a thesis submitted for the fulfilment for the requirements

226
Hollow-Bearing Trees as a Habitat Resource along an Urbanisation Gradient Author Treby, Donna Louise Published 2014 Thesis Type Thesis (PhD Doctorate) School Griffith School of Environment DOI https://doi.org/10.25904/1912/1674 Copyright Statement The author owns the copyright in this thesis, unless stated otherwise. Downloaded from http://hdl.handle.net/10072/367782 Griffith Research Online https://research-repository.griffith.edu.au

Upload: others

Post on 03-Jan-2022

4 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: A thesis submitted for the fulfilment for the requirements

Hollow-Bearing Trees as a Habitat Resource along anUrbanisation Gradient

Author

Treby, Donna Louise

Published

2014

Thesis Type

Thesis (PhD Doctorate)

School

Griffith School of Environment

DOI

https://doi.org/10.25904/1912/1674

Copyright Statement

The author owns the copyright in this thesis, unless stated otherwise.

Downloaded from

http://hdl.handle.net/10072/367782

Griffith Research Online

https://research-repository.griffith.edu.au

Page 2: A thesis submitted for the fulfilment for the requirements

Hollow-bearing Trees as a Habitat Resource

along an Urbanisation Gradient

Donna Louise Treby

MPhil

(The University of Queensland)

Environmental Futures Centre.

Griffith School of Environment, Griffith University, Gold Coast.

A thesis submitted for the fulfilment for the requirements of the degree of

Doctor of Philosophy.

December 2013.

Page 3: A thesis submitted for the fulfilment for the requirements

i

“If we all did the things we are capable of doing.

We would literally live outstanding lives.

I think; if we all lived our lives this way, we would truly create an amazing world.”

Thomas Edison.

Page 4: A thesis submitted for the fulfilment for the requirements
Page 5: A thesis submitted for the fulfilment for the requirements
Page 6: A thesis submitted for the fulfilment for the requirements

iv

Acknowledgements:

It would be remiss of me if I did not begin by acknowledging my principal supervisor Dr Guy

Castley, for the inception, development and assistance with the completion of this study.

Your generosity, open door policy and smiling face made it a pleasure to work with you. I

owe you so much, but all I can give you is my respect and heartfelt thanks. Along with my

associate supervisor Prof. Jean-Marc Hero their joint efforts inspired me and opened my mind

to the complexities and vagaries of ecological systems and processes on such a large scale.

To my volunteers in the field, Katie Robertson who gave so much of her time and help in the

early stages of my project, Agustina Barros, Ivan Gregorian, Sally Healy, Guy Castley,

Katrin Lowe, Kieran Treby, Phil Treby, Erin Wallace, Nicole Glenane, Nick Clark, Mark

Ballantyne, Chris Tuohy, Ryan Pearson and Nickolas Rakatopare all contributed to the

collection of data for this study. A huge thanks also goes out to my research assistants Clay

Simpkins, Greg Lollback and James Bone for carrying all the gear and following me around

in the bush at night; we shared many laughs and I appreciated your company on some of

those ‘tough’ or ‘itchy’ nights. Dr Greg Lollback, thank you for your patience and assistance

with ‘R’ and statistical analysis in general. Thank you to Professor Clyde Wild and Michael

Arthur for statistical support. I would also like to thank Greg Ford from Balance

Environmental for helping me with my many questions about identifying bat call frequencies

with Anabat. Thanks also to Kevin Wormington for supplying tree growth data and Michael

Ngugi for tree mortality data.

Diana a.k.a. DD Virkki you have been a pleasure to work with, hang out with, and laugh

with, thank you for being you. You also managed to help me overcome my statistical

dyslexia which will stay with me forever. Special mention must go to Kat Lowe; we have

shared a lot of laughs, fears and tears; but mostly laughs together, thank you for being such a

true friend. To my fellow HDR comrades in Room 3.26, I thank you all for just being there

as we all shared our struggles and frustrations, I wish you all the very best in your future

careers. A heart-felt thank you must go to Dr Clare Morrison for providing valuable

feedback on draft chapters of this thesis. Your ability to reduce my stress levels was

significant (p < 0.05).

Thank you to the many people that supplied me with GIS data. Without it I would not have

been able to complete my study in such fine detail. Chris Woodman at Energex, Peter Dunn

Page 7: A thesis submitted for the fulfilment for the requirements

v

at Power Link, Annie Kelly at DERM, Blair Prince and Rodney Anderson from Gold Coast

City Council. Additional thanks go to Rebecca Castley for helping me make sense of the GIS

layers.

The staff from the Natural Areas Management Unit and the Land for Wildlife team within

Gold Coast City Council was always extremely helpful: Jason Searle, Dr Tim Robson,

Howard Taylor, Wayne Abbott, Steve Vigar, Brett Galloway, Rachael Niotakis, Darryl

Larsen and Lexie Webster. To the many private land owners that allowed open access to

their properties in order to undertake this research I am deeply grateful. Wal and Heather

Mayr, Brit and Clay Badenock, Heather Smart, Jennie and Michael Tippet, Carolyn Reid,

Steve and Karen Dalton, Julia Edwards, Peter Lyons and well as John Palmer and Sharon

Kolka from Gwinganna Lifestyle Retreat, Richard Laws from Gainsborough Greens Golf

Course and Steve Rowell from Jacobs Well Education Centre.

Last, but in no ways least, I cannot thank and praise my husband enough for giving me the

freedom to explore and learn. Phil Treby you are one in a million, I love and admire the man

that you are.

Funding for the project as well as a top up scholarship was generously supplied by Gold

Coast City Council. This study was completed under the following DERM permits: Scientific

Purposes Permits WITK06680910, WISP06680710 and Permit to Collect biological or

geological material from Queensland State Forest, Timber Reserves and other State Lands:

Permit number TWB/01/2010. Ethics approval was obtained from Griffith University

Animal Ethics Committee: Ethics approval number ENV/07/09/AEC to Dr Guy Castley.

In order to complete this thesis I have had to delve, inquire and compare; spending hours,

months and years of my life observing the many changing faces of nature. My work has

forced me to question nature and approach her from a different aspect in order to understand

her secrets; if that is at all possible. Thus, mainly through reading, this journey has led me

into the company of men and women of far greater knowledge than I could ever hope to

acquire.

Page 8: A thesis submitted for the fulfilment for the requirements
Page 9: A thesis submitted for the fulfilment for the requirements

vii

Abstract:

Habitat fragmentation due to rapid urbanisation is occurring globally and results in small and

often isolated patches of remnant bushland. These small, forest remnants contribute to the

conservation of biodiversity within urban areas, as they can, and often do, contain suitable

habitat structures. Habitat structures such as hollow-bearing trees are recognised as

important features of forests globally. Therefore, the abundance of hollow-bearing trees

within a landscape may be a controlling factor for many biota where no other habitat

resources provide a feasible substitute. Therein, this thesis investigates five aspects in

relation to the ecology of hollow-bearing trees as a habitat resource within a rapidly

urbanising landscape.

In the first instance a review of the literature was undertaken. This presents a summary of the

importance of hollow-bearing trees globally, and identifies threats associated with their

conservation. Focus is centred on the importance of this habitat resource within the

Australian context, concluding with a description of the study area and thesis aims. A

hierarchical analysis of the current legislation that exists within Australia to protect hollow-

bearing trees was then undertaken. The evaluation revealed, that despite Federal and State

legislation acknowledging the importance of hollow-bearing trees to biodiversity, there are

insufficient mechanisms in place at all levels of government to halt the decline of hollow-

bearing trees across various landscapes. A subsequent quantitative review of local councils’

policies and websites in regards to biodiversity conservation and specifically the management

of hollow-bearing trees across four eastern Australian states revealed that few (< 5%) had

specific plans in place to protect this habitat resource. While existing legislation identifies

the need to retain ‘forests’, ‘regional ecosystems’, ‘remnant habitat’ or vegetation

communities’, it masks the fact that it is largely ineffective in delivering real on-ground

conservation action for finer scale habitat features such as hollow-bearing trees, particularly

those found within urban landscapes and non-protected forests.

Page 10: A thesis submitted for the fulfilment for the requirements

viii

To gain a greater understanding of the importance of this habitat resource at the local scale

the third Chapter undertook an inventory of hollow-bearing trees along an urbanisation

gradient within a rapidly urbanising landscape on the Gold Coast in south-east Queensland.

A total of 6048 trees from 45 remnant forest sites along the urbanisation gradient were

recorded, of which there were 916 hollow bearing trees containing 2143 hollows. Twenty-

four species of eucalypt were hollow-bearing, with standing dead trees containing the most

hollows (15%). Hollow type was found to be dominated by small (≤ 10 cm) bayonet type

hollows. The stands of remnant forest patches in this study were found to be of an even age,

with 79% of all trees being within the 10-40 cm diameter at breast height (dbh) range. The

urbanisation gradient did not influence hollow type, however, it was found to influence the

numbers of each individual hollow type found. No relationship was found to occur between

environmental variables, disturbance and the urbanisation gradient on the number of hollow-

bearing tree species per site, nor did they have any influence on hollow density per site.

Consistent with previous research, this study confirmed that considerable variation exists in

the availability of hollow-bearing trees across the urban matrix.

The functional ecology of habitat remnants along the urbanisation gradient was then

investigated in Chapter 4. This chapter quantified the impacts of urbanisation, stand density,

landscape variables such as fire and logging history as well as environmental variables on

hollow-bearing tree density within urban remnant forests. Logging history was found to be

the driving factor influencing hollow-bearing tree density along the urbanisation gradient on

the Gold Coast. Historical land-use practices and their effects leave long lasting impacts on

the environment driving habitat structure and function over time. These findings highlight

the significance of incorporating historical land use practices into current and future urban

planning, as these impact on remaining biodiversity values in rapidly changing landscapes

such as those found along urbanisation gradients. These findings, will therefore be of benefit

to managers and planners when making decisions about where, and how to best to manage for

hollow-bearing trees in an urban matrix.

Once the factors influencing the density of hollow-bearing trees in urban remnants had been

identified in Chapter 4 it was then possible to quantify the impact of urbanisation on the

persistence and availability of hollow-bearing trees. Chapter 5 therefore addressed the

persistence of the hollow-bearing tree resources by modelling the influences of certain tree,

disturbance and environmental variables on hollow formation, as there have been very few

Page 11: A thesis submitted for the fulfilment for the requirements

ix

studies investigating these relationships within urban remnant forest patches. The

relationship between dbh and hollow type was also investigated and demonstrated that the

probabilities of trees having hollows increased with dbh and that this pattern was consistent

across all hollow types. These results therefore confirm previous studies highlighting the

value of large trees within the landscape as habitat refugia. Modelling the formation and

persistence of hollow-bearing trees from the recruitment cohort of trees on the Gold Coast

over the long-term (150 years) revealed that ongoing land clearing would have substantial

impact on the hollow-bearing tree resource along the gradient. These findings provide the

foundations for understanding how this resource is likely to persist across the landscape and

supports suggestions for an adaptive management paradigm to be put in place to ameliorate

the expected loss of hollow-bearing trees due to the urbanisation process.

Finally, the richness and relative occupancy of hollow-using vertebrate fauna found within

remnant forest patches along an urban gradient was quantified by using a number of survey

methods. Forty-two species were recorded, representing approximately 33.6% of hollow-

using species that are known to occur within south-east Queensland. The majority of species

observed were classified as common, falling within the ‘urban exploiter’ or ‘urban adapter’

categories. Spotlighting surveys recorded the most species, with birds and mammals being

the most abundant species found along the urbanisation gradient. There was no difference in

the relative occupancy of birds and mammals among sites. There was however, a difference

in relative occupancy of both birds and mammals along the urbanisation gradient, where

control sites had lower occupancy estimates than all other gradient categories. Hollows with

a diameter ≤10 cm were those with the greatest wildlife utilisation and occurred at a height

between 0–5 m. For all taxonomic groups combined site area and tree species diversity were

the most important variables influencing the distribution of hollow-using fauna along the

urbanisation gradient. Common species such as the Australian Owlet Nightjar, squirrel glider,

common brushtail possum and Gould’s wattled bat accounted for 26.9% of the variation in

the relative abundance of fauna among sites. Microbats appeared to be influenced more by

the urbanisation gradient than other fauna as they were more prevalent in control and peri-

urban sites than rural or urban sites. Furthermore, the urbanisation gradient, past logging and

the density of hollow-bearing trees were identified as important drivers of microbat richness

in remnant patches.

Page 12: A thesis submitted for the fulfilment for the requirements

x

To quantify and manage habitat quality for wildlife in modified landscapes it is necessary to

consider the often differing matrix elements found at patch level within the entire landscape.

The preceding findings provide valuable information about the distribution and abundance of

hollow-bearing trees along an urbanisation gradient. Factors that affect this availability are

also highlighted and the potential implications for hollow-using fauna are identified. This

study is the first to quantify variation in hollow-bearing trees along an urbanisation gradient

and has thus improved our understanding of this dynamic system. This thesis has

demonstrated the value of undertaking an inventory of hollow-bearing trees along an

urbanisation gradient in order to understand remnant forest condition and the subsequent

associated hollow-using vertebrate fauna occupying forest remnants. The findings have

important implications for the future conservation of biodiversity in urban landscapes. Urban

planning efforts need to acknowledge all the relevant biodiversity elements (e.g. habitat

features) within the urban landscape. Furthermore, conservation managers and planners need

to recognise that the persistence and functional value of these elements may require future

planning consideration to be in the order of hundreds of years.

Page 13: A thesis submitted for the fulfilment for the requirements

xi

Table of Contents:

Table of Contents: .............................................................................................................

Statement of Access: ........................................................................................................ i

Statement of Contribution by Others: .......................................................................... ii

Statement of Contribution by Others: ......................................................................... iii

Acknowledgements: ....................................................................................................... iv

Statement of Originality: ............................................................................................... vi

Abstract: ........................................................................................................................ vii

List of Figures: ............................................................................................................. xiv

List of Tables: .............................................................................................................. xvii

Chapter 1: An overview of hollow-bearing trees and their conservation and management in urban landscapes. ................................................................................ 1

Introduction ................................................................................................................... 1

Urban forest patches ...................................................................................................... 2

Hollow-bearing trees ..................................................................................................... 4

Tree hollow formation ................................................................................................... 4

Threats to hollow-bearing trees ..................................................................................... 5

Timber harvesting ......................................................................................................... 5

Agriculture .................................................................................................................... 5

Firewood collection....................................................................................................... 6

Altered fire regimes ....................................................................................................... 6

Urbanisation.................................................................................................................. 6

Conservation of hollow-bearing trees ........................................................................... 7

Wildlife use of tree hollows in Australia ...................................................................... 8

Tree-hollow selection by wildlife .................................................................................. 8

The impacts of urbanisation on biodiversity ............................................................... 12

A brief history of human settlement and timber harvesting in south-east Queensland13

Study area .................................................................................................................... 14

Thesis aims and structure ............................................................................................ 17

Page 14: A thesis submitted for the fulfilment for the requirements

xii

Chapter 2: Forest conservation policy implementation gaps: Consequences for the management of hollow-bearing trees in Australia. Conservation and Society 12, 1. 2014.......................................................................................................................................... 20

Abstract ....................................................................................................................... 20

Introduction ................................................................................................................. 21

Methods ....................................................................................................................... 23

Results ......................................................................................................................... 25

Legislation and the conservation of Australian forests and hollow-bearing trees .... 25

Local level ranking of the importance of biodiversity assets, in particular hollow-bearing trees ............................................................................................................................ 30

Discussion ................................................................................................................... 35

Conservation challenges and the way forward .......................................................... 37

Conclusion ................................................................................................................... 39

Chapter 3: Hollow-bearing trees as a habitat resource within an urbanising landscape.......................................................................................................................................... 41

Introduction ................................................................................................................. 41

Methods ....................................................................................................................... 43

Study sites .................................................................................................................... 43

Plot sampling protocol ................................................................................................ 44

Landscape variables .................................................................................................... 49

Urbanisation gradient ................................................................................................. 49

Data analysis ............................................................................................................... 52

Results ......................................................................................................................... 53

Hollow bearing trees ................................................................................................... 53

The impact of environmental variables, disturbance and the urbanisation gradient on hollow-bearing tree species abundance and hollow type ........................................... 62

Discussion ................................................................................................................... 64

Chapter 4: The impacts of landscape variables on hollow-bearing trees along an urbanisation gradient. .................................................................................................. 68

Introduction ................................................................................................................. 68

Methods ....................................................................................................................... 70

Data analysis ............................................................................................................... 75

Results ......................................................................................................................... 76

Discussion ................................................................................................................... 78

Chapter 5: The retention of recruitment trees and time to hollow formation ....... 81

Page 15: A thesis submitted for the fulfilment for the requirements

xiii

Introduction ................................................................................................................. 81

Research aims ............................................................................................................ 83

Methods ....................................................................................................................... 83

Plot sampling protocol ............................................................................................... 83

Growth rates for eucalypt species .............................................................................. 84

Growth rates in the eucalypt communities in urban fragments ................................. 85

Mortality rates for eucalypt trees ............................................................................... 86

Recruitment trees and hollow development ............................................................... 87

Loss of hollow-bearing trees due to habitat loss from land clearing ........................ 88

Diameter at breast height and bark type as a predictor of hollow type .................... 89

Data analysis ............................................................................................................... 90

Univariate determinants of hollow presence by dbh .................................................. 90

Estimating hollow type probability ............................................................................ 90

Tree, disturbance and environmental variables as predictors of a tree containing a hollow .................................................................................................................................... 90

Results ......................................................................................................................... 91

Estimating hollow and hollow type probability ......................................................... 93

Predictors of trees containing hollows ...................................................................... 95

Recruitment trees, hollow development and the influence of land clearing .............. 96

Discussion ................................................................................................................... 98

Factors influencing hollow formation ........................................................................ 98

Hollow-bearing tree recruitment ............................................................................. 100

Management implications of hollow-bearing tree recruitment ............................... 100

Chapter 6: Forest patch occupancy and use of hollow-bearing trees by vertebrates along the urbanisation gradient. ................................................................................ 102

Introduction ............................................................................................................... 102

Methods ..................................................................................................................... 105

Diurnal surveys ......................................................................................................... 109

Spotlighting ............................................................................................................... 109

Stagwatching ............................................................................................................. 110

Anabat surveys .......................................................................................................... 112

Bat call identification ................................................................................................ 112

Pole camera............................................................................................................... 113

Species occupancy ..................................................................................................... 113

Page 16: A thesis submitted for the fulfilment for the requirements

xiv

Statistical analysis ..................................................................................................... 114

Determinants of species richness and relative occupancy of hollow-using vertebrate fauna along an urbanisation gradient ................................................................................. 114

Results ....................................................................................................................... 116

Diurnal surveys ......................................................................................................... 119

Spotlighting ............................................................................................................... 119

Stagwatching ............................................................................................................. 119

Anabat survey ............................................................................................................ 122

Pole camera............................................................................................................... 122

Tree species, hollow type and use ............................................................................. 123

Drivers of hollow-using vertebrate species richness along an urbanisation gradient126

Vertebrate assemblages and relative occupancy of fauna along the urbanisation gradient. ................................................................................................................................... 129

Discussion ................................................................................................................. 131

General summary ...................................................................................................... 131

Tree species, hollow type and use ............................................................................. 132

The impacts of the urbanisation gradient on species richness ................................. 133

The effects of the urbanisation gradient, landscape and environmental variables on fauna ................................................................................................................................... 134

Chapter 7: General Discussion ................................................................................. 135

Policy-implementation gaps ...................................................................................... 137

Contemporary urban planning ................................................................................... 138

Natural resource management: gradients, biotic responses and temporal scales ...... 140

Recommendations and further research .................................................................... 142

List of References: .................................................................................................... 144

List of Appendices: ..................................................................................................... 183

Appendix 1: Hollow using fauna of Australia .......................................................... 184

Appendix 2: Regional ecosystems containing wet and dry sclerophyll communities and their status found on the Gold Coast (Queensland Government 2009)..................... 192

Appendix 3: Site and plot number with GPS co-ordinates. ...................................... 195

Appendix 4: Cadastre, Infrastructure and Regional Ecosystem data used to calculate urbanisation gradient. ................................................................................................ 195

Appendix 5: Model selection .................................................................................... 196

Appendix 6: Tree growth rates ................................................................................. 200

Page 17: A thesis submitted for the fulfilment for the requirements

xv

List of Figures:

Figure 1.1: Map showing the 45 research sites studied within the Gold Coast, Brisbane,

Logan and Redland Shires, south-east Queensland, Australia………………………………15

Figure 2.1: The relative importance placed on the environment by local councils based on

the ranking of ‘environment’ tabs in relation to all other tabs on primary council web

pages…………………………………...…………………………………………..……….32

Figure 3.1: Tree form characteristics and senescence adapted from Whitford (2002). S1.

Mature living tree S2 Mature, living tree with a dead or broken top S3. Living tree with most

branches still intact S4. Living tree with the top 25% broken off S5. Living tree with the top

25-50% broken away S6. Living tree with the top 50-75% broken away S7. A solid dead tree

with 75% of the top broken away S8. Hollow stump……………………………..….……...47

Figure 3.2: Tree hollow types (Lindenmayer et al. 2000)…………………………………48

Figure 3.3: The 250 m buffer placed around sites to capture housing, cadastre and regional

ecosystem data………………………………………………………………………….…...51

Figure 3.4: The number of hollows per hollow size category……………………………..59

Figure 3.5: The relationship between hollow type and size (mean ±S.E). The number of

hollows for each hollow type are; Butt 269, Branch-end 735, Bayonet 914, Fissure 48, Trunk-

main 98, Trunk-top 54, Branch-main 25……………………………………………………60

Figure 3.6: Number of trees surveyed in each dbh size category. All trees represent the

combination of hollow-bearing and non-hollow bearing trees. Results for hollow-bearing

trees are then presented separately………………………………………………………….61

Figure 3.7: The relationship between hollow type and the urbanisation gradient (mean

±S.E.)………………………………………………………………………..………….......62

Page 18: A thesis submitted for the fulfilment for the requirements

xvi

Figure 3.8: Spatial arrangement of sites in the multidimensional space based on the density

of hollow-bearing trees. The urbanisation gradient and site area account for 45.5% (PCO1)

and 33.1% (PCO2) of the variation respectively. Legend: u = urban, r = rural, c = control, p

= peri-urban………………………………………………………………………..….…….63

Figure 5.1: The dbh range of recruitment trees compared to all trees surveyed.….……...91

Figure 5.2: Probability of a tree having a hollow for (a) all hollow types; (b) bayonet; (c)

butt; (d) branch end; (e) fissure; (f) trunk main; (g) trunk top and (h) branch main hollows.

Values represent the number of hollow-bearing trees in each dbh size class

category.…...………………………………………………………………………….…...94

Figure 5.3: The predicted number of trees by dbh category over 10, 50, 100 and 150 years

showing the progression through time of the current cohort of recruitment trees into hollow-

bearing………………………………………………………………………..…………….96

Figure 5.4: Cumulative loss (mean ± SD) of hollow bearing trees from the recruitment

cohort of trees within urban forest patches due to loss of habitat due to land clearing for urban

development and infrastructure over a 150 year time frame…………………….…...……..97

Figure 6.1: The number of bird and mammal species found to occupy all sites as a function

of cumulative survey effort during spotlighting surveys……………………………..……117

Figure 6.2: Trichosurus vulpecula found occupying a Eucalyptus siderophloia

hollow……………………………………………………………..…………..…………..120

Figure 6.3: The eggs of Ninox novaeseelandiae found in a large E. tereticornis

hollow…………………………………………………………………………..…………121

Figure 6.4: Litoria caerulea in water filled hollow…………………………..……….…121

Figure 6.5: The relationship between tree species diversity and hollow-using vertebrate

species richness across all sites……………………………………………………………126

Page 19: A thesis submitted for the fulfilment for the requirements

xvii

Figure 6.6: The relationship between tree species diversity and hollow using vertebrate

species rates of occupancy across all sites………………………………………….………126

Figure 6.7: The amount of variation in species relative abundance along the urbanisation

gradient. The Australian Owlet Nightjar and squirrel glider account for 14.1% (PCO1) while

the common brushtail possum and Gould’s wattled bat accounted for 12.8%

(PCO2)……………………………………………………………………………………..128

Figure 6.8: Occupancy rates of birds and mammals along the urbanisation

gradient……..………………………………………………………………………………129

Page 20: A thesis submitted for the fulfilment for the requirements

xviii

List of Tables:

Table 1.1: Studies undertaken across Australia on the minimum diameter at breast height

(dbh) of trees found to be suitable for hollow development and occupation by arboreal

marsupials………………………………………………………………………….………..10

Table 1.2: A summary of the characteristics of trees used by a range of hollow dependent

taxa in Australia. Modified from Gibbons and Lindenmayer (2002)………………………..11

Table 2.1: Data from the web pages of local councils in relation to how they place

themselves environmentally in a social context. The number of environmental strategies for

each council are those listed within each primary ‘Environment’ tab on council web

pages…………………………………………….…………………...……………………...31

Table 2.2: Councils contacted to request information on strategies and policies targeted to

hollow-bearing trees at the local level, and number of respondents by

state…………………………………………………………………………………………..33

Table 2.3: Specific measures implemented by two urban local councils to protect and

preserve hollow-bearing trees on public and private land in Queensland and New South

Wales………………………………………………………………………………………...34

Table 3.1: Summary of tree, tree-hollow, environmental and disturbance variables and the

urbanisation gradient quantified at each plot………………….…………………………..…46

Table 3.2: A summary of species dominance and hollow-bearing tree prevalence within

eucalypt communities recorded from 45 forest remnants along an urbanisation gradient.

HBT’s = hollow-bearing trees………………………………………………………………54

Table 3.3: A description of hollow-bearing (HBT) tree attributes found across all sites

sorted by gradient then area. Gradient: R= rural, U=urban, P=peri-urban,

C=control………………………………………………………..………………………….56

Page 21: A thesis submitted for the fulfilment for the requirements

xix

Table 3.4: Summary of trees species and hollow type and total number of hollows found

across all sites………………………………………………..……………………………..58

Table 4.1: The urbanisation index, hollow-bearing tree, landscape and environmental

variables quantified at each site………………………………………..…………………...71

Table 4.2: Landscape variables examined in association with hollow-bearing tree density

from 45 urban remnant sites. Values in brackets represent where a site is located along the

urbanisation gradient. Standard errors are supplied for sites with multiple plots. Urbanisation

index: urban = high value, rural = low value…………………………………………..……72

Table 4.3: Environmental variables examined for associations with hollow-bearing trees in

each of the 45 remnant sites. Standard errors are supplied for sites with multiple

plots………………………………………………………………………..………………...74

Table 4.4: Best approximating model comparisons of explanatory landscape and

environmental variables associated with the density of hollow-bearing trees. Akaike models

with ∆i values < 4 are presented. The next model (AICc > 4) demonstrates the distance

between the two best models and the third……………………………….…………………77

Table 4.5: Parameter estimates for the landscape and environmental variables associated

with the density of hollow-bearing trees per hectare across all sites. Variables for each

explanatory analysis are ordered by their relative importance…………………………….....77

Table 5.1: Annual dbh growth rates (cm/yr) in E. pilularis, E. microcorys and E. racemosa

(Wormington and Lamb 1999) and Corymbia citriodora (Ross 1999) in dry sclerophyll

forests of south-east Queensland………………………………………………………..…..84

Table 5.2: Annual incremental dbh growth rates calculated by bark type. Growth rates for

two common species are also provided…………………………………………….……….86

Table 5.3: The probability of a recruitment tree having a hollow as a function of

dbh…………………………………………………………………..……………………...88

Page 22: A thesis submitted for the fulfilment for the requirements

xx

Table 5.4: Risk assessment of sites due to land clearing for urban development and

infrastructure. Risk increases down and across (left–right) the table………..…..………….89

Table 5.5: The classification of the seven available hollow types as used in this

study……………………………………………………………………….……………….89

Table 5.6: The 33 species of recruitment trees sorted by bark type and absolute abundance

in all sites. Bold numbers represent the cumulative total for each bark

type………………...………………………………………..…………….……….……….92

Table 5.7: Model comparisons for explanatory tree, disturbance and environmental variables

for assessing probability of a tree having a hollow. Only models with a ∆i < 2 are

presented………………………………………………………………………..……….…93

Table 5.8: Parameter estimates for explanatory tree, disturbance and environmental

variables used to model the probability of a tree having a hollow. Variables are ordered by

their relative importance………………………………………………………..…….…….95

Table 5.9: Summary of the current cohort of recruitment trees transitioning to hollow-

bearing trees over 150 years from the 45 sites within the study area. The loss of hollow-

bearing trees due to natural mortality are presented as well as the loss of remnant habitat over

time……….…………………………………..……………………………………..……….96

Table 6.1: Area and hollow-bearing tree density of sites used for fauna surveys. Standard

errors are supplied for sites with multiple plots. c = control, p = peri-urban, r = rural, u =

urban. ………………………………………………………............................…….…….106

Table 6.2: Landscape, environment and tree variables quantified at each plot…………...107

Table 6.3: Features that can be used to distinguish the different species of gliders and

possums (Menkhorst and Knight 2001; Lindenmayer 2002)…………………………..…..109

Page 23: A thesis submitted for the fulfilment for the requirements

xxi

Table 6.4: The number of sites where vertebrate species were observed using each survey

method. The total number of sites where a species was found as well as the total number of

individuals recorded (in brackets) is also presented. Relative species occupancy across all

sites is also estimated. Roosting sites or large numbers of individuals recorded at night are

estimates and have a ~ next to their values….………………………………………….…..115

Table 6.5: The number of species recorded during surveys at each site; sorted by site with

the highest number of species found. Some species were found by more than one method.

Therefore, in some instances, taxon totals will be different from method totals. Species

tallies during spotlighting include those observed transit to and from plots……………..…118

Table 6.6: Tree species, hollow type and size that hollow-using fauna were found in, from

all surveys. Highlighted rows represent the highest number of hollow-using species found in

each category. Numbers are a representation of the number of each species found in each

hollow...…………………………………………………………………………………....122

Table 6.7: Best approximating model comparisons of landscape and the urbanisation

gradient variables associated with the number of hollow-using vertebrate species found

across all sites. Only Akaike models with ∆i values ≤ 4 are presented……………....…….124

Table 6.8: Parameter estimates of the fragment, landscape and urbanisation gradient

variables associated with the number of species found at each site. Variables for each

analysis are ordered by their relative importance………….……………………………….125

Page 24: A thesis submitted for the fulfilment for the requirements

1

Chapter 1: An overview of hollow-bearing trees and their conservation and management in urban landscapes.

Introduction

Habitat is the suite of resources and environmental conditions that determine the presence,

survival and reproduction of a population (Caughley and Sinclair 1994). Habitat structures

such as hollow-bearing trees are globally recognised as important features of forests for the

conservation of wildlife (Spies et al. 1988; Newton 1994; Lindenmayer et al. 1997), but also

provide important structural heterogeneity in natural and recently modified landscapes

(Mawson and Long 1994; Wormington et al. 2002; Aitken and Martin 2004). The low

abundance of hollow-bearing trees within a landscape may be a limiting/controlling factor for

many biota as no other habitat resource represents a feasible substitute (Gibbons and

Lindenmayer 2002). Arthropods, reptiles, birds and mammals all utilise hollow-bearing trees

with their reliance inextricably linked to these trees for food, shelter and nesting.

Consequently, species diversity in forests is often strongly linked to this resource (Saunders

et al. 1982; Lindenmayer et al. 1997; Fitzgerald et al. 2003; Tews et al. 2004; Smith et al.

2007).

Australia has approximately 300 vertebrate species that are recognised hollow-users (Gibbons

and Lindenmayer 2002) with the use of hollow-bearing trees by fauna well studied within the

‘natural’ areas of the Australian bush (Fleay 1947; Kitching and Callaghan 1982; Kehl and

Borsboom 1984; Belcher 1995; Comport et al. 1996; Krebs 1998; Lamb et al. 1998;

Lindenmayer et al. 2000a; Gibbons and Lindenmayer 2002; Gibbons et al. 2002; Luck 2002;

Lumsden et al. 2002; Rowston et al. 2002; Fitzgerald et al. 2003; Eyre 2005; Wormington et

al. 2005; Cameron 2006; van der Ree et al. 2006; Smith et al. 2007). Fewer studies, have

investigated these relationships in urban landscapes. Those that have, demonstrate that

urbanisation has a significant impact on the distribution and abundance of hollow-bearing

Page 25: A thesis submitted for the fulfilment for the requirements

2

trees (Harper et al. 2005a; van der Ree and McCarthy 2005; Harper et al. 2008) and that in

some areas, the loss of hollow-bearing trees from the urban landscape threatens the

persistence of biodiversity in forest patches (Harper et al. 2005a). Previous work on hollow-

bearing trees in Australian urban environments is confined to regions within the dry

sclerophyll forests of Victoria (Harper et al. 2005a; van der Ree and McCarthy 2005; Harper

et al. 2008) and New South Wales (Davis et al. 2013). It is uncertain whether the patterns

observed in these regions apply in sub-tropical urban environments which prompted this

investigation. To gain an understanding of the context underpinning this study this chapter

provides a review of urban forest fragments and their role in protecting biodiversity in urban

areas. It focuses on the role of hollow-bearing trees within these environments and threats

posed to this habitat resource focusing on the impacts of the urbanisation gradient.

Urban forest patches

Urban forest patches are described as either all vegetation in a city or town or, if it has been

altered, as a representation of the fauna and structure and floristics of the natural vegetation

that once dominated a site (Webb and Foley 1996). These forest patches can be found in the

vegetation along urban streets, in urban parks, remnant forest patches, abandoned sites and

residential areas. Given the current rate of urbanisation globally, these urban forest areas will

continue to grow in value (De Wet et al. 1998; Alvey 2006). In the past, the arrangement of

urban habitat remnants has not been related to urban wildlife management, but simply

happens to be what remains after the urbanisation process (Lunney and Burgin 2004a).

However, these days urban development will in some cases have strategic plans in place for

the retention of biodiversity (Niemelä 1999; Snep and Opdam 2010). Traditionally, urban

habitats have been considered depauperate in biodiversity, containing high numbers of

introduced pest species and/or generalist species of both flora and fauna. Evidence is now

revealing that urban and suburban forest patch habitats can and often do contain relatively

high levels of biodiversity (McKinney 2002; Cornelius and Hermy 2004; Williams et al.

2005a; Williams et al. 2005b; Alvey 2006; Zhao et al. 2006; Harper et al. 2008; McKinney

2008). Not only can urban habitats support an abundance of native species, they sometimes

have a greater diversity than those found in the surrounding landscape (Jim and Liu 2001;

Cornelius and Hermy 2004; Stewart et al. 2004) and can include endangered species (Alvey

2006).

Page 26: A thesis submitted for the fulfilment for the requirements

3

Therein, a number of studies have found a positive correlation between human population

density and plant, mammal, reptile and amphibian species richness (Balmford et al. 2001;

Araújo 2003) including widespread, narrow range endemics and threatened species

(Balmford et al. 2001). These relationships exist because the ecological factors contributing

to the suitability of an area for human habitation may be similar to the factors which attract

other species to the area (Araújo 2003). Alternatively, direct and indirect human action could

increase overall species richness through the introduction of native and non-native species

and increased landscape heterogeneity (Araújo 2003; Williams et al. 2006). Whether this is

acceptable in relation to biodiversity and conservation is debatable. Constant pressure from

human populations will continue to influence the current and future persistence of indigenous

urban flora and fauna (Lunney and Burgin 2004a) highlighting the need for the retention and

ongoing management of urban forest patches.

Land managers and biologists generally focus on the retention of large forest patches as

valuable biodiversity refugia. Much of this is driven by early landscape ecology theory,

particularly the single large reserves over several small reserves (SLOSS) debate (Tjørve

2010). While the conservation benefits of larger areas is generally accepted, the best design

is possibly a combination of both small and large reserves, as neither will be the better

strategy alone (Tjørve 2010). As highlighted by Daily (1995), Lindenmayer and Fischer

(2006) and Taylor and Goldingay (2009), smaller fragments of natural habitat, such as urban

forest patches were generally dismissed, as they were not considered to contain sufficient

resources to meet the needs of many individual species. Today they are no longer

overlooked, as they can contribute to the conservation of biodiversity within urban areas

(Cowling and Bond 1991; Wood 1993; Godefroid and Koedam 2003; Catterall 2004). The

value of remnant forest patches in urban landscapes depends on their ability to provide

resources for indigenous flora and fauna (Matlack 1993; Godefroid and Koedam 2003;

Harper et al. 2005a; Alvey 2006; Smith et al. 2006). The availability of one such resource,

tree hollows, will limit the distribution and abundance of obligate hollow-bearing tree users

and species richness in general, particularly in Australia (Gibbons and Lindenmayer 2002).

Therefore, the conservation of this resource is paramount to the maintenance of ecosystem

integrity in urban environments (Harper et al. 2005a). The following sections present an

overview of hollow-bearing trees as an important habitat resource, the threats they face and

their importance for wildlife within an Australian context.

Page 27: A thesis submitted for the fulfilment for the requirements

4

Hollow-bearing trees

Tree hollow formation

Within Australia, tree hollow formation can develop in eucalypts at any stage of growth

(Gibbons et al., 2000; Adkins, 2006) but hollows suitable for occupation by vertebrate fauna

in Queensland usually require a tree to be approximately 150 years of age for this to occur

(Ross 1998). Therein, the presence of tree hollows is related to tree age, with the older trees

that are hundreds of years old generally having a greater prevalence of hollows (Saunders et

al. 1982; Whitford 2002; Goldingay 2009; Loyn and Kennedy 2009; Ranius et al. 2009;

Webala et al. 2011). However, individual eucalypt species have differing rates of hollow

formation. For example, Mackowski (1984) suggested that in blackbutt (Eucalyptus

pilularis) hollow formation commenced at approximately 150 years of age, while between

170 and 200 years are required for species such as tallowwood (Eucalyptus microcorys) and

scribbly gum (Eucalyptus racemosa) (Ross 1998). Hollow-bearing trees can also be long

lived, and the Long-Billed Corella, Cacatua tenuirostris have been observed utilising hollows

in Jarrah, Eucalyptus marginata communities of south Western Australia in trees ranging

between 680 and 1333 years (Mawson and Long 1994). Apart from tree age and species, the

formation of hollows is also influenced by diameter at breast height (dbh), tree health, growth

habits, stand features, site productivity, fire and management history (Gibbons and

Lindenmayer 2002; Adkins 2006).

Australia lacks vertebrate species that excavate tree hollows (Gibbons and Lindenmayer

2002; Blakely et al. 2008), many species of parrots (Higgins 1999) and arboreal marsupials

(Gibbons and Lindenmayer 2002) will enlarge existing hollow openings and carry out

internal modifications while nesting. Therefore, for eucalypt species, there are also three

essential preconditions for hollow formation to occur; (i) a tree must be under some type of

physiological stress or subjected to injury, (ii) heartwood decay must be present and (iii) trees

must be of sufficient size to endure once decayed. Physiological weakness will also increase

in a tree with age which will also promote hollow formation (Gibbons and Lindenmayer

2002). Within eucalypts the heartwood is more vulnerable to decay by living organisms that

gain access after the tree has been subjected to injury e.g. fire-scars or limb breakage (Harper

et al. 2005a) allowing fungi followed by invertebrates, usually termites to excavate the

heartwood leaving the living sapwood as the walls of the hollow (Gibbons and Lindenmayer

2002). Fire can also play an integral role in hollow formation by reducing the time taken for

Page 28: A thesis submitted for the fulfilment for the requirements

5

hollow development by approximately 100 years (Inions et al. 1989b). It has been suggested

that fire disturbance alone will influence the number of tree hollows found in a forest and that

this will vary between wet and dry sclerophyll forests (Gibbons and Lindenmayer 2002).

Threats to hollow-bearing trees

The main threats to the retention and longevity of hollow-bearing trees comes from landscape

modifications by humans including forestry, agriculture, firewood collection, urban

development, fragmentation and altered fire regimes (Lamb et al. 1998; Gibbons and

Lindenmayer 2002). Within Australia, the clearing and disturbance of natural vegetation and

subsequent habitat fragmentation is recognised as one of the principle drivers of biodiversity

declines (Williams et al. 2001). Protection for hollow-bearing trees and the species that are

reliant upon them in Australia is supposedly provided for by legislation, policy and strategic

management plans. However, disagreement surrounding forest management has resulted in

the degeneration of national plans as interpreted by states and territories, as well as individual

stakeholders (Musselwhite and Herath 2005).

Timber harvesting

In sites managed for wood production, there has been a global recognition of the reduction in

tree hollows, which in turn leads to reduced biodiversity (Shukla et al. 1990; Poonswad et al.

2005; Holloway et al. 2007; Politi and Hunter 2009; Shearman et al. 2009: Law et al. 2013).

Within Australia, hollow-bearing trees are often considered to be of low commercial value

and are generally believed to suppress forest regeneration (Gibbons and Lindenmayer 2002).

In Queensland, if they are not harvested as saw logs they are removed for firewood or fence

posts (Lamb et al. 1998). The harvesting of timber reduces the availability of tree hollows by

reducing the diversity of hollow types, hollow abundance, altering the spatial arrangement of

tree hollows and also increases the rate of destruction of retained trees (Catterall and

Kingston 1993; Lindenmayer et al. 1997; Smith et al. 2009: Law et al. 2013).

Agriculture

Agriculture is the world’s most extensive form of land use occupying approximately 40% of

the global land surface (Foley et al. 2005) and contributing significantly to deforestation

(Bowen et al. 2007). Deforestation as a result of expanding subsistence agriculture threatens

Page 29: A thesis submitted for the fulfilment for the requirements

6

many natural habitats in developing nations (Politi and Hunter 2009; Shearman et al. 2009),

while across Australia it is estimated that there has been a 80-90% reduction in tree hollow

resources due to commercial agriculture alone (Gibbons and Lindenmayer 2002).

Firewood collection

Firewood collection impacts negatively on forests and biodiversity in general (du Plessis

1995; Williams and Shackelton 2002; Berg and Dunkerley 2004). There has been a

considerable amount of research focusing on the widespread use of firewood throughout

countries on the African and Asian subcontinents (Fox 1984; Shackleton 1993; Vermeulen et

al. 1996; Tabuti et al. 2003). The problems associated with firewood collection, are not

restricted to these developing nations. For example, the Murray Darling Basin of Australia,

an agricultural area of 1 million km2, contains approximately 29% native forest and produces

one third of the 607 million oven dry tonne of firewood burnt annually from its forests (West

et al. 2008).

Altered fire regimes

The influence of natural and management fires on forests and woodlands can impact

biodiversity at the landscape scale. Fire is a frequent and widespread phenomenon within

Australian environments and many vegetation communities have become adapted to fire

(Wardell-Johnson 2000). While fire is an important factor in the hollow-formation process in

eucalypt communities (Adkins 2006), research has found that there is a notable depletion of

hollow-bearing trees in areas that are being burnt under more frequent fire regimes. This is

primarily because trees that were dead before a fire, are destroyed at a higher rate than living

trees (Inions et al. 1989b; Gibbons and Lindenmayer 2002). Dead trees or stags represent a

large proportion of hollow-bearing trees in some forests types (44%) (Eyre and Smith 1997;

Ross 1999), making them susceptible to inappropriate fire regimes.

Urbanisation

Urbanisation dramatically alters the biota of areas that become cities and towns (McKinney

2002; Williams et al. 2006), from the decline and destruction of native grasslands (Williams

et al. 2004), to the loss of forests (Bowen et al. 2007) and a reduction in faunal assemblages

(van der Ree and McCarthy 2005). In addition to the ongoing risk of further habitat loss for

Page 30: A thesis submitted for the fulfilment for the requirements

7

development and the alteration of natural ecological processes, habitat remnants and their

resources also face threats unique to the urban landscape (Zhao et al. 2006; Harper et al.

2008; McKinney 2008). For example, regulatory controls in urban areas may classify a tree

as hazardous if there is potential for harm to people and property and are therefore removed

(Terho 2009). As one of the key focal areas of this research the impacts of urbanisation are

discussed in more detail later in this chapter.

Conservation of hollow-bearing trees

Hollow-bearing trees provide habitat for diverse taxonomic groups and as such has prompted

the recognition of the importance of hollow-bearing trees globally. Conservation efforts take

place within production landscapes (e.g. forestry) (Mazurek and Zielinski 2004; Holloway et

al. 2007; Smith et al. 2009) but also in areas designated as reserves (Simberloff et al. 1992;

Pirnat 2000; Rhodes et al. 2006; Munks 2009). Within Australia, forest managers have long

recognised the need to retain hollow-bearing trees in eucalypt forests. There is, however,

considerable debate surrounding the spatial arrangement of hollow-bearing trees within

production forests (Lindenmayer et al. 1990; Gibbons and Lindenmayer 1996). Early

forestry practices subscribed to the theory that the clumping of trees would be of greater

benefit to wildlife (Queensland Government 1996), while others argued that an even

distribution of trees across the landscape was preferable (Lindenmayer et al. 1990; Gibbons

and Lindenmayer 1996). However, these recommendations may not apply outside of

production forests, highlighting the need for local government bodies to implement programs

aimed at the protection and retention of hollow-bearing trees on commercial, public and

private lands.

It has been suggested that live hollow-bearing trees are of greater importance to wildlife in

Australia, since they are longer lived than standing dead trees (Moloney et al. 2002). Stags

(standing dead trees with hollows) comprise a significant component of the hollow-bearing

tree resource within a landscape, in some cases making up almost 50% of hollow-bearing tree

availability (Moloney et al. 2002). In general, standing dead trees are more likely to contain

hollows than dead trees (Eyre 2005; Harper et al. 2005a). Stags therefore provide a valuable

additional resource, with certain fauna displaying a preference for stags as denning and nest

sites (Lindenmayer et al. 1991a; Eyre 2004; Cameron 2006).

Page 31: A thesis submitted for the fulfilment for the requirements

8

Wildlife use of tree hollows in Australia

The availability of tree hollows in living or dead trees is fundamental to the survival of a

significant number of Australian species across various taxa (Lindenmayer 2002; Rowston et

al. 2002). In Australia 303 native vertebrate species and an additional 10 introduced species

are hollow users. Excluding introduced species, this figure represents 27 terrestrial

amphibian species (13%), 79 reptiles (10%), 114 birds (15%) and 83 mammals (31%)

(Gibbons and Lindenmayer 2002) (Appendix1).

Tree hollows perform important functional roles within the landscape by providing fauna

with microclimates suitable for occupancy (Gibbons and Lindenmayer 2002). Hollows

ameliorate ambient temperatures (Lumsden et al. 1994), collect water (Kitching and

Callaghan 1982), provide nest sites (Goldingay 2009) and forage resources (Mansergh and

Husley 1985) for a wide range of taxa. Despite the high prevalence of hollow dependence

among Australian fauna, quantitative data reporting on hollow use by vertebrate fauna are

scarce. It is difficult to characterise species as obligate hollow-users or those that exploit this

resource opportunistically, as the relationship some species have with tree hollows can

change throughout their distribution depending on climatic conditions (Gibbons and

Lindenmayer 2002). For example, the sugar glider Petaurus breviceps has been found to nest

in piles of timber on the ground (Fleay 1947), under tin sheeting and in thickets of black-

berry (Morrison 1978) as well as hollow-bearing trees (Traill and Lill 1997). This suggests a

lack of available tree hollows in the landscape as well as a certain degree of opportunism by

vertebrate hollow users. The common ringtail possum Pseudocheirus peregrinus regularly

uses tree hollows but also constructs nests or dreys, and is one of the few arboreal marsupials

to do so (Gibbons and Lindenmayer 2002).

Tree-hollow selection by wildlife

Hollow selection by fauna is non-random, and most choose hollows based on site and hollow

attributes (Gibbons et al. 2002; Eyre 2005; van der Ree et al. 2006). Some species only

occupy tree hollows in certain parts of the landscape, while others determine hollow

suitability based on the proximity of predators and conspecifics (Gibbons and Lindenmayer

2002). Furthermore, occupancy of tree hollows is positively correlated with diameter at

breast height (dbh) (Gibbons et al. 2002). Within Australia, there is a positive relationship

between tree diameter and the number of hollows present, with different eucalypt species

Page 32: A thesis submitted for the fulfilment for the requirements

9

having higher or lower rates of tree hollow development (Pausas et al. 1995; Lindenmayer et

al. 2000a). While trees with larger dbh are likely to contain more occupied hollows, they are

less likely to contain a high number of hollows compared with trees from within the mid

diameter class (Gibbons and Lindenmayer 2002). This highlights the variability in dbh

considered suitable for tree hollow formation and occupation by arboreal marsupials,

throughout Australia (Table 1.1). Other factors such as hollow diameter, tree form or degree

of senescence (Table 1.2) have also been shown to influence hollow use (Gibbons et al.

2002). While some species may prefer tall trees in early successional stages, others prefer

trees in the late stages of decay (Lindenmayer et al. 1991a), or those that have already died

(stags) (Eyre 2004; Cameron 2006).

It is important to note, that not all hollows and hollow-bearing trees are used by fauna.

Studies within Australia have found that roughly 50% of all hollows and all hollow-bearing

trees contained evidence of prior hollow occupancy (Saunders et al. 1982; Gibbons and

Lindenmayer 2002). In highly developed regions occupancy levels may be even further

depressed, with less than 10% occupancy being recorded in some urban forests (Ogden

2009). Some species also display site-specific fidelity (Frith 1976; Murphy et al. 2003). For

example, small mammals such as the agile antechinus Antechinus agilis use particular

hollows for mating or nest sites by successive generations (Lazengy-Cohen and Cockburn

1991). Despite the dependence on hollows by Australian fauna, occupancy rates and tree

hollow use by individuals are rarely quantified in the literature.

Page 33: A thesis submitted for the fulfilment for the requirements

10

Table 1.1: Studies undertaken across Australia on the minimum diameter at breast height

(dbh) of trees found to be suitable for hollow development and occupation by arboreal

marsupials.

Study Location Species Minimum

dbh required

Mackowski 1984 New South Wales possums/gliders 100 cm

Inions et al. 1989 Western Australia brush-tailed possum 40-50 cm

Smith and Lindenmayer 1992 Victoria Leadbeater’s possum >50 cm

Gibbons and Lindenmayer 2002 Victoria not specified 30 cm

Harper et al. 2005b Victoria not specified >40 cm

Wormington et al. 2003;

Wormington et al. 2005

Queensland not specified >20 cm

Eyre 2006 Queensland greater glider >20 cm

Smith et al. 2007 Queensland greater glider >50 cm

Koch et al. 2008c Tasmania not specified >56 cm

Page 34: A thesis submitted for the fulfilment for the requirements

11

Table 1.2: A summary of the characteristics of trees used by a range of hollow dependent

taxa in Australia. Modified from Gibbons and Lindenmayer (2002).

Species Variables influencing occupancy Source

Arboreal marsupials Tree shape, number of hollows Lindenmayer et al. 1991a

Leadbeater’s possum Tree size, tree shape, number of hollows, surrounding vegetation

Lindenmayer et al. 1991a

Mountain brushtail possum

Number of hollows, surrounding vegetation type

Lindenmayer et al. 1996

Common brushtail possum Tree diameter, number of hollows Wood and Wallis 1998

Sugar glider Number of fissures Lindenmayer et al. 1991a

Squirrel glider Tree species, tree health, tree diameter, hollow entrance size and dead trees

Rowston 1998;

Beyer et al. 2008

Greater glider Tree size Lindenmayer et al. 1991a

Brush-tailed phascogale Tree diameter, tree health Rhind 1996

Brown antechinus Tree shape Lindenmayer et al. 1991a

Brown antechinus Tree diameter, tree health Cockburn and Lazenby-Cohen 1992

Australian Owlet-nightjar Height of entrance above ground, number of hollows, proximity to other trees with hollows.

Bringham et al. 1998

Barking Owl Tree diameter, diameter of entrance to hollow

Taylor and Kirstin 1999

Gouldian Finch Tree diameter, hollow entrance width, steep slopes, single stemmed trees

Tideman et al. 1992

Striated Pardalote Number of hollows, hollow depth, fire scarring, tree species

Taylor and Haseler 1993

Parrot species generally Dead trees Goldingay 2009

Broad-headed snake Tree health, tree diameter, number of hollows

Webb and Shine 1997

Vertebrate fauna generally number of hollows, tree health, tree diameter

Gibbons et al. 2002

Page 35: A thesis submitted for the fulfilment for the requirements

12

The impacts of urbanisation on biodiversity

Natural systems play an important ecological, social and economic role within suburban

landscapes, providing habitat for wildlife as well as being places for community recreation

and stormwater management (McWilliam et al. 2010). The maintenance of such functional

connectivity across the landscape is central to managing biodiversity within the urban matrix

(Catterall and Kingston 1993; Goldingay and Sharpe 2004a; Damschen et al. 2006). The

urban landscape however, is characterised by a complex mosaic of land uses; with the end

result a combination of developed land and remnant patches of vegetation (Kendle and

Forbes 1997; Taylor and Goldingay 2009; McWilliam et al. 2010). Until recently urban

ecosystems have been largely ignored in terms of rigorous scientific research (McDonnell et

al. 1993; Catterall 2009). However, as urban areas often contain novel combinations of

environmental variables, they will provide a whole suite of new ecological processes

affecting biodiversity (Catterall 2009). Urban regions settled hundreds of years ago support

small proportions of the original fauna and flora with low probabilities of persistence;

subsequently, the urbanisation process is pivotal to the retention of urban biodiversity

(McKinney 2002; van der Ree and McCarthy 2005). It has recently been proposed, that the

urbanisation gradient is more than a scientific model, but is a valid research methodology

when studying urban ecosystems (Qureshi et al. 2013), as urbanisation acts as a powerful

filter by influencing species composition (Brady et al. 2009), highlighting the relationship

between habitat loss and the urbanisation gradient (McKinney 2002).

Faunal species that are heavily affected by urbanisation are those that are sensitive to human

persecution and habitat disturbance and are usually large mammals, avifauna and species that

rely on stands of old forest for food and shelter (Gibbons and Lindenmayer 2002; McKinney

2002). Urban adapters are often referred to as edge species, able to exploit the surrounding

anthropogenic habitat and often attain a biomass greater than those found in natural

surroundings (McKinney 2002; Low 2003). These urban avoiders are often habitat

specialists (McKinney 2002; McKinney 2006). Conversely some species are urban exploiters

and are often totally dependent on human resources for survival and are composed of early

successional species from nearby ecosystems that are well adapted to intensely modified

urban environments (McKinney 2002; Loss et al. 2009). This implies that some species do

not perceive a modified landscape as isolated, but simply another environment in which to

obtain resources (Low 2003; Harper et al. 2008) and have learnt to adapt to, or exploit the

urban environment.

Page 36: A thesis submitted for the fulfilment for the requirements

13

Decisions made in regards to development and management of urban forest patches are often

made without an adequate understanding of the working of natural ecosystems. Invariably

this means that irreversible environmental damage has and will continue to occur (Webb and

Foley 1996). Therefore, in modified urban landscapes where resource availability differs

across the various landscape elements e.g. natural, residential and industrial; human

intervention may be required to ensure that ecological processes continue to operate (Harper

et al. 2008), particularly in regards to timber harvesting and the availability of hollow-bearing

trees.

A brief history of human settlement and timber harvesting in south-east Queensland

Since 1824 south-east Queensland has been appreciated for its eucalypt forests, stands of

hoop pine (Araucaria cunninghamii) and red cedar (Toona ciliate) (Catterall and Kingston

1993; Kowald 1996; Powell 1998a). The discovery of red cedar in the region marked the

beginning of the timber industry and settlement in Queensland (Kowald 1996). By 1839

timber was required to meet the immediate needs for housing and firewood in the greater

Brisbane area and surrounding region (Powell 1998b). Timber soon became an important

export commodity resulting in increased settlement by timber getters particularly from the

Tweed region in New South Wales (Longhurst 1996). In addition to the timber getters,

pastoralists settled on the land and cleared it for grazing.

Due to the rapid settlement and associated deforestation of south-east Queensland, regulation

was soon established in order to better manage and control land clearing practices (Kowald

1996; Powell 1998a). While some regulations for the protection of forest resources were

heeded, it soon became apparent that licensing could not be enforced due to the lack of

manpower. Therefore, by 1879 pressures for the regulation of forests were ignored by the

immediate need for land (Kowald 1996; Powell 1998a). The demand for timber continued

for both local and export markets, with the use of timber for the local construction of homes

increasing after both world wars (Kowald 1996; Ross 1998). By the 1980’s timber

production shifted to plantation pine and the 1990’s saw the attitude towards native forests

shift to one of conservation (Powell 1998a).

Despite the downward trend in hardwood production timbers, large scale clearing continued

throughout Queensland, between 1990 and 1995 the rate of clearing increased by 700%,

Page 37: A thesis submitted for the fulfilment for the requirements

14

driven by the need to clear land for agriculture (Powell 1998a). Within south-east

Queensland, this era also saw greater public concern over the deterioration of the

environment with public disputes arising over the conservation of remnant bushland

(Catterall and Kingston 1993). Today remnant forests in south-east Queensland are protected

in a number of National Parks and World Heritage sites. However, the conservation reserve

system, like most others in the world (Pressey 1994), does not effectively represent or protect

the biodiversity within the region (McAlpine et al. 2007). Many of these national parks and

reserves protect the rainforests of the region with less attention being given to the protection

of hardwood eucalypt forests (McAlpine et al. 2007), particularly in rapidly developing urban

areas (Catterall and Kingston 1993).

Study area

This study was primarily conducted on the Gold Coast, south-east Queensland, Australia.

The Gold Coast is located approximately 70 km south of Brisbane occupying an area of 1451

km2 stretching from the New South Wales border in the south to the Logan River in the north

and from the coastline to the McPherson and Darlington Ranges in the west (Figure 1.1).

The Gold Coast is the seventh largest city in Australia with a rapidly increasing population,

currently estimated at around 500,000 (Pickering et al. 2010). Upon the development of the

first regional plan for south-east Queensland in 2005, the State Government recognised that

urban growth was proceeding in an extraordinarily haphazard manner (Spearbritt 2009).

While the Gold Coast retains more than 45% of its natural habitat (Pickering et al. 2010),

land clearing due to the urbanisation process has resulted in a heterogeneous landscape of

natural habitat remnants, some of very high conservation value, within a variable urbanisation

matrix (Pickering et al. 2010). The Gold Coast shire contains a higher level of remnant

bushland than any other shire in south-east Queensland and has a large number of vegetation

communities represented in the remaining fragments (Catterall and Kingston 1993). Despite

45% of the Gold Coast persisting as native vegetation, only 19% is protected on public land

managed by either Gold Coast City Council or State Government. Private land that is

managed under voluntary conservation agreements accounts for an additional 3% (Gold

Coast City Council 2009).

Page 38: A thesis submitted for the fulfilment for the requirements

15

Figure 1.1: Map showing the 45 research sites studied within the Gold Coast, Brisbane,

Logan and Redland Shires, south-east Queensland, Australia.

Page 39: A thesis submitted for the fulfilment for the requirements

16

This cumulative figure of 22% does not meet with the recommended guidelines for 30%

preservation of all native vegetation to retain an adequate retention of biodiversity within the

shire (Gold Coast City Council 2009). Furthermore, approximately 14% of remaining native

vegetation on the Gold Coast occurs within the urban footprint, with approximately 70% of

this vegetation being unprotected, or on private land (Gold Coast City Council 2009) and

prone to development pressures. The loss of some threatened regional ecosystems, such as

Blackbutt forests, has been particularly severe where there has been more than a 90%

reduction in habitat (Pickering et al. 2010). Projected land use commitments across the city

for future urban, industrial and infrastructure requirements, amount to 44000 ha of native

vegetation that will be cleared between 2009-2019 (Gold Coast City Council 2009). This

will result in a significant reduction in the extent and connectivity of the remaining habitats,

which are important for the retention of urban biodiversity (Koh and Sodhi 2004; Tratalos et

al. 2007; Marzluff and Rodewald 2008).

The Gold Coast comprises a myriad of landscapes including a coastal plain that includes

beaches, dunes, river deltas, bays, estuaries and wetlands; rolling foothills and low mountain

ranges supporting heath-land, melaleuca swamps; wet and dry sclerophyll forest and

rainforest vegetation communities. The topography rises from sea level up to 1010 m on the

mountain escarpments of the hinterland (Gold Coast City Council 2007). The main

geological unit is the Neranleigh-Fernvale beds of Tertiary origin which make up 62.3% of

the shire (Whitlow 2000). The most common soils are the red and yellow Podzolics found on

the hills. These are duplex soils which have a distinct loamy surface-soil and clay sub-soil

which are quite acidic (pH 4.5) and therefore makes them less fertile. The more urbanised

areas of the City would be; metamorphic hills, fine textured alluvia, or dune sands depending

on the elevation, topography and proximity to the ocean. This range of altitudes and soil

types combine to produce a unique set of environmental habitats resulting in the Gold Coast

being a regional biodiversity hotspot (Catterall and Kingston 1993). The average annual

rainfall on the Gold Coast varies from 1500 mm on the coast to 3000 mm in the hinterland

ranges (Gold Coast City Council 2007) with a temperature range between 0°C- 35°C (Bureau

of Meteorology 2009).

Page 40: A thesis submitted for the fulfilment for the requirements

17

The region supports over 1550 species of native plants, approximately 323 bird species, 105

species of reptiles and amphibians and over 70 species of mammals (Gold Coast City Council

2007), some of which are listed as threatened or endangered (Queensland Government 1992).

Thesis aims and structure

Due to the extent of land clearing for urbanisation and forestry, the resulting depletion of

hollow-bearing trees will lead to a shortage of this resource for many years to come. This

study of the distribution, abundance and use of the hollow-bearing tree resources in remnant

forest patches of urban bushland will improve our understanding of the importance of this

resource along an urban gradient. This could potentially lead to a re-evaluation of forest

patches not just as places of refuge, but more importantly as potential functioning

ecosystems, which until now have not been adequately investigated. Consequently, this

study will increase our knowledge and lead to a heightened understanding of the resources

and biodiversity found within these sites and subsequently to a greater ability to manage them

more effectively.

Biotic homogenisation along the urban-rural gradient is one of the key issues affecting urban

biodiversity (Alvey 2006; McKinney 2006). Here the loss of species richness is attributed to

the increase in generalist species at the expense of those with more specialised habitat

requirements, such a dependence on hollow-bearing trees (Smith et al. 2007; Harper et al.

2008). Within the urban landscape the effects of anthropogenic activities are ever present

and ongoing. Therein, the current study will assess the dynamics of hollow-bearing trees as a

habitat resource along an urbanisation gradient in south-east Queensland, Australia. No

similar studies have been carried out in south-east Queensland, an area of rapid expansion

and urban growth (Graymore et al. 2008; Gurran 2008).

The overall aim of this study was to identify the biotic and abiotic factors influencing hollow-

bearing tree availability, density and use along an urbanisation gradient at the landscape

scale. Hypotheses were derived from the current literature where the urban gradient was

identified as a powerful filter influencing biodiversity in general. Tree hollows provide

essential habitat for wildlife; yet despite the acknowledged importance of this resource,

within the current literature there exists a paucity of information about the use and

Page 41: A thesis submitted for the fulfilment for the requirements

18

availability of tree hollows found within remnant urban forest patches (Crooks et al. 2004;

Goldingay and Sharpe 2004a; Beyer et al. 2008; Brearley et al. 2010).

The following chapters describe the relationship between the urbanisation gradient, hollow-

bearing trees and their use. However, the first objective was to undertake a review and some

primary data analysis of the current legislation in place within Australia to protect hollow-

bearing trees, with a particular focus on urban centres. This was undertaken to understand

what measures are in place to protect hollow-bearing trees from the landscape scale down to

individual tree level (Chapter 2).

Before any hollow-bearing tree management options can be considered, information is

required on the availability of hollow-bearing trees as well as hollow number, type and size.

An inventory of hollow-bearing trees along an urbanisation gradient was undertaken to

quantify the current stock of hollow-bearing trees (Chapter 3). The distribution and

abundance of hollow-bearing trees was hypothesised to differ along the urban gradient in

response to the rate of urbanisation and historical logging such that the expectation was that

heavily urbanised patches would have fewer hollow-bearing trees than those in more rural

areas.

Chapter 4 investigated if there was a relationship between landscape dynamics and hollow-

bearing trees and tested the hypothesis that the urbanisation gradient was having an effect on

hollow-bearing trees. This was done to understand what the impacts of the urbanisation

gradient, individual tree disturbances and past anthropogenic activities such as timber

harvesting and land clearing are having on hollow-bearing trees.

Chapter 5 assesses the dynamics of the hollow-bearing tree resource itself and then asks the

following questions; (i) How long until hollow formation occurs in the current cohort of

recruitment trees (i.e. non-hollow-bearing trees of those species that have the potential to

form hollows in the future, see Chapter 5 for further definition)? And (ii) Does dbh influence

hollow type? The loss of hollow-bearing trees along the gradient due to projected land

clearing rates for the urbanisation process over the next 10, 50, 100 and 150 years are

predicted. This information is vital for managers and planners in order to gain an

understanding of the long term management of hollow-bearing trees.

Page 42: A thesis submitted for the fulfilment for the requirements

19

Finally, to better understand the functional role of urban forest patches, the relationships

between hollow-using vertebrate fauna found within remnant forest patches and a

combination of landscape, hollow-bearing tree variables and disturbance parameters along

the urban gradient were quantified in Chapter 6. It is hypothesized that the species richness

and occupancy of hollow-using vertebrate fauna differs along the urban gradient.

Chapter 7 provides a synopsis of this study and summarises the implications of Chapters 2–6

for hollow-bearing trees along an urbanisation gradient and offers suggestions for future

research.

Page 43: A thesis submitted for the fulfilment for the requirements

20

Chapter 2: Forest conservation policy implementation gaps: Consequences for the management of hollow-bearing trees in Australia. Conservation and Society 12, 2014.

Abstract

Hollow-bearing trees in native forests and woodlands are significant habitat resources for

many Australian fauna. However, habitat removal, from commercial harvesting and urban

development continue to threaten these ecosystems. Protection for these habitats and their

species is purportedly provided for in legislation, policy and strategic management plans.

Yet public debate and disagreement surrounding forest management has resulted in the

disintegration of national plans as interpreted by states and territories as well as individual

stakeholders. The implications of this on managing forests for biodiversity conservation have

in some cases been extreme. This Chapter presents a hierarchical review of the current

legislation and policy mechanisms underpinning forest conservation in Australia paying

specific attention to important habitat features such as hollow-bearing trees. Apart from

Federal and State legislation acknowledging the importance of hollow-bearing trees to

biodiversity, sufficient mechanisms to halt the ongoing loss of this resource from Australian

landscapes at the local level seem to be lacking. Hollow-bearing tree conservation strategies

from 46 local council areas in Victoria, New South Wales, Tasmania and Queensland were

reviewed. Few (<5%) respondents indicated that they have specific plans for the

conservation and management of hollow-bearing trees, highlighting a policy implementation

gap at the local level. Furthermore, apparent environmental management strategies and

actions rank relatively low on local council priorities. Therefore, a stronger focus on

conservation actions geared towards management of critical habitat features across the

landscape supported by robust local, national and international policy is required.

Page 44: A thesis submitted for the fulfilment for the requirements

21

Introduction

Sound environmental policy is the cornerstone of global conservation efforts, directing action

at local, regional, national and international levels (Thomas 2007). However, over the past

five decades environmental policy decisions in Australia have caused considerable conflict

within all levels of government. Consequently national plans as interpreted by states and

territories as well as individual stakeholders have collapsed, resulting in far reaching policy-

implementation gaps, particularly for forest management (Musselwhite and Herath 2007).

The inability to resolve conflicts nationally has led to a number of environmental assets

important for the retention of biodiversity being at risk. The national attention given to

forestry and forest management within Australia (see McAlpine et al. 2007), underpins the

conservation of these natural resources, particularly large hollow-bearing trees. This in turn

provides a useful platform for the analysis of the potential consequences of policy-

implementation gaps at the local level.

Living or dead, hollow-bearing trees are important features of forests (Spies et al. 1988;

Newton 1994), providing structural heterogeneity in natural, cultural and recently modified

landscapes. Hollow-bearing trees are also globally acknowledged for their value to conserve

wildlife (Mawson and Long 1994; Ball et al. 1999; Wormington et al. 2002; Aitken and

Martin 2004; Beaven and Tangavi 2012). They provide essential wildlife habitat by offering

protection from predators; providing roosting and breeding sites; are used for feeding

(Holloway et al. 2007); and offer a stable micro-environment that ameliorates ambient

weather conditions (Gibbons and Lindenmayer 2002). Within Australia, the use of this

resource by native fauna has received considerable research attention (Comport et al. 1996;

Lamb et al. 1998; Gibbons et al. 2000a; Lumsden et al. 2002; Eyre 2005; van der Ree et al.

2006), with many of these studies focusing on threatened species dependence on hollow-

bearing trees. Consequently, the continued loss of these habitat features from the landscape

is seen as an important conservation management problem (Gibbons and Lindenmayer 1996;

Eyre 2007; Goldingay 2009).

The retention and longevity of hollow-bearing trees within the landscape depends on the

synergistic effects of multiple factors. These include natural ecological processes

contributing to hollow formation, and the extent to which hollow-bearing trees are threatened

Page 45: A thesis submitted for the fulfilment for the requirements

22

by anthropogenic activities, but also the manner in which forests are conserved and managed.

Hollow formation is influenced by a variety of environmental factors related to tree damage

and decay forming agents such as termites and fungi (Gibbons and Lindenmayer 2002;

Adkins 2006), along with tree characteristics such as age and size (Lindenmayer et al.

2000a). Thus, Australian eucalypts require at least 150 years before hollows are likely to be

suitable for occupation by vertebrate fauna (Ross 1998). Threatening processes facing

hollow-bearing trees include agriculture (Cogger et al. 2003), firewood collection (Driscoll et

al. 2000), urban development (Garden et al. 2007b), altered fire regimes (Ross 1999), as well

as forestry and timber harvesting (Gibbons and Lindenmayer 1996; Lamb et al. 1998; Adkins

2006). Correspondingly, political inertia in the translation of forest management policy to

management actions (Riley et al. 2003; Prest 2004; Musselwhite and Herath 2007), also

potentially threatens the conservation of hollow-bearing trees. Therefore, an understanding

of the strategic frameworks and policies in place to protect hollow-bearing trees at the local

level is required to demonstrate how these achieve regional, state and national biodiversity

conservation objectives.

Australia’s natural forest estate is made up of a variety of forest types and land tenures with a

corresponding array of management practices (Aenishaenslin et al. 2007). Forestry practices

and interests can vary immensely amongst Federal, State and local private landholders, with

divergent ideologies resulting in ad hoc programs with no long-term planning or direction.

This creates difficulty in formalising conservation objectives that are often incompatible with

other potentially competing land uses such as timber harvesting, urban development or

agriculture (Dovers 2003). The diversity of historical, cultural, political and physical

circumstances within Australia’s states and territories, contributes to decentralised, confusing

and often overlapping commercial versus conservation forestry initiatives. However, some

consistency arising from the development of ‘best practice’ policies among States has been

achieved, giving rise to five basic forestry conservation initiatives. These initiatives allow for

1) afforestation, 2) the creation of wildlife habitats, 3) prevention or reversal of land

degradation, 4) establishment of shelter for crops and wildlife as well as, 5) general amenity

(Herbohn et al. 2000). However, while some attention has been directed towards retaining

habitat features in natural production forests within Australia (Lamb et al. 1998), there is

relatively little information available on the implementation of federal conservation policies

at a local government level. This failure to amalgamate policies across the nation, is also

Page 46: A thesis submitted for the fulfilment for the requirements

23

found at a state level. For example, a previous review of stakeholder involvement in

Victorian forest policy (Musselwhite and Herath 2007) revealed that local governments were

not included in the survey, thus highlighting the lack of commitment to include local

government agencies.

Given the importance of essential habitat features such as hollow-bearing trees, managing

landscapes for their persistence should be considered critical to the success of conservation

initiatives. How such conservation actions are implemented depends on the interpretation of

policies at various levels of government, but particularly at the local level. This Chapter

provides an overview of the current ‘best practice’ policy and guidelines for forest

management in eastern Australia. It compares the relative importance of environmental

management at the local level and then compares the scalar translation of policy to

implementation. It does so by investigating specific actions to conserve hollow-bearing trees

and/or habitat trees as essential habitat features within the landscape.

Methods

The degree to which the conservation of hollow-bearing trees within Australia is captured

within existing legislative frameworks was assessed by reviewing available scientific

literature, as well as forestry and nature conservation related legislation, to document the

current status of forest management. This process summarised the hierarchy of Federal and

State legislation followed by ‘best practice’ forestry policies down to local government policy

and planning guidelines. These documents were then searched for any mention of hollow-

bearing or habitat trees.

The transfer of guiding policies and recommendations from higher levels of government that

inform strategies at the local level are imperative. With this in mind, a two-stage process was

followed to assess local level environmental management objectives, and specifically how

these translate into the protection of hollow-bearing trees. Information was accessed at the

local council level by contacting 86 randomly selected local council areas throughout eastern

Australia (Tasmania, Victoria, New South Wales and Queensland). These local councils

were categorised at the broad level as either: Urban, Regional, or Rural based on the

Page 47: A thesis submitted for the fulfilment for the requirements

24

classification system used by the Department of Regional Australia Local Government, Arts

and Sport (2001). In the first stage, web pages of local councils were assessed to determine

their environmental emphasis within a broader social responsibility context. This was

undertaken across three hierarchical levels. Level 1 entailed enumerating the number of

primary council strategies and objectives listed on their respective web pages. This was then

taken to a finer scale by searching for any strategies pertaining to the importance of

environmental assets and biodiversity and thus hollow-bearing trees by association. Level 2

calculated the relative importance of environmental actions. The total number of menu tabs

on each primary council web page was counted and the placement of 'environment' within

this structure was determined by recording its order within them. The relative importance

was then calculated by dividing the order of the 'environment' tab by the total of all tabs for

each council and these figures were normalised by the maximum number of menu tabs across

all councils (n=18). Theoretical minima and maxima for the relative importance of

‘environment’ ranged between 1 (lowest score) and 324 (highest score). Finally, Level 3

analyses compared the number of sub-categories within each ‘environment’ tab to those

within all other tabs on the primary web page. Collectively, these results give an overall

representation of how councils place themselves in a socio-environmental context. As

Flemming (1998) explains, in web page design the client is aiming to supply information to

their target audience, as well as the message that they are attempting to convey about their

business. Therein, it can be seen that by evaluating council web pages placement of tabs on

their respective web pages does offer a valid method for analysis. Comparisons of the

number of environmental strategies among councils based on their location category were

made using a one-way ANOVA.

For the second stage, copies of any policies and/or strategies that captured the conservation of

hollow-bearing trees as one of their management actions were requested from council

representatives. Documents received from council representatives were then ranked based on

the nature of the conservation and management frameworks and actions that councils have in

place to conserve hollow-bearing trees. Three broad categories were recognised; (i) council

policies that rely upon the Environmental Protection and Biodiversity Conservation Act 1999

(EPBC Act) and other Federal and State legislation with no specific reference to hollow-

bearing trees; (ii) council policies that rely on the EPBC Act and other Federal and State

legislation with specific reference to the importance of hollow-bearing trees, but with no

Page 48: A thesis submitted for the fulfilment for the requirements

25

specific protection offered; (iii) council nature conservation strategies and management plans

with specific mention of the retention/protection of hollow-bearing trees and actions to

achieve these objectives. The implications of these findings are discussed in the context of

forest conservation and management in Australia with specific reference to the threats facing

hollow-bearing trees.

Results

Legislation and the conservation of Australian forests and hollow-bearing trees

At the Rio Earth Summit in 1992 the use and management of forests received considerable

attention globally, culminating in the production of a number of agreements including the

‘Global Statement of Principles of Forests’ (GSPF) and ‘Agenda 21’. The GSPF made

several recommendations for the management, conservation and sustainable development of

all forest types globally. Australia endorsed the GSPF and signed a number of conventions

relating to biological diversity and climate change, in order to achieve the full range of

benefits available from forests (Commonwealth of Australia 1992). Furthermore, Agenda 21

called for global action in regards to sustainable development and was regarded by many as

the centrepiece of the Rio accords (Prado-Lorenzo and Sanchez-Garcia 2007). From Agenda

21, Local Agenda 21 was developed that offers support for environmental issues at the local

level (Thomas 2007). Local Agenda 21 systems were implemented with the aim of managing

for the future while achieving sustainability through integrating planning and policy, focusing

on long-term outcomes and involving all sectors of the community (Stenhouse 2004; Thomas

2007).

Australian Federal strategies and objectives were then outlined in the ‘National Forests Policy

Statement 1992’ (NFPS) which lay the foundations for forest management in Australia for

the next century. The NFPS, signed by the Commonwealth, States, Territories and local

governments, provides the framework recognising the need to initiate processes required for

changes to occur in order to protect Australia’s forests for ecologically sustainable growth

and management (Commonwealth of Australia 1992). At this level however, there are no

explicit provisions made that stipulate the management interventions or actions required to

protect specific habitat features such as hollow-bearing trees.

Page 49: A thesis submitted for the fulfilment for the requirements

26

Following the endorsement of the NFPS, the Commonwealth and State governments

established the framework for Regional Forestry Agreements (RFAs). Twelve RFAs were

progressively signed by the Commonwealth and State governments between 1997 and 2001

(Mobbs 2003). Each RFA is a 20 year intergovernmental agreement concerning the security

of forests resources, as well as the conservation of environmental heritage and social values

(Lane 1999; Aenishaenslin et al. 2007). The key outcomes of RFAs are the allocation of

approximately 30% of publicly owned commercial native forests to an expanded reserve

system, the strengthening of the codes of practice and increased resource management for the

timber industry (McAlpine et al. 2007). Despite the provisions for increasing the extent of

native forests within the national reserve system the RFAs do not highlight any measures

required to conserve specific structural features within these forests. Furthermore, there has

not been widespread adoption of the RFAs by State governments, exacerbating

inconsistencies in forest management practices among states and territories. For example,

Queensland did not initially sign an agreement and developed its own Queensland Forest

Agreement outside of the Federal Government’s RFA guidelines. This was done in

conjunction with the Australian Rainforest Conservation Society, the Queensland

Conservation Council, The Wilderness Society and the Queensland Timber Board (Mobbs

2003). This alternative agreement, initiated in 1999, failed to fully encompass all forest types

in south-east Queensland, favouring rainforest and wet sclerophyll forests in the conservation

reserve system (McAlpine et al. 2007). Virtually no allowance was made for the protection

of dry eucalypt forests, thereby failing to preserve those ecosystems and species with the

highest vulnerability to land clearing (Norman et al. 2004; McAlpine et al. 2007). Similarly,

dry sclerophyll forests in New South Wales received little or no recognition and are poorly

represented in the New South Wales RFA (Flint et al. 2004). Nevertheless, it has been

suggested that RFAs under the NFPS have achieved some positive outcomes, if only for

native forests within the national reserves system following the implementation of a higher

level of sustainable forestry practices in regards to conservation (Meek 2004).

Nationally, RFAs require that within individual public forest estates a minimum of 15% of

each forest type, 60% of remaining old growth forest and at least 90% of high quality

wilderness is reserved (Aenishaenslin et al. 2007). While the RFA process has seen a

significant reduction in the volumes of timber harvested from within the public estate, in

some states this has led to an increase in the importance of private native forests as a source

Page 50: A thesis submitted for the fulfilment for the requirements

27

of saw log’s (Norman et al. 2004). Private forests are therefore potentially under greater

threat as a consequence of the implementation of the RFAs in various states. This highlights

that further engagement at the local level is needed as the current system only ‘encouraged’

land owners of private forests to participate in the policy. Local authorities such as councils

are also not readily included in national reserve system planning or management (Lunney

2004).

Private forests have typically been placed in the ‘too hard’ basket by policy makers,

legislators, conservationists and foresters as being of low conservation and economic value.

For example, the management of private forests has been accompanied by a combination of

public ignorance and political and agency inertia, where the laws applicable to private native

forests have been formed by inaction (Prest 2004). For example, while the forestry code of

New South Wales is highly restrictive in terms of forest harvesting on public land, and

management is heavily focused on environmental objectives, little importance is placed on

the contribution of private native forests. In Queensland and Victoria codes are commercially

driven, while Tasmania attempts to achieve a balance between environmental and

commercial objectives (Aenishaenslin et al. 2007).

Equally, small forest patches isolated by fragmentation through urbanisation have been

abandoned and considered to be of low conservation value (Alvey 2006; Harper et al. 2008).

However, small and often isolated forest patches are valuable to conservation as they are

representative of former habitat that was once common to any given area (van der Ree et al.

2003), and also serve to connect habitats within a larger landscape (Lindenmayer and Fischer

2006). Ongoing land conversion will increase the value of private and urban forests (Alvey

2006), as there is a clear link between the condition of private land and biodiversity

conservation in Australia (Fitzsimmons and Westcott 2001). The importance of local

councils in directing conservation efforts on private land is paramount. For example within

Queensland these agencies already administer a number of private conservation initiatives

such as the Land for Wildlife schemes which are also available in Victoria, New South Wales

and Queensland (Fitzsimmons and Westcott 2001). Local councils therefore act as the

interface in providing private landholders with clear policy directions on the conservation and

management of forest features such as hollow-bearing trees. Despite these objectives,

Page 51: A thesis submitted for the fulfilment for the requirements

28

conflicts continue to plague the industry and both public and private native forests continue to

shrink, emphasising the urgent need for sustainable management of all forests to conserve

biodiversity (McAlpine et al. 2007).

Regardless of their long-standing recognition as important forest features, the loss of hollow-

bearing trees due to forestry practices was nominated for listing as a key threatening process

in 2005 under the Commonwealth Environment Protection and Biodiversity Act 1999. The

nomination recognised two parts to the process: activities which remove or destroy hollow-

bearing trees and processes which affect tree and seedling recruitment. However, the

Endangered Species Scientific Sub-committee found that, where an RFA was in place, a

nationally co-ordinated threat abatement plan was neither feasible nor an effective and

efficient way to limit the loss of hollow-bearing trees. For regions where no RFA is in place,

a range of State forestry management prescriptions, nature conservation and threatened

species legislation, vegetation clearance controls and voluntary measures apply. This

suggests that RFAs and other legislation have explicit details and plans in place for the

conservation of hollow-bearing trees. This is not the case. These documents largely

comprise broad sweeping terminologies that cover conservation issues in general, with no

specific references to hollow-bearing trees or other important forest structures. Thus, hollow-

bearing trees are not specifically protected under the Commonwealth Environment Protection

and Biodiversity Act 1999; as there is a belief that sufficient mechanisms are in place within

individual state legislation to address the situation. The time frame for management and

system reform as determined by the RFAs under the NFPS draws to its conclusion in 2017

and the direction of forestry reform and conservation in Australia thereafter should be closely

monitored. Herein lies an opportunity exists for local councils to put policy measures in

place to ensure the ongoing conservation of hollow-bearing trees.

Forestry practices within Australia generally identify hollow-bearing trees as being of low

commercial value, while also suppressing forest regeneration (Gibbons and Lindenmayer

2002). Despite this, they continue to be harvested primarily as saw logs or at a later stage for

fire wood or fence posts (Lamb et al. 1998). In some regions, provisions for the retention of

hollow-bearing trees under certain land uses are in place. For example, in Queensland the

‘Habitat Tree Technical Advisory Group’ (HTTAG), was formed by the Environmental

Page 52: A thesis submitted for the fulfilment for the requirements

29

Protection Agency to assist in the development of guidelines for the management of hollow-

bearing trees as habitat for wildlife in relation to silviculture (Lamb et al. 1998). Under these

guidelines, habitat and recruitment trees are identified as those requiring retention for the

purpose of wildlife conservation and may be of either merchantable or non-merchantable

value (Queensland Government 2007). Habitat trees are defined as those having one or more

hollows >10 cm in entrance diameter, while recruitment trees are suitable trees within a 40

cm diameter at breast height class (Lamb et al. 1998). Recommendations from the HTTAG

have also been incorporated into the Code of Practice for Native Timber Production on State

Lands (Eyre 2005; Queensland Government 2007). Of those recommendations, the retention

of a minimum of six live habitat trees and two recruitment trees per hectare is mandatory

throughout harvesting areas (Eyre 2005). Under standard harvesting practice, where habitat

trees occur uniformly in an area subject to clearing, additional recruitment trees must be

retained where >50% of the basal stand area is to be removed (Lamb et al. 1998). However,

these recommendations and guidelines are not applicable outside of production forests, again

highlighting the need for local government bodies to establish and implement conservation

programs aimed at the maintenance and recruitment of hollow-bearing trees on commercial,

public and private lands.

The preceding review highlights that all Acts, Codes, agreements and policies at a Federal

and State level within Australia strive for ecological sustainability and the retention of certain

vegetation communities at a national scale. Yet, apart from the HTTAG in Queensland there

appears to be little documentation regarding the specific management and retention of

hollow-bearing trees as a habitat resource to be conserved within the landscape. Therefore,

while Federal and State legislation acknowledges the importance of hollow-bearing trees to

biodiversity, sufficient mechanisms to halt the ongoing loss of this valuable habitat resource

from Australian landscapes appear to be lacking. Without adequate legislation at the national

or state levels there is little hope in achieving conservation objectives for hollow-bearing

trees at the local level. This study assessed the nature of hollow-bearing tree management

strategies at the local level to determine if national level policies are implemented at this

scale.

Page 53: A thesis submitted for the fulfilment for the requirements

30

Local level ranking of the importance of biodiversity assets, in particular hollow-bearing

trees

Of the 86 councils contacted only 46 (53.4%) responded to calls for information on their

environmental strategies and plans for the conservation of hollow-bearing trees. These

councils were subsequently used in the analysis of broader council social responsibility and

how ‘environment’ placed within this operational framework. Of the 46 respondents: 12

were classified as being ‘urban, 23 ‘regional’ and 11 ‘rural’ (Table 2.1).

Level 1 searches of council web pages revealed that only 46.7% of councils had overarching

strategies and guiding principles in place relating to environmental protection and the

importance of biodiversity assets. The relative importance of ‘environment’ among councils

was also highly variable with ranked importance values ranging between 2.3 and 108.

Overall, 85% of councils ranked 'environment' low, very low or not at all. Only 2.1% of

councils ranked ‘environment’ highly and a further 13% placed a medium emphasis on

'environment’ (Figure 2.1). Finally, Level 3 analyses revealed that the number of

environmental programs run by councils in general were significantly less than the mean

numbers captured across all other council programs (t=4.5; d.f.=45; P <0.001) (Table 2.1).

There was also no difference in the emphasis placed on ‘environment’ when compared

among ‘urban’, ‘regional’, or ‘rural’ locations (F=2.32; d.f=2,43; P=0.11).

Information on the management strategies for hollow-bearing trees at the local level provided

by 46 councils provided the basis for assessing the degree to which higher level policies are

taken up by implementing agencies. Local councils providing information were spread along

the south-eastern seaboard of Australia and were represented by Queensland (30.4%), New

South Wales (36.9%), Victoria (26.1%) and

Page 54: A thesis submitted for the fulfilment for the requirements

31

Table 2.1: Data from the web pages of local councils in relation to how they place

themselves environmentally in a social context. The number of environmental strategies for

each council are those listed within each primary ‘Environment’ tab on council web pages.

Council by State

*Council Classification

No: of tabs

Placement of 'Environment' tab on web page

Mean n tabs (excluding environment)

No: of environmental strategies

NSW 1 Rg 8 4 7.4 1 NSW 2 U 9 4 12.6 28 NSW 3 Rg 6 4 5.6 9 NSW 4 R 5 0 14.5 0 NSW 5 R 13 9 5.3 10 NSW 6 Rg 7 7 11.8 3 NSW 7 U 8 8 9.0 0 NSW 8 R 8 4 5.1 0 NSW 9 Rg 7 1 8.5 1 NSW 10 Rg 5 3 10.8 1 NSW 11 R 8 5 9.6 0 NSW 12 U 3 3 29.0 14 NSW 13 Rg 8 5 11.6 5 NSW 14 Rg 7 3 12.0 8 NSW 15 Rg 5 5 9.5 0 NSW 16 U 7 7 10.2 0 NSW 17 R 9 7 21.9 8 QLD 1 U 8 4 30.3 0 QLD 2 U 7 3 38.8 6 QLD 3 U 8 5 14.4 5 QLD 4 U 7 0 32.5 10 QLD 5 U 6 0 11.2 0 QLD 6 U 9 4 9.3 0 QLD 7 R 7 4 13.3 1 QLD 8 Rg 8 4 9.0 0 QLD 9 Rg 8 4 9.6 3 QLD 10 Rg 8 5 12.9 1 QLD 11 Rg 7 5 23.5 0 QLD 12 Rg 8 0 5.1 1 QLD 13 R 5 0 9.5 3 QLD 14 U 6 0 12.2 4 TAS 1 Rg 5 4 6.5 0 TAS 2 R 9 0 7.1 1 TAS 3 R 7 0 5.8 1 VIC 1 Rg 8 8 22.7 3

*U Urban: Rg regional: R Rural.

Page 55: A thesis submitted for the fulfilment for the requirements

32

Table 2.1 continued: Data from the web pages of local councils in relation to how they

place themselves environmentally in a social context. The number of environmental

strategies for each council are those listed within each primary ‘Environment’ tab on council

web pages, continued.

Council by State

*Council Classification

No: of tabs

Placement of 'Environment' tab on web page

Mean n tabs (excluding environment)

No: of environmental strategies

VIC 2 Rg 6 2 12.2 3 VIC 3 U 17 0 8.0 0 VIC 4 U 7 4 8.3 13 VIC 5 Rg 8 0 24.7 5 VIC 6 U 18 9 10.2 14 VIC 7 Rg 6 0 15.2 2 VIC 8 U 6 3 13.2 8 VIC 9 Rg 14 9 7.2 1 VIC 10 R 5 0 26.0 16 VIC 11 Rg 18 11 7.2 3 VIC 12 R 7 4 9.2 20

*U Urban: Rg regional: R Rural.

Figure 2.1: The relative importance placed on the environment by local councils based on

the ranking of ‘environment’ tabs in relation to all other tabs on primary council web pages.

0

5

10

15

20

25

Zero Very low Low Medium High Very High

Num

ber

of C

ounc

ils

Relative importance

Page 56: A thesis submitted for the fulfilment for the requirements

33

Tasmania (6.5%) (Table 2.2). Significantly, 95.6% of councils reported that they had no

specific conservation, management policies or guidelines in place for hollow-bearing trees,

and that they relied primarily on either State or Federal Acts and legislation to protect

vegetation communities within their shires.

Only two councils (4.4%), one each from New South Wales and Queensland had specific

measures in place to protect hollow-bearing trees, while a further seven councils (15.2%)

mentioned the ecological importance of hollow-bearing trees but did not have any specific

conservation measures in place. In the case of the New South Wales council these guidelines

were embedded within development control measures along with a significant tree registry.

By comparison, the council in Queensland had a series of guidelines targeted to hollow-

bearing tree retention. The guidelines implemented by the Queensland local council is a

comprehensive document that is entirely committed to the identification, conservation,

retention and management of hollow-bearing trees within its boundaries with reference to the

Integrated Planning Act 1974 (Table 2.3). The guidelines offer ecological recognition and

value to habitat trees within the landscape supported by the ability to undertake on-ground

implementation, including measures that consider the implications following the removal of

hollow-bearing trees.

Table 2.2: Councils contacted to request information on strategies and policies targeted to

hollow-bearing trees at the local level, and number of respondents by state.

State Queensland New South Wales Victoria Tasmania Number contacted 25 26 25 10 Number providing information 14 15 11 2

Page 57: A thesis submitted for the fulfilment for the requirements

34

Table 2.3: Specific measures implemented by two urban local councils to protect and

preserve hollow-bearing trees on public and private land in Queensland and New South

Wales.

Queensland Council New South Wales Council

Guidelines for the Retention of Hollow-bearing Trees.

1. Tree Management Report

1.1 Ecological Assessment Report

1.2 Risk assessment

1.3 Hollow-bearing tree assessment

1.4 Selection Criteria: Hollows, Tree health and DBH

2. Annual Management Report

2.1 Maintenance of hollow-bearing trees

2.2 Data reporting and recording

3. Tree Removal Plan

3.1 Removal of wildlife

3.2 Compensatory actions

Development Control Plan.

1. To protect old growth and significant hollow-bearing trees.

2. Significant Tree Register: Trees of high ecological value as well as other social values.

Page 58: A thesis submitted for the fulfilment for the requirements

35

Discussion

Within Australia, amnesia and ad hoc policy formulation is considered to be an ongoing

problem at the institutional and organisational foundations of forest policy and management

(Dovers 2003). This presents a significant problem for land managers in regards to the

implementation of natural resource policy and legislation. The difficulties are in translating

the intentions and aspirations of parliamentary statutes, which are intended to protect the

environment, into meaningful on-ground actions (Shepheard and Martin 2011). This

supported the results here that have revealed how some councils appear to give little

consideration for environmental issues within the broader socio-environmental context, and

that few gave specific attention to hollow-bearing trees. This is demonstrated by the fact that

eight of the 46 councils surveyed appeared to have no information relating to the environment

and/or the importance of biodiversity assets displayed on their public websites. Furthermore,

when councils did provide information about the environment, this information was often

hidden or nested within larger programs or objectives. Nevertheless, results from the three-

tiered investigation into council’s web pages suggest that approximately 50% of local

councils generally considered the environment and biodiversity as being of some importance.

To interpret these results further, they need to be taken in context with the information

pertaining to hollow-bearing trees supplied by councils. In doing so the overwhelming lack

of translation of overarching national and regional policy and legislation into local

conservation action is telling. This lack of local policy implementation is highlighted by the

fact that only two councils (4%); both of which were classified as being ‘urban’; had

dedicated measures in place to protect hollow-bearing trees. The majority of local councils

rely heavily on both existing State and Federal legislation, despite these being vague in their

provisions for the preservation of hollow-bearing trees. Thus, the dependence on overarching

policies may not ultimately achieve conservation objectives simply because these policies

frequently do not have the level of support and detail required to implement targeted

conservation strategies.

In the years following the introduction of Local Agenda 21, 117 of Australia’s then 750

councils had either established or were developing local sustainability strategies (McDonald

1998). While this national trend saw the transfer of power and responsibility from Federal

government to local authorities. As Local Agenda 21 is a voluntary option for councils, by

November 1999 local council involvement had dropped to 75 (Mercer and Jotkowitz 2000).

Page 59: A thesis submitted for the fulfilment for the requirements

36

Whittaker (1997) obtained a deeper insight as to why Australian councils were slow or

reluctant to embrace Local Agenda 21; reporting that many had difficulty in deciding whether

they were in fact developing a Local Agenda 21 or not and if their management plans could

be incorporated into a Local Agenda 21. Ultimately, Local Agenda 21 requires changes to

values and patterns of consumption both from within local councils and the community.

However, many councils stated that they had great difficulty in raising awareness of Local

Agenda 21 and getting the community involved at the time (Whittaker 1997). Furthermore,

the apparent lack of support offered to conservation strategies within local councils, as

recorded in this study, is further exacerbated by the limited transfer of scientific findings to

policy makers or the general public (Pitman et al. 2007). Ecologists generally convey their

findings to a limited audience (e.g. other academics) often resulting in key decisions relating

to the management of biodiversity and conservation being made without the full benefit of

science (Shanley and Lopez 2009). Ironically, the rapid loss of biodiversity continues despite

the plethora of information available and the considerable financial investment in research,

highlighting the breakdown in communication between science and policy makers. Based on

the findings presented here, even where research may have informed the development of

higher level conservation policy, these policies are not being enacted at a local level

demonstrating the cascading effects of limited information flow pathways.

This study demonstrates this policy implementation breakdown by using hollow-bearing trees

as an example and suggests that the problem itself is more deep-seated; thus compromising

the effectiveness of a myriad of conservation programs at the local level. Conservation

managers at the local level therefore need to make conscious efforts to address ongoing

threats to natural habitat and their associated fauna and flora through the development of

targeted action plans along with the implementation of hollow-bearing tree protection

legislation (e.g. development control measures). The extent to which these problems are

entrenched more broadly within other Australian local councils remains to be seen as only

53% of councils responded to this survey. Nonetheless, it is troubling that this initial review

may be indicative of a prevailing status quo, particularly in urban regions.

This investigation generated considerable interest amongst councils, with some council

respondents expressing their frustration at the lack of protection they felt they were able to

Page 60: A thesis submitted for the fulfilment for the requirements

37

give for hollow-bearing trees and remnant vegetation in general. This is perhaps due to the

variable land uses that occur within council jurisdictions such as public parks, bushland,

reserve networks and recreational reserves, as well as the range of environmental and social

values operating across the landscape (Snep and Opdam 2010). For example, in urban areas

a tree is also considered to be hazardous if there is potential for harm to people and property.

Accordingly, trees that contain hollows that are readily detectable, and are then judged to be

hazardous, are therefore removed (Terho 2009). However, an underlying cause perpetuating

inaction appears to be linked with the slow and often hesitant rates of transition to an adaptive

management paradigm by local councils. Active adaptive management balances the

requirements of management with the need to learn about the system being managed thus

leading to better decision making (McCarthy and Possingham 2006). For crisis disciplines

such as conservation (Burgman et al. 1993), conservation managers must deal with direct

environmental threats in dynamic socio-political environments. The broader social

responsibility targets being pursued by local councils, as demonstrated by this study, can be

captured within contemporary active adaptive management paradigms to improve

management (Cundill and Fabricius 2010). Fostering an adaptive management approach for

the conservation of hollow-bearing trees at the local level is dependent on a number of

factors. These include an understanding of the conservation development objectives of

multiple stakeholders (e.g. councils, landowners, NGOs, traditional landowners), knowledge

of the status of the resource, and the identification of defined approaches to achieve these

objectives. Calls for conservation planners to develop strategies that will enable them to

engage with decision makers in order to integrate biodiversity conservation initiatives into

land-use planning (Lagabrielle et al. 2010) are supported by this study.

Conservation challenges and the way forward

Australian urban forests contain hollow-bearing trees (Goldingay and Sharpe 2004a; Harper

et al. 2005a), providing critical refugia for a variety of native fauna. These forests, however,

are characterised by significant time spans separating current hollow-bearing trees and a

succession of trees (of suitable sizes) for recruitment and the continuous replacement of

hollows (Gibbons and Lindenmayer 2002). Therefore, prioritising forest management

policies to preserve a suitable age-structure supply of hollow-bearing trees is required if the

biodiversity values of these urban habitats are to be retained in the long-term. This can be

achieved through a number of actions that would arguably form the basis of future hollow-

Page 61: A thesis submitted for the fulfilment for the requirements

38

bearing tree conservation strategies; particularly for those councils where these are currently

lacking. The following strategies and actions are recommended:

1. Definition of clear management objectives and conservation targets at the outset that

will inform future adaptive monitoring efforts;

2. Compiling an inventory of hollow-bearing trees at a local scale to establish a baseline

for identifying potential interventions required to meet predefined objectives. These

data can then be incorporated with other known data sets on the urbanisation process,

such as land clearing, land use, population density etc., to inform holistic land use

planning and policy development for the retention of hollow-bearing trees; and

3. Most importantly there is a need to facilitate on-ground action plans for the retention,

recruitment and management of hollow-bearing trees at the local level and

particularly within rapidly urbanising regions.

Such actions will in turn maintain forest structure and faunal communities as well as

increasing their value at a landscape scale. Activities include those that manage the residual

natural resources (e.g. minimising fire impacts on large dead hollow-bearing trees), but also

those that propose active facilitation to enhance or improve habitat conditions (e.g.

supplementation, accelerated hollow formation). While short-term supplementation e.g.

erection of nest boxes can assist species conservation efforts in some situations (Beyer and

Goldingay 2006; Durant et al. 2009), they cannot replace naturally occurring hollows and

their usefulness has been vigorously debated (Spring et al. 2001; Lindenmayer et al. 2002).

For hollow-bearing trees any mitigation action should be implemented with a complete

understanding of the current status and availability of this resource within the landscape.

Artificial hollow formation has also been suggested as an alternative to the use of nest-boxes

(Gibbons and Lindenmayer 1996). Accelerating the formation of hollows could be achieved

by deliberate attempts to kill or injure a tree (Bull and Partridge 1986), the injection of

growth hormones and inoculation with fungi (Connor et al. 1981), tree girdling (Connor et al.

1981), the use of explosives (Smith and Lindenmayer 1988), fire (Adkins 2006) and the

combination of these techniques with natural decay agents associated with hollow formation

Page 62: A thesis submitted for the fulfilment for the requirements

39

(Lindenmayer et al. 1993). However, Goldingay (2009) points out there are currently no

published studies from Australia that describe experiments to promote or create natural

hollows in trees or indeed whether these are successful. Furthermore, the time to accelerated

hollow formation in Australian hardwood forests may also be longer than the 3 - 5 years

reported from North American softwood forests (Bull and Partridge 1986; Arnett et al. 2010).

Conclusion

There are a number of implications for the ongoing management of natural resources at the

local level arising from this work, particularly for hollow-bearing trees. Firstly, this study

provides empirical evidence of a policy-implementation gap in connection with the

management of hollow-bearing trees in Australia. Despite the national emphasis placed on

the value of these habitat features within native forests there is little on-ground

implementation of conservation actions aimed at their persistence within the landscape.

Higher level policy is poorly translated into on-ground action and this requires a shift in

current management strategies to adopt a more proactive adaptive approach.

Secondly, a review of, and change in environmental policy is required in order to amalgamate

the varying, convoluted and disjointed policies currently in place to provide greater focus to

the management of forest resources at various scales. Continuing under current legislative

frameworks will be insufficient in preventing the further loss of key resources such as

hollow-bearing trees. Furthermore, the potential implications of such largely inadequate and

ineffective efforts extend beyond that of hollow-bearing trees and could compromise a

number of other forest assets; which is of particular concern in rapidly developing regions.

Thirdly, the role of local government in biodiversity conservation operates at both a

regulatory and advisory level in that local councils direct both short and long-term influences

on land management and development processes across varied land tenures. As such a

review of existing policy is also necessary to acknowledge and guide the conservation

contribution within the private sector.

Page 63: A thesis submitted for the fulfilment for the requirements

40

In closing, this review of local council policy provides new evidence demonstrating the lack

of connectivity between Federal policy and local implementation. While existing legislation

identifies the need to retain ‘forests’, ‘regional ecosystems’, ‘remnant habitat’ or ‘vegetation

communities’, it masks the fact that it is largely ineffective in delivering real on-ground

conservation actions for finer scale habitat features such as hollow-bearing trees, particularly

for those found within urban landscapes and non-protected forests.

Page 64: A thesis submitted for the fulfilment for the requirements

41

Chapter 3: Hollow-bearing trees as a habitat resource within an

urbanising landscape.

Introduction

The destruction and removal of natural habitats is proceeding globally at a rapid rate

(Rankmore and Price 2004; Lewis et al. 2009). Within Australia, the clearing and

disturbance of natural vegetation and the subsequent habitat fragmentation is recognised

as one of the principle drivers behind declines in biodiversity (Williams et al. 2001).

Approximately 50% of Australia’s forests have been lost or severely modified since

European settlement, with over 80% of eucalypt forests in particular having been lost to

anthropogenic activities (Bradshaw 2012). Consequently, much of the remaining forest

cover is severely fragmented (Bradshaw 2012). Small and often isolated habitats are

recognised as being valuable to conservation, as they are representative of former

habitat that was once common to any given area (van der Ree et al. 2003) and form part

of a larger landscape of habitats inclusive of patches and remaining matrix. The

landscape is therefore never uniform and disturbances to this are determined by the type

and intensity of the land uses found within it (Brady et al. 2009). These disturbances

and impacts to patches vary with the time since isolation, distance from other remnants,

and the degree of connectivity with other remnants (Saunders et al. 1991; Lindenmayer

and Fischer 2006). Within urbanising regions the matrix may be considerably less

permeable for biodiversity, thereby reducing the functional potential of natural habitat

within these landscapes (Ewers and Didham 2006b). The retention of such habitats as

refugia therefore becomes increasingly important in these situations (Harper et al.

2005a).

Within urban landscapes refugia are found in forest patches as well as urban parks and

gardens, with their value depending on their ability to provide resources for indigenous

flora and fauna (Godefroid and Koedam 2003; Harper et al. 2005a; Alvey 2006). The

availability or paucity of one such resource, tree hollows, will generally limit the

distribution and abundance of obligate hollow-bearing tree users and general species

Page 65: A thesis submitted for the fulfilment for the requirements

42

richness of an area (Gibbons and Lindenmayer 2002). Consequently, the spatial and

temporal abundance of hollow-bearing trees within a forest will vary in response to

disturbance and stand age (Lindenmayer et al. 1997), with the additional influence of

adjacent non forested areas impacting on forest structure and species composition

(Harper et al. 2005a). The importance of hollow-bearing trees for a large range of fauna

has been demonstrated globally (Braithwaite 1996; Ball et al. 1999; Gibbons and

Lindenmayer 2002; Bai et al. 2003; Aitken and Martin 2004; Cameron 2006;

Monterrubio-Rico and Escalante-Pliego 2006; Holloway et al. 2007; Blakely et al.

2008; Goldingay 2009). This is particularly so for Australian faunal communities,

given the large number of species that are reliant upon them as habitat trees (Braithwaite

1996; Ball et al. 1999; Gibbons and Lindenmayer 2002; Cameron 2006; Goldingay and

Stevens 2009). The distribution and abundance of hollow-bearing trees varies across

Australian landscapes, suggesting that other factors beyond the tree level are operating

to influence the occurrence of trees with hollows (Gibbons and Lindenmayer 2002).

These include large scale processes such as anthropogenic disturbance through timber

harvesting (Lamb et al. 1998), agriculture (Gibbons and Lindenmayer 2002) and

urbanisation (van der Ree and McCarthy 2005). Thus, depending on the prevalence of a

range of disturbance processes the density of hollow-bearing trees may vary

considerably. Generally, undisturbed forests and woodlands support a higher density of

hollow-bearing trees per hectare than disturbed forests (Lindenmayer et al. 1991b; Ross

1999; Koch 2008). Hollow-bearing tree density also appears to be positively correlated

with older stands, gullies, areas with no previous logging, and flat terrain (Lindenmayer

et al. 1991b).

In urban regions there is a distinct transition in housing density along an urban-rural

gradient (van der Ree and McCarthy 2005), with the gradient model being a useful tool

for examining the ecological consequences of urbanisation (McDonnell et al. 1997).

Previous studies along urban gradients have demonstrated significant effects of such

gradients on floral and faunal assemblages where communities are generally

depauperate in the more heavily urbanised areas as opposed to those in rural settings

(Brady et al. 2009; Hahs and McDonnell 2008; Johnson et al. 2008; Moffat et al. 2004).

However, while these studies focus on the species composition along the urban gradient

no studies have investigated the impact of the urban gradient on habitat resources such

as hollow-bearing trees. Therein, management practices, landscape variables and

ecological processes along the gradient will differ with the potential to affect resource

Page 66: A thesis submitted for the fulfilment for the requirements

43

availability. Due to the rate of urbanisation urban areas have had less exposure to

historical land practices such as logging for a longer period than the more rural areas.

As such it is hypothesised that there will be a difference in the size and availability of

hollow-bearing trees across the urban gradient where urban patches are expected to have

a greater number of these resources. Processes that occur in the urban matrix are

poorly understood primarily because forest patches may have been, or currently are,

subjected to increasing external pressures. There is also a paucity of information on the

distribution of habitat resources, such as hollow-bearing trees, along urbanisation

gradients. Since hollow-bearing trees are a highly valued resource within an Australian

context an investigation of the availability of hollow-bearing trees would provide a

deeper understanding of the impacts of the urbanisation gradient on urban forest

patches.

This chapter therefore aims to quantify the distribution and abundance of hollow-

bearing trees along an urbanisation gradient by answering the follow questions:

1. How does the urban gradient influence the density of hollow-bearing trees?

2. Is there uniformity in the availability of different types of hollows along the

urban gradient?

3. Does the size of the tree influence the availability of hollows? and

4. How important is the urbanisation gradient in influencing the density of

hollow-bearing trees compared to other disturbance parameters?

Compiling an inventory of both hollow-bearing and non-hollow-bearing trees and

evaluating these patterns against the impacts of disturbance and landscape variables

along the urbanisation gradient will answer the above questions. Specifically, it is

hypothesised that there will be more hollow-bearing trees within the urban forest

remnants due to the reduced time since exposure to historical land practices. Through

the empirical analysis of the aforementioned questions this Chapter is then able to

expand on the findings to discuss options for management prescriptions for the retention

of hollow-bearing trees along an urbanisation gradient.

Page 67: A thesis submitted for the fulfilment for the requirements

44

Methods

Study sites

The study was undertaken in patches of natural forest habitat on the Gold Coast,

Australia (for a full description see Chapter 1 pages 14-17). Study sites were confined

to wet and dry sclerophyll forest types from 21 Regional Ecosystems (Taylor 2003), of

which three are threatened and seven are of concern (Appendix 2). Wet and dry

sclerophyll forest types were selected as they are widespread across the study area and

would guarantee an adequate number of replicate sites. Natural forest patches ≥ 2 ha

were identified from within council Conservation Areas, National Parks, State Forests

and on private land. Ground-truthing of 50 sites that were deemed to be suitable was

undertaken to exclude those that may have access issues, intensive livestock grazing or

a mowed understory. Sites with either grazing or mowing (e.g. parklands) were

excluded as these were not representative of natural vegetation communities. Forty-five

sites were subsequently selected from the highly developed coastal region through to

the more natural environments of the hinterland located ≤ 500 m above sea level. The

500 m altitudinal threshold was chosen as rain forests generally dominate the landscape

above this elevation. Five large contiguous forest patches (> 500 ha) were chosen as

control sites in order to allow for variations found within large forest patches. Three of

the controls were located in peri-urban areas while two were within urban areas. Three

of the five control sites were established within the southern suburbs of Brisbane, Logan

and Redland Shires; to supplement the small number of large contiguous forest patches

below 500 m on the Gold Coast. The Gold Coast has 9669.2 ha of open space and

reserves that were available as study sites; of which 6650.9 ha was captured within the

45 sites. A systematic grid was placed over all sites with sampling points (plots)

identified at all intersection points. The number of plots to be surveyed at each site was

determined by patch size using a hierarchical grid cell size, ranging from 250 m in small

forest patches to 2000 m in larger sites to allow for sufficient environmental variation.

In total 90 plots were selected from all 45 sites (Appendix 3 for GPS locations)

composed of 13 urban sites with 20 plots, 14 peri-urban sites with 23 plots, 13 rural

sites with 21 plots and five control sites with 31 plots. Sites varied in size (2- 1699 ha),

shape, topographic location as well as position along an urban gradient. At each plot a

fixed area methodology was used (sensu Gibbons and Lindenmayer 2002), to sample

relevant parameters (described below). Fixed area methodology allows for ease in

determining if a tree should be sampled or not; plots can be permanently marked and all

Page 68: A thesis submitted for the fulfilment for the requirements

45

trees within a plot can be catalogued enabling the estimation of density. A number of

environmental and disturbance variables were also quantified in each plot (Table 3.2).

Plot sampling protocol

This study quantified the abundance and distribution of all species from within five

genera belonging to the family Myrtaceae; Angophora, Corymbia, Eucalyptus,

Lophostemon and Syncarpia, hereafter referred to as ‘eucalypts’. At each plot all

eucalypt trees within a 30 m radius from the plot midpoint and with a diameter at breast

height over bark (dbh) ≥10 cm were identified and measured to determine vegetation

community structure and quantify the status of hollow-bearing trees. Eucalypt

identification followed Brooker and Kleinig (2004) and Euclid (Centre for Plant

Biodiversity 2006), on the basis of bark, fruit bud and leaf characteristics. The area of

each circular plot was 0.28 ha resulting in a total survey area of 25.4 ha for all 90 plots.

Hollow-bearing trees detected on each plot were categorised into one of seven

senescence types using the framework of Whitford (2002) (Figure 3.1). The senescence

types were modified to focus on living trees as opposed to dead trees; therefore, S1 is a

completely healthy tree, S2-S6 varying degrees of senescence similar to Whitford

(2002) but replacing ‘dead’ with ‘living’. The justification for this was based on the

higher value of living trees to fauna than dead trees (Moloney et al. 2002). Tree-

hollows in each tree were assessed using binoculars and recorded as one of seven

hollow forms (Figure 3.2). A 10 cm diameter entrance threshold was selected based on

previous studies that found that tree hollows with entrances of this size were large

enough to accommodate the larger species of gliders such as the yellow-bellied glider

Petaurus australis and the greater glider Petauroides volans (Kehl and Borsboom 1984;

Mackowski 1984; Ross 1998; Eyre 2005; Wormington et al. 2005). While vertebrate

fauna do utilise hollows <10 cm, these hollows may not be easily detected using ground

based surveys (Harper et al. 2005a). One way to counter this problem would be to fell

trees (Gibbons and Lindenmayer 2002) or undertake double sampling in the form of tree

climbing (Harper et al. 2004). The first option was not considered appropriate or

possible for this study, while time and financial constraints made the latter option

unfeasible. It is also possible to overestimate the number of trees hollows due to the

fact that a tree will often contain many openings which are blind and not of a suitable

depth for occupation (Lindenmayer et al. 1990). Despite the potential bias in estimating

Page 69: A thesis submitted for the fulfilment for the requirements

46

hollow presence in trees, ground based surveys are widely used in a number of different

Australian environments (Harper et al. 2004; Eyre 2005; Koch 2008).

Table 3.1: Summary of tree, tree-hollow, environmental and disturbance variables and

the urbanisation gradient quantified at each plot.

Variable Description of variable

Tree variables Tree species Angophora, Corymbia, Eucalyptus, Lophostemon,

Syncarpia. dbh ≥10 cm Diameter at breast height for all trees. Tree form & senescence See Figure 3.2. Number of hollows Number of hollows observable in a tree from the ground. Hollow variables Hollow height (m) Height of hollow in tree. Hollow type See Figure 3.3. Hollow entrance size (cm) Approximation of entrance width. Hollow density Equation 3.1 Environmental variables Angle of slope θ Calculated from GIS layers (Equation 3.2). Aspect (°) Directional orientation. Elevation (m) Obtained from GIS mapping layers. Tree disturbance variables

Fire scar Presence or absence of fire scars on trees. Termite damage Presence or absence of termite activity. Wind damage Presence or absence of broken limbs. Epicormic growth Presence or absence of epicormic growth. Disturbance index Presence/absence fire and logging history from historical

aerial imagery. Urbanisation index Calculated from GIS data on infrastructure, cadastre and

regional ecosystems (Equation 3.3).

Page 70: A thesis submitted for the fulfilment for the requirements

47

Figure 3.1: Tree form characteristics and senescence adapted from Whitford (2002).

S1. Mature living tree S2 Mature, living tree with a dead or broken top S3. Living tree

with most branches still intact S4. Living tree with the top 25% broken off S5. Living

tree with the top 25-50% broken away S6. Living tree with the top 50-75% broken away

S7. A solid dead tree with 75% of the top broken away S8. Hollow stump.

S5 S6 S7 S8

Alive/dying Alive/dying Dead standing tree Dead stump

S1 S2 S3 S4

Alive/healthy Alive/healthy with Alive with Alive/dying minimal crown damage damaged crown

Page 71: A thesis submitted for the fulfilment for the requirements

48

Figure 3.2: Tree hollow types (Lindenmayer et al. 2000).

The diameter of each hollow entrance was estimated to the nearest 5 cm, using 10 cm as

the minimum size. Fissure hollow size estimates were only based on the width

(smallest dimension) of each fissure. Tree height and hollow height were measured

using a rangefinder (True Pulse 200, Laser Technology Inc.). In some instances (i.e.

steep slopes and rugged terrain) it was not possible to use the rangefinder and in these

cases tree and hollow heights were estimated using prior knowledge of tree heights

measured in other sites. GPS coordinates for each hollow-bearing tree were recorded

and each tree was fitted with an aluminium tag carrying a unique identification number.

Hollow-bearing tree density D (number of trees / ha) per site was calculated as the sum

of all hollow-bearing trees HBTs from all plots (p) at a site divided by the sum of the

total plot area Ap surveyed at each site (Equation 3.1).

Page 72: A thesis submitted for the fulfilment for the requirements

49

=

== n

pp

n

pp

A

HBTD

1

1 (3.1)

Landscape variables

Angle of slope (Ө) was calculated from GIS layers using Equation 3.2, where opposite

represents the change in elevation (m) from the lowest to highest contour line at each

plot, and adjacent represents the straight line distance (m) between these contour lines.

Thus the higher the value the steeper the slope.

𝑡𝑎𝑛θ = 𝑜𝑝𝑝𝑜𝑠𝑖𝑡𝑒𝑎𝑑𝑗𝑎𝑐𝑒𝑛𝑡

(3.2)

Landscape disturbance history (logging and fire) was extrapolated from historical aerial

images at 1:100000 supplied by the Queensland Government (Queensland Government

2008; Queensland Government 2011). These data span a 54 year period (1955-2009)

and contain 22 years of data related to the current study area (i.e. photography was not

available for each year). A disturbance index was calculated from the number of

logging/fire events at each site as a function of the 22 years with available site relevant

data. Logging was classified where images showed the removal of trees due to clear

felling or large canopy gaps from where selective logging had occurred. The same

principle was applied to fires across the landscape; where large scale burnt areas were

observable from images. These data were then square root transformed before analysis.

Urbanisation gradient

It is evident from the published literature that the classification of urban gradients

generally weight the importance of commonly used parameters differently. Some rely

upon road type and density (Brady et al, 2009), vegetation cover and land development

(Germaine and Wakeling 2001), human population density and land-use category

(Guntenspergen and Levenson 1997) and the ratio of the density of people divided by

the proportion of urban land cover (PEOP/%URB) (Hahs and McDonnell 2009).

Therein, for this study the classification of sites along the urbanisation gradient was

based on the spatial analysis (using ArcGIS) of a series of land use and development

Page 73: A thesis submitted for the fulfilment for the requirements

50

parameters supplied by the Gold Coast, Brisbane, Redland and Logan Councils.

Spatially explicit measures of housing density on individual properties (cadastre), land

use, service infrastructure networks (e.g. roads, rail, canals, electricity) and remnant

vegetation were used to derive an urbanisation index (Appendix 4) (Figure 3.3). The

urbanisation index (U) was determined for each site by first placing a 250 m buffer

around the periphery of the site and then calculating firstly, the proportional areal extent

of transformed habitat (C) within the 250 m buffer (as prescribed by Biodiversity

Planning unit 2002). Secondly, the housing density data extracted from the cadastre

information within each buffer were summed and then normalised by dividing this total

by the maximum housing density across all sites. The transformed extent for each site

was then adjusted based on the normalised housing density for built up areas within the

sites buffers to calculate the urbanisation index (Equation 3.3), where a greater index

value represents a more urbanised site.

𝑈 = ∑𝑑𝑒𝑛𝑠𝑖𝑡𝑦𝑀𝐴𝑋 𝑑𝑒𝑛𝑠𝑖𝑡𝑦

× 𝐶 (3.3)

Once the urbanisation index had been calculated quantile breaks in ArcGIS were used to

identify three primary categories to represent the urban gradient namely; urban, peri-

urban and rural. Control sites were assigned a priori but were also subsequently

assigned an urban gradient category based on their individual urbanisation indices.

Values 0-20 were rural, 21-360 peri urban and >360 urban. Site and tree predictor

variables were also included in analyses while landscape disturbance history (logging)

was extrapolated from historical 1:100000 aerial images (Queensland Government

2008) to derive a logging index (Table 3.2).

Page 74: A thesis submitted for the fulfilment for the requirements

51

Figure 3.3: The 250 m buffer placed around sites to capture housing, cadastre and

regional ecosystem data.

Page 75: A thesis submitted for the fulfilment for the requirements

52

Data analysis

Predictor variables known to have a strong biological influence on tree hollows were

analysed based on a priori investigations. Spearman rank correlation measures were

used to determine similarities between sites using BIOENV in Primer E version 6.1.11

(Clarke and Warwick 2001a). A correlation coefficient value |r| > 0.7 was chosen to

identify pairs of highly correlated variables (Garden et al. 2010). BIOENV uses all

of the available environmental variables to find the combination that ‘best explains’ the

patterns in the biological data (Clark and Ainsworth 1993). No explanatory variables

were strongly correlated and thus all were retained. The decision to retain or remove

variables was also based on the known or perceived ecological significance of the

variables. Significance levels were set at 0.05 for all analyses.

The use of different types of transformations is recommended where explanatory

variables are of dissimilar nature (Zuur et al. 2010); such as those used in this analysis.

An arcsine square root transformation was undertaken for explanatory variables with

percentage values, while a log transformation was undertaken for all other explanatory

variables requiring transformation due to large differences in scale. No transformations

were undertaken on the dependent variable.

The relationship between dbh and the number of hollows a tree contained used a

standard linear regression. Univariate analysis using MANOVA with Tukey’s post-hoc

analysis was undertaken in order to compare the variation between the number of

different hollow types and urbanisation gradient. Statistical analysis were conducted

using Statistica Version 7 (Statsoft. 2005).

The effects of environmental variables, disturbance, and the urbanisation gradient on the

density of hollow-bearing trees were analysed using non-metric Multidimensional

Scaling (nMDS) and Principle Co-ordinate Analysis (PCO). Non-metric

Multidimensional Scaling was used to assess the similarities amongst sites. The nMDS

represents the sites in a multidimensional space, so that the distances between the sites

accurately represent the original proximity measures. The similarity matrix required for

the nMDS was first calculated using the Bray-Curtis similarity index. A Pearson

correlation was then applied to the data to find the tree species which were accounting

for most of the variation in the density of hollow-bearing trees along the gradient. A

Kruskal’s stress test for measuring of goodness of fit was performed prior to conducting

Page 76: A thesis submitted for the fulfilment for the requirements

53

the PCO to measure the distances between samples on the ordination to match the

corresponding dissimilarities in community structure. Univariate analysis using

ANOVA with Fisher Least Significant Difference (LSD) post-hoc analysis was

undertaken in order to compare the variations between predictor variables (dbh) and the

dependant variables (number of hollow-bearing trees, hollow type, hollow size, number

of hollows, tree form and the urbanisation gradient). A Students t-test was used to

investigate the relationship between the dbh of trees with or without hollows. Statistical

analyses were conducted using Primer E (Clarke and Warwick 2001a) and Statistica

(Statsoft. 2005).

Results

Hollow bearing trees

A total of 6048 trees from 34 eucalypt species were recorded from the 45 sites (90

plots) (Table 3.2). More than 95% of all trees were one of 17 species, with only five

species Lophostemon confertus, Corymbia intermedia, Eucalyptus carnea, Eucalyptus

crebra and Eucalyptus propinqua making up 51% of the woody community in the

urban forest patches. Corymbia intermedia was the most common species and was

found at 98% of all sites. A total of 916 (14.9%) hollow-bearing trees were recorded,

containing 2159 hollows across all sites, comprised of 24 eucalypt species including

unidentifiable dead trees, with each site containing an average of six species of hollow-

bearing tree species (Table 3.3). Most of which were L.confertus (142) followed by

standing dead trees (124) and E. crebra (91) (Table 3.2). Along the urbanisation

gradient 36% of hollow-bearing trees were contained within rural sites, 29% in control

sites, 22% in urban and 11% in peri-urban sites (Table 3.2). Site area, the number of

plots per site, number of hollows, hollow-bearing tree species and hollow density per

site, reflected the variation in hollow availability along the urbanisation gradient (Table

3.3). Total area of all sites surveyed was 6650.9 ha comprising five control, seven peri-

urban, 20 rural and 13 urban sites. Control and rural sites were found to have lower

densities of hollow-bearing trees than both peri-urban and urban sites (Table 3.3)

however, this was not significant (p = 0.967). Mean hollow-bearing tree density was

found to be 37.47 (± 3.13 S.E.) hollow-bearing trees per hectare across all sites.

Standing dead trees had the most hollows (15.4%) followed by L.confertus (11%) and

E. propinqua (10.1%) (Table 3.4). Lophostemon confertus also represented 45% of all

butt-hollows (Table 3.4).

Page 77: A thesis submitted for the fulfilment for the requirements

54

Table 3.2: A summary of species dominance and hollow-bearing tree prevalence within eucalypt communities recorded from 45 forest

patches along an urbanisation gradient. HBT’s = hollow-bearing trees.

n HBT’s by Gradient Category

Tree species Common name N trees N sites N HBT’s Control Peri urban Rural Urban L.confertus Brush box 968 35 142 28 36 44 34 C.intermedia Pink bloodwood 778 44 64 6 5 35 18 E.carnea Broad-leaved white mahogany 563 30 76 21 1 42 12 E.crebra Narrow-leaved iron-bark 527 33 91 22 8 21 40 E.propinqua Small-fruited grey-gum 477 33 80 25 11 34 10 C.citriodora.v Spotted gum 417 23 54 10 4 29 11 C.gummifera Red bloodwood 347 16 35 23 2 10 0 E.tindaliae Queensland white stringy-bark 299 15 36 29 1 4 2 E.microcorys Tallowwood 293 26 71 22 5 38 6 L.suaveolens Swamp box 277 14 20 2 8 4 6 E.pilularis Blackbutt 216 17 29 0 0 8 21 E.siderophloia Grey-leaved iron-bark 162 15 15 7 1 3 4 E.racemosa Scribbly gum 114 6 36 14 6 0 16 Dead Dead 124 39 124 45 11 48 20 E.resinifera Red mahogany 96 10 1 1 0 0 0 A.woodsiana Rough barked apple 61 3 0 0 0 0 0 E.grandis Flooded gum 40 5 2 0 0 2 0 E.longirostrata Grey gum 40 4 0 0 0 0 0 C.trachyphloia Brown bloodwood 36 3 2 1 0 1 0 E.acmenoides White mahogany 30 6 8 8 0 0 0 E.eugenioides Thin-leaved stringy-bark 31 5 4 0 3 0 1 E.tereticornis Forest red gum 25 8 9 0 2 3 4 E.baileyana Bailey’s stringy-bark 26 1 4 4 0 0 0 E.major Grey gum 20 7 9 1 0 4 4

Page 78: A thesis submitted for the fulfilment for the requirements

55

Table 3.2 continued: A summary of species dominance and hollow-bearing tree prevalence within eucalypt communities recorded

from 45 forest patches along an urbanisation gradient. HBT’s = hollow-bearing trees.

n HBT’s by Gradient Category

Tree species Common name N trees N sites N HBT’s Control Peri urban Rural Urban E.robusta Swamp mahogany 24 1 0 0 0 0 0 E.saligna Sydney blue-gum 14 2 1 0 1 0 0 C.henryi Large-leaved spotted-gum 11 3 1 0 1 0 0 E.moluccana Gum topped box 10 4 0 0 0 0 0 E.biturbinata Large-fruited grey-gum 10 4 1 0 0 1 0 C.tessellaris Moreton Bay ash 4 1 0 0 0 0 0 E.fibrosa Broad-leaved red iron-bark 3 1 0 0 0 0 0 E.seeana Narrow-leaved red-gum 3 2 0 0 0 0 0 E.dura Eucalyptus dura 1 1 0 0 0 0 0 S.glomulifera Turpentine 1 1 0 0 0 0 0 Totals 34 species 6048 916 269 244 194 209

Page 79: A thesis submitted for the fulfilment for the requirements

56

Table 3.3: A description of hollow-bearing (HBT) tree attributes found across all sites sorted by gradient then area. Gradient: R= rural, U=urban, P=peri-

urban, C=control. Control sites have also been classified in relation to their position along the urbanisation gradient.

Site name Gradient No:

plots Area (ha) N trees tagged

Dominant species

Dominant HBT Species

Mean dbh of HBT's

N HBT species N HBT's

N Hollows

HBT density per hectare

Nerang Forest Reserve C (p) 5 1668.76 387 E.carnea E.crebra 42.6 12 86 232 61.4 Karawatha Forest Reserve C (p) 7 824.33 400 E.tindaliae E.tindaliae 38.1 13 53 88 27.0 Bonogin Conservation Area C (u) 6 725.07 301 L.confertus L.confertus 51.2 8 57 99 33.9 Venman Bushland National Park C (p) 6 434.24 236 E.propinqua E.propinqua 52.5 11 40 152 24.4 Daisy Hill Forest Reserve C (u) 5 432.06 235 E.tindaliae Dead 41.6 8 33 67 23.6 Pimpama Conservation Area P 2 565.87 152 C.intermedia C. intermedia 21.5 1 1 1 1 Mt Nathan Conservation Area P 3 266.07 135 C.citriodora.v C.citriodora.v 41.0 6 28 63 33.3 Syndicate Road P 3 185.31 164 C.gummifera E.propinqua 60.4 11 33 73 39.2 Kirken Road P 1 176.87 68 E.racemosa E.racemosa 48.9 4 14 37 50.0 Woody Hill, Hart's Reserve P 2 91.08 117 L.confertus L.confertus 51.0 7 25 44 44.6 Galt Road P 1 33.67 41 E.microcorys E.propinqua 39.6 7 15 32 53.6 Elanora Wetland P 1 26.61 85 L.suaveolens L.suaveolens 42.8 3 8 11 28.6 The Plateau Reserve P 1 26.53 83 L.confertus L.confertus 28.1 8 26 55 92.9 Aqua Promenade P 2 13.10 188 C.intermedia E.propinqua 52.4 8 34 83 60.7 Elanora Conservation Area P 2 13.06 210 L.confertus L.confertus 46.3 8 31 68 55.4 Hellman Street P 1 9.15 53 C.intermedia C.citriodora.v 43.6 4 8 13 28.6 Reserve Road Parklands P 1 7.66 73 C.intermedia E.crebra 43.7 4 5 7 17.9 Piggabeen Road P 1 2.97 72 E.pilularis E.microcorys 45.5 3 4 5 14.3 Johns Road P 1 2.11 85 C.intermedia E.propinqua 38.5 6 12 16 42.9 Numinbah Road R 2 157.55 92 C.gummifera E.microcorys 43.1 6 16 34 28.6 Austenville Road R 3 24.98 318 L.confertus E.microcorys 64.2 6 29 67 34.5 Behms Road R 1 14.29 71 C.intermedia C.intermedia 58.9 4 11 41 39.3 Andrews Reserve R 1 10.82 72 E.pilularis E.pilularis 69.6 4 8 27 28.6 Carrington Road R 1 5.99 47 L.confertus E.propinqua 58.4 5 11 22 39.3

*highlighted areas represent the congruency between the dominant tree species found on a site and the dominant HBT species found on a site.

Page 80: A thesis submitted for the fulfilment for the requirements

57

Table 3.3 continued: A description of hollow-bearing tree (HBT) attributes found across all sites sorted by gradient then area. Gradient: R= rural, U=urban, P=peri-urban, C=control. Control sites have also been classified in relation to their position along the urbanisation gradient.

Site name Gradient

No:

plots Area (ha)

N trees

tagged

Dominant

species

Dominant HBT

Species

Mean dbh of

HBT's

N HBT

species

N

HBT's

N

Hollows

HBT density per

hectare

Wongawallen Conservation Area R 2 383.67 71 E.microcorys L.confertus 51.3 5 24 65 42.8

Upper Mudgeeraba Conservation Area R 2 193.24 144 C.citriodora.v Dead 40.2 4 5 12 8.9

Trees Road Conservation Area R 2 53.32 203 E.crebra E.carnea 47.4 7 26 47 46.4

Stanmore Park R 1 50.87 111 L.suaveolens E.siderophloia 51.0 2 2 6 7.1

Reedy Creek Conservation Area R 3 44.85 178 L.confertus E.carnea 52.5 8 22 46 26.1

Tallowwood Road R 1 14.24 93 L.confertus E.microcorys 47.8 6 13 32 46.4

Tomewin Road R 1 5.10 51 E.microcorys E.microcorys 77.3 5 16 70 57.1

Gilston Road R 1 4.04 69 L.suaveolens C.citriodora.v 36.2 6 10 14 35.7

Olsen Avenue U 2 38.94 134 A.woodsiana E.pilularis 60.0 2 2 3 3.6

Burleigh Headland National Park U 1 27.32 86 L.confertus L.confertus 57.3 4 26 68 92.8

Burleigh Ridge U 3 17.89 159 L.confertus E.crebra 52.8 5 34 102 40.5

Pacific Highway, Currumbin U 2 16.89 205 L.confertus E.pilularis 60.1 8 33 92 58.9

Brushwood Ridge Reserve U 2 16.87 140 C.citriodora.v E.carnea 38.5 6 22 50 39.2

Herbert Park U 3 16.45 198 C.intermedia E.propinqua 50.3 8 23 56 27.4

Tugun Conservation Area U 1 14.38 67 E.pilularis E.pilularis 78.4 4 9 14 32.1

Summerhill Court Reserve U 1 9.53 52 E.microcorys E.microcorys 54.3 6 14 28 50.0

Kincaid Road Reserve U 1 7.52 50 L.confertus L.confertus 41.8 3 5 5 17.9

Pacific Pines Reserve U 1 5.89 82 E.crebra L.confertus 20.2 2 6 6 21.4

Miami Conservation Area U 1 4.41 126 E.carnea L.confertus 34.9 4 5 6 17.9

Burleigh Knoll National Park U 1 4.15 66 E.racemosa E.racemosa 67.0 3 25 91 89.3

Lakala Street Reserve U 1 3.23 78 C.intermedia E.pilularis 39.3 5 6 9 21.4

Total 45 92 6650.9 6048 14 dominant species 13 HBT species

Mean dbh across all sites 48.5 (±1.84 S.E.)

Mean n of HBT species across all sites 5.7 (±0.4 S.E.) 916 2159

Mean of HBT density per hectare across all sites 37.47 (±3.13 S.E.)

*highlighted areas represent the congruency between the dominant tree species found on a site and the dominant HBT species found on a site.

Page 81: A thesis submitted for the fulfilment for the requirements

58

Table 3.4: Summary of trees species and hollow type with the total number of hollows found across all sites.

Tree species Bayonet Branch end Branch main Butt Trunk main Trunk top Fissure Total

Dead 113 119 1 27 24 34 12 330

L.confertus 51 38 0 122 12 4 9 236

E.propinqua 104 93 4 6 8 2 0 217

E.crebra 106 74 0 4 3 0 4 191

E.microcorys 94 57 5 20 6 1 5 188

E.carnea 80 51 1 25 9 4 4 174

C.citriodora.v 81 49 3 10 7 5 2 157

E.racemosa 38 42 6 8 8 2 1 105

C.intermedia 38 34 0 7 4 0 1 84

C.gummifera 49 21 0 3 5 2 0 80

E.tindaliae 40 26 1 8 3 1 1 80

E.pilularis 20 36 1 11 3 0 2 73

E.tereticornis 17 36 1 3 2 0 1 60

L.suaveolens 23 12 0 9 2 0 1 47

E.siderophloia 24 11 0 0 2 0 0 37

E.major 16 6 0 1 0 0 0 23

E.grandis 3 15 0 0 0 1 0 19

E.acmenoides 3 5 2 3 1 1 0 15

E.resinifera 8 3 0 2 0 0 0 13

E.baileyana 2 2 0 0 1 0 0 5

C.trachyphloia 1 4 0 0 0 0 0 5

E.eugenioides 2 0 0 0 0 0 0 2

E.moluccana 1 0 0 0 0 0 0 1

C.henryi 0 1 0 0 0 0 0 1

Total 914 735 25 269 100 57 43 2143

Page 82: A thesis submitted for the fulfilment for the requirements

59

Small hollows within the 10 cm size category were the most abundant in all hollow-bearing

trees comprising 50.4% of all hollows detected (Figure 3.4). The relationship between

hollow size and type was also significant (F = 114.4; d.f. = 6, 2136; p < 0.01), with all hollow

types differing from each other (Figure 3.5). Trunk top hollows were larger than fissures and

bayonets were smaller than trunk-main, branch-main, butt and branch end hollows. The

relationship between tree form and the presence of a hollow was significant (F = 3.4; d.f. = 6,

5917; p < 0.01) in that forms one and two were significantly different and contained more

hollows than forms 3-7. Over 69% of hollow-bearing trees were found to range between dbh

30-60 cm, mean 49.3 cm (± 0.75 S.E.), while 78% of non-hollow bearing trees ranged

between dbh 20-40 cm, mean 26.8 cm (± 0.18 S.E.) and the mean dbh range for all trees was

30 cm (± 0.22 S.E.) range 10-159 cm (Figure 3.6). This difference between the dbh of trees

with hollows and those with no hollows present was significant (t = - 28.75; d.f. = 879; p <

0.01). Trees with a larger dbh were also found to have more hollows (R2 0.137; p < 0.05).

Figure 3.4: The number of hollows per hollow size category.

0

200

400

600

800

1000

1200

10 15 20 25 30 35 40 45 50 60 65 70 80 90 100 150

No.

of h

ollo

ws

Hollow size (cm)

Page 83: A thesis submitted for the fulfilment for the requirements

60

Butt Branch-end Bayonet Fissure Trunk-main Trunk-top Branch-main

Hollow type

5

10

15

20

25

30

35

40

45H

ollo

w si

ze (c

m)

Figure 3.5: The relationship between hollow type and size (mean ±S.E). The number of

hollows for each hollow type are; Butt 269, Branch-end 735, Bayonet 914, Fissure 48, Trunk-

main 98, Trunk-top 54, Branch-main 25.

Page 84: A thesis submitted for the fulfilment for the requirements

61

Figure 3.6: Number of trees surveyed in each dbh size category. All trees represent the

combination of hollow-bearing and non-hollow bearing trees. Results for hollow-bearing

trees are then presented separately.

The impact of environmental variables, disturbance and the urbanisation gradient on hollow-

bearing tree species abundance and hollow type

Hollow type was not influenced by the urbanisation gradient (F = 0.474; d.f = 18, 280; p =

0.97), however, there was a difference in the number of hollow types more generally in that

there were significantly more butt, branch end and bayonet hollow types found (F = 35.8; d.f

= 18, 280; p = < 0.05) (Figure 3.7). Site area influenced the pattern within hollow-bearing

tree communities. Area of forest patches was strongly linked to control sites (PCO axis 2),

while the urbanisation gradient appeared to discriminate urban sites from non-urban sites

(PCO axis1) (Figure 3.8). There was no significant difference in the density of hollow-

bearing trees among sites along the urbanisation gradient (F = 0.118; d.f. = 3, 41; p = 0.95).

0

200

400

600

800

1000

1200

1400

1600

1800

2000N

o of

tree

s sur

veye

d

dbh category (cm)

Hollow-bearing treesAll trees

Page 85: A thesis submitted for the fulfilment for the requirements

62

ButtBranch end

BayonetFissure

Trunk mainTrunk top

Branch main

Hollow type

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7R

elat

ive

abun

danc

e of

hol

low

type

Control Peri urban Rural Urban

Figure 3.7: The relationship between hollow type and the urbanisation gradient (mean ±

S.E.).

Page 86: A thesis submitted for the fulfilment for the requirements

63

Figure 3.8: Spatial arrangement of sites in the multidimensional space based on the density

of hollow-bearing trees. The urbanisation gradient and site area account for 45.5% (PCO1)

and 33.1% (PCO2) of the variation respectively. Legend: u = urban, r = rural, c = control, p

= peri-urban.

Discussion

The distribution and abundance of habitat features is an important driver of community

diversity particularly in areas where species are heavily dependent on structurally complex

features (Woinarski et al. 1997; Tanabe et al. 2001). Therefore, the prevalence of hollow-

bearing trees in an urbanising landscape may be pivotal to ensuring the persistence of urban

diversity. This study found that the density of hollow-bearing trees across all sites averaged

approximately 37.47 HBTs/ha but ranged from 1-92.9 HBTs /ha. This figure is considerably

higher than the minimum number (4–6 /ha) recommended for managed forests in south-east

Queensland (Lamb et al. 1998). This figure is also higher than the 5.8 HBTs/ha found in the

only other Australian study on hollow-bearing tree density within an urban landscape (Harper

et al. 2005a). However, this average density of hollow-bearing trees may overestimate the

Page 87: A thesis submitted for the fulfilment for the requirements

64

availability of hollows, particularly in terms of their use by fauna, as small, bayonet hollows

accounted for 50.4% of all hollows found. While small hollows are a valuable resource

(Kehl and Borsboom 1984; Mackowski 1984; Ross 1998; Eyre 2005; Wormington et al.

2005), hollows with a diameter > 10 cm are necessary to meet the requirements of larger

hollow-using species (Murphy et al. 2003).

Of the 916 hollow-bearing trees found, standing dead trees were represented by 13.5% (n124)

of hollow-bearing trees found, which is substantially lower than the 50% reported for

production forests in south-east Queensland (Moloney et al. 2002). The high representation

of standing dead trees found during that particular study was likely to have been in response

to extensive silvicultural practices including poisoning and ringbarking carried out in these

forests (Moloney et al. 2002). While Moloney et al. (2002) suggest that live hollow-bearing

trees are of greater importance to wildlife in Australia, standing dead trees are important

habitat features. This is widely supported more generally with certain species displaying a

preference for stags as denning and nest sites (Lindenmayer et al. 1991a; Eyre 2004;

Cameron 2006). Tree-form was found to be related to a tree being hollow-bearing, as forms

one and two were significantly different to forms 3-7. This study is consistent with previous

studies established that hollows are most likely to occur in large trees with moderately

senescent crowns (Whitford 2002). However, fauna have been observed on more occasions

(45%) leaving hollows found in the crowns of trees in forms one and two than other forms

(Smith and Lindenmayer 1988; Lindenmayer et al. 1991a). This study found that live

hollow-bearing trees were found to contain more hollows than standing dead trees. This

could be due to the modification of tree senescence from Whitford’s (2002) classification to

focus on living trees as opposed to dead trees.

In general, standing dead trees have a higher likelihood of containing hollows than live trees

(Eyre 2005; Harper et al. 2005a; Beyer et al. 2008). The persistence of these dead trees

across the landscape, however, is likely to be less than that of the living cohort of hollow-

bearing trees due to the impacts of fire and wind. Within the dry sclerophyll forests of south

east Queensland, the longevity of standing dead trees has been estimated at ~50 years

(Moloney et al. 2002), which is significantly less than the time taken for living trees to reach

maturity and form hollows (Wormington et al. 2003). Dead trees are also highly susceptible

to destruction by fires in dry forests (Lamb et al. 1998). Fire did not appear to be an

important factor in influencing distribution of hollow-bearing trees in this study and may

Page 88: A thesis submitted for the fulfilment for the requirements

65

actually be suppressed in urban forest patches given the safety concerns associated with fires

and human settlement in Australia (Shea et al. 1981; Jurskis et al. 2003). Consequently, even

though standing dead trees are less abundant in urban forest patches they may persist for

longer periods than those in managed forests.

Lophostemon confertus dominated the woody tree communities across urban habitat forest

patches and were also well represented as hollow-bearing trees. There was a significantly

greater prevalence of butt hollows in L.confertus than any other species. Lophostemon

confertus is wind dispersed (Lee et al. 2008) and is an early coloniser in rainforests where

fire occurrence is low (Burrows 2002). As fire has largely been removed from the urban and

peri-urban landscape it is a species that appears to be dominating these sites. Unlike eucalypt

species where bark thickness ranges between 13-35 mm, L. confertus with a maximum bark

thickness of 3 mm has a reduced regenerative potential, making them susceptible to fire

damage (Burrows 2002). This may facilitate the formation of butt hollows in larger

L.confertus that have been repeatedly burnt. Furthermore, given that butt-hollows are located

at ground level, the likelihood of predation on hollow-using vertebrate fauna would be high,

thus making them a less than desirable choice for denning. However, two common brushtail

possums Trichosurus vulpecula and a Scaly-breasted Lorikeet Trichoglossus chlorolepidotus

were observed leaving small butt-hollows located within L.confertus and Corymbia

gummifera respectively, during the course of this study.

Diameter at breast height for hollow-bearing trees ranged from 10-160 cm across all sites, but

the mean dbh for hollow-bearing trees was significantly larger than that for the entire wooded

community within urban forest patches. These findings support those of previous studies

confirming that larger, older trees have a greater likelihood of containing hollows

(Wormington et al. 2003; Cameron 2006; Koch et al. 2008b), as well as having a greater

number of hollows per tree (Whitford 2002; Wormington et al. 2003). The size class

distribution of hollow-bearing trees was also different to that of non-hollow trees with the

latter appearing to be of a relatively even age, (78% of all trees found within the 20-40 cm

dbh range). This potentially reflects the intense logging history of the region (Chapter 1)

where historical forestry practices removed most large mature trees from forest patches and

entire fragments were cleared for housing within urban areas. With 92% of Australians

living in cities, urban fragments play a critical role in conserving biodiversity (Manning et al.

2006). Town planners should then be able to gain a basic understanding of the patterns and

Page 89: A thesis submitted for the fulfilment for the requirements

66

processes that affect biodiversity as a whole when investigating urban ecosystems (Alvey

2006). Additional options will then become available for the effective management of urban

bushland in order to develop an integrated management approach that places conservation

reserves in the context of the overall landscape (Saunders et al. 1991). A historical

perspective is critical at the site-specific scale as sites that have undergone anthropogenic

modifications require an understanding of their current ecology to predict future trends (De

Wet et al. 1998).

In this current study, the ‘urbanisation gradient’ (45.5%) was found to be having a stronger

effect on hollow-bearing tree density than ‘site area’ (33.1%) supporting the original

hypothesis put forward for this chapter, that the urbanisation gradient was having an impact

on hollow-bearing tree density. These results are consistent with other studies assessing

biodiversity patterns across urban gradients, in that the gradient strongly influences biota

(Williams et al. 2005a; McKinney 2006; Brady et al. 2009), and, sites in urban centres

appeared more depauperate of species than those in peri-urban or rural areas (Blair 1999;

Cornelius and Hermy 2004; Alvey 2006). A structurally complex matrix within a human-

modified landscape, however, can provide habitat resources and increase species richness in

modified landscapes (Brady et al. 2011).

The urbanisation gradient was found not to have an effect on the type of hollow found.

Given that small bayonet type hollows represented 42% of all hollow types along the

gradient, it is conceivable that there is a shortage of the larger hollow types required to meet

the needs of some vertebrate species. Therein, the somewhat controversial option to address

this shortage would be the installation of species appropriate artificial nest-boxes (Goldingay

et al. 2007). This is a hotly debated issue (Lindenmayer et al. 1991c; Spring et al. 2001;

Harley and Spring 2003; Lindenmayer et al. 2009) as the use of nest boxes has been

questioned due to installation and maintenance costs, monitoring and nest-box deterioration.

The use of next-boxes in large scale forestry situations has been questioned due to significant

logistic constraints and here the retention of hollow-bearing trees is likely to be more viable

in the long-term (Lindenmayer et al. 1997; 2002). In contrast, nest-box use has been found to

be higher in urban remnants (18.2%) (Harper et al. 2005b; Davis et al. 2013) than in large

contiguous forests (7%) (Menkhorst 1984). There are two possible explanations for nest box

use patterns. The first is that it reflects a higher abundance of natural hollows in larger and

more contiguous forests than urban forest patches. The second is that fauna have become

Page 90: A thesis submitted for the fulfilment for the requirements

67

relatively isolated in fragmented urban forest patches resulting in community overcrowding.

There is some support for this from recent work documenting aggression in hollow-nesting

birds in the Sydney region of Australia (Davis et al. 2013). However, an understanding of

hollow-bearing tree resource availability, the presence of species that require hollow-bearing

trees and their species specific hollow requirements is required prior to introducing nest-

boxes into any environment (Harper et al. 2005b; Davis et al. 2013).

The Gold Coast as a city with its own unique set of geographical, historical and economic

factors, will have different results than cities located in different regions not only in Australia

but globally as well. This variation in results between study sites have been well documented

in a review of 105 studies across urbanisation gradients (McKinney 2008). From results

presented here it seems that many forest patches on the Gold Coast have been left relatively

in-tact across the gradient, and have over time, gradually been reduced in size rather than

cleared. This may be due to the rapid urbanisation that the Gold Coast has undergone in the

last few decades (Spearbritt 2009).

With the understanding of the complexity of hollow-bearing tree management and

amelioration across the landscape, inevitably comes the acknowledgement of the time lag

between hollow formation and availability. Therefore, retaining the current inventory of

hollow-bearing trees is paramount to maintaining biodiversity (Eyre et al. 2010). However,

the conservation of hollow-bearing trees alone will only answer the immediate need for this

resource, it is therefore imperative to plan for future needs by conserving regrowth as

recruitment trees for the long-term sustainability of hollow-bearing trees (Gibbons and

Lindenmayer 2002).

Page 91: A thesis submitted for the fulfilment for the requirements

69

Chapter 4: The impacts of landscape variables on hollow-bearing trees along an urbanisation gradient.

Introduction

Second only to agriculture, urbanisation is the single most damaging, persistent and

confounding form of anthropogenic pressure exerted upon natural systems globally

(Vitousek 1997; McKinney 2006; Gaston 2010a). Within Australia, approximately

50% of forests have been lost or severely modified since European settlement, with over

80% of eucalypt forests having suffered from anthropogenic activities, leaving much of

Australia’s remaining forest severely fragmented (Bradshaw 2012). In heavily settled

regions, landscapes continue to be fragmented due to urbanisation. Consequently,

urban bushlands become ‘stepping stones’ within the landscape allowing certain species

to move between urban, suburban, rural and forested areas (Manning et al. 2006). The

processes driving fragmentation are greater in smaller patches, as they are more likely to

be altered by changed environmental parameters (such as urbanisation) to a greater

degree than large patches (e.g. edge effects). Such shifts may have implications for

habitat structure, ecosystem function and food web interactions in small remnant forest

patches (Morgan and Farmilo 2012). Thus, the dynamics of forest patches are

principally driven by features from the surrounding landscape (Rowntree 1988; Ewers

and Didham 2006a; Morgan and Farmilo 2012). Therefore, the management of, and

research on, fragmented ecosystems should be directed at understanding and controlling

these external influences as much as the biota of forest patches themselves. This is

particularly significant for habitats containing species that utilise specialised resources

such as hollow-bearing trees (Lindenmayer 2002; Rowston et al. 2002; Manning et al.

2006).

The urban landscape however, is never uniform, with impacts determined by the type

and intensity of the land uses found within it (Brady et al. 2009). These impacts vary

with the time since isolation (Saunders et al. 1991), distance from other remnants (Luck

and Daily 2003) and degree of connectivity with other remnants (Lindenmayer and

Page 92: A thesis submitted for the fulfilment for the requirements

70

Fischer 2006). Consequently, the conservation of hollow-bearing trees within the urban

landscape is essential for maintaining ecosystem function by providing areas of refugia

and ecosystem services through the conservation of biodiversity in general (McKinney

2006).

Urbanisation intensity correlates with increased disturbance and the structural

simplification of remaining vegetation (McKinney 2008). However, it has been found

that a complex matrix within a human-modified landscape can provide supplementary

habitat resources (Brady et al. 2011; Threlfall et al. 2011). Thus, the urban landscape

can be of benefit to biodiversity when each site is assessed individually (McKinney

2008). Individual species traits also determine how well a species adapts to

urbanisation (Luck and Smallbone 2010). So it is important to consider differing

species responses to landscape change, to move beyond focusing primarily on spatial

attributes (size, isolation), and recognise that landscape change is the most important

variable to consider when examining functionality (Holland and Bennett 2009).

The functional ecology of the urbanisation gradient, and specifically that pertaining to

the availability of hollow-bearing trees, is what is being investigated in this chapter. In

doing so, the ecology in and the ecology of urban areas was explored. The ecology in

urban areas is closely affiliated with traditional ecology, in that it investigates ecological

patterns and processes in urban areas. The ecology of urban areas is concerned with

how those systems function as a whole. Thus, this system-oriented approach will

deliver a deeper insight into the role that urbanisation gradients play in conservation

management (Gaston 2010a). This chapter quantifies the impacts of urbanisation,

landscape and environmental variables on hollow-bearing trees within urban forest

patches. The significance of these findings in regards to rapidly changing landscapes,

such as those found along the urbanisation gradient, will be of benefit to managers and

planners when making decisions about where, and how to best manage for hollow-

bearing trees in an urban matrix.

Methods

Site establishment and the categorisation of sites along the urbanisation gradient were

completed as outlined in Chapter 3. The urbanisation gradient was used to understand

Page 93: A thesis submitted for the fulfilment for the requirements

71

the impacts of urbanisation on hollow-bearing trees. Landscape and environmental

variables used for this study along this gradient were chosen to assess the importance of

these potential driving forces on forest structure within forest patches (Tables 4.1 - 4.3).

Table 4.1: The urbanisation index, hollow-bearing tree, landscape and environmental

variables quantified at each site.

Variable Description of variable

Urbanisation Index Calculated as a percentage of site buffer transformed by housing and infrastructure and normalised by the density of housing (Chapter 3 Equation 3.3).

Hollow-bearing tree variables Density of hollow-bearing trees per hectare per site. Stand density Stand density was calculated by dividing the sum of

all trees per plot by the surveyed area of each plot (0.28 ha).

Landscape variables Fire history Number of fire events from historical aerial imagery

over time (1955-2009). Logging history Number of logging events from historical aerial

imagery over time (1955-2009). Connectivity (%) Percentage of site perimeter connected to other

forested areas. Environmental variables Angle of slope θ Calculated from GIS layers (Chapter 3 Equation 3.2) Aspect (°) Directional orientation. Elevation (m) Obtained from GIS mapping layers.

Page 94: A thesis submitted for the fulfilment for the requirements

72

Table 4.2: Landscape variables examined in association with hollow-bearing tree

density from 45 urban forest patches sites. Values in brackets represent where a site is

located along the urbanisation gradient. Standard errors are supplied for sites with

multiple plots. Urbanisation index: urban = high value, rural = low value.

Site name No: of plots

HBT density (ha)

Urbanisation Index

Area (ha)

Connectivity (%)

Andrews Reserve (r) 1 28.57 5.22 10.8 44.35

Aqua Promenade (p) 2 60.71 ± 10.15 32.29 13.1 38.80

Austenville Road (r) 3 34.52 ± 8.5 0.02 24.9 71.95

Behms Road (r) 1 39.29 6.94 14.2 0.00

Bonogin Conservation Area (c/u) 6 33.92 ± 7.8 157.16 725.1 73.01

Brushwood Ridge (u) 2 39.28 ± 3.51 522.96 16.8 0.00

Burleigh Headland NP (u) 1 92.85 441.57 27.3 0.00

Burleigh Knoll NP (u) 1 85.71 1412.99 4.1 0.00

Burleigh Ridge (u) 2 53.56 ± 35.71 3195.78 17.89 0.00

Carrington Road (r) 1 39.29 4.18 5.9 77.38

Daisy Hill Forest Reserve (u) 5 23.56 ± 6.6 133.43 432.1 49.97

Elanora Conservation Area (p) 2 71.48 ± 7.14 150.88 13.1 50.62

Elanora Wetlands (u) 1 28.57 90.08 26.6 0.00

Galt Road (p) 1 53.57 115.88 33.6 73.38

Gilston Road (r) 1 35.71 10.23 4.04 85.62

Hellman Street (p) 1 28.57 212.51 9.1 0.00

Herbert Park (u) 3 28.56 ± 7.43 393.57 16.4 0.00

Johns Road (r) 1 39.28 32.50 2.1 45.10

Karawatha Forest Reserve (c/p) 7 25.5 ± 4.63 1299.21 824.3 7.61

Kincaid Park (u) 1 17.86 83.34 7.5 0.00

Kerkin Road (r) 1 42.85 845.68 176.9 17.55

Lakala Street Reserve (u) 1 21.43 760.06 3.2 0.00

Miami Conservation Area (u) 1 10.71 1178.50 4.4 0.00

Mt Nathan Conservation Area (r) 3 33.33 ± 7.8 21.84 266.1 16.52

Nerang Forest Reserve (c/p) 5 61.43 ± 7.93 382.66 1669 15.08

Numinbah Road (r) 2 28.57 ±10.71 3.98 157.6 53.25

Olsen Avenue (u) 2 7.14 807.23 38.9 0.00

Pacific Highway, Currumbin (u) 2 53.57 ± 0 1741.30 16.8 0.00

Pacific Pines Parkland (u) 1 21.42 415.38 5.89 0.00

Piggabeen Road (p) 1 14.28 32.21 2.9 41.60

Pimpama Conservation Area (r) 2 3.57 16.99 565.8 1.28

Reedy Creek Conservation Area (r) 3 26.19 ± 11.72 19.69 44.8 63.74

Reserve Road Parklands (p) 1 17.85 276.44 7.6 0.00

Stanmore Park (r) 1 7.14 15.72 50.8 2.66

Summerhill Court Reserve (u) 1 50 793.33 9.5 0.00

Page 95: A thesis submitted for the fulfilment for the requirements

73

Table 4.2 Continued: Landscape variables used to quantify the impacts of

urbanisation on hollow-bearing tree density from 45 urban forest patches. Values in

brackets represent where a site is located along the urbanisation gradient. Standard

errors are supplied for sites with multiple plots. Urbanisation index: urban = high value,

rural = low value.

Site name No: of plots

HBT density (ha)

Urbanisation Index

Area (ha)

Connectivity (%)

Syndicate Road (p) 3 39.28 ± 10.71 22.25 185.3 84.93

Tallowwood Road (r) 1 50 2.46 14.2 86.08

The Plateau Reserve (p) 1 92.85 21.06 26.5 24.83

Tomewin Road (r) 1 57.14 17.17 5.1 47.14

Trees Road Conservation Area (r) 2 44.64 ± 12.49 13.31 53.3 81.82

Tugun Conservation Area (u) 1 32.14 684.35 14.3 0.00

Upper Mudgeeraba Conservation Area (r) 2 8.93 ± 1.78 3.02 193.2 92.33

Venman Bushland National Park (c/p) 6 25 ± 5.9 49.52 434.2 86.96

Wongawallen Conservation Area (r) 2 30.35 ±26.78 19.12 383.7 83.28

Woody Hill, Hart's Reserve (p) 2 44.6 ± 12.5 358.72 91.1 0.00

Page 96: A thesis submitted for the fulfilment for the requirements

74

Table 4.3: Environmental variables examined for associations with hollow-bearing trees in each of the 45 forest patches. Standard errors are supplied for sites with multiple plots.

Site Slope (radians) Aspect Elev.

(m) Stand density (# trees ± S.E.)

Fire index

Logging index

Andrews Reserve 0.24 -0.17 50 257.14 4.55 0.38 Aqua Promenade Reserve 0.38 -0.04 62 323 ± 41 2.27 0.38 Austenville Road 0.44 0.60 266 370.23 ± 44 4.55 0.21 Behms Road 0.05 0.98 10 242.86 0 0.38 Bonogin Con Area 0.24 0.17 50 175 ± 25 13.64 0.31 Brushwood Ridge Parklands 0.00 1.00 0 242.85 ± 68 2.27 0.38 Burleigh Heads National Park 0.34 -0.51 210 303.57 0 0 Burleigh Knoll Con Park 0.31 -0.21 45 232.14 2.27 0 Burleigh Ridge Park 0.43 0.70 77 273.21 ± 45 2.27 0.31 Carrington Road 0.03 0.64 225 167.86 6.82 0.21 Daisy Hill Con Area 0.08 -0.24 95 158.57 ± 5.5 4.55 0.21 Elanora Con Reserve 0.24 0.00 62 373.21 ± 112.5 0 0.38 Elanora Wetlands 0.00 1.00 0 303.57 0 0.31 Galt Road Park 0.19 -0.64 200 142.86 4.55 0.44 Gilston Road 0.12 -0.17 70 242.86 2.27 0 Hellman Street 0.12 0.98 50 182.14 2.27 0.38 Herbert Park 0.31 0.10 36 233.33 ± 28 2.27 0.21 Johns Road 0.40 0.98 40 300 2.27 0 Karawatha Forest Reserve 0.08 0.00 53 200 ± 40 4.55 0.6 Kincaid Park 0.17 0.94 80 185.7 0 0.38 Kerkin Road 0.05 -0.77 5 225 4.55 0.38 Lakala Street Reserve 0.16 1.00 15 275 0 0.21 Miami Bushland Con Park 0.05 -0.98 20 450 0 0 Mount Nathan 0.41 0.13 193 153.5 ± 9 0 0.38 Nerang State Forest 0.19 0.01 80 271.4 ± 45 2.27 0.31 Numinbah Road 0.32 0.36 180 157.1 ± 11 13.64 0.21 Olsen Avenue 0.07 0.50 20 239.2 ± 11 15.9 0.5 Pacific Highway 0.22 -0.34 47 360.7 ± 3.6 4.55 0.31 Pacific Pines Parkland 0.10 1.00 10 239.86 2.27 0 Piggabeen Road 0.00 1.00 2.5 257.14 0 0 Pimpama Con Park 0.31 0.71 65 271.43 ± 78 6.82 0.44 Reedy Creek 0.33 0.17 60 208.33 ± 56 2.27 0.38 Reserve Road Parklands 0.30 -0.50 69 260.71 0 0.5 Stanmore Park 0.17 0.64 75 396.43 0 0.5 Summerhill Court Reserve 0.05 -0.99 30 178.57 0 0 Syndicate Road 0.21 -0.17 55 192.85 ± 29 0 0.21 Tallowwood Road 0.18 1.71 223 321.43 2.27 0.31 The Plateau Reserve 0.24 -0.17 100 275 2.27 0.38 Tomewin Road 0.44 -0.34 135 178.57 0 0 Trees Road Con Area 0.27 0.01 112 360.7 ± 50 0 0.31 Tugun Conservation Park 0.14 0.50 55 239.29 2.27 0.31 Upper Mudgerabah Cons Area 0.38 0.20 165 253.57 ± 46 2.27 0.21 Venman National Park 0.02 0.48 83 134.52 ± 17 4.55 0.44 Wongawallen 0.45 -0.47 285 125 ± 7 4.55 0.38 Woody Hill 0.01 -0.41 8 205.35 ± 30 0 0

Page 97: A thesis submitted for the fulfilment for the requirements

75

Data analysis

Predictor variables known to have strong biological influences were presented for

analysis based on a priori investigations from within the literature (Table 4.1). A

Spearman’s Rank correlation matrix examined colinearity among variables. Variables

that had a correlation coefficient value |r| > 0.7 were considered to be highly correlated

(Garden et al. 2010). Mean distance to nearest fragment edge was highly correlated to

area (|r| = 0.768) and was removed from further analysis. The decision to retain area

over mean distance to nearest fragment edge from plot for inclusion was based on the

assumption that area would be more ecologically meaningful for analysis. All other

variables were included in subsequent analysis.

Transformation of data was undertaken for two reasons (i) to meet the requirements for

homogeneity of variance, i.e. to limit the dominance of outliers and (ii) due to differing

scales of data. Arcsine square root transformations were undertaken for explanatory

variables with percentage values; while a logarithmic (x+1) transformation was

undertaken for all other explanatory variables requiring transformation due to scale

differences. Aspect data was transformed using Tan θ and slope was converted to

radians prior to analysis. No transformations were made to the response variable

hollow-bearing tree density.

The general rule of n/3 (where n = number of sites sampled) was used to determine the

maximum number of predictor variables to use for all models (Crawley 2007). The

relationship between the response variable (hollow-bearing tree density) and the

explanatory variables was examined using pairwise plots to ascertain the strength of the

relationship and whether it was linear. Eight predictor variables were modelled to

examine their relationships with hollow-bearing tree parameters and those were; stand

density, fire history, logging history, connectivity, angle of slope, aspect, elevation and

the urbanisation index. Poisson Generalised Linear Mixed Effects Models (GLMM)

were used for the analysis; as plots were nested within sites. Interactions between

predictor variables were therefore also analysed. The Poisson distribution was chosen

as it is generally used for count data. The main advantages are (i) the probability for

negative values is 0 and (ii) the mean variance relationship allows for heterogeneity

(Zuur et al. 2009).

Page 98: A thesis submitted for the fulfilment for the requirements

76

Twenty-eight candidate models including the global model were run to assess landscape

and environmental parameters (Appendix 5). An auto-correlation structure was

included into the models to account for site variation. Models with random effect of

site were then compared using Akaike’s Information Criterion (AIC) (Akaike 1973)

with the ‘best’ model having the lowest AIC value (Burnham and Anderson 2002). The

interaction of logging and elevation were included as a model as they were the two

strongest performing variables when modelled individually. An interaction represents

non-additivity in the combined effects of the two interacting factors. The net outcome

of the two processes is significantly more (a synergistic effect) or less (an antagonistic

effect) than the sum of the two processes operating independently (Didham et al.

2007b).

Models were ranked according to their Δi values. The higher the Δi value, the less

accurate the model for the given data. The best approximating models that had a Δi ≤ 4

are presented. Because of uncertainty in some of the final model selections, a model

averaging approach was applied and variables were ranked according to their relative

overall importance by summing the Akaike weight (wi) from all top model

combinations where each of the variables occurred. Statistical analyses were conducted

using R 3.1 statistical software (R Core Team 2012) and the glmmML (Brøstrom 2012)

and MuMIn package (Barton 2012).

Results

Analysis of landscape and environmental variables revealed that the model containing

the logging index, elevation and the interaction of the logging index*elevation had the

best fit (Table 4.4). Logging and elevation were both negatively correlated with the

density of hollow-bearing trees (Table 4.4). Parameter estimates for the approximate

importance of landscape and environmental variables associated with the density of

hollow-bearing trees per hectare revealed that logging was the most important variable.

Overall the only coefficient estimates that represented a good fit was the interactive

model of the logging index*elevation; all other coefficient estimates did not perform as

well (Table 4.5). The urbanisation index was not identified in any of the best

performing models, nor did it rank highly in the assessment of the relative importance

of parameters influencing hollow-bearing tree density.

Page 99: A thesis submitted for the fulfilment for the requirements

77

Table 4.4: Best approximating model comparisons of explanatory landscape and

environmental variables associated with the density of hollow-bearing trees. Akaike

models with ∆i values < 4 are presented. The next model (AICc > 4) demonstrates the

distance between the two best models and the third.

Model AICc ∆i wi

logging index + elevation + (logging index * elevation) 247.54 0.00 0.49

logging index + elevation 250.70 3.15 0.10

logging index + elevation + stand density 251.87 4.33 0.05

Table 4.5: Parameter estimates for the landscape and environmental variables

associated with the density of hollow-bearing trees per hectare across all sites.

Variables for each explanatory analysis are ordered by their relative importance.

Parameter Estimate

Standard

error

Confidence interval

2.5% 97.5%

Relative variable

importance

logging index -2.900 1.857 -6.541 0.740 0.91

elevation -0.063 0.143 -0.344 0.217 0.77

logging history*elevation 0.857 0.355 0.160 1.554 0.50

stand density 0.000 0.000 -0.000 0.001 0.12

fire index -0.008 0.014 -0.035 0.019 0.11

slope 0.250 0.502 -0.734 1.236 0.09

aspect 0.095 0.073 -0.047 0.239 0.05

connectivity 0.002 0.003 -0.004 0.009 0.01

urbanisation index 0.057 0.061 -0.063 0.177 0.01

Page 100: A thesis submitted for the fulfilment for the requirements

78

Discussion

Plant community composition is a product of environmental and anthropogenic

disturbance regimes (Williams et al. 2005a). Therein, the logging index was found to

be a driving factor influencing hollow-bearing density along the urbanisation gradient

on the Gold Coast. When the interaction model, logging index*elevation was included

it was found to have the greatest effect, in that hollow-bearing tree density decreased

where there was increased logging at lower elevations. Given the intensive logging

history of south-east Queensland (Chapter 1) it is not surprising that the logging index is

the variable of most importance when assessing hollow-bearing trees across the

landscape. Previous studies consistently show that the number of hollow-bearing trees

is negatively correlated with logging (Lindenmayer et al. 1990; Gibbons and

Lindenmayer 1996; Ball et al. 1999; Ross 1999; Lloyd et al. 2006; Law et al. 2013).

Importantly, the logging index as well as a number of other environmental parameters

was more influential than the urbanisation index. This suggests that on the Gold Coast

the effects of urbanisation are not yet discernible within the hollow-bearing tree

communities. Previous studies reporting on the impacts of urbanisation gradients tend

to focus on specific landscape parameters such as; individual patch dynamics (Blair

1999), land use type (Brady et al. 2009), fragmentation (Crooks et al. 2004) and

vegetation communities (Hahs and McDonnell 2007) to name a few. Although

historical land use is often mentioned as a variable of importance for biodiversity at a

landscape scale (McDonnell et al. 1997; Moffatt et al. 2004; Hahs and McDonnell

2007; Brady et al. 2009), the current study is the first of its kind to quantify the impacts

of logging history on hollow-bearing trees, as well as its relative importance compared

to other variables along an urbanisation gradient.

These findings are consistent with the view that land use history will significantly

impact on the function and dynamics of forest patches (Gibbons and Lindenmayer

2002); with the density and formation of hollow-bearing trees being dependent on site

logging history. Logging history in general has resulted in a complex matrix of remnant

forest patches (Gibbons and Lindenmayer 2002). At a site level, many of the smaller,

highly urbanised sites on the Gold Coast contain high densities of hollow-bearing trees

which have remained relatively undisturbed over time (e.g. Burleigh Knoll National

Park). They have, however, contracted in size, thus leaving many mature hollow-

bearing trees in small remnant forest patches along the gradient. This suggests that for

Page 101: A thesis submitted for the fulfilment for the requirements

79

hollow-bearing trees there is a lag response to disturbance given that these resources are

relatively long lived. As such the impacts of more recent urbanisation processes may

not yet be evident. Conversely, all control sites in this study are ex-forestry lots and

have a relative recent logging history, with some sites being logged as recently as the

1990’s (unpublished data. The Department of Environment and Heritage). This is

reflected by low numbers of hollow-bearing trees and an evenness in the size class of

trees found within these forests, with a mean dbh of 48.5 cm (± 1.84 S.E.) (Chapter 3).

After accounting for logging history and elevation; stand density was the next most

important variable in the final models. The stand of trees is the scale at which specific

forestry management actions are usually implemented (O'Hara 1998). Therefore,

assessing how stands respond to either external or internal factors may be central to the

development of adaptive management strategies to improve the functional integrity of

urban forest remnants. This could be done by enhancing forest condition through the

implementation of protection and restoration initiatives (Garden et al. 2010).

Additionally, the arrangement of regrowth forest in the landscape may also aid in

remnant forest patch recovery, as connectivity to larger contiguous forests improves

stand structure (Bowen et al. 2009; Garden et al. 2010) but also movement of fauna

between fragments (Laita et al. 2011). Regrowth and recruitment are therefore both

vitally important in ensuring the persistence of particular habitat features such as

hollow-bearing trees at the stand level and was the focus of Chapter 5.

The general importance of landscape and environmental variables as drivers of

biodiversity has been acknowledged in the literature (Saunders et al. 1991;

Lindenmayer et al. 2000b; Ewers and Didham 2006a; Lindenmayer and Fischer 2006;

Gaston 2010a). The theoretical management of fragmented landscapes therefore has

two components; i) management of the internal dynamics of remnant forest patches, and

ii) management of the external influences. It has been suggested that the management

of larger reserves focus on the internal dynamics and for smaller forest patches

management should focus on the external influences (Saunders et al. 1991). It has also

been hypothesised, that external influences are important no matter the size of the forest

patch (Janzen 1986). Which supports the findings from this study, in that logging

Page 102: A thesis submitted for the fulfilment for the requirements

80

history was what has been found to determine the vegetation structure and therefore,

hollow-bearing trees along the urbanisation gradient on the Gold Coast.

Historical land use practices and their effects have long-lasting impacts on the

environment, driving habitat structure and function over time. Whether hollow-using

vertebrate fauna utilise these forest patches depends on the availability of specific

resources but also on the complexity of the surrounding matrix (Chapter 6). Thus, the

long-term viability of native flora and fauna relies upon the ability of local government

planners and policy makers to meet the many and varied challenges that are presented

along the urbanisation gradient (Chapter 2).

Page 103: A thesis submitted for the fulfilment for the requirements

81

Chapter 5: The retention of recruitment trees and time to hollow

formation

Introduction

Tree hollow development is a complex process and can take up to 150 years for large

hollows to form. In Australia the time taken for hollow formation to occur differs

substantially among eucalypt species (Gibbons and Lindenmayer 2002). Large, old

trees with a greater diameter at breast height (dbh) are known to contain significantly

more hollows than smaller, younger trees (Lindenmayer et al. 1993) with a positive

relationship between the diameter of a tree and the presence of hollows reported in

many studies (Saunders et al. 1982; Nelson and Morris 1994; Ross 1998; Ross 1999;

Wormington and Lamb 1999; Koch 2008; Munks 2009; Goldingay 2011).

The loss of hollow-bearing trees is not easily reversed, as hollow replacement is a slow,

incremental process (Gibbons and Lindenmayer 2002). The distribution and abundance

of hollow-bearing trees is therefore a function of the retention of trees, but also of the

processes facilitating hollow formation in non-hollow-bearing trees. From the few

studies examining the long-term trends in numbers of hollow-bearing trees within

managed forests; all have predicted a gradual decline (Gibbons and Lindenmayer 1996;

Wormington et al. 2005; Koch et al. 2008b; Koch et al. 2009). Failure to halt the loss

of hollow-bearing trees can only lead to a prolonged absence of this resource for

wildlife, which could result in the possible extinction of species, e.g. Leadbeater’s

possum (Smith and Lindenmayer 1992; Harley 2006). Therefore, to assess the impact

of such losses for hollow-dependant fauna, it is necessary to gain an understanding of

the trajectories of hollow formation as well as patterns of hollow-bearing tree

availability. This may be particularly important in regions undergoing rapid habitat

transformation, such as within urban areas.

Page 104: A thesis submitted for the fulfilment for the requirements

82

Hollow-bearing eucalypt trees are extremely long-lived, typically living for 300-500

years (Brookhouse 2006) and potentially providing hollows for a further 100 years once

the tree had died (Gibbons and Lindenmayer 2002). The age at which tree-hollows are

produced usually occurs between 100-230 years (Gibbons et al. 2000b; Whitford 2002;

Goldingay 2011). Furthermore, hollows are unlikely to be suitable for fauna if trees are

< 120 years old, as large hollows are rare in eucalypts < 220 years old (Gibbons and

Lindenmayer 2002). There are also other factors related to age that will influence the

presence of hollows in a tree:

• stand competition, equalling suppressed growth (Wormington and Lamb 1999);

• the shedding of large branches is associated with the age at which a tree’s

accumulative growth rates slow and begin to decline (Jacobs 1955; Mackowski

1984), the shedding of large branches will increase the potential number of sites

on a tree where hollows can form (Gibbons and Lindenmayer 2002);

• the slowing of tree growth rates (ageing) reduces the ability of a tree to occlude

wounds, increasing the potential for hollow formation (Jacobs 1955);

• sapwood thickness may decline with age (Florence 1996) thereby increasing a

trees suseptability to attack from termites and fungi (Werner and Prior 2007) and

fire damage (Adkins 2006) leading to the potential for an increase in hollow

formation;

• the ratio of heartwood to sapwood also declines with age, leading to a greater

incidence of heartwood decay (Gibbons and Lindenmayer 2002);

• older trees have a greater frequency of exposure to events such as fire and

windstorms (Gibbons and Lindenmayer 2002).

While these relationships have all been studied within Australian production forests,

how they apply to urban forest patches has received little attention. Furthermore,

despite this previous research there is still ongoing debate about the management

requirements to achieve a perpetual supply of hollow-bearing trees within the

production forests of Australia (Goldingay 2011). In contrast, forest patches within the

urban landscape have few, or no, management guidelines for the retention or

recruitment of hollow-bearing trees (Chapter 2). However, forest patches within the

Page 105: A thesis submitted for the fulfilment for the requirements

83

urban matrix continue to be subjected to variables associated with anthropogenic

activities such as land clearing (Davis et al. 2013). Until recently a dismissive

management attitude towards small forest patches was taken, as they were not

considered to contain sufficient resources to meet the needs of many individual species

(Daily 2001). Public safety concerns have also seen the removal of many hollow-

bearing trees within urban areas (Rhodes et al. 2005), resulting in a negative social

attitudes towards trees from certain sectors of the community (Kirkpatrick et al. 2013).

Despite the acknowledged significance of the loss of hollow-bearing trees in urban

areas, relatively few studies have investigated the ecological impacts that this will have

on faunal communities (Goldingay and Sharpe 2004a; Harper et al. 2005a; Davis et al.

2013) (Chapter 6).

Research aims

This chapter aimed to quantify the impact of urbanisation on the persistence and

availability of hollow-bearing trees by modelling the influences of selected tree,

disturbance and environmental variables on hollow formation from the recruitment

eucalypt population in urban forest patches. In particular, the relationship between dbh

and hollow type was investigated, as faunal preferences for particular hollow types have

been well documented in Australia (Gibbons et al. 2002; Lindenmayer 2002; Goldingay

2009; Koch et al. 2009) but not within urban forest patches. These analyses were

completed in conjunction with an investigation into the potential for hollow-bearing tree

recruitment and their ability to persist across the landscape.

Methods

Plot sampling protocol

To quantify the effects of urbanisation on hollow-bearing trees and the availability of

hollows; 45 forest patches along an urbanisation gradient were surveyed. At each site a

varying number of plots were sampled to quantify the status of hollow-bearing trees.

For a full description of methodologies and site establishment see Chapter 3.

Page 106: A thesis submitted for the fulfilment for the requirements

84

This study evaluated five genera belonging to the family Myrtaceae; Angophora,

Corymbia, Eucalyptus, Lophostemon and Syncarpia; hereafter referred to as eucalypts.

In each plot all eucalypt trees within a 30 m radius from the plot midpoint and with a

dbh ≥ 10 cm were identified and measured to determine vegetation community

structure. For each tree the presence of hollows was noted to classify the tree as

hollow-bearing or not. In addition, the number and type of hollows in all hollow-

bearing trees was recorded (Chapter 3). While dead trees continue to provide a habitat

resource for approximately 50 - 100 years (Moloney et al. 2002), they were removed

from the data set, as the analyses in this chapter focused on the living cohort of both

recruitment trees and existing hollow-bearing trees.

Growth rates for eucalypt species

Recruitment trees were classified as being eucalypt trees with a dbh ≥ 10 cm without a

hollow. To determine the time frame for recruitment modelling of hollow-bearing-trees

it is necessary to estimate tree age at the onset of hollow formation. Tree age can be

predicted according to the disturbance history of a site (Bradshaw and Rayner 1997), by

radiocarbon dating (Turner 1984), by tree ring counting (Banks 1997) or by tree

diameter increment data used in growth models (Wormington and Lamb 1999; Gibbons

et al. 2000b; Moloney et al. 2002). Growth models are also widely used as a tree-

ageing technique as they are non-destructive and easily applied (Koch 2008). Tree age

estimates from wet and dry sclerophyll forests of south-east Queensland reveal the

intraspecific variation that occurs between species grown within the same sites (Table

5.1).

Table 5.1: Annual dbh growth rates (cm/yr) in E. pilularis, E. microcorys and E.

racemosa (Wormington and Lamb 1999) and Corymbia citriodora (Ross 1999) in dry

sclerophyll forests of south-east Queensland

Species dbh (cm) 0 - 24.9 25 - 49.9 50 - 74.9 75 - 99.9 100 - 125 125+

E.pilularis 1.2 0.8 0.6 0.5 0.5 0.4 E.microcorys 0.9 0.6 0.45 0.35 0.35 0.3 E.racemosa 0.6 0.5 0.3 0.25 0.24 0.2 C.citriodora 0.4 0.4 0.4 - - -

Page 107: A thesis submitted for the fulfilment for the requirements

85

In growth and yield modelling, dbh is the primary measure of tree age (Avery and

Burkhart 1989). Individual tree diameter growth models are used to predict growth

rates of trees within a stand (Subedi and Mahadev 2011). Estimates of eucalypt growth

rates are available from south-eastern Australia (Bauhus et al. 2002), Tasmania (Koch et

al. 2008a) and south-east Queensland (Ross 1999; Wormington and Lamb 1999). For

the purpose of this study only data from the managed forests of south-east Queensland

(Ross 1999; Wormington and Lamb 1999) were used.

Growth rates in the eucalypt communities in urban fragments

The species growth rate data (Table 5.1) were then applied to trees found across the

study area. In total 850 trees (16.5%) were represented by the above four species.

Remaining species were then classified by bark type i.e. half-bark, tessellated bark,

smooth-bark, stringy-bark or iron-bark; to best fit with the data available on the known

growth rates of species with similar bark types (Appendix 6). In the case of the iron-

barks = Eucalyptus crebra and E. fibrosa; tessellated-barks = Corymbia intermedia and

C. trachyphloia; stringy-barks = E. microcorys and E. acmenoides; the mean growth

rates of each species per bark type was calculated and then used as the predicted growth

rates for other iron, tessellated and stringy-bark species. Smooth-bark growth rates

were calculated using C.citriodora as a proxy species. Incremental annual dbh growth

rates were then calculated (Table 5.2). To allow for variation in growth rate estimates

due to site location, a sensitivity analysis ranging from -5% to 20% was applied

(Equation 5.1) where G is growth rate, A is the annual incremental dbh growth rates and

s is the sensitivity. It is acknowledged that this is unlikely to give precise growth rate

estimates; however, given the lack of data available on the tree species in this study area

it offers a justifiable solution.

𝐺 = (𝐴 × 𝑠) (5.1)

Page 108: A thesis submitted for the fulfilment for the requirements

86

Table 5.2: Annual incremental dbh growth rates calculated by bark type. Growth rates

for two common species are also provided.

dbh size

class (cm) E. microcorys E. racemosa

Half-

bark

Iron-

bark

Smooth-

bark

Stringy-

bark

Tessellated-

bark

0 0.9 0.6 1.2 0.6 0.6 0.6 0.6

25 0.6 0.5 0.8 0.5 0.5 0.5 0.5

50 0.45 0.3 0.6 0.3 0.3 0.3 0.3

75 0.35 0.25 0.5 0.25 0.25 0.25 0.25

100 0.35 0.24 0.5 0.24 0.24 0.24 0.24

125+ 0.3 0.2 0.4 0.2 0.2 0.2 0.2

Mortality rates for eucalypt trees

Tree mortality has a fundamental role to play in forest dynamics (Franklin 1987).

However, tree mortality remains a very misunderstood process in ecology (Hartman

2010; Sala 2010). The ability to reliably estimate tree mortality rates is difficult due to

trees long life span, their large effective population size, geographical ranges (Petit

2006) and to the limitations of data availability (Csilléry 2013). Permanent forest plots

and dendrochronological data have been used for probability estimates of mortality for

individual trees i.e. the number of dead trees over the total number of individual trees

(Peng 2011; Taylor 2007; Wolf 2004) known as the ‘volume mortality rate’. Results

from these data reflect mortality from the natural patterns of tree life history, such as

competition or senescence (Csilléry 2013) and are well justified (Kurz 2008). Therein,

volume mortality rates are used in this current study.

Mortality data from within Native Forest Permanent Plots (NFPP) held by the

Queensland Herbarium Department of Science, Information Technology, Innovation

and the Arts (unpublished data1) have been used. These data consist of permanently

monitored plots in uneven-aged mixed species native forests in Queensland’s State

Forests and National Parks. These plots were established between 1936 and 1998 and

remeasured every two to 10 years up to 2011. These data were collected from 99 State

Forests and National Parks over a 62 year time frame. Data were collected on the

species name, date of measure, measure interval in years, a count of all trees measured

per site, a count of trees by species and each trees status; being either living, dead by

1 Mortality data from The Queensland Herbarium from unpublished work 1936-2011.

Page 109: A thesis submitted for the fulfilment for the requirements

87

natural causes or dead by human causes i.e. logging, treatment or fire. Only trees that

had died by natural causes were used for this analysis. As all species of flora are

included in this data; initial data exploration required eucalypt species to be extracted;

this was further refined to reflect the range of species found within this current study

area.

Thus, over 82% of species found within the Gold Coast are represented. The remaining

18% of species were classified by bark type as per the methods outlined in ‘Tree growth

rates’. Annual mortality was then calculated at each site for each species (Equation

5.2). Where M is annual mortality, nd is the number of trees dying from natural causes;

nl the number of living trees and t represents the number of years between consecutive

measurements. The mean mortality rate for each species was then calculated and was

found to range between 0.01 - 0.05 (±SE 0.002). This estimate does not take into

account stochastic events such as climate or disease.

𝑀 = 𝑛d

(𝑛l)𝑡 (5.2)

Recruitment trees and hollow development

The differential growth rates (Table 5.2) were then applied to the current cohort of

recruitment trees (n = 5132) over 150 years using VLOOKUP tables in Excel based on

their known dbh at Year 1. The proportion of the recruitment tree population with

hollows was then quantified at 10, 50, 100 and 150 years to monitor trends in hollow-

bearing trees over time. Classification of recruitment trees as hollow-bearing or not (i.e.

presence/absence of a hollow), in each year was determined based on the probability of

a tree having a hollow as a function of dbh. These hollow probabilities were derived

from the current hollow-bearing tree cohort using log-linear logistic models (Table 5.3).

Loss of trees due to clearing for urban development was also factored into calculations

based on the perceived risk at sites along the gradient. The analysis is restricted to the

current cohort of recruitment trees (i.e. all non-hollow-bearing trees) as no data are

available on recruitment rates or natural mortality to quantify trends in the community

as a whole.

Page 110: A thesis submitted for the fulfilment for the requirements

88

Table 5.3: The probability of a recruitment tree having a hollow as a function of dbh.

dbh (cm)

Hollow probability

10 - 20 0.05 20 - 30 0.15 30 - 40 0.25 40 - 50 0.42 50 - 60 0.5 60 - 70 0.52 70 - 80 0.65 80 - 90 0.8 90 - 100 0.8 100 - 110 0.9 110 - 120 1 120 - 130 1 130 - 140 1 140 - 150 1 150 - 160 1 160 - 170 1

Loss of hollow-bearing trees due to habitat loss from land clearing

Modelling was undertaken to predict the loss of hollow-bearing trees due to habitat loss

from land clearing for housing and infrastructure over a 10, 50, 100 and 150 year

period. To facilitate this, a risk matrix was constructed to assign risk based on the

location of sites along the urban gradient as well as their size as represented by the four

primary classes (ha). The total area of remnant vegetation on the Gold Coast (including

the three control sites located in Brisbane and Redland Shires) is 63,678 ha. The extent

of forest patches was calculated by interrogating ArcGIS regional ecosystem and land

use layers supplied by Gold Coast City Council, Brisbane City Council and Redland

Shires. Risk values were assigned a priori on a scale from - 0.5 (less risk) to + 0.5

(high risk) along the urbanisation gradient (Table 5.4). Land clearing rates of 442

ha/year were applied uniformly over the study area as this figure has been reported as

the Gold Coast City Council clearing rate until 2019 (Gold Coast City Council 2009).

The modelling presented here extends well beyond 2019 and predictions should

therefore be seen as a minimum threshold as clearing rates could conceivably increase.

Therefore, the net rate of remnant habitat loss for the region amounts to 69% per year.

This figure was used in all subsequent calculations after adjusting the loss from each

Page 111: A thesis submitted for the fulfilment for the requirements

89

site by the relevant risk factor. For example, a small, highly urbanised forest patch may

be at a very low risk of clearing given its proximity to high density housing, i.e. narrow

streets. By contrast, the larger control sites, some of which have been recently

harvested, are at greater risk of further clearing for urbanisation and thus have been

weighted accordingly (Table 5.4).

Table 5.4: Risk assessment of sites due to land clearing for urban development and

infrastructure. Risk increases down and across (left–right) the table.

Size category for research sites

Gradient Small <5 ha Medium 6-50 ha Large 51-100 ha Very Large > 100 ha

Urban very low (-0.5) low (-0.25) medium low (-0.1) medium (base value)

Peri-urban low (-0.25) medium low (-0.1) medium (base value) medium high (+0.1)

Rural medium low (-0.1) medium (base value) medium high (+0.1) high (+0.25)

Control medium (base value) medium high (+0.1) high (+0.25) very high (+0.5)

Diameter at breast height and bark type as a predictor of hollow type

From the current inventory of hollow-bearing trees the probability of a tree containing a

hollow of a certain type was calculated based on known dbh and hollow type. Hollow

type (Table 5.5) was chosen over hollow size, as apart from bayonet and fissures,

hollows in all other categories can be either very large or very small.

Table 5.5: The classification of the seven available hollow types as used in this study.

Hollow type classification Hollow type

1 butt 2 branch end 3 bayonet 4 fissure 5 trunk main 6 trunk top 7 branch main

Page 112: A thesis submitted for the fulfilment for the requirements

90

Data analysis

Univariate determinants of hollow presence by dbh

A paired t-test was used in order to determine if the dbh of hollow-bearing trees was

greater than non-hollow bearing trees.

Estimating hollow type probability

Fifty iterations of a non-linear logistic least squares regression analysis were performed

on the hollow-bearing trees sample from all sites to ascertain the relationship between

the dependant variable hollow type probability and the independent variable dbh size

class. Significance levels were set at 0.05 for all analyses. Calculated probability

values for all trees were subsequently used in recruitment cohort modelling to estimate

the proportion of trees with hollows over time. Statistical analysis was conducted using

Statistica Ver. 7 (Statsoft. 2005).

Tree, disturbance and environmental variables as predictors of a tree containing a

hollow

Eight predictor variables presented for analysis were selected a priori from those known

to have strong biological influences on hollow formation. A correlation matrix was

performed to determine colinearity between variables using Spearman’s Rank

correlation. A correlation coefficient value |r| > 0.7 was chosen to identify pairs of

highly correlated variables (Garden et al. 2010). As expected, tree age was found to be

highly correlated to dbh (0.752) and was thus removed from analysis while all other

variables were retained. An inspection of the relationship between the response variable

(probability of a tree containing a hollow) and the explanatory variables; fire damage,

termite damage, epicormic growth, wind damage, dbh, bark type, slope, and elevation

was undertaken using pairwise plots to ascertain the strength of the relationships and if

they were linear. Poisson Generalised Linear Mixed Effects Models (GLMM) was run

as plots were nested within sites.

Interaction models between these variables were included in the analysis and 27

candidate models were run (Appendix 5). Models were ranked according to their Δi

Page 113: A thesis submitted for the fulfilment for the requirements

91

values. The higher the Δi value, the less accurate the model for the given data. The best

approximating models that had a Δi ≤ 2 are presented (Burnham and Anderson 2002).

Because of uncertainty in some of the final model selections, a model averaging

approach was applied. Variables were ranked according to their relative overall

importance by summing the Akaike weight (wi) from all model combinations where

each of the variables occurred. Statistical analysis was conducted in R version 3.1

statistical software (R Core Team 2012) using the glmmML (Brøstrom 2012) and

MuMIn (Barton 2012) packages.

Results

A total of 5132 recruitment trees representing 33 eucalypt species was recorded from all

sites. Lophostemon confertus accounted for the most recruitment trees (826), followed

by Corymbia intermedia (714), Eucalyptus carnea (487), E.crebra (436) and E.

propinqua (397) (Table 5.6). Most recruitment trees (85.2%) were within the 10-40 cm

dbh size class (Figure 5.1). Trees with hollows had a significantly greater dbh (49.3 ±

0.75 S.E.) than recruitment trees (23.5 ± 0.18 S.E.) (F = 23.08 d.f. = 10.36; p < 0.05).

Figure 5.1: The dbh range of recruitment trees compared to all trees surveyed.

0

250

500

750

1000

1250

1500

1750

2000

Num

ber

of tr

ees

dbh category (cm)

All treesRecruitment Trees

Page 114: A thesis submitted for the fulfilment for the requirements

92

Table 5.6: The 33 species of recruitment trees sorted by bark type and absolute

abundance in all sites. Bold numbers represent the cumulative total for each bark type.

Species Common name Bark type n N E.microcorys Tallowwood Stringy-bark 222

L.suaveolens Swamp box Stringy-bark 257 E.carnea Broad-leaved white mahogany Stringy-bark 487 E.tindaliae Queensland white stringy-bark Stringy-bark 263 E.resinifera Red mahogany Stringy-bark 94 E.eugenioides Thin-leaved stringy-bark Stringy-bark 27 E.robusta Swamp mahogany Stringy-bark 24 E.acmenoides White mahogany Stringy-bark 22 E.baileyana Bailey's stringy-bark Stringy-bark 22 1418

C.intermedia Pink bloodwood Tessellated bark 714 C.gummifera Red bloodwood Tessellated bark 311 C.trachyphloia Brown bloodwood Tessellated bark 34 A.woodsiana Rough-barked apple Tessellated bark 61 S.glomulifera Turpentine Tessellated bark 1 1121 E.pilularis Blackbutt Half-bark 187

E.moluccana Gum-topped box Half-bark 10 E.tessellaris Moreton Bay ash Half-bark 4 L.confertus Brush box Half-bark 826 1027

C.citriodora.v Spotted gum Smooth-bark 363 E.propinqua Small-fruited grey gum Smooth-bark 397 E.racemosa Scribbly gum Smooth-bark 78 E.longirostrata Grey gum Smooth-bark 40 E.grandis Flooded gum Smooth-bark 38 E.tereticornis Forest red-gum Smooth-bark 16 E.saligna Sydney blue-gum Smooth-bark 14 E.major Grey gum Smooth-bark 11 C.henryi Large-leaved spotted gum Smooth-bark 10 E.biturbinata Grey gum Smooth-bark 9 E.seeana Narrow-leaved red gum Smooth-bark 3 979 E.siderophloia Ironbark Iron-bark 147

E.fibrosa Broad-leaved red iron-bark Iron-bark 3 E.dura Grey iron-bark Iron-bark 1 587

Total 5132

Page 115: A thesis submitted for the fulfilment for the requirements

93

Estimating hollow and hollow type probability

For the entire cohort of hollow-bearing trees the chances of a tree having a hollow

increased sharply between 50–100 cm dbh. The probability that a tree would definitely

have a hollow was reached at a dbh of 140 cm (F = 113444.1; d.f. = 2; p < 0.001). The

probability estimate of a hollow-bearing tree containing a certain hollow type based on

dbh revealed that the chances for butt (F = 299; d.f. = 2; p < 0.001), branch end (F=

1063.2; d.f. = 2; p < 0.001), bayonet type hollows (F= 706.5; d.f. = 2; p < 0.001), fissure

(F= 34.0; d.f. = 2; p < 0.001), trunk main (F= 121.7; d.f. = 2; p < 0.001), trunk top (F2=

20.7; d.f. = 2; p < 0.001) and branch main (F= 58.4; d.f. = 2; p < 0.001) to be present in

a tree increased significantly with dbh (Figure 5.2).

Predictors of trees containing hollows

Within all models, dbh and tree disturbance variables (wind damage, epicormic growth

and termite damage) influenced the potential of a tree to contain a hollow more than

environmental variables (Table 5.7). Individually, dbh and disturbance are the most

important variables. Elevation and fire were found to be negatively correlated with the

presence of hollow-bearing trees. Confidence intervals performed weakly for all

variables except for dbh, epicormic growth, wind and termite damage (Table 5.8).

Table 5.7: Model comparisons for explanatory tree, disturbance and environmental

variables for assessing probability of a tree having a hollow. Only models with a ∆i < 2

are presented.

Model AICc ∆i wi

dbh + epicormic growth + fire scar + bark type + termite

damage + wind damage 4443.7 0.00 0.407

dbh + epicormic growth + fire scar + slope + termite damage

+ wind damage 4443.8 0.14 0.379

Page 116: A thesis submitted for the fulfilment for the requirements

94

Figure 5.2: Probability of a tree having a hollow for (a) all hollow types; (b) bayonet; (c) butt; (d) branch end; (e) fissure; (f) trunk main; (g) trunk top and (h) branch main hollows. Values represent the number of hollow-bearing trees in each dbh size class category.

0 20 40 60 80 100 120 140 160-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Prob

abili

tyof

at r

eeha

ving

aho

ll ow

47 80

160

171

134

7841

32

18

138 6 2 1 1

(a)

0 20 40 60 80 100 120 140 160-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Prob

a bili

tyo f

bayo

neth

ollo

ws

234

8095

9455

24 16

16

6

2

3

0 0

1

(b)

0 20 40 60 80 100 120 140 160-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Prob

abili

tyfo

rbu

ttho

ll ow

s

21 40 4040

31 1812

8

11

0

54

0 0

1

(c)

0 20 40 60 80 100 120 140 160-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Prob

a bili

tyo f

bra n

c he n

dho

llow

s3 5 35

56

60

40

19

25

15

12

4

62 1 1

(d)

0 20 40 60 80 100 120 140 1600.0

0.2

0.4

0.6

0.8

1.0

1.2

6 4 7 2 4 4 31

2

0

1

0 0 0 0

Prob

abili

tyo f

fi ssu

reh o

ll ow

s

(e)

0 20 40 60 80 100 120 140 160-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

212 10 7 6 68 4

6

1

2

0

1

0 0Prob

abili

tyof

trun

km

ain

h oll o

ws

(f)

0 20 40 60 80 100 120 140 160

dbh (cm)

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Prob

abili

tyo f

t run

kto

pho

llow

s

1 2 3 3 2 13

0

11

0 0 0 0 0

(g)

0 20 40 60 80 100 120 140 160

dbh (cm)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 0 3 2 2 23

41

3

0 0 0 0 0

Prob

abili

t yof

b ran

chm

ain

h ollo

ws

(h)

Page 117: A thesis submitted for the fulfilment for the requirements

95

Table 5.8: Parameter estimates for explanatory tree, disturbance and environmental

variables used to model the probability of a tree having a hollow. Variables are ordered

by their relative importance.

Parameter Estimate

Standard

error

Confidence

interval 2.5%

97.5%

Relative variable

importance

wind damage 0.570 0.139 0.297 0.843 1.00

termite damage 0.224 0.094 0.040 0.409 1.00

epicormic growth 0.062 0.135 -0.202 0.327 1.00

dbh 0.026 0.002 0.022 0.030 1.00

fire scar -0.050 0.082 -0.212 0.111 0.81

bark type 0.013 0.009 -0.004 0.032 0.62

slope 0.043 0.331 -0.606 0.693 0.21

elevation -0.022 0.040 -0.102 0.057 0.06

Recruitment trees, hollow development and the influence of land clearing

Data are presented on the current cohort of 5132 recruitment trees (i.e. non hollow-

bearing trees) showing a change over time in the numbers of trees within each dbh

category (Figure 5.3). At Year 1, the majority of trees are ≤ 40 cm dbh (85%), with

only five trees >90 cm dbh. The mean dbh is 26.9 cm (± 0.19 S.E.). The number of

hollow-bearing trees increases to almost 1200 by year 10 and peaks at 1643 in year 50

before declining to 1184 at year 150 (Table 5.9). Due to the predicted land clearing

rate of 69%, approximately 71% of habitat will be lost along the urbanisation gradient

over the next 150 years (Figure 5.4).

Page 118: A thesis submitted for the fulfilment for the requirements

96

Figure 5.3: The predicted number of trees by dbh category over 10, 50, 100 and 150

years showing the increasing dbh through time of the current cohort of recruitment

trees.

Table 5.9: Summary of the current cohort of recruitment trees transitioning to hollow-

bearing trees over 150 years from the 45 sites within the study area. The loss of hollow-

bearing tree due to natural mortality are presented as well as the loss of remnant habitat

over time.

Year Recruitment

trees (n)

Hollow bearing trees (n)

Less Tree

Mortality (0.01)

Less Tree

Mortality (0.05)

Habitat remaining

(ha)

1 5132 0 0 0 6650.96 10 3555 1164 (25%) 1152 1106 6048.07 50 1737 1643 (49%) 1627 1561 4142.37 100 782 1450 (65%) 1436 1378 2590.66 150 295 1184 (80%) 1173 1125 1627.15

0

400

800

1200

1600

2000

20 30 40 50 60 70 80 90 100 110 120 130 140 150 160

No.

tree

s

dbh category (cm)

Year 1

Year 10

Year 50

Year 100

Year 150

Page 119: A thesis submitted for the fulfilment for the requirements

97

Figure 5.4: Cumulative loss (mean ± SD) of hollow bearing trees from the recruitment

cohort of trees within urban forest patches due to loss of habitat due to land clearing for

urban development and infrastructure over a 150 year time frame.

Discussion

Factors influencing hollow formation

The distribution and abundance of hollow-bearing trees within the landscape is

influenced by various disturbance processes (Lindenmayer et al. 1997; van der Ree and

McCarthy 2005; Adkins 2006) which alter their availability. Therefore, knowledge of

the factors affecting the presence of hollows in trees in the face of disturbance is

important for their management. Results indicate that the presence of hollows in trees

was significantly influenced by dbh, where larger trees were more likely to have

hollows. Estimations of the probability of a tree containing a certain hollow type based

on dbh, demonstrated that all hollow types are potentially present in any given dbh size

class but that this probability generally increased with dbh. While this is consistent with

previous findings (Fox et al. 2008; Koch et al. 2009; Goldingay 2011; Lindenmayer et

al. 2012), the models run in this study found that within urban systems a complex mix

Year 10 Year 50 Year 100 Year 150

Time scale

0

10

20

30

40

50

60

70

80

Perc

enta

ge o

f hol

low

-bea

ring

tree

s lo

st

Page 120: A thesis submitted for the fulfilment for the requirements

98

of tree disturbance and environmental variables influence hollow development.

Hollows are more likely to form in trees with a large dbh consistent with other studies

(Mackowski 1984; Lindenmayer et al. 1993; Gibbons et al. 2000b); although the

probability of a tree with a large dbh containing a hollow is in part a function of decay

resistance in ageing trees (Rudman 1965). Diameter at breast height was the most

important predictor of hollow-bearing trees. However, the probabilities of hollow

formation in fissure, trunk-main, trunk-top and branch-main hollows was less than other

hollow types. This pattern is likely due to the relatively small number of recorded

observations for these hollow types (combined n = 225) compared to butt, branch end

and bayonet hollow types (combined n = 1918).

The significance of hollow types and probability of formation is important as

preferences for specific hollow types by vertebrate fauna have been well documented in

the literature (Lindenmayer 2002; Goldingay 2009; Koch et al. 2009; Goldingay 2011).

Findings from this current study support this, in that most species were observed using

small (10 cm) bayonet type hollows (Chapter 6). This study also found that butt, branch

end and bayonet hollows have a strong relationship to dbh; i.e. with increased dbh there

was a greater probability a tree containing one of these hollow types. However, tree

species differ in their tendency for hollow formation (Wormington et al. 2003;

Todarella and Chalmers 2006; Fox et al. 2009), as a consequence of differing growth

rates, susceptibility to decay and site management history (Eyre 2005). For example,

45% of butt-hollows were found in Lophostemon confertus. Lophostemon confertus is a

fire sensitive species; which may facilitate the formation of butt-hollows in larger trees

that have been repeatedly burnt (Chapter 3).

The disturbance variables; termite damage, wind damage and epicormic growth (as a

surrogate measure of disturbance) also influenced the presence of hollows in trees and

were equally important as dbh. Typically termites gain entry at the base of a tree where

damage has occurred or via a fungal infection (Werner and Prior 2007). Epicormic

growth is a response by trees to disturbances such as fire (Waters et al. 2010), logging

(Hooper and Sivasithamparam 2005), insect (Stone and Coops 2004) and wind damage

(Staben and Evans 2008) with wind damage being a primary cause of hollow formation

in urban areas (Harper et al. 2005a).

Page 121: A thesis submitted for the fulfilment for the requirements

99

Within the literature fire has been found to play an important role in hollow-formation

(Inions et al. 1989b; Adkins 2006; Haslem et al. 2012) but the exclusion of fires from

urban forest patches may explain why fire was not an important variable in the models

analysed here. Alternatively, the influence of fire may have been underestimated in the

models as it has also been found to remove large numbers of hollows from the

landscape, depending on fire intensity (Inions et al. 1989). Furthermore, smooth-barked

eucalypts shed their fine outer layer of bark annually (Brooker and Kleinig 2004)

removing signs of non-destructive fires. Trees with this bark type account for 73% of

hollow-bearing trees recorded in this study; possibly resulting in the underestimation of

fire damage and hence its importance in hollow formation in urban forest patches. The

acknowledgment and understanding of fire ecology is significant for all forest types

across Australia. However, there have been relatively few studies conducted within the

forests of south-east Queensland. Therefore, currently there are no sound published

recommendations for fire management for any vegetation community within the region

(Tran and Wild 2000). However, hazard reduction burns are carried out on the Gold

Coast on a regular basis (Gold Coast City Council 2009).

Hollow-bearing tree recruitment

The Gold Coast City council has committed to making land available on the Gold Coast

for housing and infrastructure until 2019 (Gold Coast City Council 2009). The amount

of land cleared could be as much as 442 ha per annum across the shire based on past

historical clearing rates on the Gold Coast. There is no mention if this is a target to be

reached, or if it is a maximum allowed; nor is there any reference as to where clearing

will/can take place. Therein, while there is some recruitment of hollow-bearing trees

into the population in forest patches (the majority of the current recruitment cohort

having hollows by the year 150), there is also considerable loss of trees through land

clearing during this time (up to 70%). With these predicted rates of loss, the

implications to urban biodiversity in general may in some cases be extreme. The

immediate consequences of habitat loss results in the formation of forest patches of

varying size and shape (Fahrig 2002; Aguilar et al. 2008) such as those found along the

urbanisation gradient within this study. Loss of habitat coincides with a reduction in

population size and an increase in the degree of isolation across the landscape, with

these phenomena recognised as being globally significant driving forces in biodiversity

loss (Cook et al. 2002; McGarigal and Cushman 2002; Villard 2002; Lindenmayer and

Page 122: A thesis submitted for the fulfilment for the requirements

100

Fischer 2006). This also appears to be a limiting factor for hollow-bearing trees along

the gradient in the future. However, habitat loss is only one disturbance factor and this

may act in synergy with others with corresponding management implications.

Management implications of hollow-bearing tree recruitment

It is generally accepted that declines in the distribution and abundance of species

coincide with habitat loss (Cameron 2006; Lindenmayer and Fischer 2006; Didham et

al. 2007a). As demonstrated above, these patterns also hold for structural habitat

features such as hollow-bearing trees. Ensuring the persistence of those features in

urban landscapes may therefore require active management.

Without adaptive management plans in place, hollow-bearing trees and tree-hollows

will continue to decline. Coinciding with this decline in hollow-bearing trees will be

the decline in the fauna that relies upon this resource (Goldingay and Stevens 2009;

Lindenmayer et al. 2009; Manning et al. 2012). The question should not be how many

trees to retain but, how many hollow-bearing trees to manage for (Mackowski 1984)?

This current study has shown that tree disturbance and environmental variables are

important predictors of hollows in trees, as well as type of hollow. However, the long

length of time taken for hollow formation may be superseded by other factors of a much

shorter time frame e.g. changes in priorities by managers as well as changing legislation

and policy (Chapter 2). Therefore, in formulating active adaptive management

strategies for hollow-bearing trees in urban landscapes managers will firstly need to

define a series of clear objectives in relation to the density, type and size requirements

for hollow-bearing trees. These objectives should be based on the known distribution

and abundance of the resource within forest patches (as identified by this study) as well

as green space and gardens. Monitoring of the distribution and dynamics of the hollow-

bearing tree resources as well as their use by fauna will therefore inform the review of

objectives over time. The measures needed to define goals for the conservation of the

resource in the long term, may require active intervention such as artificial hollow

formation, erection of nest boxes and habitat restoration. Subsequent monitoring of

natural and artificial hollow use as well as the ability to adapt and change policy and

management guidelines is required. Progressive silviculture management techniques

may promote hollow-bearing tree formation. For example, planting eucalypt trees

Page 123: A thesis submitted for the fulfilment for the requirements

101

within a mixed species stand has been found to facilitate growth rates e.g. interactions

between E.pilularis and E.grandis which were found to have increased yields ranging

between 10-30% in production forests (Forrester and Smith 2010). As the findings of

this study have demonstrated that the presence of hollows in trees is significantly

influenced by dbh; therein, the ability to increase growth rates would decrease the time

taken to hollow formation.

Page 124: A thesis submitted for the fulfilment for the requirements

102

Chapter 6: Forest patch occupancy and use of hollow-bearing trees by vertebrates along the urbanisation gradient.

Introduction

In Australia and elsewhere, increased awareness of conservation and biodiversity

objectives being solely restricted to National Parks has been reassessed and urban

environments are now being recognised for their ability to provide a range of ecosystem

functions and services with the potential to enhance and broaden conservation

objectives (Lunney and Burgin 2004a; Davison and Ridder 2006). In 2001, the

Australia State of the Environment Report, stated that, ‘Overall, the condition of

biodiversity in Australia today is poorer than it was in 1996’ (Lunney and Burgin

2004b). The Australia State of the Environment Report 2011, delivers a similar

message, ‘Human activities have the potential to further reduce genetic, species and

ecosystem biodiversity, which will seriously affect the delivery of environmental

benefits to Australians and reduce our quality of life’(Hatton et al. 2011). With 90% of

the Australian population living in urban centres by 2050 (Zipperer and Pickett 2001),

urban ecology in regards to biodiversity and conservation should be an important

consideration for policy makers and city planners now and into the future (Lunney and

Burgin 2004b). As to whether policy makers and planners will take up the challenge

remains to be seen. The results from Chapter 2 suggest otherwise, in that biodiversity

conservation and hollow-bearing trees in particular did not feature prominently on

council websites potentially reflecting a lower order priority.

Within Australia, areas that once provided habitat for many species have often become

major centres of urbanisation (Lindenmayer and Fischer 2006). It is widely

acknowledged that urbanisation has caused many species to be displaced (Gaston

2010b), species diversity has decreased (Luck and Smallbone 2010) and localised

extinctions have occurred and will continue to do so (Gaston and Fuller 2007). Since

European settlement, certain species of Australian native fauna have proven to be

resilient to habitat disturbance (Braithwaite 2004), as predicted by McKinney (2006),

Page 125: A thesis submitted for the fulfilment for the requirements

103

these species are urban ‘winners’, able to exploit the surrounding anthropogenic

landscape.

Urbanisation typically results in a highly fragmented landscape containing a complex

matrix of remnant vegetation that is sometimes connected to larger contiguous forested

areas (Bunnell 1999; Davis et al. 2013; Harrisson et al. 2013). Faunal responses to

urbanisation and the associated fragmentation of natural habitats can be at times

difficult to interpret as cities are not generally known for their wild animals or places.

This can, and often does, lead to false social and political perspectives about urban

wildlife (Lunney and Burgin 2004b). Since most Australians live in urban

environments, urban wildlife is what will be encountered on a daily basis (Lunney

2004). However, ’urban wildlife’ may elicit different responses from different people,

depending on the circumstances surrounding any interactions (Lunney 2004).

Perceptions of how urban wildlife responds to urbanisation may therefore be influenced

by these perspectives, and a change in perspective, may be what is required to conserve

urban wildlife.

Previous studies have reported both positive and negative responses by fauna to the

effects of urbanisation (Low 2003; Ross 2004; Durant et al. 2009; Taylor and

Goldingay 2009; Cardilini et al. 2013). Recent interest in species adaptation to

urbanisation has revealed a significant divergence between urban and rural conspecific

populations demonstrating a wide range of responses depending on the traits being

investigated (Evans 2010). Nevertheless, species responses to the impacts of

urbanisation have been broadly categorised as being either ‘matrix-occupying’

(winners) or ‘matrix-sensitive’ (losers) based on similarities between species and their

ability to persist within the urban environment as well as the surrounding natural

environments (McKinney 2002; Catterall 2004; Garden et al. 2006; Didham et al.

2007a). Within urban environments there are four main mechanisms driving spatial

variation which are likely to generate intraspecific responses (Evans et al. 2010) and

these include;

• Differences in the history of urban development;

• biotic factors, such as potential colonial species;

Page 126: A thesis submitted for the fulfilment for the requirements

104

• the quality of urban habitats i.e. human attitudes to urban wildlife as well as

environmental factors; and

• the ability of populations (e.g. rural) to acquire specific traits that increase their

ability to adapt to urban environments.

Faunal responses to urbanisation are often strongly associated with changes in their

habitats (Rankmore and Price 2004; Luck and Smallbone 2010). Urbanisation has

resulted in the fragmentation of native vegetation into small isolated remnants in many

regions (Alberti and Marzluff 2004) with the urban matrix being the dominant, most

extensive and often the most modified landscape type (Lindenmayer and Fischer 2006).

The resultant fragmented landscape alters species interactions, the trophic structure of

communities, and movement of individuals through the landscape as well as resource

flow between habitats (Lindenmayer et al. 2000b; Ewers and Didham 2006b; Brearley

et al. 2010). Natural habitat remnants are therefore important features of fragmented

urban landscapes with parks or reserves having direct value for urban biodiversity

(Cornelius and Hermy 2004).

The ecological value of remnants to local fauna is frequently dependent upon the

provision of habitat resources (Harper 2005; Harper et al. 2008; Goldingay 2011). The

distribution and abundance of these resources are affected by urbanisation, as has been

shown for hollow-bearing trees (Harper et al. 2005a). Studies conducted in Australia

have shown that there is a general paucity of hollow-bearing trees in Australia’s urban

environments (Harper et al. 2005a; Harper et al. 2008; Luck and Smallbone 2010).

Consequently, urban areas within Australia do not generally support a high diversity of

hollow-using species (Garden et al. 2006). There is however, a knowledge gap in

relation to studies of fauna and their relationship with hollow-bearing trees along an

urban gradient. Only two published papers were found within Australia (Harper et al.

2005a; Harper et al. 2008), which highlights the need for further research in this area.

Against this backdrop this chapter quantifies the richness and relative occupancy of

hollow-using vertebrate fauna found within forest patches along an urban gradient on

Page 127: A thesis submitted for the fulfilment for the requirements

105

the Gold Coast in south-east Queensland, a rapidly urbanising region of Australia. This

chapter tests the hypothesis that smaller, isolated urban habitats would support more

depauperate communities dominated by common species, compared to larger and more

contiguous forest patches. Specifically, this chapter will determine whether hollow

dependent fauna are able to persist within the urban environment, and whether their

richness and occupancy rates are a function of the features of forest patches along an

urban gradient. This information will contribute to our understanding of forest patches

in various urban settings and their ability to support habitat specialist fauna.

Methods

A full description of methodologies for the establishment of sites and their classification

along the gradient can be found in Chapter 3. For this particular Chapter, 34 sites

(including the five control sites), were selected from the original 45 sites established for

the hollow-bearing trees and vegetation assessment component of this project Chapter 3

(Table 6.1 and Appendix 3 for GPS locations). The subset of 34 sites (78 plots) was

chosen from within the original data due to varying access and safety issues that were

revealed during the establishment of sites.

Various survey methodologies have previously been advocated to ensure that

representative samples of multiple taxonomic groups are captured (Garden et al. 2007a).

Many studies report on the limitations or relative success of different methods for

detecting mammal and/or reptiles; such as pit-fall traps (Catling et al. 1997), Elliot traps

(Clemann et al. 2005), wire cage traps (Catling et al. 1997), direct observations and

active searches (Ryan et al. 2002), hair tubes, vocalisations and direct signs such as

tracks, scats, diggings or scratches (Mills et al. 2002). All studies unanimously agreed

that different survey methods are useful for sampling particular fauna species, and that

no single approach will capture all species within a community. The selection of

methods used however, is a decision that will influence the accuracy of the survey

outcomes (Garden et al. 2007a).

A number of different survey methods were used in this study to capture the range and

taxonomic diversity of hollow using vertebrate fauna along the urbanisation gradient.

These include; (i) point count diurnal surveys, (ii) spotlighting, (iii) stagwatching, (iv)

Page 128: A thesis submitted for the fulfilment for the requirements

106

Anabat and (v) pole camera surveys. Broader landscape variables such as logging and

fire indices as well as a tree disturbance index that captured estimates of epicormic

growth, wind, termite and fire damage were quantified at a site level. An urbanisation

index was calculated along the gradient (see Chapter 3) as well as environmental and

tree variables (Table 6.2).

Table 6.1: Area and hollow-bearing tree density of sites used for fauna surveys.

Standard errors are supplied for sites with multiple plots. c = control, p = peri-urban, r

= rural, u = urban.

Site No: of plots

Urban gradient

Area (ha)

HBT density (ha)

Nerang Forest Reserve 5 c/p 1668.76 61.43 ± 7.93 Karawatha Forest Reserve 7 c/p 824.33 25.5 ± 4.63 Bonogin Conservation Area 6 c/u 725.07 33.92 ± 7.8 Venman Bushland National Park 6 c/p 434.24 25 ± 5.9 Daisy Hill Forest Reserve 5 c/u 432.06 23.56 ± 6.6 Mt Nathan Conservation Area 3 p 266.07 33.33 ± 7.8 Syndicate Rd 3 p 185.31 39.28 ± 10.71 Kerkin Rd 1 p 176.90 42.85 Woody Hill, Hart's Reserve 2 p 91.08 44.6 ± 12.5 The Plateau Reserve 1 p 26.53 92.85 Aqua Promenade 2 p 13.10 60.71 ± 10.15 Elanora Conservation Area 2 p 13.06 71.48 ± 7.14 Piggabeen Rd 1 p 2.97 14.28 Wongawallen Conservation Area 2 r 383.67 30.35 ± 26.78 Numinbah Rd 2 r 157.55 28.57 ± 10.71 Trees Road Conservation Area 2 r 53.35 44.64 ± 12.49 Stanmore Park 1 r 50.87 7.14 Reedy Creek Conservation Area 3 r 44.85 26.19 ± 11.72 Austenville Road 3 r 24.98 34.52 ± 8.5 Behms Road 1 r 14.29 39.29 Tallowwood Rd 1 r 14.24 50 Andrews Reserve 1 r 10.82 28.57 Carrington Rd 1 r 5.99 39.29 Tomewin Rd 1 r 5.10 57.14 Gilston Rd 1 r 4.04 35.71 Olsen Ave 2 u 38.94 7.14 Burleigh Ridge 2 u 17.89 53.56 ± 35.71 Pacific Highway 2 u 16.89 53.57 ± 0 Brushwood Ridge 2 u 16.87 39.28 ± 3.51 Herbert Park 3 u 16.45 28.56 ± 7.43 Tugun Hill Conservation Area 1 u 14.38 32.14 Summerhill Court Reserve 1 u 9.53 50 Miami Conservation Area 1 u 4.41 10.71 Burleigh Knoll National Park 1 u 4.15 85.71

Page 129: A thesis submitted for the fulfilment for the requirements

107

Table 6.2: Landscape, environment and tree variables quantified at each plot.

Variable Description of variable

Landscape variables

Site area Area in hectares for each forest patch. Logging Index Presence/absence of logging history from historical

aerial imagery. Fire Index Presence/absence of fire history from historical aerial

imagery Connectivity (%) Percentage of site perimeter connected to other

forested areas. Environmental variables

Angle of slope (°) Calculated from GIS layers (Chapter 3. Equation 3.2).

Aspect (°) Directional orientation.

Elevation (m) Obtained from GIS mapping layers.

Urbanisation Index Calculated from GIS data on infrastructure, cadastre and regional ecosystems (Chapter 3. Equation 3.3).

Density of hollow-bearing trees

Density of hollow-bearing trees per hectare per site.

Tree disturbance index Combined tree disturbance parameters i.e. fire, termites, wind damage and epicormic growth.

Stand density Density of all trees per hectare per site.

Tree species diversity Tree species diversity calculated using the Shannon Wiener diversity index H = – ∑ (pi) (ln pi).

Diurnal surveys

The point count method was used to identify hollow-using avian fauna observed or

heard at each plot. Point count methods are generally highly efficient in terms of data

quality and survey effort. This method allows for data to be collected in relation to

habitat use and can be used to measure relative and absolute abundance. Point counts

are also frequently used to survey bird communities in various habitats (Sewell and

Catterall 1998; Luck and Daily 2003; Sutherland 2006; Elliott et al. 2010). Surveys

were undertaken between 24/11/2009 and 14/03/2011. Start times at each site were

staggered, but always took place between 9.00 a.m. and 11 a.m. Surveys were

undertaken for 30 minutes from the central point of each plot. There was an initial

settling in period of 10 minutes prior to each survey to allow birds to become

accustomed to human presence. One observer would then identify birds by sight and

Page 130: A thesis submitted for the fulfilment for the requirements

108

individual calls heard. Any other hollow-using vertebrate fauna observed were also

recorded.

Spotlighting

Spotlighting is frequently used to survey nocturnal fauna in Australian environments

(Ward 2000; Eyre 2006) and can be applied to both plot (Garden et al. 2007a) and

transect surveys (Eyre 2006). For this study all plots were surveyed a minimum of three

times between 12/07/2011 and 29/05/2012 with spotlighting start times being rotated

across each site and plot per visit. To account for variable species behaviour plots were

extensively searched for hollow-using fauna by walking slowly through the plot for 10

minutes using a 50-100 watt variable spotlight. Travel times and any species observed

while walking to plot points were also recorded. All hollow-using mammalian fauna

observed were identified to species level assisted by individual variations in eyeshine,

as well as size, colour and markings, gliding style and individual character traits (Table

6.3). If an animal was observed leaving a hollow the species of tree was also recorded.

Hollow-using fauna heard were also recorded at each plot. As the resultant survey

effort for each site was a combination of travel and survey times, fauna abundance

measures were standardised to the number of observations per hour of survey effort

prior to analysis.

Stagwatching

Stagwatching is commonly used to observe or detect hollow using vertebrate fauna in

Australia (Smith et al. 1989; Lindenmayer et al. 1990; Lindenmayer et al. 1994a; van

der Ree and Loyn 2002). Previous stagwatching studies make use of large numbers of

observers (Smith et al. 1989), which were not available for this study. Two observers

were stationed at the plot centroid, each scanning one half of the plot for signs of

wildlife emerging from hollows. Stagwatching began approximately 10 minutes before

sunset and ended 15 minutes after sunset with each plot having a total survey effort of

50 person minutes. Any hollow-using fauna observed or heard were identified to

species level. Tree species and hollow characteristics that animals were observed

emerging from were also recorded. All plots were surveyed once between 12/07/2011

and 30/05/2012.

Page 131: A thesis submitted for the fulfilment for the requirements

109

Table 6.3: Features that can be used to distinguish the different species of gliders and

possums (Menkhorst and Knight 2001; Lindenmayer 2002).

Common name Scientific name Distinguishing features Feathertail glider Acrobates pygmaeus Small size, pen-quill-like tail, white underbelly

fur and rapid movements. Eyeshine: tiny brilliant white.

Sugar glider Petaurus breviceps Medium-sized body,' yap-yap' call. Tail is often white-tipped. Smaller than the Squirrel glider and face is shorter. Ears are shorter and more oval than the Squirrel glider. Eyeshine: pale red.

Squirrel glider Petaurus norfolcensis Similar to Sugar glider but larger and the tail is larger and more thickly furred. Tail is sometimes black-tipped but never white-tipped. Belly fur is always white; deep guttural gurgle call reminiscent of the last part of the Yellow-bellied glider call. Eyeshine: pale red.

Yellow-bellied glider

Petaurus australis Highly active, loud gurgling call, pale-red eyeshine. Ears not heavily furred like the Greater glider. Body fur is olive-grey and black on the limbs. Eyeshine: pale red.

Greater glider Petauroides volans Pendulous tails, bright white eye-shine, slow movements and prolonged periods of inactivity. Largely silent. Ears heavily furred. Eyeshine: brilliant white.

Common brushtail possum

Trichosurus caninus A cat sized arboreal and terrestrial possum: ears obviously longer than broad. Fur colour, length and density variable in E. Australia mostly uniform silver grey above with cream underparts; belly often stained yellow-orange; tail black, only slightly brushy. Call a characteristic loud series of rattling nasal coughs and hisses. Eyeshine: red/orange.

Mountain brushtail possum

Trichosurus vulpecula A large heavy-set possum typically of wet forests; upper parts uniformly dark grey, but can be blackish, dark brown or reddish: underparts cream or yellow-orange: tail thick and black. Ears broadly oval. Call a short sequence of sharp grunts of coughs, not a rattling series as in the common brushtail possum. Eyeshine: red.

Common ringtail possum

Pseudocheirus peregrinus Species in S.E Qld have bright orange face, limbs and flanks with rufous underparts and white patches behind and below ears; blackish flecked rufous back. Tail rufous with white terminal quarter, short-haired, tapering, prehensile, often carried in a coil. Call, soft, high-pitched, insect-like chirruping and harsher ‘zip zip’. Eyeshine: red.

Page 132: A thesis submitted for the fulfilment for the requirements

110

Anabat surveys

A ‘space for time’ approach (Fahrig et al. 1995; Germaine and Wakeling 2001) was

used to rapidly sample micro-bat species richness from all sites between August 2011

and November 2011 using an Anabat II™ detector (Titley Electronics, Ballina,

Australia). Active monitoring with Anabat II™ recorded echolocation calls digitally to

a 2 Mb flash card. The detector was placed at ground level, angled at 45° with a

researcher monitoring and saving calls as they were detected for a period of 20 minutes.

No sampling was conducted during rain events or periods of high winds. Plots were

surveyed on three occasions with each plot having a total survey effort of 1 hour. Start

times were rotated across each plot per visit; the earliest surveys starting at dusk 17:00

hours and the latest surveys commencing at 20:00 hours.

Microbat surveys are routinely undertaken using harp traps (Hourigan et al. 2008) or

echolocation recording (Duffy et al. 2000) and subsequent identification. While some

studies suggest that restricting methods to acoustic surveys alone has certain limitations

(Barclay 1999; Reardon 2009; Adams et al. 2010), others have found that these provide

reliable measures of micro-bat communities (Dixon 2012). During a study undertaken

in Brisbane, a highly urbanised region in south-east Queensland, significantly more

species were identified using acoustic surveys alone than either harp traps alone or the

two methods used in conjunction (Hourigan et al. 2008). The space for time

methodology used in this survey is expected to provide a valid assessment of

insectivorous bat assemblages within the urban landscape.

Bat call identification

Echolocation calls, stored as individual files, were identified to species level using

Analook™, Bat Calls of New South Wales (Pennay et al. 2004), and Key to Bat Calls

of south-east Queensland and north-east New South Wales (Reinhold et al. 2001).

Positive identifications were only made where a minimum of three pulses classified to

the same species were identified. As part of this process pulses were excluded if they

were unable to be identified to species level. Species richness was calculated by the

assessment of presence/absence data from each site for all nights surveyed.

Page 133: A thesis submitted for the fulfilment for the requirements

111

Pole camera

Previous studies have used pole mounted cameras to monitor wildlife in a variety of

elevated microhabitats e.g. nest boxes, nests and tree hollows (Purcell 1997; Richardson

et al. 1999; Huebner and Hurteau 2007; Luneau and Noel 2010). For this study a small,

lightweight infra-red security camera (model: CCG8110 from OzSpy Ltd.) hard wired

to a rechargeable 9 volt power source was mounted to a flexible arm at the top of a 12 m

fiberglass telescopic pole. Video footage (3GP format) from the camera was

transmitted via a cable to a monitor and digital recording device (2.4" TFT-LCD

monitor with 640 x 240 pixel resolution). At each accessible hollow from all sites

surveyed (i.e. those ≤ 12 m n = 500 hollows) the pole was extended to the height of the

hollow and the camera was manoeuvred into position to obtain a clear view of the

hollow cavity. For each hollow surveyed, the presence or absence of any vertebrate

fauna was recorded. Any animals found were identified to species level, as well as the

number of occupants. Other signs of hollow-use such as eggs, feathers or fur were also

recorded. Surveys were conducted between 16/01/2012 and 11/05/2012 with each

hollow being surveyed once.

Species occupancy

The presence of a species at a site over time provides information about the relative

value of the habitat for each species. Therefore, the occupancy of sites i.e. the

proportion of those where a species is present (MacKenzie et al. 2006), provides a

useful means to assess the value of these sites for hollow-dependent fauna. To evaluate

the presence or absence of a species at any given location with any confidence requires

more than two surveys be undertaken (Wintle et al. 2005). A minimum of eight surveys

per site were undertaken during this study, determined by site area and the number of

plots within each site. Site occupancy was determined by quantifying the presence of

individual species from all surveys which were then grouped as either birds or mammals

to broadly compare the occupancy between these two taxonomic groups.

Page 134: A thesis submitted for the fulfilment for the requirements

112

Statistical analysis

Determinants of species richness and relative occupancy of hollow-using vertebrate

fauna along an urbanisation gradient

Eleven predictor variables were modelled; site area, the urbanisation index,

connectivity, slope, aspect, elevation, individual tree disturbance index (i.e. the

presence/absence of fire, termite, wind damage and epicormic growth), tree species

diversity, hollow-bearing tree density per hectare, fire and logging indices. Predictor

variables used in the models were selected a priori from those known to have strong

biological influences. A correlation matrix was performed to determine colinearity

between variables using Spearman’s Rank correlation. Variables with a correlation

coefficient value |r| > 0.7 were considered to be highly correlated (Garden et al. 2010).

The decision to retain or remove variables was based on its known/perceived ecological

significance as well as the degree of colinearity. Aspect was removed as an

environmental variable for two reasons. Firstly, it was found to perform weakly in the

models and secondly there was insufficient data in the case of wildlife use of tree

hollows to undergo rigorous analysis. Remaining predictor variables (Table 6.3) were

retained. Data transformation is recommended where explanatory variables are of

dissimilar nature (Zuur et al. 2010); such as those used in this analysis. An arcsine

square root transformation was undertaken for proportional explanatory variables while

a logarithmic transformation (ln[x+1]) was undertaken for all other explanatory

variables due to scale differences. No transformations were made to the response

variable.

As plots were nested within sites, Poisson Generalised Linear Mixed Effects Models

(GLMM) were used to assess the importance of tree, landscape and environmental

variables on the relative abundance of hollow-using species. Interaction models

between predictor variables were included in analyses and a total of 26 candidate

models were run including the global model (Appendix 5). Models were ranked

according to their Δi AIC values. The higher the Δi value, the less accurate the model

for the given data, thus the best approximating models that had a Δi ≤ 4 are presented

(Burnham and Anderson 2002). Model averaging was applied to remove uncertainty in

some of the final model selections. Variables were ranked according to their relative

overall importance by summing the Akaike weight (wi) from all top model

Page 135: A thesis submitted for the fulfilment for the requirements

113

combinations where each of the variables occurred. Statistical analyses were conducted

using R 3.1 statistical software (R Core Team 2012) and packages glmmML (Brøstrom

2012) and MuMIn (Barton 2012).

A paired t-test was used to compare the difference between mammal and bird

occupancy across all sites. Only these two taxonomic groups were chosen as there were

insufficient data for other faunal groups. Analysis was conducted using the SPSS

statistical program (IBM. Corporation. 2012). Linear regressions were performed to

determine whether mammal and bird occupancy by site was a function of tree species

diversity. A MANOVA comparing four factors (occupancy, taxon, urbanisation index

and site) was then run in Statistica (Statsoft. 2005) to compare the difference in rates of

occupancy between birds and mammals along the urbanisation gradient.

A Chi-square (χ²) goodness-of-fit test compared avian and mammalian species

occupancy observed along the urbanisation gradient to expected occupancy based on the

assumption that each part of the gradient was utilised in proportion to habitat

availability, α was set at 0.05 for all occupancy analyses.

Non-metric Multidimensional Scaling (nMDS) assessed similarities in species

occupancy along the urbanisation gradient using the Bray-Curtis similarity index. A

Kruskal’s stress test for measure of fit was performed prior to undertaking a Principle

Coordinates Analysis (PCO) to measure the distances between sites on the ordination to

match the corresponding dissimilarities in community structure. A Pearson rank

correlation with a threshold |r| value of 0.6 was then used to identify those species

accounting for the variation in occupancy across the gradient. A total of nine sites had

less than four species and were excluded from analysis. This analysis was performed

using the statistical program Primer-E (Clarke and Warwick 2001b).

Page 136: A thesis submitted for the fulfilment for the requirements

114

Results

Forty-two native hollow-using vertebrate species were observed during this study.

These species included 13 mammals, 19 birds, five frogs and four reptiles (Table 6.6).

The Rainbow Lorikeet, Trichoglossus haematodus (250) and the Little Corella, Cacatua

sanguinea (230) were found in very high numbers at two sites due to the presence of

roosts. When individual counts of species are taken into consideration the common

brushtail possum, Trichosurus vulpecula (58) and Laughing Kookaburra, Dacelo

novaeguineae (39) represented the highest number of hollow-using species across all

sites. Spotlighting as a survey method recorded the greatest number of species (26)

followed by diurnal surveys (21) (Table 6.4). Karawatha Forest Park and Venman

Bushland National Park had the highest diversity of hollow-using fauna with 21 and 18

species respectively (Table 6.5), while 11 other sites (including four control sites) had

10 or more species recorded.

Page 137: A thesis submitted for the fulfilment for the requirements

115

Table 6.4: The number of sites where vertebrate species were observed using each survey method. The total number of sites where a species

was found as well as the total number of individuals recorded (in brackets) is also presented. Relative species occupancy across all sites is also

estimated. Roosting sites or large numbers of individuals recorded at night are estimates and have a ~ next to their values.

Species Common name

Survey method

Spotlighting Stagwatching Diurnal Pole Camera Anabat Total No. sites where found

Relative species occupancy

Mammals Petaurus breviceps Sugar glider 5 4 0 1 - 10 (13) 0.21

Petaurus norfolcensis Squirrel glider 5 3 0 0 - 8 (17) 0.18 Trichosurus vulpecula Common brushtail possum 10 0 0 1 - 11 (58) 0.47 Trichosurus caninus Mountain brushtail possum 2 0 1 0 - 3 (5) 0.09 Petauroides volans Greater glider 1 1 0 0 - 2 (3) 0.06 Pseudocheirus peregrinus Common ringtail possum 3 0 0 0 - 3 (3) 0.09 Rattus fuscipes Bush rat 2 0 0 0 - 2 (~21) 0.06 Insectivorous bats

Chalinolobus gouldii Gould's wattled bat 0 0 0 0 13 13 (22) 0.32 Vespadelus pumilus Eastern forest bat 0 0 0 0 5 5 (9) 0.12 Scotorepens greyii Little broad-nosed bat 0 0 0 0 4 4 (5) 0.15 Saccolaimus flaviventris Yellow-bellied sheath-tail bat 0 0 0 0 4 4 (3) 0.19 Austronomus australis White-striped free-tail bat 0 0 0 0 3 3 (1) 0.03 Vespadalus darlingtoni Large forest bat 0 0 0 0 2 2 (3) 0.09 Birds

Ninox strenua Powerful Owl 1 0 0 0 - 1 (2) 0.03 Ninox conivens Barking owl 2 0 0 0 - 1 (2) 0.06 Ninox novaeseelandiae Southern Boobook Owl 12 1 0 1 - 14 (29) 0.50 Aegotheles cristatus Australian Owlet Nightjar 6 0 0 0 - 6 (6) 0.18 Colluricincla harmonica Grey Shrike Thrush 0 0 1 0 - 1 (2) 0.06 Cacatua galerita Sulphur Crested Cockatoo 4 1 3 0 - 8 (25) 0.26 Calyptorhynchus funereus Yellow-tailed Black Cockatoo 0 0 1 0 - 1 (6) 0.03 Calyptorhynchus lathami Glossy Black Cockatoo 0 0 2 0 - 2 (7) 0.06 Cacatua roseicapilla Galah 1 1 2 0 - 4 (~29) 0.15 Cacatua sanguinea Little Corrella 2 0 0 0 - 2 (~230) 0.06 Alisterus scapularis King Parrot 0 1 3 0 - 4 (10) 0.02 Trichoglossus chlorolepidotus Scaly Breasted Lorikeet 0 0 2 0 - 2 (4) 0.09

Page 138: A thesis submitted for the fulfilment for the requirements

116

Table 6.4 continued: The number of sites where vertebrate species were observed using each survey method. The total number of sites where

a species was found as well as the total number of individuals recorded (in brackets) is also presented. Relative species occupancy across all

sites is also estimated. Roosting sites or large numbers of individuals recorded at night are estimates and have a ~ next to their values.

Species Common name

Survey method

Spotlighting Stagwatching Diurnal Pole Camera Anabat Total No. sites where found

Relative species occupancy

Trichoglossus haematodus Rainbow Lorikeet 2 0 3 0 - 5 (~250) 0.24 Glossopsitta pusilla Little Lorikeet 0 0 1 0 - 1 (2) 0.06 Dacelo novaeguineae Laughing Kookaburra 11 0 2 0 - 13 (39) 0.47 Cormobates leucophaeus White Throated Tree Creeper 1 0 1 0 - 2 (4) 0.12 Todiramphus sanctus Sacred Kingfisher 1 0 3 0 - 4 (11) 0.09 Platycercus adscitus Pale Headed Rosella 0 0 2 0 - 2 (7) 0.03 Platycercus eximus Eastern Rosella 3 0 1 0 - 4 (11) 0.06 Chenonetta jubata Australian Wood Duck 1 1 1 0 - 2 (10) 0.09 Frogs

Litoria fallax Eastern dwarf tree frog 8 0 0 0 - 8 (9) 0.63 Litoria tyleri Tyler's tree frog 4 0 0 0 - 5 (3) 0.38 Litoria caerulea Green tree frog 1 0 1 1 - 3 (2) 0.25 Litoria peronii Emerald tree frog 2 0 0 0 - 2 (5) 0.13 Litoria gracilenta Graceful tree frog 1 0 0 0 - 1 (1) 0.00 Reptiles

Morelia spilota Carpet python 1 0 0 0 - 1 (1) 0.13 Dendrelaphis punctulatus Green tree snake 0 0 2 0 - 2 (2) 0.25 Boiga irregularis Brown tree snake 0 0 1 0 - 1 (1) 0.13 Varanus varius Lace monitor 0 0 2 0 - 2 (3) 0.25 Pogona barbata Eastern bearded dragon 0 0 2 0 - 2 (2) 0.25 Total number of species for each survey method 26 8 21 4 6

Page 139: A thesis submitted for the fulfilment for the requirements

117

Diurnal surveys

For diurnal surveys 21 hollow-using species were recorded across all sites (Table 6.5) the

Rainbow Lorikeet T. haematodus, Sacred Kingfisher Todiramphus sanctus, Australian King-

Parrot Alisterus scapularis, and Sulphur-crested Cockatoo Cacatua galerita, were all seen at

3 sites. The greatest diversity of reptiles was also recorded during diurnal surveys, with only

one mammal and one frog species being detected (Table 6.5).

Spotlighting

Twenty-two species of hollow-using vertebrate fauna were found across all sites from 108

hours of survey effort (Figure 6.1). Species richness for mammals and birds reached a

plateau after 62.5 and 94 hours respectively of spotlighting effort. The common brushtail

possum T.vulpecula, Southern Boobook Owl Ninox novaeseelandiae and Laughing

Kookaburra D. novaeguineae were the species most often recorded across all sites (Table

6.6).

Figure 6.1: The number of bird and mammal species found to occupy all sites as a function

of cumulative survey effort during spotlighting surveys.

Stagwatching

Eight hollow-using species were found during stagwatching surveys, with the sugar glider,

Petaurus breviceps being the most common species found (4 sites in total) (Table 6.5). The

greatest number of species found on one site during stagwatching was two; the common

brushtail possum T.vulpecula and the squirrel glider Petaurus norfolcensis. The majority of

sites (71%) recorded zero species using this method.

0

1

2

3

4

5

6

7

02468

101214161820

0 10 20 30 40 50 60 70 80 90 100 110

Mam

mal

ric

hnes

s

Bir

d ri

chne

ss

Effort (hours)

bird

mammal

Page 140: A thesis submitted for the fulfilment for the requirements

118

Table 6.5: The number of species recorded during surveys at each site; sorted by site with the highest number of species found. Some species

were found by more than one method. Therefore, in some instances, taxon totals will be different from method totals. Species tallies during

spotlighting include those observed transit to and from plots.

Site Urban

gradient Diurnal Spot-

lighting Stag-

watching Pole

Camera Anabat Total Birds

(n)

Mammals inc. bats

(n) Reptiles

(n) Amphibians

(n) Karawatha Forest Reserve c 4 13 1 0 3 21 11 6 1 1 Venman Bushland National Park c 2 13 1 0 2 18 8 6 1 2 Woody Hill, Hart's Reserve p 2 4 2 3 3 14 3 6 0 1 Nerang Forest Reserve c 4 6 0 0 3 13 6 4 1 1 Daisy Hill Forest Reserve c 1 6 2 0 3 12 6 6 0 0 Austenville Road r 1 9 1 0 0 11 5 2 0 2 Syndicate Rd p 3 6 1 0 1 11 5 1 0 1 Burleigh Knoll National Park u 5 2 1 0 1 9 8 0 0 0 Mt Nathan Conservation Area p 4 2 0 0 1 7 5 1 1 0 Numinbah Rd r 2 5 0 0 0 7 3 1 0 1 Kerkin Rd p 0 4 0 0 2 6 3 3 0 0 Miami Conservation Area u 4 2 0 0 0 6 4 1 0 0 Bonogin Conservation Area c 2 3 0 0 1 6 4 1 1 0 Brushwood Ridge u 2 4 0 0 0 6 3 1 1 1 Reedy Creek Conservation Area r 1 3 1 0 0 5 4 0 0 0 Aqua Promenade p 0 4 0 0 1 5 3 2 0 0 Elanora Conservation Area p 1 3 0 0 1 5 2 1 1 0 Behms Road r 2 3 0 0 0 5 1 1 1 2 Tallowwood Rd r 0 4 0 0 0 4 2 2 0 0 Tomewin Rd r 3 1 0 0 0 4 2 1 0 0 Carrington Rd r 0 3 0 0 1 4 0 1 0 2 The Plateau Reserve p 1 2 0 0 1 4 2 2 0 0 Andrews Reserve r 1 1 1 0 1 4 1 1 0 0

Page 141: A thesis submitted for the fulfilment for the requirements

119

Table 6.5 continued: The number of species recorded during surveys at each site; sorted by site with the highest number of species found.

Some species were found by more than one method. Therefore, in some instances taxon totals will be different from method totals. Species

tallies during spotlighting include those observed in transit to and from plots.

Site Urban

gradient Diurnal Spot-

lighting Stag-

watching Pole

Camera Anabat Total Birds

(n)

Mammals inc. bats

(n) Reptiles

(n) Amphibians

(n) Pacific Highway u 0 3 0 0 0 3 3 0 0 0 Trees Road Conservation Area r 0 2 0 0 1 3 2 1 0 0 Piggabeen Rd p 0 4 0 0 0 4 2 0 0 2 Burleigh Ridge u 0 0 1 0 1 2 1 1 0 0 Wongawallen Conservation Area r 1 1 0 1 0 3 1 1 0 0 Herbert Park u 0 2 0 0 0 2 0 1 0 1 Gilston Rd r 0 2 0 0 0 2 0 0 0 2 Olsen Ave u 0 2 0 0 0 2 1 0 0 1 Summerhill Court Reserve u 0 2 0 0 0 2 1 1 0 0 Stanmore Park r 0 1 0 0 0 1 1 1 0 0 Tugun Hill Conservation Area u 0 1 0 0 0 1 0 1 0 0 Total species richness 21 22 8 4 6 20 13 5 5

Page 142: A thesis submitted for the fulfilment for the requirements

120

Anabat survey

The detector recorded echolocation frequencies over 75 nights for 75 hours of survey effort.

Six hollow-using insectivorous bat species were recorded across all sites; an additional three

species of non-hollow using insectivorous bat were also found along the urbanisation

gradient. Gould’s wattled bat, Chalinolobus gouldii was the most common species with 22

echolocations recorded, followed by the eastern forest bat, Vespadelus pumilus with nine

echolocations recorded. All other species had five or fewer echolocations recorded (Table

6.5).

Pole camera

A total of 196 hollows (9% of all hollows) were investigated. Only four observations of four

species were made at two sites using the pole camera. A common brushtail possum, T.

vulpecula, was observed occupying a 15 cm trunk-top hollow at 4.7 m in a Eucalyptus

siderophloia (Figure 6.2) and a sugar glider, P. breviceps was found in a 15 cm trunk-main

hollow at 6.2 m in a L. confertus. Two eggs, from a pair of Southern Boobook Owl, N.

novaeseelandiae were found in a large ~40 cm branch-end hollow at 9.6 m in a Eucalyptus

tereticornis (Figure 6.3). Eggs were identified upon observing the parents guarding the

hollow during spotlighting surveys. A green tree-frog, Litoria caerulea was observed in a

water-filled bayonet hollow at 1 m located in a dead tree (Figure 6.4). Overall, the results

from the pole camera survey were poor with 97% of all hollows checked being unoccupied

when surveyed. Ten hollows of the 196 investigated were found to be false hollows, in that

they did not have openings, or where openings were present, these did not extend beyond one

or two centimetres.

Figure 6.2: Trichosurus vulpecula found occupying a Eucalyptus siderophloia hollow.

Page 143: A thesis submitted for the fulfilment for the requirements

121

Figure 6.3: The eggs of Ninox novaeseelandiae found in a large E. tereticornis hollow.

Figure 6.4: Litoria caerulea in water filled hollow.

*The frog was discovered while using the pole camera. Given the location of the hollow (1m) this image was taken with an SLR camera (Photo: N. Rakotopare).

Tree species, hollow type and use

Of the 916 available hollow-bearing trees 24 (2.6%) were found to be occupied. Hollow

using fauna were observed entering or leaving 11 species of eucalypt trees including standing

dead trees. Tree species use was dominated by three species (48%) in which fauna were

observed entering or leaving hollows. Standing dead trees were found to be the most utilised

tree type for vertebrate fauna (n = 8; 24%), followed by C. citriodora (12%) and L.confertus

(12%). Trunk main (36%) was the most commonly used hollow type followed by trunk top

(24%). A hollow entrance diameter of 10 cm was the size in which most wildlife (33%) were

observed utilising, at a height between 0-5 m (36%). Seven species of birds, six mammals

(including bats) and one amphibian were observed using hollows (Table 6.6). No reptiles

were observed using hollows in this study.

Page 144: A thesis submitted for the fulfilment for the requirements

122

Table 6.6: Tree species, hollow type and size that hollow-using fauna were found in, from all surveys. Highlighted rows represent the highest number of

hollow-using species found in each category. Numbers are a representation of the number of each species found in each hollow.

Hollow using fauna

Galah

Pale-headed Rosella

Boobook Owl

Sulphur-crested Cockatoo

Rainbow Lorikeet

Scaly-breasted Lorikeet

Laughing Kookaburra

Common brushtail possum

Squirrel glider

Mt. brushtail possum

Sugar glider

Greater glider Bats

Green tree-frog Total

Tree species E. racemosa 2

2

C. intermedia

1

1

1 3 L. confertus

2

1 1

4

E. grandis

1 1

2 C. citriodora.v

1 1

2

4

E. tereticornis

2

2 E. microcorys

1

1

1

3

dead 1 6 1 8 C.gummifera

1

1

E. carnea

1

1 E. pilularis

2 1

colony

3

Hollow type

Bayonet

0

Butt hollow

1

2

3 Branch end

1

3

1

1 6

Branch main 2 1

1

4 Trunk main 3 2 2 1 1 1 1 1 12 Trunk top

1 5

2 colony

8

Fissure 0 Hollow size (cm)

10 2 2 2 2 2 10 15 2 1

2

1

6

20

1 2 2

2

7 30

2

4 1

7

50 1 2 colony 3

Page 145: A thesis submitted for the fulfilment for the requirements

123

Table 6.6: The tree species, hollow type and size that hollow-using fauna were found in, from all surveys. Highlighted rows represent the highest number of

hollow-using species found in each category. Numbers are a representation of the number of each species found in each hollow.

Hollow using fauna

Galah

Pale-headed Rosella

Boobook Owl

Sulphur-crested Cockatoo

Rainbow Lorikeet

Scaly-breasted Lorikeet

Laughing Kookaburra

Common brushtail possum

Squirrel glider

Mt. brushtail possum

Sugar glider

Greater glider Bats

Green tree-frog Total

Hollow height (m)

0-5

2

2 4 1 1

2 12 6-10 2 1 2

2

1

1

9

11-15

1

1

2

4 16-20

1

4

5

21-25

1 2

colony

3

Page 146: A thesis submitted for the fulfilment for the requirements

124

Drivers of hollow-using vertebrate species richness along an urbanisation gradient

Site area was the most important variable influencing the distribution of hollow-using

species. Of all 26 models tested, the model with the combined variables site area and

tree species diversity was the best fit when explaining the distribution of hollow-using

species, followed by the model containing site area as a single variable (Table 6.7). Site

area and tree species diversity were both positively correlated with the species richness

of hollow using vertebrate fauna, whereas logging, fire, stand density and slope were

negatively correlated with species richness. Apart from site area and tree species

diversity, confidence intervals were poor in regards to all other variables along the

urbanisation gradient (Table 6.8). As a reflection of the urbanisation gradient the

urbanisation index did not explain any patterns in species richness along the gradient.

Table 6.7: Best approximating model comparisons of landscape and the urbanisation

gradient variables associated with the number of hollow-using vertebrate species found

across all sites. Only Akaike models with ∆i values ≤ 4 are presented.

Model AIC ∆i wi site area + tree species diversity 70.2 0.00 0.455

site area 71.8 1.62 0.203

site area + fire index + tree species diversity 72.8 2.64 0.120

tree species diversity 73.0 2.85 0.110

tree disturbance + tree species diversity 73.9 3.76 0.069

Page 147: A thesis submitted for the fulfilment for the requirements

125

Table 6.8: Parameter estimates of the fragment, landscape and urbanisation gradient

variables associated with the number of species found at each site. Variables for each

analysis are ordered by their relative importance.

Parameter Estimate Standard error

Confidence interval 2.5% 97.5%

Relative variable importance

site area 0.167 0.066 0.037, 0.298 0.78

tree species diversity 0.687 0.316 0.066, 1.308 0.77

fire index -0.007 0.026 -0.058, 0.044 0.13

tree disturbance index 1.337 1.053 -0.726, 3.402 0.08

logging index -0.470 0.877 -2.191, 1.249 0.02

hbt density/ha 0.044 0.171 -0.291, 0.379 0.02

slope -0.824 0.856 -2.502, 0.853 0.01

urbanisation gradient 0.057 0.072 -0.085, 0.200 0.01

stand density -0.001 0.001 -0.004, 0.001 0.01

elevation 0.000 0.002 -0.003, 0.005 0.01

connectivity 0.005 0.005 -0.004, 0.015 0.00

Page 148: A thesis submitted for the fulfilment for the requirements

126

The species richness of hollow-using vertebrate fauna increased with tree species

diversity (R2 0.577; p < 0.05) (Figure 6.5). Conversely, occupancy rates of hollow-

using vertebrate fauna decreased with tree species diversity (R2 0.623; p < 0.05) (Figure

6.6).

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8

Tree species diversity

0

2

4

6

8

10

12

14

16

18

20

22

24

26

Spec

ies

rich

ness

Figure 6.5: The relationship between tree species diversity and hollow-using

vertebrate species richness across all sites.

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8

Tree species diversity

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

Spec

ies

occu

panc

y

Figure 6.6: The relationship between tree species diversity and hollow using

vertebrate species rates of occupancy across all sites.

Page 149: A thesis submitted for the fulfilment for the requirements

127

Vertebrate assemblages and relative occupancy of fauna along the urbanisation

gradient.

The Australian Owlet Nightjar, squirrel glider, common brushtail possum and Gould’s

wattled bat account for 26.9% of species variation observed among sites, with all of

these species corresponding to non-urban areas of the urbanisation gradient (Figure 6.7).

The relative abundance of birds and mammals was high along the urbanisation gradient

compared to all other taxa but there was no difference in the occupancy rates between

birds and mammals across all sites (paired t-test = 1.3; d.f. = 33; p = 0.197). There was

however, a difference in occupancy rates between birds and mammals along the

urbanisation gradient (F = 11.66; d.f. = 3, 110; p < 0.01). Control sites had lower

occupancy rates than all other sites along the urbanisation gradient with occupancy

being similar between urban, rural and peri-urban sites (Figure 6.8). Despite area being

a significant determinant of species richness in the models, there was no difference in

observed occupancy rates along the urbanisation gradient compared to expected rates

for either taxon (χ² = 0.0037, d.f. = 3, 0.25 > p > 0.10). There was also no relationship

between bird (R2 = 0.005; p = 0.7807) or mammal (R2 = 0.004; p = 0.7501) occupancy

and site area.

Page 150: A thesis submitted for the fulfilment for the requirements

128

Figure 6.7: The amount of variation in species relative abundance along the

urbanisation gradient. The Australian Owlet Nightjar and squirrel glider account for

14.1% (PCO1) while the common brushtail possum and Gould’s wattled bat accounted

for 12.8% (PCO2).

Page 151: A thesis submitted for the fulfilment for the requirements

129

Control Peri urban Rural Urban

Urban Gradient

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

Rat

e of

occ

upan

cy Birds Mammals

Figure 6.8: Occupancy rates of birds and mammals along the urbanisation gradient.

Discussion

General summary

This study recorded 42 hollow-using vertebrate species along an urbanisation gradient,

representing 34% of hollow-using species that are known to occur regionally in south-

east Queensland and 15% nationally. Overall more species were found during

spotlighting surveys (n = 22), confirming that spotlighting is an effective technique

when surveying arboreal mammals (Catling et al. 1997; Goldingay and Sharpe 2004b;

Goldingay and Sharpe 2004c). As no further species were found after 94 (birds) and

62.5 (mammals) hours of spotlighting effort this suggests that the species accumulation

threshold for forest patches along the urbanisation gradient has been reached. While

stagwatching was not overly successful during this study, other studies have found

stagwatching to be effective when examining the relationship between tree species used

as nest sites, the height of hollows and the type of hollow used by arboreal marsupials

(Smith et al. 1989; Lindenmayer et al. 1990; Lindenmayer et al. 1994b; Lindenmayer

2002). In this regard stagwatching was able to provide in indication of the type of

hollow used by vertebrate fauna for this study.

Page 152: A thesis submitted for the fulfilment for the requirements

130

The application of extendable pole cameras to investigate tree hollows has received

mixed reviews. All studies report similar problems associated with height restrictions,

precluding many hollows from investigation thus impacting results (Purcell 1997;

Richardson et al. 1999; Huebner and Hurteau 2007; Luneau and Noel 2010). During

the current study 500 hollows were identified as potential survey hollows for use with

the pole camera. However, many were inaccessible due to the location of the

tree/hollow; or the hollow entrance was at an awkward angle, while some were false

hollows. As a result only 39% of the 500 potential hollows could actually be surveyed.

False hollows are tree deformities (particularly branch hollows located on the top side

of branches) or had openings that did not have a depth beyond one or two centimetres.

In addition, the classification of hollows in this study was made during ground based

surveys where the limitations of such surveys have been acknowledged (Harper et al.

2004; Koch 2008; Stojanovic et al. 2012). Nevertheless, they are useful for providing

relative, rather than true hollow abundance surveys (Koch 2008).

Tree species, hollow type and use

Standing dead trees are recognised as an important structural feature for hollow-using

fauna within Australian forests (Moloney et al. 2002; Koch et al. 2009) and this study

supports these findings in that standing dead trees accounted for the most observations

of use by fauna (24%). This is likely to be due to the fact that there is a greater

likelihood of dead trees to contain hollows (Eyre 2005; Harper et al. 2005a; Beyer et al.

2008). The findings from this study concurred with these previous studies, in that dead

trees accounted for the most hollow-bearing trees across all sites (13.5%), as well as

containing the most hollows (15.4%) (Chapter 3).

Most hollows in eucalypts form in branches (i.e. bayonet and branch end) due to the

process of branch shedding and consequently these hollow types account for most

hollows found in open forests (Gibbons and Lindenmayer 2002). Hollows with a

relatively small entrance size i.e. 10 cm are often preferred by hollow-using vertebrate

fauna and were found to have 44% occupancy rates in the forests of south-eastern

Australia (Gibbons et al. 2002) and 34% in Tasmania (Koch et al. 2008b). A similar

pattern in hollow sizes was found in this study where hollows with an entrance size of

10 cm were used by 33% of species found utilising hollows, while the majority (36%)

Page 153: A thesis submitted for the fulfilment for the requirements

131

of hollows used by fauna in this study were located at a height ranging between 0-5 m.

Apart from one frog and bird species, all species utilising hollows at this height were

possums and gliders. This suggests that hollow height is not a constraint for these

species, the findings of which have been reported in a similar study (Traill and Lill

1997). This provides valuable information when evaluating the suitability of hollow-

bearing trees for retention (Gibbons et al. 2002; Koch et al. 2008b; Goldingay 2011).

Ogden (2009) undertook a study of hollow-use in Karawatha Forest Park (a control site

in this current study) and found the occupancy rate to be 10%. A similar pattern was

found in this study with an overall low occupancy rate of 2.6% across all sites.

The impacts of the urbanisation gradient on species richness

Within urban environments common species are often more prevalent than specialists

(Pearman and Weber 2007; Gaston 2010b; Luck and Smallbone 2010). A similar

pattern was observed in this study where the two species most frequently found across

the urban gradient were the common brushtail possum and Laughing Kookaburra.

These species are known to be locally ‘common’ and would be categorised as urban

winners. Common species have the ability to exclude other species from resources

(Laurance 1991) as well as being able to move through, and live within, the urban

matrix, therefore tending to dominate these environments (Garden et al. 2006).

Common species however, contribute much of the structure, biomass, and energy

turnover to the majority of terrestrial systems and can exert a significant influence on

environmental conditions; thus enabling other species to co-exist (Pearman and Weber

2007; Gaston 2010b). Gaston and Fuller (2007) argue that even a small decline in the

abundance of common species can result in large absolute losses of individuals and

biomass in general, and as such, common rather than rare species are the principle

drivers of spatial variation and species richness. Potentially, this is more of a problem

in environments that have a greater degree of fragmentation and isolation (Tilman et al.

1994; Gaston 2010b) such as urban ecosystems.

Studies of species along urbanisation gradients often report a peak in diversity at

moderate levels of development with reduced species richness occurring at urban

centres (Sewell and Catterall 1998; Blair 2004; Smith and Wachob 2006). For hollow-

using fauna it appears that this pattern does not hold, as this study found no relationship

Page 154: A thesis submitted for the fulfilment for the requirements

132

between species richness overall and the degree of urbanisation. It is also important to

note that comprehensive bird surveys were not completed in this study, and results may

therefore not provide a useful comparison with other studies assessing bird communities

along urban gradients. Findings from the current study also suggest that some species

are better able to exploit the urban gradient, in particular Gould’s wattled bat,

Chalinolobus gouldii. Chalinolobus gouldii favours large flyways and low levels of

forest clutter (Lloyd et al. 2006) and is therefore highly adaptable to foraging in cities,

making them one of the most common city bats in Australia (Richards et al. 2012). The

study did find, however, that micro-bats responded differently to general predictions, as

species that occupy the forest niche high in the canopy, fly fast and have little need for

great manoeuvrability, adapt well to the uncluttered structure of urbanised areas

(Threlfall et al. 2011) (unpublished data). Micro-bats are common in urban

environments and many species (60%) are dependent on hollow-bearing trees as roost

sites (Richards et al. 2012), highlighting the importance of this resource for this class of

mammals alone. They are however; still highly reliant on hollow-bearing trees and thus

the retention of forest patches is crucial to their survival.

The effects of the urbanisation gradient, landscape and environmental variables on

fauna

The urbanisation gradient was not found to have an impact on species richness, despite

control sites having the greatest species richness overall. This is due in part to all

species found during the course of this study being classified as ‘common’ and being

able to adapt to the urban landscape. Control sites were larger forest patches with some

having greater connectivity to surrounding areas and it could be expected that these sites

would therefore support more species. Landscape ecology theory predicts that such

sites would not only have a greater number of species but also greater abundance (van

der Ree et al. 2003; Rhodes et al. 2006). This is because larger patches may be less

disturbed from matrix variables such as edge effects, isolation or urbanisation and thus

capture more environmental variability (Lindenmayer and Fischer 2006). In the current

study, site area was found to be the most important variable in relation to species

richness along the urbanisation gradient. Previous studies have unanimously agreed

that site area is the driver behind species richness (Brotons et al. 2003; Lindenmayer

and Fischer 2006; Ewers et al. 2007; Catterall et al. 2010). This highlights that the

Page 155: A thesis submitted for the fulfilment for the requirements

133

retention of species rich communities in urbanising landscapes is inextricably linked to

the retention of larger, contiguous forest patches.

Other deterministic factors beyond the site area may also be in operation, particularly at

the site scale. Plant communities frequently regulate ecological processes, community

structure and environmental services (Hooper and Vitousek 1997; Symstad et al. 1998;

Kahmen et al. 2005; Young et al. 2011). Previous research has identified that

vegetation structure, composition and plant species richness strongly influences overall

species richness and species rates of occupation along the urbanisation gradient

(Gibbons and Lindenmayer 2002; Ross et al. 2002; Catterall 2004; McElhinny et al.

2005). Vegetation is an important resource for many faunal species, with the drivers of

species diversity for both flora and fauna changing depending on unique site-specific

environmental conditions (Tews et al. 2004). This is particularly pertinent for

Australian environments where fauna are heavily dependent on the availability of

hollow-bearing trees (Goldingay 2009; Ball et al. 2011; Goldingay 2011; Davis et al.

2013). Within this study, site area and tree species diversity were the most significant

variables when accounting for species richness.

Tree species diversity had a strong positive influence on the species richness of hollow-

using fauna. Conversely, tree species diversity had a negative influence on occupancy

rates of fauna along the urbanisation gradient. This may however, be due to the low

numbers of individuals found on many sites. These results highlight the importance of a

holistic approach when planning conservation and biodiversity management strategies,

based on the individual site requitements (Felton et al. 2010) and not focusing on spatial

elements such as area and isolation alone (Holland and Bennett 2009).

Hollow-bearing trees as a resource face ongoing pressures through urbanisation

(Chapter 5) and the effects of this pressure will be obvious at the landscape scale. For

example, fauna in urban forest patches may be isolated to such a degree that relict

populations now exist at higher densities than in more contiguous environments. The

data presented here provide some support for this theory given that the occupancy rates

of fauna where generally greater in smaller more urbanised forest patches than larger

control sties. Fauna were therefore more readily detected in smaller urban fragments.

Page 156: A thesis submitted for the fulfilment for the requirements

134

Disturbance in the form of human landscape modification typically leads to local or

regional declines in species richness (Shochat et al. 2006). This is also greatly

influenced by human land-use policy along the gradient which may be solely

responsible for species richness at any given location (Chapter 2) (Luck and Smallbone

2010). In general, the results presented here concur with the ‘habitat heterogeneity

hypothesis’ (Wintle et al. 2005), which assumes that structurally complex habitats may

provide more environmental resources and in turn increase species diversity. The

presence of keystone structures, such as hollow-bearing trees, is therefore positively

correlated to the species diversity-habitat heterogeneity relationship (Tews et al. 2004).

This chapter set out to quantify the richness and occupancy of hollow-using vertebrate

fauna found within forest patches along an urban gradient. In the sampled population

the hypothesis was; that smaller, isolated, urban habitats would support more

depauperate communities, dominated by common species. While the predictions about

the dominance of common species along the gradient hold, richness was not strongly

influenced by the gradient for these hollow-using fauna. It is important to realise that

common species can sometimes be at risk of rapid decline and that complacency in their

conservation can be problematic (Lindenmayer et al. 2011). There is a need to identify,

monitor and alleviate significant events that lead to declines in species numbers (such as

the loss of hollow-bearing trees) which would involve paying more attention to the

needs of common species (Gaston and Fuller 2007). Therefore, in terms of the

implications for conservation it is paramount to undergo continual assessment of current

management strategies to ensure the constant supply of tree hollows into the future for

all fauna. The findings here have highlighted the range of variables that are at play

along the urbanisation gradient. Therein, caution is necessary to avoid making

generalised statements about the value of forest patches to hollow-using vertebrate

fauna based purely on their size and location on the urbanisation gradient.

Page 157: A thesis submitted for the fulfilment for the requirements

135

Chapter 7: General Discussion

The importance of hollow-bearing trees to faunal communities is generally acknowledged

globally (Wormington et al. 2002; Aitken and Martin 2004; Eyre 2005; Wormington et al.

2005; Ball et al., 2011; Beaven and Tangayi 2012), although there is scant reference to this

resource in relation to urban forest patches. Therefore, this study aimed to identify the biotic

and abiotic factors influencing hollow-bearing tree availability, density and use within forest

patches along an urbanisation gradient at a landscape scale. While I acknowledge that

hollow-bearing trees do exist in other parts of the landscape such as, parks, back yards and

within non-indigenous tree species, this thesis has been restricted to trees belonging to the

Myrtaceae family within forest patches. Therein, this thesis had a number of objectives and

the findings presented in the discussions of Chapters 2-6 provide the details about how these

objectives have been met. It is the purpose of this chapter to expand on these findings and

how they pertain to wider ecological and urban planning theory.

In the first instance a literature review was undertaken to highlight the importance of hollow-

bearing trees globally and the threats associated with their conservation. The review then

focused on the importance of this habitat resource within the Australian context. While the

importance of hollow-bearing trees is documented in the production forests of Australia (Eyre

2005; Koch et al. 2008b; Johnstone et al. 2013), there is relatively little literature available

from urban landscapes, with only a handful of studies investigating the impacts of

urbanisation on hollow-bearing trees (Harper et al. 2005a; Harper et al. 2008). As such, the

findings from the current study provide an important contribution to our understanding of the

relationship between hollow-bearing trees as a habitat resource and impacts of urban

development.

Page 158: A thesis submitted for the fulfilment for the requirements

136

A number of general themes emerge from this work, building on previous studies assessing

biotic responses to urbanisation. The first key outcome reported in Chapter 2 highlighted the

policy/implementation gap in providing for the conservation and management of hollow-

bearing trees at the local level, particularly for non-production forests (Treby et al. 2014).

This is an important issue, as the failure to apply conservation actions that preserve important

habitat features contributing to the functional integrity of urban ecosystems at the local level

fundamentally undermines the purpose of broader overarching legislative provisions. While

many council respondents voiced their frustrations about not being able to ‘do much’ for

hollow-bearing trees this appears to be driven more by institutional inertia than by the lack of

policy (McAlpine et al. 2002; Calver and Wardell-Johnson 2004). Institutional inertia is

widely acknowledged as a hurdle to conservation efforts globally (Karr 1990; Norman et al.

2004; Beunen 2006). Further implications of such ineffective policy measures pertain to the

future planning provisions in urban regions. Urban planners need to consider various

elements in their appraisal and delivery of development guidelines for urban areas, to meet

the needs of present and future generations (Niemelä 1999; Albers 2006; McDonnell 2007;

Gordon et al. 2009; Snep and Opdam 2010). Recognising the need to retain natural habitat

remnants may be just one of these elements for sustainable urban development.

Linked to the concept of sustainable urban development is the need to understand how

species, communities and specific habitat features respond to urbanisation. This is necessary

to determine if there is any change in the functional capacity of the urban landscape to

support biodiversity. Here the urban gradient has been proposed as a theoretical concept to

facilitate the monitoring of these changes (McDonnell 2007) and much work has been done

to assess how faunal communities respond to these gradients (Germaine and Wakeling 2001;

Crooks et al. 2004; Johnson et al. 2008; Pillsbury and Miller 2008). These studies highlight

some general patterns, such as biotic homogenisation (Knight 1999; Hooper and

Sivasithamparam 2005; McKinney 2006). Homogenisation results in a reduction of available

resourse within the landscape (Mckinney 2006), therefore the retention of hollow-bearing

trees will provide structural complexity to the environment. The concepts of urban winners

and losers in relation to the sensitivity to, and resilience of, some species to the urban

transformation process (Threlfall et al. 2012; Cardilini et al. 2013; Davis et al. 2013) has also

been revealed. The current study provides a novel contribution in that it quantified the

distribution and abundance of an important habitat feature (i.e. hollow-bearing trees)

Page 159: A thesis submitted for the fulfilment for the requirements

137

highlighting that some of the patterns reported for wildlife may not hold for habitat features

that are remnants of past land use practices.

Providing for the retention of functional elements of biodiversity in urban landscapes is

complex. An important point for consideration is the time-scale at which various elements

including policy, conservation practice and ecological processes operate (Fallding 2004).

This study highlights the mismatch between these elements and considers the implications of

these disparities for urban biodiversity. Each of the aforementioned issues is discussed in

further detail below.

Policy-implementation gaps

The analysis of the existing forms of legislative provisions and statutory regulations at

National, State and local government levels completed in Chapter 2 revealed an

overwhelming lack of policy transition into conservation action. It also found considerable

amounts of confusion surrounding biodiversity, conservation policy and legislation at all

levels of government within Australia, but particularly in relation to the conservation and

management of hollow-bearing trees. This failure of the Australian administrative system to

integrate broader level policy into local level urban planning has previously been noted by

(Humphries 1989) and the current findings suggest that little has changed in the more than 20

years that have passed. These problems are in no way restricted to the Australian

environment and similar situations have been reported from other countries. For example, in

North America Karr (1990) revealed that the narrowly defined goals of the legislation

designed to protect endangered species and ecosystems were largely inefficient in reducing

the impacts of humans at a landscape scale. Elsewhere, a break down in the communication

of legislation into on-ground implementation was also found to cause problems for

conservation efforts in the Netherlands (Beunen 2006). In the United Kingdom, Brown and

Grant (2005), list the divide between policy and implementation as one of the major barriers

limiting the integration of biodiversity values into urban planning.

It should be noted that there were a small number of local councils surveyed in the current

study that did have specific provisions in place for the conservation and management of

Page 160: A thesis submitted for the fulfilment for the requirements

138

hollow-bearing trees. This suggests that it is possible for local councils to enact broader

policy and legislative provisions. This has also been demonstrated in the past where urban

planning provisions in local council areas have assigned remnant habitats with high

ecological values ‘special environmental amenity’ status to provide protection for these

environments within urbanising landscapes (Humphries 1989). Therefore, while the current

research has shown that policy implementation at the local level seems to be a perpetual and

pervasive problem in certain areas, this may not be due to the lack of specific provisions such

as tree preservation orders (Llewellyn 1984) or similar development controls.

Humphreys (1989) suggested a number of reasons behind the failure of statutory regulations

to protect trees in urban environments. Amongst these is the policy-implementation gap, as

demonstrated here, for the conservation of hollow-bearing trees. Additional issues were the

failure of planning provisions to account for the ecological functioning of remnant habitats as

well as the failure of regulatory procedures to engage with communities. The current study

also provides a useful basis offering some perspectives to the long-term impacts of

dysfunctional or weak legislation that fails to account for ecosystem functioning. Within

Australia, the age at which a tree produces a large hollow usually occurs at 150 years (Ross

1998). During this time there could potentially be over 35 changes in Federal government,

with as many, or more, changes in governance at the local level. This temporal scale

mismatch between development planning and ecological analysis has previously been

reported (Fallding 2004). This study highlights that; conservation legislation at all levels of

government will need to be robust enough to defend itself from radical changes brought about

by individual party policies (e.g. the proposed reintroduction of logging concessions within

national parks in Queensland and world heritage sites in Tasmania). Furthermore, future

legislation will also need to focus more on implementation and decision making processes,

particularly at the local level, to achieve positive conservation outcomes within contemporary

urban planning. While it appears that Local Agenda 21 provides the legislative backdrop for

local action, the results from Chapter 2 suggest that far greater engagement with this process

is required by local councils.

Page 161: A thesis submitted for the fulfilment for the requirements

139

Contemporary urban planning

Global human populations are non-randomly distributed, with a bias towards urban centres

(Zipperer and Pickett 2001; Gaston et al. 2010). Human population projections for Australia

are also substantial and this is expected to have significant effects on both human quality of

life and natural environments (Zipperer and Pickett 2001; Byrne et al. 2010; Bekessy et al.

2012). Despite these projections, there appears to be considerable ad hoc planning in regards

to both the integration of biodiversity values (Fallding 2004; Bekessy et al. 2012) as well as

the retention of urban resident amenity values (Byrne et al. 2010).

Chapter 2 highlighted the poor implementation of conservation actions targeting hollow-

bearing trees at the local level within Australia. However, the specific issues raised above,

extend beyond hollow-bearing trees and are symptomatic of a larger biodiversity planning

problem. As Niemelä (1999) argued, there are three key pre-requisites needed to capture

biodiversity into planning provisions. The first is an improved basic knowledge of urban

nature. The second is an understanding of urban ecological processes and the third is the

need of ecosystem-specific management schemes. As such the current study provides

valuable information pertaining to the value of urban forests in each of these three areas.

First, it provides the managers and planners of the Gold Coast with an inventory of hollow-

bearing trees, their distribution and relative abundance but also how these resources are

affected by urbanisation (Chapters 3 and 4). Second, it demonstrates how this particular

keystone resource is likely to change over time considering future urban growth projections

and highlights the importance of considering these longer time frames in future planning

(Chapter 5). Finally, it highlighted the value of urban forest patches for the protection of not

only hollow-bearing trees but also the fauna that are reliant upon these (Chapter 3 and 6).

While these findings can now be used in future urban planning decision making it is

important that they are considered early in the development process as this has been a key

limitation in the past (Fallding 2004). However, it is important to note that the current study

focuses on the values of only one type of urban landscape element and its associated biota,

i.e. natural forest patches in various urban settings. Within the urban context there is marked

variation in the concept of urban greenspace and various typologies of park lands (Byrne and

Sipe 2010). Additionally, there are differing social values placed on the importance of trees

Page 162: A thesis submitted for the fulfilment for the requirements

140

and natural areas where these are perceived as being sacred, utilitarian, decorative or

hazardous with some sectors of the community being indifferent to them (Breuste 2004;

Kirkpatrick et al. 2013).

These results highlight the various socio-political and ecological management facets that

local councils need to address when managing for trees and natural habitats in general, within

an urbanising landscape. Recent advances advocate the use of systematic conservation

planning, or reverse conservation planning (i.e. identification of the areas most suited to

urban development) to assist decision making for urban planners (Gordon et al. 2009;

Bekessy et al. 2012). However, as Bekessey et al. (2012) highlight these systematic planning

tools should be used to assist in decision making rather than being seen to provide

prescriptive recommendation for urban planning. They also emphasised the importance of

robust baseline data that are used to develop the planning models. It is important to note that

an element not yet captured in these conservation planning tools is the human dimension, and

in particular the social and amenity value placed on natural remnants by urban residents.

Undeniably improved urban planning is required. One way forward lies with strategic and

active adaptive management (McCarthy and Possingham 2006; Armitage et al. 2009). Here

the various and often contradictory values from differing stakeholders can be addressed in the

early stages of the planning domain in order to derive common objectives.

The opportunity exists for local Australian councils, with the support of overarching policy

directives, to implement a more community based model for the management of urban

forests. Environmental planners could therefore potentially follow a set of prescribed targets

offering ecological justification for specific planning goals within urban environments that

would enhance biodiversity as well as accommodating the interests and needs of community

stakeholders.

Natural resource management: gradients, biotic responses and temporal scales

The current study has revealed that urban ecosystems are important to biodiversity, and more

importantly, that people and their activities create a unique set of circumstances for natural

resource management within the urban landscape. These findings confirm the value of using

Page 163: A thesis submitted for the fulfilment for the requirements

141

the urban-rural gradient as a model to test the effects of the urbanisation process (McDonnell

et al. 1993; McDonnell et al. 1997; Hahs and McDonnell 2006; Qureshi et al. 2013).

Urbanisation creates a number of shifting effects depending on patch size, the loss and

degradation of habitat and the amount and type of human disturbance (McKinney 2008). The

urbanisation process has seen a reduction in species richness for birds, reptiles and

amphibians, while for mammals there has been a shift to medium sized, generalist species

globally (Ǻs 1999; Germaine and Wakeling 2001; Devictor et al. 2007; Evans 2010).

Generally, plant communities have shown in an increase in species richness, depending on

the level of urbanisation (Guntenspergen and Levenson 1997; Ross et al. 2002; McKinney

2008). Importantly, the responses detected for hollow-bearing trees did not follow predicted

patterns (e.g. intermediate disturbance hypothesis) in that there was little variability in the

density of this resource along the gradient. Conversely, the faunal responses reported were

similar to those published previously (Germaine and Wakeling 2001; Crooks et al. 2004;

Johnson et al. 2008) suggesting that the urbanisation gradient is exerting a much stronger

influence on these taxa than their specific habitat requirements (i.e. for tree hollows).

In addition to the changing pace of urbanisation as a driver behind the variations detected in

natural forest patches, was the importance of historical land use practices. The logging

history of any given region results in a disjointed array of forest patches across the landscape

(Gibbons and Lindenmayer 2002; Andrieu et al. 2011), and also alters the availability of

resources within these disturbed landscapes (Flynn et al. 2011; Law et al. 2013). The

importance of past logging on the Gold Coast was evident in this study and these practices

were the driving factor influencing the density of hollow-bearing trees in urban forests

(Chapter 4). The enduring effects of these past logging practices will need to be addressed by

natural resource managers well into the future given the length taken before hollow formation

occurs in eucalypt trees.

Previous work has shown that the combined effects of urbanisation and logging results in a

reduction in overall stand density in urban remnants (Pickett et al. 2001). However, the mean

density of hollow-bearing trees found in the current study (37.47 trees/ha) was greater than

that reported from other urban and managed forests in Australia (Lamb et al. 1998; Moloney

et al. 2002; Harper et al. 2005a; Eyre et al. 2010). Despite the greater prevalence of hollow-

Page 164: A thesis submitted for the fulfilment for the requirements

142

bearing trees on the Gold Coast this figure may overestimate the functional value of the

resources since the majority of these hollows were small, potentially limiting their use by

hollow-dependant fauna, as larger hollows are generally more valuable for wildlife (Saunders

et al. 1982; Goldingay 2009; Johnstone et al. 2013). This result indicates that the absolute

hollow-bearing tree densities reported here should not simply be taken at face-value but that

their functional significance within the landscape should be considered. This has

implications for management and future research directions.

Recommendations and further research

The loss of hollow-bearing trees from across the landscape has been documented globally

(Lindenmayer et al. 1991b; Mazurek and Zielinski 2004; Holloway et al. 2007; Politi and

Hunter 2009; Smith et al. 2009). Past land use practices and urbanisation continue to

threaten the persistence and availability of hollow-bearing trees in urban landscapes. As

noted previously the density of hollow-bearing trees and hence individual hollows may be

inflated on the Gold Coast due to the dominance of small and possibly unusable hollows

within urban forest patches. While these trees may yet develop larger hollows, this process is

a lengthy one (Whitford 2002; Goldingay 2011) and may not meet the immediate wildlife

requirements. Future biodiversity planning for the retention of forest patches and their

habitat features should therefore consider the need to supplement the existing resource.

Nest-boxes that target species-specific requirements are already being used within Australian

urban landscapes as a means, of sustaining biodiversity until there are enough large hollows

to support native fauna (Harper et al. 2005b; Durant et al. 2009). In some areas the

deployment of nest-boxes did not appear to improve conditions for arboreal marsupials (Ball

et al. 2011) highlighting that local conditions may play an important factor in nest-box use.

On the Gold Coast nest-boxes are already being used as a means of remedial mitigation in the

development of residential estates in peri-urban regions (Castley, pers. comm.). Therefore,

nest-boxes may have a role to play in the management and conservation of hollow-using

vertebrate fauna within the urban context (see Chapter 3). Further research is required that

quantifies the success of such measures, or indeed, whether they are necessary in certain

regions.

Page 165: A thesis submitted for the fulfilment for the requirements

143

Future research is also required in order to understand the relationship between hollow type

and use by faunal species along the urbanisation gradient. While a number of previous

studies have investigated hollow use in the managed forests of Australia (Lindenmayer et al.

1990; Lindenmayer 2002; Koch et al. 2009), very few have investigated the number and type

of hollows required by individual wildlife species along the urban gradient. More research is

required to improve our understanding of (i) the abundance and distribution of existing

hollows, (ii) what species are using them, (iii) whether intra- and interspecific co-occupancy

occurs, and (iv) the levels of intra and interspecific competition for hollows.

The current study focused on the ecological value of one particular feature of urban forest

patches, i.e. hollow-bearing trees. However, as noted previously these habitat features

comprise only one of the myriad urban landscape elements (Byrne and Sipe 2010). Research

that quantifies the attitudes of local communities towards these forest patches as well as other

urban green-spaces are urgently required to capture the needs of residents within a larger

active adaptive management framework. Furthermore, Watts and Selman (2004) have shown

that an improved understanding of the barriers limiting the integration of biodiversity aspects

into urban planning can lead to better local implementation. Therefore, further research

should strive to determine what limits such integration in the Australian urban context.

Finally, the advent of social media may provide a ready means to obtain opinion and

feedback about urban planning projects (Schroeter and Houghton 2011). Social media

platforms not only provide opportunities for research but also wider engagement among

stakeholders. For example, sites such as ‘LinkedIn’, readily make the qualifications and

attributes of individual stakeholders publically available. With this comes the potential

ability to network and share ideas between all stakeholders, theoretically resulting in strong,

cohesive active adaptive management plans. The utility of using these new technologies in

guiding co-management strategies requires further study.

Page 166: A thesis submitted for the fulfilment for the requirements

144

References:

Adams, M. D., Law, B. S., Gibson, M. S., (2010) Reliable automation of bat call

identification for eastern New South Wales, Australia, using classification trees and

AnaScheme software. Acta Chiropterologica 12, 231-245.

Adkins, M., (2006) A burning issue: using fire to accelerate tree hollow formation in

Eucalyptus species. Australian Forestry 69, 107-113.

Aenishaenslin, S., Convery, K., Gua, B., Spain, M., Tunstall, L., (2007) Private native forest

policies in New South Wales, Queensland, Victoria and Tasmania. Small-Scale Forestry 6,

141-155.

Aguilar, R., Quesada, M., Ashworth, L., Herrerias-Diego, Y., Lobo, J., (2008) Genetic

consequences of habitat fragmentation in plant populations: susceptible signals in plant traits

and methodological approaches. Molecular Ecology 17, 5177-5188.

Aitken, K., Martin, K., (2004) Nest cavity availability and selection in aspen-conifer groves

in a grassland landscape. Canadian Journal of Forest Research 34, 2099-2110.

Akaike, H., (1973) Information theory and an exstension of the maximum likelihood

principle. Proceedings of the 2nd International symposium on information theory. Budapest,

Hungary.

Albers, G., (2006) Urban development, maintenance and conservation: planning in Germany

– values in transition. Planning Perspectives 21, 45-65.

Alberti, M., Marzluff, J. M., (2004) Ecological resilience in urban ecosystems: Linking urban

patterns to human and ecological functions. Urban Ecosystems 7, 241-265.

Alvey, A. A., (2006) Promoting and preserving biodiversity in the urban forest. Urban

Forestry and Urban Greening 5, 195-201.

Andrieu, E., Ladet, S., Heintz, W., Deconchat, M., (2011) History and spatial complexity of

deforestation and logging in small private forests. Landscape and Urban Planning 103, 109-

117.

Araújo, M. B., (2003) The coincidence of people and biodiversity in Europe. Global Ecology

and Biogeography 10, 5-12.

Page 167: A thesis submitted for the fulfilment for the requirements

145

Armitage, D. R., Plummer, R., Berkes, F., Arthur, R. I., Charles, A. T., Davidson-Hunt, I. J.,

Diduck, A. P., Doubleday, N. C., Johnson, D. S., Marschke, M., McConney, P., Pinkerton, E.

W., Wollenberg, E. K., (2009) Adaptive Co-Management for Social-Ecological Complexity.

Frontiers in Ecology and the Environment 7, 95-102.

Arnett, E. B., Kroll, A. J., Duke, S. D., (2010) Avian foraging and nesting use of created

snags in intensively-managed forests of western Oregon, USA. Forest Ecology and

Management 260, 1773-1779.

Ǻs, S., (1999) Invasion of matrix species in small habitat patches. Conservation Ecology 3, 1.

Australian Government, (2001) Appendix F: Australian Classification of Local Governments

Available at: http://www.regional.gov.au/local/publications/reports/2001_2002/appendix_f.aspx#tablef_1

(Accessed on 7/1/2013)

Avery, T., Burkhart, H., (1989) Forest Measurements. McGraw-Hill Book Company, New

York

Bai, M. L., Wichmann, F., Mühlenberg, M., (2003) The abundance of tree-holes and their

utilisation by hole-nesting birds in primeval boreal forests of Mongolia. Acta Ornithologica

38, 95-102.

Ball, I. R., Lindenmayer, D. B., Possingham, H. P., (1999) A tree hollow dynamics

simulation model. Forest Ecology and Management 123, 179-194.

Ball, T., Goldingay, R. L., Wake, J., (2011) Den trees, hollow-bearing trees and nest boxes:

management of squirrel glider Petaurus norfolcensis nest sites in tropical Australian

woodland. Australian Mammalogy 33, 106-116.

Balmford, A., Moore, J. L., Brooks, T., Burgess, N., Hansen, L. A., Williams, P., Rahbek, C.,

(2001) Conservation conflicts across Africa. Science 291, 2616-2619.

Banks, J. C. G., (1997) Tree ages and ageing in yellow box. In 'Australia's Ever-Changing

Forests III. Proceedings of the third national conference on Australian forest history. (Ed. J.

Dargavel) pp. 17-28. Centre for Resource and Environmental Studies, Australian National

University: Canberra

Page 168: A thesis submitted for the fulfilment for the requirements

146

Barclay, R., (1999) Bats are not birds: a cautionary note on using echolocation calls to

identify bats. Journal of Mammalogy 80, 290-296.

Barton, K. (2012) MuMIn: Multi-model inference 1.7.2

Bauhus, J., McElhinny, C., Alcorn, P., (2002) Stand structure and tree growth in uneven-aged

spotted gum, Corymbia maculata forests: some implications for management. Forestry 75,

451-455.

Beaven, U., Tangavi, M., (2012) Availability of hollow-bearing trees and their utilisation by

small animals in Cawston Ranch, Zimbabwe. African Journal of Ecology 50, 1-5.

Bekessy, S. A., White, M., Gordon, A., Moilanen, A., McCarthy, M. A., Wintle, B. A.,

(2012) Transparent planning for biodiversity and development in the urban fringe. Landscape

and Urban Planning 108, 140-149.

Belcher, C. A., (1995) Diet of the Tiger Quoll, Dasyurus maculatus in East Gippsland,

Victoria. Wildlife Research 22, 341-357.

Berg, S. S., Dunkerley, D. L., (2004) Patterned Mulga near Alice Springs, central Australia,

and the potential threat of firewood collection on this vegetation community. Journal of Arid

Environments 59, 313-350.

Beunen, R., (2006) European nature conservation legislation and spatial planning: For better

or for worse? Journal of Environmental Planning and Management 49, 605-619.

Beyer, G., Goldingay, R. L., (2006) The value of next boxes in the research and management

of Australian hollow-using arboreal marsupials. Wildlife Research 33, 161-174.

Beyer, G., Goldingay, R. L., Sharpe, D. J., (2008) The characteristics of squirrel glider,

Petaurus norfolcensis den trees in subtropical Australia. Australian Journal of Zoology 56,

13-21.

Blair, R. B., (1999) Birds and butterflies along an urban gradient: surrogate taxa for assessing

biodiversity. Ecological Applications 9, 164-170.

Blair, R. B., (2004) The effects of urban sprawl on birds at multiple levels of biological

organisation. Ecology and Society 9, 2.

Page 169: A thesis submitted for the fulfilment for the requirements

147

Blakely, T., Jellyman, P., Holldaway, R., Young, L., Burrows, B., Duncan, P., Thirkettle, D.,

Simpson, J., Ewers, R., Didham, R., (2008) The abundance, distribution and structural

characteristics of tree-holes in Nothofagus forest, New Zealand. Austral Ecology 33, 963-

974.

Bowen, M., McAlpine, C., Seabrook, L., House, A., Smith, G., (2009) The age and amount of

regrowth forest in fragmented brigalow landscapes are both important for woodland

dependant birds. Biological Conservation 142, 3051-3059.

Bowen, M. E., McAlpine, C. A., House, A. P. N., Smith, G. C., (2007) Regrowth forests on

abandoned agricultural land: A review of their habitat values for recovering forest fauna.

Biological Conservation 140, 273-296.

Bradshaw, C. J. A., (2012) Little left to lose: deforestation and forest degradation in Australia

since European colonisation. Journal of Plant Ecology 5, 109-120.

Bradshaw, F. J., Rayner, M. E., (1997) Age structure of the karri forest: 2. Projections of

future forest structure and impications for management. Australian Forestry 60, 188-195.

Brady, M., McAlpine, C., Miller, C., Possingham, H., Baxter, C., (2009) Habitat attributes of

landscape mosaics along a gradient of matrix development intensity: matrix management.

Landscape Ecology 24, 879-891.

Brady, M., McAlpine, C., Possingham, H., Miller, C., Baxter, G., (2011) Matrix is important

for mammals in landscapes with small amounts of native forest habitat. Landscape Ecology

26, 617-628.

Braithwaite, L. W., (1996) Conservation of arboreal herbivores: The Australian scene.

Australian Journal of Ecology 21, 21-30.

Braithwaite, W., (2004) Do current forestry practices threaten forest fauna? a perspective. In:

Conservation of Australia's Forest Fauna. (Ed.) D. Lunney, Royal Zoological society of New

South Wales, Australia, pp. 513-536.

Brearley, G., Bradley, A., Bell, S., McAlpine, C. A., (2010) Influence of contrasting urban

edges on the abundance of arboreal mammals: A study of squirrel gliders, Petaurus

norfolcensis in south east Queensland, Australia. Biological Conservation 143, 60-71.

Page 170: A thesis submitted for the fulfilment for the requirements

148

Breuste, J. H., (2004) Decision making, planning and design for the conservation of

indigenous vegetation within urban development. Landscape and Urban Planning 68, 439-

452.

Bringham, M., Debus, S., Geiser, F., (1998) Cavity selection for roosting, and roosting

ecology of forest-dwelling Australian Owlet-nightjars, Aegotheles cristatus. Australian

Journal of Ecology 23, 424-429.

Brooker, M. I. H., Kleinig, D. A., (2004) Field Guide to Eucalypts: Northern Australia. 2 nd

Edn. Bloomings Books, Melbourne

Brookhouse, M., (2006) Eucalypt dendrochronology: past, present and potential. Australian

Journal of Botany 54, 435-449.

Brøstrom, G. (2012) glmmML: Generalised Linea Models with random intercept 0.82-1

Brotons, L., Mönkkönen, M., Martin, L. J., (2003) Are fragments islands? Landscape context

and density-area relationships in boreal forest birds. American Naturalist 162, 343-357.

Brown, C., Grant, M., (2005) Biodiversity and human health: What role for nature in healthy

urban planning? Built Environment 31, 326-338.

Bull, E. L., Partridge, A. D., (1986) Methods of killing trees for use by cavity nesters.

Wildlife Society Bulletin 14, 142-146.

Bunnell, F., (1999) Foreword. Let's kill a panchestron: giving fragmentation meaning In:

Forest Wildlife and Fragmentation. (Eds) J. Rochelle, L. Lehmann and J. Wisniewski, Brill,

Leiden, Germany, pp. vii-xiii.

Bureau of Meteorology, (2009) 12 month maximum and minimum temperatures of Australia

Available at http://www.bom.gov.au/cgibin/silo/temp_maps.cgi?variable=maxhigh&area=nat&period=12month&time=latest

(Accessed on 17/6/2009)

Burgman, M., Ferson, S., Akcakaya, H., (1993) Risk Assessment in Conservation Biology.

In: Population and Community Biology (Eds) D. Angelis and R. Kitching, Chapman & Hall,

London,

Page 171: A thesis submitted for the fulfilment for the requirements

149

Burnham, K. P., Anderson, D. R., (2002) Model selection and multimodel inference: A

practical information - theoretical approach. Springer, New York

Burrows, G. E., (2002) Epicormic strand structure in Angophora , Eucalyptus and

Lophostemon (Myrtaceae) – implications for fire resistance and recovery. New Phytologist

153, 111-131.

Byrne, J., Sipe, N., (2010) Green and open space planning for urban consolidation: A review

of the literature in best practice. Urban Research Program.

Byrne, J., Sipe, N., Searle, G., (2010) Green around the gills? The challenge of density for

urban greenspace in SEQ. Australian Planner 47, 162-177.

Calver, M., Wardell-Johnson, 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. In: The Conservation of Australia's Forest Fauna. (Ed.)

D. Lunney, Royal Zoological Society of New South Wales, Sydney,

Cameron, M., (2006) Nesting habitat of the Glossy Black-cockatoo in central New South

Wales. Biological Conservation 127, 402-410.

Cardilini, A. P. A., Weston, M. A., Nimmo, D. G., Dann, P., Sherman, C. D. H., (2013)

Surviving in sprawling suburbs: Suburban environments represent high quality breeding

habitat for a widespread shorebird. Landscape and Urban Planning 115, 72-80.

Catling, P. C., Burt, R. J., Kooyman, R., (1997) A comparison of techniques used in a survey

of the ground-dwelling and arboreal mammals in forests in north-eastern New South Wales.

Wildlife Research 24, 417-432.

Catterall, C. P., (2004) Birds, garden plants and suburban bush-lots: where good intentions

meet unexpected outcomes. In: Urban wildlife more than meets the eye. (Eds) D. Lunney and

S. Burgin, Royal Zoological Society of New South Wales, Sydney, pp. 21-31.

Catterall, C. P., (2009) Responses of faunal assemblages to urbanisation: global research

paradigms and an avian case study. In: Ecology of Cities and Towns: A Comparative

Approach. (Eds) M. J. McDonnell, J. Breuste and A. K. Hahs, Cambridge University Press,

Cambridge,

Page 172: A thesis submitted for the fulfilment for the requirements

150

Catterall, C. P., Cousin, J. A., Piper, S., Johnson, G., (2010) Long-term dynamics of bird

diversity in forest and suburb: decay, turnover or homogenization? Diversity and

Distributions 16, 559-570.

Catterall, C. P., Kingston, M., (1993) Remnant bushland of Southeast Queensland in the

1990's. Griffith University Institute of Applied Environmental Research, Brisbane.

Caughley, G., Sinclair, A. R. E., (1994) Wildlife Ecology and Management. Blackwell

Science, London

Centre for Plant Biodiversity, (2006) Euclid: Eucalypts of Australia. 3rd Edn. CSIRO,

Melbourne

Clark, K. R., Ainsworth, M., (1993) A method of linking multivariate community structure to

envinonmental variables. Marine Ecology Progress Series 92, 205-219.

Clarke, K. R., Warwick, R. M. (2001b) Primer -E.

Clemann, N., Long, K., Skurrie, D., Dzuris, J., (2005) A trapping survey of small, ground-

dwelling vertebrates in the Little Desert National Park, Victoria. Australian Zoologist 33,

119-126.

Cockburn, A., Lazenby-Cohen, K., (1992) Use of nest trees by Antechinus stuartii, a

semelparous lekking marsupial. Journal of Zoology 226, 657-680.

Cogger, H., Ford, H., Johnson, C., Holman, J., Butler, D., (2003) Impacts of Land Clearing

on Australian Wildlife in Queensland. World Wide Fund for Nature: Australia 4.

Commonwealth of Australia, (1992) National Forest Policy Statement 1992: A New Focus

For Australia's Forests. In. ' 2nd Edn. Commonwealth of Australia

Comport, S., Ward, S., Foley, W., (1996) Home ranges, time budgets and food-tree use in a

high-density tropical population of greater gliders, Petauroides volans minor

(Pseudocheiridae: Marsupialia). Wildlife Research 23, 401-419.

Connor, R. N., Dickson, J. G., Locke, B. A., (1981) Herbicide killed trees infected by fungi.

Potential for cavity sites for woodpeckers. Wildlife Society Bulletin 9, 308-310.

Cook, W. M., Lane, K. T., Foster, B. L., Holt, R. D., (2002) Island theory, matrix effects and

species richness patterns in habitat fragments. Ecology Letters 5, 619-623.

Page 173: A thesis submitted for the fulfilment for the requirements

151

Cornelius, J., Hermy, M., (2004) Biodiversity relationships in urban and suburban parks in

Flanders. Landscape and Urban Planning 69, 385-401.

Cowling, R. M., Bond, W. J., (1991) How small can reserves be? An empirical approach in

Cape Fynbos, South Africa. Biological Conservation 58, 243-256.

Crooks, K., Suarez, A., D., B., (2004) Avian assemblages along a gradient of urbanisation in

a highly fragmented landscape. Biological Conservation 115, 451-462.

Csilléry, K., Seignobosc, M., Laofond, V., Kunster. G., Courbaud, B., (2013) Estimating

long-term tree mortaility rate time series by comgining data from periodic inventories and

harvest reports in a Bayesian state-space model. Forest Ecology and Management 292, 64-

74.

Cundill, G., Fabricius, C., (2010) Monitoring the governance dimension of natural resource

co-management. Ecology and Society 15, 15.

Daily, G., (1995) Restoring value to the world's degraded lands. Science 269, 350-355.

Daily, G. C., (2001) Ecological forecasts. Nature, 245.

Damschen, E. I., Haddad, N. M., Orrock, J. L., Tewksbury, J. J., Levey, D. J., (2006)

Corridors Increase Plant Species Richness at Large Scales. Science 313, 1284-1286.

Davis, A., Major, R. E., Taylor, C. E., (2013) Housing shortages in urban regions:

Aggressive interactions at tree hollows in forest patches. PLoS One 8, e59332.

Davison, A., Ridder, B., (2006) Turbulent times for urban nature: conserving and re-

inventing nature in Australian cities. Australian Zoologist 33, 306-314.

De Wet, A., Richardson, J., Olympia, C., (1998) Interactions of land-use history and current

ecology in a recovering "urban wildland". Urban Ecosystems 2, 237-262.

Devictor, V., Godet, L., Julliard, R., Couvet, D., Jiguet, F., (2007) Can common species

benefit from protected areas? Biological Conservation 139, 29-36.

Didham, R., Tylianakis, J., Gemmell, N., Rand, T., (2007a) Interactive effects of habitat

modification and species invasion on native species decline. Trends in Ecology and Evolution

22, 489-496.

Page 174: A thesis submitted for the fulfilment for the requirements

152

Didham, R. K., Tylianakis, J. M., Gemmell, N. J., Rand, T. A., Ewers, R. M., (2007b)

Interactive effects of habitat modification and species invasion on native species decline.

Trends in Ecology & Evolution 22, 489-496.

Dixon, M. D., (2012) Relationship between land cover and insectiverous bat activity in an

urban landscape. Urban Ecosystems 15, 683-695.

Dovers, S., (2003) Reflecting on Three Decades: A synthesis. In: Managing Australia's

Environment. (Eds) S. Dovers and S. Wild River, The Federation Press, Sydney, pp. 515-

535.

Driscoll, D., Milkovits, G., Freudenberger, D., (2000) Impact and use of Firewood in

Australia. CSIRO

du Plessis, M., (1995) The effects of fuel-wood removal on the diversity of some cavity-using

birds and mammals in South Africa. Biological Conservation 74, 77-82.

Duffy, A. M., Lumseden, L. F., Caddle, C. R., Chick, R. R., Newell, G. R., (2000) The

efficacy of Anabat ultrasonic detectors and harp traps for surveying microchiropterans in

south-eastern Australia. Acta Chiropterologica, 127-144.

Durant, R., Luck, G. W., Matthews, A., (2009) Nest-box use by arboreal mammals in a peri-

urban landscape. Wildlife Research 36, 565-573.

Elliott, G. P., Wilson, P. R., Taylor, R. H., Beggs, J. R., (2010) Declines in common,

widespread native birds in a mature temperate forest. Biological Conservation 143, 2119-

2126.

Evans, K. L., (2010) Individual species and urbanisation. In: Urban Ecology. (Ed.) K. J.

Gaston, Cambridgy University Press, London, pp. 53-87.

Evans, K. L., Hatchwell, B. J., Parnell, M., Gaston, K. J., (2010) A conceptual framework for

the colonisation of urban areas: the blackbird, Turdus merula as a case study. Biological

Reviews 85, 643-667.

Ewers, R., Didham, R., (2006a) Confounding factors in the detection of species responses to

habitat fragmentation. Biological Review 81, 117-142.

Page 175: A thesis submitted for the fulfilment for the requirements

153

Ewers, R., Didham, R., (2006b) Continuous response functions for quantifying the strength of

edge effects. Journal of Applied Ecology 43, 527-536.

Ewers, R., Thorpe, S., Didham, R., (2007) Synergistic interactions between edge and area

effects in a heavily fragmented landscape. Ecology 88, 96-106.

Eyal, S., Pairge, S. W., Faeth, S. H., McIntyre, N. E., Hope, D., (2006) From patterns to

emerging processes in mechanistic urban ecology. Trends in Ecology & Evolution 4, 186-

191.

Eyre, T. J., (2004) Distribution and conservation status of the possums and gliders of

southern Queensland In: The Biology of Australian Possums and Gliders. (Eds) R. L.

Goldingay and S. M. Jackson, Surrey Beatty, Sydney, pp. 1-25.

Eyre, T. J., (2005) Hollow-bearing trees in large glider habitat in south-east Queensland,

Australia: Abundance, spatial distribution and management. Pacific Conservation Biology 11,

23-37.

Eyre, T. J., (2006) Regional habitat selection of large gliding possums at forest stand and

landscape scales in southern Queensland, Australia: I. Greater glider Petauroides volans.

Forest Ecology and Management 235, 270-282.

Eyre, T. J., (2007) Regional habitat selection of large gliding possums at forest stand and

landscape scales in southern Queensland, Australia: II. Yellow-bellied glider Petaurus

australis. Forest Ecology and Management 239, 136-149.

Eyre, T. J., Butler, D. W., Kelly, A. L., Wang, J., (2010) Effects of forest management on

structural features important for biodiversity in mixed-age hardwood forests in Australia's

subtropics. Forest Ecology and Management 259, 534-546.

Eyre, T. J., Smith, A. P., (1997) Floristic and structural habitat preferences of yellow-bellied

gliders (Petaurus australis) and selective logging impacts in southeast Queensland, Australia.

Forest Ecology and Management 98, 281-295.

Fahrig, L., (2002) Effect of habitat fragmentation on the extinction threshold: A synthesis.

Ecological Applications 12, 346-353.

Fahrig, L., Pedlar, J. H., Pope, S. E., Taylor, P. D., Wegner, J. F., (1995) Effect of road traffic

on amphibian density. Biological Conservation 73, 177-182.

Page 176: A thesis submitted for the fulfilment for the requirements

154

Fallding, M., (2004) Planning for biodiversity: Can we do better? Australian Planner 41, 45-

50.

Felton, A., Knight, E., Wood, J., Zammit, C., Lindenmayer, D., (2010) A meta-analysis of

fauna and flora species richness and abundance in plantations and pasture lands. Biological

Conservation 143, 545-555.

Fitzgerald, M., Shine, R., Lemckert, F., (2003) A reluctant heliotherm: thermal ecology of the

arboreal snake Hoplocephalus stephensii (Elapidae) in dense forest. Journal of Thermal

Biology 28, 515-524.

Fitzsimmons, J., Westcott, G., (2001) The role and contribution of private land in Victoria to

biological conservation and the protected area system. Australian Journal of Environmental

Management 9, 142-187.

Fleay, D., (1947) Gliders of the Gum Trees: The Most Beautiful and Enchanting Australian

Marsupials. Bread and Cheese Club, Melbourne.

Flemming, J., (1998) Interface and Interaction Design. Web Navigation: Designing the User

Experience, 1st Edn. Chapter 5. O'Reilly on line Catalogue.

Flint, C., D., P., Beaver, D., (2004) The good, the bad and the ugly: science, process and

politics in forestry reform and the implications for conservation of forest fauna in north-east

New South Wales. In: Conservation of Australia's Forest Fauna. 2nd Edn. (Ed.) D. Lunney,

Royal Zoological Society of New South Wales, Sydney, pp. 222-256.

Florence, R. G., (1996) Ecology and Silviculture of Eucalypt forests. CSIRO, Canberra

Flynn, E. M., Jones, S. M., Jones, M. E., Jordan, G. J., Munks, S. A., (2011) Characteristics

of mammal communities in Tasmanian forests: exploring the influence of forest type and

disturbance history. Wildlife Research 38, 13-29.

Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., Chapin, F.

S., Coe, M. T., Daily, G. C., Gibbs, H. K., Helkowski, J. H., Holloway, T., Howard, E. A.,

Kucharik, C. J., Monfreda, C., Patz, J. A., Prentice, I. C., Ramankutty, N., Snyder, P. K.,

(2005) Global consequences of land use. Science 309, 570-574.

Page 177: A thesis submitted for the fulfilment for the requirements

155

Forrester, D. I., Smith, R. G. B., (2010) Faster growth of Eucalyptus grandis and Eucalyptus

pilularis in mixed species stands than monocultures. Forest Ecology and Management 286,

81-86.

Fox, C., Hamilton, F., Ades, P., (2008) Models of tree-level hollow incidence in Victorian

State forests. Forest Ecology and Management 255, 2846-2857.

Fox, J., (1984) Firewood consumption in a Nepali village. Environmental Management 8,

243-249.

Fox, J. C., Hamilton, F., Occhipinti, S., (2009) Tree hollow incidence in Victorian state

forests. Australian Forestry 72, 39-48.

Franklin, J., Shugart, H., Harmon, M., (1987) Tree death as an ecological process. BioScience

37, 550-556.

Frith, H. J., (Ed.) (1976) Birds in the Australian High Country A.H. and A.W. Reed, Sydney

Garden, J., McAalpine, C., Possingham, H., Jones, D., (2007a) Using multiple survey

methods to detect terrestrial reptiles and mammals: what are the most successful and cost-

efficient combinations? Wildlife Research 34, 218-227.

Garden, J., McAlpine, C., Peterson, A., Jones, D., Possingham, H., (2006) Review of the

ecology of Australian urban fauna: A focus on spatially explicit processes. Austral Ecology

31, 126-148.

Garden, J., McAlpine, C., Possingham, H., (2010) Multi-scaled habitat considerations for

conserving urban biodiversity: native reptiles and small mammals in Brisbane, Australia.

Landscape Ecology 25, 1013-1028.

Garden, J., McAlpine, C., Possingham, H., Jones, D., (2007b) Habitat structure is more

important than vegetation composition for local-level management of native terrestrial reptile

and small mammal species living in urban remnants: A case study from Brisbane, Australia.

Austral Ecology 32, 669-685.

Gaston, K. J., (Ed.) (2010a) Urbanisation Urban Ecology Cambridge University Press,

London

Gaston, K. J., (2010b) Valuing common species. Science & Culture 327, 154-155.

Page 178: A thesis submitted for the fulfilment for the requirements

156

Gaston, K. J., Davies, Z. G., Edmondson, J. L., (2010) Urban environments and ecosystem

functions. In: Urban Ecology. (Ed.) K. J. Gaston, Cambridge University Press, London, pp.

35-52.

Gaston, K. J., Fuller, R. A., (2007) Commonness, population depletion and conservation

biology. Trends in Ecology & Evolution 23, 14-19.

Germaine, S. S., Wakeling, B. F., (2001) Lizard species distributions and habitat occupation

along an urban gradient in Tucson, Arizona, USA. Biological Conservation 97, 229-237.

Gibbons, P., Lindenmayer, D., (1996) Issues associated with the retention of hollow-bearing

trees within eucalypt forests managed for wood production. Forest Ecology and Management

83, 245-279.

Gibbons, P., Lindenmayer, D., (2002) Tree hollows and wildlife conservation in Australia.

CSIRO Publishing, Collingwood, Victoria

Gibbons, P., Lindenmayer, D., Barry, S. C., Tanton, M. T., (2000a) The effects of slash

burning on the mortality and collapse of trees retained on logged sites in south-eastern

Australia. Forest Ecology and Management 139, 51-61.

Gibbons, P., Lindenmayer, D., Barry, S. C., Tanton, M. T., (2000b) The formation of hollows

in eucalypt forests from temperate forests. Pacific Conservation 6, 218-228.

Gibbons, P., Lindenmayer, D., Barry, S. C., Tanton, M. T., (2002) Hollow selection by

vertebrate fauna in forests of south eastern Australia and implications for forest management.

Biological Conservation 103, 1-12.

Godefroid, S., Koedam, N., (2003) How important are large vs. small forest patches for the

conservation of the woodland flora in an urban context? Global Ecology and Biogeography

12, 287-298.

Gold Coast City Council, (2007) Our Living City Report: A description of the Gold Coast

area. Gold Coast City Council, Gold Coast. 14-22.

Gold Coast City Council, (2009) Gold Coast City Council Nature Conservation Strategy

2009. Gold Coast.

Page 179: A thesis submitted for the fulfilment for the requirements

157

Goldingay, R. L., (2009) Characteristics of tree hollows used by Australian birds and bats.

Wildlife Research 36, 394-409.

Goldingay, R. L., (2011) Characteristics of tree hollows used by Australian arboreal and

scansorial mammals. Australian Journal of Zoology 59, 277-294.

Goldingay, R. L., Grimson, M. J., Smith, G. C., (2007) Do feathertail gliders show a

preference for nest box design? Wildlife Research 34, 484-490.

Goldingay, R. L., Sharpe, D. J., (2004a) How do we conserve the squirrel glider in Brisbane's

urban matrix? In: Conservation of Australia's Forest Fauna. (Ed.) D. Lunney, Royal

Zoological Society of New South Wales, Sydney, pp. 663-675.

Goldingay, R. L., Sharpe, D. J., (2004b) How effective is spotlighitng for detecting the

squirrel glider? Wildlife Research 31, 443-449.

Goldingay, R. L., Sharpe, D. J., (2004c) Spotlights versus nestboxes for detecting feathertail

gliders in northeast New South Wales. In: The Biology of Australian Possums and Gliders.

(Eds) R. L. Goldingay and S. M. Jackson, Surrey Beatty, Sydney,

Goldingay, R. L., Stevens, J. R., (2009) Use of artificial tree hollows by Australian birds and

bats. Wildlife Research 36, 81-97.

Gordon, A., Simondson, D., White, M., Moilanen, A., Bekessy, S. A., (2009) Integrating

conservation planning and landuse planning in urban landscapes. Landscape and Urban

Planning 91, 183-194.

Graymore, M. L. M., Sipe, N. G., Rickson, R. E., (2008) Regional sustainability, how useful

are current tools of sustainability assessment at a regional scale. Ecology and Economics 13,

391-414.

Guntenspergen, G. R., Levenson, J., (1997) Understory plant species composition in remnant

stands along an urban-to-rural land-use gradient. Urban Ecosystems 1, 155-169.

Gurran, N., (2008) The turning tide: amenity migration in coastal Australia. Integrated

Planning Study 13, 391-414.

Hahs, A. K., McDonnell, M. J., (2006) Selecting independent measures to quantify

Melbourne's urban–rural gradient. Landscape and Urban Planning 78, 435-448.

Page 180: A thesis submitted for the fulfilment for the requirements

158

Hahs, A. K., McDonnell, M. J., (2007) Composition of the plant community in remnant

patches of grassy woodland along an urban-rural gradient in Melbourne, Australia. Urban

Ecosystems 10, 355-355.

Harley, D. K. P., Spring, D. A., (2003) Reply to the comment by Lindenmayer et al. on

"Economics of a nest-box program for the conservation of an endangered species: a re-

appraisal". Canadian Journal of Forest Research 33, 752-753.

Harper, M., McCarthy, M., van der Ree, R., (2005a) The abundance of hollow-bearing trees

in urban dry sclerophyll forest and the effect of wind on hollow development. Biological

Conservation 122, 181-192.

Harper, M., McCarthy, M., van der Ree, R., (2008) Resources at the landscape scale

influence possum abundance. Austral Ecology 33, 243-252.

Harper, M., McCarthy, M., van der Ree, R., Fox, J., (2004) Overcoming bias in ground-based

surveys of hollow-bearing trees using double-sampling. Forest Ecology and Management

190, 291-300.

Harper, M. J., (2005) Home range and den use of common brushtail possums, Trichosurus

vulpecula in urban forest patches. Wildlife Research 32, 681-687.

Harper, M. J., McCarthy, M. A., Van der Ree, R., (2005b) The use of nest boxes in urban

natural vegetation remnants by vertebrate fauna. Wildlife Research 32, 509-516.

Harrisson, K. A., Pavlova, A., Amos, J. N., Takeuchi, N., Lill, A., Radford, J. Q., Sunnucks,

P., (2013) Disrupted fine-scale population processes in fragmented landscapes despite large-

scale genetic connectivity for a widespread and common cooperative breeder: the superb

fairy-wren (Malurus cyaneus). Journal of Animal Ecology 82, 322-333.

Hartman, H., (2010). Will a 385 million year-struggle for light become a struggle for water

and for carbon? How trees may cope with more frequent climate change-type drought events.

Global Change Biology 17, 642-655.

Haslem, A., Avitable, S. C., Raylor, R. S., Kelly, L. T., Watson, S. J., Nimmo, D. G., Kenny,

S. A., Callister, K. E., Spence-Bailey, L. M., Bennett, A. F., Clarke, M. F., (2012) Time-since

fire and inter-fire interval influence hollow availabilify for fauna in a fire-prone system.

Biological Conservation 152, 212-221.

Page 181: A thesis submitted for the fulfilment for the requirements

159

Hatton, T. J., Cork, S., Harper, P., Robert, J., Kanowski, P., MacKay, R., McKenzie, N.,

Ward, T., (2011) State of the Environment 2011. E. Australian Government Minister for

Sustainability, Water, Population and Communities Canberra. Australian Government.

Herbohn, K. F., Harrison, S. R., Herbohn, J. L., (2000) Lessons from small-scale forestry

initiatives in Australia: The effective integration of environmental and commercial values.

Forest Ecology and Management 128, 227-240.

Higgins, P. J., (Ed.) (1999) Handbook of Australian, New Zealand and Antarctic Birds

Parrots to Dollarbird Oxford University Press, Melbourne

Holland, G., Bennett, A., (2009) Differing responses to landscape change: implications for

small mammal assemblages in forest fragments. Biodiversity and Conservation 18, 2997-

3016.

Holloway, G. L., Caspersen, J. P., Vanderwel, M. C., Naylor, B. J., (2007) Cavity tree

occurrence in hardwood forests of central Ontario. Forest Ecology and Management 239,

191-199.

Hooper, D. U., Vitousek, P. M., (1997) The effects of plant composition and diversity on

ecosystem processes. Science 277, 1302-1305.

Hooper, R. J., Sivasithamparam, K., (2005) Characterization of damage and biotic factors

associated with the decline of Eucalyptus wandoo in southwest Western Australia. Canadian

Journal of Forest Research 35, 2589-2602.

Hourigan, C. L., Catterall, C. P., Jones, D., Rhodes, M., (2008) A comparison of the

effectiveness of bat detectors and harp traps for surveying bats in an urban landscape.

Wildlife Research 35, 768-774.

Huebner, D. P., Hurteau, S. R., (2007) An economical wireless cavity-nest viewer. Journal of

Field Ornithology 78, 87-92.

Humphries, J., (1989) Urban planning and vegetation conservation: A case for forward

planning? Urban Planning and Policy 7, 106-116.

IBM. Corporation., (2012) SPSS. 21. Edn. IBM Corporation

Page 182: A thesis submitted for the fulfilment for the requirements

160

Inions, G. B., Tanton, M. T., Davey, S. M., (1989) Effect of fire on the availability of hollows

in trees used by the Common Brushtail Possum, Trichosurus vulpeca (Kerr, 1972), and the

Ringtail Possum, Pseudocheirus peregrinus (Boddaerts, 1785). Wildlife Research 16, 449-

458.

Jacobs, M. R., (1955) Growth Habits of the Eucalypts. Forest Timber Bureau Australia,

Canberra

Janzen, D. H., (1986) The eternal external threat. In: Conservatio biology. The science of

scarcity and diversity. (Ed.) M. E. Soulé Sinauer Associates, Massachusetts,

Jim, C. Y., Liu, H. T., (2001) Species diversity of three major urban forest types in

Guangzhou City, China. Forest Ecology and Management 146, 99-114.

Johnson, J. B., Gates, J. E., Ford, M. W., (2008) Distribution and activity of bats at local and

landscape scales within an urban-rural gradient. Urban Ecosystems 11, 227-242.

Johnstone, R. E., Kirkby, T., Sarti, K., (2013) 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.

Jurskis, V., Bridges, B., de Mar, P., (2003) Fire management in Australia: the lessons of 200

years. In 'Joint Australia and New Zealand Institute of Forestry Conference Proceedings. '

pp. 353 -368. Ministry of Agriculture and Forestry: Queenstown, New Zealand

Kahmen, A., Perner, J., Audorff, V., Weisser, W., Buchmann, N., (2005) Effects of plant

diversity, community composition and environmental parameters on productivity in montane

European grasslands. Oecologica 142, 606-615.

Karr, J. R., (1990) Biological integrity and the goal of environmental legislation: Lessons for

conservation biology. Conservation Biology 4, 244-250.

Kehl, J., Borsboom, A., (1984) Home-range, den tree use and activity patterns in the greater

glider (Petauroides volans) In: Possums and Gliders. (Eds) A. P. Smith and I. D. Hume,

Surrey Beatty & Sons, Sydney, pp. 229-236.

Kendle, T., Forbes, S., (1997) Urban Nature Conservation. E&FN Spon, London

Page 183: A thesis submitted for the fulfilment for the requirements

161

Kirkpatrick, J. B., Davison, A., Harwood, A., (2013) How tree professionals perceive trees

and conflicts about trees in Australia's urban forest. Landscape and Urban Planning 119,

124-130.

Kitching, R. L., Callaghan, C., (1982) The fauna of water-filled tree-holes in box forest in

south-east Queensland. Australian Entomologist 8, 61-70.

Knight, R. L., (1999) Private Lands: The Neglected Geography. Conservation Biology 13,

223-224.

Koch, A. J., (2008) Errors associated with two methods of assessing tree hollow occurrence

and abundance in Eucalyptus oblique forest, Tasmania. Forest Ecology and Management

255, 674-685.

Koch, A. J., Driscoll, D., Kirkpatrick, J. B., (2008a) Estimating the accuracy of tree ageing

methods in mature Eucalyptus obliqua forest, Tasmania. Australian Forestry 71, 147-159.

Koch, A. J., Munks, S. A., Driscoll, D., (2008b) The use of hollow-bearing trees by

vertebrate fauna in wet and dry Eucalypt obliqua forest, Tasmania. Wildlife Research 35,

727-746.

Koch, A. J., Munks, S. A., Driscoll, D., Kirkpatrick, J. B., (2008c) Does hollow occurrence

vary with forest type? A case study in wet and dry Eucalyptus obliqua forest. Forest Ecology

and Management 255, 3938-3951.

Koch, A. J., Munks, S. A., Woehler, E. J., (2009) Hollow-using vertebrate fauna of

Tasmania: distribution, hollow requirements and conservation status. Australian Journal of

Zoology 56, 323-349.

Koh, L. P., Sodhi, N. S., (2004) Importance of reserves, fragments and parks for butterfly

conservation in a tropical urban landscape. Ecological Applications 14, 1695-1708.

Kowald, M., (1996) Historical Overview of the South East Queensland Biogeographic

Region: with particular reference to forested areas. Department of Environment, Brisbane.

Krebs, E. A., (1998) Breeding biology of Crimson rosellas, Platycercus elegans on Black

Mountain, Australian Capital Territory. Australian Journal of Zoology 46, 119-136.

Page 184: A thesis submitted for the fulfilment for the requirements

162

Kurz, W. A., Dymond, C.C., Stinson, G., Rampley, G.J., Neilson, E.T., Carroll, L., Ebata,

T., Safranyik, I., (2008) Mountain pine beetle and forest carbon feedback to climate change.

Nature 452, 987-990.

Lagabrielle, E., Botta, A., Dare, W., David, D., Aubert, S., Fafricius, C., (2010) Modelling

with stakeholders to integrate biodiversity into land-use planning: Lessons learned in

Reunion Island (Western Indian Ocean). Environmental Modelling & Software 25, 1413-

1427.

Laita, A., Mönkkönen, M., Kotiaho, J. S., (2011) Assessing the functional connectivity of

reserve networks in continuously varying nature under the constraints imposed by reality.

Biological Conservation 144, 1297-1298.

Lamb, D., Loyn, R., Smith, A., Wilkinson, G., (1998) Managing habitat trees in Queensland

forests. Queensland Department of Natural Resources, Brisbane. 1-62.

Lane, B., (1999) Regional Forest Agreements: Resolving Resource Conflicts or Managing

Resource Politics? Australian Geographical Studies 37, 142-153.

Law, B., Chidel, M., Britton, A., Brassil, T., (2013) Response of eastern pygmy possums,

Cercartetus nanus, to selective logging in New South Wales: Home range, habitat selection

and den use. Wildlife Research 40, 470-481.

Lazenby-Cohen, K. A. and A. Cockburn (1991) Social and foraging components of the home

range in Antechinus stuartii (Dasyuridae: Marsupialia). Australian Journal of Ecology 16,

301-307.

Laurance, W. F., (1991) Ecological correlates of extinction proneness in Australian tropical

rain forest mammals. Conservation Biology 5, 79-89.

Lee, E., Hau, B., Corlett, R., (2008) Seed rain and natural regeneration in Lophostemon

confertus plantations in Hong Kong, China. New Forests 35, 119-130.

Lewis, D. J., Plantinga, A. J., Wu, J., (2009) Targeting incentives to reduce habitat

fragmentation. American Journal of Agriculture and Economics 91, 1080-1096.

Lindenmayer, D., (2002) Gliders of Australia: A Natural History. University of New South

Wales, Sydney

Page 185: A thesis submitted for the fulfilment for the requirements

163

Lindenmayer, D., Cunningham, R., Donnelly, C., (1997) Decay and collapse of trees with

hollows in eastern Australian forests: Impacts on arboreal marsupials. Ecological

Applications 7, 625-641.

Lindenmayer, D., Cunningham, R., Pope, M., Donnelly, C., Gibbons, P., (2000a) Cavity sizes

and types in Australian eucalypts from wet and dry forest types: A simple rule of thumb for

estimating size and number of cavities. Forest Ecology and Management 137, 139-150.

Lindenmayer, D., Cunningham, R. B., Donnelly, C. F., (1994a) The conservation of arboreal

marsupials in the montane ash forests of the central highlands of Victoria, south-eastern

Australia, VI. The performance of statistical models of the nest tree and habitat requirements

of arboreal marsupials applied to new survey data. Biological Conservation 70, 143-147.

Lindenmayer, D., Cunningham, R. B., Donnelly, C. F., Tanton, M. T., Nix, H. A., (1993) The

abundance and development of cavities in Eucalyptus trees: A case in the montane forests of

Victoria, south-eastern Australia. Forest Ecology and Management 60, 77-104.

Lindenmayer, D., Cunningham, R. B., Donnelly, C. F., Triggs, B. E., Belvedere, M., (1994b)

Factors influencing the occurrence of mammals in retained linear strips (wildlife corridors)

and contiguous stands of montane ash forest in the Central Highlands of Victoria, south-

eastern Australia. Forest Ecology and Management 67, 113-133.

Lindenmayer, D., Cunningham, R. B., Tanton, M. T., Smith, A. P., Nix, H. A., (1990) The

conservation of arboreal marsupials in the Montane ash forests of the central highlands of

Victoria, South-east Australia: I. Factors influencing the occupancy of trees with hollows.

Biological Conservation 54, 111-131.

Lindenmayer, D., Cunningham, R. B., Tanton, M. T., Smith, A. P., Nix, H. A., (1991a)

Characteristics of hollow-bearing trees occupied by arboreal marsupials in the montane ash

forests of the Central Highlands of Victoria, south-east Australia. Forest Ecology and

Management 40, 289-308.

Lindenmayer, D., Cunningham, R. B., Tanton, M. T., Smith, A. P., Nix, H. A., (1991b)

Predicting the abundance of hollow-bearing trees in montane forests of south-eastern

Australia. Journal of Ecology 16, 91-98.

Lindenmayer, D., Fischer, J., (2006) Habitat Fragmentation and Landscape Change. CSIRO,

Melbourne

Page 186: A thesis submitted for the fulfilment for the requirements

164

Lindenmayer, D., McCarthy, M., Parris, K., Pope, M., (2000b) Habitat fragmentation,

landscape context, and mammalian assemblages in south-eastern Australia. Journal of

Mammalogy 81, 787-797.

Lindenmayer, D., Welsh, A., Donnelly, C., Crane, M., Michael, D., Macgregor, C.,

McBurney, L., Montague-Drake, R., 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., Welsh, A., Donnelly, C., Cunningham, R., (1996) Use of nest trees by the

mountain brushtail possum, Trichosurus caninus (Phalangeridae, Marsupialia). 2.

Characteristics of occupied trees. Wildlife Research 23, 531-545.

Lindenmayer, D. B., MacGregor, C. I., Gibbons, P., (2002) Comment-Economics of a nest-

box program for the conservation of an endangered species: A re-appraisal. Canadian

Journal of Forest Research 32, 2244-2247.

Lindenmayer, D. B., Tanton, M. T., Cunningham, R., (1991c) A critique of the use of nest

boxes for the conservation of Leadbeater's possum, Gymnobelideus leadbeateri McCoy.

Wildlife Research 18, 619-624.

Lindenmayer, D. B., Wood, J. T., McBurney, L., MacGregor, C., Youngentob, K., Banks, S.

C., (2011) How to make a common species rare: A case against conservation complacency.

Biological Conservation 144, 1663-1672.

Lindenmayer, D. L., Blanchard, W., McBurney, L., Blair, D., Banks, S. C., Likens, G. E.,

Franklin, J. F., Laurance, W. F., Stein, J., Gibbons, P., (2012) Interacting factors driving a

major loss of large trees with cavitites in a forest ecosystem. PloS One 7, e41864.

Llewellyn, J., (1984) Tree preservation orders in New South Wales: some legal and economic

aspects. Environmental and planning law journal 1, 178.

Lloyd, A., Law, B., Goldingay, R., (2006) Bat activity on riparian zones and upper slopes in

Australian timber production forests and the effectiveness of riparian buffers. Biological

Conservation 129, 207-220.

Longhurst, R., (1996) From Tallebudgera to the Tweed. Gold Coast City Council, Gold

Coast

Page 187: A thesis submitted for the fulfilment for the requirements

165

Loss, S. R., Ruiz, M. O., Brawn, J. D., (2009) Relationships between avian diversity,

neighbourhood age, income, and environmental characteristics of an urban landscape.

Biological Conservation 142, 2578-2585.

Low, T., (2003) The New Nature: Winners and Losers in Wild Australia. Penguin Books,

Camberwell

Loyn, R., Kennedy, S., (2009) Designing old forest for the future: Old trees as habitat for

birds in forests of Mountain Ash, Eucalyptus regnans. Forest Ecology and Management 258,

504-515.

Luck, G. W., (2002) The habitat requirements of the rufous treecreeper, Climacteris rufa. 1.

Preferential habitat use demonstrated at multiple spatial scales. Biological Conservation 105,

383-394.

Luck, G. W., Daily, G. C., (2003) Tropical countryside bird assemblages: Richness,

composition, foraging differ by landscape context. Ecological Applications 13, 235-247.

Luck, G. W., Smallbone, L. T., (2010) Species diversity and urbanisation: patterns, drivers

and implications. In: Urban Ecology. (Ed.) K. J. Gaston, Cambridge University press,

London, pp. 88-119.

Lumsden, L., Bennet, A., Silins, J., Krasna, S., (1994) Fauna in a remnant vegetation-

farmland mosaic: movements, roosts and foraging ecology of bats Department of

Conservation and Natural Resources, Melbourne.

Lumsden, L., Bennett, A., Silins, J., (2002) Location of roosts of the Lesser long-eared ba,t

Nyctophilus geoffroyi and Gould's wattled bat, Chalinolobus gouldii in a fragmented

landscape in south-eastern Australia. Biological Conservation 106, 237-249.

Luneau, J. M. D., Noel, B. L., (2010) A wireless video camera for viewing tree cavities.

Journal of Field Ornithology 81, 176-185.

Lunney, D., (2004) A test of our civilisation: conserving Australia's forest fauna across a

cultural landscape. In: Conservation of Australia's Forests. (Ed.) D. Lunney, Royal

Zoological Society of New South Wales, Sydney, pp. 1-22.

Page 188: A thesis submitted for the fulfilment for the requirements

166

Lunney, D., Burgin, S., (2004a) Urban wildlife management: forming an Australian

synthesis. In: Urban wildlife: More than meets the eye. (Eds) D. Lunney and S. Burgin, Royal

Zoological Society of New South Wales, Sydney, pp. 231-241.

Lunney, D., Burgin, S., (2004b) Urban wildlife management:an emerging discipline. In:

Urban wildlife more than meets the eye. (Eds) D. Lunney and S. Burgin, Royal Zoological

Society of New South Wales, Sydney, pp. 1-7.

MacKenzie, D. I., Nichols, J. D., Royle, J. A., Pollock, K., H., Bailey, L. L., Hines, J. E.,

(2006) Occupancy Estimation and Modeling. Elesevier, London

Mackowski, C. M., (1984) The ontology of hollows in blackbutt, Eucalyptus pilularis and its

relevance to the management of forests for possums, gliders and timber. In: Possums and

Gliders. (Eds) A. P. Smith and I. D. Hume, Surrey Beatty & Sons, Sydney, pp. 553-567.

Manning, A. D., Fischer, J., Lindenmayer, D. B., (2006) Scattered trees are keystone

structures - implications for conservation. Biological Conservation 132, 311-321.

Manning, A. D., Gibbons, P., Fischer, J., Oliver, D. L., Lindenmayer, D. B., (2012) Hollow

futures? Tree decline, lag effects and hollow-dependent species. Animal Conservation.

Mansergh, I., Husley, L., (1985) Gould's wattle bat as a food item for the lace monitor.

Victorian Naturalist 103, 93.

Marzluff, J., Rodewald, A. D., (2008) Conserving biodiversity in urbanising area:

Nontraditional views from a birds perspective. Cities and the Environment 1, 1-27.

Matlack, G. R., (1993) Sociological edge effects: Spatial distribution of human impact in

suburban forest fragments. Environmental Management 17, 829-835.

Mawson, P. R., 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.

Mazurek, M. J., Zielinski, W. J., (2004) Individual legacy trees influence vertebrate wildlife

diversity in commercial forests. Forest Ecology and Management 193, 321-334.

McAlpine, C., Fensham, R. J., Temple-Smith, D. E., (2002) Biodiversity Conservation and

Vegetation Clearing in Queensland: Principles and Thresholds. Rangeland Journal 24, 36-55.

Page 189: A thesis submitted for the fulfilment for the requirements

167

McAlpine, C., Spies, T., Norman, P., Peterson, A., (2007) Conserving forest biodiversity

across multiple land ownerships: Lessons from the Northwest Forest Plan and the Southeast

Queensland regional forests agreement (Australia). Biological Conservation 134, 580-592.

McCarthy, M., Possingham, H., (2006) Active adaptive management for conservation.

Conservation Biology 21, 956-963.

McDonald, I., (1998) Ecologically sustainable development and Local Agenda 21. Significant

Environmental Speeches 7, 18-20.

McDonnell, M., (2007) Restoring and managing biodiversity in an urbanising world filled

with tensions. Ecological Management & Restoration 8, 83-84.

McDonnell, M., Pickett, S., Groffman, P., Bohlen, P., Pouyat, R., Zipperer, W., Parmelee, R.,

Carreiro, M., Medley, K., (1997) Ecosystem processes along an urban-to-rural gradient.

Urban Ecosystems 1, 21-36.

McDonnell, M. J., Pickett, S. T. A., Pouat, R. V., (1993) The application of the ecological

gradient paradigm to the study of urban effects. In: Humans as componentes of ecosystems:

the ecology of subtle human effects and populated areas. (Eds) M. J. McDonnell and S. T. A.

Pickett, Springer-Verlag, New York, pp. 175-189.

McElhinny, C., Gibbons, P., Brack, C., Bauhus, J., (2005) Forest and woodland stand

structural complexity: Its definition and measurement. Forest Ecology and Management 218,

1-24.

McGarigal, K., Cushman, S. A., (2002) Comparative Evaluation of Experimental Approaches

to the Study of Habitat Fragmentation Effects. Ecological Applications 12, 335-345.

McKinney, M. L., (2002) Urbanisation, Biodiversity, and Conservation. BioScience 52, 883-

890.

McKinney, M. L., (2006) Urbanisation as a major cause of biotic homogenization. Biological

Conservation 127, 247-260.

McKinney, M. L., (2008) Effects of urbanisation on species richness: a review of plants and

animals. Urban Ecosystems 11, 161-176.

Page 190: A thesis submitted for the fulfilment for the requirements

168

McWilliam, W., Eagles, P., Seasons, M., Brown, R., (2010) The housing forestry interface:

Testing structural approaches for protecting suburban natural systems following

development. Urban Forestry and Urban Greening 19, 149-159.

Meek, P., (2004) Compromisng science for regulatory compliance. In: Conserving Australia's

Forest Fauna. (Ed.) D. Lunney, Royal Zoological Society of New South Wales, Mosman,

NSW, Australia, pp. 256-269.

Menkhorst, P., Knight, F., (2001) A field guide to the mammals of Australia. Oxford

University Press, South Melbourne

Menkhorst, P. A., (1984) The use of nest boxes by forest vertebrates in Gippsland:

preference and demand. Australian Wildlife Research 11, 225-264.

Mercer, D., Jotkowitz, B., (2000) Local Agenda 21 and barriers to sustainability at the local

government level in Victoria, Australia. Australian Geographer 31, 163-181.

Mills, D. J., Harris, B., Claridge, A. W., Barry, S. C., (2002) Efficacy of hair-sampling

techniques for the detection of medium-sized terrestrial mammals. A comparison between

hair-funnels, hair-tubes and indirect signs. Wildlife Research 29, 379-387.

Mobbs, C., (2003) National Forest Policy and Regional Forest Agreements. In: Managing

Australia's Environment. (Eds) S. Dovers and S. Wild River, The Federation Press, Sydney,

pp. 90-114.

Moffatt, S. F., McLachlan, S. M., Kenkel, N. C., (2004) Impacts of land use on riparian forest

along an urban-rural gradient in southern Manitoba. Plant Ecology 174, 119-135.

Moloney, D., Wormington, K., DeStefano, S., (2002) Stag retention and use by arboreal

marsupials in Eucalypt forests of southeast Queensland, Australia: Implications for

management. USDA Forest Service, United States of America.

Monterrubio-Rico, T. C., Escalante-Pliego, P., (2006) Richness, distribution and conservation

status of cavity nesting birds in Mexico. Biological Conservation 128, 67-78.

Morgan, J. W., Farmilo, B. J., (2012) Community (re)organisation in an experimentally

fragmented forest landscape: insights from occupancy–scale patterns of common plant

species. Journal of Vegetation Science 23, 962-969.

Page 191: A thesis submitted for the fulfilment for the requirements

169

Morrison, R. G., (1978) An observation of ground nesting in sugar gliders. South Australian

Naturalist 52, 64.

Munks, S. A., (2009) From guiding principles for the conservation of forest biodiversity to

on-ground practice: Lessons from tree hollow management in Tasmania. Forest Ecology and

Management 258, 516-524.

Murphy, S., Legge, S., Heihnsohn, R., (2003) The breeding biology of Palm Cockatoos,

Probisciger aterimus: a case of a slow life history. The Journal of Zoology 261, 327-339.

Musselwhite, G., Herath, G., (2005) Australia's regional forest agreement process: analysis of

the potential an problems. Forest Policy and Economics 7, 579-588.

Musselwhite, G., Herath, G., (2007) Chaos theory and assessment of forest stakeholder

attitudes towards Australian forest policy. Forest Policy and Economics 9, 947-964.

Nelson, J. L., Morris, B. J., (1994) Nesting requirements of the Yellow-tailed Black-

cockatoo, Calyptorynchus funereus in Eucalyptus regnans forest and implications for forest

management. Wildlife Research 21, 267-278.

Newton, I., (1994) The role of nest sites in limiting the number of hole-nesting birds: a

review. Biological Conservation 70, 265-276.

Niemelä, J., (1999) Ecology and urban planning. Biodiversity & Conservation 8, 119-131.

Norman, P., Smith, G., McAlpine, C., Borsboom, A., (2004) South-east Queensland forests

agreement: conservaion outcomes for forest fauna. In: Conserving Austrlaia's Forest Fauna.

(Ed.) D. Lunney, Zoological Society of New South Wales, Mosman, NSW, Australia, pp.

208-221.

O'Hara, K. L., (1998) Silviculture for structural diversity: a new look at multi-aged systems.

Journal of Forestry 96, 4-10.

Ogden, C. M., (2009) Hollow bearing trees in Karawatha Forest Park in south-east

Queensland: their distribution abundance and occupancy. Honours Thesis, Griffith

University, Gold Coast.

Page 192: A thesis submitted for the fulfilment for the requirements

170

Pausas, J. G., Braithwaite, L. W., Austin, M. P., (1995) Modelling habitat quality for arboreal

marsupials in the South Coastal forests of New South Wales, Australia. Forest Ecology and

Management 78, 39-49.

Pearman, P. B., Weber, D., (2007) Common species determine richness patterns in

biodiversity indicator taxa. Biological Conservation 138, 109-119.

Peng, C., Ma, Z., Lei, Z., Zhu, Q., Chen, H., Wang, W., Liu, S., Fang, W., Zhou, X., (2011)

A drought-induced pervasive increase in tree mortality across Canada's boreal forests. Nature

Climate Change 1, 467-472.

Pennay, M., B, L., Reinhold, L., (2004) Bat calls of New South Wales: Region based guide to

the echolocation calls of Microchiropteran bats. Hurstville

Petit, R. J., Hampe, A., (2006) Some evolutionary consequences of being a tree." Annual

Review of Ecology and Systematics 37, 187-214.

Pickering, C., Castley, J. G., Hill, W., Newsome, D., (2010) Environmental, safety and

management issues of unauthorised trail technical features for mountain bicycling.

Landscape and Urban Planning 97, 58-67.

Pickett, S. T. A., Cadenasso, M. L., Grove, J. M., Nilon, C. H., Pouyat, R. V., Zipperer, W.

C., Costanza, R., (2001) Urban Ecological Systems: Linking Terrestrial Ecological, Physical,

and Socioeconomic Components of Metropolitan Areas. Annual Review of Ecology and

Systematics 32, 127-157.

Pillsbury, F. C., Miller, J. R., (2008) Habitat and landscape characteristics underlying anuran

community structure along an urban-rural gradient. Ecological Applications 18, 1107-1108.

Pirnat, J., (2000) Conservation and management of forest patches and corridors in suburban

landscapes. Landscape and Urban Planning 52, 135-143.

Pitman, N. C. A., Del, M., Azeldegul, C. L., Salas, K., Vigo, G. T., Lutz, D. A., (2007)

Written accounts of an Amazonian landscape over the last 450 years. Conservation Biology

21, 253-262.

Politi, N., Hunter, M., (2009) Nest selection by cavity-nesting birds in subtropical montane

forests of the Andes: Implications for sustainable forest management. Biotropica 41, 354-

360.

Page 193: A thesis submitted for the fulfilment for the requirements

171

Poonswad, P., Sukkasem, C., Phataramata, S., Hayeemuida, S., Plongmai, K., Chuailua, P.,

Thiensongrusame, P., Jirawatkavi, N., (2005) Comparison of cavity modification and

community involvement as strategies for hornbill conservation in Thailand. Biological

Conservation 122, 385-393.

Powell, J., (1998a) People and trees: A thematic history of south east Queensland with

particular reference to forested areas, 1823-1997. Queensland and CRA/RFA Steering

Committee, Brisbane.

Powell, J., (1998b) Travel routes, forest towns and settlements. Department of Natural

Resources, Brisbane.

Prado-Lorenzo, J. M., Sanchez-Garcia, I. M., (2007) The effect of participation in the

develoment of Local Abenda 21 in the European Union. In 'International Conference of

Territorial Intelligence. ' pp. Pp. 523-546: Huelva

Pressey, R. L., (1994) Ad hoc reservations: Forward or backward steps in developing

representative reserve systems? Conservation Biology 8, 1994.

Prest, J., (2004) The forgotten forests: The regulation of forestry on private land in New

South Wales 1997-2002. In: The Conservation of Australia's Forests. 2nd Edn. (Ed.) D.

Lunney, Royal Zoological Society of New South Wales, Mosman, pp. 297-330.

Purcell, K., (1997) Use of a fiberscope for examining cavity nests. Journal of Field

Ornithology 68, 283-286.

Queensland Government, (1992) Nature Conservation Act 1992. Brisbane. Queensland

Government.

Queensland Government, (1996) Code of Practice for Native Forest Timber Production. In. '

Ed. Department of Primary Industries). Queensland Government: Brisbane

Queensland Government, (2007) Code of practice for native forest timber production on State

lands. In. ' Environmental Protection Agency: Brisbane

Queensland Government, (2008) Beenleigh 9542: Historical aerial images over 1:100000

map area from 1955-2002. Brisbane. Natural Resources and Water.

Page 194: A thesis submitted for the fulfilment for the requirements

172

Queensland Government, (2011) Murwillumbah 9541: Historical aerial images over

1:100000. Brisbane. Natural Resources and Water.

Qureshi, S., Haase, D., Coles, R., (2013) The Theorized Urban Gradient (TUG) method—A

conceptual framework for socio-ecological sampling in complex urban agglomerations.

Ecological Indicators 36, 100-110.

R Core Team, (2012) R: A language and environment for statistical computing. In 'R:

Foundation for Statistical Computing,. ' Vienna, Austria

Ranius, T., Niklasson, M., Berg, N., (2009) Development of tree hollows in pendunculate oak

(Quercus robur). Forest Ecology and Management 257, 303-310.

Rankmore, B., Price, O., (2004) Effect of habitat fragmentation on the vertebrate fauna of

tropical woodlands, Northern Territory. In: Conservation of Australia's Forest Fauna. (Ed.)

D. Lunney, Royal Zoological Society of New South Wales, Sydney, pp. 452-471.

Reardon, T., (2009) Survey guidlelines for Australia's threatened bats; guidlines for detecting

bats listed as threatened under the Environment Protection and Biodiversity Conservation Act

1999. Department of the Environment, Water, Heritage & the Arts.

Reinhold, L., Law, B., Ford, G., Pennay, M., (2001) Key to the bat calls of south-east

Queensland and north-east New South Wales. Queensland Department of Natural Resources

and Mines: State Forests of New South Wales: University of Queensland: New South Wales

National Parks and Wildlife Service

Rhind, S., (1996) Habitat tree requirements and the effects of removal during logging on the

marsupial brush-tailed phascogale, Phascogale tapoatafa tapoatafa in western Australia. The

Western Australian Naturalist 21, 1-22.

Rhodes, J., Wiegand, T., McAlpine, C., Callaghan, J., Lunney, D., Bowen, M., Possingham,

H., (2006) Modeling species' distribution to improve conservation in semiurban landscapes:

Koala case study. Conservation Biology 20, 449-459.

Rhodes, M., Wardell-Johnson, G., Rhodes, M., Raymond, B., (2005) Applying network

analysis to the conservation of habitat trees in urban environments: a case study from

Brisbane, Australia. Conservation Biology 20, 861-869.

Page 195: A thesis submitted for the fulfilment for the requirements

173

Richards, G., Hall, L., Parish, S., (2012) A Natural History of Australian Bats. CSIRO,

Collingwood

Richardson, D. M., Bradford, J. W., Range, P. G., Christensen, J., (1999) A video probe

system to inspect red-cockaded woodpecker cavities. Wildlife Society Bulletin 27, 353-356.

Riley, S. J., Siemer, W. F., Decker, D. J., Carpenter, L. H., Organ, J. F., Berchielli, L. T.,

(2003) Adaptive impact management: An integrative approach to wildlife management.

Human Dimensions of Wildlife 8, 81-95.

Ross, G., (2004) Ibis in Urban Sydney: a gift from Ra or a Pharaoh's Curse? In: Urban

wildlife more than meets the eye. (Eds) D. Lunney and S. Burgin, Royal Zoological Society

of New South Wales, Sydney, pp. 148-152.

Ross, K. A., Fox, B. J., Fox, M. D., (2002) Changes to plant species richness in forest

fragments: frament age, disturbance and fire history may be as important as area. Journal of

Biogeography 29, 749-765.

Ross, Y., (1998) Habitat trees and hollow-dependant fauna. Queensland Department of

Natural Resources, Brisbane. 1-18.

Ross, Y., (1999) Hollow bearing trees in native forest permanent inventory plots in south-east

Queensland: Forest Ecosystem and Assessment Technical Papers. Queensland Department of

Natural Resources, Brisbane. 1-36.

Rowntree, R., (1988) Ecology of the Urban forest: Introduction to Part III. Landscape and

Urban Planning 15, 1-10.

Rowston, C., (1998) Nest and refuge tree usage by squirrel gliders Petaurus norfolcensis, in

south-eastern Queensland. Wildlife Research 25, 157-164.

Rowston, C., Catterall, C., Hurst, C., (2002) Habitat preferences of squirrel gliders, Petaurus

norfolcensis, in the fragmented landscape of southeast Queensland. Forest Ecology and

Management 164, 197-209.

Rudman, P., (1965) Causes of natural durability in timber XIX. Ageing of eucalypt

heartwoods and its effects on decay resistance. Holzforschung 19, 190-195.

Page 196: A thesis submitted for the fulfilment for the requirements

174

Ryan, T. J., Philippi, T., Leiden, Y. A., Dorcas, M. E., Wingley, T. B., Gibbons, J. W., (2002)

Monitoring herpetofauna in a managed forest landscape: effects of habitat types and census

techniques. Forest Ecology and Management 167, 83-90.

Sala, A., Piper, F., Hoch, G., (2010) Phsiological mechanisms of drought-induced tree

mortality are far from being resolved. New Phytologist 186, 272-281.

Saunders, D. A., Hobbs, R. J., Margules, C. R., (1991) Biological consequences of ecosystem

fragmentation: A review. Conservation Biology 5, 18-32.

Saunders, D. A., Smith, G. T., Rowley, I., (1982) The availability and dimensions of tree

hollows that provide nest sites for Cockatoos (Psittaciformes) in Western Australia. Wildlife

Research 9, 541-556.

Schroeter, R., Houghton, K., (2011) Neo-planning: Location based social media to engage

Australian's new digital locals. In: 'Proceedings of Planning Institute of Australia National

Conference'. Planning Institute of Australia. Hobart, Australia.

Shochat, E., Warren, P.S., Faeth, S.H., McIntyre, N.H., Hope, D. (2006). "From patterns to

emerging processes in mechanistic urban ecology." Trends in Ecology & Evolution 21, 186-

191.

Sewell, S. R., Catterall, C., (1998) Bushland modification and styles of urban development:

their effects on birds in south-east Queensland. Wildlife Research 25, 41-63.

Shackleton, C. M., (1993) Fuelwood harvesting and sustainable utilisation in a communal

grazing land and protected area of the eastern Transvaal lowveld. Biological Conservation

63, 247-254.

Shanley, P., Lopez, C., (2009) Out of the loop: Why research rarely reaches policy makers

and the epublic and what can be done. Biotropica 41, 535-544.

Shea, S. R., Peet, G. B., Cheney, N. P., (1981) The role of fire in forest management In: Fire

and the Australian biota. (Eds) A. M. Gill, R. H. Groves and I. R. Noble, Australian

Academy of Science, Canberra, pp. 443-470

Shearman, P. L., Ash, J., Mackey, B., Bryan, J., Lokes, B., (2009) Forest Conversion and

Degradation in Papua New Guinea Biotropica 41, 379-390.

Page 197: A thesis submitted for the fulfilment for the requirements

175

Shepheard, M. L., Martin, P., (2011) Using the moot court to trial legislation about land

stewardship. Land Use Policy 28, 371-446.

Shukla, J., Nobre, P., Sellers, P., (1990) Amazon deforestation and climate change. Science

247, 1322-1326.

Simberloff, D., Farr, J. A., Cox, J., Mehlman, D. W., (1992) Movement corridors:

Conservation bargains or poor investments? Conservation Biology 6, 493-504.

Smith, A. P., Lindenmayer, D., Begg, R. J., Macfarlane, M. A., Seebeck, J. H., Suckling, G.

C., (1989) Evaluation of the stagwatching technique for census of possums and gliders in tall

open forest. Australian Wildlife Research 16, 575-580.

Smith, A. P., Lindenmayer, D. B., (1988) Tree hollow requirements of Leadbeater's Possum

and other possums and gliders in timber production forests of the Victorian Central

Highlands. Wildlife Research 15, 347-362.

Smith, A. P., Lindenmayer, D. B., (1992) Forest succession and habitat management for

Leadbeater's possum in the State of Victoria, Australia. Forest Ecology and Management 49,

311-332.

Smith, C. M., Wachob, D. G., (2006) Trends associated with residential development in

riparian breeding bird habitats along the Snake River in Jackson Hole, WY USA:

implications for conservation planning. Biological Conservation 128, 431-446.

Smith, C. Y., Moroni, M. T., Warkentin, I. G., (2009) Snag dynamics in post-harvest

landscapes of western Newfoundland balsam fir-dominated boreal forests. Forest Ecology

and Management 258, 823-839.

Smith, G., Mathieson, M., Hogan, L., (2007) Home range and habitat use of a low-density

population of greater gliders, Petauroides volans (Pseudocheiridae: Marsupialia), in a

hollow-limiting environment. Wildlife Research 34, 472-483.

Smith, G., Phillips, E., Doret, G., (2006) The contribution of biodiversity conservation on

private land to Australian cityscapes. Biodiversity Conservation on Private Land ISoCaRP

Congress.

Snep, R., Opdam, P., (2010) Integrating nature values in urban planning and design. In:

Urban Ecology. (Ed.) K. J. Gaston, Cambridge University Press, London, pp. 261-286.

Page 198: A thesis submitted for the fulfilment for the requirements

176

Spearbritt, P., (2009) The 200 km City: Brisbane, the Gold Coast and Sunshine Coast.

Australian Economic History Review 49, 87-106.

Spies, T. A., Franklin, J. F., Thomas, B. T., (1988) Course woody debris in Douglas-fir

forests of western Oregon and Washington. Ecology 69, 1698-1702.

Spring, D. A., Bevers, M., Kennedy, J. O. S., 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.

Staben, G. W., Evans, K. G., (2008) Estimates of tree canopy loss as a result of Cyclone

Monica, in the Magela Creek catchment northern Australia. Austral Ecology 33, 562-569.

Statsoft. (2005) Statistica (data anlysis software system).7.1

Stenhouse, R., (2004) Local government conservation and management of native vegetation

in urban Australia. Environmental Management 34, 209-222.

Stewart, G. H., Ignatieva, M. E., Meurk, C. D., Earl, R. D., (2004) The re-emergence of

indigenous forest in an urban environment, Christchurch, New Zealand. Urban forestry and

Urban Greening 2, 149-158.

Stojanovic, D., Webb, M., Roshier, D., Saunders, D., Heinsohn, R., (2012) Ground-based

survey methods both overestimate and underestimate the abundance of suitable tree-cavities

for the endangered Swift Parrot. Emu 112, 350-356.

Stone, C., Coops, N. C., (2004) Assessment and monitoring of damage from insects in

Australian eucalypt forests and commercial plantations. Australian Journal of Entomology

43, 283-292.

Subedi, N., Mahadev, S., (2011) Individual-tree diameter growth models for black spruce and

jack pine plantations in northern Ontario. Forest Ecology and Management 261, 2140-2148.

Sutherland, W., (2006) Ecological Census Techniques. Cambridge University Press, New

York

Symstad, A. J., Tilman, D., Willson, J., Knops, J. M. H., (1998) Species Loss and Ecosystem

Functioning: Effects of Species Identity and Community Composition. Oikos 81, 389-397.

Page 199: A thesis submitted for the fulfilment for the requirements

177

Tabuti, J. R. S., Dhillion, S. S., Lye, K. A., (2003) Firewood use in Bulamogi County,

Uganda: species selection, harvesting and consumption patterns. Biomass and Bioenergy 25,

581-596.

Tanabe, S., Toda, M. J., Vinokurova, A. V., (2001) Tree shape, forest structure and diversity

of drosophilid community: Comparison between boreal and temperate birch forests.

Ecological Research 16, 369-385.

Taylor, B., Goldingay, R. L., (2009) Can road-crossing structures improve population

viability of an urban gliding mammal? Ecology and Society 14, 13-33.

Taylor, I., Kirstin, I., (1999) Woodland survivors. Wingspan 9, 8-11.

Taylor, P., (2003) Land Zones of Queensland. Brisbane. Queensland Government.

Taylor, R., Haseler, M., (1993) Occurrence of potential nest trees and their use by birds in

sclerophyll forest in north east Tasmania. Australian Forestry 56, 165-171.

Taylor, S. L., MacLean, D.A., (2007) Spatiotemporal patterns of mortality in declining

balsam fir and spruce stands. Forest Ecology and Management 253, 188-201.

Terho, M., (2009) An assessment of decay among urban Tilia, Betula, and Acer trees felled as

hazardous. Urban Forestry and Urban Greening 8, 77-85.

Tews, J., Brose, U., Grimm, V., Tielbörger, K., Wichmann, M., Schwager, M., Jeltsch, F.,

(2004) Animal species diversity driven by habitat heterogeneity/diversity: the importance of

keystone structures. Journal of Biogeography 31, 79-92.

Thomas, I., (2007) Environmental Policy: Australian Practice in the Context of Theory. The

Federation Press

Threlfall, C., Law, B. S., Penman, T., Banks, P., B., (2011) Ecological processes in urban

landscapes: mechanisms influencing the distribution and activity of insectivorous bats.

Ecography 34, 814-826.

Threlfall, C. G., Law, B., Banks, P. B., (2012) Sensitivity of insectivorous bats to

urbanization: Implications for suburban conservation planning. Biological Conservation 146,

41-52.

Page 200: A thesis submitted for the fulfilment for the requirements

178

Tideman, S., Boyden, J., Elvish, R., O'Gorman, B., (1992) Comparison of the breeding sites

and habitat of two hole-nesting estrilid finches, one endangered, in northern Australia.

Journal of Tropical Ecology 8, 373-388.

Tilman, D., May, R. M., Lehman, C. L., Nowak, M. A., (1994) Habitat destruction and the

extinction debt. Nature 371, 65-66.

Tjørve, E., (2010) How to resolve the SLOSS debate: Lessons from species-diversity models.

Journal of Theoretical Biology 264, 604-612.

Todarella, P., Chalmers, A., (2006) The characteristics of five species of hollow-bearing trees

on the New South Wales Central Coast. Proceedings of the Linnean Society of New South

Wales 128, 1-14.

Traill, B. J., Lill, A., (1997) Use of tree hollows by two sympatric gliding possums, the

squirrel glider, Petaurus norfolcensis, and the sugar glider, P. breviceps. Australian

Mammalogy, 79-88.

Tran, C., Wild, C., (2000) A Review of Current Knowledge and Literature to Assist in

Determining Ecologically Sustainable Fire Regimes for the Southeast Queensland Region,

Griffith University and The Fire and Biodiversity Consortium.

Tratalos, J., Fuller, R. A., Warren, P. H., Davies, R. G., Gaston, K. J., (2007) Urban form,

biodiversity potential and ecosystem services. Landscape and Urban Planning 83, 308-317.

Treby, D. L., Castley, J.G., Hero, J, M., (2014). Forest conservation policy implementation

gaps: Consequences for the management of hollow-bearing trees in Australia. Conservation

and Society 12, 16-26.

Turner, J., (1984) Radiocarbon dating of wood and charcoal in an Australian forest

ecosystem. Australian Forestry 47, 79-83.

van der Ree, R., Bennett, A. F., Gilmore, D. C., (2003) Gap-crossing by gliding marsupials:

thresholds for use of isolated woodland patches in an agricultural landscape. Biological

Conservation 115, 241-249.

van der Ree, R., Bennett, A. F., Soderquist, T. R., (2006) Nest-tree selection by the

threatened brush-tailed phascogale, Phascogale tapoatafa (Marsupialia: Dasyuridae) in a

highly fragmented agricultural landscape. Wildlife Research 33, 113-119.

Page 201: A thesis submitted for the fulfilment for the requirements

179

van der Ree, R., Loyn, R. H., (2002) The influence of time since fire and distance from fire

boundary on the distribution and abundance of arboreal marsupials in Eucalyptus regnans

dominated forest in the Central Highlands of Victoria. Wildlife Research 29, 151-158.

van der Ree, R., McCarthy, M. A., (2005) Inferring persistence of indigenous mammals in

response to urbanisation. Animal Conservation 8, 309-319.

Vermeulen, S., Campbell, B., Matzke, G., (1996) The consumption of wood by rural

households in Goke Communal Area, Zimbabwe. Human Ecology 24.

Villard, M.-A., (2002) Habitat Fragmentation: Major conservation issue or intellectual

attractor? Ecological Applications 12, 319-320.

Vitousek, P., (1997) Human domination of Earth's ecosystems. Science 275, 494-499.

Ward, S. J., (2000) The efficacy of nestboxes versus spotlighting for detecting feathertail

gliders. Wildlife Research 27, 75-79.

Wardell-Johnson, D., (2000) Responses of forest eucalypts to moderate and high density fire

in the Tingle Mosaic, south-western Australia: comparisons between locally endemic and

regionally distributed species. Austral Ecology 25, 409-421.

Waters, D. A., Burrows, G. E., Harper, J. D. I., (2010) Eucalyptus regnans (Myrtaceae): A

fire sensitve eucalypt with a resprouter epicormic structure American Journal of Botany 97,

546-556.

Watts, K., Selman, P., (2004) Forcing the pace of biodiversity action: a force-field analysis of

conservation effort at the landscape scale. Local Environment 9, 5-20.

Webala, P. W., Craig, M. D., Law, B. S., Armstrong, K. N., Wayne, A. F., Bradley, J. S.,

(2011) Bat habitat use in logged jarrah eucalypt forests of south-western Australia. Journal of

Applied Ecology 48, 398-406.

Webb, H., Foley, A., (1996) Urban bushland under threat: review of urban bushland and

recommendations for its protection. Nature Conservation Council of NSW, Sydney.

Webb, J., Shine, R., (1997) Out on a limb - conservation implications of tree-hollow use by a

threatened snake species Hoplocephalus bungaroides, (Serpentes, Elapidae). Biological

conservation 86, 233-242.

Page 202: A thesis submitted for the fulfilment for the requirements

180

Werner, P., Prior, L., (2007) Tree-piping termites and growth and survival of host trees in

savanna woodland of north Australia. Journal of Tropical Ecology 23, 611-622.

West, P. W., Cawsey, E. M., Stol, J., Freudenberger, D., (2008) Firewood harvest from

forests of the Murray-Darling Basin, Australia. Part 1: Long-term, sustainable supply

available from native forests. Biomass and Bioenergy 32, 1206-1219.

Whitford, K. R., (2002) Hollows in jarrah, Eucalyptus marginata and marri, Corymbia

calophylla trees: I. Hollow sizes, tree attributes and ages. Forest Ecology and Management

160, 201-214.

Whitlow, R., (2000) The Geology of the Gold Coast Region. TriMap Pty Ltd, Gold Coast.

Whittaker, S., (1997) Are Australian councils' willing and able' to implement local agenda

21? Local Environment 2, 319-328.

Williams, A., Shackelton, C., (2002) Fuelwood use in South Africa: Where to in the 21st

Century? Southern African Forestry Journal 196, 1-6.

Williams, J., Read, C., Norton, A., Dovers, S., Burgman, M., Proctor, W., Anderson, H.,

(2001) Australia State of the Environment. CSIRO, Canberra.

Williams, N., McDonnell, M., Phelan, G., Keim, L., van der Ree, R., (2006) Range expansion

due to urbanisation: Increased food resources attract grey-headed flying-fox, Pteropus

poliocephalus to Melbourne. Austral Ecology 31, 190-198.

Williams, N., McDonnell, M., Seager, E., (2004) Factors influencing the loss of an

endangered ecosystem in an urbanising landscape: a case study of native grasslands from

Melbourne, Australia. Landscape and Urban Planning 71, 35-49.

Williams, N., Morgan, J., McDonnell, M., McCarthy, M., (2005a) Plant traits and local

extinctions in natural grasslands along an urban-rural gradient. Journal of Ecology 93, 1203-

1213.

Williams, N. S. G., McDonnell, M. J., Seager, E. J., (2005b) Factors influencing the loss of

an endangered ecosystem in an urbanising landscape: a case study of native grasslands from

Melbourne, Australia. Landscape and Urban Planning 71, 35-49.

Page 203: A thesis submitted for the fulfilment for the requirements

181

Wintle, B. A., Kavanagh, R. P., McCarthy, M. A., Burgman, M. A., (2005) Estimating and

dealing with detectability in occupancy surveys for forest owls and arboreal maruspials. The

Journal of Wildlife Management 69, 905-917.

Woinarski, J. C., Recher, H. F., Majer, J. D., (1997) Vertebrates of Eucalypt Formations. In:

Eucalypt Ecology: Individuals to Ecosystems. (Eds) J. E. Williams and J. C. Woinarski,

Cambridge University Press, pp. 303-341.

Wolf, A., Iler, P. F. M., Bradshaw, R. H. W., Bigler, J., (2004) Storm damage and long-term

mortality in a semi-natural, temperate deciduous forest. Forest Ecology and Management

188, 197-210.

Wood, K. A., (1993) The avian population of an urban bushland reserve at Wollongong, New

South Wales: implications for management. Landscape and Urban Planning 23, 81-95.

Wood, M., Wallis, R., (1998) Potential competition for nest sites between feral European

honeybees, Apis mellifera and common brushtail possums Trichosurus vulpecula. Australian

Mammalogy 20, 377-388.

Wormington, K., Lamb, D., (1999) Tree hollow development in wet and dry sclerophyll

forests in south-east Queensland, Australia. Australian Forestry 62, 336-345.

Wormington, K., Lamb, D., McCallum, H. I., Moloney, D. J., (2002) Habitat requirements

for the conservation of arboreal marsupials in dry sclerophyll forests of southeast Queensland

Australia. Forest Science 48, 217-227.

Wormington, K., Lamb, D., McCallum, H. I., Moloney, D. J., (2003) The characteristics of

six species of living hollow-bearing trees and their importance for arboreal marsupials in the

dry sclerophyll forests of southeast Queensland, Australia. Forest Ecology and Management

182, 75-92.

Wormington, K., Lamb, L., McCallum, H. I., Moloney, D. J., (2005) The status of hollow-

bearing trees required for the conservation of arboreal marsupials in the dry sclerophyll

forests of south-east Queensland, Australia. Pacific Conservation Biology 11, 38-49.

Young, B., Liang, J., Chapin, F. S., (2011) Effects of species and tree size diversity on

recruitment in the Alaskan boreal forest: A geospatial approach. Forest Ecology and

Management 262, 1608-1617.

Page 204: A thesis submitted for the fulfilment for the requirements

182

Zhao, S., Da, L., Tang, Z., Fang, H., Song, F. J., (2006) Ecological consequences of rapid

urban expansion, Shanghai, China. Frontiers in Ecology and Environment 4, 341-346.

Zipperer, W. C., Pickett, S. T. A., (2001) Urban Ecology: Patterns of Population Growth and

Ecological Effects. In: Urban Ecology. John Wiley & Sons, Ltd,

Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., Smith, G. M., (2009) Mixed Effects

Models and Extensions in Ecology with R. Springer, United States of America

Zuur, A. F., Leno, E. N., Smith, G. M., (2010) Analysing ecological data: Statistics for

biology and health. Springer, New York

Page 205: A thesis submitted for the fulfilment for the requirements

183

List of Appendices:

Page 206: A thesis submitted for the fulfilment for the requirements

184

Appendix 1: Hollow using fauna of Australia

Table 1: Frogs recorded from hollows, and other arboreal or semi arboreal frogs that may use hollows.

■ indicates the species is confirmed to use hollows (Gibbons and Lindenmayer 2002). ▲ indicates fauna occurring in south-east Queensland. ♦ indicates conservation concern

Species Name Common Name

Frogs

■ Lechriodus fletcheri Fletcher's Frog

Litoria adelaidensis Slender Tree Frog

Litoria bicolor Northern Dwarf Tree Frog

■ ▲ Litoria caerulea Green Tree Frog ▲ Litoria chloris Red-eyed Tree Frog

Litoria citropa Blue Mountains Tree Frog

▲ Litoria denata Bleating Tree Frog ■ Litoria ewingii Brown Tree Frog, Whistling Tree Frog ▲ Litoria fallax Eastern Sedge Frog

Litoria genimaculata New Guinea Tree Frog

■ ▲ Litoria gracilenta Dainty Green Tree Frog

Litoria infrafrenata Giant Tree Frog, White-lipped Tree Frog

Litoria jervisiensis Jervis Bay Tree Frog

Litoria littlejohni Littlejohn's Tree Frog

Litoria nyakalensis Nyakala Frog, Mountain Mist Frog

■ ▲ Litoria peronii Peron's Tree Frog

Litoria personata Masked Frog

▲ Litoria phyllochroa Leaf Green Tree Frog ■ Litoria piperata Peppered Tree Frog ■ ▲ Litoria rothii Roth's Tree Frog ■ ▲ Litoria rubella Desert Tree Frog

Litoria subglandulosa Glandular Frog

Litoria splendida Magnificent Tree Frog

▲ Litoria tyleri Tyler's Tree Frog

Litoria xanthomera Orange-thighed Frog

Nyctimystes dayi Australian Lace-lid

▲ Cophixalus ornatus Ornate Frog

Page 207: A thesis submitted for the fulfilment for the requirements

185

Table 2: Reptiles recorded that use/may use hollows in Australia Species Name Common Name

Geckos

Cyrtodactylus louisiadensis Ring Tailed Gecko

Diplodactylus ciliaris Spiny-tailed Gecko

Diplodactylus intermedius Eastern Spiny-tailed Gecko

Diplodactylus rankini

Diplodactylus spinigerus Western Spiny-tailed Gecko

Diplodactylus strophurus

Diplodactylus taenicauda Golden-tailed Gecko

Diplodactylus wellingtonae

■ Diplodactylus williamsi

Gehyra australis Northern Dtella

Gehyra baliola

Gehyra catenata

Gehyra dubia

Gehyra oceanica Oceanic Gecko

Gehyra purpurascens

■ Gehyra variegata Tree Dtella

Hemidactylus frenatus House Gecko

Lepidodactylus lugubris Mourning Gecko

Lepidodactylus pumilus

Oedura castelnaui Northern Velvet Gecko

Oedura marmorata Marbled Velvet Gecko

Oedura monilis Ocellated Velvet Gecko

■ Oedura reticulata Reticulated Velvet Gecko

Oedura rhombifer

▲ Oedura robusta Robust Velvet Gecko ▲ Phyllurus caudiannulatus

Phyllurus cornutus Northern Leaf-tailed Gecko

Pseudothecadactylus australis

Lizards

Caimanops amhiboluroides

■ Chlamydosaurus kingii Frilled Lizard

Diporiphora australis

Diporiphora superba

Hypsilurus boydii Boyd's Forest Dragon

▲ Hypsilurus spinipes Southern Angle-headed Dragon

Lopthangus gilbertii Gilbert's Dragon

Lophognathus longirostris Long-nosed Water Dragon

▲ Pogona barbata Bearded Dragon

Pogona minor Dwarf Bearded Dragon

Pogona mitchelli

Pogona nullarbor

Pogona vitticeps

Goannas & Monitors ■ Varanus caudolineatus ■ Varanus gilleni Pygmy Mulga Monitor

■ Varanus scalaris Spotted Tree Monitor

Varanus mitchelli Mitchell's Water Dragon

Varanus semiremex Rusty Monitor

Page 208: A thesis submitted for the fulfilment for the requirements

186

Table 2: Reptiles in Australia recorded using tree hollows continued. Species Name Common Name

Goannas & Monitors

■ Varanus timorensis Spotted Tree Monitor ■ ▲ Varanus tristis

■ ▲ Varanus varius Lace Monitor

Skinks

Crytoblepharus carnabyi

Crytoblepharus plagiocephalus

■ ▲ Crytoblepharus virgatus

Egernia formosa

Egernia napoleonis ■ Egernia saxatilis Black-rock Skink

▲ Egernia striolata Tree Skink

Emoia longicauda

Eulamprus heatwolei ▲ Eulamprus martini

Eulamprus sokosoma

■ Eulamprus tenuis ■ Eulamprus tigrinus Rainforest Water Skink

Pseudemoia coventryi

Pseudemoia orocrypta ■ Pseudemoia spenderi Spencers Skin

Snakes

■ Morelia viridis Green Python ▲ Antaresia childreni Children's Python

Antaresia maculosus

Antaresia stimsoni ■ Morelia amethistina Amethystine Python

Morelia bredli Centralian Carpet Python

■ ▲ Morelia spilota Carpet/Diamond Python ■ Boiga irregularis Brown Tree Snake ■ ▲ Dendrelaphis calligastra Northern Tree Snake ■ ▲ Dendrelaphis punctulata Common Tree Snake

Hoplocephalus bitorquatus Pale-headed Snake

■ Hoplocephalus bungaroides Broad-headed snake ■ Hoplocephalus Stephensii Stephen's banded Snake

Page 209: A thesis submitted for the fulfilment for the requirements

187

Table 3: Birds that use tree hollows in Australia. The list does not include taxa that feed in hollows but do not otherwise use them (e.g. some species of corvids). Species Name Common Name

Birds

▲ Phaethon lepturus White-tailed Tropic bird ▲ Tadorna tadornoides Australian Shelduck ▲ Tadorna radjah Radjah Shelduck ▲ Nettapus pulchellus Green Pygmy-goose ▲ Nettapus coromandelianus Cotton Pygmy-goose ▲ Chenonetta jubata Maned Duck ▲ Malancorhynchus membranaceus Pink-eared Duck ▲ Anas gracilis Grey Teal ▲ Anas castanea Chestnut teal ▲ Anas superciliosa Pacific Black Duck

Anas platyrhynchos Mallard (Introduced)

▲ Anas thynchotis Australian Shoveler ▲ Gallirallus philippensis Buff-banded Rail ▲ Falco cenchroides Australian Kestrel ▲ Falco peregrinus Peregrine Falcon ▲ Geopelia cuneata Diamond Dove ▲ Probiscar aterrimus Palm Cockatoo ▲ Calyptorhynchus banksii Red-tailed Black Cockatoo ▲♦ Calyptorhynchus lathami Glossy-black Cockatoo ▲ Calyptorhynchus funereus Yellow-tailed Black Cockatoo

Calyptorhynchus latirostris Short-billed Black Cockatoo

Calyptorhynchus baudinii Long-billed Black Cockatoo

Callocephalon fimbriatum Gang Gang Cockatoo

▲ Cacatua roseicapilla Galah

Cacatua tenuirostris Long-billed Corella

Cacatua pastinator Western Corella

▲ Cacatua sanguinea Little Corella

Cacatua leadbeateri Major Mitchell's Cockatoo

▲ Cacatua galerita Sulpher-crested Cockatoo ▲ Nymphicus hollandicus Cockateil ▲ Trichoglossus haematodus Rainbow Lorikeet ▲ Psitteuteles versicolor Varied Lorikeet ▲ Trichoglossus chlorolepidotus Scaly-breasted Lorikeet ▲ Glossopsitta concinna Musk Lorikeet ▲ Glossopsitta pusilla Little Lorikeet

Glossopsitta porphyrocephala Purple-crowned Lorikeet

▲ Eclectus roratus Eclectus Parrot

Geoffrous geoffroyi Red-cheeked Parrot

▲♦ Cyclopsitts diophthalma Double-eyed Fig Parrot ▲ Alisterus scapularis Australian King Parrot ▲ Aprosmictus erythropterus Red-winged parrot ▲ Polytelis swainsonii Superb Parrot

Polytelis anthopeplus Regent Parrot

Polytelis alexandrea Princess Parrot

Platycercus caledonicus Green Rosella

▲ Platycercus elegans Crimson Rosella ▲ Platycercus eximius Eastern Rosella ▲ Platycercus adscitus Pale-headed Rosella

Page 210: A thesis submitted for the fulfilment for the requirements

188

Table 3. Birds continued. Species Name Common Name

Platycercus venustus Northern Rosella

Platycercus icterotis Western Rosella

Barnadius zonarius Australian Ringneck

Purpureicephalus spurius Red-capped Parrot

▲ Northiella haematogaster Blue Bonnet ▲♦ Lathamus discolor Swift parrot ▲ Psephotus haematonotus Red-rumped Parrot ▲ Psephotus varius Mulga Parrot ▲ Mesopsitta cusundulatus Budgerigar ▲ Neophsephotus bourkii Bourke's Parrot ▲ Neophema chrysostoma Blue-winged Parrot

Neophema elegans Elegant Parrot

Neophema chrysogaster Orange-bellied Parrot

▲ Neophema pulchella Turquoise Parrot ▲ Neophema splendid Scarlet-chested Parrot ▲♦ Ninox strenua Powerful Owl ▲ Ninox rufua Rufous Owl ▲ Ninox conivens Barking Owl ▲ Ninox novaeseelandiae Southern Boobook ▲♦ Tyto tenebricosa Sooty Owl

Tyto multipunctata Lesser Sooty Owl

▲♦ Tyto novaehollandiae Masked Owl ▲ Tyto alba Barn Owl ▲ Aegotheles cristatus Australian Owlet-nightjar ▲ Alseco pusilla Little Kingfisher ▲ Dacelo novaeguineae Laughing Kookaburra ▲ Dacelo leachii Blue-winged Kookaburra ▲ Syma torotoro Yellow-billed Kingfisher ▲ Todiramphus macleayii Forest Kingfisher ▲ Todiramphus pyrrhopygia Red-backed Kingfisher ▲ Todiramphus sanctus Sacred Kingfisher ▲ Todiramphus chloris Collared Kingfisher ▲ Eurystomus orientalis Dollarbird ▲ Cecropis nigricans Tree Martin

Hirundo ariel Fairy Martin

Zoothera heinei Russet Ground Thrush

Zoothera lunulata Australian Ground Thrush

Petroica rodinogaster Pink Robin

Petroica multicolor Scarlet Robin

▲ Petroica phoenica Flame Robin ▲ Melanodryas cucullata Hooded Robin

Melanodryas vitta Dusky Robin

▲ Colluricincla harmonica Grey Shrike-thrush

Oreocia gutturalis Crested Bellbird

▲ Aphelocephala leucopsis Southern Whiteface

Aphelocephala nigricinta Banded Whiteface

▲ Acanthiza reguloides Buff-rumped Thornbill

Acanthiza inornata Western Thornbill

▲ Acanthiza uropygialis Chestnut-rumped Thornbill

Page 211: A thesis submitted for the fulfilment for the requirements

189

Table 3. Birds continued. Species Name Common Name ▲ Cormobates leucophaea White-throated Treecreeper ▲ Climacteris melanura Black-tailed Treecreeper

Climacteris rufa Rufous Treecreeper

▲ Climacteris erythrops Red-browed Treecreeper ▲ Climacteris affinis White-browed Treecreeper ▲ Climacteris picumnus Brown Treecreeper

Paralotus quadragintus Forty-spotted Pardalote

▲ Pardalotus striatus Striated Pardalote ▲ Poephila guttata Zebra Finch ▲ Neochima phaeton Crimson Finch ▲ Erythrura gouldiae Gouldian Finch ▲ Poephila cincta Black-throated Finch ▲ Aplonis metallica Metallic Starling

Passer domesticus House Sparrow (Introduced)

Passer montanus Tree Sparrow (Introduced)

Sturnus vulgaris Common Starling (Introduced)

Acidotheres tristis Common Myna (Introduced)

▲ Artamus minor Little Woodswallow ▲ Artamus cyanopterus Dusky Woodswallow ▲ Artamus personatus Masked Woodswallow ▲ Artamus supercilliosus White-browed Woodswallow ▲ Artamus leucorhynchus White-breasted Woodswallow

Page 212: A thesis submitted for the fulfilment for the requirements

190

Table 4: Microbats that use tree hollows in Australia Species Name Common Name

Mammals/Microbats

▲ Saccolaimus flaviventris Yellow-bellied Sheathtail-bat

Saccolaimus mixtus Cape York Sheathtail-bat

Saccolaimus saccolaimus Bare-rumped Sheathtail-bat

Taphozous kapalgensis Arnhem Sheathtail-bat

Rhinolophus phillippinensis achilles Greater Horseshoe Bat*

Rhinolophus phillippinensis maros Large-eared Horseshoe Bat*

Hipposideros ater Dusky Leaf-nosed Bat

Hipposideros diadema reginae Diadem Leaf-nosed Bat*

Hipposideros semoni Semon's Leaf-nosed Bat

Rhinonycteris aurantius Orange Leaf-nosed Bat

▲ Chalinolobus gouldii Gould's Wattled Bat ▲ Chalinolobus morio Chocolate Wattled Bat ▲ Chalinolobus nigrogriseus Hoary Wattled Bat

Chalinolobus picatus Little Pied Bat

Falsistrellus mackenziei Western Falsistrelle

▲♦ Falsistrellus tasmaniensis Eastern Falsistrelle ▲♦ Myotis adversus Large-footed Myotis ▲ Nycophilus bifax Northern Long-eared Bat ▲ Nyctophilus geoffroyi Lesser Long-eared Bat ▲ Nyctophilus gouldi Gould's Long-eared Bat ▲ Nyctophilus timorensis Greater Long-eared Bat

Pipistrellus adamsi Cape York Pipistrelle

Pipistrellus murrayi Christmas Island Pipistrelle

Scoteanax ruepellii Greater Broadnosed Bat

▲♦ Scotorepens balstone Inland Broadnosed Bat ▲♦ Scotorepens greyii Little Broadnosed Bat ▲♦ Scotorepens orion Eastern Broadnosed Bat

Scotorepens sanborni Northern Broadnosed Bat

Vespadelus baverstocki Inland forest Bat

▲♦ Vespadelus darlingtoni Large Forest Bat

Vespadelus pumilus Eastern Forest Bat

▲♦ Vespadelus regulus Southern Forest Bat ▲♦ Vespadelus vulturnus Little Forest Bat ▲ Austronomus australis White-striped Freetail Bat

Charephon jobensis Northern Freetail Bat

▲ Mormopterus beccarii Beccari's Freetail Bat ▲♦ Mormopterus norfolcensis East-coast Freetail Bat

Mormopterus sp. Eastern Freetail Bat

Mormopterus sp. Hairy-nosed Freetail Bat

Mormopterus loriae Little Northern Freetail Bat*

Mormopterus planiceps (small penis sp) Inland Freetail Bat*

Mormopterus planiceps (eastern long penis sp) Southern Freetail Bat*

Mormopterus planiceps (western long penis sp) Western Freetail Bat*

Page 213: A thesis submitted for the fulfilment for the requirements

191

Table 5: Arboreal and scansorial mammals that have been recorded using hollows in Australia. Species Name Common Name

Mammals

▲♦ Dasyurus maculatus Spotted-tailed Quoll

Dasyurus geoffroii Western Quoll (Chuditch)

▲♦ Phascogale tapoatafa Brush-tailed Phascogale

Phascogale calura Red-tailed Phascogale

▲ Antichinus flavipes Yellow-footed Antechinus (Mardo)

Antichinus agilis Agile Antechinus

Antichinus bellus Fawn Antechinus

▲♦ Antechinus stuartii Brown Antechinus ▲ Antechinus swainsonii Dusky Antechinus

Antechinus leo Cinnamon Antechinus

Sminthopsis dolichura Little Long-tailed Dunnart

Sminthopsis murina Common Dunnart

Sminthopsis leocopus White-footed Dunnart

Myrmecobius fasciatus Numbat

▲♦ Pseudocheirus peregrinus Common Ringtail Possum

Pseudocheirus occidentalis Western Ringtail Possum

▲ Pseudocheirus herbertensis Herbert River Ringtail Possum ▲ Pseudocheirus archeri Green Ringtail Possum ▲ Hemibelideus lemuroides Lemuroid Ringtail Possum ▲♦ Petauroides volans Greater Glider ▲♦ Petaurus australis Yellow-bellied Glider ▲ Petaurus breviceps Sugar Glider ▲♦ Petaurus norfolcensis Squirrel Glider ▲♦ Petaurus gracilis Mahogony Glider

Gymnobelideus leadbeateri Leadbeater's Possum

▲ Dactylopsila trivirgata Striped Possum ▲ Trichosurus vulpecula Common Brushtail Possum ▲ Trichosurus arnhemensis Northern Brushtail Possum ▲ Trichosurus caninus Mountain Brushtail Possum ▲ Phalanger orientalis Grey Cuscus ▲♦ Cercartetus nanus Eastern Pygmy Possum

Cercartetus consinnus Western Pygmy Possum

Cercartetus lepidus Little Pygmy Possum

▲ Cercartetus caudatus Lont-tailed Pygmy Possum ▲ Acrobatus pygmaeus Feathertail Glider

Tarsips rostratus Honey Possum

▲ Uromys causimaculatus White-tailed Rat ▲ Conilurus penicaillatus Brush-tailed Rabbit-rat ▲ Mesembriomys gouldii Black-footed Tree-rat ▲ Rattus fuscipes Bush Rat

Rattus rattus Black Rat (Introduced)

Rattus exulans Polynesian Rat (Introduced)

Mus musculus House mouse (Introduced)

Funambulus pennati Five-striped Palm Squirrel (Introduced)

Mustela furo Ferret (Introduced)

Vulpes vulpes European red fox (Introduced)

Felis cat Cat (Introduced)

Page 214: A thesis submitted for the fulfilment for the requirements

192

Appendix 2: Regional ecosystems containing wet and dry sclerophyll communities and their status found on the Gold Coast (Queensland Government 2009).

Regional Ecosystem Status Short Description

12.3.2 Of concern E. grandis open forest 12.3.3 ENDANGERED E. tereticornis open forest 12.3.7 Not of concern E. tereticornis on alluvial plain 12.3.11 Of concern Eucalyptus spp. coastal

E. tereticornis, E.siderophloia, C.intermedia, L. suaveolens

12.5.3 ENDANGERED E. tereticornis coastal 12.8.2 Of concern E. oreades open forest 12.8.8 Of concern E. saligna open forest 12.8.8a Of concern E. siderophloia open forest 12.8.14 Not of concern Eucalyptus spp. open forest

E.eugenioides, E.biturbinata, E melliodora 12.9.4 Not of concern E. racemosa open forest 12.9/10.7a Of concern Eucalyptus spp. open forest

E.tereticornis. E.crebra, E.siderophloia, C.intermedia, L. suaveolens

12.9/10.19a Not of concern Eucalyptus spp. open forest C.henryi, E.fibrosa, C.citriodora, E.acmenoides, A. leiocarpa. E.major

12.11.2 Not of concern Eucalyptus spp. open forest E.saligna, E.grandis. E.microcorys, E.acmenoides, L. confertus

12.11.3 Not of concern Eucalyptus spp. open forest E.siderophloia, E.propinqua, L. confertus

12.11.5 Not of concern Eucalyptus spp. open forest C.citriodora, E.siderophloia, L. confertus

12.11.5a Not of concern Eucalyptus spp. open forest E.tindaliae, E.carnea, C.citriodora, E.crebra, E.major, C.henryi, C.trachyphloia, E.siderophloia, E.microcorys, E.racemosa, E.propinqua

12.11.5j Not of concern Eucalyptus spp. open forest E.seeana, C.intermedia, E.siderophloia, C.citriodora, E.pilularis. L. suaveolens

12.11.5k Not of concern Eucalyptus spp. open forest C.henryi, E.fibrosa, C.citriodora, E.carnea, E.tindaliae, E.propinqua

12.11.9 Of concern E. tereticornis open forest 12.11.18 Not of concern E. moluccana open forest 12.11.23 ENDANGERED E. pilularis coastal

Page 215: A thesis submitted for the fulfilment for the requirements

193

Appendix 3: Site and plot number with GPS co-ordinates. Site name Plot No: Point X Point Y Andrews Reserve 1 540694.906555 6889735.367500 Aqua Promenade 1 543257.308090 6883229.676150 Aqua Promenade 2 543486.139761 6883479.078970 Austenville Road 1 530666.301243 6881462.468190 Austenville Road 2 530680.325252 6881733.416220 Austenville Road 3 530400.165131 6881961.965800 Behms Road 1 533165.752583 6927484.785180 Bonogin Forest Reserve 1 534633.094556 6886460.527840 Bonogin Forest Reserve 2 534552.811936 6885552.921390 Bonogin Forest Reserve 3 535540.313051 6887455.392450 Bonogin Forest Reserve 4 535253.426648 6884450.604270 Bonogin Forest Reserve 5 536542.029379 6884461.419280 Bonogin Forest Reserve 6 535521.355264 6885454.544230 Brushwood Ridge 1 535463.239208 6908060.224510 Brushwood Ridge 2 535361.124621 6908301.810920 Burleigh Headland NP 1 544864.684547 6892205.053510 Burleigh Knoll NP 1 543154.255047 6893677.694360 Burleigh Ridge 1 544005.036810 6892144.909970 Burleigh Ridge 2 544387.718958 6892372.062960 Carrington Road 1 537423.106852 6887339.083980 Daisy Hill Forest Reserve 1 515129.010051 6944705.488550 Daisy Hill Forest Reserve 2 515274.707680 6945640.440550 Daisy Hill Forest Reserve 3 516120.289081 6944481.220770 Daisy Hill Forest Reserve 4 517053.987953 6944356.571600 Daisy Hill Forest Reserve 5 516977.879368 6943430.255700 Elanora Conservation Area 1 543351.664747 6886452.523860 Elanora Conservation Area 2 543417.600948 6886270.312100 Elanora Wetland 1 543576.186581 6889471.948350 Gelt Road 1 525705.860191 6918287.600500 Gilston Road 1 529615.672771 6898301.197930 Hellman Street 1 534782.659501 6906647.004210 Herbert Park 1 543237.979633 6891289.330760 Herbert Park 2 542997.138556 6891254.724850 Herbert Park 3 543253.691349 6891030.915890 Johns Road 1 531669.381782 6891612.371050 Karawatha Forest Reserve 1 507727.607731 6942529.452500 Karawatha Forest Reserve 2 508728.667171 6943502.908220 Karawatha Forest Reserve 3 506785.617820 6943369.089470 Karawatha Forest Reserve 4 507927.601614 6944422.802160 Karawatha Forest Reserve 5 506744.915788 6944399.440000 Karawatha Forest Reserve 6 507751.468544 6943379.635020 Karawatha Forest Reserve 7 508775.185812 6944306.744620 Kincaid Park 1 532202.735559 6900780.800900 Kirken Road 1 532162.520897 6922919.666220 Lakala Street 1 538479.489186 6905674.962470 Miami Conservation Area 1 542168.767812 6894693.885270 Mt Nathan Conservation Area 1 526332.928915 6903271.678560 Mt Nathan Conservation Area 2 525847.763866 6904131.802350 Mt Nathan Conservation Area 3 525576.926372 6905136.382180

Page 216: A thesis submitted for the fulfilment for the requirements

194

Appendix 3: continued.

Site name Plot No: Point X Point Y Nerang Forest Reserve 1 532028.965643 6905420.456540 Nerang Forest Reserve 2 530263.009672 6905505.184520 Nerang Forest Reserve 3 529955.625783 6907263.104300 Nerang Forest Reserve 4 527720.308972 6905414.606190 Nerang Forest Reserve 5 528038.795086 6907334.221670 Numinbah Road 1 521672.286258 6883086.616550 Numinbah Road 2 521275.101355 6883727.335790 Olsen Avenue 1 537341.899703 6906520.653280 Olsen Avenue 2 537247.375321 6906106.382430 Pacific Highway, Currumbin 1 547483.722075 6887359.276130 Pacific Highway, Currumbin 2 547452.517681 6887686.574000 Pacific Pines Parklands 1 531600.463863 6909873.633300 Pimpama Conservation Area 1 535199.746344 6924216.326470 Pimpama Conservation Area 2 534347.183290 6924287.300750 Reedy Creek Conservation Area 1 537562.243502 6889388.059000 Reedy Creek Conservation Area 2 538085.796694 6888847.692140 Reedy Creek Conservation Area 3 538066.910600 6889393.403100 Reserve Road Parklands 1 528448.699157 6917116.209830 Stanmore Park 1 521764.900688 6930177.034770 Summerhill Court Reserve 1 535309.338108 6894408.814870 Syndicate Road 1 539047.034743 6883822.279950 Syndicate Road 2 538528.754842 6883205.819700 Syndicate Road 3 538218.744518 6882973.473620 Tallowwood Road 1 531120.025094 6886628.940760 The Plateau Reserve 1 523029.078759 6925041.471360 Tomewin Road 1 541158.644162 6881541.226480 Trees Road 1 540575.611833 6884050.718910 Trees Road 2 541002.687432 6884233.426600 Tugun Hill Conservation Area 1 547329.453671 6886315.700510 Upper Mudgeeraba Conservation Area 1 533680.658976 6890020.044500 Upper Mudgeeraba Conservation Area 2 534350.212959 6890505.841910 Venman Bushland National Park 1 519480.294031 6944925.190920 Venman Bushland National Park 2 519408.855771 6943972.463510 Venman Bushland National Park 3 520443.786257 6943875.546520 Venman Bushland National Park 4 520398.854195 6942926.637070 Venman Bushland National Park 5 519420.690871 6942938.186230 Venman Bushland National Park 6 518520.007062 6944989.216650 Wongawallen Conservation Area 1 523358.236927 6918341.414670 Wongawallen Conservation Area 2 525058.095543 6917380.770640 Woody Hill, Hart's Reserve 1 537191.808597 6896953.136380 Woody Hill, Hart's Reserve 2 536983.990140 6896764.950610

Page 217: A thesis submitted for the fulfilment for the requirements

195

Appendix 4: Cadastre, Infrastructure and Regional Ecosystem data used to calculate urbanisation gradient.

Site name Buffer area

(m) Cadastre

(%) Infrastructure

(%) Regional Ecosystem

(%) Andrews Reserve 560779.61 50.87 6.59 42.54 Aqua Promenade Reserve 806637.29 41.40 5.86 52.74 Austenville Road 690016.43 8.10 1.53 90.37 Behms Road 674678.36 82.96 8.14 8.90 Bonogin Conservation Area 5210152.79 24.76 6.20 69.04 Brushwood Ridge Parklands 984271.29 60.99 11.23 27.78 Burleigh Heads National Park 883416.40 28.64 57.46 13.90 Burleigh Knoll Conservation Park 405382.51 80.52 19.00 0.49 Burleigh Ridge Park 714007.17 56.53 29.90 13.57 Carrington Road 452760.66 24.31 7.76 67.92 Daisy Hill Conservation Area 2765383.93 30.27 2.82 66.91 Elanora Conservation Reserve 713406.75 34.87 4.91 60.22 Elanora Wetlands 926847.43 38.79 35.06 26.15 Galt Road Park 839228.49 39.22 0.33 60.46 Gilston Road 437194.44 31.84 4.27 63.88 Hellman Street 612001.21 41.21 26.29 32.50 Herbert Park 694176.02 46.12 17.66 36.22 Johns Road 407443.15 54.95 10.22 34.83 Karawatha Forest Reserve 4270380.63 53.91 22.70 23.38 Kincaid Park 534169.22 61.30 13.90 24.79 Kirken Road 1692360.99 69.70 5.85 24.45 Lakala Street Reserve 399861.10 43.39 23.29 33.31 Miami Bushland Conservation Park 424732.47 79.81 16.05 4.14 Mount Nathan 3486450.10 35.34 10.12 54.54 Nerang State Forest 6181506.57 36.43 9.25 54.32 Numinbah Road 1610309.71 34.39 2.62 63.00 Olsen Avenue 1019031.21 47.03 29.80 23.17 Pacific Highway 699636.88 54.93 20.84 24.24 Pacific Pines Parkland 538435.24 56.93 15.28 27.78 Piggabeen Road 382837.93 77.37 5.44 17.20 Pimpama Conservation Park 3726843.55 49.05 20.95 30.00 Reedy Creek 1007140.33 21.47 1.49 77.04 Reserve Road Parklands 511487.84 57.57 11.72 30.72 Stanmore Park 990738.01 35.07 11.04 53.89 Summerhill Court Reserve 507236.42 68.27 21.24 10.48 Syndicate Road 2347051.35 33.48 4.56 61.95 Tallowwood Road 599332.43 22.85 1.91 75.24 The Plateau Reserve 1250117.45 41.70 2.86 55.44 Tomewin road 494248.65 47.11 5.50 47.39 Trees Road Conservation Area 908002.79 23.13 2.70 74.18 Tugun Conservation Park 569598.42 48.50 37.52 13.98 Upper Mudgeeraba Conservation Area 1874057.58 16.25 4.08 79.68 Venman National Park 3488388.60 27.00 3.52 69.48 Wongawallen Conservation Area 3260992.65 19.03 2.25 78.73 Woody Hill, Hart's Reserve 1710937.66 94.67 2.05 3.28

Page 218: A thesis submitted for the fulfilment for the requirements

196

Appendix 5: Model selection Model Selection for Chapter 4: M1<-glmmML(density~gradient+connectivity+fire+logging+slope+elevation+aspect+stand_den,cluster

=site.no,family=poisson,data=env)

M2<-glmmML(density~connectivity,cluster=site.no,family=poisson,data=env)

M3<-glmmML(density~logging,cluster=site.no,family=poisson,data=env)

M4<-glmmML(density~slope,cluster=site.no,family=poisson,data=env)

M5<-glmmML(density~elevation,cluster=site.no,family=poisson,data=env)

M6<-glmmML(density~fire,cluster=site.no,family=poisson,data=env)

M7<-glmmML(density~aspect,cluster=site.no,family=poisson,data=env)

M8<-glmmML(density~stand_den,cluster=site.no,family=poisson,data=env)

M9<-glmmML(density~logging+elevation+stand_den,cluster=site.no,family=poisson,data=env)

M10<-glmmML(density~logging+elevation+fire,cluster=site.no,family=poisson,data=env)

M11<-glmmML(density~logging+fire,cluster=site.no,family=poisson,data=env)

M12<-glmmML(density~fire+logging+slope,cluster=site.no,family=poisson,data=env)

M13<-glmmML(density~logging+slope+elevation,cluster=site.no,family=poisson,data=env)

M14<-glmmML(density~slope+aspect+stand_den,cluster=site.no,family=poisson,data=env)

M15<-glmmML(density~gradient+connectivity+fire,cluster=site.no,family=poisson,data=env)

M16<-glmmML(density~gradient+connectivity+fire+logging,cluster=site.no,family=poisson,data=env)

M17<-glmmML(density~slope+elevation,cluster=site.no,family=poisson,data=env)

M18<-glmmML(density~logging+elevation,cluster=site.no,family=poisson,data=env)

M19<-glmmML(density~logging+aspect,cluster=site.no,family=poisson,data=env)

M20<-glmmML(density~logging+stand_den,cluster=site.no,family=poisson,data=env)

M21<-glmmML(density~elevation+stand_den,cluster=site.no,family=poisson,data=env)

M22<-glmmML(density~logging+slope,cluster=site.no,family=poisson,data=env)

M23<-glmmML(density~fire+logging,cluster=site.no,family=poisson,data=env)

M24<-glmmML(density~fire+stand_den,cluster=site.no,family=poisson,data=env)

M25<-glmmML(density~fire+elevation,cluster=site.no,family=poisson,data=env)

M26<-glmmML(density~fire+slope,cluster=site.no,family=poisson,data=env)

M27<-glmmML(density~fire+aspect,cluster=site.no,family=poisson,data=env)

M28<-glmmML(density~logging*elevation,cluster=site.no,family=poisson,data=env)

Page 219: A thesis submitted for the fulfilment for the requirements

197

Model Selection for Chapter 5: M1<-glmmML(hollows~dbh+species+fire+termites+epicorms+wdamage+slope+elevation+aspect,cluster=site,

family=binomial)

M2<-glmmML(hollows~dbh+species+fire+termites+epicorms+wdamage+slope+elevation,cluster=site,family=

binomial)

M3<-glmmML(hollows~dbh+species+fire+termites+epicorms+wdamage+slope+aspect,cluster=site,family

=binomial)

M4<-glmmML(hollows~dbh+species+fire+termites+epicorms+wdamage+elevation+aspect,cluster=site,family

=binomial)

M5<-glmmML(hollows~dbh+species+fire+termites+epicorms+slope+elevation+aspect,cluster=site,family

=binomial)

M6<-glmmML(hollows~dbh+species+fire+termites+wdamage+slope+elevation+aspect,cluster=site,family

=binomial)

M7<-glmmML(hollows~dbh+species+fire+epicorms+wdamage+slope+elevation+aspect,cluster=site,family

=binomial)

M8<-lmmML(hollows~dbh+species+termites+epicorms+wdamage+slope+elevation+aspect,cluster=site,family

=binomial)

M9<-glmmML(hollows~dbh+termites+epicorms+wdamage+slope+elevation+aspect,cluster=site,family=

binomial)

M10<-glmmML(hollows~species+fire+termies+epicorms+wdamage+slope+elevation+aspect,cluster=site,

family=binomial)

M11<-glmmML(hollows~dbh,cluster=site,family=binomial)

M12<-glmmML(hollows~species,cluster=site,family=binomial)

M13<-glmmML(hollows~fire,cluster=site,family=binomial)

M14<-glmmML(hollows~termites,cluster=site,family=binomial)

M15<-glmmML(hollows~epicorms,cluster=site,family=binomial)

M16<-glmmML(hollows~wdamage,cluster=site,family=binomial)

M17<-glmmML(hollows~slope,cluster=site,family=binomial)

M18<-glmmML(hollows~elevation,cluster=site,family=binomial)

M19<-glmmML(hollows~aspect,cluster=site,family=binomial)

M20<-glmmML(hollows~dbh+fire,cluster=site,family=binomial)

Page 220: A thesis submitted for the fulfilment for the requirements

198

M21<-glmmML(hollows~dbh+termites,cluster=site,family=binomial)

M22<-glmmML(hollows~dbh+epicorms,cluster=site,family=binomial)

M23<-glmmML(hollows~dbh+wdamage,cluster=site,family=binomial)

M24<-glmmML(hollows~dbh+slope,cluster=site,family=binomial)

M25<-glmmML(hollows~dbh+elevation,cluster=site,family=binomial)

M26<-glmmML(hollows~dbh+aspect,cluster=site,family=binomial)

M27<-glmmML(hollow~dbh+species,cluster=site,family=binomial)

Page 221: A thesis submitted for the fulfilment for the requirements

199

Model Selection for Chapter 6: M1<glmmML(species~urban+connect+slope+elevation+disturbance+trees+hbtden+stand_den+fire+logging+area,cluster=site,family=poisson)

M2<glmmML(species~urban+connect+slope+elevation+disturbance+trees+hbtden+stand_den+fire+logging,cluster=site,family=poisson)

M3<glmmML(species~urban+connect+slope+elevation+disturbance+trees+hbtden+stand_den+fire,cluster=site,family=poisson)

M4<glmmML(species~urban+connect+slope+elevation+disturbance+trees+hbtden+stand_den,cluster=site,family=poisson)

M5<glmmML(species~urban+connect+slope+elevation+disturbance+trees+hbtden,cluster=site,

family=poisson)

M6<glmmML(species~urban+connect+slope+elevation+disturbance+trees,cluster=site,family=

poisson)

M7<-glmmML(species~urban+connect+slope+elevation+disturbance,cluster=site,family=poisson)

M8<-glmmML(species~urban+connect+slope+elevation,cluster=site,family=poisson)

M9<-glmmML(species~urban+connect+slope,cluster=site,family=poisson)

M10<-glmmML(species~urban+connect,cluster=site,family=poisson)

M11<-glmmML(species~urban,cluster=site,family=poisson)

M12<-glmmML(species~connect,cluster=site,family=poisson)

M13<-glmmML(species~slope,cluster=site,family=poisson)

M14<-glmmML(species~elevation,cluster=site,family=poisson)

M15<-glmmML(species~disturbance,cluster=site,family=poisson)

M16<-glmmML(species~trees,cluster=site,family=poisson)

M17<-glmmML(species~hbtden,cluster=site,family=poisson)

M18<-glmmML(species~stand_den,cluster=site,family=poisson)

M19<-glmmML(species~fire,cluster=site,family=poisson)

M20<-glmmML(species~logging,cluster=site,family=poisson)

M21<-glmmML(species~area,cluster=site,family=poisson)

M22<-glmmML(species~area+trees,cluster=site,family=poisson)

M23<-glmmML(species~area+fire+trees,cluster=site,family=poisson)

M24<-glmmML(species~area+disturbance+stand_den,cluster=site,family=poisson)

M25<-glmmML(species~trees+fire+disturbance,cluster=site,family=poisson)

M26<-glmmML(species~fire+disturbance+stand_den,cluster=site,family=poisson)

Page 222: A thesis submitted for the fulfilment for the requirements

200

Appendix 6: Tree growth rates

Table 1: The predicted age of all tree species categorised by dbh. Species are further categorised as being either a recruitment tree (Rec) or

hollow-bearing tree (HBT).

Size class (dbh in cm)

Tessellated C. intermedia C.trachyphloia E. pilularis Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

10-20 38.49 109 1 110 44.19 221 6 227 32.79 15 0 15 24 27 1 28 21-30 76.98 110 5 115 88.38 257 4 261 65.57 18 0 18 36 55 2 57 31-40 117.67 54 7 61 131.79 128 13 141 103.55 1 1 2 42 33 1 34 41-50 164.15 23 5 28 190.84 64 9 73 137.45

0 0 46 38 3 41

51-60 225.52 7 8 15 251.09 30 14 44 199.95

1 1 61 10 4 14 61-70 263.24 3 4 7 293.19 6 6 12 233.28

0 78 18 3 21

71-80 317.72 4 2 6 376.52 4 5 9 258.92

0 92 4 4 8 81-90 383.55 1 2 3 473.70 3 2 5 293.40

0 109 2 2 4

91-100 458.74

1 1 584.71 1 1 2 332.77

0 135

2 2 101-110 543.30

1 1 709.57

1 1 377.04

0 155

2 2

111-120 638.74

851.48

1 1 426.20

175

4 4 121-130 737.50

1021.78

1 1 453.22

196

1 1

131-140 885.00

1226.14

0 0 543.87

218 141-150 1062.00

1471.36

0 0 652.64

245

151-160 1274.50

1765.64

1 1 783.17

270 160+ 1529.50 2118.76 939.80 283 Total 311 36 347 714 64 778 34 2 36 187 29 216

Page 223: A thesis submitted for the fulfilment for the requirements

201

Table 1 continued: The predicted age of all tree species categorised by dbh. Species are further categorised as being either a recruitment tree

(Rec) or hollow-bearing tree (HBT).

Size class (dbh in cm)

Half-bark Iron-bark E. fibrosa E. crebra Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

10-20 24 420 20 440 33.33 37 0 37 23.80 1 0 2 66.67 145 3 148 21-30 36 244 26 270 90.31 56 0 56 47.62 2 0 1 133.33 124 5 129 31-40 42 95 31 126 132.39 23 3 26 75.61

189.17 83 20 103

41-50 46 42 22 64 173.05 24 5 29 101.08

245.01 49 33 82 51-60 61 13 12 25 226.60 4 6 10 129.25

323.96 25 17 42

61-70 78 8 14 22 290.14 3 1 4 157.29

422.53 9 7 16 71-80 92 4 8 12 363.72 1

1 186.70

540.74 1 3 4

81-90 109 2 2 4 450.95

223.29

678.61

2 2 91-100 135 2 3 5 542.53

265.06

820.38

1 1

101-110 155

1 1 648.01

312.03

984.45 111-120 175

1 1 772.59

364.19

1181.25

121-130 196

2 2 919.27

421.55

1417.45 131-140 218

0 1080.04

459.08

1700.85

141-150 245

0 1295.95

550.90

2041.05 151-160 270

1555.04

661.08

2449.25

160+ 283 1866.14 793.29 2939.05 Total 830 142 972 148 15 163 3 0 3 436 91 527

Page 224: A thesis submitted for the fulfilment for the requirements

202

Table 1 continued: The predicted age of all tree species categorised by dbh. Species are further categorised as being either a recruitment tree

(Rec) or hollow-bearing tree (HBT).

Size class (dbh in cm)

Stringy-bark E. microcorys E. acmenoides C. citriodora Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

10-20 39.34 456 11 467 43 70 1 71 35.68 5 0 5 34.72 154 2 156 21-30 59.18 403 16 419 47 58 3 61 71.37 7 0 6 69.43 122 8 130 31-40 81.52 212 32 244 53 38 12 50 110.04 4 4 3 103.39 53 8 61 41-50 104.50 106 32 138 61 28 16 44 148.07 5 1 6 137.49 24 11 35 51-60 134.43 41 23 64 80 13 15 28 188.86 1 1 1 172.47 7 12 19 61-70 165.13 10 10 20 102.5 6 8 14 227.77

2

216.01 2 6 8

71-80 198.75 5 9 14 126.5 5 3 8 271.01

261.36 1 2 3 81-90 233.78 1 7 8 154 3 7 10 313.57

314.79

1 1

91-100 272.33 1 1 2 182.5 0 2 2 362.16

375.82

3 3 101-110 313.90 0 0 0 211 1 3 4 416.80

444.45

1 1

111-120 463.49 1 1 2 239.5

0 0 477.49

520.68 121-130 390.18

269.5

1 1 510.86

554.22

131-140 457.50

302

613.03

665.06 141-150 535.31

335

735.63

798.08

151-160 625.66

368.5

882.83

957.69 160+ 722.50 386 1059.00 1149.23

Total 1236 142 1378 222 71 293 22 8 21 363 54 417

Page 225: A thesis submitted for the fulfilment for the requirements

203

Table 1 continued: The predicted age of all tree species categorised by dbh. Species are further categorised as being either a recruitment tree

(Rec) or hollow-bearing tree (HBT).

Size class (dbh in cm)

Smooth-bark E. propinqua E. racemosa Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

Age (yrs) n Rec n HBT N

10-20 34.72 40 0 40 38.46 156 3 159 49 13 0 13 21-30 69.43 51 0 51 76.92 129 9 138 68.6 25 2 27 31-40 103.39 33 5 38 116.53 72 17 89 72 15 6 21 41-50 137.49 18 4 22 154.99 27 21 48 82 12 9 21 51-60 172.47 6 4 10 198.94 7 14 21 109 8 9 17 61-70 216.01 2 3 5 231.73 5 5 10 142.5 1 3 4 71-80 261.36 0 1 1 273.40 0 3 3 177.5 3 1 4 81-90 314.79 1 1 2 321.96 1 5 6 215.5 1 1 2 91-100 375.82

2 2 377.42

1 1 255.5

2 2

101-110 444.45

1 1 439.78

0 0 295.5

2 2 111-120 520.68

1 1 453.63

0 0 335.5

0 0

121-130 554.22

543.60

1 1 356

1 1 131-140 665.06

653.20

1 1 427.2

141-150 798.08

738.88

512.6 151-160 957.69

940.65

615.2

160+ 1149.23 1128.78

738.2 Total 151 22 173 397 80 477 78 36 114

Page 226: A thesis submitted for the fulfilment for the requirements

204