsically

less ‘risky’ (J Lee, Murdoch University, pers. obs.). Predation risk may be important

consideration within landscapes as open as the RDAs. Its is likely that the presence of

vegetation, particularly trees or taller shrubs, near standing water may encourage

drinking activity because birds will be able to roost in the vegetation during drinking

bouts.

214

Chapter 10 General Discussion and Synthesis

This chapter summarises the key findings from earlier chapters in relation to the seven

main research aims for this thesis (see Chapter 1). On the basis of that information, I

then examine how the results of this study can be applied on site at NBG and in the

diverse array of land uses and vegetation forms present in the southwest through an

adaptive management framework. An adaptive management framework treats

management interventions as experimental manipulations to test hypotheses about their

effectiveness. Such an approach is particularly appropriate for production landscapes,

given their ecological importance to black cockatoos and socio-economic value for

humans.

10.1 Overview of Principal Findings

This broad study of two black cockatoo species (Carnaby’s Cockatoo Calyptorhynchus

latirostris and Baudin’s Cockatoo C. baudinii) and a subspecies (Forest Red-tailed

Black Cockatoo C. banksii naso; FRTBC) addresses the conservation of black

cockatoos within landscapes devoted to mining. The field research was undertaken at

the Newmont Boddington Gold (NBG) mine, a mine-site approximately 130 km

southeast of Perth, Western Australia, within the Jarrah-Marri forest of southwestern

Australia. The area is significant because all three black cockatoos occur there and

because it lies at the ecotone between the high rainfall Jarrah-Marri forest of the

southwest and the lower rainfall Wandoo (E. wandoo) woodland to the east.

Conservation efforts for black cockatoos at NBG have been hampered by poor baseline

information on basic ecology at the site, as well as uncertainty about the value of

rehabilitated mine pits as feeding habitat for black cockatoos. Other specific problems

identified were: (1) the possible lack of suitable nest hollows following past forest

management and agricultural clearing, and more recent clearing for mining, and (2) the

interactions of black cockatoos with residue disposal areas (RDAs) containing cyanide-

bearing materials.

These issues were addressed in the seven main research aims of the thesis. These aims

aligned with two recovery actions identified in the Forest Black Cockatoo Recovery

Plan ‘Determine and implement ways to minimise the effects of mining on habitat loss’

215

and ‘Map feeding and breeding habitat critical to survival and important populations,

and prepare management guidelines for these habitats’ (Chapman 2008), and with the

general recovery objectives identified in the 2002 - 2012 Recovery Plan for Carnaby’s

Cockatoo (‘ensuring the species persists within its present range and increasing

population numbers within its present range and by expansion into its former range’)

(Cale 2003). Similar concerns were also identified in the most recent Action Plan for

Australian Birds (Garnett et al. 2011) and in new, recently released (October 2012)

Recovery Plan for Carnaby’s Cockatoos (DEC 2012).

The principal findings in relation to the thesis aims were as follows:

Aim 1: Describe the ecology of the three black cockatoos at NBG, particularly

patterns in group size, site occupancy, habitat use, and food plant use; including

seasonal and inter-annual changes (Chapter 3).

Fields studies of black cockatoos at NBG led to these important findings:

All three black cockatoos were encountered in remnant native forests as well

as human-modified habitats, with Carnaby’s Cockatoo using the broadest

range of habitats including native forest, mine-site rehabilitation, and pine

plantations.

Carnaby’s Cockatoo occurred frequently at NBG and in relatively larger

flocks that displayed seasonal changes in group size and occupancy, compared

to FRTBC that occurred less frequently and in smaller family groups with

occupancy patterns that suggests year-round residency and lack of seasonality

in numbers and movements. Baudin’s Cockatoo were also present in large

flocks, but displayed higher abundances in autumn and winter and fewer

numbers in spring and summer.

Carnaby’s Cockatoo was the most generalist of the three black cockatoos and

fed on 10 of the 16 feed plant species recorded on site, which comprised of

proteaceous shrubs (Banksia and Hakea spp.) and introduced Pinus species,

while Baudin’s Cockatoo and FRTBC fed primarily on Marri. Carnaby’s

Cockatoo and Baudin’s Cockatoo were also observed feeding on grubs.

Overall, these features show that within a single landscape, these three black cockatoos

can exhibit differences in group size, seasonal and annual occupancy, habitats and food

216

plants used. Therefore, they are likely to differ in their response to disturbance and in

their use of anthropogenic resources. These differences have important implications,

and should be considered when assessing the relative vulnerability of the three

cockatoos to land clearing and other impacts, and for evaluating the potential

effectiveness of proposed management measures or offsets.

Aim 2: Examine the effectiveness of ground-based hollow surveys, post-felling

inspections of hollows, and behavioural observations as approaches for assessing

black cockatoo breeding habitat in the Jarrah-Marri forest (Chapter 4).

Assessing the potential value of a site as breeding habitat for threatened black cockatoos

requires field methodologies that can identify potentially suitable nesting hollows and

probable nest sites. The clearing of 1 203 ha of Jarrah-Marri forest at the NBG site

provided an opportunity for the evaluation of three approaches to locate potentially

suitable hollows and probable nesting hollows. They were: (1) ground-based surveys for

hollows, (2) post-felling inspections of hollows, and (3) behavioural observations of

black cockatoos.

The studies identified 11 probable black cockatoo nest hollows (n = 7 for Carnaby’s

Cockatoos, n = 3 for FRTBC and n = 1 for an unknown black cockatoo species).

Behavioural observations of black cockatoos (through visual and acoustic cues that

indicated the potential presence of a nest and was accompanied by ‘tree knocking’)

were the most effective approach to identify probable nest hollows (n = 10 hollows).

Despite the large sample sizes and extensive field survey periods, ground-based surveys

(n = 1 hollows) and post-felling inspections (n = 0 hollows) were ineffective at

identifying nest hollows. Ground-based surveys identified 149 trees with potentially

suitable hollows from Jarrah, Marri and Wandoo (of which 119 trees survived felling

intact enough for inspection). Few of the potential hollows identified in Jarrah trees

were considered large enough to be potentially suitable for black-cockatoos (n = 28 of

89 trees inspected). In contrast, large hollows occurred more frequently in Marri (n = 14

of 22 trees inspected) and Wandoo (n = 8 of 12 trees inspected). Of the probable nest

hollows, six hollows were in Marri, four in Wandoo, and one in Jarrah.

217

These findings suggest that:

The most effective approach for identifying nest sites is targeted behavioural

observations at dawn and dusk during known breeding seasons.

Ground-based surveys may significantly overestimate the presence of

potentially suitable hollows, but may provide useful assessments of relative

hollow abundance if biases are identified and corrected.

Similarly, post-felling inspections appear ineffective at characterising hollow

occupancy, but provide opportunities for the collection of other data.

Aim 3: Assess the successional stage of the rehabilitated mine pits and

characterise variation in the structure and floristics of pits, in order to identify

features that might influence the availability of food resources for black cockatoos

(Chapter 5).

From the revegetation sampling, both the forest tree species and understorey vegetation

were successfully established at the rehabilitation pits. Most of the rehabilitation pits at

NBG have advanced beyond the initial ‘scrub’ successional stage and reached an

interim stage in which the canopy-forming trees are beginning to predominate and

become a significant structural feature, although high densities of proteaceous shrubs

remained. Two structural and floristic characteristics indicate that pits are at an

intermediate successional stage: (1) the stem density for live stems was higher than the

stem density for dead stems, indicating that the revegetation has become established at

these sites rather than failed and (2) the height differences between the canopy and non-

canopy species indicate a demarcation between the canopy and non-canopy (shrub- or

ground-level vegetation) layers. The mean heights of the canopy species ranged

between 2 and 6 m, while the mean heights of the non-canopy species were all less than

1 m.

Overall the findings suggest few edge effects at the pits, as the interior and exterior

plots differed only in the frequency of dead stems and in canopy cover, both of which

were higher in interior plots (which may be consistent with the greater canopy cover

shading and killing understorey species and late developing individuals of canopy

species). It is possible that the similarity of the exterior and interior vegetation relates to

the generally small sizes (with lesser interior area) of the rehabilitation pits (which

218

ranged 0.7 ha to 14.9 ha). However, it is more likely that: (1) the proximity (<50 m in

most cases) of the pits to remnant native forest vegetation limits the influence of edge-

related effects and (2) uniformity in the rehabilitation prescriptions for each pit (e.g.

ripping, seeding) leads to similar vegetation establishing across each pit.

Vegetation varied structurally and floristically across pits despite the use of similar seed

mixes and a similar age range, and these pit-scale differences in structure are common

in mine-site rehabilitation within the Jarrah-Marri forest. This suggests that the

trajectory of each pit was influenced by pit-scale factors relating to the rehabilitation

landscape, environment, soil properties, or some artifact of the rehabilitation process

such as differences in fertiliser application, seeding rates. Some of the differences

between pits are readily observable such as more open woodland-like appearances in

the K pits.

As succession proceeds and the structure (and species composition) of the vegetation

becomes closer to that of the native Jarrah-Marri forest (with lower stem densities, more

well-developed overstorey, and moderately- to poorly-developed understoreys), the

availability of habitat resources is likely to change. It is likely that the abundance and

diversity of the proteaceous shrub species and therefore availability of these under-

storey food sources will decline and become thinner, as canopies of Jarrah and Marri

increase in volume (thereby reducing the availability of light to the mid- and under-

storey vegetation). Evidence of structural and floristic differences between pits likely

means that differences in food availability between pits also occurs, such as variation in

the abundance and quality of Marri fruits. Such differences in availability of food

resources is the most likely explanation for the variation in feeding activity found across

pits.

Aim 4: Document feeding activity by all three black cockatoos within

rehabilitated mine pits and any associations with structural or floristic features

of the vegetation (Chapter 6).

Studies of black cockatoos at NBG rehabilitation pits led to these findings:

From the behavioural observations, Carnaby’s Cockatoo and Baudin’s

Cockatoo were the two most frequently encountered species in mine-site

219

revegetation. FRTBC were not observed using the revegetation, but were

detected in forest trees adjacent to mine-site rehabilitated pits. Group sizes for

both White-tailed species were similar to those observed in remnant forest

areas.

Behavioural observations suggest that feeding is the predominant activity

undertaken by black cockatoos while in rehabilitation vegetation. Also, it

appears likely that Carnaby’s Cockatoo accounted for most of the feeding on

proteaceous plants in rehabilitation vegetation. Baudin’s Cockatoo fed on

Marri in the rehabilitation plots and are also likely to have fed on proteaceous

plants, although this was not directly observed.

Plot sampling of feeding residues confirmed that all three black cockatoos

used rehabilitated mining pits for feeding, even although FRTBC were never

seen feeding in rehabilitated pits.

The counts of feeding residues suggest that two proteaceous species (H.

undulata and B. squarrosa) and the myrtaceous Marri were the predominant

food plants for black cockatoos. Baudin’s Cockatoo and FRTBC ate mainly

Marri. However, as feeding residues of proteaceous plants could not be

assigned positively to individual species, it is possible that all three black

cockatoos may have eaten some proteaceous plants.

While feeding activity varied from pit to pit, comparisons of interior and

exterior plots within a pit demonstrated no difference in feeding activity; this

suggests that black cockatoos do not have a preference for feeding location

within these rehabilitated mining pits. The differences in feeding activity

between pits could be accounted for by the variation in vegetation structure

and in species composition (i.e. the food plants present) across pits.

Feed and non-feed plots were similar structurally and floristically, so

particular structural or floristic variables associated with a higher likelihood of

feeding could not be identified. However, the effect of any structural or

floristic variables may be overshadowed by the variability across individual

rehabilitation pits in their proximity to water or roosting sources. As the

canopy species mature, a greater dependence on myrtaceous overstorey plants

is more likely.

Overall, this study confirmed that rehabilitated mine pits at NBG offer a food resource

for all three black cockatoos in the form of proteaceous and myrtaceous food plants, and

220

indicated that food resources can start to become available as early as six years post-

establishment. The youngest rehabilitation pits were revegetated in 2002, and I found

feeding residues of proteaceous and myrtaceous species in 2008. This suggests that if

those species continue to be included in the seed mix for restoration, then the

rehabilitation pits will supply food for black cockatoos within a few years.

Aim 5: Trial the implementation of artificial nest hollows to support breeding

on-site and compensate for the loss of natural hollows cleared by mining

(Chapter 7).

Studies of black cockatoos at NBG and the installation of artificial nest hollows (ANHs)

led to these findings:

In this trial, black cockatoos at NBG were not observed inspecting or using

either cockatube or wooden box ANHs implemented at the site.

Black cockatoos were observed in the vicinity of ANHs at NBG but were not

observed to use them. The lack of use may be attributed to: (1) unfamiliarity

with ANHs; (2) structural features of ANHs that might be unappealing (e.g.

ANH shape, dimensions or material); (3) ANH position (i.e. too low in the

tree) or location (i.e. far from an established nesting site); and (4) the presence

of sufficient natural hollows in the NBG landscape to support current breeding

populations of black cockatoos.

Despite the lack of use, future modifications to the NBG study area that lead to loss of

natural hollows may increase the attractiveness and utilisation of these ANHs. The

establishment of ANHs at the site should take into consideration factors such as

durability and placement, maintenance and monitoring, as well as control of potential

nest hollow competitors.

221

Aim 6: Conduct a review of the use of artificial nest hollows for black cockatoos

to assess their value as a tool for mitigating natural hollow loss (Chapter 8).

The results from a state-wide survey provided these major findings:

The range of individuals and organisations involved in the use of ANHs for

black cockatoos in Western Australia belonged to what presumably are all of

the key stakeholder categories for black cockatoo conservation in the State,

including independent lay-people and professionals. This suggests there is

considerable interest in the issue of whether ANHs can be used to support

black cockatoo conservation efforts.

The survey determined that at least 157 ANHs were established from as early

as 1996 in both natural and captive settings. They were placed within a range

of land tenures (e.g. private land properties, remnant vegetation, public or

State land) and comprised three general types of ANHs used (i.e. cockatubes,

hollowed-out log sections and wooden box-type ANHs).

ANHs were attached to the mesh in cage settings at heights of less that 2 m,

while ANH installation in the wild required equipment such as vehicles,

elevated work platforms and rope-pulley systems assisted by personnel on

harnesses. These ANHs were usually attached to the main trunks or forks of

living trees (i.e. usually the dominant nest tree species in the area) at heights

of 2 - 17 m, or a metal pole fixed into the ground and adjacent to trees. In

natural settings, monitoring occurred opportunistically and was associated

with the nesting season of black cockatoos, while ANHs in aviary settings

facilitated weekly to monthly monitoring. Monitoring involved climbing to

check inside of ANHs, looking for external signs of use from chewing and

tapping the base of trees to flush nesting birds. Maintenance requirements

involved replacing sacrificial wooden posts, re-filling the woodchip nesting

material and filling the gaps in wood-based ANHs. The total cost of ANH

construction, transportation and establishment ranged from $400 - $700

overall.

Black cockatoos nested and reared young in all designs of ANHs, with most

observations of black cockatoos using ANHs involving Carnaby’s Cockatoos.

FRTBC accounted for a smaller number of records. To date, there are no

records of Baudin’s Cockatoos using ANHs in the wild, but they do in captive

222

settings. ‘Successful ANHs’ were reported by multiple, independent groups,

and hence success was not attributable to unique features of one group at one

site.

Some potential hazards presented to black cockatoos by ANHs included:

occupation by non-target competitor species (e.g. feral bees and Galahs),

suspected predation of chicks by cursorial species (e.g. feral cats and monitor

lizards). Only a small number of injuries were caused by ANH use and were

associated with the ladder in captive settings. However, damage to ANHs by

the birds and environmental conditions (i.e. tree fall or fire) were seldom

serious and few ANHs were removed permanently (i.e. due to lack of use).

Other concerns from ANH workers include the costs of ANH implementation,

maintenance and monitoring, and associated safety and access issues.

Overall, the existing ANH designs all consider the important principles of thermal

insulation, durability, adequate internal dimensions, drainage, accessibility, safety and

provision of chewing material. However, some important considerations gathered from

the study include: (1) improved monitoring to assess use by target and non-target

competitor or predator species; (2) establishing safety procedures to guide installation;

(3) longevity and associated maintenance and costs; (4) ANH positioning (i.e. height

and location) proximate to known nesting grounds with reduced natural hollow

availability to increase breeding or at non-breeding sites with feeding/watering

supplementation to encourage breeding; (5) future studies in to differences in

preferences of ANH designs across the three black cockatoos or conditions of the ANH

itself; (6) application of ANHs in other areas with high canopies (e.g. Karri forest) to

encourage breeding by less known species such as Baudin’s Cockatoo; as well as (7) the

consideration of other methods to encourage more rapid hollow development in natural

standing trees.

The information drawn from the survey suggests that ANHs may, in some

circumstances, present a feasible short-term mitigation option, especially in areas

deficient in natural hollows. However, to sustain current populations of black

cockatoos, they cannot substitute for natural hollows, so the major focus of conservation

should be the retention of breeding and feeding habitats. Also, information exchange

among ANH workers is critical.

223

Aim 7: Investigate the use of natural and artificial water sources at NBG and

assess the potential for interactions with residue disposal areas (Chapter 9).

Studies of black cockatoo populations at watering locations in the NBG site led to these

major findings:

All three black cockatoos were observed using water sources at NBG.

Drinking was observed at 15 distinct locations, including natural (e.g. streams

and wetlands) and man-made sources (i.e. faunal drinking points, sumps,

dams, and reservoirs) that held water year-round or nearly so. Ephemeral

water sources were also utilised.

Most drinking occurred within a few hours of sunrise or sunset. Black

cockatoos appear to be more wary of potential predators or disturbance when

on the ground and drinking.

Black cockatoos appear to prefer water sources with firm and gently inclined

edges for access, surrounded by vegetation.

Black cockatoos infrequently came into proximity with potentially hazardous

sections of RDAs, and were primarily observed flying overhead. Several

factors may make these RDAs less preferable than the other drinking present

at NBG including: greater perceived predation risk associated with the lower

and more open landscape of RDAs and their surrounds; frequent traffic and

human activity; and the availability of suitable alternative water sources

nearby.

Finally, black cockatoos were also observed using the newly established

faunal drinking points (FDPs) around the RDAs.

Both natural and man-made water sources in the NBG study area are able to provide

adequate, permanent watering resources for all three black cockatoos across the year,

especially in the dry season. Furthermore, the establishment of FDPs around the

potentially hazardous RDAs appears, based on the initial observations in this study, to

provide a useful measure for minimising black cockatoo-RDA interactions.

To help connect the findings from the various chapters and bring them all together, refer

to the summary table (Table 10.1) on the next page.

224

Table 10.1 Summary table comparing all three black cockatoos for each of the key social or ecological parameters measured.

Black Cockatoo species

Carnaby’s Cockatoo Baudin’s Cockatoo Forest Red-tailed Black Cockatoo

General Ecology on-site

- Used the broadest habitat range – native forests/woodlands, mine-site rehabilitation and pine plantations and paddocks

- Occurred frequently on-site in relatively large flocks (10.1 ± 1.3, range: 1 – 90)

- Seasonal changes in group size and occupancy observed (smaller flocks observed in autumn and winter)

- A generalist – used 10 of 16 feed plant species including Pinus and proteaceous species

- Used native forests/woodlands, mine-site rehabilitation and paddocks

- Occurred in large flocks (14.5 ± 3.5, range: 2 – 107) - Higher abundances in autumn and winter and lower

numbers in spring and summer - A specialist – used primarily Marri

- Used native forests/woodlands, paddocks and remnant vegetation

- Occurred less frequently and in smaller groups (7.4 ± 0.6, range: 1 – 45)

- Occupancy patterns suggest year-round residency and lack of seasonality in movements

- A specialist – used primarily Marri

Effectiveness of breeding hollow assessment

- Seven probable nest hollows were identified - Targeted behavioural observations at dawn/dusk in

the breeding season was most effective for identifying nest sites.

- Ground-based surveys and post-felling inspections were less effective

- No probable nest hollows were identified - Targeted behavioural observations at dawn/dusk in the

breeding season was most effective for identifying nest sites.

- Ground-based surveys and post-felling inspections were less effective

- Three probable nest hollows were identified - Targeted behavioural observations at dawn/dusk in

the breeding season was most effective for identifying nest sites.

- Ground-based surveys and post-felling inspections were less effective

Use of rehabilitated mine pits

- Floristic and structural variations between pits likely results in food availability differences and therefore variations in feeding activity across pits

- Frequently encountered in mine-site vegetation - Feeding was predominant activity - Fed mostly on proteaceous plants - Evidence of feeding from residue - No differences in feeding observed in interior and

exterior - Rehabilitated pits offer a food resource as early as six

years - No observed differences between feed and non-feed

plots suggesting particular structural or floristic variables associated with higher feeding likelihood could not be identified

- Floristic and structural variations between pits likely results in food availability differences and therefore variations in feeding activity across pits

- Frequently encountered in mine-site vegetation - Feeding was predominant activity - Fed mostly on myrtaceous marri - Evidence of feeding from residue - Marri residues observed in sampling - No differences in feeding observed in interior and

exterior - Rehabilitated pits offer a food resource as early as six

years - No observed differences between feed and non-feed

plots suggesting particular structural or floristic variables associated with higher feeding likelihood could not be identified

- Floristic and structural variations between pits likely results in food availability differences and therefore variations in feeding activity across pits

- Not observed in revegetation, but in adjacent forest areas

- Evidence of feeding from residue - Marri residues observed in sampling - No differences in feeding observed in interior and

exterior - Rehabilitated pits offer a food resource as early as six

years - No observed differences between feed and non-feed

plots suggesting particular structural or floristic variables associated with higher feeding likelihood could not be identified.

Use of artificial nest hollows

- No observations of ANH use - May be attributed to unfamiliarity, ANH features,

placement or presence of sufficient hollows in the

- No observations of ANH use - May be attributed to unfamiliarity, ANH features,

placement or presence of sufficient hollows in the area

- No observations of ANH use - May be attributed to unfamiliarity, ANH features,

placement or presence of sufficient hollows in the

225

(ANHs) on-site area - Future site modifications may encourage appeal and

utilisation of ANHs

- Future site modifications may encourage appeal and utilisation of ANHs

area - Future site modifications may encourage appeal and

utilisation of ANHs

Use of artificial nest hollows across WA

- Recorded to successfully use all three types of ANHs – cockatube, hollowed-out log sections and wooden box-type

- ANH used were in captive and wild settings - Setup used a variety of methods. - Heights varied from less than 2 – 17 m. - Monitoring and maintenance were mostly

opportunistic in the wild, but systematic in captive settings.

- Potential hazards include competition, predation, ANH-associated injuries and safety to birds and workers, and damages to ANHs.

- Issues to consider include ANH durability and placement, maintenance and monitoring, control of predators/competitors, and implementation costs.

- Costs ranged from $400 – 700.

- Recorded to use only cockatube ANH - ANH used were in captive setting - ANH used were in captive and wild settings - Setup used a variety of methods. - Heights varied from less than 2 – 17 m. - Monitoring and maintenance were mostly

opportunistic in the wild, but systematic in captive settings.

- Potential hazards include competition, predation, ANH-associated injuries and safety to birds and workers, and damages to ANHs.

- Issues to consider include ANH durability and placement, maintenance and monitoring, control of predators/competitors, and implementation costs.

- Costs ranged from $400 – 700.

- Recorded to successfully use all three types of ANHs – cockatube, hollowed-out log sections and wooden box-type

- ANH used were in captive and wild settings - Setup used a variety of methods. - Heights varied from less than 2 – 17 m. - Monitoring and maintenance were mostly

opportunistic in the wild, but systematic in captive settings.

- Potential hazards include competition, predation, ANH-associated injuries and safety to birds and workers, and damages to ANHs.

- Issues to consider include ANH durability and placement, maintenance and monitoring, control of predators/competitors, and implementation costs.

- Costs ranged from $400 – 700.

Use of natural and artificial water sources

- Observed using both natural and man-made water sources

- Drinking occurred around sunrise and sunset - Water sources appear able to provide a drinking

resource year round - Observed infrequently in vicinity of RDAs - Observed using established FDPs - FDPs appear to be more appealing than RDAs - FDPs may provide useful measure for minimising

black cockatoo-RDA interactions

- Observed using both natural and man-made water sources

- Drinking occurred around sunrise and sunset - Water sources appear able to provide a drinking

resource year round - Observed infrequently in vicinity of RDAs - Observed using established FDPs - FDPs appear to be more appealing than RDAs - FDPs may provide useful measure for minimising

black cockatoo-RDA interactions

- Observed using both natural and man-made water sources

- Drinking occurred around sunrise and sunset - Water sources appear able to provide a drinking

resource year round - Observed infrequently in vicinity of RDAs - Observed using established FDPs - FDPs appear to be more appealing than RDAs - FDPs may provide useful measure for minimising

black cockatoo-RDA interactions Summary With the careful evaluation and consideration of

mitigation options or conservation measures, and based on detailed research of the specific ecology and natural life history of each of the three black cockatoos at the study area, there is potential to integrate the objectives of conservation of Carnaby’s Cockatoo with that of production at the NBG mining landscape

With the careful evaluation and consideration of mitigation options or conservation measures, and based on detailed research of the specific ecology and natural life history of each of the three black cockatoos at the study area, there is potential to integrate the objectives of conservation of Baudin’s Cockatoo with that of production at the NBG mining landscape

With the careful evaluation and consideration of mitigation options or conservation measures, and based on detailed research of the specific ecology and natural life history of each of the three black cockatoos at the study area, there is potential to integrate the objectives of conservation of Forest Red-tailed Black Cockatoo with that of production at the NBG mining landscape

226

10.2 Original Contributions of the Research

All three black cockatoos vary in their ecologies. These ecological differences have

important implications for assessing the impact of mining, determining how the mining

tenements can be managed to conserve black cockatoo habitat, and examining how

black cockatoo habitat may be ‘restored’ through rehabilitation of mining areas or

establishment of ANHs. Before the study, there was a lack of detailed information about

the use of mining production landscapes by the three threatened southwestern species of

black cockatoos. We did not know how black cockatoo used these landscapes in relation

to a number of ecological attributes:

- The distributions or abundances, group sizes and site occupancy of black

cockatoo across the mining tenement

- Patterns in and types of habitats or food plants utilised within the mining

tenement

- The effectiveness and certainty over hollow assessment for habitat trees in a

mining context

- How and when black cockatoos begin using mine-site rehabilitation for feeding,

and the plants they feed on, as well as any association with revegetation features

(structural or floristic) that may influence the availability of food resources for

black cockatoos

- Whether black cockatoos would use (wood or plastic) ANHs on-site, or the

effectiveness of ANHs in other trials state-wide

- Whether black cockatoos would use artificial and natural water sources or the

tailings dams on-site, or whether water supplementation would minimise the risk

posed by tailings dams.

Now we know that all three black cockatoos utilise the mining tenement at NBG to

varying degrees. All three species were observed at the study site all year round. Both

White-tailed black cockatoo species displayed seasonal differences in flock size and site

occupancy, while FRTBC were more sedentary and occurred in generally smaller

flocks. All three species used a range of habitat types and plant types for roosting and

feeding. Only one known tree (Wandoo) was recorded to be used for breeding by

Carnaby’s Cockatoo and it was identified from a ground-based survey using

behavioural observations and follow up surveys. Both White-tailed black cockatoos

were observed using six to eight year old rehabilitated mine pits in rehabilitation for

feeding, and FRTBC feeding was confirmed using evidence from feed residue

227

sampling. Thus food plants may be restored in less than a decade. No structural or

floristic associations with the revegetation were identified as particularly influencing

black cockatoo feeding of rehabilitated mine-pits.

The state-wide ANH survey confirmed used of ANHs by all three species of black

cockatoos in both natural and captive settings. However in the trial at NBG, no black

cockatoos were observed inspecting or using either wood- or plastic-based ANH on-

site. All three black cockatoos also used both natural and artificial watering sources at

NBG. Water supplementation through the provision of faunal drinking points attracted

black cockatoos. Black cockatoos infrequently came into the vicinity of the tailings and

initial observations suggest that water supplementation may minimise that interaction.

In conclusion, mine-sites may contribute to conserving black cockatoos by restoring

feeding habitat and providing water sources, but also via continuous monitoring

programs and support of research initiatives. Breeding resources in the form of natural

hollows, however, can only be restored in the time scale of centuries, highlighting the

importance of retaining large (hollow-bearing) trees by protecting remnant native

vegetation areas.

10.3 An Integrated Adaptive Management Approach to Conserving Black

Cockatoos at the Newmont Boddington Gold Site and More Broadly in the

Forested Southwest

10.3.1 Adaptive Management

Black cockatoos range widely over many different management tenures and habitats in

southwestern Australia, posing diverse challenges for their management and

conservation (Garnett et al. 2011). Argent (2009, p. 12) listed several key elements of

complex environmental problems that are clearly applicable to black cockatoo

conservation in this region, including:

Multiple uses and multiple objectives;

A mix of scales of interest and boundaries of responsibility;

Divergent needs and desires of stakeholder groups;

Tight economic imperatives around ecosystem exploitation;

Reduced ecosystem health and ecosystem services;

228

Significant technical information on parts of the system, with information gaps on

other parts; and

Competing or open mandates, with different policy options and system targets.

To address these issues, linkages must be developed between research and management.

This can be achieved effectively through adaptive management, which seeks to learn

from management interventions in a planned, systematic way to better inform future

management. Walters and Holling (1990) recognised three different types of adaptive

management: ‘trial and error’, passive adaptive management and active environmental

management. Trial and error involves noting responses to ad hoc management actions,

but the results could be subject to substantial chance. Passive adaptive management

improves on this by using advice from experts to modify management activities and to

evaluate monitoring data continually to refine and improve actions. Active adaptive

management, which is most likely to yield robust results, considers different

management actions across space or time as treatments in an experimental design,

which is a powerful means to increase understanding of the effects of different

management actions on the system (Boyce 1997; Parma et al. 1998; Lee 1999).

That is not to say that developing such experiments is easy. As Krebs (1998, p. 145)

observed: Field experiments are inherently attractive and impossibly messy. They are attractive because they include a ‘control’ which becomes the yardstick by which the experimental treatment can be judged…Field experiments are messy because it is difficult to match sites so that all relevant conditions are equal at the start of the manipulation, and replication is rarely as frequent as anyone would like.

But difficulty is no reason not to try, particularly given the risks to biodiversity at stake.

Krebs (1988; 1999) and Green (1979) outlined the basic principles of experimental

design in field ecology. These principles provide a suitable framework for the design of

active adaptive management programs, and involve:

A deliberative manipulation of the system, called an experimental treatment;

A control, which varies from the treatment in only one variable to allow

comparison of effect; and

Independent replication of either (or ideally both) of control and treatment

groups/areas.

229

Wider stakeholder consultation in choosing questions to investigate and the

interventions to be tested is also relevant. The need for wide consultation is strongly

endorsed in published examples of successful environmental management in marine

(e.g. Hughes et al. 2007), terrestrial (e.g. Hoffmann et al. 2012) and inland aquatic

systems (e.g. Allan et al. 2009). Indeed, critics of adaptive management such as Cundill

et al. (2012) argue that the paradigm’s focus on objectives rather than process limits its

applicability in complex real world systems where social interactions are vital.

While conceding these points, the situation of a mine-site such as NBG or, indeed, the

wider tenure of State Forest in southwestern Australia, facilitates planning for adaptive

management, in contrast to management occurring over large landscapes and involving

multiple landholders. Although there are multiple mining companies operating in the

southwest (Dell et al. 1989; Ward et al. 1990), their common interest in combining

mining and conservation should enhance co-operation among stakeholders and

managers. I have therefore chosen the adaptive management approach as an appropriate

framework for black cockatoo conservation within the production landscapes of the

Jarrah-Marri forest.

Choosing areas where adaptive management can be applied should not be guided by

empirical findings alone (e.g. the results of this thesis), but also by the aims and actions

identified in relevant forest management, species recovery and action plans (Cale 2003;

Conservation Commission of Western Australia 2003; Chapman 2008; Garnett et al.

2011), which are summarised by topic in Table 10.2. I have grouped these into topics

related to: (1) management of orchards, (2) captive breeding, (3) general population

ecology and baseline data collection, (4) issues related to nesting hollows, (5)

management issues specific to mine-sites, and (6) management issues relevant to

forested areas of the southwest, in particular State Forest. I will not consider points (1)

and (2) despite their importance, because they are not of immediate concern to mine-site

management or State Forest. Instead, the issues developed below relate to points (3) –

(6).

230

Table 10.2 Specific recommendations for black cockatoo conservation from Cale (2003) and Chapman (2008).

Species Class of action Action recommended

Baudin’s Cockatoo

Administrative and finance

Seek the funding required to implement future recovery actions

Management of orchards

Determine and promote non-lethal means of mitigating fruit damage by Baudin’s Cockatoo in orchards Eliminate illegal shooting Develop and implement strategies to allow for the use of noise emitting devices in orchards

Nesting hollows

Determine and implement ways to remove Honeybees from nesting hollows Identify factors affecting the number of breeding attempts and breeding success and manage nest hollows to increase recruitment

Mining Determine and implement ways to minimise the effects of mining on habitat loss

Forest Management Determine and implement ways to manage forests for the conservation of Forest black cockatoos

Further Research Endeavour

Identify and manage important groups of each species and protect from threatening processes Map feeding and breeding habitat critical to survival and important populations, and prepare management guidelines for these habitats Determine population numbers and distribution Determine the patterns and significance of movement Maintain the Cockatoo Care program and use other opportunities to promote the recovery of Forest Black Cockatoos

FRTBC Same as for Baudin’s Cockatoo Same as for Baudin’s Cockatoo

231

Table 10.2 (continued).

Carnaby’s Cockatoo

Habitat management within priority areas (i.e. identify priority areas, management of breeding habitat within priority areas and management of feeding habitat within priority areas)

Determine known breeding sites Assess the size and health of specific breeding populations Assess the relative proportions of breeding and feeding habitat for specific breeding populations Select priority areas and formulate specific plans for each area Monitor and maintain nesting hollows, fire prevention Minimise illegal poaching throughout breeding distribution Fence woodland remnants Control nest competitors Design, construct and erect nest boxes and logs

Management of feeding habitat in non-breeding areas

Develop and publish planting guidelines in consultation with stakeholders Develop management guidelines for woodland regeneration Manage existing heath/shrubland reserves Prepare and implement revegetation strategies for each priority area, taking into consideration the range of planning and operational requirements needed for a successful revegetation program Manage heath/shrubland in land managed by the Department Develop guidelines for management of feeding habitat Surveys of Carnaby's Black-Cockatoos’ use of Gnangara Park Develop a log book, and/or use current District Fauna Files in a Maritime Pine Program Develop monitoring procedures for landholders, print log

Monitoring of population Monitoring of nestling health

Community involvement

Design and production of a 'Carnaby’s Cockatoo Recovery Kit'

Captive breeding programs Maintain captive breeding program

Species Class of action Action recommended

232

10.3.2 Adaptive Management and the Newmont Boddington Gold Site

The data in this thesis fall mainly within the passive adaptive management paradigm,

with some elements of trial and error. There was careful monitoring of the movements,

numbers and general behaviour of black cockatoos on-site to infer relative abundances,

habitat use, diet, and seasonality in accordance with passive adaptive management.

Similarly, the work on ANHs on site involved systematic monitoring of two types of

ANHs for use by black cockatoos. The survey of ANH users across the southwest was a

systematic documentation of trial and error approaches by different groups. The detailed

data collected on use of rehabilitation pits was also passive adaptive management,

monitoring and describing feeding by black cockatoos within the pits rather than testing

the birds’ response to differing rehabilitation protocols.

Lastly, the data on use of watering points also centered on monitoring of existing

watering points and black cockatoo activity at the RDAs, without the deliberate

experimental provision of control and treatment groups to test particular hypotheses

about what attracts or repels birds at different types of watering point. While much

useful data have been generated in this way, there are now opportunities to design active

adaptive management investigations that employ experimental protocols to explicitly

test hypotheses about black cockatoo management on-site. These can be combined

effectively with some ongoing passive adaptive management strategies.

10.3.3 The Newmont Boddington Gold Site

Active Adaptive Management Initiatives:

- Artificial Nest Hollow Work

Recovery and action plans for the black cockatoos raise concerns for identifying,

enhancing and protecting breeding habitat (Cale 2003; Chapman 2008; Garnett et al.

2011), and ANH work fits within these broad concerns. The work completed at NBG, in

association with the experience of groups using ANHs for black cockatoos in

southwestern Australia (also reported in Groom 2010), suggests that the design of the

ANH is less important than their positioning in the environment. This is in keeping with

the results of numerous other trials of ANHs for a range of arboreal fauna in Australia

(Menkhorst 1984; Harper et al. 2005; Beyer and Goldingay 2006; Goldingay and

Stevens 2009) and overseas (Beissinger et al. 1998; Fernández-Juricic and Jokimäki

233

2001; Bolton et al. 2004; Robaldo-Guedes 2004). The trial at NBG used a design that

was successful elsewhere and the ANHs were positioned close to used natural hollows,

water and food, which are requirements for success (Saunders 1980; Groom 2010).

However, the ANHs were not used, and it is hypothesised that this was because they

were positioned too low on the trees in relation to the canopy at the site and the position

of natural hollows at the site.

To test this hypothesis, I propose the following design. A minimum of two sites should

be identified, each with similar access to good feeding resources and to permanent

water, and with a record of successful breeding in the near locality (positive records

within a 1 km radius). At each site, a control group of 5 – 10 ANHs would be positioned

between 5 - 10 m high in suitable trees. An experimental group of 5 – 10 ANHs would

be positioned at heights between 12 - 17 m, which is just below the average canopy

height of approximately 20 m on-site. Placing ANHs at this height would require

specialist equipment (e.g. cherry picker) and some occupational health and safety

negotiations and guidance would be required. The individual ANHs in each group

would be interspersed across the landscape, to avoid any bias that might arise if they

were in two discrete clumps. All ANHs would be of the same design, either cockatubes

or wooden box-type design. The hypothesis would be supported if the higher ANHs

were used by black cockatoos at a greater frequency than the lower ones at both sites.

- Water Work

Cameron (2007) notes that cockatoos in general need to drink daily. Therefore good

habitat provides food and water in close proximity to each other (Saunders 1980).

Failure to find adequate water during abnormal heat conditions was a likely cause of a

mass death event of a flock of Carnaby’s Cockatoos near Hopetoun, on the south coast

of Western Australia (Saunders et al. 2011). Lack of nearby water may also render nest

hollows unsuitable (Saunders 1977).

Black cockatoos were observed drinking at a number of man-made watering sites at

NBG, particularly at the water sumps and FDPs. Black cockatoos were not observed

drinking at the RDAs. It is hypothesised that the ‘open’ environment of the RDAs may

lead birds to perceive a greater predation risk at such sites and therefore deter black

cockatoos from using them (Cameron and Cunningham 2006). Since the sumps and

234

FDPs are located near the forest-line; this produces a more ‘covered’ environment may

provide a sense of security to drinking birds.

The following experimental design on watering location/placement could be used to test

this hypothesis: a control group of 5 – 10 watering points could be established at

various locations (approximately 250 m apart), but in close proximity to a forest-line

(<25 m). With the experimental group comprising of 5 – 10 watering points

(approximately 250 m apart) established in open, cleared areas (>2.5 km from the

forest-line). The design of watering points at both control and experimental sites should

be accessible by black cockatoos and ought to include features such as firm and gently

sloping water edges for ease of access, and be free of obstructions that may impede

black cockatoos from drinking at the water edges safely. The hypothesis would be

supported if watering points near the forest-line were used more frequently than

watering points in the open.

10.3.4 The Wider Southwest

Passive Adaptive Management Initiatives:

Passive adaptive management may be more appropriate in the wider southwest, because

of the need to co-ordinate action across multiple land tenures and landowners (i.e.

multiple rehabilitation sites that vary in age across different mining tenements).

- Use of Different-Aged Rehabilitation Pits

In the Jarrah-Marri forest, FRTBC feed primarily on myrtaceous plants, especially

Marri (Abbott 1998; Johnstone and Kirkby 1999). Marri and Jarrah provide a high-

energy return for foraging effort in retrieving the seeds relative to some other food

plants (Cooper et al. 2002). Baudin’s Cockatoo also feed extensively on Marri (Higgins

1999; Johnstone and Kirkby 2008), possibly for the same energetic reasons as FRTBC.

They also feed on a range of proteaceous plants, especially the genera Hakea and

Banksia (Johnstone and Storr 1998; Johnstone and Kirkby 2008; Johnstone and Kirkby

2009) and are noted for searching trunks and branches for wood-boring grubs (Saunders

1974). Carnaby’s Cockatoo feeds even more extensively on proteaceous species, and

also feed on Marri and Jarrah (Saunders et al. 1985, Johnstone and Storr 1998). Where

pine plantations are available, they also feed on the pinecones (Saunders 1974).

235

Given this background, rehabilitation plots that include proteaceous and myrtaceous

species should provide feeding resources quite quickly because many proteaceous and

myrtaceous plants will flower and produce fruit or seed within the first six years

(Abbott 1985; Abbott and Loneragan 1986; Mawson 1995; Lamont and Groom 1998;

Lamont et al. 1998; Lee et al. 2010). However, the slow growing, long-lived eucalypt

species dominate the overstorey in the forested ecosystems of southwestern Australia,

although proteaceous shrubs may form an overstorey component in some communities

(Burrows and Wardell-Johnson 2003). Many proteaceous plants tend to be faster

growing but have short or moderate lifespans (Auld and Myerscough 1986; Hansen et

al. 1992), and become less abundant as an overstorey of myrtaceous trees develops.

Therefore the main food resources in rehabilitation plots are predicted to shift from

proteaceous to myrtaceous species as the vegetation ages. In contrast to food resources,

which may become available within a decade, breeding cockatoos requires old, mature

or senescent trees that have large hollows, features which may not return until a time

scale of a century or more (Abbott and Whitford 2002).

Rehabilitation sites at NBG demonstrated feeding by all three black cockatoos as early

as eight years post-rehabilitation (Lee et al. 2010). However, it is not clear just how

early revegetation might able to start providing a food source or, at the other end, what

food plants will be available two or more decades after revegetation is established.

Although it is preferable to document differences between black cockatoo feeding in

selected rehabilitation pits over time (e.g. over a timespan of decades), it is logistically

and financially impractical to do so. It is also an unfeasibly long time to wait for

feedback on the effectiveness of restoration methods. However, one way to overcome

this would be via the use of chronosequences or the space-for-time substitution study

approach (Walker et al. 2010), although care must be taken in the statistical analyses of

results (Tyre et al. 2000).

Space-for-time studies are examples of passive adaptive management (e.g. Shearer et al.

2009). In this case, they would involve selecting a series of rehabilitation sites that

differ in age and sampling the different-aged sites simultaneously. It is assumed that

only one factor differs across sites, and that is their age. All sites are assumed to share

similar starting points and experience similar conditions and disturbances, although the

precise assumptions should be stated and verified as much as possible at the start of the

study (Johnson and Miyanishi 2008). By comparing features between sites, any

236

differences could be used to approximate change at a single site over time, thus

allowing inferences as to how the vegetation communities in rehabilitation areas change

as they age (Burrows and Wardell-Johnson 2003; Orr and Stanley 2006). This gives

information far more quickly than long-term monitoring at one or more sites over a

decade (Burrows and Wardell-Johnson 2003). For example, rehabilitation pits could be

divided into age groups: >5 years, 5 - 10 years, 10 - 15 years, 15 - 20 years and >20

years post-establishment. Surveys on black cockatoo feeding and feed residue sampling

could be performed at these different-aged sites to observe if feeding activity differed

by age of rehabilitation, which can be associated with corresponding changes in

vegetation structure, composition and complexity with time.

Obtaining a suitable range of pit ages will require cooperation between several different

mining operations. However, this will address the specific concern identified in

Chapman (2008) about the impact of mining on black cockatoo habitat and recognise

the importance of identifying and protecting food resources for all three black cockatoos

(Cale 2003; Chapman 2008; Garnett et al. 2011). Pine plantations of different ages

could be investigated as well, in keeping with a concern for the value of non-native

vegetation for feeding habitat (Cale 2003).

- Fire Studies

Fire management is an important practice for many different production areas in

southwestern Australia (McCaw et al. 2003; Pyne 2003; Morley et al. 2004), as a result

of a fire-prone landscape where annually conditions are favourable for bushfires for up

to eight months (McCaw and Hanstrum 2003). Fuel reduction burns are aimed at

minimising the risk of wildfires that can cause considerable damage to natural forest

habitats and production landscapes (Hopper 2003). Frequent but low-intensity

prescribed burning in autumn and spring that creates a ‘patchy’ environment composed

of areas burnt at different frequencies and in different seasons should reduce fuel loads

and minimise risk of undesirable bushfires (Abbott 2003). Nonetheless, attention must

also be given to the needs of the biota for fire regimes of varying lengths (Burrows and

Abbott 2003; Burrows and Wardell-Johnson 2003). While fire management is not listed

explicitly as a management action in the recovery plans for the three black cockatoos, it

is implicit in managing feeding and breeding habitat because of fire’s effects on

vegetation succession and the longevity of hollow-bearing trees (Whelan et al. 2009,

Parnaby et al. 2010).

237

Fire may influence black cockatoos through changes in food plants and availability of

nest hollows (Whelan et al. 2009). Fire is a key factor in altering the structure of plant

communities (Bond and van Wilgen 1996), and stimulates many plant species to

reproduce by enhancing flowering, as well as seed production, dispersal and

germination (Burrows and Wardell-Johnson 2003). Fire-adapted plants persist and, in

some cases depend, on a particular range of fire regimes (Gill 1981; Lamont et al. 2000;

Bowen and Pate 2004). Old stags with hollows suitable for cockatoo breeding may be

particularly susceptible to fire damage and tree fall. For example, Parnaby et al. (2010)

reported in an opportunistic, post-fire assessment of the proportion of burnt, hollow-

bearing trees that collapsed following low-intensity prescription burns in the mixed-

eucalypt Pilliga forest in New South Wales that for a total of 329 burnt hollow-bearing

trees, collapsed hollow-bearing trees were predominantly older with 40% of senescent

trees and 44% of live stags collapsing in their plots. It is reasonable to suppose that

similar problems could occur in southwestern Australia, where old trees are critical for

cockatoo breeding (Saunders et al. 1982).

Fire management is an operational objective at NBG. For black cockatoos, the main

features likely to be affected by bushfires are nest trees (especially dead hollow-bearing

stags) and feeding vegetation in rehabilitation or forest sites. However, the effects of

fuel reduction burning on black cockatoo habitat are poorly understood. Prescribed

burning of rehabilitation may favour the retention and growth of desirable plant species.

A passive adaptive management study could be performed to study the shorter-term

effects of burning of vegetation in either rehabilitation pits or remnant forest areas on

black cockatoo activity. By taking advantage of natural ignitions and fuel reduction

burns, monitoring for black cockatoo activity could be undertaken to observe when

birds return to a burnt site. Alternatively, to study longer-term effects of burning, space-

for-time studies could be performed on selected sites that differ by time (age) post-

burning to monitor usage by black cockatoos.

10.4 Implications for Psittacine Conservation Globally

10.4.1 Global Threats to Psittacines

The Order Psittaciformes comprises 375 species worldwide across both tropical and

temperate regions. Nineteen species are ‘extinct’, 16 ‘critically endangered’, 34

238

‘endangered’, 54 ‘vulnerable’ and 48 ‘near threatened’ (Collar and Juniper 1992; IUCN

2013). They are amongst the most highly threatened group of birds globally (Wright et

al. 2001). Various factors that work alone or in combination have been implicated in the

highly threatened status as well as global decline of psittacine populations, and these

vary geographically, temporally and with specific characteristics of the species involved

(Snyder et al. 2000). However, two are of primary significance: progressive and

widespread habitat loss through destruction and fragmentation (i.e. for agriculture,

logging, mining, urbanisation and tourism) and the capture or trapping (including

poaching) for the bird trade (Collar and Juniper 1992; Collar 2000; Snyder et al. 2000;

Marini et al. 2010; White et al. 2012; Clarke and de By 2013). Psittacines' colourful

plumage and sociable natures make them popular avian companions (Collar 2000;

González 2003; White et al. 2012). Of the 95 species considered in the Action Plan for

Parrots, 78 species are affected by habitat loss, while the avian trade affects 36 species.

However, 29 species are affected by both forms of pressures concurrently.

For a more comprehensive list of threats contributing to the decline of parrot

populations worldwide refer to Table 10.3.

239

Table 10.3 Summarised list of global causes of decline and examples.

Introduced predators The Kaka Nestor meridionalis in New Zealand is predated upon by

introduced Brush-tail Possums Trichosurus vulpecula and feral cats Felis

catus (Collar and Juniper 1992; Wilson et al. 1998; Snyder et al. 2000;

Wright et al. 2001)

Competition for resources

with introduced or native

species

Competition for nesting sites with woodpeckers, bats and toucans and the

Blue-throated Macaw Ara glaucogularis (Collar and Juniper 1992; Wilson

et al. 1998; Collar 2000; Snyder et al. 2000; Wright et al. 2001)

Human overexploitation

or persecution in the form

of hunting for feathers and

food

The Palm Cockatoo Probosciger aterrimus for food, and Pesquet’s Parrot

Psittrichas fulgigus for its red and black plumage in New Guinea (Pacific

Islands) for ceremonial use by indigenous people. This can become a more

serious threat when taken outside the normal cultural context and to supply

tourist markets (e.g. Hyacinth Macaw Andorhynchus hyacinthinus plumes

in the Araguaia catchment (Collar and Juniper 1992; Collar 2000; Snyder et

al. 2000; Wright et al. 2001)

Disturbance from natural

causes or disasters

The Thick-billed Parrot Rhynchopsitta pachyrhyncha by wildfires, and the

Puerto Rican Amazon Amazona vittata, Black-billed Amazon Amazona

agilis and Cuban Amazon Amazona leucocephala by hurricanes (Collar and

Juniper 1992; Collar 2000; Wright et al. 2001; U.S. Fish and Wildlife

Service 2013)

Culling or shooting to

protect crops or as a

pastime

The Philippine Cockatoo Cacatua haematuropygia is shot to protect

maize/rice fields, and the Blue-winged Racquet-tail Prioniturus verticalis

are common target practice for locals (Snyder et al. 2000)

Use of chemicals Pesticides in the case of the Red-breasted Parakeet Psittacula alexandri

(Snyder et al. 2000)

Diseases or parasitism Introduced diseases such as viscertropic velogenic Newcastle disease from

released infected captive birds may affect wild members of the Philippine

Cockatoo. Parasites include bot or soldier flies (Snyder et al. 2000; Wright

et al. 2001)

Tourism and associated

activities

The Yellow-shouldered Amazon Amazona barbadensis on Margarita

(Snyder et al. 2000)

These threats are compounded by certain natural or life history traits/parameters or

strategies of the species (Collar 2000), refer to Table 10.4.

240

Table 10.4 List of natural and life history traits/parameters or strategies of psittacines that make them susceptible to threats.

a) Having a naturally restricted range such as those confined to island habitats that make them ‘single-

country endemics’ (Saint Lucia Amazon Amazona versicolor and Saint Vincent Amazon A.

guildingii) (Collar and Juniper 1992)

b) The specialisation of certain psittacine species to particular food types (e.g. pine-cone specialists:

Maroon-fronted Parrot Rhynchopsitta terrisi, palm specialists: Lear’s Macaw Anodorhynchus leari

and the Yellow-eared Parrot Ognorhynchus icterotis, grass-seed specialist: the Golden-shouldered

Parrot Psephotus chrysopterygius, and the Kaka Nestor meridionalis that depends on honeydew

from a scale insect) makes them vulnerable as their food plants are cleared as a result of

deforestation (Collar 2000). Or when their feeding habits make them competitors for agricultural

crops (e.g. Black-cheeked Lovebird Agapornis nigrigenis) (Snyder et al. 2000).

c) The reproductive biology of psittacines is generally characterised by low reproductive rates (i.e.

small clutch size, single clutch per year), low recruitment (i.e. low chick or fledgling survival rates),

delayed sexual maturity, longevity and aging populations (i.e. large proportions of non-breeding

adults) and restrictive nesting requirements (Saunders 1986; Snyder et al. 1987; Emison et al. 1994;

Wright et al. 2001), which may render parrot populations less able to recover from reductions in

population size caused by anthropogenic disturbances (Wright et al. 2001). This is exacerbated by

the vulnerability of particularly the larger parrot species to the loss of nesting sites (usually larger

trees). Since most psittacines are cavity-nesters and land clearing may lead to a bottleneck in nesting

site availability that may last several centuries (e.g. Spix’s Macaw Cyanopsitta spixii tied to Caraiba

Tabebuia caraiba woodland for nesting) (Collar 2000).

10.4.2 What is being done about the threats?

The conservation of psittacines is challenging because parrots exhibit ranging

behaviours that extend beyond the limits of actual or feasible protected areas (Collar

2000), which are often quite extensive (Collar and Juniper 1992). Additionally, the long

history of keep parrots in captivity gives them a strong attraction as pets and status

symbols. This resulted in the public’s general inability to distinguish endangered from

non-endangered species, and also makes private ownership of even endangered parrots

socially acceptable (Snyder et al. 2000).

The overarching goal of psittacine conservation should be the maintenance (or increase)

of stable viable wild populations and subpopulations of all species within their native

ranges and natural ecosystems and across long time-scales (Collar 2000; Snyder et al.

2000). This requires genetic and demographic considerations such as variations among

species in range, natural population fluctuations, life history parameters and sensitivity

241

to environmental threats. This may involve the reduction of fundamental causes of

endangerment in the wild, preservation of essential habitat and determining a particular

minimum population size as a goal for each species (Collar and Juniper 1992; Collar

2000; Snyder et al. 2000).

Conservation solutions have to be locally tailored and resource allocation has to

appropriately distributed across the landmass (e.g. target conservation resources to

richest parrot areas – endemic island species in Indonesia). Programs that benefit

sympatric species (e.g. protection areas) are more productive than those targeting only

one (e.g. captive breeding) (Collar 2000; Snyder et al. 2000). Most parrot conservation

programmes will utilise more than one conservation technique, and the optimal

combination for a species will be determined on a case-by-case basis. Conservation

measures should consider both advantages and disadvantages as well as be socio-

economically effective and politically viable (Collar and Juniper 1992; Collar 2000;

Snyder et al. 2000). In the long-term, there is a need to widen the scope from single-

species, crisis-driven intervention to broad-based, sustained efforts to create conditions

for psittacine assemblages to flourish in increasingly human-dominated landscapes

(Collar 2000; Snyder et al. 2000) - the problems confronting black cockatoos in

southwestern Australia. Conservation solutions have been mostly summarised from the

Parrots - Status Survey and Conservation Action Plan 2000-2004 edited by Snyder et al.

(2000), refer to Table 10.5.

242

Table 10.5 Summary of the list of conservation measures for psittacines globally.

Status assessment and

protection

For long-lived, low-fecundity monogamous psittacine species that display strong site-fidelity, specifying demographic parameters including population

trends, reproductive success and mortality is fundamental to identify causes of population decline. These along with identifying factors such as the

frequency, severity and effects of disturbances are essential to determine the vulnerability of the species (e.g. poaching for Lear’s Macaw

Anodorhynchus leari). The basic demographic information can be provided through population viability analyses, nest-monitoring efforts and artificial

marking techniques (e.g. banding, palatial tagging or radio-telemetry). This means that the measures of population recovery are tailored to the species

and severity of the crisis. Though various methods exist for monitoring parrot populations, the utility of these techniques differ as a result of inter-

species variation in behaviour and ecology. For example, roost counts work for the Bahama Parrot Amazona leucocephala bahamensis, while nest

enumeration is better for the Maroon-fronted Parrot Rhynchopsitta terrisi).

Increasing current

knowledge base

The limited or lack of sound information and synthesis of the natural history, status and pressures threatening psittacids is often cited as a major obstacle

to conservation efforts. Much of the material often comprises of anecdotal reports and the need for more data is an overriding issue (e.g. Grey-headed

Lovebird Agapornis cana in Madagascar). This requires establishing a network of people dedicated to conservation (e.g. recovery teams) and to collect

detailed information on the species (e.g. population trends, food resource utilisation, nest-site availability, direct/indirect human influences, local

population numbers and densities). This will (i) confirm the species’ conservation status with confidence, (ii) determine the best practice principles,

mitigation options or goals that is compatible with conservation as well as the political, economical and social environment, (iii) help inform

management decisions to propose realistic and cost-effective conservation measures (e.g. increasing and improving natural hollows or implementing

artificial hollows – Red-fronted Parakeet Cyanoramphus novaezelandiae), and (iv) ultimately proper and long-term species-specific management and

legal interventions, as well as measure progress through conservation programmes (e.g. reserve design, function and establishment, control of limiting

factors, habitat enhancement and promotion of local awareness programs) (Collar 2000).

243

Habitat preservation and

restoration

Habitat change, destruction and fragmentation are a major threatening process for psittacids globally. Therefore creating, extending or modifying

protected areas for endangered psittacine species such as nature parks (e.g. Echo Parakeet Psittacula eques echo in Mauritius) or game reserves (e.g.

Palawan in Philippines) where it is illegal to capture wild birds may aid in parrot conservation. This may be done by recreating critical habitat features

in the wild such as expanding food and water supplies, as well as nest and roost sites. The successful design of a habitat protection plan requires

information on breeding and wintering habitat as well as habitat used along the migratory route in migratory species in their annual cycles (e.g. Great

Green Macaw Ara ambigua). Site selection targeting geographically related suites to create more refined and well-defined protected area priorities for

wider ranging, but especially restricted range species can use these birds as valuable flagship species for these initiatives. Another way to achieve this is

through - (i) increasing the stabilisation and equitability of land tenure to reduce agricultural exploitation, (ii) increasing inter-reserve connectivity, (iii)

outright purchasing of conservation easements (e.g. biosphere reserves in the Andes of Colombia) or protecting priority areas (e.g. endemic bird areas),

(iv) eliminate subsidies for land clearing and (v) provide incentives for wildland maintenance. This must be equitably, effectively and economically

supported by the appropriate management strategies, especially of publicly owned land. Its also needs to be enforced and properly integrated in

conjunction with other conservation actions in managing other stress factors (e.g. hunting, trade) or development objectives on private lands (e.g.

multiple-use areas, buffer zones) that will link biodiversity conservation with social and economic betterment of local communities through best

practices in land use.

Laws, law enforcement

and education

Despite the inclusion of many psittacid species to the Appendices of CITES (Convention on International Trade in Endangered Species) that aided in

reducing the national and international exportation of parrots worldwide through species-specific protection and bans from trade (Collar and Juniper

1992). It has also made a number of the CITES Appendix I species highly prized and notable targets (e.g. blue macaws), because they combine beauty

with rarity and prestige value for collectors. Illegal capture and trade (especially domestic), even while trading under quotas, is likely to decline with the

introduction of locally-based economically viable alternatives that also lessens the threats to the birds and habitat (e.g. financial rewards to those who

report and protect nest holes of the Blue-naped Parrot Tanygnathus lucionensis in the Philippines) or through manned controls at major transport

terminals to enforce CITES legislation for wild parrots (e.g. Australia’s Quarantine and Inspection Service AQIS). This control of trade via total ban

was successful in Australia given its remote island nature. Changing the attitudes of people may require comprehensive law enforcement, changing the

regulations or laws or willingness of people to obey the laws. Penalties and wide publicity given to their transgressions for those breaking the law need

244

to be deterrent and also meaningful and applied in a just manner. It must be coupled with the understanding of the need of such legalities. This

especially so for dealing with illegal pet-bird or plume-bird trade and hunting by prohibiting such activities. While easy to pass, enforcing them is often

difficult, as there are profits to parrot smuggling and also because private ownership remains socially acceptable at the local level. Along with a

perceived overabundance of birds, another challenging area is crop depredation by parrots. While control efforts have been instituted at local levels

against pest parrot species, much of it is based on exaggerated perceptions of the damage, and rare parrot species are often a target of these efforts.

Finding solutions that permit both parrot survival and satisfactory minimisation of depredations is often hard. For example, crop substitutions (e.g.

seedless instead of seeded fruit plants) are adequate but not economically attractive. IUCN/UNEP (United Nations Environment Programme)/WWF

(World Wide Fund for Nature) had urged countries to legislate against ownership of internationally threatened species except under tightly controlled

conditions. CITES is another example of a system implemented to curtail and regulate import and export of birds, and all parrots are listed there.

In terms of education, psittacines make excellent emblematic species for use in campaigns through concerted education efforts (e.g. avicultural

societies) to raise public awareness of conservation issues and reduce social acceptability of private ownership. This is because they inspire human

empathy, as well as local or national pride (e.g. Saint Lucia Amazon Amazona versicolor). This makes local involvement and awareness campaigns a

requirement of any parrot conservation strategy when appropriately orientated. Educational programs can be effective and economical, but must be

locally implemented and recognise cultural diversity and that the values and perceptions may differ from Western conservationist thinking. They should

aim to target all age and socio-economic groups through a diverse range of appropriate activities (e.g. curriculum in schools or official meetings for

resource-users and leaders in the public sector). Politically viable solutions are more productive in gaining widespread public support, but have to be

carefully planned in the social context to gain significant local participation in the conservation programmes. Programmes should maximise financial

and technical resources, promote positive attitudes and cooperation, but should also be replicable and locally sustainable through training and the

provision of new problem-solving skills/techniques and be based on local knowledge and existing philosophies. Awareness of the cruelty involved in

the capture of parrots for the trade may eventually lead to a steady decline in consumption of wild-caught parrots in the developed world. This may be

coupled with an increase of the protection of wild birds in the developing world through nurturing local sympathy and understanding, and creating a

long-lived environmental consciousness.

Ecotourism Parrots may be good potentials to serve as foci for ecotourism given their spectacular natures. For example, a parrot-centered ecotourism is the viewing

245

of spectacles of large macaws and other parrot species at clay licks in southeastern Peru. Tourism operations that increase parrot conservation and add

local economic value to the birds that also benefits the local communities by becoming major income-producing revenue and reducing the financial

value brought in by the trading of these birds in many rural economies and becoming and incentive, decreasing the economic counter-incentive to retain

forest (Collar 2000). By giving economic benefit to local communities (e.g. tour guide or park fees), it may also cause them to value and preserve the

local ecosystems, and also raise funds for their conservation. However, ecotourism as a conservation option needs to be implemented properly and

minimise degradation of the very resources on which it depends, and the disruption of local cultures, but also exploitation and corruption. Another

consideration is that it is vulnerable to unpredictable fluctuations in international economies, currency rates and the ever-changing perceptions of the

risks of visiting such areas (e.g. illicit drug cultivation and associated issues on safety).

Captive breeding Experiments with captive breeding programs are in their infancy but in-situ and ex-situ conservation initiatives may be improved through internationally

coordinated programmes (e.g. WWF, BirdLife), and endorsed by local governing systems with regard to biological and economical realities. It has a

crucial function in the recovery of critically endangered parrots, especially if integrated into a species recovery program where it can act as a safety net

and serve as a boost for some severely threatened populations of steeply declining species that require intensive intervention. It can greatly increase

reproduction rate for species that breed readily in captivity through multiple clutching. For example, it can play an important role in breeding birds

where most if not all of the individuals of the species are already in captivity (e.g. Spix’s macaw Cyanopsitta spixii) but is a case of captive breeding is a

forced circumstance upon them as oppose to them needing it. Release of captive-bred birds can speed recovery of wild populations by increasing

population extant, correcting sex-ratio imbalances, minimise losses of genetic diversity and chances of catastrophic population declines, as well as

reestablish extirpated populations or establish new populations in altered habitats (e.g. Echo Parakeet Psittacula eques echo). However, is advisable as

only a short-term measure when other preferred conservation options are not immediately available or if there is no way to sustain wild populations, and

should be implemented only after comprehensive evaluation. Other values it may have include providing birds for exhibit, conservation education; fund-

raising purposes and also provides a source of fundamental biological research unachievable with wild individuals (e.g. especially in zoos where their

wealth of husbandry and veterinary experience needs to be explored further). Commercial captive breeding may be important in supplying the market in

cage birds as captive bred birds are reputed to be more docile and manageable as pets and have suffered no trauma equivalent to that experienced by

wild caught birds in reaching domesticity. They may be more expensive due to the high cost of captive breeding but are highly desirable alternative to

246

drawing off wild populations. This may have the potential to reduce trade profits (e.g. Naretha blue bonnets Psephotus haematogaster narethae). There

are a number of problems associated with it that may accumulate project costs such as (i) in husbandry or domestication where there are difficulties in

breeding certain species, (ii) difficulties in reintroducing species to the wild (e.g. hand-reared birds versus parent-reared birds), (iii) high costs in

personnel and facilities (i.e. input-intensive), (iv) disease risks (i.e. >30 diseased conditions are widespread in mixed captive collections and are hard to

detect in carrier birds and exotic, novel diseases may pose a threat to wild birds), (v) managing genetic and behavioural changes (i.e. need to preserve

extant genetic variation by minimising genetic inbreeding through preparation of pedigree systems and proper species-by-species behavioural

management in captive environments, (vi) problems in ensuring continuity of the programmes (e.g. changes in personnel, institutions and financial

resources can lead to unsustainability and instability of programmes that need several decades for species recovery).

Reintroductions This is associated with increasing the overall security and sustainability of the species through reestablishing or boosting viable wild populations with

release of captive birds (e.g. wild-caught or captive-reared stocks) Snyder et al. 1996 White et al. 2012. It may also help reinvigorate the genetic

diversity, correct imbalanced sex ratios and establish wild populations in new habitat thus extending the range (e.g. Kakapo Strigops habroptilus).

While reintroductions may be beneficial as a conservation goal, there are difficulties in re-establishing wild populations from captivity. They include –

behavioural deficiencies in captive-bred versus wild-raised stocks. This is a result of the modification, lost or lack of appropriate teaching or learning of

social matters and survival skills through social interaction and cultural transmission from parents, siblings or flock members (e.g. foraging, essence of

tool-use, tracking ripening food, roosting assemblages, recognising predators). Wild-caught birds should be the preferred method as they fared better

than the more naïve captive-reared birds in this respect (e.g. Thick-billed Parrots Rhynchopsitta pachyrhyncha). This may be avoided by the fostering of

captive-bred eggs/nestlings into wild nests. Unintended ecological effects are also a concern especially with reintroductions into non-native regions.

Another consideration is the extent of predator pressure faced by released birds, and reintroductions may be maximised in low-predator environments

hosting existing wild populations. This requires thorough evaluation of the potential release site. A major issue faced in reintroductions is exotic disease

introduction and transmission from captive to wild birds (especially high in psittacines), so any release should be made with screened or quarantined,

disease-free individuals or healthy wild-caught birds that are relatively safer from a disease standpoint. Finally, cultural or genetic pollution of wild

populations of non-adaptive traits evolved in captivity from captive-bred individuals through learning and interbreeding with their wild counterparts.

This may be overcome by the implementation of adequate pre-release acclimatisation and training. Overall, reintroductions should not be considered a

247

conservation option without careful evaluations (e.g. sufficient numbers to give reasonable hope of success), and post-release results should be

comprehensively monitored (e.g. banding/microchipping/radiotagging released birds), followed by conducting quantitatively adequate follow up

releases.

Confiscated birds from the

parrot trade

There are a few options post-confiscation of parrots. For example, the preferred way adopted is that birds are quarantined and rehabilitated for release

back into the wild (e.g. Yellow-naped Amazon Parrot Amazona ochrocephala auropalliata in Costa Rica) (P Martinez, Murdoch University, 2013, pers.

comm.). Birds may be donated or sold to zoos or research institutions, auctioned for purchase or even euthanised. Issues with donations or sales to zoos

or similar institutions may be that these bodies lack the space and finance to absorb the large numbers of birds, and most of the birds accepted are those

with exhibit potential. Euthanasia is simple and no-risk method of ‘disposing’ confiscated birds, especially with disease-compromised birds, however

this may be oppose by opposition groups on principle.

Sustainable harvest The ability of parrots to imitate human speech and adaptability to captive conditions lends to their popularity and long historical record of being kept as

pets. With increasing human population, wild parrot populations have been unsustainably overharvested with many to the point of extinction. Parrots are

particularly sensitive to overharvesting given their low reproductive potentials and long life spans. Prices for the more desirable species often serve as a

significant source of revenue for rural communities - in some cases the supplementary or even sole source of income. When conservatively and properly

controlled and implemented and through the restructuring of the socio-economic environment, economic value of parrots and also their habitat may be

used to promote a truly sustainable utilisation of wild populations that will provide advantages for conservation, aviculturists, the pet industry and local

human communities. From a conservation standpoint, this will not only maintain healthy wild parrot populations, but also make substantial forested

areas of extractive reserves a commodity more valuable than clearing for other land uses that may result in the conservation of many other associated

species. From birds harvested sustainably, aviculturists would be able to purchase new genetic stock for breeding programmes and the pet industry

would have a small steady flow of legally imported birds already conditioned to captive conditions. Lastly, the profits and benefits from the trade of

sustainable harvesting would be directed to the local communities as means of financial support. Birds may be harvested a number of ways such as –

harvesting only nestlings and not adults, specifically nestlings produced in excess of natural productivity through management programmes where

nesting pairs are close to the carrying capacity of the environment. And practicing early partial brood removal (i.e. last-hatched chicks) in species where

brood reduction occurs regularly so as not to greatly affect productivity or in species where the overall productivity may be increased through nest box

248

use. Sustainable harvests of parrots need to be biologically justified by detailed biological studies of the demography, natural history, population trends

and movements of the species to be harvested before setting a quota. The harvest and trade must also be effective and regulated nationally and

internationally, meaning this conservation option may only be suitable for countries with the funds and administrative and enforcement capabilities. A

number of potential problems with sustainable parrot harvest that will exacerbate the conservation issue are - the illegal laundering of non-sustainably

harvested birds through these legal programmes, the continued poaching outside of these programmes and the tendency to be parsimonious with the

costs of wild population monitoring and to overharvest to maximise short-term profits as oppose to ensuring long-term sustainability, as the birds

become increasingly viewed as items of legal trade. This may also lead to a distortion of wildlife management efforts to selectively favour only those

high trade value species. This is because of the economic incentive - ‘non-ranched’ parrots are less costly to harvest than sustainably ranched birds (i.e.

no nest box monitoring costs versus locating nests). Also illegal non-ranched birds, harvested as nestlings, are of equal trade quality as legally ranched

parrots, and are also hard to detect in trade. Although the attractiveness of laundering parrots may be reduced through bird identification systems based

on DNA-type techniques (e.g. Carnaby’s Cockatoo in Australia), such operations are invariably expensive and logistically difficult considering the large

number of individuals and species. Thus, these are unlikely to be borne by the ranchers themselves, without subsidies from the government. Another

issue with the live bird trade is that the value of trade products tends to fluctuate greatly through time, as it is dependent on the vagaries of supply and

demand (e.g. species availability). Other potential risks associated with this include the continued export of wildlife diseases to trade countries, the

establishment of feral pest populations of exotic parrots.

249

10.4.3 What are the implications of my findings for conservation of psittacines

elsewhere?

Like many land use activities, exploration and land clearing by mining companies clear

large tracts of vegetation. This removes habitat and is a threatening process for many

faunal species, including psittacines (Snyder et al. 2000). For example, three psittacid

species listed as ‘vulnerable’ on the IUCN list - the Red-faced Parrot Hapalopsittaca

pyrrhops, the Golden-plumed Parakeet Leptosittaca branickii and the White-breasted

Parakeet Pyrrhura albipectus - all inhabit the tropical rainforest of Podocarpus National

Park in southeast Ecuador. Gold mining activities in and around the park pose a major

threat to these birds through habitat loss and mercury poisoning of rivers (Toyne et al.

1992; Vallée 1992; Snyder et al. 2000). Two other areas in which the White-breasted

Parakeet is found (i.e. Cordillera de Cutucú and Cordillera del Condor) are also

threatened by gold mining (Collar et al. 1992; Toyne et al. 1992). It is of importance

that all mining activities in the Podocarpus National Park are halted in order to secure

habitat for these three globally threatened parrots.

There are implications from my findings at NBG for the case of the Podocarpus

National Park psittacines, particularly in relation to habitat loss and mercury poisoning.

Similarly, at the study site in Western Australia, land clearing for gold mining has led

resulted in a loss of habitat for all three southwestern black cockatoo species (Lee et al.

2010). Although the impacts of the habitat loss on the three psittacines in Ecuador were

not specified, land clearing for mining generally removes breeding and feeding habitat,

causing population decline.

From my study, one way in which a loss of feeding habitat may be mitigated is through

the rehabilitation of mine pits post-mining. Younger-aged revegetation may play a role

in attracting parrot feeding, so this measure begins early. Secondly, although it was not

stated if mercury poisoning of the river systems had a direct impact on the three

psittacine populations, it is likely (Donato et al. 2007). One way in which this may be

avoided is through the establishment of artificial water sources that would hopefully

attract psittacines away from the potentially hazardous water systems. This is shown in

the NBG study where faunal drinking points (FDPs) were established away from the

RDAs. Black cockatoos used these FDPs and did not drink from the RDAs.

250

A further example concerns the impact of loss and fragmentation of natural forests on

the Black-billed Parrot Amazona agilis in Jamaica. Deforestation and habitat

degradation has reduced shelter from adverse weather (e.g. hurricanes), food resources

and nesting sites in the form of tree hollows (Koenig 2001, 2008). The species inhabits

the karst limestone region of Cockpit Country (Koenig 2001). This is a region that

contains 27 of Jamaica’s 28 endemic bird species and is the highest-ranked Important

Bird Area on the island (Koenig 2008). Although the rough terrain has hindered large-

scale exploitation, the periphery has been disturbed and modified for various land uses,

creating a mosaic of forest, selective cutting of hardwood and agricultural (e.g. sugar

cane and cattle production) patches (Koenig 2008). Alcoa Minerals of Jamaica LLC and

Clarendon Alumina Producers were issued a licence in 2004 (renewed in 2005 and

2006) by the Government of Jamaica to prospect for bauxite more than 50% of the

Cockpit Country Landscape (Koenig 2008). However, after an intensive public

awareness campaign opposing any mining in the landscape, the licence was suspended

in December 2006, and a second prospecting licence was also surrendered. Despite the

Government agreeing not to mine the interior ‘heart’ of Cockpit Country, the exterior

regions, which were covered in the surrendered licence, are still open to future

prospecting (Koenig 2008).

Currently, mining areas exist east of Cockpit Country including the central region of

Mount Diablo, which has been mined in the past four decades (Koenig 2008). However,

much is still not known about the impacts of mining. A population viability analysis

highlighted that breeding pairs in edge habitat had a poorer reproductive potential

compared to interior ‘core’ sites. The model also predicted that the effects of bauxite

mining would increase the ‘edge’ habitat through land clearing and fragmentation, and

illegal forestry and poaching will continue to be advanced by mining infrastructure

(Koenig 2008). The study also emphasised that these threatening factors along with

their mitigation costs should be evaluated in association with bauxite mining (Koenig

2008).

The situation of the Black-billed Parrot in Jamaica is similar to that of the black

cockatoos in southwestern Australia in that both landscapes are affected by mining and

logging. While my study did not look at breeding habitat and psittacine reproductive

potential at interior versus ‘edge’ sites, the habitat fragmentation (from mining and

logging) leading to the increasing ‘edge’ habitat and loss of breeding resources in the

251

study is similar to that existing at NBG. Therefore, two implications of my findings at

NBG to this case study is the addition of artificial nest hollows (ANHs) and efforts to

rehabilitate. Black cockatoos display site-fidelity. Despite not observing breeding with

ANHs setup on-site at NBG, the results of an ANH survey in Western Australia found

that the birds were more likely to use ANH established in known breeding sites

regardless if they are cleared or uncleared. Black-billed Parrot has been suggested to

display site-fidelity (Koenig et al. 2007) and therefore the establishment of ANH may

be successful as increasing breeding potential at the Jamaican mine site. However, more

detailed studies on this trait, as well as their nesting requirement and breeding grounds

is required if ANH addition is to be implemented as a conservation measure.

Another way is through the rehabilitation of natural nest sites. Due to the extensive time

lag in which breeding habitat may be returned to a landscape (Vesk et al. 2008), the

study at NBG was not able to examine the effects of restoring breeding habitat for black

cockatoos. However, by revegetating a site, naturally with time and in the absence of

any obstruction or disturbance, a functional forest ecosystem will ultimately be

established and this has the potential to return not only nesting sites on a long-term, but

also feeding resources on a shorter time-scale. However, as indicated in my study, this

should be mitigated in conjunction with the preservation of existing native forest.

10.5 Concluding Remarks

Lunney (2004a) called the conservation of forest fauna ‘a test of our civilisation’. This

is because we now understand clearly that conservation cannot be achieved through

reserves alone, but requires attention to fauna management in production landscapes as

well (Lindemayer and Franklin 2002). Given this fundamental principle, conservation

of forest fauna becomes a moral issue – whether our society is prepared to value

conservation and to manage accordingly.

There are few long-term studies of black cockatoos in production landscapes, an

information deficiency which has impeded our understanding of the dynamics of black

cockatoo populations and their response to environmental and anthropogenic factors.

This thesis, in combination with the work of other researchers, has increased our

understanding of how to minimise the impacts of mining on black cockatoos and, in

252

particular, how to restore mine-sites to provide suitable habitat for black cockatoos on a

scale of decades rather than centuries.

Nevertheless, our understanding of best practice for the restoration of black cockatoo

habitat is still developing. Studies at NBG could help establish best practice in this area

by supporting a long-term study of habitat restoration techniques over the life of the

mine and at closure that would allow for the trial and monitoring of different habitat

restoration approaches. This would integrate our knowledge on the feeding, breeding

and drinking behaviours of black cockatoos, not only at NBG, but also throughout

southwestern Australia.

The will to apply that understanding will be critical. As Lunney (2004b) observed about

Australian forest fauna in general: In 2104, a subsequent generation may look back and smile at (our) earnest efforts…but that smile will surely be a grimace if there was a failure by their forebears to reverse the continuing loss and fragmentation of forest habitats and the simplifying of those forests that remain.

The same can certainly be said if we fail to act on our understanding to conserve the

black cockatoos of southwestern Australia.

  253  

References

Abbott, I., 1985. Rate of growth of Banksia grandis Willd. (Proteaceae) in Western

Australian forest. Australian Journal of Botany 33: 381 - 391.

Abbott, I., 1998. Conservation of the forest red-tailed black cockatoo, a hollow

dependent species, in the eucalypt forests of Western Australia. Forest Ecology and

Management 109: 175 - 185.

Abbott, I., 2001. Karrak watch: a summary of the information about the Forest Red-

tailed Black cockatoo (FRTBC) of south-west Western Australia. [Online]. Department

of Environment and Conservation, Perth, Western Australia. Available from:

http://www.dec.wa.gov.au/content/view/2384/1971/. Accessed Thu, 1 Feb 2012.

Abbott, I., 2003. Aboriginal fire regimes in south-west Western Australia: evidence

from historical documents. Pp. 119 - 146 in Fire in ecosystems of south-west Western

Australia: impacts and management. Ed by I. Abbott and N. Burrows. Backhuys

Publishers, Leiden, The Netherlands.

Abbott, I., 2008. Historical perspectives of the ecology of some conspicuous vertebrate

species in south-west Western Australia. Conservation Science Western Australia 6: 1 -

214.

Abbott, I. and Loneragan, O. W., 1986. Ecology of Jarrah (Eucalyptus marginata) in the

northern Jarrah forest of Western Australia. Department of Conservation and Land

Management, Perth, Western Australia.

Abbott, I. and Whitford, K., 2002. Conservation of vertebrate fauna using hollows in

forests of south-west Western Australia: strategic risk assessment in relation to ecology,

policy, planning, and operations management. Pacific Conservation Biology 7: 240 -

255.

  254  

Abbott, I., Dell, B. and Loneragan, O., 1989. The jarrah plant. Pp. 41 - 51 in The jarrah

forest: a complex mediterranean ecosystem. Ed by B. Dell, J. J. Havel and N.

Malajczuk. Kluwer, Dordrecht, Netherlands.

Adkins, M. F., 2006. A burning issue: using fire to accelerate tree hollow formation in

eucalypt species. Australia Forestry 69: 107 – 113.

Allan, C., Watts, R. J., Commens, S. and Ryder, D. S., 2009. Using adaptive

management to meet multiple goals for flows along the Mitta Mitta River in south-

eastern Australia. Pp. 59 - 72 in Adaptive environmental management: a practitioner’s

guide. Ed by C. Allan and G. H. Stankey. Springer and CSIRO Publishing, Dordrecht,

the Netherlands and Collingwood, Australia.

Altmann, J., 1974. Observational study of behavior: sampling methods. Behaviour 49:

227 - 267.

Argent, R. M., 2009. Components of adaptive management. Pp. 11 - 36 in Adaptive

Environmental Management: a practitioner’s guide. Ed by C. Allan and G. H. Stankey.

Springer and CSIRO Publishing, Dordrecht, the Netherlands and Collingwood,

Australia.

Armstrong, J. A. and Abbott, I., 1995. Sustainable conservation - a practical approach to

conserving biodiversity in Western Australia. Pp. 21 - 28 in Conservation through

sustainable use of wildlife. Ed by G. C. Grigg, P. T. Hale and D. Lunney. Centre for

Conservation Biology, the University of Queensland, Brisbane, Queensland.

Armstrong, K. N. and Nichols, O. G., 2000. Long-term trends in avifaunal

recolonisation of rehabilitated bauxite mines in the jarrah forest of southwestern

Australia. Forest Ecology and Management 126: 213 - 225.

Auld, T. D. and Myerscough, P. J., 1986. Population dynamics of the shrub Acacia

suaveolens (Sm.) Willd.: seed production and predispersal seed predation. Australian

Journal of Ecology 11: 219 - 234.

  255  

Bamford, M. J., Wilcox, J. A. and Davis, R. A., 2004. Worsley Alumina extension area:

northern mining envelopes. Preliminary fauna assessment. Unpublished report prepared

for Worsley Alumina Pty Ltd by Bamford Consulting Ecologist and Western Wildlife.

Barrett, G., Silcocks, A., Barry, S., Cuningham, R. and Poulter, R., 2003. The new atlas

of Australian birds. Royal Australasian Ornithologists’ Union, Victoria, Australia.

Bartle, J. R. and Riches, J. R. H., 1978. Rehabilitation after mining in the Collie

coalfield. Pp. 1 - 5 in Rehabilitation of mined lands in Western Australia: proceedings

of a meeting held in Perth on 11th October 1978. Ed by E. D. Fox. Western Australian

Institute of Technology, Bentley.

Bartle, J. and Slessar G. C., 1989. Mining and rehabilitation. Pp. 357 - 377 in The jarrah

forest: a complex Mediterranean ecosystem. Ed by B. Dell, J. J. Havel and N.

Malajczuk. Kluwer, Dordrecht.

Bednarz, J. C., Martin, J. H., Benson, T. J. and Varland, D. E., 2013. The efficacy of

fungal inoculation of live trees to create wood decay and wildlife-use trees in managed

forests of western Washington, USA. Forest Ecology and Management 307: 186 – 195.

Begon, M., Townsend, C. R. and Harper, J. L., 2006. Ecology: From individuals to

ecosystems. Blackwell Publishing, Oxford.

Beissinger, S. R., Tygielski, S. and Elderd, B., 1998. Social constraints on the onset of

incubation in a neotropical parrot: a nestbox addition experiment. Animal Behaviour 55:

21 - 32

Bennett, A. F., Kimber, S. L. and Ryan, P. A., 2000. Revegetation and wildlife - a guide

to enhancing revegetated habitats for wildlife conservation in rural environments.

Bushcare National and Research and Development Program Research Report 2/00.

Bentley, P. J., 2002. Endocrines and osmoregulation: a comparative account in

vertebrates. Springer, Berlin, London.

  256  

Berry, P. F., 2008. Counts of Carnaby’s cockatoo (Calyptorhynchus latirostris) and

records of flock composition at an overnight roosting site in metropolitan Perth. The

Western Australian Naturalist 26: 1 - 11.

Beyer, G. L. and Goldingay, R. L., 2006. The value of nest boxes in the research and

management of Australian hollow-using arboreal marsupials. Wildlife Research 33: 161

- 174.

Biggs, E., Finn, H., Taplin, R. and Calver, M., 2011. Landscape position predicts

distribution of eucalypt feed trees for threatened black-cockatoos in the northern jarrah

forest, Western Australia. Journal of the Royal Society of Western Australia 94: 541 -

548.

BirdLife International 2010a. Calyptorhynchus latirostris. [Online]. IUCN Red List of

Threatened Species. Version 2011.2. Available from:

http://www.iucnredlist.org/apps/redlist/details/106001391/0.

Accessed, Fri 16 Dec 2011.

BirdLife International 2010b. Calyptorhynchus baudinii [Online]. IUCN Red List of

Threatened Species. Version 2011.2. Available from:

http://www.iucnredlist.org/apps/redlist/details/106001390/0.

Accessed, Fri 16 Dec 2011.

Bohner, F., 1984. First breeding of the white-tailed black cockatoo. Bird Keeping in

Australia 27: 17 - 18.

Bolton, M., Medeiros, R., Hothersall, B. and Campos, A., 2004. The use of artificial

breeding chambers as a conservation measure for cavity-nesting Procellariiform

seabirds: a case study of the Madeiran storm petrel (Oceanodroma castro). Biological

Conservation 116: 73 - 80.

Bond, W. J. and van Wilgen, B. W., 1996. Fire and plants. Chapman & Hall Ltd.

  257  

Bowen, B. J. and Pate, J. S., 2004. Effect of season of burn on shoot recovery and post-

fire flowering performance in the resprouter Stirlingia latifolia R. Br. (Proteaceae).

Austral Ecology 29: 145 - 155.

Boyce, M. S., 1997. Population viability analysis: adaptive management for threatened

and endangered species. Pp. 226 - 238 in Ecosystem management: applications for

sustainable forest and wildlife resources. Ed by M. S. Boyce and A. Haney. Yale

University Press, USA.

Bray, J. R. and Curtis, J. T., 1957. An ordination of the upland forest communities of

southern Wisconsin. Ecological monographs 27: 325 - 349.

Brasel, J. M., Cooper, R. C. and Pritsos, C. A., 2006. Effects of environmentally

relevant doses of cyanide on flight times in Pigeons, Columba livia. Bulletin of

environmental contamination and toxicology 76: 202 - 209.

Brawn, J. D., 1988. Selectivity and ecological consequences of cavity nesters using

natural versus artificial nest sites. Auk 105: 789 - 791.

Brennan, K. E., Nichols, O. G. and Majer, J. D., 2005. Innovative techniques for

promoting fauna return to rehabilitated sites following mining. MERIWA Report, 248.

Brooks, S. S. and Lake, P. S., 2007. River restoration in Victoria, Australia: change is in

the wind, and none too soon. Restoration Ecology 15: 584 - 591.

Bryan, B. A., 2002. Reserve selection for nature conservation in South Australia: past,

present and future. Australian Geographical Studies 40: 196 - 209.

Bull, E. L. and Partridge, A. D., 1986. Methods of killing trees for use by cavity nesters.

Wildlife Society Bulletin 14: 142 – 146.

Bull, E. L., Partridge, A. D. and Williams, W. G., 1981. Creating snags with explosives.

Research Note PNW-393, Pacific Northwest Forest and Range Experiment Station,

USDA Forest Service.

  258  

Burns, B., Barker, G. M., Harris, R. J. and Innes, J., 2000. Conifers and cows: forest

survival in a New Zealand dairy landscape. Pp. 81 - 89 in Nature Conservation 5:

Conservation in Production Landscapes: Managing the matrix. Ed by J. Craig, N.

Mitchell and D. A. Saunders. Surrey Beatty and Sons, Chipping Norton, Sydney,

Australia.

Burrows, N. and Abbott, I., 2003. Fire in south-west Western Australia: synthesis of

current knowledge, management implications and new research directions. Pp. 437 -

452 in Fire in ecosystems of south-west Western Australia: impacts and management.

Ed by I. Abbott and N. Burrows. Backhuys Publishers, Leiden, The Netherlands.

Burrows, N. and Wardell-Johnson, G. W., 2003. Fire and plant interactions in forested

ecosystems of south-west Australia: a review. Pp. 225 - 268 in Fire in ecosystems of

south-west Western Australia: impacts and management. Ed by I. Abbott and N.

Burrows. Backhuys Publishers, Leiden, The Netherlands.

Burrows, N. D., Ward, B. and Robinson, A. D., 1995. Jarrah forest fire history from

stem analysis and anthropological evidence [Eucalyptus marginata; Western Australia].

Australian Forestry 58: 7 - 16.

Cale, B., 2003. Carnaby's Black-Cockatoo (Calyptorhychus latirostris) Recovery Plan

2002 - 2012. Department of Conservation and Land Management, Perth Western

Australia.

Calver, M. C. and Dell, J., 1998. Conservation status of mammals and birds in

southwestern Australian forests. I. Is there evidence of direct links between forestry

practices and species decline and extinction? Pacific Conservation Biology 4: 296 - 314.

Calver, M. C. and Wardell-Johnston, G., 2004. Sustained unsustainability? An

evaluation of evidence for a history of overcutting in the jarrah forests of Western

Australia and its consequences for fauna conservation. Pp. 94 - 114 in Conservation of

Australia's forest fauna, 2nd edition. Ed by D Lunney. Royal Zoological Society of New

South Wales, Mosman, New South Wales, Australia.

  259  

Calver, M. C., Bradley, J. S. and Wright, 1. W., 1999. Towards scientific contributions

in applying the Precautionary Principle: An example from Southwestern Australia.

Pacific Conservation Biology 5: 63 – 72.

Calver, M. C., Dickman, C. R., Feller, M. C., Hobbs, R. J., Horwitz, P., Recher, H. F.

and Wardell-Johnson, G., 1998. Towards resolving conflict between forestry and

conservation in Western Australia. Australian Forestry 61: 258 - 266.

Cameron, M., 2005. Group size and feeding rates of Glossy Black-Cockatoos in central

New South Wales. Emu 105: 299 - 304.

Cameron, M., 2006. Nesting habitat of the glossy black-cockatoo in central New South

Wales. Biological Conservation 127: 402 - 410.

Cameron, M., 2007. Cockatoos. CSIRO Publishing, Collingwood, Victoria, Australia.

Cameron, M. and Cunningham, R. B., 2006. Habitat selection at multiple spatial scales

by foraging glossy black-cockatoos. Austral Ecology 31: 597 - 607.

Carey, A. B., 1993. The forest ecosystem study: experimental manipulation of managed

stands to provide habitat for spotted owls and to enhance plant and animal diversity: a

summary and background for the interagency experiment at Fort Lewis, Washington,

Forestry Sciences Laboratory.

Carey, A. B. and Sanderson, H. R., 1981. Routine to accelerate tree cavity formation.

Wildlife Society Bulletin 9: 14 – 21.

Catterall, C. P., Kanowski, J., Wardell-Johnson, G. W., Proctor, H., Reis T., Harrison,

D. and Tucker, N. I. J., 2004. Quantifying the biodiversity values of reforestation:

perspectives, design issues and outcomes in Australian rainforest landscapes. Pp. 359 -

393 in Conservation of Australia’s Forest Fauna, 2nd edition. Ed by D. Lunney. Royal

Zoological Society of New South Wales, Mosman, New South Wales, Australia.

Caughley, G. and Gunn, A., 1996. Conservation biology in theory and practice.

Blackwell Science, Cambridge, UK.

  260  

Cawthen, L., Munks, S., Richardson, A. and Nicol, S. C., 2009. The use of temperature

loggers to monitor tree hollow use by mammals. Ecological Management and

Restoration 10: 153 - 155.

Chapman, A. and Dell, J., 1985. Biology and Zoogeography of the Amphibians and

Reptiles of the Western Australian Wheatbelt. Records of the Western Australian

Museum 12: 1 - 46.

Chapman, T. F., 2007. An endangered species that is also a pest: a case study of

Baudin's Cockatoo Calyptorhynchus baudinii and the pome fruit industry in south-west

Western Australia. Journal of the Royal Society of Western Australia 90: 33 - 40.

Chapman, T. F., 2008. Forest Black Cockatoo (Baudin's cockatoo Calyptorynchus

baudinii and Forest Red-tailed Black Cockatoo Calyptorynchus banksii naso) Recovery

Plan. Department of Conservation and Land Management, Perth Western Australia.

Chapman, T. F. and Paton, D. C., 2005. The glossy black-cockatoo (Calyptorhynchus

lathami halmaturinus) spends little time and energy foraging on Kangaroo Island, South

Australia. Australian Journal of Zoology 53: 177 - 183.

Chapman, T. F. and Paton, D. C., 2006. Aspects of drooping sheoaks (Allocasuarina

verticillata) that influence glossy black-cockatoo (Calyptorhynchus lathami

halmaturinus) foraging on Kangaroo Island. Emu 106: 163 - 168.

Charles, S. P., Silberstein, R., Teng, J., Fu, G., Hodgson, G., Gabrovsek, C., Crute, J.,

Chiew, F. H. S., Smith, I. N., Kirono, D. G. C., Bathols, J. M., Li, L. T., Yang, A.,

Donohue, R. J., Marvanek, S. P., McVicar, T. R., Van Niel, T. G. and Cai, W., 2010.

Climate analyses for south-west Western Australia. A report to the Australian

Government from the CSIRO South-West Western Australia Sustainable Yields Project.

CSIRO, Australia. Pp. 83.

Chazdon, R. L., Peres, C. A., Dent, D., Sheil, D., Lugo, A. E., Lamb, D., Stork, N. E.

and Miller, S. E., 2009a. The potential for species conservation in tropical secondary

forests. Conservation Biology 23:1406 - 1417.

  261  

Chazdon, R. L., Harvey, C. A., Komar, O., Griffith, D. M., Ferguson, B. G., Martínez-

Ramos, M., Morales, H., Nigh, R., Soto-Pinto, L., Van Breugel, M. and Philpott, S. M.,

2009b. Beyond reserves: a research agenda for conserving biodiversity in human-

modified tropical landscapes. Biotropica 41: 142 - 153.

Clarke K. R., 1993. Non-parametric multivariate analysis of changes in community

structure. Australian Journal of Ecology 18: 117 - 143.

Clarke, K. R. and Warwick, R. M., 2001. Change in marine communities: an approach

to statistical analysis and interpretation. Primer-E Ltd, Plymouth, UK.

Clarke, K. R., Somerfield, P. J. and Chapman, M. G., 2006. On resemblance measures

for ecological studies, including taxonomic dissimilarities and a zero-adjusted Bray-

Curtis coefficient for denuded assemblages. Journal of Experimental Marine Biology

and Ecology 330: 55 - 80.

Clarke, R. V. and de By, R. A., 2013. Poaching, habitat loss and the decline of

neotropical parrots: a comparative spatial analysis. Journal of Experimental

Criminology 9: 333 – 353.

Clout, M. N., 1989. Forgaing behavior of glossy black-cockatoos. Australian Wildlife

Research 16: 467 - 473.

Cockle, K. L., Martin, K. and Robledo, G., 2012. Linking fungi, trees, and hole-using

birds in a neotropical tree-cavity network: Pathways of cavity production and

implications for conservation. Forest Ecology and Management 264: 210 – 219.

Collar, N. J., 2000. Globally threatened parrots: criteria, characteristics and cures.

International Zoo Yearbook 37: 21 – 35.

Collar, N. J. and Juniper, A. T., 1992. Dimensions and causes of the parrot conservation

crisis. Pp.1 - 24 in New World parrots in crisis: solutions from conservation biology. Ed

by S. R. Beissinger and N. F. R. Snyder, Smithsonian Institution Press, Washington,

D.C..

  262  

Collar, N. J., Gonzaga, L. P., Krabbe, N. J., Madroño Nieto, A., Naranjo, L. G., Parker

III, T. A. and Wege, D. C., 1992. Threatened birds of the Americas. The ICBP/ IUCN

red data book, Third edition (Part 2). BirdLife International Publications, Cambridge

UK.

Commonwealth of Australia, 2009a. Matters of national environmental significance.

Significant impact guidelines 1.1. Environment Protection and Biodiversity

Conservation Act 1999, Australia.

Commonwealth of Australia, 2009b. Significant impact guidelines for the vulnerable

western ringtail possum (Pseudocheirus occidentalis) in the southern Swan Coastal

Plain, Western Australia. Nationally threatened species and ecological communities

Environmental Protection and Biodiversity Act (EPBC) policy statement 3.10.

Department of the Environment, Water, Heritage and the Arts, Canberra, ACT.

Conservation Commission of Western Australia, 2003. Forest management plan 2004-

2013. Conservation Commission of Western Australia, Perth, Western Australia.

Conservation Commission of Western Australia, 2012. Draft Forest Management Plan

2014-2023. Conservation Commission of Western Australia, Perth.

Conner, R. N., Dickson, J. G. and Locke, B. A., 1981. Herbicide-killed trees infected by

fungi: potential cavity sites for woodpeckers. Wildlife Society Bulletin 9: 308 - 310.

Conner, R. N., Kroll, J. C. and Kulhavy, D. L., 1983. The potential of girdled and 2,4-

D-injected southern red oaks as woodpecker nesting and foraging sites. Southern

Journal of Applied Forestry 7: 125 – 128.

Connors, N. and Connors, E., 2005. A guide to black cockatoos as pet and aviary

birds. ABK Publications, South Tweed Heads, NSW.

Cooper, C., 2000. Food manipulation by southwest Australian cockatoos. Eclectus 8: 3 -

9.

  263  

Cooper, C. E., Withers, P. C., Mawson, P. R., Bradshaw, S. D., Prince, J. and

Robertson, H., 2002. Metabolic ecology of cockatoos in the south-west of Western

Australia. Australian Journal of Zoology 50: 67 - 76.

Cooper, C. E., Withers, P. C., Mawson, P. R., Johnstone, R., Kirkby, T., Prince, J.,

Bradshaw, S. D., and Robertson, H., 2003. Characteristics of Marri (Corymbia

callophylla) fruits in relation to the foraging behaviour of the Forest Red-tailed Black

Cockatoo (Calyptorhynchus banksii naso). Journal of the Royal society of Western

Australia 86: 139 - 142.

Craig, M. D., 2007. The short-term effects of edges created by forestry operations on

the bird community of the jarrah forest, south-western Australia. Austral Ecology 32:

386 - 396.

Craig, M. D. and Roberts, J. D., 2005. The short-term impacts of logging on the jarrah

forest avifauna in south-west Western Australia: implications for the design and

analysis of logging experiments. Biological Conservation 124: 177 - 188.

Craig, M. D., Hobbs, R. J., Grigg, A. H., Garkaklis, M. J., Grant, C. D., Fleming, P. A.

and Hardy, G. E. S. J., 2010. Do thinning and burning sites revegetated after bauxite

mining improve habitat for terrestrial vertebrates? Restoration Ecology 18: 300 - 310.

Croton, J. T. and Reed, A. J., 2007. Hydrology and bauxite mining on the Darling

Plateau. Restoration Ecology 15: S40 - S47.

Cundill, G., Cumming, G. S., Biggs, D. and Fabricius, C., 2012. Soft systems thinking

and social learning for adaptive management. Conservation Biology 26: 13 - 20.

Danks, A., 1994. Noisy scrub-bird translocations: 1983-1992. Pp. 129 – 134 in

Reintroduction Biology of Australian and New Zealand Fauna. Ed by M Serena. Surrey

Beatty.

Danks, A., 1998. Conservation of the noisy scrub-bird: a review of 35 years of research

and management. Pacific Conservation Biology 3: 341 - 349.

  264  

Davies, S. J. J. F., 1966. The movements of the white-tailed black-cockatoos

(Calyptorhynchus baudinii) in south-western Australia. Western Australian Naturalist

10: 33 - 42.

Davies, S. J. J. F., 1982. Behavioural adaptations of birds to environments where

evaporation is high and water is in short supply. Comparative Biochemistry and

Physiology 71: 557 - 566.

Davies, S., 2003. Some practical steps for enhancing the status of Carnaby’s black

cockatoo in Conserving Carnaby’s black cockatoo, Future Directions: Proceedings from

a conservation symposium. Ed by C. Gole. Birds Australia and Department of

Conservation and Land Management, Perth, Western Australia.

Davies, S., 2005. Usage of artificial nest boxes by Carnaby’s cockatoo. Assessment of

status 2004. Report to World Wide Fund for Nature, Threatened Species Program and

Men of the Trees.

Davis, S. K., 2004. Area sensitivity in grassland passerines: effects of patch size, patch

shape, and vegetation structure on bird abundance and occurrence in southern

Saskatchewan. The Auk 121: 1130 - 1145.

Dell, B., Havel, J. J. and Malajczuk, N. (Ed), 1989. The Jarrah forest: a complex

mediterranean ecosytem. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Department of Environment and Conservation (DEC), 2012. Carnaby’s Cockatoo

(Calyptorhynchus latirostris) Recovery Plan. Western Australian Wildlife Management

Program No. 52, Government of Western Australia.

Department of Environment, Heritage, Water and Arts (DEWHA), 2010. Survey

guidelines for Australia's threatened birds, Environment Protection and Biodiversity

Conservation Act survey guidelines 6.2. Canberra, ACT, Australia.

Department of Sustainability, Environment, Water, Population and Communities

(DSEWPaC), 2011. Environmental Protection and Biodiversity Conservation Act 1999

draft referral guidelines for three threatened black cockatoo species: Carnaby’s

  265  

cockatoo (endangered) Calyptorhynchus latirostris, Baudin’s cockatoo (vulnerable)

Calyptorhynchus baudinii, Forest red-tailed black cockatoo (vulnerable)

Calyptorhynchus banksii naso.

Department of Sustainability, Environment, Water, Population and Communities

(DSEWPaC), 2012a. Calyptorhynchus banksii naso. [Online]. Species Profile and

Threats Database, Department of Sustainability, Environment, Water, Population and

Communities, Canberra, Australian Commonwealth Territory. Available from:

http://www.environment.gov.au/sprat. Accessed Wed, 25 Jan 2012.

Department of Sustainability, Environment, Water, Population and Communities

(DSEWPaC), 2012b. Calyptorhynchus baudinii. [Online]. Species Profile and Threats

Database, Department of Sustainability, Environment, Water, Population and

Communities, Canberra, Australian Commonwealth Territory. Available from:

http://www.environment.gov.au/sprat. Accessed Wed, 25 Jan 2012.

Department of Sustainability, Environment, Water, Population and Communities

(DSEWPaC), 2012c. Calyptorhynchus latirostris. [Online]. Species Profile and Threats

Database, Department of Sustainability, Environment, Water, Population and

Communities, Canberra, Australian Commonwealth Territory. Available from:

http://www.environment.gov.au/sprat. Accessed Wed, 25 Jan 2012.

Dixon, V., Glover, H. K., Winnell, J., Treloar, S. M., Whisson, D. A. and Weston, M.

A., 2009. Evaluation of three remote camera systems for detecting mammals and birds.

Ecological Management and Restoration 10: 156 - 157.

Donato, D. B., Nichols, O., Possingham, H., Moore, M., Ricci, P. F. and Noller, B. N.,

2007. A critical review of the effects of gold cyanide-bearing tailings solutions on

wildlife. Environment International 33: 974 - 984.

Donato, D., Ricci, P. F., Noller, B., Moore, M., Possingham, H. and Nichols, O., 2008.

The protection of wildlife from mortality: hypothesis and results for risk assessment.

Environment International 34: 727 - 736.

  266  

Durant, R., Luck, G. W. and Matthews, A., 2009. Nest-box use by arboreal mammals in

a peri-urban landscape. Wildlife Research 36: 565 - 573.

Eisler, R. and Wiemeyer, S. N., 2004. Cyanide hazards to plants and animals from gold

mining and related water issues. Reviews of Environmental Contamination and

Toxicology 183: 21 - 54.

Emison, W. B., 1996. Use of supplementary nest hollows by an endangered subspecies

of red-tailed black-cockatoo. Victorian Naturalist 113: 262 - 263.

Emison, W. B., Beardsell, C. M. and Temby, I. D., 1994. The biology and status of the

long-billed corella in Australia. Proceedings of the Western Foundation of Vertebrate

Zoology 5: 209 – 247.

Environmental Protection Authority (EPA), 1994. Boddington gold mine: rehabilitation

strategy. Bulletin 766. Report and recommendations of the Environmental Protection

Authority, Perth, Western Australia.

Fahrig, L., 2001. How much habitat is enough? Biological Conservation 100: 65 - 74.

Faith, D. P., Minchin, P. R. and Belbin, L., 1987. Compositional dissimilarity as a

robust measure of ecological distance. Plant Ecology 69: 57 - 68.

Fernández-Juricic, E. and Jokimäki, J., 2001. A habitat island approach to conserving

birds in urban landscapes: case studies from southern and northern Europe. Biodiversity

and Conservation 10: 2023 - 2043.

Fierke, M. K., Kinney, D. L., Salisbury, V. B., Crook, D. J. and Stephen, F. M., 2005.

Development and comparison of intensive and extensive sampling methods and

preliminary within-tree population estimates of red oak borer (Coleoptera:

Cerambycidae) in the Ozark Mountains of Arkansas. Environmental Entomology

34:184 - 192.

Finn, H., 2011. Assessment of habitat values for black cockatoos within the eastern

acquired lands at Newmont Boddington Gold. Report to Newmont Boddington Gold.

  267  

School of Biological Sciences and Biotechnology, Murdoch University, Perth, Western

Australia.

Finn, H., Stock, W. and Valentine, L., 2009. Pines and the ecology of Carnaby’s Black-

Cockatoos (Calyptorhyncus latirostris) in the Gnangara Sustainability Strategy study

area. Technical report for the Forest Products Commission (Perth, Western Australia) in

support of the Gnangara Sustainability Strategy (GSS). Available from:

http://ro.ecu.edu.au/snsc_papers/.

Finnie, B., Stuart, J., Gibson, L. and Zabriskie, F., 2009. Balancing environmental and

industry sustainability: a case study of the US gold mining industry. Journal of

Environmental Management 90: 3690 - 3699.

Fischer, J., Lindenmayer, D. B. and Manning, A. D., 2006. Biodiversity, ecosystem

function, and resilience: ten guiding principles for commodity production landscapes.

Frontiers in Ecology and the Environment 4: 80 - 86.

Fisher, C. D., Lindgren, E. and Dawson, W. R., 1972. Drinking patterns and behavior of

Australian desert birds in relation to their ecology and abundance. The Condor 74: 111 -

136.

Fitzsimons, J. A. and Robertson, H. A., 2005. Freshwater reserves in Australia:

directions and challenges for the development of a comprehensive, adequate and

representative system of protected areas. Hydrobiologia 552: 87 - 97.

Ford, J., 1980. Morphological and ecological divergence and convergence in isolated

populations of the red-tailed black-cockatoo. Emu 80: 103 - 120.

Forests Department, 1971. Forestry in Western Australia, Forests Department of

Western Australia, Perth.

Forshaw, J. M. and Cooper, W. T., 1934-2002. Australian parrots (3rd revised edition),

Alexander Editions, Robina, Queensland.

Foster, W. A., Snaddon, J. L., Turner, E. C., Fayle, T. M., Cockerill, T. D., Farnon

Ellwood, M. D., Broad, G. R., Chung, A. Y. C., Eggleton, P., Khen, C. V. and Yusah,

  268  

K. M., 2011. Establishing the evidence base for maintaining biodiversity and ecosystem

function in the oil palm landscapes of South East Asia.

Philosophical Transactions of the Royal Society B: Biological Sciences 366: 3277 -

3291.

Gardner, J. H. and Bell, D. T., 2007. Bauxite mining restoration by Alcoa world

alumina Australia in Western Australia: social, political, historical, and environmental

contexts. Restoration Ecology Supplement 15: S3 - S10.

Garkaklis, M. J., Calver, M. C., Wilson, B. A. and Hardy, G. E., 2004. Habitat

alteration caused by an introduced plant disease, Phytophthora cinnamomi: a potential

threat to the conservation of Australian forest fauna. Pp. 899 - 913 in Conservation of

Australia’s Forest Fauna, 2nd edition. Ed by D. Lunney. Royal Zoological Society of

New South Wales, Mosman, New South Wales, Australia.

Garnett, S. T., Pedler, L. P. and Crowley, G. M., 1999. The breeding biology of the

glossy black-cockatoo Calyptorhynchus lathami on Kangaroo Island, South Australia.

Emu 99: 262 - 279.

Garnett, S. T., Szabo, J. and Dutson, G., 2011. The Action Plan for Australian Birds

2010. CSIRO Publishing, Collingwood, Victoria.

Gentilli, J., 1989. Climate of the jarrah forest. Pp. 23 - 40 in The Jarrah Forest: A

Complex Mediterranean Ecosystem. Ed by B. Dell, J. J. Havel and N. Malajczuk.

Kluwer, Dordrecht, Netherlands.

George, T. L. and Zack, S., 2001. Spatial and temporal considerations in restoring

habitat for wildlife. Restoration Ecology 9: 272 - 279.

Gibbons, P., 1999. Habitat-tree retention in wood production forests. PhD thesis, The

Australian National University, Canberra.

Gibbons, P. and Lindenmayer, D. B., 1996. Issues associated with the retention of

hollow-bearing trees within eucalypt forests managed for wood production. Forest

Ecology and Management 83: 245 – 279.

  269  

Gibbons, P. and Lindenmayer, D. B., 1997a. Developing tree retention strategies for

hollow-dependent arboreal marsupials in the wood production eucalypt forests of

eastern Australia. Australian Forestry 60: 29 – 45.

Gibbons, P. and Lindenmayer, D. B., 1997b. A review of prescriptions employed for the

conservation of hollow-dependent fauna in wood production forests of eastern

Australia. Pp. 497 – 505 in Conservation Outside Nature Reserves. Ed by P. Hale and

D. Lamb. Centre of Conservation Biology, The University of Queensland, Brisbane.

Gibbons, P. and Lindenmayer, D. B., 2002. Tree hollows and wildlife conservation in

Australia. CSIRO Publishing, Sydney, New South Wales.

Gibbons, P. and Lindenmayer, D. B., 2007. Offsets for land clearing: No net loss or the

tail wagging the dog? Ecological Management and Restoration 8: 26 - 31.

Gibbons, P., Lindenmayer, D. B., Barry, S. C. and Tanton, M. T., 2000. Hollow

formation in eucalypts from temperate forests in southeastern Australia. Pacific

Conservation Biology 6: 218 - 228.

Gibbons, P., Lindenmayer, D. B., Barry, S. C. and Tanton, M. T., 2002. Hollow

selection by vertebrate fauna in forests of southeastern Australia and implications for

forest management. Biological Conservation 103: 1 - 12.

Gill, A. M., 1981. Adaptive responses of Australian vascular plant species to fire. Pp.

243 - 272 in Fire and the Australian Biota. Ed by A. M. Gill, R. H. Groves and I. Noble.

Australian Academy of Science, Canberra.

Goldingay, R. L., 2009. Characteristics of tree hollows used by Australian birds and

bats. Wildlife Research 36: 394 - 409.

Goldingay, R. L. and Stevens, J. R., 2009. Use of artificial tree hollows by Australian

birds and bats. Wildlife Research 36: 81 - 97.

González, J. A., 2003. Harvesting, local trade, and conservation of parrots in the

Northeastern Peruvian Amazon. Biological Conservation 114: 437 – 446.

  270  

Goosem, M. and Jago, R., 2006. Weed penetration, edge effects and rehabilitation

strategy success in weedy swathes of the Palmerston powerline clearing. Pp. 51 - 67 in

Weed incursions along roads and powerlines in the Wet Tropics World Heritage Area:

the potential of remote sensing as an indicator of weed infestations. Ed by M. W.

Goosem and M. Turton (research report). Cooperative Research Centre for Tropical

Rainforest Ecology and Management, Cairns.

Gould, S., 2011. Does post-mining rehabilitation restore habitat equivalent to that

removed by mining? A case study from the monsoonal tropics of northern Australia.

Wildlife Research 38: 482 - 490.

Grant, C., 2006. State-and-transition successional model for bauxite mining

rehabilitation in the jarrah forest of Western Australia. Restoration Ecology 14: 28 - 37.

Grant, C. D., Norman, M. A. and Smith, M. A., 2007. Fire and silvicultural

management of restored bauxite mines in Western Australia. Restoration Ecology

Supplement 15: S127 - S136.

Green, R. H., 1979. Sampling design and statistical methods for environmental

biologists. Wiley Interscience, Chichester, England.

Griffiths, S. R., Smith, G. B., Donato, D. B. and Gillespie, C. G., 2009. Factors

influencing the risk of wildlife cyanide poisoning on a tailings storage facility in the

eastern Goldfields of Western Australia. Ecotoxicology and Environmental Safety 72:

1579 - 1586.

Grigg, G. C., Hale, P. T. and Lunney, D. (Ed), 1995. Conservation through sustainable

use of wildlife. Centre for Conservation Biology, The University of Queensland,

Brisbane.

Groom C., 2010 Artificial hollows for Carnaby’s black cockatoo: an investigation of the

placement, use, monitoring and maintenance requirements of artificial hollows for

Carnaby’s black cockatoo. Project report. Department of Environment and

Conservation.

  271  

Gunarso, P. and Davie, J., 2000. Can Indonesian production forests play a nature

conservation role? Pp. 50 - 59 in Nature Conservation 5: Conservation in Production

Landscapes: Managing the matrix. Ed by J. Craig, N. Mitchell and D. A. Saunders.

Surrey Beatty and Sons, Chipping Norton, Sydney, Australia.

Hale, P. and Lamb, D. (Ed), 1997. Conservation outside nature reserves. Centre for

Conservation Biology, University of Queensland, Brisbane.

Hallett, J. G., Lopez, T., O’Connell, M. A. and Borysewicz, M. A., 2001. Decay

dynamics and avian use of artificially created snags. Northwest Science 75: 378 – 386.

Hammer, Ø, Harper, D. A. T. and Ryan, D., 2001. PAST: Paleontological Statistics

Software Package for Education and Data Analysis. Palaeontologia Electronica 4: 9.

Available from: http://palaeo-electronica.org/2001_1/past/issue1_01.htm.

Hansen, A., Pate, J. S. and Hansen, A. P., 1992. Growth, reproductive performance and

resource allocation of the herbaceous obligate seeder Gompholobium marginatum R.Br.

(Fabaceae). Oecologia 90: 158 - 166.

Harley, D. K. P., 2006. A role for nest boxes in the conservation of Leadbeater's possum

(Gymnobelideus leadbeateri). Wildlife Research 33: 385 - 395.

Harper, M. J., McCarthy, M. A. and Van Der Ree, R., 2005. The use of nest boxes in

urban natural vegetation remnants by vertebrate fauna. Wildlife Research 32: 509 - 516.

Harper, M. J., McCarthy, M. A., van der Ree, R. and Fox, J. C., 2004. Overcoming bias

in ground-based surveys of hollow-bearing trees using double-sampling. Forest Ecology

and Management 190: 291 - 300.

Harris, J. B. C., Fordham, D. A., Mooney, P. A., Pedler, L. P., Araújo, M. B., Paton, D.

C., Stead, M. G., Watts, M. J., Akçakaya, H. R. and Brook, B. W., 2012. Managing the

long-term persistence of a rare cockatoo under climate change. Journal of Applied

Ecology 49: 785 - 94.

  272  

Havel, J. J., 1975a. Site vegetation mapping in the northern jarrah forest (Darling

Range) I. Definition of site vegetation types. Forests Department Western Australia,

Bulletin 86, Perth, Western Australia.

Havel, J. J., 1975b. Site vegetation mapping in the northern jarrah forest (Darling

Range) II. Location and mapping of site vegetation types. Forests Department Western

Australia, Bulletin 87, Perth, Western Australia.

Havel, J. J., 2000. Ecology of the forests of south-western Australia in relation to

climate and landforms. PhD thesis, Murdoch University, Perth, Western Australia.

Heberle, G., 1997. Timber harvesting of crown land in the south-west of Western

Australia: an historical view with maps. CALMScience 2: 203 - 224.

Hennon, P. E. and Loopstra, E. M., 1991. Persistence of western hemlock and western

redcedar trees 38 years after girdling at Cat Island in southeast Alaska. Research Note

PNW-RN-507, USDA Forest Service.

Higgins, P. J., 1999. Handbook of Australian, New Zealand and Antarctic birds Volume

4: Parrots to Dollarbird. Oxford University Press, Melbourne.

Hobbs, R. J., 1993. Effects of landscape fragmentation on ecosystem processes in the

Western Australian wheatbelt. Biological Conservation 64: 193 - 201.

Hobbs, R. J., 2000. Repair versus despair: hope and reality in ecological management

and restoration. Ecological Management and Restoration 1: 1 - 2.

Hobbs, R. J. and Harris, J. A., 2001. Restoration ecology: repairing the Earth’s

ecosystems in the new millennium. Restoration Ecology 9: 239 - 246.

Hobbs, R. J. and Lambeck, R. J., 2002. Landscape management and restoration: new

models for integrating science and action. Pp. 412 - 430 in Integrating landscape

ecology into natural resource management. Ed by J. Liu and W. W. Taylor. Cambridge

University Press, Cambridge, United Kingdom.

  273  

Hobbs, R. J. and Saunders, D. A. (Ed), 1993. Regenerating fragmented landscapes.

Towards sustainable production and conservation. Springer-Verlag, New York, USA.

Hobbs, R. J. and Saunders, D. A., 2000. Nature conservation in agricultural landscapes:

real progress or moving deckchairs? Pp. 1 - 12 in Nature Conservation 5: Conservation

in Production Landscapes: Managing the matrix. Ed by J. Craig, N. Mitchell and D. A.

Saunders. Surrey Beatty and Sons, Chipping Norton, Sydney, Australia.

Hobbs, R. J., Arico, S., Aronson, J., Baron, J. S., Bridgewater, P., Cramer, V. A.,

Epstein, P., Ewel, J. J., Klink, C. A., Lugo, A. E., Norton, D., Ojima, D., Richardson, D.

A., Sanderson, E. W., Valladares, F., Vilà, M., Zamora, R. and Zobel, M., 2006. Novel

ecosystems: theoretical and management aspects of the new ecological world order.

Global Ecology and Biogeography 15: 1 - 7.

Hofmann, M. E., Hinkel, J. and Wrobel, M., 2011. Classifying knowledge on climate

change impacts, adaptation, and vulnerability in Europe for informing adaptation

research and decision-making: a conceptual meta-analysis. Global Environmental

Change 21: 1106 - 1116.

Hopper, S. D., 2003. An evolutionary perspective on south-west Western Australian

landscapes, biodiversity and fire: a review and management implications. Pp. 9 - 35 in

Fire in ecosystems of south-west Western Australia: impacts and management. Ed by I.

Abbott and N. Burrows. Backhuys Publishers, Leiden, The Netherlands.

Hughes, T. P., Gunderson, L. H., Folke, C., Baird, A. H., Bellwood, D., Berkes, F.,

Crona, B., Helfgott, A., Leslie, H., Norberg, J., Nyström, M., Olsson, P., Österblom, H.,

Scheffer, M., Schuttenberg, H., Steneck, R.S., Tengö, M., Troell, M., Walker, B.,

Wilson, J. and Worm, B., 2007. Adaptive management of the Great Barrier Reef and the

Grand Canyon World Heritage Areas. Ambio 36: 586 - 592.

Hunter, J. E. and Bond, M. L., 2001. In my opinion - Residual trees: wildlife

associations and recommendations. Wildlife Society Bulletin 29: 995 – 999.

  274  

Hutto, R. L., 1995. Composition of bird communities following stand replacing fires in

northen Rocky Mountains (USA) conifer forests. Conservation Biology 10: 1041 –

1058.

International Union for Conservation of Nature (IUCN), 2011. The IUCN red list of

threatened species. 2001 Categories and Criteria (version 3.1). Available from:

http://www.iucnredlist.org/apps/redlist/static/categories_criteria_3_1. Accessed Fri, 16

Dec 2011.

International Union for Conservation of Nature (IUCN) 2013. The IUCN Red List of

Threatened Species. Version 2013.2. Available from: http://www.iucnredlist.org.

Downloaded on 21 November 2013.

Jarmyn, B., 2000. Nest predation of cockatoos in south-west Victoria: with special

reference to the endangered sub-species of red-tailed black cockatoo, Calyptorhynchus

banksii graptongyne. B.Sc. (Hons) thesis, University of Adelaide, South Australia.

Jarvis, N. T., 1981. Western Australia: an atlas of human endeavour, 1829-1979.

Education and Lands and Surveys Departments of Western Australia, Perth, Western

Australia.

Johnson, E. A. and Miyanishi, K., 2008. Testing the assumptions of chronosequences in

succession. Ecology Letters 11: 419 - 431.

Johnson, G. A. and Donato, D., 31 October - 4 November 2005. Avoidance of wildlife

fatalities: hard lessons from the African Sahel in SD05 Sustainable Development

Conference, Alice Springs, Australia. Available from:

http://www.minerals.org.au/__data/assets/pdf_file/0008/10124/Johnson_Graham6C1.pd

f.

Johnstone, R. E., 1997. Current studies on three endemic Western Australian cockatoos.

Eclectus 3: 34 - 35.

  275  

Johnstone, R. E. and Kirkby, T., 1999. Food of the Forest Red-tailed Black Cockatoo

Calyptorynchus banksii naso in south-west Western Australia. Western Australian

Naturalist 22: 167 - 177.

Johnstone, R. E. and Kirkby, T., 2007. Feral European honey bees: a major threat to

cockatoos and other tree hollow users. Western Australian Naturalist 25: 252 - 254.

Johnstone, R. E. and Kirkby, T., 2008. Distribution, status, social organisation,

movements and conservation of Baudin's Cockatoo (Calyptorhynchus baudinii) in

South-west Western Australia. Records of the Western Australian Museum 25: 107 -

118.

Johnstone, R. E. and Kirkby, T., 2009. Birds of the Wungong Dam catchment,

Bedfordale, Western Australia. Western Australian Naturalist 26: 219 - 274.

Johnstone, R. and Kirkby, T., 2010. Carnaby’s food list. Unpublished report. Western

Australian Museum, Perth.

Johnstone, R. E. and Storr, G. M., 1998. Handbook of Western Australian birds.

Volume I - Non-passerines (Emu to Dollarbird). Western Australian Museum, Perth,

Western Australia.

Johnstone, R. E., Johnstone, C. and Kirkby, T., 2004. Report on Carnaby's cockatoo

Calyptorhynchus latirostris breeding season 2003-2004 within the Cataby Project Area.

Prepared for Iluka Resources Limited. Unpublished report.

Johnstone, R. E., Johnstone, C. and Kirkby, T., 2005. Report on Carnaby's Cockatoo

Calyptorhynchus latirostris nest monitoring within the Cataby Project Area breeding

season 2004-2005. Prepared for Iluka Resources Limited. Unpublished report.

Johnstone, R. E., Johnstone, C. and Kirkby, T., 2008. White-tailed black cockatoos on

the northern Swan Coastal Plain (Lancelin-Perth) Western Australia. Report for

(former) Department of the Environment, Water, Heritage and the Arts (DEWHA).

  276  

Johnstone, R. E., Johnstone, C. and Kirkby, T., 2010. Carnaby’s Cockatoo

(Calyptorhynchus latirostris), Baudin’s Cockatoo (Calyptorhynchus baudinii) and the

Forest Red-tailed Black Cockatoo (Calyptorhynchus banksii naso) on the Swan Coastal

Plain (Lancelin-Dunsborough), Western Australia. Studies on distribution, status,

breeding, food, movements and historical changes. Report for the Department of

Planning, Western Australia.

Johnstone, R. E., Kirkby, T. and Sarti, K., 2013a. The breeding biology of the forest

red-tailed black cockatoo Calyptorhynchus banksii naso Gould in south-western

Australia. I. Characteristics of nest trees and nest hollows. Pacific Conservation Biology

19: 121 – 142

Johnstone, R. E., Kirkby, T. and Sarti, K., 2013b. The breeding biology of the forest

red-tailed black cockatoo Calyptorhynchus banksii naso Gould in south-western

Australia. II. Breeding behaviour and diet. Pacific Conservation Biology 19: 143 – 155

Kenneally, K. F., 2002. Carnaby's cockatoos feeding on Liquid Amber. Western

Australian Naturalist 23: 224 - 225.

Koch, A. J., 2008. Errors associated with two methods of assessing tree hollow

occurrence and abundance in Eucalyptus obliqua forest, Tasmania. Forest Ecology and

Management 255: 674 - 685.

Koch, J. M., 2007. Alcoa's mining and restoration process in south Western Australia.

Restoration Ecology 15 (Supplement): S11 - S16.

Koch, J. M. and Hobbs, R., 2007. Synthesis: is Alcoa successfully restoring a jarrah

forest ecosystem after bauxite mining in Western Australia? Restoration Ecology 15

(Supplement): S137 - S144.

Koch, J. M. and Samsa, G.P., 2007. Restoring Jarrah forest trees after bauxite mining in

Western Australia. Restoration Ecology 15 (Supplement): S17 - S25.

  277  

Koenig, S. E., 2001. The breeding biology of black-billed parrot Amazona agilis and

yellow-billed parrot Amazona collaria in Cockpit Country, Jamaica. Bird Conservation

International 11: 205 - 225.

Koenig, S. E., 2008. Black-billed parrot (Amazona agilis) population viability

assessment (PVA): a science-based prediction for policy makers. Ornitologia

Neotropical 19: 135 – 149.

Koenig, S. E., Wunderle (Jr), J. M. and Enkerlin-Hoeflich, E. C., 2007. Vines and

canopy contact: a route for snake predation on parrot nests. Bird Conservation

International, 17: 79 - 91.

Krebs, C. J., 1988. The experimental approach to rodent population dynamics. Oikos

52: 143 - 149.

Krebs, C. J., 1999. Ecological methodology. Addison-Wesley Longman, Menlo Park,

California.

Lambeck, R., 2003. Farming for the future: designing agricultural landscapes for

conservation and production. Pacific Conservation Biology 9: 68 - 82.

Lamont B. B. and Groom P. K., 1998. Seed and seedling biology of the woody-fruited

Proteaceae. Australian Journal of Botany 46: 387 - 406.

Lamont B. B., Olesen J. M. and Briffa P. J., 1998. Seed production, pollinator

attractants and breeding system in relation to fire response — are there reproductive

syndromes among co-occurring proteaceous shrubs? Australian Journal of Botany 46:

377 - 385.

Lamont, B. B., Swanborough, P. W. and Ward, D., 2000. Plant size and season of burn

affect flowering and fruiting of the grasstree Xanthorrhoea preissii. Austral Ecology 25:

268 - 272.

Lee, K. N., 1999. Appraising adaptive management. Conservation ecology 3: 3.

[Online]. Available from: http://www.consecol.org/Journal/vol3/iss2/art3/.

  278  

Lee, J., Finn, H. and Calver, M. C., 2010. Mine-site revegetation monitoring detects

feeding by threatened black-cockatoos within 8 years. Ecological Management and

Restoration 11: 141 - 143.

Lee, J., Finn, H. and Calver, M., 2013a. Feeding activity of threatened black cockatoos

in mine-site rehabilitation in the jarrah forest of south-western Australia. Australian

Journal of Zoology 61: 119 - 131.

Lee, J. G. H., Finn, H. C. and Calver, M. C., 2013b. Ecology of black cockatoos at a

mine-site in the eastern jarrah-marri forest, Western Australia. Pacific Conservation

Biology 19: 76 - 90

Legislative Council of Western Australia, 2010. Western Australia Parliamentary

Debates. Legislative Assembly (Hon Norman Moore), Parliaments of Western

Australia, Western Australia 2223c - 2225a.

Lewis, J. C., 1998. Creating snags and wildlife trees in commercial forest landscapes.

Western Journal of Applied Forestry 13: 97 – 101.

Lindenmayer, D., 2009. Old forests, new perspectives - insights from the mountain ash

forests of the central highlands of Victoria, south-eastern Australia. Forest Ecology and

Management 258: 357 - 365.

Lindenmayer, D. and Fischer, J., 2006. Habitat fragmentation and landscape change,

Island Press, Washington.

Lindenmayer, D. and Franklin, J., 2002. Conserving forest biodiversity: a

comprehensive multiscaled approach, Island Press, Washington.

Lindenmayer, D. B. and Possingham, H. P., 1995. The conservation of arboreal

marsupials in the montane ash forests of the central highlands of Victoria, south-eastern

Australia - VII. Modelling the persistence of Leadbeater's possum in response to

modified timber harvesting practices. Biological Conservation 73: 239 - 257.

  279  

Lindenmayer, D., Cunningham, R. and Donnelly, C., 2002. Effects of forest

fragmentation on bird assemblages in a novel landscape context. Ecological

Monographs 72: 1 – 18.

Lindenmayer, D. B., Cunningham, R. B., Pope, M. L., Gibbons, P. and Donnelly, C. F.,

2000. Cavity sizes and types in Australian eucalypts from wet and dry forest types - a

simple of rule of thumb for estimating size and number of cavities. Forest Ecology and

Management 137: 139 - 150.

Lindenmayer, D. B., Welsh, A., Donnelly, C., Crane, M., Michael, D., Macgregor, C.,

McBurney, L., Montague-Drake, R. and Gibbons, P., 2009. Are nest boxes a viable

alternative source of cavities for hollow-dependent animals? Long-term monitoring of

nest box occupancy, pest use and attrition. Biological Conservation 142: 33 - 42.

Lindenmayer, D. B., Wood, J., McBurney, L., Michael, D., Crane, M., Macgregor, C.,

Montague-Drake, R., Gibbons, P. and Banks, S. C., 2011. Cross-sectional versus

longitudinal research: a case study of trees with hollows and marsupials in Australian

forests. Ecological Monographs 81: 557 - 580.

Long, R. A., MacKay, P., Zielinski, W. J. and Ray, J. C., 2008. Non-invasive survey

methods for carnivores. Island Press, Washington DC, USA. Pp. 385.

Loyn, R. H., 2000. Managing the forest matrix: regrowth forests as bird habitat. Nature

Conservation 5: 111 - 119.

Lunney, D., 2004a. A test of our civilisation: concerning Australia’s forest fauna across

a cultural landscape. Pp. 1 - 22 in Conservation of Australia’s Forest Fauna, 2nd edition.

Ed by D. Lunney. Royal Zoological Society of New South Wales, Mosman, New South

Wales, Australia.

Lunney, D., 2004b. The future of Australia’s forest fauna revisited. Pp. 1059 - 1070 in

Conservation of Australia’s Forest Fauna, 2nd edition. Ed by D. Lunney. Royal

Zoological Society of New South Wales, Mosman, New South Wales, Australia.

  280  

MacMillen, R. E. and Baudinette, R. V., 1993. Water economy of granivorous birds:

Australian parrots. Functional Ecology 7: 704 - 712.

Manning, A. D., Gibbons, P., Fischer, J., Oliver, D. L. and Lindenmayer, D. B., 2013.

Hollow futures? Tree decline, lag effects and hollow-dependent species. Animal

Conservation 16: 395 – 403.

Manning, A. D., Lindenmayer, D. B. and Barry, S. C., 2004. The conservation

implications of bird reproduction in the agricultural "matrix": a case study of the

vulnerable superb parrot of south-eastern Australia. Biological Conservation 120: 367 -

378.

Manning, A. D., Lindenmayer, D. B., Barry, S. C. and Nix, H. A., 2006. Multi-scale site

and landscape effects on the vulnerable superb parrot of south-eastern Australia during

the breeding season. Landscape Ecology 21: 1119-1133.

Manning, A. D., Lindenmayer, D. B., Barry, S. C. and Nix, H. A., 2007. Large-scale

spatial and temporal dynamics of the vulnerable and highly mobile superb parrot.

Journal of Biogeography 34: 289-304.

Marzluff, J. M. and Ewing, K., 2001. Restoration of fragmented landscapes for the

conservation of birds: a general framework and specific recommendations for

urbanizing landscapes. Restoration Ecology 9: 280 - 292.

Marini, M. A., Barbet-Massin, M., Martinez, J., Prestes, N. P. and Jiguet, F., 2010.

Applying ecological niche modelling to plan conservation actions for the red-spectacled

Amazon (Amazona pretrei). Biological Conservation 143: 102 – 112.

Mast, A. R. and Thiele, K. R., 2007. The transfer of Dryandra R.Br. To Banksia L.f.

(Proteaceae). Australian Systematic Botany 20: 63 – 71.

Mattiske Consulting, Pty. Ltd., 2005. Review of the flora and vegetation located in the

Boddington Gold Mine and Hedges Lease Areas. Report prepared for Worsley Alumina

Pty Ltd by Mattiske Consulting Pty. Ltd., Kalamunda, Western Australia.

  281  

Mawson, P. R., 1995. Observations of nectar feeding by Carnaby’s cockatoo

Calyptorhynchus latirostris. Western Australian Naturalist 20: 93 - 96.

Mawson, P. R. and Long, J. L., 1994. Size and age parameters of nest trees used by four

species of parrot and one species of cockatoo in south-west Australia. Emu 94: 149 -

155.

Maxwell, S., Burbidge, A. A. and Morris, K., 1996. The 1996 Action Plan for

Australian Marsupials and Monotremes. Australasian Marsupial and Monotreme

Specialist Group, IUCN Species Survival Commission: Gland, Switzerland.

McCaw, W. L. and Burrows, N. D., 1989. Fire management. Pp. 317 - 334 in Fire in

ecosystems of south-west Western Australia: impacts and management. Ed by I. Abbott

and N. Burrows. Backhuys Publishers, Leiden, The Netherlands.

McCaw, W. L. and Hanstrum, B., 2003. Fire environment of Mediterranean south-west

Western Australia. Pp. 87 - 106 in Fire in ecosystems of south-west Western Australia:

impacts and management. Ed by I. Abbott and N. Burrows. Backhuys Publishers,

Leiden, The Netherlands.

McCaw, L., Cheney, P. and Sneeuwjagt, R., 2003. Development of a scientific

understanding of fire behaviour and use in south-west Western Australia. Pp. 171 - 187

in Fire in ecosystems of south-west Western Australia: impacts and management. Ed by

I. Abbott and N. Burrows. Backhuys Publishers, Leiden, The Netherlands.

McComb, W., Faunt, K., Bradshaw, F. J. and Biggs, P., 1994. Hollow availability for

selected vertebrates in jarrah forests, Western Australia. Pp. 37 in Mapping Potential

Habitat for Vertebrates in Forests of Western Australia: Final report. Ed by W.

McComb (unpublished report), Department of Conservation and Land Management,

Perth, Western Australia.

McIntyre, S. and Hobbs, R. J., 2000. Human impacts on landscapes: matrix condition

and management priorities. Pp. 301 - 307 in Nature Conservation 5: Conservation in

Production Landscapes: Managing the matrix. Ed by J. Craig, N. Mitchell and D. A.

Saunders. Surrey Beatty and Sons, Chipping Norton, Sydney, Australia.

  282  

McKechnie, A. E. and Wolf, B. O., 2010. Climate change increases the likelihood of

catastrophic avian mortality events during extreme heat waves. Biology Letters 6: 253 –

256

McNab, B. K., 2002. The physiological ecology of vertebrates: a view from energetics,

Cornell University Press, New York, USA.

Meine, C., Soulé, M. and Noss, R. F., 2006. “A mission-driven discipline”: the growth

of conservation biology. Conservation Biology 20: 631 - 651.

Menkhorst, P. W., 1984. Use of nest boxes by forest vertebrates in Gippsland:

acceptance, preference and demand. Australian Wildlife Research 11: 255 - 264.

Moore, B., 1993. Tourists, scientists and wilderness enthusiasts: early conservationists

of the south west. Pp. 110 - 111 in Portraits of the south west: aborigines, women and

the environment. Ed by B. de Garis. UWA Press, Crawley.

Morgan, T. and Fernández-Juricic, E., 2007. The effects of predation risk, food

abundance, and population size on group size of brown-headed cowbirds (Molothrus

ater). Ethology 113: 1173 - 1184.

Morley, S., Grant, C., Hobbs, R. and Cramer, V., 2004. Long-term impact of prescribed

burning on the nutrient status and fuel loads of rehabilitated bauxite mines in Western

Australia. Forest Ecology and Management 190: 227 -239.

Munks, S., Wapstra, M., Corkrey, R., Otley, H., Miller, G. and Walker, B., 2007. The

occurrence of potential tree hollows in the dry eucalypt forests of south-eastern

Tasmania, Australia. Australian Zoologist 34: 22 - 36.

Munro, N. T., Lindenmayer, D. B. and Fischer, J., 2007. Faunal response to

revegetation in agricultural areas of Australia: a review. Ecological Management and

Restoration 8: 199 - 207.

  283  

Munro, N. T., Fischer, J., Wood, J. and Lindenmayer, D. B., 2009. Revegetation in

agricultural areas: the development of structural complexity and floristic diversity.

Ecological Applications 19: 1197 - 1210.

Munro, N. T., Fischer, J., Barrett, G., Wood, J., Leavesley, A. and Lindenmayer, D. B.,

2011. Bird's response to revegetation of different structure and floristics - are

“restoration plantings” restoring bird communities? Restoration Ecology 19: 223 - 235.

Nathan, A., Legge, S. and Cockburn, A., 2001. Nestling aggression in broods of a

siblicidal kingfisher, the laughing kookaburra. Behavioral Ecology 12: 716 - 725.

Nelson, J. L. and Morris, B. J., 1994. Nesting requirements of the yellow-tailed black-

cockatoo, Calyptorhynchus Funereus, in Eucalyptus regnans forest, and implications

for forest management. Wildlife Research 21: 267 - 278.

Newmont Mining Corporation, 2007. Newmont Annual Report 2006. Newmont Mining

Corporation, Denver, USA.

Nichols, O. G. and Watkins, D., 1984. Bird utilisation of rehabilitated bauxite minesites.

Biological Conservation 30: 109 - 131.

Nichols, O. G. and Grant, C. D., 2007. Vertebrate fauna recolonisation of restored

bauxite mines - key findings from almost 30 years of monitoring and research.

Restoration Ecology 15 (Supplement): S116 - S126.

Nichols, O. G. and Muir, B., 1989. Vertebrates of the jarrah forest. Pp. 133 - 153 in The

jarrah forest: a complex mediterranean ecosystem. Ed by B. Dell, J. J. Havel and N.

Malajczuk. Kluwer, Dordrecht, Netherlands.

Nichols, O. G. and Nichols, F. M., 2003. Long-term trends in faunal recolonization after

bauxite mining in the jarrah forest of southwestern Australia. Restoration Ecology

11: 261 - 272.

  284  

Ninox Wildlife Consulting, 2003. The vertebrate fauna of the Boddington gold mine,

unpublished report prepared for Boddington Gold Mine (BGM) Management Company

Pty Ltd.

Norman, M. A., Koch, J. M., Grant, C. D., Morald, T. K. and Ward, S. C., 2006.

Vegetation succession after bauxite mining in Western Australia. Restoration Ecology

14: 278 - 288.

Orr, C. H. and Stanley, E. H., 2006. Vegetation development and restoration potential

of drained reservoirs following dam removal in Wisconsin. River Research and

Applications 22: 281 - 295.

Paap, T., 2006. The incidence, severity and possible causes of canker disease of

Corymbia calophylla (Marri) in the southwest of Western Australia. PhD thesis, School

of Biological Sciences and Biotechnology, Murdoch University, Perth, Western

Australia.

Pallas, A. M., Entwisle, D. R., Alexander, K. L. and Cadigan, D., 1987. Children who

do exceptionally well in first grade. Sociology of Education 60: 257 - 271.

Parks, C. G., Bull, E. M. and Filip, G. M., 1995. Using artificial inoculated decay fungi

to create wildlife habitat. Pp. 175 – 177 in Partnerships for sustainable forest ecosystem

management. USDA Forest Service General Technical Report RM-GTR-266. Ed by C.

Aguirre-Bravo, L. Eskew, A. B. Vilal-Salas and C. E. Gonzalez-Vicente, USDA Forest

Service.

Parma, A. M., Amarasekare, P., Mangel, M., Moore, J., Murdoch, W. W., Noonburg,

E., Pascual, M. A., Possingham, H. P., Shea, K., Wilcox, C. and Yu, D., 1998. What can

adaptive management do for our fish, forests, food and biodiversity? Integrative

Biology, Issues, News, and Reviews 1:16 - 26.

Parnaby, H., Lunney, D., Shannon, I. and Fleming, M., 2010. Collapse rates of hollow-

bearing trees following low intensity prescription burns in the Pilliga forests, New

South Wales. Pacific Conservation Biology 16: 209 - 220.

  285  

Pedler, L., 1996. Artificial nest hollows for black cockatoos. Eclectus 1: 13.

Pell, A. S. and Tidemann, C. R., 1997. The impact of two exotic hollow-nesting birds

on two native parrots in savannah and woodland in eastern Australia. Biological

Conservation 79: 145 - 153.

Pepper, J. W., Male, T. D. and Robers, G. E., 2000. Foraging ecology of the South

Australian glossy black-cockatoo (Calyptorhynchus lathami halmaturinus). Austral

Ecology 25: 16 – 24.

Perry, D. H., 1948. Black cockatoos and pine plantations. Western Australian Naturalist

1: 133 - 135.

Pressey, R. L. and Taffs, K. H., 2001. Scheduling conservation action in production

landscapes: priority areas in western New South Wales defined by irreplaceability and

vulnerability to vegetation loss. Biological Conservation 100: 355 - 376.

Pryke, J. S. and Samways, M. J., 2012. Conservation management of complex natural

forest and plantation edge effects. Landscape Ecology 27: 73 – 85.

Pryor, L. D., 1959. Species distribution and association in Eucalyptus. Monographiae

Biologicae 8: 461 - 471.

Pyne, S. J., 2003. Introduction - fire’s lucky country. Pp. 1 - 8 in Fire in ecosystems of

south-west Western Australia: impacts and management. Ed by I. Abbott and N.

Burrows. Backhuys Publishers, Leiden, The Netherlands.

Quinn, G. P. and Keogh, M. J., 2002. Experimental design and data analysis for

biologists. Cambridge University Press, Cambridge, UK.

Rayner, L., Ellis, M. and Taylor, J. E., 2010. Double sampling to assess the accuracy of

ground-based surveys of tree hollows in eucalypt woodlands. Australia Ecology.

Published on-line. DOI: 10.1111/j.1442-9993.2010.02145.x.

  286  

Rayner, G. A., Scott P. R. and Piper, R. J., 1996. Environmental management at the

Boddington Gold Mine—an operation in a multiple land use catchment. Pp. 348 - 363

in Environmental management in the Australian minerals and energy industries. Ed by

D. R. Mulligan. University of New South Wales Press, Sydney, New South Wales.

Recher, H. F., 1996. Conservation and management of eucalypt forest vertebrates. Pp.

339 - 388 in Conservation of faunal diversity in forested landscapes. Ed by R. M.

DeGraff and R. I. Miller. Chapman and Hall, London, UK.

Recher, H. F., 2002. The past, future and present of biodiversity conservation in

Australia. Pacific Conservation Biology 8: 8 - 11.

Recher, H. F., 2004. Eucalypt forest birds: the role of nesting and foraging resources in

conservation and management. Pp. 23 - 35 in Conservation of Australia’s Forest Fauna,

2nd edition. Ed by D. Lunney. Royal Zoological Society of New South Wales, Mosman,

New South Wales, Australia.

Regional Forest Agreement (RFA), 1998. Assessment of mineral and hydrocarbon

resources in the south-west forest region of Western Australia. Joint Commonwealth

and Western Australian Regional Forest Agreement Steering Committee,

Commonwealth of Australia and Western Australian Government.

Reynolds, A. M., 2012. Fitness-maximizing foragers can use information about patch

quality to decide how to search for and within patches: optimal Lévy walk searching

patterns from optimal foraging theory. Journal of The Royal Society Interface 9: 1568 –

1575.

Rhind, S. G., 1998. Ecology of the brush-tailed phascogale in jarrah forest of

southwestern Australia. PhD thesis, Division of Science, Murdoch University, Western

Australia.

Robaldo-Guedes, N. M., 2004. Management and conservation of the large macaws in

the wild. Ornitologia Neotropical 15: 279 - 283.

  287  

Roberts, J. D., Wardell-Johnson, G. and Barendse, W., 1990. Extended descriptions of

Geocrinia vitellina and Geocrinia alba (Anura: Myobatrachidae) from south-western

Australia, with comments on the status of G. lutea. Records of the Western Australian

Museum 14: 427 - 437.

Robinson, A., 1960. The importance of the marri as a food source to south-western

Australian birds. Western Australian Naturalist 7: 109 - 115.

Robinson, A., 1965. Feeding notes on the white tailed black cockatoo. Western

Australian Naturalist 9: 169 - 170.

Rundle, G. E., 1996. History of conservation reserves in the south-west of Western

Australia. Journal of the Royal Society of Western Australia 79: 225 - 240.

Ryan, M. F., 1996. Sub-lethal poisoning of two HEMS eucalypts to induce seedfall: a

case study. North Eastern Mixed Species Silviculture Project Report No. 4, Department

of Natural Resources and Environment.

Samways, M. J. (1995) Insect conservation biology. Chapman and Hall, London.

Saunders, D. A., 1974a. The occurrence of the white-tailed black cockatoo,

Calyptorhynchus baudinii, in Pinus plantations in Western Australia. Australian

Wildlife Research 1: 45 - 54.

Saunders, D. A., 1974b. Subspeciation in the White-tailed Black Cockatoo,

Calyptorhynchus baudinii, in Western Australia. Australian Wildlife Research 1: 55 -

69.

Saunders, D. A., 1977. Red-tailed black cockatoo breeding twice a year in the south-

west of Western Australia. Emu 77: 107 - 110.

Saunders, D. A., 1979. The availability of tree hollows for use as nest sites by white-

tailed black cockatoos. Australian Wildlife Research 6: 205 - 216.

  288  

Saunders, D. A., 1980. Food and movements of the short-billed form of the White-tailed

Black Cockatoo. Australian Wildlife Research 7: 257 - 269.

Saunders, D. A., 1982. The breeding behaviour and biology of the short-billed form of

the white-tailed black cockatoo Calyptorhynchus funereus. Ibis 124: 422 - 455.

Saunders, D. A., 1986. Breeding season, nesting success and nestling growth in

Carnaby's cockatoo, Calyptorhynchus funereus latirostris, over 16 years at Coomallo

Creek, and a method for assessing the viability of populations in other areas. Australian

Wildlife Research 13: 261 - 273.

Saunders, D. A., 1990. Problems of survival in an extensively cultivated landscape: the

case of Carnaby's cockatoo Calyptorhynchus funereus latirostris. Biological

Conservation 54: 277 - 290.

Saunders, D. A. and Dawson, R., 2009. Update on longevity and movements of

Carnaby's black cockatoo. Pacific Conservation Biology 15: 72.

Saunders, D. A. and Ingram, J. A., 1987. Factors affecting survival of breeding

populations of Carnaby's cockatoo Calyptorhynchus funereus latirostris in remnants of

native vegetation. Pp. 249 - 258 in Nature Conservation - The Role of Remnants of

Native Vegetation. Ed by D. A. Saunders, G. W. Arnold, A. A. Burbidge and A. J. M.

Hopkins. Surrey Beatty, Sydney, New South Wales.

Saunders, D., and Ingram, J., 1995. Birds of southwestern Australia: an atlas of changes

in distribution and abundance of the wheatbelt fauna. Pp. 296 in Birds of Eucalypt

Forests and Woodlands: Ecology, Conservation and Management. Ed by A. Keast, H. F.

Recher, H. Ford and D. Saunders. Surrey Beatty, Sydney, New South Wales.

Saunders, D. A. and Ingram, J. A., 1998. Twenty-eight years of monitoring a breeding

population of Carnaby's cockatoo. Pacific Conservation Biology 4: 261 - 270.

Saunders, D. A., Hobbs, R. J. and Margules, C. R., 1991. Biological consequences of

ecosystem fragmentation: a review. Conservation Biology 5: 18 - 32.

  289  

Saunders, D. A., Mawson, P. and Dawson, R., 2011. The impact of two extreme

weather events and other causes of death on Carnaby's black cockatoo: a promise of

things to come for a threatened species? Pacific Conservation Biology 17: 141 - 148.

Saunders, D. A., Rowley, I. and Smith, G. T., 1985. The effects of clearing for

agriculture on the distribution of cockatoos in the southwest of Western Australia. Pp.

309 - 321 in Birds of Eucalypt Forests and Woodlands: Ecology, Conservation and

Management. Ed by A. Keast, H. F. Recher, H. Ford and D. Saunders. Surrey Beatty,

Sydney, New South Wales.

Saunders, D. A., Smith, G. T. and Rowley, I. 1982. The availability and dimensions of

tree hollows that provide nest sites for cockatoos (Psittaciformes) in Western Australia.

Australian Wildlife Research 9: 541 - 556.

Schofield N. J. and Bartle, J. R., 1984. Bauxite mining in the jarrah forest: impact and

rehabilitation: a report. Department of Conservation and Environment, Western

Australia, Bulletin 169. Pp. 55.

Schofield, N. J., Stoneman, G. L. and Loh, I. C., 1989. Hydrology of the jarrah forest.

Pp. 179 - 201 in The jarrah forest: a complex mediterranean ecosystem. Ed by B. Dell,

J. J. Havel and N. Malajczuk. Kluwer, Dordrecht, Netherlands.

Scott, R., 2009. Artificial hollow trial on the south coast. Western Australian Birds

Notes 131: 1 - 2.

Scott, J. K. and Black, R., 1981. Selective predation by White-tailed Black Cockatoos

on fruit of Banksia attenuata containing the seed-eating weevil Alphitopis nivea.

Australian Wildlife Research 8: 421 - 430.

Sedgwick, E. H., 1949. Bird movements in the wheatbelt of Western Australia. Western

Australian Naturalist 2: 25 - 33.

Shah, B., 2006. Conservation of Carnaby's Black-Cockatoo on the Swan Coastal Plain,

Western Australia Project Report, Birds Australia Western Australia, Perth, Western

Australia.

  290  

Shearer, B. L., Crane, C. E., Fairman, R. G. and Dunne, C. P., 2009. Ecosystem

dynamics altered by pathogen-mediated changes following invasion of Banksia

woodland and Eucalyptus marginata forest biomes of south-western Australia by

Phytophthora cinnamomi. Australasian Plant Pathology 38: 417 - 436.

Shea, P. J., Laudenslayer (Jr), W. F., Ferrell, G. and Borys, R., 2002. Girdled versus

bark beetle-created Ponderosa pine snags: utilization by cavity-dependent species and

differences in decay rate and insect diversity. Pp. 145 – 153 in Proceedings of the

symposium on ecology and management of dead wood in western forests. General

Technical Report PSW-GTR-181. USDA Forest Service, Albany, California, USA

Shepherd, K. R., 1957. Some aspects of ringbarking in alpine ash stands. Australian

Forestry 21: 70 – 75.

Smith, G. B. and Donato, D. B., 2007. Wildlife cyanide toxicosis - monitoring of

cyanide-bearing tailing and heap leach facilities - compliance with the international

cyanide management code. Pp. 149 - 155 in Australasian Institute of Mining and

Metallurgy Publication Series.

Smith, G. B., Donato, D.B. and Adams, M., 2010. Interim report: identification and

development of protective mechanisms at the residual disposal area: wildlife cyanosis

risks and the International Cyanide Management Code; Newmont Boddington Gold and

Donato Environmental Services.

Smith, G. B., Donato, D. B., Gillespie, C. G., Griffiths, S. R. and Rowntree, J. R., 2007.

Ecology of a goldmining tailings facility in the Eastern Goldfields of Western Australia:

a case study. International Journal of Mining, Reclamation and Environment 22: 154 -

173.

Smith, G. T. and Saunders, D. A., 1986. Clutch size and productivity in three sympatric

species of cockatoo (Psittaciformes) in the south-west of Western Australia. Australian

Wildlife Research 13: 275 - 285.

  291  

Smith, P., 2004. Artificial hollows and their use in the rehabilitation of sand mined

areas on North Stradbroke Island: progress report June 2004. Consolidated Rutile

Limited, Dunwich.

Spring, D. A., Bevers, M., Kennedy, J. O. S. and Harley, D., 2001. Economics of a nest-

box program for the conservation of an endangered species: a reappraisal. Canadian

Journal of Forest Research 31: 1992 - 2003.

Stoneman, G. L., 2007. Ecological forestry and eucalypt forests managed for wood

production in south-western Australia. Biological Conservation 137: 558 - 566.

Stoneman, G. L., Rayner, M. E. and Bradshaw, F. J., 1997. Size and age parameters of

nest trees used by four species of parrot and one species of cockatoo in South-Western

Australia: Critique. Emu 97: 94 - 96.

Storr, G. M. and Johnstone, R. E., 1988. Birds of the Swan Coastal Plain and adjacent

seas and islands. Records of the Western Australian Museum Supplement 28: 1 - 76.

Snyder, N. F. R., Derrickson, S. R., Beissinger, S. R., Wiley, J. W., Smith, T. B., Toone,

W. D. and Miller. B., 1996. Limitations of captive breeding in endangered species

recovery. Conservation. Biology l0: 338 – 348.

Snyder, N., McGowan, P., Gilardi, J. and Grajal, A. (Eds.), 2000. Parrots - Status

Survey and Conservation Action Plan 2000–2004. IUCN, Gland, Switzerland and

Cambridge, UK. x + 180 pp.

Snyder, N. F. R., Wiley, J. W. and Kepler, C. B., 1987. The parrots of Luquillo: natural

history and conservation of the Puerto Rican Parrot. Western Foundation of Vertebrate

Zoology, Los Angeles.

Thackway, R. and Cresswell, I. D., 1995. An interim biogeographic regionalisation for

Australia: a framework for setting priorities in the National Reserves System

Cooperative Program, Version 4.0. Australian Nature Conservation Agency, Canberra.

  292  

Treves, A., Mwima, P., Plumptre, A. J. and Isoke, S., 2010. Camera-trapping forest-

woodland wildlife of western Uganda reveals how gregariousness biases estimates of

relative abundance and distribution. Biological Conservation 143: 521 - 528.

Tscharntke, T., Tylianakis, J. M., Rand, T. A., Didham, R. K., Fahrig, L., Batáry, P.,

Bengtsson, J., Clough, Y., Crist, T. O., Dormann, C. F., Ewers, R. M., Fründ, J., Holt,

R. D., Holzschuh, A., Klein, A. M., Kleijn, D., Kremen, C., Landis, D. A., Laurance,

W., Lindenmayer, D., Scherber, C., Sodhi, N., Steffan-Dewenter, I., Thies, C., van der

Putten, W. H. and Westphal, C., 2012. Landscape moderation of biodiversity patterns

and processes - eight hypotheses. Biological Reviews 87: 661 - 685.

Toyne, E. P., Jeffcote, M. T. and Flanagan, J. N. M., 1992. Status, distribution and

ecology of the White-breasted Parakeet Pyrrhura albipectus in Podocarpus National

Park, southern Ecuador. Bird Conservation International 2: 327 – 339.

Tyre, A. J., Tenhumberg, B., McCarthy, M. S. and Possingham, H. P., 2000. Swapping

space for time and unfair tests of ecological models. Austral Ecology 25: 327 - 331.

Underwood, R. J. and Murch, J. H., 1984. Hygienic logging in the northern jarrah forest

[Western Australia; fungal eucalypt dieback]. Australian Forestry 47: 39 - 44.

U.S. Fish and Wildlife Service, 2013. Thick-billed Parrot (Rhynchopsitta

pachyrhyncha) Recovery Plan Addendum. U.S. Fish and Wildlife Service, Southwest

Region. Albuquerque, New Mexico.

Valentine, L. and Stock, W., 2008. Food resources of Carnaby’s Black-Cockatoos in the

Gnangara Sustainability Study Area. Technical report for the Forest Products

Commission (Perth, Western Australia) in support of the Gnangara Sustainability

Strategy (GSS). Available from: http://ro.ecu.edu.au/snsc_papers/.

Vallée, D., 1992. Environmental impacts of gold-mining in Podocarpus National Park,

southern Ecuador. MSc thesis, Imperial College.

  293  

Vesk, P. A. and Mac Nally, R., 2006. The clock is ticking—revegetation and habitat for

birds and arboreal mammals in rural landscapes of southern Australia. Agriculture,

ecosystems & environment 112: 356 - 366.

Vesk, P. A., Nolan, R., Thomson, J. R., Dorrough, J. W. and MacNally, R., 2008. Time

lags in provision of habitat resources through revegetation. Biological Conservation

141: 174 - 186.

Vitousek, P. M., Mooney, H. A., Lubchenco, J. and Melillo, J. M., 1997. Human

domination of Earth’s ecosystems. Science 277: 494 - 499.

Wali, M. K., 1999. Ecological succession and the rehabilitation of disturbed terrestrial

ecosystems. Plant and Soil 213: 195 - 220.

Walker, L. R., Walker, J. and Hobbs, R. J. (Ed), 2007. Linking restoration and

ecological succession. Springer, New York, USA.

Walker, L. R., Wardle, D. A., Bardgett, R. D. and Clarkson, B. D., 2010. The use of

chronosequences in studies of ecological succession and soil development. Journal of

Ecology 98: 725 - 736.

Wallace, W. R., 1965. Fire in the jarrah forest environment. Journal of the Royal

Society of Western Australia 49: 33 - 44.

Walters, C. J. and Holling, C. S., 1990. Large-scale management experiments and

learning by doing. Ecology 71: 2060 - 2068.

Warburton, L. S. and Perrin, M. R., 2005. Conservation implications of the drinking

habits of black-cheeked lovebirds Agapornis nigrigenis in Zambia. Bird Conservation

International 15: 383 - 396.

Ward, S. C., Koch, J. M. and Nichols, O. G., 1990. Bauxite mine rehabilitation in the

Darling Range, Western Australia. Proceedings of the Ecological Society of Australia

16: 557 - 565.

  294  

Wardell-Johnson, G. and Horwitz, P., 1996. Conserving biodiversity and the

recognition of heterogeneity in ancient landscapes: a case study from south-western

Australia. Forest Ecology and Management 85: 219 - 238.

Wardell-Johnson, G. and Roberts, J. D., 1991. The survival status of the Geocrinia

rosea (Anura: Myobatrachidae) complex in riparian corridors: biogeographical

implications. Nature Conservation 2: The Role of Corridors. Pp. 167 - 175.

Wardell-Johnson, G., Calver, M. C., Saunders, D. A., Conroy, S. and Jones, B. A.,

2004. Why the integration of demographic and site-based studies of disturbance is

essential for the conservation of jarrah forest fauna. Pp. 394 - 417 in Conservation of

Australia’s Forest Fauna, 2nd edition. Ed by D. Lunney. Royal Zoological Society of

New South Wales, Mosman, New South Wales, Australia.

Webala, P. W., Craig, M. D., Law, B. S., Wayne, A. F. and Bradley, J. S., 2010. Roost

site selection by southern forest bat Vespadelus regulus and Gould's long-eared bat

Nyctophilus gouldi in logged jarrah forests; south-western Australia. Forest Ecology

and Management 260: 1780 - 1790.

Webala, P. W., Craig, M. D., Law, B. S., Armstrong, K. N., Wayne, A. F. and Bradley,

J. S., 2011. Bat habitat use in logged jarrah eucalypt forests of south-western Australia.

Journal of Applied Ecology 48: 398 - 406.

Western Australia Department of Environment and Conservation (DEC), 2012.

Conservation codes for Western Australian flora and fauna. Available from:

http://www.dec.wa.gov.au/content/view/852/2010/1/1/.

Western Australia Threatened Species Scientific Committee, 2012. Current threatened

and priority fauna rankings. Available from:

http://www.dec.wa.gov.au/content/view/852/2010/. Accessed Mon, 19 Mar 2012.

Weerheim, M. S., 2008. Distribution patterns and habitat use of black cockatoos

(Calyptorhynchus spp) in modified landscapes in the south-west of Western Australia.

MSc thesis, Edith Cowan University, Perth, Western Australia.

  295  

Wells, J. V., Rosenberg, K. V., Dunn, E. H., Tessaglia-Hymes, D. L. and Dhondt, A. A.,

1998. Feeder counts as indicators of spatial and temporal variation in winter abundance

of resident birds. Journal of Field Ornithology 69: 577 - 586.

Whelan, R. J., Collins, L. and Loemker, R., 2009. Predicting threatened species

responses to fuel reduction for asset protection. Proceedings of the Royal Society of

Queensland 115: 77 - 83.

Whisenant, S.G., 1999. Repairing damaged wildlands: a process- orientated, landscape-

scale approach. Cambridge University Press, Cambridge, United Kingdom.

White, N. E., Phillips, M. J., Gilbert, M. T. P., Alfaro-Núñez, A., Willerslev, E.,

Mawson, P. R., Spencer, P. B. S. and Bunce, M., 2011. The evolutionary history of

cockatoos (Aves: Psittaciformes: Cacatuidae). Molecular Phylogenetics and Evolution

59: 615 - 622.

White (Jr), T. H., Collar, N. J., Moorhouse, R. J., Sanz, V., Stolen, E. D. and

Brightsmith, D. J., 2012. Psittacine reintroductions: common denominators of success.

Biological Conservation 148: 106 – 115.

Whitford, K. R., 2001. Dimensions of tree hollows used by birds and mammals in the

jarrah forest: improving the dimensional description of potentially usable hollows.

CALMScience 3: 499 - 511.

Whitford, K., 2002a. Forest hollows: wildlife homes. Landscope 17: 20 - 27.

Whitford, K. R., 2002b. Hollows in jarrah (Eucalyptus marginata) and marri (Corymbia

calophylla) trees I: hollow sizes, tree attributes and ages. Forest Ecology and

Management 160: 201 - 214.

Whitford, K. and Stoneman, G., 2004. Management of tree hollows in the jarrah

Eucalyptus marginata forest of Western Australia. Pp. 807 - 829 in Conservation of

Australia’s Forest Fauna, 2nd edition. Ed by D. Lunney. Royal Zoological Society of

New South Wales, Mosman, New South Wales, Australia.

  296  

Whitford, K. R., Stoneman, G. L., Freeman, I. A., Reynolds, M. J. and Birmingham, T.

C., 1995. Mortality of Eucalyptus marginata (jarrah) and E. calophylla (marri) trees

following stem injection: effects of herbicide, dose, season, and spacing of injections.

Australian Forestry 58: 172 – 178.

Whitford, K. R. and Williams, M. R., 2001. Survival of jarrah (Eucalyptus marginata

Sm.) and marri (Corymbia calophylla Lindl.) habitat trees retained after logging. Forest

Ecology and Management 146: 181 - 197.

Whitford, K. R. and Williams, M. R., 2002. Hollows in jarrah (Eucalyptus marginata)

and marri (Corymbia calophylla) trees II. Selecting trees to retain for hollow dependent

fauna. Forest Ecology and Management 160: 215 - 232.

Williams, J. B., Withers, P. C., Bradshaw, S. D. and Nagy, K. A., 1991. Metabolism

and water flux of captive and free-living Australian parrots. Australian Journal of

Zoology 39: 131 - 142.

Williams, J. E. and Woinarski, J. C. Z. (Ed), 1997. Eucalypt ecology: individuals to

ecosystems. Cambridge University Press, Cambridge, UK.

Williamson, J., Moore, S. and Warren, C., 2005. 1919 to 1935: a pivotal period for the

forests of the south west of Western Australia. Pp. 503 - 513 in A forest conscienceness:

Proceedings 6th National Conference of the Australian Forest History Society Inc. Ed

by M. Calver et al., Millpress Science Publishers, Rotterdam.

Wilson, P. R., Karl, B. J., Toft, R. J., Beggs, J. R. and Taylor, R. H., 1998. The role of

introduced predators and competitors in the decline of kaka (Nestor meridionalis)

populations in New Zealand. Biological Conservation 83: 175 - 185.

Worsley Alumina Pty Ltd, 1985. Worsley Alumina project: flora and fauna studies.

Phase two. Worsley Alumina, Collie, WA.

Worsley Alumina Pty. Ltd., 1999. Worsley Alumina Boddington Gold Mine Project.

Flora and Fauna Studies. Worsley Alumina Pty. Ltd., Perth, Western Australia.

  297  

Wright, T. F., Toft, C. A., Enkerlin-Hoeflich, E., Gonzalez-Elizondo, J., Albornoz, M.,

Rodríguez-Ferraro, A., Rojas-Suárez, F., Sanz, V., Trujillo, A., Beissinger, S. R.,

Vicente Berovides, A., Xiomara Gálvez, A., Brice, A. T., Joyner, K., Eberhard, J.,

Gilardi, J., Koenig, S. E., Stoleson, S., Martuscelli, P., Michael Meyers, J., Renton, K.,

Rodríguez, A. M., Sosa-Asanza, A. C., Vilella, F. J. and Wiley, J.W., 2001. Nest

poaching in neotropical parrots. Conservation Biology 15: 710 – 720.

Zar, J. H. 2010. Biostatistical Analysis (5th edition). Pearson Prentice Hall, New Jersey,

USA.

BGM BLACK COCKATOO SURVEY FORM CS # June 2008 Revision 5.0 HCF/JL Entered:

298

Appendix I

Date

Time Start

5 10 Observed from?

Vehicle Foot Sight Sound

Other

Day Time End

Location & Description Weather

Latitude [50H] Temp

clear overcast fog/mist light rain

heavy rain light wind medium wind heavy wind

Longitude [UTM]

Elevation

Species

FRTBC Carnaby Baudin Unknown Probable Other

Comp Size FU

Vegetation Used Microhabitat

M P marri jarrah wandoo Eucalypt sp. Persoonia M deadstag J deadstag

Banksia Grevillea Hakea Dryandra grass tree U deadstag Capex

sheoak pine grubs grass shrub unknown other:

canopy open branch floor

bush/shrub hollow other

F P J P Estimate U P Min Environmental factors Unknown Max Infrastruct.

Vehicles Personnel Road Log track Log activity Bees Burn scar Bark burn Land clear

Plantation Rehab veg Rem veg Paddo veg Dieback Sump/Pond WSR RDA Other:

Habitat forest woodland heath swamp rehab paddock mod-land other:

# / % obs Best All? Y

N ?

Height Use dead branch

Y N Landscape Position

<10m 10-20m >20m E:

Tree Dead/Alive

D A streambed swamp valley floor swamp edge hilltop

ridge lower-slope mid-slope upper-slope mod-land

Edge Habitat

Hum Nat

Distance to edge

Dieback area Y N Activity Circle: PDA Check: Non-PDA Underline: after scan Behaviours Circle: scan Check: after scan

Roost-short-term Roost-rest Roost-overnight Feed Fly Preen/Allo-preen Feed juvenile Drink Socialise Nest/Inspt Hollow Unknown

Behaviours observed roost alone roost proximate self-preen allo-preen M preen F F preen M preen other sex preen same sex A/J preen

bush/shrub forage ground forage consume marri consume jarrah consume grub consume other: branch/stick break clean beak other feed:

M feed juvenile F feed juvenile juvenile plea male display other social: avoid predator drink vigilance/alert squabble/tiff inspect hollow

Vocalisations Y N Contact Juvenile plea Male-display

Chatter Alarm Other

Disturbance Y N ? 0 none 1 moved 1-10m 2 1-2 fly off 3 group fly off

Veg. Notes

marri jarrah wandoo sheoak Persoonia

Gastrolobium Xanthorrhoea Macrozamia Dryandra Banksia

Hakea Shrub/ Herb Fern Wattle Melaleuca

Flag tree

Husks Feed Notes Feed tree?

Y N

Clr:

Grey Roan Green Red Orange

Y N

other Multiple feed? Y N ?

PHOTOS Frames Blank(s) Photo Notes

Vocalisation Notes Recordings Y N

Y N

Aggregation Y N ? SG

Resight

Y N ? Different location Different members

Method Dawn / Dusk Sector Search (foot /vehicle) FF # Other birds/wildlife:

Site Survey FF Other: (ad-hoc)

Notes Arrivals:

Departures:

BGM BLACK COCKATOO SURVEY FORM CS # June 2008 Revision 5.0 HCF/JL Entered:

299

  300  

Appendix II

Questionnaire (survey form) for Chapter 8 – A Survey of the Users of Artificial Nest

Hollows in Western Australia

Black Cockatoo Artificial Nest Hollow Research Questions

Set up and inspection of the ANH

1) When were the ANH set up?

2) How long have they been up? Were any taken down? Why and how long after? Did they

fall?

3) How were they set up (i.e. ladder, elevated work platform)? What mechanism was used

to attach it?

4) How often were they checked and how was this done (e.g. video/camera, ladder, EWP,

ground-based)?

Design of the ANH and materials used

5) What materials were used to build ANH? (Draw rough sketch) If the ANH was of a

wooden design – was the wood used treated or untreated?

6) What design did they use? Dimensions of interior/exterior and entrance hole?

Maintenance of ANH and costs

7) What efforts did you do to maintain the ANH? - How long and how often were they

maintained?

8) How much did it cost to make, set-up and maintain a single ANH (estimation)?

Location of the ANH and tree characteristics

9) What was the general area in which they were placed (e.g. geography, paddocks,

housing)?

10) What species of trees was used? How were they selected? (I.e. what characteristics?)

Was the tree dead or alive?

11) Where in the tree were they placed – trunk/branch/fork? Height from ground?

  301  

12) If more than one ANH was set up, were they regularly spaced out in the landscape or

clumped?

13) How far apart were they set up from another ANH or a tree with a natural hollow?

14) Were any ANH placed on the same tree (black cockatoos and other species)?

Use of the ANH by black cockatoos and other species

15) Were any of the ANH used by black cockatoos – which species?

16) Were the ANH used by species other then black cockatoos? If so, what species?

17) How long after setting up did black cockatoos or other species start inspecting or using

the ANH?

18) If black cockatoos were observed going in and out, was any evidence that eggs were

laid/hatched or chicks fledged?)

19) Did any pair return to the same ANH or another one in close proximity in subsequent

breeding seasons?

20) Were there any cases of injury or fatality associated with ANH? (e.g. gnawing and

swallowing plastic pieces)

21) Did you notice anything unusual, in terms of behaviour, when black cockatoos use

ANH compared to using natural hollows?

22) Did black cockatoos damage or manipulated ANH or parts of it?

Other outcomes

23) Were there any cases of ANH-related predation? By what predatory species? Did you

notice any changes in numbers or presence of predators in the surrounding area? Were

predation levels different to that associated with natural hollows?

24) Were feral bees observed in or present near ANH?

25) Did you personally have or do you know anyone with any issues associated with the

manufacturing, transportation, installation, maintenance and monitoring of the ANHs?

26) In your personal opinion, do you think ANHs are a good, alternative long-term source

of nesting for black cockatoos?