Can Legislation Deliver Conservation?: an Assessment of The

University of Wollongong Research Online

University of Wollongong Thesis Collection University of Wollongong Thesis Collections

2002 Can legislation deliver conservation?: an assessment of the Threatened Species Conservation Act 1995 (NSW) using two threatened plant species as case studies Claire Louise Brown University of Wollongong

Recommended Citation Brown, Claire Louise, Can legislation deliver conservation?: an assessment of the Threatened Species Conservation Act 1995 (NSW) using two threatened plant species as case studies, Doctor of Philosophy thesis, Department of Biological Sciences, University of Wollongong, 2002. http://ro.uow.edu.au/theses/1038

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: [email protected]

CAN LEGISLATION DELIVER CONSERVATION?

An assessment of the Threatened Species Conservation Act 1995 (NSW) using two threatened plant species as case studies.

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

Doctor of Philosophy

from the

UNIVERSITY OF WOLLONGONG

by

CIAIRE LOUISE BROWN, B.Sc. (HONS)-B.A.

DEPARTMENT OF BIOLOGICAL SCIENCES and FACULTY OF LAW November 2002 DECLARATION

I, Claire Louise Brown, declare that this thesis, submitted in fulfilment of the requirements for the award of Doctor of Philosophy, in the Department of Biological Sciences and the Faculty of Law, University of Wollongong, is wholly my own work unless otherwise referenced or acknowledged. The document has not been submitted for qualifications at any other academic institution.

Claire L. Brown

11th November 2002 Table of Contents

TABLE OF CONTENTS

page no.

List of Tables I List of Figures IV List of Plates (Photographs) Vlll Abstract x Acknowledgments XIV

SECTION A

Chapter 1: General Introduction

1.1 Biodiversity 1 1.1.1 The Impact of Agriculture 3 1.1.2 The Impact of Urban Development 3 1.1.3 Threatened species in New South Wales 6 1.1.4 Overall review of conservation paradigms 7 Vegetation Corridors 9 Translocation of Species 11 1.2 Policy, legislation and management instruments 12 1.2.1 The role of local government 13 1.3 Importance of the science/law interface 15 1.3.1 The role of scientific uncertainty and the precautionary principle 16 1.4 Specific Aims 21 1.5 General outline of methodology for the study as a whole 21

SECTIONB

Chapter 2: Legislating for biodiversity using the Threatened Species Conservation Act 1995 (NSW)

2.1 Introduction 25 2.2 Legislating for Conservation in New South Wales 26 2.3 The Threatened Species Conservation Act 1995 (NSW) and Environmental Planning andAssessmentAct 1979 (NSW) Part 4 32 2.3.1 Listing 32 2.3.2 Eight Part Test 35 2.3.3 Species Impact Statements 36 2.3.4 Other elements of the legislation 40 2.4 Two case studies 41 2.4.1 Eight part test (Green Road, Glenhaven) 42 2.4.2 Species Impact Statement (Aquatic Drive, Allambie Heights) 42 2.5 Conclusions 43

Chapter 3: Case Studies

3.1 Introduction 44 Methodology 49 3.2 Aquatic Drive, Allambie Heights (33Q40'30"S 151 Ql4'20"E) 50 Table of Contents

3.2.1 Vegetation attributes of the site 50 3.2.2 The development 54 3.2.3 The issues 55 Identification and presence of Tetratheca glandulosa 55 Ardel Limited v Warringah Council (No. 10606of1994 NSWLEC) 55 Hassell Pty Ltd v Warringah Council (41NSWLEC1998) 58 Hassell Pty Ltd v Warringah Shire Council (49 NSWLEC 2000) 60 Consultation with Council and with National Parks and Wildlife Service 60 Monitoring programme 62 3.2.3 Analysis 62 3.2.4 Outcomes 64 3.3 Green Road, Glenhaven (33Q42'S, 150Q157'E) 64 3.3.1 Vegetation attributes of the site 64 3.3.2 The issues 67 Eight part test and Species Impact Statement 67 National Parks and Wildlife Service concurrence and Council consultation 67 Development Controls 69 Monitoring Programme 69 3.3.3 Analysis 69 3.4 Carnarvon Drive, Frenches Forest 72 3.4.1 Vegetation attributes of the site 72 3.4.2 The development 74 3.43 The issues 74 Species Impact Statement 74 National Parks and Wildlife Service concurrence and Council consultation 77 Development Controls/Conditions of Consent 78 Monitoring Programme 78 3.4.3 Analysis 78 3.5 Grosvenor Street, North Wahroonga (33Q42'S, 151Q07'30"E) 79 3 .5 .1 Vegetation attributes of the site 81 3.5.2 The development 82 3.5.3 The issues 82 Zoning issues 82 Peter William Lean v Ku-Ring-Gai Council (No. 10457of1996 and 40265 of 1996) 82 3.5.3 Analysis 85

SECTIONC

Chapter 4: Species Biology And Site Descriptions

4.1 Introduction 87 4.2 Tetratheca glandulosa Smith 89 4.2.1 Taxonomic Description 89 4.2.2 Recovery Plan Status 94 4.3 Darwinia biflora (Cheel) B. Briggs 94 4.3.1 Taxonomic description 94 Table of Contents

4.3.2 Recovery plan status 98 4.4 Site descriptions 98

Chapter 5: Pollination Biology of Tetratheca glandulosa and Darwinia biflora

5.1 Introduction 106 5.2Aims 110 5.3 Materials and Methods 110 5.3.1 Determination of the breeding system 110 5.3.2 Seed fitness (weight and length) 113 5.3.3 Pollinator identification 113 5.3.4 Test for clonality in Tetratheca glandulosa 115 5.4 Results 117 5.4.1 Determination of the breeding system 117 Percentage offlowers setting viable fruit 117 Seed development 119 5.4.2 Seed length and weight 123 5.4.3 Pollinator identification 125 5.4.4 Test for clonality in Tetratheca glandulosa 125 5.5 Discussion and Conclusions 127

Chapter 6: Pollination Success of Tetratheca glandulosa and Darwinia biflora. Association with fragmentation.

6.1 Introduction 131 6.2Aims 134 6.3 Materials and methods 134 6.3.1 Flowering and fruiting densities 134 6.3.2 Pollinator behaviour and pollen removal 135 Pollinator visits 135 Pollen removal 136 6.4 Results 137 6.4.1 Flowering and fruit densities 137 Flower intensities 137 Fruit."flower ratio 139 6.4.2 Insect visitation and pollen removal 142 Visits 142 Pollen removal 144 6.5 Discussion and Conclusions 144

Chapter 7: The viable seed store for Tetratheca glandulosa and Darwinia biflora (The hidden science).

7.1 Introduction 150 7.2Aims 151 7.3 Materials and Methods 152 7.3.1 Seeds of Tetratheca glandulosa found in the soil 152 7.3.2 Estimation of annual inputs to the seed banks 154 7.3.3 Seed germination 156 7.3.4 Seed dormancy 157 7.4 Results 160 Table of Contents

7.4.1 Seeds of Tetratheca glandulosa found in the soil 160 7.4.2 Estimation of annual inputs to the seed banks 161 7.4.3 Seed germination 161 7.4.4 Seed dormancy 164 7.4 Discussion and Conclusions 166

SECTIOND

Chapter 8: General discussion

8.1 Introduction 170 8.2 Summary of thesis 171 8.3 When scientific uncertainty exists and how is it dealt with? 174 8.3.1 Tetratheca glandulosa (Aquatic Drive, Allambie Heights) 174 8.3.2 Darwinia bif/,ora (Green Road, Glenhaven) 175 8.4 How could the legislation be improved? 177 8.4.1 Recovery planning 179 8.4.2 Incorporating the precautionary principle into decision-making 184 8.4.3 Adaptive management 185 8.4.4 Improved pathways for decision-making in Local Councils 188

References 192

Appendices

Appendix 2.1 History of the Threatened Species Conservation Act 1995 (NSW) 219

Appendix 2.2 Contents of recovery plans 222

Appendix 5.1 Sonication (buzz) pollination 223

Appendix 5.2 Pollen library 225

Appendix 8.1 An examination of recovery plans which have been written for listed threatened species 226 List of Tables I

LIST OF TABLES

page no. Table 1.1: Summary of the number of species listed on the Threatened Species ConservationAct 1995 (NSW). 6

Table 1.2: Summary of the parks system in N.S.W., showing the different categories. 8

Table 1.3: History of wildlife conservation state legislation in New South Wales 14

Table 1.4: There are many different classifications of scientific uncertainty and the sources it originates from (Source Deville & Harding 1997). 20

Table 1.5: Summary of specific questions in the study and how each question was approached. 22

Table 2.1: An eight part test determines if a significant effect will occur from a proposed development or activity upon a listed species. 37

Table 2.2: Species Impact Statements provide a site assessment for a development or activity in relation to threatened species. 39

Table 3.1: Ecological communities represented on the site found at Aquatic Drive, Allambie Heights. 53

Table 3.2: Summary of the SIS and outcomes for the population of Tetratheca glandulosa at Aquatic Drive, Allambie Heights. 61

Table 3.3: Summary of the Eight Part Test and outcomes for the population of Darwinia biflora at Green Road, Glenhaven. 68

Table 4.1: There are summary of characteristics of the threatened plant species in the Sydney region, which have been considered during the decisions on development consent. 88

Table 4.2: Known and unknown ecological attributes of and threats to Tetratheca glandulosa. 95

Table 4.3: Known and unknown ecological attributes of and threats to Darwinia biflora. 99

Table 4.4: Descriptions, population size and management options for sites that are conserved in National Parks or Council Reserves. 101

Table 5.1: Selected studies between 1986-2000 (taken from a variety of sources), showing a variety of preferred mating systems found in Australian flora. 107 List of Tables n

Table 5.2: The number of individual plants used at each site for Tetratheca glandulosa and Darwinia biffora for pollen supplementation and pollinator exclusion experiments. 111

Table 5.3: Table showing where donor pollen for Tetratheca glandulosa came for outcross-pollination experiments at each site. 112

Table 5.4: Primer pairs that were tested. Asterisk (*) indicates the pair that was eventually used for analysis. 116

Table 5.5: Two-factor ANOVA for seed weight for three sites across two treatments (open- and cross-pollination) for Tetratheca glandulosa (significance p<0.05). 123

Table 5.6: Two-factor ANOVA for seed length at three different sites across two treatments (open- and cross-pollination) for Tetratheca glandulosa (significance p<0.05). 123

Table 5.7: Two-factor ANOVA for seed weight at four different sites across three treatments (open-, self- and cross-pollination) for Darwinia biflora (significance p<0.05). 125

Table 5.8: Two-factor ANOVA for seed length at four different sites across three treatments (open-, self- and cross-pollination) for Darwinia biflora (significance p<0.05). 125

Table 6.1: Summary of a two-factor ANOVA for flowering densities at five sites across two years (1999 and 2000) for Tetratheca glandulosa. 139

Table 6.2: Summary of a two-factor ANOVA for flowering densities at four sites across two years (1999 and 2000) for Darwinia biflora. 139

Table 6.3: Summary of a two-factor ANOVA for the mean proportion of flowers setting fruit per plant for two years (1999 and 2000) for Tetratheca glandulosa at three sites. 141

Table 6.4: Summary of a two-factor ANOVA for the mean proportion of flowers setting fruit per plant for two years (1999 and 2000) for Darwinia biflora at three sites. 141

Table 6.5: Summary of a two-factor ANOVA for the mean number of visits by insects per flower per plant per hour of observation for Tetratheca glandulosa over two years (1999, 2000). 142

Table 6.6: Summary of a two-factor ANOVA for the mean number of visits by insects per flower per plant per hour of observation for Darwinia biflora over two years (1999, 2000). 144 List of Tables m

Table 7.1: Summary of a two-factor ANOVA for seed germination three sites across two treatments (open and cross pollination) for Tetratheca glandulosa. 164

Table 7.2: Summary of a two-factor ANOVA for seed germination at three sites across three treatments (open, self, cross pollination) for Darwinia biflora. 164

Table 7.3 Two-factor ANOVA for germination at four sites for two treatments (scarified and unscarified) for Tetratheca glandulosa. 166

Table 7.4 Two-factor ANOVA for germination at three sites for two treatments (scarified and unscarified) for Darwinia biflora. 166

Table 8.1: Summary of the specific questions in the study and a brief summation of the answers and/or outcomes obtained. 172

Table 8.2: A revised summary of the SIS and outcomes for the population of Tetratheca glandulosa at Aquatic Drive, Allambie Heights. 176

Table 8.3: A revised summary of the eight-part test and outcomes for the population of Darwinia biflora, at Green Road Glenhaven. 178 List of Figure lV

LIST OF FIGURES

page no. Figure 1.1: Schematic diagram displaying how a detailed examination of the Threatened Species Conservation Act 1995 (NSW) was carried out. 23

Figure 1.4: Schematic diagram showing methodology for how case studies were approached. 24

Figure 2.1: Pathways for gaining development consent in New South Wales under the Environmental Planning and Assessment Act 1979 (NSW). 28

Figure 2.2: Obtaining development consent under Part 4 of the Environmental Planning and Assessment Act 1979 (NSW). 30

Figure 3.1: Flow chart showing an idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part). 45

Figure 3.2: The location of sites discussed in the case studies. 48

Figure 3.3: Approximate location of Tetratheca glandulosa (shaded) on the ARDEL site at Aquatic Drive, Allambie Heights (adapted from Smith & Smith 1996). 52 Figure 3.4: Flow chart showing the decision-making process for housing sub­ division at Aquatic Drive, Allambie Heights (Warringah Council) (Hassell Pty Ltd v Warringah Council). 56

Figure 3.5: Approximate location of Darwinia biflora on the site at Green Road, Kellyville (Adapted from the Draft Local Environment Plan 1991). 66

Figure 3.6: Flow chart showing the decision-making process for a subdivision at Green Road, Glenhaven. 71

Figure 3.7: Approximate location of Tetratheca glandulosa at Carnarvon Drive, Frenches Forest (adapted from development plan). 73

Figure 3.8: Flow chart showing the decision-making process for a playing field at Carnarvon Drive, Frenches Forest (Warringah Council). 76

Figure 3.9: Flow chart showing the decision-making process for a housing sub- division at Grosvenor St, North Wahroonga (Ku-Ring-Gai Municipal Council). 83

Figure 4.1: Approximate known distribution of Tetratheca glandulosa. 91

Figure 4.2: Approximate known distribution of Darwinia biflora. 97 List of Figure v

Figure 4.3: Location of sites that are conserved in National Parks or Council Reserves. 100

Figure 5.1: The percentage of flowers setting fruit for Tetratheca glandulosa in 1999 (a) and 2000 (b) and the percentage of the fruit which was viable in 1999 (c) and 2000 (d) at five different sites where populations are different sizes. 118

Figure 5.2: The percentage of flowers setting fruit for Darwinia biflora in (a) 1999 and (b) 2000 and the percentage of the fruit which was viable in (c) 1999 and (d) 2000 at four different sites where populations are different sizes. 120

Figure 5.3: The mean percentage of seed (averaged over plants for Tetratheca glandulosa that were aborted following pollination ( • ), eaten by insects before dispersal (1§1), viable whole seed (o) and seed that was whole but not viable '1). 121

Figure 5.4: The mean percentage of seed (averaged over plants) for Darwinia biflora that were aborted following pollination ( • ), eaten by insects before dispersal (~),viable whole seed (o) and seed that was whole but not viable (0). 122

Figure 5.5: The mean weights (gm) of individual seeds of (a) Tetratheca glandulosa and (b) Darwinia biflora and the mean length (mm) of an individual seed for (c) T. glandulosa and (d)D biflora. 124

Figure 5.6: A PCA plot for repeat AFLP profiles for Tetratheca glandulosa. 126

Figure 5.7: A PCA plot showing AFLP profile data for Tetratheca glandulosa for four populations (ARDEL, Cliff Oval, KCNP and Marramarra NP). 126

Figure5.8: Frequency distribution of the similarity index calculated for pairs of individual plants for three different populations (ARDEL, KCNP and Marramarra N.P.) for Tetratheca glandulosa. 128

Figure 6.1: Floral structure of Darwinia biflora, showing position of the oily pollen drop attached to the anthers. 137

Figure 6.2: The mean number of flowers per plant for Tetratheca glandulosa at five sites where populations were different sizes. 138

Figure 6.3: The mean number of flowers per plant for Darwinia biflora at four sites. 138

Figure 6.4: Mean proportion of flowers setting fruit per plant at five different sites over two years, 1999 (•)and 2000 (o) for Tetratheca glandulosa. 140

Figure 6.5: Mean proportion of flowers setting fruit per plant at four different sites over two years, 1999 (•)and 2000 (o) for Darwinia biflora. 140 List ofFigure v1

Figure 6.6: Mean number of visits by insects per a flower per plant for an hour of observation for populations of Tetratheca glandulosa at five sites. 143

Figure 6. 7: Mean number of visits by insects per a flower per plant for an hour of observation for populations of Darwinia biflora at four sites 143

Figure 6.8: Pollen removal from anthers of Tetratheca glandulosa at five different sites during a one hour period in 2000. 145

Figure 6.9: Pollen removal from anthers of Darwinia biflora at four different sites during a one hour period in 2000. 145

Figure 7.1: Soil cores (ten) where taken from sites and then sieved twice before being sorted by hand in a soil tray for seeds. 153

Figure 7.2: The viable seed store for any plant species can be estimated by using five different parameters (flower development, fruit development, seed development, impact of pre-dispersal seed predation and seed viability). 155

Figure 7.3: To gain the highest amount of seed germination, without knowing the germination cue, this hierarchical methodology was established for treatment of seeds. 157

Figure 7 .4: The mean number of seeds of Tetratheca glandulosa found per core at five different sites in 2000. 160

Figure 7.5: The modelled viable seed store per plant for two different years 1999 (•)and 2000 (o) for Tetratheca glandulosa at five different sites. 162

Figure 7 .6: The modelled viable seed store per plant for two different years 1999(•) and 2000 (o) for Darwinia biflora at four different sites. 162

Figure 7.7: The percentage of seed germinating of Tetratheca glandulosa at five sites where populations are different sizes. 163

Figure 7.8: The percentage of seed germinating of Darwinia biflora at four sites where populations are different sizes. 163

Figure 7.9: The mean germination for seed of Tetratheca glandulosa for two different treatments, scarified seed coat(•) and unscarified seed coat (o). 165

Figure 7.10: The mean germination of seed of Darwinia bifiora for two different treatments, scarified seed coat(•) and unscarified seed coat (o). 165

Figure 8.1: Generic pathway for decision-making when completing and assessing eight part tests, including when scientific evidence is required. 186 List of Figure vn

Figure 8.2: Generic pathway for decision-making when completing and assessing Species Impact Statements, including when scientific evidence is required. 187

Figure 8.3: A flow chart demonstrating potential paths for deciding the importance of a population of Tetratheca glandulosa within the decision-making process of development application. 189

Figure 8.4: Proposed alternative model for the Threatened Species Conservation Act 1995 (NSW). 191 List of Plates Vlll

LIST OF PLATES (PHOTOGRAPHS)

page no. Plate 3.1: Site at Aquatic Drive, Allambie Heights owned by ARDEL Limited. 51

Plate 3.2: Population of Tetratheca glandulosa at Aquatic Drive Allambie Heights (identified by pink flowers). 51 Plate 3.3: The site at Aquatic Drive, Allambie Heights was cleared for development. Population of Tetratheca glandulosa occurs in vegetation on the left of the photograph. 65

Plate 3.4: During clearing of the site at Aquatic drive, Allambie Heights, between 10-20% of the population of Tetratheca glandulosa was destroyed. 65

Plate 3.5 (above): A fence was placed around an area where a population of Darwinia biflora occurs at Green Road, Glenhaven. 70

Plate 3.6 (left): A bushland corridor was included in the development controls for Green Road, Glenhaven. 70

Plate3.7: The site at Carnarvon Drive, Frenches Forest was zones open space due to the powerlines that crossed the site. 75

Plate 3.8: Carnarvon Drive, Frenches Forest. 75

Plate 3.9: Grosvenor Street, North Wahroonga. 80

Plate 3.10: Grosvenor Street, North Wahroonga. 80

Plate 4.1: Tetratheca glandulosa is a small spreading shrub growing to between 10 to SOcm (plant can be identified by the mass of pink flowers). 90

Plate 4.2: Tetratheca glandulosa produces solitary flowers that are deep lilac to pink in colour. 90

Plate 4.3 (a): The persisting sepals distinguish Fruit produced by Tetratheca glandulosa. Open fruit are also shown on the right hand side of the photograph. 93

Plate 4.3 (b): A developed Tetratheca glandulosa fruit surrounded by persisting sepals (photograph taken under the microscope). 93

Plate 4.4: Darwinia biflora grows in an erect or spreading habit 96

Plate 4.5: Flowers produced by Darwinia biflora are .small, green in colour and surrounded by red bracteoles. 96

Plate 4.6 (left): Cobah Track, Marramarra National Park. 102 List of Plates ix

Plate 4.7 (below): Long Track, Ku-Ring-Gai Chase National Park. 102

Plate 4.8: Bobbin Head Track, Ku-Ring-Gai Chase National Park. 103

Plate 4.9: Murra Trail, Ku-Ring-Gai Chase National Park. 103

Plate 4.10: Tumbledown Dick Hill, Terry Hills. 104

Plate 4.11 (above): Cliff Oval, North Wahroonga. 105

Plate 4.12 (left): Cliff Oval, North Wahroonga following management burn in 1999. 105

Plate 5.1: A 'sticky' (insect) trap used to capture potential pollinators at an individual of Tetratheca glandulosa. 114

Plate 5.2: A microscope slide covered in 'bird-off' and attached to a bamboo pole to form an insect trap to capture potential pollinators at an individual of Tetratheca glandulosa. 114

Plate 7.1: Tetratheca glandulosa seeds have a chalazal, which is an elaiosome-like appendage indicated by the arrow. 158

Plate 7.2: Seeds were germinated in petri dishes on cotton wool and filter paper. 158 Abstract x

.ABSTRACT

Maintaining biodiversity is a basic theme in conservation in Australia and around the world. Biodiversity is defined to encompass the variation and abundance of species, genes, populations and ecosystems/habitats and is fundamental to maintaining many ecological processes. Human activities have caused loss of habitat and fragmentation of the remaining habitat, ultimately reducing biodiversity. Isolated patches of remnant bushland often form the last refuges for threatened species.

Attempts to conserve biodiversity have taken many directions including, in most countries, the use of a reserve system. However, it is unlikely that a reserve system can, by itself, achieve conservation and therefore other strategies must be employed. Legislation has taken a prominent role in the protection of biodiversity, mostly focusing on threatened species. The nature of such legislation in N.S .W. has varied over the last century, ultimately leading to the current Threatened Species Conservation Act 1995 (NSW) (TSCA) which is tightly intertwined with planning legislation, through the Environmental Planning and Assessment Act 1979 (NSW) (EPAA). This thesis focuses on the ability of the TSCA and EPAA Part 4, as they relate to development applications made to local councils, to achieve effective conservation.

There were four main components to the study. First, I critically reviewed the relevant parts of the legislation to assess the nature of the demands on scientific knowledge made by the legislation, second, I use two threatened species: Tetratheca glandulosa and Darwinia biflora as case studies to assess how local councils applied the TSCA and the EP AA. Thirdly, I conducted ecological field studies of the pollination ecology and reproductive success of T. glandulosa and D. bifl.ora: to determine how more complete scientific knowledge might alter decision-making. Finally, I used the integration of the scientific studies and the review of the legislation to identify ways in which legislation, and it application by decision-makers, could be altered so as to improve conservation outcomes. Abstract Xl

Many different interest groups have seen the TSCA (in its interrelationship with development assessment and control under the EP AA Part 4), as a fundamental step towards threatened species conservation and therefore biodiversity conservation. In fact, the TSCA is a mixture of proactive and reactive approaches to conservation. It has four main mechanisms; (i) listing, (ii) the so-called "eight part test" (determining if an action has a significant impact upon a listed species in order to determine whether an SIS and the concurrence of NPWS is required), (iii) the species impact statement (SIS), and (iv) the recovery plan. Listing and recovery planning occur under the TSCA and eight part tests, SISs and concurrence by National Parks and Wildlife Service are required under the EPAA. The legislation has attracted many criticisms, a primary one being that it offers inadequate protection to biodiversity and threatened species, because it assumes that comprehensive scientific evidence exists or can readily be obtained.

Four case studies were used to examine the existing framework, as a basis for proposing an ideal framework. First, a housing subdivision at Aquatic Drive, Allambie Heights was the centre of three decisions by the Land and Environment Court. At one stage, Tetratheca glandulosa was mis-identified. This case study highlighted the fact that scientific uncertainty can surround even the simple question of whether a species occurs on a site, and the importance of completing field surveys at the appropriate time of year. Second, a subdivision at Green Road, Glenhaven demonstrated how an amelioration process can be built into the eight-part test to ensure a smooth passage for a development application, despite the fact that the amelioration was based on questioned theory and lacked any empirical evidence. Third, the role of the NPWS · and the acceptance of scientific uncertainty were highlighted in the proposed council development at Camarvon Drive, Frenches Forest. Illustrating the responsibilities of council in deciding whether a development application should be approved. Fourth, a comprehensive debate between experts on the existence of a possible seed bank characterises the case study based around a proposed housing development at Grosvenor Street, North W ahroonga. This illustrates the potential ecological importance of a component of a plant's life-history that can not be readily detected. Abstract Xll

Ecological data were gathered to test the appropriateness of decisions on the two species: Tetratheca glandulosa and Darwinia bifiora, given the original paucity of knowledge. These species occur in a range of sites: some large, protected populations in National Parks and some small, more isolated, disturbed populations in land that has been developed or was the subject of a development application. I examined pollination biology, pollination success and viable seed stores. To examine the pollination biology of both species a bagging and cross-pollination experiment was carried out. The fitness of the resulting seed was inferred from seed weight and length. Potential pollinators were identified using a trapping regime and potential clonality of Tetratheca glandulosa was tested using Amplified Fragment Length Polymorphisms (AFLP) markers. The pollination success for T. glandulosa and D. bifiora at different sites was examined using flowering and fruiting densities, pollinator behaviour and pollen removal rates. The viable seed store was assessed by looking for seeds in the soil seed bank, estimating the annual input into the seed store and testing for seed germination and dormancy.

Tetratheca glandulosa displayed a strong preference for out-crossed pollen, while Darwinia bifiora exhibited self-compatibility, but with a preference for out-crossed pollen. No difference was found in seed weight or length for either species across sites or across pollination treatments. No pollinators were identified for either species. Genetic analysis using AFLP markers suggested that T. glandulosa is potentially clonal but results were inconclusive.

Fragmentation of habitats is assumed to be detrimental to the life cycles of many species through edge effects, invasion of non-native species, disease and interruptions to gene flow. Such impacts have led to what is known as a "pollinator crisis scenario", which results in the loss of pollinator guilds or communities through habitat alteration, invasive species and pesticides. Individuals of Tetratheca glandulosa in the large National Park population produced more flowers but less fruit than those in small, disturbed sites. There was no fruit set or pollen removal from flowers in the two smaller, isolated sites. In contrast, flower production and fruit set, pollinator visitation and pollen removal did not vary significantly among sites for D. bifiora. Seed banks of both T. glandulosa and D. biflora showed very Abstract Xlll

little or no seed in the soil. Seeds were easy to germinate (little dormancy), suggesting short-lived seed banks. Therefore, neither of these species could depend on the seed bank for survival through a sequence of disturbances. Maintaining reproductive success will require the pollination processes are sustained.

I conclude that there are significant problems with the present approach taken in the legislation and its implementation. These include: (i) lack of ecological knowledge, (ii) inadequate use of ecological knowledge during the decision-making process and (iii) the lack of a formal pathway for incorporating the precautionary principle and adaptive management. I propose a more appropriate model, which includes (i) incorporation into decision-making processes of more basic survey work and ecological experiments, (ii) incorporation of recovery plans into Local Environmental Plans and Regional Environmental Plans, (iii) integration of recovery planning into the SIS process, and (iv) the introduction of a peer review system to reduce inconsistencies in the interpretation of scientific knowledge and theory.

The ideal decision-making framework proposed consists of six elements; (1) the formal production of a development application, (2) consultation with local councils and independent scientists, (3) an accredited peer review system, ( 4) formal and informal roles for the N.S.W. National Parks Wildlife Service, (5) the use of development conditions, development modifications and adaptive management and ( 6) appeal to the Land and Environment Court. Acknowledgments XIV

ACKNOWLEDGMENTS

I begin by thanking my supervisors Rob Whelan and David Farrier for their advice, support, wisdom and patience at all stages of this study. A University of Wollongong Postgraduate Award supported this work. Further support was received from the Linnean Society of New South Wales via the Joyce W. Vickery Research Fund.

I also thank NPWS for allowing access to Ku-Ring-Gai Chase National Park and Marramarra National Park and other information. Jennie Powell (from Warringah Council), and Ku-Ring-Gai Municipal Council helped with access documents and access to sites. Access was gained to the site at Aquatic drive, Allambie Heights thanks to ARDEL Limited. A special thanks to Peter Myerscough how helped with the location of sites of Darwinia biflora in Ku-Ring-Gai Chase National Park and Chris Lacey (NPWS) for location of sites for Tetratheca glandulosa.

I thank David Ayre and Annette Usher for help in the lab and analysis of my genetic studies, Lou Rodgerson, and Mike Dunlop for help with analysing some of the data sets. Dr Chandra Gulati (UOW Statistical Consulting Service) helped with three-dimensional chi-squared tests.

To members of Rob's ecology lab, a special thanks for helping to provide the perfect academic environment for the completion of this study. Other members of the Institute for Conservation Biology and Law provided helpful comments on the study at various stages. Thanks to Dr. Adrienne Nicotra and members of the CROSAP lab at ANU for supplying a safe haven during 2001 and 2002.

A big thanks to family and friends who have supported me over the last couple of years. The encouragement and laughter has been important in aiding me to complete this study. And lastly, the endless love and support given by Stephen Humphrey should be acknowledged including all those endless hours in the field patiently taking notes for me, commenting on drafts, fixing computer problems and all those endless domestic duties. Chapter 1 General Introduction 1

CHAPTER 1 GENERAL INTRODUCTION

With the threat of Australia's sixth mass extinction, conservation has been fast gaining prominence within both scientific and public arenas. There have been calls for the legal system to cooperate. Environmental legislation is often viewed as being constructed and interwoven from social, economic and political histories (Dovers 1994). Intriguingly science is never considered to have contributed significantly to its development. The application of environmental legislation has had to struggle with the scanty knowledge base of science, which consists often of broad theories and lists of "possible outcomes", rather than detailed species-specific and site-specific predictions. Ecologists involved in environmental assessment have had to confront unrealistic expectations in law and policy partly because they have been removed from the decision making process or have become reluctant to become involved. This chapter establishes the background knowledge, sets out certain background information, explains the importance of a study examining how science and the law interact, and canvasses the specific questions asked during the study.

1.1 Biodiversity

Biodiversity and biological diversity are terms that are used freely in science, political and public arenas around the world, providing a focal point for discourse relating to conservation. The concept of biodiversity revolves around variation in the entirety of

1 life on the planet , with differing dimensions and levels (Gaston 1996; Powledge 1998). These levels are often defined m terms of species, genes and ecosystem/habitat/communities diversity m a given area, region, country or even globally (Randell 1991; Blay & Piotrowicz 1993; Franklin 1993; Dixon 1994; Risser 1995; Gaston 1996; Mamouney 2000). It is difficult to formulate a working definition2

1 First coined by E.O. Wilson, in the 1980s

2 The first formal definition came from the Office of Technology Assessment in the U.S.A. in 1987, and states that biodiversity is "the variety and variability among living organisms and the ecological complexes in which they occur". In addition, because items are organised at many [biological] levels, biodiversity "encompasses different ecosystems, species, genes and their relative abundance". Later discussion moved to include genes as an important component (Angermeier and Karr 1994). Chapter 1 General Introduction 2

of biodiversity, as there is the potential for the concept to embody such a breadth of knowledge (and interdisciplinary study) (Gaston 1996). Hence biodiversity is a complex idea involving issues that are moral, philosophical, spiritual, environmental, scientific, political, legal and economic (Bradsen 1992). All of these issues (e.g. a nation's right to exploit biodiversity (Bradsen 1992)) are driven by a range of concerns but most importantly the increase in species extinction and habitat destruction. In other words biodiversity is not simply a scientific concept (Lunney et al. 1998). In this thesis a species approach has been taken when discussing biodiversity.

The full array of the world's biodiversity will never be known (Gaston 1996). However, it has been estimated that the world supports well over six million different species, of which approximately seven percent are found in Australia. Of these, 80% are endemic to this continent. European settlements, with the development of agriculture and increased urban spread in Australia have had dramatic impacts upon biodiversity, demonstrated in the marked decline of biodiversity since 1788 (ANZECC 2001). Australia has one of the worst records for mammal extinction rates in the world (over the past 200 years) and almost a quarter of the nation's 22,000 plant species are now classified as threatened (Allan 1995). Pressure to conserve biodiversity has come from both the domestic and international arenas in light of the increased species extinction rates (Dawson 1996). A number of strategies have been developed internationally and applied locally with the major objective of conserving biodiversity3.

One justification for biodiversity conservation is the ethical argument for a species' right to exist (Beattie 1995). More commonly, the importance of biodiversity is embodied in the contribution of species to essential ecological processes (e.g. the nutrient cycle, carbon cycle and water quality) and ecosystem resilience (Holling 1986; Nystrom & Folke 2001). The wealth of biodiversity and complexity of ecological systems makes it

3 Global Biodiversity Strategy: Guidelines for Action to Save, Study and use Earth's Biotic Wealth Sustainably and Equitably (1992) published by World Resources Institute, The World Conservation Union (IUCN), and the United Nations Environment Programme in consultation with the Food and Agriculture Organisation of the United Nations and the United Nations Educational Scientific and Cultural Organisation; The National Strategy for the Conservation of Australia's Biological Diversity (1993) published by Federal Department of the Environment, Sport and Territories, Canberra, Australia; The New South Wales Biodiversity Strategy (1999) published by Environmental Policy Division, National Parks and Wildlife Service, Hurstville, NSW. Chapter 1 General Introduction 3

virtually impossible to assess the role that each individual species may play in the maintenance of an ecosystem (Lyons & Schwartz 2001; Steininger et al. 2001).

Recent losses of biodiversity have attracted attention from the media, public, government and conservation groups, with such losses being attributed to many processes, including: (a) habitat loss (including degradation and fragmentation); (b) pollution; (c) over-harvesting; and (d) the introduction to an area of non-native species (e.g. McNeely 1992; Western 1992; Powledge 1998; Smith 1997).

1.1.1 The Impact of Agriculture

Agriculture contributes to the loss of biodiversity both directly and indirectly by fragmenting remnant native vegetation (Steffan-Dewenter & Tscharntke 1999). The direct effects stem from clearing habitat and converting it to other uses (Wilson et al. 1997). Indirect effects include: the introduction of exotic species to the ecosystem (O'Dwyer & Attiwill 1999; Woolley & Kirkpatrick 1999), disruption of plant-pollinator interactions (Steffan-Dewenter & Tscharntke 1999), increased nutrient inputs (Compton & Boone 2000) and overgrazing (Lovich & Bainbridge 1999). The specific impacts of agriculture on biodiversity are not always predictable. Weiss (1999) argued that, although poorly managed cattle grazing can significantly disrupt native ecosystems, in some cases, moderate, well managed grazing may be essential for maintaining such ecosystems when the threat of exotic weed species is present. Burel et al. (1998) have also shown that an intensification of agriculture does not always lead to a decrease in biodiversity but is dependent on the taxonomic grouping that is examined. The impacts of agriculture in particular grazing may also be site specific and is an issue that is being discussed by managers (e.g. Anderson et al. 1996; Earl & Jones 1996' Lunt 1997, Lunt & Morgan 1999)

1.1.2 The Impact of Urban Development

As with agriculture, irreversible loss of diversity through species extinction has often been the result of the destruction of natural habitats (Wilson 1988) to make way for urban development. Urbanisation of a landscape is considered to be a major cause of Chapter 1 General Introduction 4

loss of biodiversity around the world (Tabarelli et al. 1999). The increase in fragmentation 4 of urban vegetation has been linked to the existing patterns of land ownership, with private ( developable) land existing beside reserves and national parks (Swenson & Franklin 2000). It has been suggested that to maintain urban biodiversity, there is a need for ecological knowledge to be integrated into urban planning (Niemela 1999). Urban development often leads to inappropriate management regimes. Inappropriate management can include changes to fire regimes, accelerated introduction of exotic species, changes to water flow direction and nutrient levels, and altered pollination and seed dispersal processes.

Studies in the northern hemisphere (Europe, U.S.A. and Canada) have found that the impact of fragmented vegetation due to urban development varies depending on which taxonomic group is examined. Suarez et al. (1998) showed that native ant communities in coastal southern California are relatively vulnerable to habitat fragmentation by urban development, allowing an increase in the invasion of exotic species such as the Argentine ant (which out-competes native ants). On the other hand, a study in Greater Manchester (Hardy & Dennis 1999) found that urban development (percentage of land cover) did not strongly influence butterfly species richness and species incidences. Distribution patterns could be explained by a butterfly's diet. Adult butterflies are often opportunistic nectar users and nectar sources can be widely spread in urban areas (gardens and park land). Therefore, generalist butterfly species will be less influenced by urban development compared to specialist feeders, which require specific butterfly host plants. However, it is common for biodiversity to decrease in urban areas (Clergeau et al. 1998; Blair 1999; Bolger et al. 2000). This raises some interesting questions, such as, should exotic species be included in an assessment of biodiversity.

Some studies in Australia have concentrated upon the impact of urban development, particularly on avian communities (Catterall et al. 1998; Martin & Catterall 2000) and

4 Fragmentation is used to describes a patchy distribution of suitable habitats, sometimes thought as 'ecological islands" surrounded by a matrix of inhospitable or inadequate habitats of varying permeability. Fragmentation means more than the mere existence of isolated or patchy habitats, it also implies that a more continuous habitat has been subdivided or broken up by some (often anthropogenic) process (Cane 2001). Chapter 1 General Introduction 5

mammal communities (Bentley et al. 2000). Many of these studies show that habitat loss and fragmentation affect species, but the specific response depends on the species. Martin & Catterall (2001) found that coastal heathland birds in subtropical eastern Australia were intolerant to a matrix habitat (residential and cane cropland interspersed). Heathland birds were tolerant of decreased remnant area (this however is dependent on floristic associations in the remnants, which were themselves dependent on the environmental regime in the remnants (e.g. water availability)). In another study, Catterall et al. (1998) identified two groups of native birds (bushland birds and developed-land birds) when looking at the impact of urban development on bird communities, supporting the argument that responses to urban development are species­ specific. As a result, the exact effects on overall biodiversity are very difficult to predict.

Few studies have closely examined the impact of urban development on threatened plant species and communities. Such studies usually concentrate on remnant vegetation as a whole (Clements 1983; Lambert & Turner 1987; Leishman 1990; Rose 1997; Rose & Fairweather 1997; King & Buckney 2000) rather than threatened plant species in particular. For example, King & Buckney (2000) examined the impact of urbanisation upon the northern Sydney streams, using invasion of exotic plants as an indication of the impact of urban development. Other studies have examined changes in soil and water nutrients due to increased run-off from urban areas and floristic changes to vegetation (Leishman 1990; Clements 1983)

Potential impacts from urban development on biodiversity within remnant/fragmented vegetation can be classified into the following groupings. These impact often result in a reduction in population sizes so that they become unviable (Young et al. 1996): • Size effects, an area of habitat may be too small, causing changes to pollinator movement which affects the breeding system of a plant (Buchmann & Nabhan 1996); • Edge effects, including changes to the micro-habitat through stronger winds, higher temperatures, higher or lower humidity and higher light levels (Kelly et al. 2000); • Increased levels of nutrients from stormwater run-off, sewage overflow and diverted water flow resulting from developments (King & Buckney 2000); Chapter 1 General Introduction 6

• Increased numbers of competitors and herbivores (Major et al. 1996); • Changes to fire regimes, affecting upon the evolutionary potential and demography of a species (Gill & Williams 1996); • Changes to grazing regimes, such as the removal of cattle from an area that has become an urban reserve (Lunt & Morgan 1999), and; • Interruptions to gene flow, resulting in outbreeding or inbreeding depressions (Kronfrost & Fleming 2001 ).

1.1.3 Threatened species in New South Wales

Since the colonisation of Australia, a number of species have become extinct. These extinctions include 125 species or sub-species of plants and animals, including seven percent of Australia's mammal species. Thus, Australia has the highest extinction rate of mammals from around the world. Australia still has a large number of species that are considered threatened (www.ea.gov.au - for a full list) despite recent conservation efforts. New South Wales, in particular has a number of species that are classified as endangered, extinct or vulnerable5 (Table 1.1).

Table 1.1: Summary of the number of species listed under the Threatened Species Conservation Act 1995 (NSW) (Source: www.npws.nsw.gov.au 13th July 2001).

Endangered Endangered Species presumed Vulnerable species population extinct species Amphibians 9 (4) 1 (0) 0 (0) 14 (5) Reptiles 6 (2) 0 (0) 1 (0) 25 (8) Birds 26 (5) 3 (0) 12 (10) 76 (3) Mammals 13 (3) 7 (0) 27 (7) 41 (2) Marine mammals 1 (1) 0 (0) 0 (0) 7 (3) Invertebrates 10 (0) 1 (0) 0 (0) 0 (0) Plants 282 (115) 11 (0) 37 (11) 214 (181) Total 347 (130) 23 (0) 77 (28) 377 (302)

Note: The numbers of listed species that were also listed under the Environment Protection and Biodiversity Conservation Act 1999 (Cth) (as at 13th July 2001) are in brackets. There is also no provision to list populations at a Commonwealth level.

5 For definitions of endangered species, endangered populations, species presumed extinct and vulnerable species see section 2.2.1. Chapter 1 General Introduction 7

Many threatened species that are listed on the schedules to the Threatened Species Conservation Act 1995 (NSW) occur around urban areas (Mokany & Adam 2000). As another example, of the nearly 800 plant species listed on the Rare or Threatened Australian Plants (ROTAP) list, 20% occur in the region from Newcastle to Wollongong including Sydney.

1.1.4 Overall review of conservation paradigms

Throughout history, human activities have driven the response to conservmg biodiversity (Christie 1993). Use of the land has led to habitat degradation and fragmentation. Responses to the impacts of habitat fragmentation and degradation are now reflected in future planning and development proposals (Christie 1993) as the emphasis in conservation shifts towards protection of ecological integrity. It has long been said that conservation of threatened species needs a holistic approach, an integration of scientific or technical data and social or community values (Christie 1991). The new regimes or paradigms6 of threatened species conservation have been designed to identify and provide stronger protection, integrated with management options to ensure species survival (Kelly 1994).

Conservation efforts initially revolved around the protection of land in the form of the creation of national parks, nature reserves, Crown reserves, state recreation areas, flora conservation reserves and other icon areas on public land (Farrier & Tucker 1998). The protection of such parcels of land within N.S.W. give the appearance of covering the range of natural, in some cases conserving populations of threatened species that occur nowhere else. This is not always the case, for example on the north coast there are significant areas needed for a comprehensive and representative reserve system that occur on private land. Australia's first national park was established in 1879 (as a "people's park") (Pressey 1995), making it the world's second oldest national park. Since the enactment of the National Parks and Wildlife Act 1974 (NSW), the parks

6 Kuhn first coined the term paradigm in 1970 as a disciplinary matrix, which sets the standards for legitimate work within a particular field of research. A paradigm consists of the theoretical assumption, laws and techniques, used for the application of research by members of a particular community to define their work at a given period in time (Chalmers 1991). Chapter 1 General Introduction 8

system has expanded to include seven different categories of protection; national parks, nature reserves, historic sites, aboriginal areas, state game reserves, karst conservation areas, state recreation areas and regional parks (National Parks and Wildlife Act 1974 (NSW); State of Parks, 2001). In N.S.W. a total of 6.7% of land area is in the reserve

1 system, as of the 30 h June 2001 (Table 1. 2).

Table 1.2: Summary of the parks system in N.S.W., showing the different categories (Source: www.npws.nsw.gov.au).

Category How many Area (hectares) National parks 161 4,442,200 Nature reserves 359 794,877 Historic sites 13 2,635 Aboriginal sites 11 11,643 State recreation areas 22 126,368 Regional parks 10 4,970 Karst conservation areas 4 4,409 Total 580 5,387,102

The parks system has moved away from the sole objective of providing park land for recreational use by society to a three pronged approach: (1) conserving and protecting areas with natural and cultural significance, ensuring the survival of natural systems, (2) providing appropriate and sustainable opportunities for recreation and (3) contributing to the development of a global network of protected areas (State of Parks, 2001). These objectives have been lost in the acquisition and allocation of land to reserve systems. Land that has been allocated to reserves has often been left over or adjacent to already formed reserves often not providing adequate conservation in terms of biodiversity (Pressey 1995). Recently there has been a move towards integrating national park and off-reserve management (Morton et al. 1995), through a number of initiatives (such as voluntary conservation agreements, recovery planning, and more stringent development control processes).

For many decades, Australian conservation has been dominated by the assumption that conservation can be achieved through a reserve system (Recher 1997). In N.S.W., reserves fall into two main types with different objectives; nature reserves which are set aside because of scientific significance (small in size) and national parks (Farrier 1996). Ecologists have argued that using a reserve-based system is limited and not Chapter 1 General Introduction 9

representative, as often many ecological communities are not conserved (Pressey 1995). Rather, national parks tend to occur close to cities, for immediate access by the public (37 reserves in Sydney and surrounds c.f. Central N.S.W. with 18 reserves and the Outback with 9 reserves). Arid and semi-arid zone communities are neglected in conservation and are the most at threat from extinction. The next two sections are about corridors and translocation, representing aspects of off-reserve management.

Off-reserve management or conservation on private lands has become a dominant theme in the conservation literature (Farrier 1996). This change in conservation paradigms is due to the increased voice of ecologists demanding that the landscape be recognised as a "shifting mosaic", of continuous biotic change and heterogeneity (Farrier 1996). Voluntary conservation agreements (made between a landholder and the Minister for the Environment) are one way in which conservation is being achieved on private lands. So far, in NSW 106 VCAs have been signed to incorporate land that contains significant threatened species, important habitat types not within national parks, historic sites, remnant vegetation, linking native habitat, and geological features (and in the future critical habitat). The signing of a group of VCAs in the Eden/Bega area, developed from the Regional Forestry Agreement process, allowed adjoining landholders to create a corridor between two new parks (pers. com. with T. Celebrezze, senior project officer, NPWS). The creation of such a corridor meets one of the objectives of VCAs, creating connectivity between vegetation.

Vegetation Corridors

As a result of fragmentation of natural vegetation by either urban or agricultural development, small islands of habitat have been isolated, forming what is known as remnant vegetation often providing a last refuge for threatened species (Hogbin et al. 1998). Remnant vegetation has been shown to play an important function in the conservation of biodiversity, becoming a central conservation issue in many developed countries (Bowers 1999). However, the conservation value of patches of relatively small and isolated remnant vegetation has been questioned over the years, as small isolated populations often have a reduced reproductive potential and genotypic diversity (Hogbin et al. 1998). The relationship between remnant integrity and size, age, shape, Chapter 1 General Introduction 10

environment and disturbance are not well established (Gilfedder & Kirkpatrick 1998). However, a number of studies have shown that the worth of remnant populations is going to be dependent on the ecology of the species that is central to the conservation of the area. For example, Hogbin et al. (1998) found that populations of Grevillea macleayana within remnant vegetation had a high conservation value because it was as fecund and genetically diverse as populations occurring in connected vegetation. On the other hand Abenspergtraun et al. (1996) found that habitat disturbance associated with remnant vegetation (within Eucalyptus woodland in Western Australia) greatly influenced the arthropod communities, causing low variation, abundance and diversity of species. Furthermore, the degree of habitat disturbance was correlated with size and connectivity of the vegetation (e.g. small, poorly connected habitat had a higher degree of disturbance) (Abenspergtraun et al. 1996). In general, retaining remnant vegetation, particularly in agricultural systems, has been viewed as beneficial in terms of maintaining non-threatened populations size and reducing site extinction of threatened species (Blarney et al. 2000).

With the rates of habitat loss and fragmentation increasing, the inclusion of animal movement or habitat corridors in urban development, reserve design and conservation programmes has received mixed support (e.g. Forney & Gilpin 1989; Lindenmayer et al. 1993; Bentley & Catteral 1997). While strong skepticism is apparent from some ecologists (e.g. Simberloff & Cox 1987; Simberloff et al. 1992), movement corridors are receiving strong support from other ecologists (e.g. Noss 1987; Beier & Noss 1998) as essential conservation instruments (Haddad 2000; Haddad et al. 2000). However, empirical studies have shown that taking an intermediate position on the inclusion of corridors in management is more appropriate (e.g. Hobbs 1992; Andreassen et al. 1996) as results have shown that the effectiveness of corridors are species and landscape specific (Beier & Noss 1998).

Gaps in knowledge of corridor effectiveness as highlighted by Mann & Plummer (1995) have lead to scientific skepticism of corridor effectiveness. Firstly, the theory assumes that a species exists as a metapopulation. Natural fragmentation of a species is also not recognised within the theoretical framework. Most importantly, there is a substantial lack of empirical evidence to support corridors. The idea of corridors in landscape Chapter 1 General Introduction 11

management comes from the meshing together of island biogeography and metapopulation theory (Haddad 1999). The use of corridors in management is based upon the hypothesis that the retention of a linear strip of habitat between at least 2 patches will increase animal movement (Hobbs 1992; Rosenberg et al. 1997; Haddad 1999; Haddad et al. 2000). Therefore the best evidence for corridor usage by an animal is through mark and recapture experiments (Haddad 1999). The patches can have been historically connected with naturally occurring corridors (e.g. riparian strips, shelter belts) or artificially created through revegetation or the leaving of remnant habitat on agricultural lands and cleared forest.

Four rationales have been identified for creating corridors; (a) decreasing the potential extinction rate, (b) reducing the consequences of demographic stochasticity, (c) reducing inbreeding depression, and (d) enabling movement (Simberloff et al. 1992) of animals. A number of studies have predicted that there are costs to corridors as well; (a) transmitting disease readily, (b) fire spread, and ( c) predation by domestic animals (Simberloff & Cox 1987).

Empirical data are being used to establish the value of corridors as conservation instruments. However, research has concentrated exclusively upon animal movement (e.g. Dunning et al. 1995). There is no data on whether corridors are beneficial for plants (e.g. to facilitate pollinator movement or seed dispersal).

Translocation of Species

Translocation is the process whereby individuals are removed from an area and released or planted in another area. Translocation has been widely used throughout the world to conserve threatened fauna species. However, its success is questionable as there are very little data available post-translocation (Pierre 1999). Translocation of plant species has received very little attention (e.g. Coates & Atkins 2001) but it is increasingly being pressed as a "solution" in the urban development context. Translocation has been used to re-introduce a species to an area (e.g. Short & Turner 2000), remove a species that is under threat from a development, such as a hydroelectric reservoir (e.g. Richard-Hansen et al. 2000) or to bring back a species from the brink of extinction. With high extinction

3 0009 03317227 6 Chapter 1 General Introduction 12

rates of mammals in Western Australia, translocation has proved to be an important conservation tool especially in arid and semi-arid regions (Coates & Atkins 2001), following the successful translocation of the burrowing bettong from its island population to the mainland (Short & Turner 2000). The use of islands and peninsulas in conjunction with controlling exotic predators, herbivores and appropriate fire regimes have contributed to the success of translocation management plans in Western Australia (Short & Smith 1994; Abbott 2000; Short & Turner 2000).

The translocation of the endangered Gould's Petrel in N.S.W. to form a new colony has proven a success and is important for the long-term conservation of the species. This management option incorporates long term monitoring to assess the viability of the translocated colony (Priddle and Carlile 2001).

Translocation of threatened plant species can take place as re-stocking (increasing population size by adding individuals), re-introduction (establishing a population at a site where one previously existed), introduction (establish a population where one has never existed but the area occurs within the known distribution) or conservation introduction (establishing a population outside the known distribution). Many different factors need to be taken into consideration when translocating threatened plant species (e.g. understanding of population biology, genetic assessment). These are set out in Guidelines for the Translocation of Threatened Plants in Australia (Australian Network for Plant Conservation 1998).

1.2 Policy, legislation and management instruments

Historically, wildlife law in New South Wales is derived from the British legal system. Hence, Anglo-Saxon ideals were embedded within early N.S.W. environmental law. New South Wales wildlife law can be divided into three distinct stages: (i) exploitative pioneering (1800s), (ii) development of nature and 'wise use' of resources (1900- 1960s), and (iii) modern environmentalism (1960s - present) (Frawley 1994). It is within the final stage that two major ideologies compete (conservation and resource use). Chapter 1 General Introduction 13

Exploitative pioneering in Australia was an era in which the colony's efforts were directed towards expansion and identification of useful resources in their surroundings. Early wildlife laws of New South Wales were developed within the paradigm of the wise use of resources (stage (ii)) (Table 1.3). Subsequently the focus of wildlife law was on regulation, protected areas and the prohibition of killing. The creation of lists of species in legislation has traditionally been used as a means of triggering regulation such as the prohibition of killing, with lists containing a mixture of both native and introduced species.

Within the third stage of wildlife law, developers placed a large amount of pressure on the legal system to allow pro-development ideas to continue to dominate despite small concessions to conservation. Between 1975 and 1982, 102 pieces of environmental legislation were passed (79 at the state level in N.S.W. and 23 at the Commonwealth), with approximately 80% of these laws directed towards conservation and environmental planning (Frawley 1994).

1.2.1 The role of local government

Local councils are the traditional providers of property-related services (Kelly & Farrier 1996). However, local councils have recently been given a comprehensive role in biodiversity conservation on both private and public land (Kelly 1995; Kelly & Farrier 1996), via legal obligations under a number of pieces of legislation, primarily the TSCA7 and EPAA. Major threats to biodiversity that local councils are now involved in dealing with include clearing of native vegetation, building construction and design, land filling and earthworks, bush fire management, stock grazing, alteration to hydrological systems (e.g. increased nutrients and salinity), climate change and global warming, roads and traffic, soil (e.g. erosion, sedimentation and compaction), waste disposal and rubbish dumping (Fallding et al. 2001). Planning instruments are used to

7 This thesis is concerned with terrestrial issues and therefore the Fisheries Management Act will not be discussed. The Rural Fires Act is also not discussed in depth but mentioned where appropriate in the text. The Native Vegetation Clearence Act does not apply in urban areas and therefore is also not discussed. Table 1.3: History of wildlife conservation legislation specific to New South Wales (Sourced from the legislation). g -§ ~ "'I Legislation Listing Regulations and fines Protected Areas Closed Seasons Licences ...... ~ ~ ;::s ~ Birds Protection Act Under the direction of the A fine of £10 for killing, or taking Can be either an area From the 1st August Authorisation from ~ 1901 Colonial Secretary. an individual nest or eggs of a listed of water, an island or to 31st January. Colonial Secretary for - ....~ species during a closed season. enclosed areas. natural history collections. ~ ~ ("") Native Animals Under the direction of the A fine of £5 for killing capturing or ...... c:. Protection Act 1903 Colonial Secretary. injuring any listed species during a Not specified. From the 1st August Authorisation from ;::s closed season. Also a £5 fine for to 31st January. Colonial Secretary for selling a live individual or a skin of natural history collections. any listed species during a closed season.

Birds and Animals Under the direction of the A £20 fine for taking or killing a Declared by the Minister may issue Protection Act 1918 Minister. Districts can also listed species during a closed Not specified Minister, it may licences to take or kill for be listed. season by means of hunting, cover a state or purposes of sale, or if the shooting, killing, poisoning, netting, area. Alternatively animal is injurious to snaring, spearing, pursuing, taking, an open season may property, or for scientific disturbing or injuring. be declared. exploration.

Wildflowers and Under the guidance of the A fine of £5 (1st offence), £10 (2nd Not specified. Time periods for Minister may issue a Native Plants Governor for the first time offence) or £20 (3rd or subsequent protection can be licence to pick for Protection Act 1927 plants could be listed. offence) for picking or selling a specified. scientific purposes. listed species.

Fauna Protection Under the guidance of the Fauna! Reserves Act 1948 Minister. Districts can also recommended by the be listed, and rare species Panel and declared by can be declared by the the Governor. Each governor reserve has a working plan approved by the Minister.

National Parks and The Governor-General of It is an offence to harm, pick, buy Areas within National NIA Licences can be obtained Wildlife Act 1974 NPWS can list protected sell or posses any species listed on Parks and Service from the Director-General and Unprotected species, the TSCA or NPWA or damage the estates are protected ofNPWS. as well as endangered habitat associated with a listed areas. fauna. Lists associated with species. the TSCA are also considered under the NPWA.

Threatened Species Vulnerable, endangered This act is linked to the EP AA and Protected areas take NIA Licences are granted un Conservation Act and extinct species, NPWS, which contain the the form of either section 91 of the TSCA. 1995 endangered populations regulatory mechanisms. The TSCA critical habitat or Generally licences are and ecological is linked to the development control endangered obtained under the NPWA communities, and systems. ecological and approvals under the threatening process can be communities.Areas of EP AA. If an action has listed by the scientific ceitical habitat are approval under the EP AA committee. Critical habitat recommended by the a licence from NPWS is can also be identified. scientific committee not required. to the Director- ..... general of NPWS. +'- Chapter 1 General Introduction 15

8 9 regulate the above threats, through the use of policy plans , framework plans and 10 management plans • Local councils can influence private landholders in conserving threatened species, through the use of local planning instruments (Kelly & Farrier 1996).

1.3 Importance of the science/law interface

There are many reasons why science and law do not complement each other. Scientific and legal paradigms have their own terminologies, methodologies and philosophies (Stewart 1992). These values present in the paradigms are seen to contribute to the fundamental conflict between the legal system, with its complex set of rules, which are used for inquiry in a retrospective manner, and science, which is constructed by a predictive methodology to test hypotheses and develop models (Stewart 1992). Despite their fundamental differences scientific knowledge has fast been incorporated into environmental legislation, in particular legislation that aims to protect threatened species. Often scientists are not consulted on how their paradigm is incorporated into a legal paradigm, often leading to unusable statutory mechanisms, policy and management options.

In some situations, the law has borrowed terminology from science, even though there is no intention of using the concepts in the way they have been constructed. Instead, they have been firmly locked into the common law paradigm with an emphasis on the adversarial process and the pursuit of victory in disputes (Stewart 1992). It is therefore important to have an understanding of how science and the law interact, ultimately leading to an understanding of how legislation, policy and other statutory mechanisms are used and how they differ from their intended usage.

8 Policy plans express broad principles for the future such as those expressed in the National Strategy for Conserving Biodiversity.

9 Framework plans determine which activities may be carried out at particular sites and are regulated by the EP AA through planning instruments such as SEPPs and LEPs.

10 Management plans specify how activities are to be undertaken and usually cover lands such as national parks1 community land and crown reserves. These usually go beyond development control to include active management. Chapter 1 General Introduction 16

1.3.1 The role of scientific uncertainty and the precautionary principle.

Since the 1980's, legislation, policy and management documents have embraced the concepts of scientific uncertainty and the precautionary principle. The precautionary principle first arose in Germany in the 1930's as the concept Vorsogeprinzip, which literally means 'foresight principle' (Gullett 1997; Meurling 1999). Originally concerned with the problem of pollution in the fast-developing industrial nation of West Germany, by the 1970's it was found extensively in West German legislation (Gullet 1997). The precautionary principle first emerged on the international stage as part of the Second International North Seas Conference in Breman, Europe in 1987 (Deville & Harding 1997). The concerns of the conference delegates centred around regulation of emissions into the North Sea, not just of those chemicals that had a proven scientific effect but also of those with no clear scientific link to impact on fisheries (scientific uncertainty) (Deville & Harding 1997). In the 1980's, the environment became a major theme in politics (Stein 2000), sparking the incorporation of the precautionary principle into the objectives of international treaties and Australian domestic policy and legislation11 (Gullet 1997; Meurling 1999). Despite this legal recognition of the concept, there is very little or no indication of what role the precautionary principle should serve. Hence, the precautionary principle has become easily endorsed as a keystone principle (Santillo et al. 1998) by all levels of government but has proven difficult to apply (Recher 1999). The precautionary principle entered the environmental discourse as a way of counteracting anthropogenic impacts upon the environment by providing philosophical authority to deal with scientific uncertainty (Gullet 1997; Santillo et al. 1998).

11 Definitions of, and suggestions for the implementation of the precautionary principle have been found in a number of major policy documents in Australia, including The National Strategy for Ecological Sustainable Development (NSESD) (1992) and The Intergovernmental Agreement on the Environment (IGAE) (1992). These are broad national agreements (policy documents) for environmental protection and are not legally binding The precautionary principle has further been incorporated into legislation such as the Protection of the Environment Administration Act 1991 (NSW), Environment Protection Act 1993 (SA), National Environment Protection Council Act 1994 (Cth), Environment Protection Act 1994 (OLD), Fisheries Management Act 1994 (NSW), Threatened Species Conservation Act 1995 (NSW) and Living Marine Resources Management Act 1995 (Tas). Chapter 1 General Introduction 17

The precautionary principle is an extensively used concept but has no clear or agreed 12 upon definition • The definition of the precautionary principle in the Protection of the Environment Administration Act 1991 is actively used in N.S.W. and is incorporated into various pieces of legislation by cross-reference, including the TSCA but, surprisingly not the EP AA. The precautionary principle is used as follows in this thesis: when the outcome from an activity (e.g. development) has an apparent significant (including irreversible) impact on the environment then scientific uncertainty cannot be used as a reason for not implementing appropriate environmental protection.

In many of the definitions of the precautionary principle, science lies at the core but administrative or legislative factors complicate the definition (Harding & Fisher 1999) by obscuring the meaning (Santillo et al. 1998; Meurling 1999). With such varying definitions of the precautionary principle, the scientific community has stayed away from what it sees as essentially a policy concept. The precautionary principle has taken on the role of forcing a decision in an attempt to by pass scientific "fence sitting". Suspicion has arisen from confusion surrounding the role of scientific data and its translation into a philosophical-political process (Calver et al. 1999). It has been argued that by placing the onus of proof on to industry or the developer, the precautionary principle provides a general framework for discussion and decision making in regards to environmental issues. However, the precautionary principle is deemed too vague to serve as a regulatory standard (compared with those set down in legislation and policy), on the level of caution to be used (Bodansky 1991).

12 Legislative and policy definition examples include: (l)"If there are threats of serious or irreversible environmental damages, lack of full scientific uncertainty should not be used as a reason for postponing measures to prevent environmental damage." (Protection of the Environment Administration Act 1991 (NSW)), (2) "Where there are threats of serious or irreversible environmental damage, lack of full scientific certainty should not be used as a reason for postponing measures to prevent environmental damage. In the application of the precautionary principle, public and private decisions should be guided by: (i) careful evaluation to avoid, wherever practicable, serious or irreversible damage to the environment; and (ii) an assessment of the risk-weighted consequences of various options." (The Intergovernmental Agreement on the Environment, May 1992, Tasmanian State Coastal Policy 1996) and (3) "In order to protect the environment, the precautionary approach shall be widely applied by States according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation." (Principle 15 Rio Declaration on the Environment). Chapter 1 General Introduction 18

Confusion over a standard definition has led to many interpretational conflicts. Is the precautionary principle a guiding or operational concept or even a yardstick to measure political decisions (Stein 2000)? Stein (2000) argues that the lack of a concise definition is at the heart of the lack of interest in implementing the precautionary principle. Forming a definition with precision is indeed problematic, because the concept is rapidly evolving to suit different scenarios. Stein (2000) further argues that ambiguity in conceptualisation at a policy level leads to divergent meanings and can act as a fundamental barrier to successful implementation. Advocates for the implementation and enforcement of the precautionary principle argue that the principle design allows for its use as a guiding force, not to provide the outcome for a decision (Deville & Harding 1997). Holms and Harris (1999) argues that the element that prohibits the precautionary principle from being successfully implemented is the lack of a workable definition. A majority of definitions (see footnote 11) tell us to balance evidence (scientific knowledge) in a specific way (eg. to what degree to believe the evidence or data) against moral choices and judgments. This illustrates one of the fatal weaknesses in the precautionary principle, the attempt to convert moral choices into legislation. This leads to the temptations of looking for overarching principles and rules.

Deville & Harding (1997) believe that the precautionary principle places the onus of proof on proponents of technological and industrial developments to demonstrate they are ecologically sustainable. Calver et al. (1999) further cultivates this idea by arguing that the precautionary principle redistributes the onus of proof onto the developer to provide guidance for preventive management. It is apparent in the literature that the distribution or allocation of onus of proof is essential to the implementation of the precautionary principle. Calver et al. (1999b) argue that scientific proof is irrelevant if applying the precautionary principle, as the concept revolves around whether there are reasonable grounds for concern. Meurling (1999) believes that we should always err on the side of caution, show respect for the environment, and place the onus on the proponent of development. Beder (1993), Deville & Harding (1997) and Meurling (1999) support these arguments that policy should always err on the side of caution and argue that technological and industrial developments should be forced to demonstrate their ability to be ecologically sustainable. Chapter 1 General Introduction 19

The precautionary principle is a culturally framed concept, and takes its cue from changing social conceptions (Frangos 1999). This redistributes the onus of proof on to the developer/proponent to bring about preventive management (Calver et al. 1999). Despite reference in N.S.W. legislation, there is a lack of support (Meurling 1999) within environmental case history and the onus of proof has often been placed on proving there is going to be significant effect.

Ideally, decision-making processes (e.g. determining whether a development will impact upon the environment) require uncontested information such as scientific knowledge. This form of decision-making is acknowledged as forming the basis for environmental regulation (Gullet 1997). This approach does not recognise scientific uncertainty. Uncertainty arises when the baseline data are not available or when resources are constrained, or even when the environmental factors requiring information are unable to be measured (Gullet 1997).

The methodology of ecology has always placed emphasis on the importance of rigorous experimental design, involving hypothesis testing and confirmation with statistics (Krebs 1989). It is important to note that ambiguity, subjectivity and assumptions are inherent in scientific methods and interpretations (Gullet 1997) (Table 1.4). And scientific uncertainty arises as a complex web of risk, uncertainty, ignorance and indeterminacy (Harding and Fisher 1999). MacGarvin (1999) argues that major environmental disasters historically have been the result of ignorance not uncertainty. Science is constructed with old beliefs entrenched in methodologies, resulting in continued attempts to assess points at which serious damage has been happening to the environment, and often this is impossible (MacGarvin 1999).

The role of scientific uncertainty and how it arises has been neglected in policy making. Scientific uncertainty should only become a problem when assumptions are institutionalised into policy-making or used as a basis for decisions (Barton 1998) placing restrictions on the appropriateness of long-term predictions (e.g. the debate regarding the ozone hole where science discovered but failed to predict its emergence) (Gullet 1997). Chapter 1 General Introduction 20

Table 1.4: There are many different classifications of scientific uncertainty and its sources (Deville & Harding 1997).

Type of Uncertainty Origination

Technical or theoretical Accuracy of the data is questioned.

Methodological The appropriateness of methods employed is in doubt.

Epistemological Focuses upon whether the knowledge framework in which a resolution came about is correct.

Pragmatic Lack of funds or time, leading to inadequate investigation into potential answers.

Complexity The unpredictable nature of the environment due to the complex relationship (e.g. between species and their environment).

Intangible damage Damage may not be easily observed and policy makers are yet to set relevant baselines and standards.

The precautionary principle forms part of the objectives of the TSCA13 contained within a statement about promoting ecologically sustainable development. However, the precautionary principle is not identified as an enforceable or usable component of the decision-making process (in particular during the assessment of development applications). The NPWS however, need to consider the principles of ESD when deciding whether or not to give concurrence (Farrier et al. in prep). The application of the precautionary principle has not increased the impact of the TSCA (Sperling 1999).

While the precautionary principle is referred to in many policy documents attempting to translate science into policy (Santillo et al. 1998), the courts have been slow to accept the precautionary principle as directly relevant for decision-makers within the legal community. However, the precautionary principle has been referred to as a statement of commonsense in relation to threatened species in the judgement of Leatch v National Parks and Wildlife Service and Shoalhaven City Council (1993) 81LGERA270.

13 " ••• conserve biodiversity and promote ecologically sustainable development" (s. 3a, TSCA) Chapter 1 General Introduction 21

1.4 Specific aims

The above literature review has highlighted the 'gaps' in the knowledge relating to the conservation of threatened species. These 'gaps'or objectives which my thesis aims to address include the need: (1) for a review of legal and policy instruments relating to threatened species; (2) to understand how scientific knowledge is used and where it is needed; (3) to evaluate whether science and policy can continue to deliver conservation, and; (4) to reveal weaknesses in the assumptions associated with threatened species legislation.

To answer the general question of how policy and science interacts, a study was established using two threatened14 flora species, Tetratheca glandulosa and Darwinia biflora. The study had two different but related objectives. Firstly, it was designed to determine how scientific knowledge and theory has interacted with environmental policy. Secondly, it was designed to assess in what way policy and law have attempted to protect species, as measured by development (human) impacts on the persistence of populations/individuals. To achieve these aims, five specific questions were asked (Table 1.5).

1.5 General outline of methodology for the study as a whole

The study was divided into four components; (1) analysis of the TSCA and the EPAA (the TSCA and EPAA are linked closely to one another), (2) a detailed examination of a number of case studies involving urban development proposals, (3) detailed scientific experiments examining the fundamental biology of the two threatened species and (4) an analysis of the adequacy of the scientific information used in the decision-making process in light of the research findings. Initially the study began with an examination and breakdown into components of the TSCA and EPAA (Figure 1.1), in which a number of key questions were identified. The components of the TSCA that required scientific knowledge were then incorporated into a process that allowed for the identification of case studies and study species (Tetratheca glandulosa and Darwinia

14 Listed on the schedules of the Threatened Species Conservation Act 1995 (NSW). Chapter 1 General Introduction 22

biflora) (Figure 1.2). An appropriate method with which to examine the case studies was established (see Chapter 3).

Table 1.5: Summary of specific questions in the study and how each question was approached.

Specific Question Approach

How has the wording of the legislation An examination of the wording of the influenced decisions (by councils, legislation, relating this to development developers, NPWS and the Land and applications and the outcomes of the Environment Court) on development decisions made by consent authorities. applications for sites in which there are populations of threatened flora species? (Chapters 2 and 3)

What is the role and understanding of An examination of the TSCA and EP AA, scientific knowledge and processes in and other policies and decisions dealing decision-making processes relating to with species conservation in relation to: threatened species conservation? (Chapter • site specificity, the importance of remnant 3 and 8) vegetation and reserves; • lack of scientific knowledge and use of the precautionary principle; and, • pollinating systems.

What is the pollination biology of T. An examination of the mating system of glandulosa and D. biflora? (Chapter 5) each species and its impact upon the population, using flower fruit ratios, the effect of pollen parentage on seed quality, identification of pollinators and the potential of T. glandulosa displaying clonality.

What is the impact of urban development The examination of natural flowering and (through habitat fragmentation) upon fruiting densities, pollinator movement and pollination regimes? (Chapter 6) seed germination in relation to population size and connectivity.

What is the seed bank profile (recruitment An examination of the level of seed potential and success) for the two study dormancy for each study species and the species and how is this altered through size of the soil seed bank, which was then human disturbances? (Chapter 7) contrasted with models of the potential seed bank each year. Chapter 1 General Introduction 23

EPAA How does the TSCA interact with other NPWA legislation Rural Fires Act

What are the major components of the TSCA? What can be listed?

How is science incorporated into the Listing process?

How does Usting contribute to biodiversity conservation?

How does the Eight Part Test operate?

What are the ecological constraints of the Eight Part Test? Eight Part Tests What are the problems in the application of the Eight Part Test?

How do Eight Part Tests contribute to biodiversity conservation?

How does the SIS operate?

What are the ecological constraints of the SIS? Species Impact Statements (SIS) What are the problems in the application of the SIS?

How do SIS contribute to biodiversity conservation?

How do Recovery Plans operate?

What are the ecological constraints of Recovery Plans? Recovery Plans What are the problems in the application of Recovery Plans?

How do Recovery Plans contribute to biodiversity conservation?

Figure 1.1: Schematic diagram displaying how a detailed examination of the Threatened Species Conservation Act 1995 (NSW) was carried out. Chapter 1 General Introduction 24

Is the species in the area?

N0 --Yes -----, ,, Is the site Is the site part of a Is this correct? ----developed or to be No reserve system? developed? I Ye s l i Yes Inappropriate Species never survey there technique Was the development Case study application No rejected processed using ,. the TSCA?

Species extinct in the area Yes

Did existing science play a role in determining development consent?

Yes No

How would urbanisation affect .__ -+ natural ecological processes?

~, Is the population going to remain viable following development of the site?

Figure 1.2: Schematic diagram showing methodology for how case studies presented in Chapter 3 were approached. Chapter 2 Legislation and Threatened Species 25

CHAPTER 2 LEGISLATING FOR BIODIVERSITY CONSERVATION IN NEW SOUTH WALES

"The objects of this Act are as follows: to conserve biological diversity and promote · ecologically sustainable development, and.. " TSCA (1995) (s 3a)

2.1 Introduction

Governments tend to measure their environmental success based on increased numbers of National Parks and area of National Parks (Lunney et al. 2000). However, conservation measures have not readily been applied to off-reserve conservation. During the late 1980s and early 1990s, environmental pressure groups caused the community and government in NSW to take notice of the 1 2 activities of the Forestry Commission through a series of court cases • Following the successful injunction against further logging activities in the 1990 Chaelundi judgment (see Appendix 2.1), the N.S.W. State government had to face the fundamental issue of increased awareness and pressure from the community for conservation off reserve, including privately owned land. The Threatened Species Conservation Act 1995 (NSW) (TSCA) was a legislative response addressing this issue. Scientific knowledge of the ecology of threatened3 species had very little influence in the creation of the legislation.

1 The Forestry Commission will be referred to as 'the Commission'.

2 Kivi v. Forestry Commission (1982) 47 LGRA 38, Prineas v. Forestry Commission (1984) 53 LGRA 160, Jarasius v. Forestry Commission [No.1] (1989) 69 LGRA 156; [No.2] (1990) 71 LGRA 116, Corkill v Hope (1991) 74 LGRA 33. Court cases collated by Prest (1995).

3 The TSCA not only protects species but populations and ecological communities, although the focus in this thesis is on species, unless specifically talking about listed species, populations and ecological communities. Chapter 2 Legislation and Threatened Species 26

2.2 Legislating for Conservation in New South Wales

Using legislation4 to protect biodiversity has become the tool of choice of many interest groups for dictating the future of land use (Dixon 1994; Beattie 1996; Lim 1997). The TSCA is viewed as a major governmental initiative in New South Wales (Adam et al. 1997) despite its forced entrance into the political and conservation arenas (Brunton 1996; Kelly 1996). There has been a shift in conservation from protection of single species (e.g. Baker & Whelan 1994; Smith et al. 1994; Auld & Denham 1999; Hogbin & Peakall 1999; Abbott 2000) to ecosystem management (e.g. Grumbine 1994; Stanley 1995; Grumbine 1997; Brunner & Clark 1997; Norton & Reid 1997; Shachak et al. 1998; Richards et al. 1999). Beattie (1996) made the important point that protecting single species is still essential. However, maintaining ecological processes may be more efficient in achieving biodiversity conservation than concentrating on individual species. The conservation of individual species is still favoured by legislation5 as a number of interest groups argue that the protection of single species is the best approach. This approach is viewed as achievable as its aim is directed towards the protection of the basic building blocks of diversity, individual species (Dixon 1994).

The TSCA is a legislative attempt to minimise biodiversity losses and meet the concerns of the various interest groups (Smith 1997). Biodiversity principles6 have not only been incorporated as the primary objective of the TSCA (s 3a) but they are incorporated into the planning system (Smith 1997) through the Environmental Planning and Assessment Act 1979 (NSW) (EPAA).

4 It is often argued that a legislative framework for species protection has increased benefits compared to non-legislative conservation. Dixon (1994) comments that non-legislative conservation groups (such as community groups) have the undue risk of being overwhelmed by the economic constraints of inter-government dissensions.

5 Such as Threatened Species Protection Act 1995 (TAS), Flora and Fauna Guarantee Act 1988 (VIC) and Nature Conservation Act 1992 (QLD).

6 There are nine biodiversity principles which include: (i) conservation in-situ, (ii) responsibility at all governmental levels, (iii) removal of threats to biodiversity and (iv) lack of knowledge should be incorporated into management (The National Strategy for the Conservation of Australia's Biological Diversity 1996). The TSCA does not adopt these specifically. Chapter 2 Legislation and Threatened Species 27

The TSCA does not operate in isolation. It is linked with the EPAA the National Parks and Wildlife Act 1974 (NSW) (NPWA) and the Rural Fires Act 1997 (NSW) (RFA). The TSCA and the EP AA, the primary land use planning legislation, are closely integrated. Threatened species are now considered at all stages of the planning system through a wider coverage of listed species, populations, ecological communities, and threatening processes, and their incorporation into a series of quite intricate environmental assessment procedures (i.e. "eight part tests" determining likely significant affect and species impact statements) and controls (Lim 1996).

Most urban development by private developers is considered under Part 4 of the EP AA, and it is this path to development consent that this study focuses upon. Part 4 of the EPAA incorporates links with the TSCA (eight part tests and SIS), with consideration of threatened species, populations and ecological communities during the evaluation of development applications.

State governments are empowered to manage, develop and conserve land for future generations. In New South Wales, the Environmental Planning and Assessment Act 1979 (NSW) (EPAA) (Figure 2.1) provides forward planning mechanisms to manage development and development controls. Planning in N.S.W. is controlled by Environmental Planning Instruments (EPis), which is the collective name for: State Environmental Planning Policies (SEPPsf, Regional 8 9 Environmental Plans (REPs) , and Local Environmental Plans (LEPs) (which

7 SEPPs (EPAA Div. 2 s. (37)-(39)) deal with issues, which are considered to be of state significance (e.g. SEPP No. 14 Coastal wetlands and SEPP No. 62 Sustainable Aquaculture SEPP no. 34 Major employment-generating industrial development SEPP No. 47 Moore Park Showground).

8 REPs (EPAA Div. 3 s. (40)-(52)) are concerned with issues that go beyond particular local government areas (e.g. REP 20 Hawkesbury-Nepean River) or provide a framework for detailed planning regimes over large areas (e.g. REP 1 Hunter Valley) or over small areas of regional significance (e.g. REP 16 Walsh Bay, REP 24 Homebush Bay Area).

9 LEPs (EPAA Div. 4 s. (53)-(72)(L)) are prepared by local councils and guide decisions in local planning by using zoning and development controls. Hence each local government area can adopt codes for development. 9 May specify when .§ SEPPs (Div. 2 s.(37)-(39)) Zoning ...,~ 1------~Idevel~pment consent REPS (Div. 3 s.(40)-(52)) N (EPAA part 3) 1s needed LEPs (Div. 4 s.(53}-(72)(L)) t'-.; (EPAA s. 76 (1 )-(3)) ~ ~· $S"" .....

~·

~ ~ EPAA Part 5, if significant impact occurs ~ then concurrence required (from NPWS or • ~ environmental minister) (EP AA s. 76 & s.112) N o d eve Iopmen t consen t . . . Development consent required ~ e.g. development by public bodies or private ~ development which requires an approval a other than development consent. ~ ("\) ("')

~· • • Licensing ~ Exempt development -.. (TSCA Part 6) EPAA Part 4 e.g. Private Forestry

v I i Fiaure 2.2 I ._, ' ! No environmental !...... ··················-···-·· assessment

N Figure 2.1: Development consent is obtained in New South Wales under the Environmental Planning and Assessment Act 1979 (NSW). There are a number of paths to gain 00 development consent or to determine if the development is exempt from the process. Chapter 2 Legislation and Threatened Species 29

includes zoning10 mechanisms). Development Control Plans (DCPs )11 can also be made, but they are not classified as environmental planning instruments.

Development that does not require development consent falls into two types: (1) Exempt development which does not require any environmental assessment, for example, if the development has minimal impact on the local environment such as small fences, BBQs or pergolas, providing it is also classified as exempt in EPis. Often these types of development must comply with other conditions (e.g. height restrictions) specified by the local council. (2) Traditionally, some activities can be completed without consent from local council (e.g. private forestry, agriculture and major public works including motorways and electricity transmission lines). Where they require an approval from another government agency or a public authority is the developer (Farrier et. al 1999), they are subject to environmental assessment under Part 5 of the EP AA. Part 5 development is not covered in this thesis.

Complying development is development, which requires consent and includes common or routine activities described under planning instruments (Part 4 EPAA) (Figure 2.2). If the impact of a development or activity is predictable or minor, then the local council must certify it (EPAA s. 76A(l)) but requires no assessment on the merits. However, for other development covered by Part 4 EPAA, a development application must be lodged and an assessment of the development is made on the merits (EPAA s. 78A). Development consent is usually obtained from council and usually has conditions attached. In some situations, development consent is obtained from the Minister for Urban Affairs and Planning. These situations include state significant development (e.g. major industrial sites that generate employment, large waste disposal facilities and developments that affect

10 Zoning provisions offer a form of predictability to owners, developers and conservationists by prohibiting development or permitting development in particular areas (Farrier et al., 1999) (EPAA Part 3).

11 DCPs institute specific and comprehensive guidelines for individual types of development or over specific areas. They are not legally binding but their provisions must be taken into account by councils in deciding whether or not to grant development consent. Figure 2 1 9 .§ development consent required by the provisions of ~ a LEP, REP or (SEPP) (EPAA Part4) ""i N l t-; ~ development: local (EPAA s. 76A(1)) or of state ~· significance (EPAA s. 76A(7)-(9)) iiS"" ...... l §. $::) 8 part test (EPAA s. 79C), requirements set out under ;::s (TSCA s. 94 (2)) ~ ~ l ""i ~ no significant development application (EPAA s. 78A) . effect (no SIS §' required) ~ l ~ significant effect found, following application of B consent• no consent• ('1:) ("") part test I I I I

~· ~ ..__ listing concurrence or consultation with NPWS (TSCA Part 2) SIS required (EPAA s. 77(3) , 79A(B)(b), 90(1)(c2),(c3). Contents required if (1) likely significant impact or (i1} of SIS set out in TSCA s. 110(2) critical habitat occurs (EPAA s. 798(3)-(7)) I l l l public exhibttion (30 days) (EPAA granting of concurrence refusal of concurrence (must s.79A) (uncondltional or subject to give reasons) conditions) (EPAA 79B(8)-(12)) (EPAA 79B(8)-(12)) l

evaluation of developmen ,_ recovery plans application (EPAA s. 79CJ (TSCA s.69(2)) • I • consent l I no consent I Figure 2.2: Most significant development in New South Wales requires consent under Part 4 of the Environmental Planning and Assessment Act 1979 (NSW). This allows for the consideration of threatened species, populations and ecological communities during evaluation of a development application. The italicised box identifies steps that require VJ 0 scientific knowledge to make a decision. Chapter 2 Legislation and Threatened Species 31

important natural environments) (EPAA 76A(7)-(9)) or where specific policies or REPs identify the Minister as the consent authority.

There are three types of development requiring consent where additional issues are covered as part of the consent process. These include integrated development (requires a permit or licence from a state government authority in addition to development consent under the EP AA), development that requires concurrence (requires agreement from a state agency such as NPWS before development consent can be given) and designated development12 listed in schedule 3 of the EP AA Regulation 2000 (due to its potential impact on the environment, such as highly polluting industries near wetlands), which requires the preparation and consideration of an environmental impact statement.

Local councils (the major consent authorities) have become especially important players in conservation and have a new level of responsibility for planning and development under the EPAA (Kelly 1996; Lim 1996) and biodiversity management.

The TSCA has attracted many criticisms, largely stemming from the procedural nature of the legislation (Smith 1997). The TSCA largely establishes procedures 13 for discretionary decision-making . These decisions are sometimes impossible to track and will almost certainty lead to inconsistency for biodiversity protection. On the one hand, the Act is criticised for being a complicated piece of legislation. On the other, it is criticised for being a poor instrument for achieving conservation. Specific criticisms include: (i) mechanisms are too bureaucratic and discretionary (Hartcher 1993; West 1991); (ii) too much responsibility is given to local government; (iii) the licensing system is both lengthy and covers a range of actions; and (iv) eight part tests and SISs are expensive for proposers of actions.

12 Designated development is not discussed in Figure 2.2 and are mentioned, as a development type to be aware exists in the legislation.

13 The exercise of discretion occurs when a decision is made by a decision-maker exercising judgement after taking into account criteria specified in the legislation (Farrier 1999). Chapter 2 Legislation and Threatened Species 32

The purpose of this chapter is firstly to question whether the TSCA and EP AA (Part 4) does (or can) achieve successful biodiversity conservation and secondly to examine whether any legislation could be an effective tool in biodiversity conservation. I begin by examining the individual components: listing, eight part tests, species impact statements and other elements such as recovery plans and threat abatement plans. Two case studies are outlined to illustrate the biodiversity conservation issues and criticisms of the TSCA. These case studies involve: (1) urban development of a site containing a population of Darwinia biflora at Green Road, Glenhaven, and (2) urban development of a site containing a population of Tetratheca glandulosa at Aquatic Drive, Allambie Heights.

2.3 The Threatened Species Conservation Act 1995 (NSW) and Environmental Planning and Assessment Act 1979 (NSW) Part 4

There are five mam components m the TSCA and EP AA that influence the conservation of threatened species. The five components are: (i) listing of threatened species (TSCA Part2), (ii) "8 part tests" of likely significant affect (TSCA s 94, EP AA s SA), (iii) Species Impact Statements (SIS) (TSCA Part 6 Div. 2), (iv) recovery planning for listed species (TSCA Part 4), and (v) approval of development (EPAA Part 4) and activities (EPAA Part 5; TSCA Part 6; NPWA Part 8A). Listing of a threatened species triggers the various other components (i.e. eight part tests, SISs, and recovery plans) (Smith 1997).

2.3.1 Listing

The listing process relies on current scientific understanding of the evolutionary development of species and ecological communities, and the genetic diversity of populations. While populations14 and ecological communities15 are offered

14 Population means a group of organisms, all of the same species, occupying a particular area (TSCA s. 4(1)).

15 Ecological community means an assemblage of species occupying a particular area (TSCA s. 4(1)). Chapter 2 Legislation and Threatened Species 33

protection only if they are endangered16 (TSCA s. 6(2), (3), listed in Schedule 1, Parts 2 & 3), species17 can be listed as either endangered (TSCA s. 6(1), listed in Schedule 1, Part 1), vulnerable (TSCA s. 7, listed in Schedule 2) or extinct (TSCA s. 6(4), listed in Schedule 1, Part 4). A species can be listed as endangered if there is a likelihood of it becoming extinct, through factors threatening its survival or evolutionary development (TSCA s. lO(a)) or the reduction of individuals or loss of habitat to what is considered a critical level (TSCA s. lO(b)). A 'vulnerable' species is classified as a species that is likely to become endangered unless the factors that are affecting its survival and evolutionary development are removed (TSCA s.14). An endangered population is characterised by being disjunct, near the limit of its geographical range (TSCA s. ll(a)), likely to be genetically distinct (TSCA s. ll(b)) or of significant conservation value (TSCA s. ll(c)). At least one of these core criteria needs to be met, as proving that a species is at the limit of geographical range is difficult in the absence of historical records. Eligibility for listing an endangered ecological community revolves around two issues. Firstly determining whether the definition of an ecological community is met in a particular case and secondly, proving that the community is likely to become extinct if the factors threatening the survival and evolutionary development are not removed (TSCA s. 12(a)).

A process may be listed (Schedule 3) (TSCA s. 8) as a key threatening process (e.g. clearing of native vegetation; high fire frequency, resulting in the disruption of life cycle processes in plants and animals and loss of vegetation structure and composition; and invasion of native plant communities by Chrysanthemoides monilifera (see www.npws.nsw.gov.au for more detail) if it is adversely affecting two or more listed species, populations or ecological communities (TSCA s.

16 Endangered ecological community means an ecological community specified in Part 3 of Schedule 1. An endangered population means a population specified in Part 2 of Schedule 1 (TSCA s. 4(1)).

17 Species of animal or plant includes any defined sub-species and taxon below sub-species and any recognizable variant of sub-species or taxon (TSCA s. 4(1)). Fish and marine vegetation are the only organisms that are not listed under the TSCA. They are covered by their own legislation, the Fisheries Management Act 1994 (NSW) (FMA) (Schedule 4) (FMA s. 220C) with the Fisheries Minister having greater involvement of the listing process (FMA s. 220G). Chapter 2 Legislation and Threatened Species 34

15(a)) or if it will cause species, populations or ecological communities not yet threatened to become threatened (TSCA s. 15(b) ).

Threatened species can be listed on schedules of both the TSCA and the Commonwealth threatened species legislation, the Environment Protection and 18 Biodiversity Conservation Act 1999 (Cth) (EPBC) (Division 1, subdivision A) • If a species or community (indigenous to NSW) is listed as a Commonwealth threatened species or community, it must be considered for listing under the TSCA, and the scientific committee19 must initiate the listing (TSCA s. 9). The scientific committee can initiate any other listings (TSCA s. 18) but members of the public can trigger the listing process (TSCA s. 19). The Minister may also initiate the listing process (TSCA s. 18 (2)(a)). Each listing must contain the information required by the regulations (TSCA s. 19 (5), 20). However, currently there are no regualtions so TSCA s. 19(5) is inoperative. Once the scientific committee has made a preliminary determination (TSCA s. 22(1)), a notice must be published announcing the determination and reasons for it (TSCA s. 22 (2)(c)). The public are allowed to comment and the scientific committee must consider all written submissions (TSCA s. 22 (5)). However, the scientific committee makes the final determination (TSCA s. 23(1)).

The listing of species has historically been viewed in Australia and overseas as a powerful mechanism forming a key part of the protection of biodiversity. The TSCA retains listing as its central feature, acting as a trigger for other mechanisms. The strength of listing is believed to come from placing limitations on whether certain pieces of land can be disturbed by a development or what

18 At a Commonwealth, level species may be listed in Schedule 1 as extinct, extinct in the wild, critically endangered, endangered, vulnerable and conservation dependent. Ecological communities are listed in Schedule 2 as critically endangered, endangered or vulnerable. Key threatening processes are listed in Schedule 3.

19 The Scientific Committee is a technical committee and is not controlled or influenced by the Minister. Neither Local and State government nor the wider community or interest groups is represented. Under the Fauna Interim Protection Act, the Scientific Committee used a precautionary principle approach to listing species. Listings were initiated through surveys sent to others scientists studying a particular species. Chapter 2 Legislation and Threatened Species 35

activity can take place there (Lim 1997). As listing can have an impact on land development, the process is bureaucratic by nature. The USA has struggled with this issue. Under the Endangered Species Act 1979 (USA), for example, some species have become extinct while waiting for listing to take place (Smith 1997). N.S.W. has also struggled with the issue of listing threatened species. Even though this new listing regime, under the TSCA, allows for the recognition of endangered populations, ecological communities and critical habitat, it would appear that the 20 regime is currently geared towards individual species . Initially the schedules contained only species, however the listing of communities, especially in the Sydney basin is proving to have a practial conservation impact. The listing of threatened species requires many criteria, in particular the recognition that individual numbers and habitat have been reduced or the 'evolutionary development' has been compromised. There are two important underlying assumptions in the listing process that are strongly questioned by the lack of ecological knowledge. First, there is an assumption that data on changes in species distribution and numbers, for any species, are well enough known or can be readily gathered. Further there is an assumption here that scientists can define the measurable factors that would constitute limits to 'evolutionary development'. Second it is assumed that data of this sort are available and have been examined for all species.

2.3.2 Eight Part Test

The "eight part test" (TSCA (s. 94) and EPAA (s. SA)), forms a package of assessment tools, along with species impact statements (SIS) (see section 2.3.3). The eight part test is essentially a checklist to help determine if a particular development is likely to have a significant impact on a listed species. If this is the case, an SIS must be completed.

20 As of August 2000, 766 species had been listed compared with 17 populations, 28 ecological communities and 7 key threatening processes. Chapter 2 Legislation and Threatened Species 36

The "eight part test" operates within all situations where approval for development or activities is required (i.e. licence under the TSCA, EP AA Part 4 and EPAA Part 5). Eight different criteria are taken in to account when assessing likely significant effect (TSCA s. 94(2); EP AA s. SA) (Table 2.1 ). Any amelioration proposed is taken into account in applying the eight-part test and should take the form of a technical report (Smyth v Nambucca SC [1999] NSWLEC 226; Kelly 1996). The eight part test though not entirely focused on species enforces a species-based approach, because of the nature of the eight questions. As it does not require a detailed scientific analysis, the information applied is often generic (e.g. extrapolated across species, genera and even families), neglecting important site and habitat specific factors that may be essential for the survival of a species within a particular site. Responsibility for who should complete an eight-part test is ambiguous in the legislation, often leading to an amelioration process between local council and developers.

A thorough application of the eight part test presents many "challenges" to ecological knowledge (Table 2.1). In most cases, scientific studies are required to address many of the questions in the eight part test. For example, part (a) asks whether the life cycle of the threatened species is likely to be disrupted. To address this criterion demographic studies are needed not only to determine the life cycle of the threatened species in question but what factors might disrupt it and importantly what constitutes a viable population. This information is virtually certain to vary between species and, as argued above is lacking for most species.

2.3.3 Species Impact Statements

A Species Impact Statement is required when the eight part test indicates that there is likely to be a significant effect (TSCA s. 110). An SIS must be in writing and signed by the principal author and the applicant (TSCA s. 109). The SIS can be in any format but must cover the issues set out in section 110 of the TSCA. Table 2.1: An eight part test determines if a significant effect will occur from a proposed development or activity upon a listed species. An eight part test needs to consider eight Q different components of species ecology with each presenting a different ecological challenge. -§ ~ Part of the Eight Part Test Ecological "Challenge" Problems with the ""'! N Application of this Part of the Test t°""" (a) in the case of a threatened species, whether the Zife cycle of A complete understanding of all aspects of the life Very little information about life cycles for threatened species is ~ the species is likely to be disrupted such that a viable local cycle, the factors that might disrupt it, and what available and what constitutes a viable population differs between ~· population the species is likely to be placed at risk constitutes a viable population for that particular species. Assumptions are often made across families and habitats. E;" of of ..... extinction species (demography studies). Extent of 'Risk of extinction' is difficult to define due to varying data in the 15· knowledge needed will depend on development. literature and PVA modelling. ~ $::i ~ (b) in the case of an endangered population, whether the life NIA NIA ~ cycle of the species that constitutes the endangered population ~ is likely to be disrupted such that the viability of the population is likely to be significantly compromised ~ ~ ( c) in relation to the regional distribution of the habitat of a To determine how much habitat is going to be Deciding if the modification of this amount of habitat is 'significant' ~ threatened species, population or ecological community, removed and what ratio this is to the total habitat is a subjective judgment. No guidelines have been set down. whether a significant area of known habitat is to be modified or can be determined by the use of distribution maps. Significance of the impact will vary between species and will ~ removed However is a region an IBRA region or another increase as remnant sites are successfully developed. ~ measure. ~· (d) whether an area of known habitat is likely to become Determining if a habitat is isolated involves a The notion of isolation differs between species; lOOm can be a isolated from currently interconnecting or proximate areas of complete understanding of the life cycle. significant distance for a seed, insect or bird depending on the habitat for a threatened species, population or ecological intervening lOOm (e.g. bare lands, road or water might deter some community species).

( e) whether critical habitat will be affected There should be no problems because critical habitat will be defined and mapped. There are no grounds for challenging.

(j) whether a threatened species, population or ecological Often this information is obtained from a map of Deciding if a threatened species or habitat is adequately represented community, or their habitats, are adequately represented in the species distribution. The adequacy and in conservation areas is subjective. Once again this information will conservation reserves (or other similar protected areas) in the effectiveness of the surveying methods is often vary between species and no guidelines have been established. region questioned.

(g) whether the action proposed is of a class of action that is Determining threatening processes is a general While key threatening processes must be listed this provision is not recognised as a threatening process ecological challenge. confined to KTPs. Other threatening processes cannot be readily identified.

(h) whether any threatened species or ecological community is see (f) This information is often obtained from a distribution map of the VJ -.....) at the limit of its known distribution threatened species whose accuracy and completeness are questionable. Chapter 2 Legislation and Threatened Species 38

These include site-specific effects that the activity or development will have (Table 2.2).

The SIS is assessed by the Director-General NPWS and used as the basis for determining whether concurrence21 will be given (TSCA s. 95). The Director­ General is required by legislation to institute arrangements for the accreditation of suitably qualified and experienced persons to prepare assessment reports on SISs for the purposes of the Act (TSCA s. 113(1)). No accreditation system has ever been introduced. However, there is no requirement for those preparing SISs in the first place to be accredited. Responsibility for the SIS rests with the development applicant who submits the application to the consent authority (EP AA s. 77). If an Environmental Impact Statement (EIS) is also required, the SIS may form part of this (TSCA s. 109(2)(b); EPAA s. 79C).

When a development application does not contain an SIS, but is nevertheless approved, the development application can then be challenged in the Land and Environment Court on the basis that an SIS should have been prepared (Timbarra Protection Coalition Inc v Ross Mining NL [1998] NSWLWC 19; Kelly 1996). An SIS can also be challenged if the document is deemed inadequate, even if Council has already approved the development.

Where there is likely to be a significant effect, the concurrence of the NPWS is required. This means that the NPWS can veto consent being given to the proposal (TSCA s. 111). In determining concurrence, the Director-General no longer considers just the conservation issues but needs to take into consideration all social and economic consequences of accepting or rejecting a development application. However, this process has been viewed by Kelly (1996) as simply a means of displaying environmental diligence, hence legitimating the existing pro­ development factors (economic and social benefits) embedded in the EP AA.

21 Concurrence is required if critical habitat is present at the site or there is a likely significant effect on a listed species, population or ecological community (EPAA s. 79B (3)-(7)). Concurrence is also required from the Environment Minister if a significant impact occurs when no development consent is required under Part 5 of the EPAA (i.e. development by public bodies or private development requiring an approval other than development consent). Q Table 2.2: Species Impact Statements provide a site assessment for a development or activity in relation to threatened species. Each component of an SIS provides an -§ ecological challenge with many problems depending on available information . ~ N Legislation Section of SIS Ecological "Challenge" Problems with Application of this t--< Section of an SIS ~ (1) a fell description of activity proposed in the development 1;;· NIA NIA S' application, including nature, extent, location, timing and layout of ..... 5· the proposed development or activity ;:::

s:::i ;::: (2a) a general description of the threatened species known or likely to None, a general descritpion of the ecological The issue of interpreting the likelihood s:::i... be present in the area that is the subject of the action and in any area characteristics of the species is required. of a species being in an area. Such as it is ~ that is likely to be affected by the action often difficult to determine seedbank ~ information. ~ ;::: (b) an assessment of which threatened species known or likely to be A complete knowledge of the life history of Often complete life histories are ~ present in the area are likely to be affected by the action the species is required. unobtainable, therefore predicting the ~ effect is subjective/guesswork p;

~· ( c) for each species likely to be affected, detail of its local, regional To determine what threatening processes are Generalisations from examples within and State-wide conservation status, the key threatening processes affecting a species. the same genus or family are likely to generally affecting it, its habitat requirements and any recovery plan occur but this may not be appropriate. or threat abatement plan applying to it Also the recovery plan may not be written.

( d) an estimate of the local and regional abundance of those species This information could be obtained from a This relies on adequate surveying for the distribution map, data are usually poor threatened species in all likely habitat

( e) a general description of the threatened species known or likely to see (2a) NIA be present in the area that is the subject of the action and in any area that is likely to be affected by the action

(j) a description of type, location, size and condition of the habitat Assessing the condition of habitats can be Assessing condition of habitats is and details of the distribution and condition of similar habitats in the subjective from the perspective of the species difficult. Species disturbance regimes region in question, and disturbance regimes, which need to be known (e.g. some species affect these conditions, are difficult to assess thrive in highly disturbed areas). immediately.

(g) a fell assessment of the likely effect of the action on those species, Detailed life histories need to be known. Ecological experiments are needed to including, if possible, the quantitative effect of local populations in However, the effects are going to be site­ obtain this information. To properly the cumulative effect in the region. specific and activity-specific. Assessing assess cumulative impacts a long term cumulative impacts before a development is monitoring programme is required. also subjective. (h) a description of any feasible alternatives to the action that are Needs to be decided on a case by case basis. Needs to be decided on a case by case likely to be a lesser effect and the reason justifying the carrying out of basis. the action in the manner proposed, having regard to the biophysical, economic and social considerations and the principles of ecologically sustainable development,

(i) a fell description and justification of the measures proposed to How an action is going to affect a species is Generalisations might occur, as adaptive mitigate any adverse effect of the action on the threatened species not always apparent. Often mitigation occurs management does not have a set place using development controls (e.g. corridors) within the legislation. used on questionable science.

(j) a list of any approvals that must be obtained under any other Act NIA NIA or law before the action may be lawfelly carried out, including details of the conditions of any existing approvals that are relevant to the species or EQ[J_Ulations. t>.l \0 Chapter 2 Legislation and Threatened Species 40

Technically, an SIS demands a high level of ecological knowledge (e.g. detailed knowledge of the ecology of a threatened species, and its likely responses to the proposed activity) and expertise to evaluate any impact (Table 2.2, Ecological "Challenge"). However, ecology often does not have the answers to the questions, which flow from an attempt to follow requirements set out in the legislation. Nor, in practice, is science given the time to explore further, thus forcing a subjective decision and leaving considerable margin for poor performance. For example, part ( d) asks for an estimate of the local and regional abundance of a species, which should rely on adequate surveying in all likely habitat at an appropriate time (e.g. flowering time for cryptic plant species). Due to the lack of ecological data, the adequacy of the SIS is usually based on "reasonableness" rather than on an expectation of scientifically defensible methodologies.

When the protection of threatened species outweighs the potential economic outcomes of a development or activity, amelioration between developers and consent authorities can occur (TSCA s. 110 2(h)). Like all previous planning and environmental legislation the TSCA is merely a way of assessing proposals, which are going to affect high-profile organisms (Kelly 1994).

2.3.4 Other elements of the legislation

Listing of a threatened species also triggers compulsory recovery planning. The Director-General must prepare recovery plans within 5 years of listing for endangered species, and within 10 years of listing for vulnerable species at the time of proclamation (TSCA s. 56(3)). A successfully implemented recovery plan should remove a species from its present listing and ensure its viability in nature in the long term (TSCA s. 56(1)) (Brebach 1997; Smith 1997). Once a recovery plan has been signed off, Ministers and public authorities must take appropriate action to implement the measures in the plan and make no decisions inconsistent with the plan (TSCA s. 69(1)-(3)).

Recovery planning is outlined in Part 4 of the TSCA with four requirements: (1) recovery plans must meet the six objects of the Act (TSCA s. 57(a)); (2) all social Chapter 2 Legislation and Threatened Species 41

and economic consequences need to be considered (TSCA s. 57(b)); (3) all available resources need to be used effectively and efficiently (TSCA s. 57 (c)) and, ( 4) if social and economic consequences are to occur, these need to be minimised (TSCA s. 57(d)). The requirements for what a recovery plan is to contain are set out in section 59 of the Act (Appendix 2.2).

Key threatening processes are listed in schedule 3 of the TSCA. A threatening process can be listed if it impacts upon two or more threatened species, populations or ecological communities (TSCA s. 15(a)) or causes a species, population or ecological community to become threatened (TSCA s. 15(b)). The listing of a threatening process follows the same procedures as listing a threatened species. Once a threatening process has been listed, a threat abatement plan must be prepared by the Director-General to abate, ameliorate or eliminate any adverse effects on threatened species (TSCA s. 74(1)). Threat abatement plans must take into consideration the objects of the act (TSCA s. 75(1)(a)), any social and economic consequences (TSCA s. 75 (l)(b)), efficient and effective use of resources (TSCA s. 75 (l)(c)) and the desirability of minimising any social and economic consequences (TSCA s. 75 (l)(d)). The requirements for what a threat abatement plan must contain are set out in section 77 of the TSCA. Public authorities (including the Director-General and local councils) are to take the appropriate action to implement the plan for which they are responsible (TSCA s. 86(1)). Local councils can be identified as responsible for implementing measures included in the plan and must report on these in an annual report (TSCA s. 87 (2)).

2.4 Two case studies

The following two case studies are set out to examine and to integrate the themes discussed in this chapter. These themes include how the eight part test and SIS are implemented and ecological knowledge is incorporated into decision-making. Chapter 2 Legislation and Threatened Species 42

2.4.1 Eight part test (Green Road, Glenhaven)

A housing sub-division at Green Road Glenhaven (see section 3.3 for background information) contains a population of Darwinia biflora, listed as vulnerable, comprising 80 individuals within a conservation zone linked to surrounding bushland by a corridor. Provision for a conservation zone and corridor were incorporated into the eight part test following consultation with the local council. With such an amelioration process built into the test, local council determined that there would be no significant effect from the sub-division on D. biflora locally, regionally or state wide. Therefore, no SIS was required. The use of bushland corridors in conservation programmes has attracted mixed reviews (see section 1.1.3 for a comprehensive discussion) from the scientific community over the last ten years (e.g. Forney & Gilpin 1989; Lindenmayer et al. 1993; Andreassen et al. 1996; Bentley & Catteral 1997), producing an extensive body of literature mainly from North America. The available scientific knowledge is therefore important when assessing the effectiveness of the decision to create a bushland corridor on the site. The bushland corridor was used to address part (d) of the eight part test (Table 2.1) i.e. it was assumed that it allowed for cross pollination via insects associated with the species, even though the habitat was unsuitable for D. biflora. This assumption was made without any scientific data about the use of corridors by pollinators and plant population viability. This example highlights the danger of basing management strategies on inappropriate scientific data contained within an eight-part test, leading to an inadequate decision making process.

2.4.2 Species Impact Statement (Aquatic Drive, Allambie Heights)

A housing subdivision (28 lots) at Allambie Heights, NSW (see section 3.2 for background information), has been at the centre of extensive debate between the local council, developers and local conservation groups. Tetratheca glandulosa, a species listed as vulnerable on the TSCA, occurs at the site, representing one of the most southern known populations. An eight part test was completed by a survey team from CSIRO with the conclusion that there would be no significant effect of the development on any threatened species likely to be present. The Chapter 2 Legislation and Threatened Species 43

existence of T. glandulosa on the site was disputed by the developer as no individual or population was discovered during surveying of the site. Tetratheca glandulosa is a cryptic species when not in flower, making it difficult for individuals and estimate population sizes. A botanist employed by the local council identified T. glandulosa on the site. The developer therefore produced an SIS. The SIS concluded that even if a population of T. glandulosa occurred at the site it would be non-critical to the survival of the species due to its presence in eight different reserves, which cover the full geographic extent, therefore conserving the species on a local, regional and state wide level. The SIS came to this conclusion without investigating any of the fundamental ecology of the species.

2.5 Conclusions

In this chapter, I have shown that for two particular instances the TSCA and its combination with the EP AA (Part 4) have not favours biodiversity outcomes. I have set out in some detail the extent to which the law relies on ecological knowledge. And the two case studies highlight this inadequacy in conservation by demonstrating the pressure ecological knowledge is placed under to give answers and solutions to problems. Chapter 3 Case studies 44

CHAPTER 3 CASE STUDIES

3.1 Introduction

Decision-making frameworks for planning and threatened species conservation are determined by legislation: the Threatened Species Conservation Act 1995 (NSW) (TSCA) and the Environmental Planning and Assessment Act 1979 (NSW) (EPAA). These pieces of legislation place pressure on science to operate in a particular manner, potentially providing inappropriate outcomes. To overcome the many problems in the existing decision-making framework, an input from science would be required early in the planning process (Figure 3.1). In an ideal world, scientific knowledge and the process of science would be incorporated into the framing of legislation and its application to policy. Science could thereby contribute effectively to the approval and decision-making process for the conservation of threatened species, populations and ecological communities. I have identified six elements that need revision or consideration in an improved decision-making process (Figure 3.1): (I) the formal submission of development applications; (II) consultation between/with Council and independent scientists; (III) the use of accreditation and peer review systems; (IV) the formal and informal role of the NPWS; (V) the use of development conditions, development modifications, and adaptive management; and (VI) the role of the Land and Environment Court.

The process of making a development application is comprehensively set out in legislation (TSCA and EPAA). Having a strong legislative backbone to the decision-making process allows for a direct path for the submission of a development application. It provides a structure from which to build an ideal decision-making process without a complete change in legislation (Figure 3.1, section I).

Requiring consultation between the proponent (applicant) and Council at the preliminary scoping stage of a development proposal would allow the developer to familiarise themselves with the strategic planning (e.g. zoning (EP AA Part 3), SEPPs (EP AA s. (37)- Chapter 3 Case Studies 45

Planning instruments such as: Zoning (EPAA Strategic Planning , SEPPs (EPAA s.(37)-(39)) ~~ , Part 3)* (EPAA s.5(a)(i))* REPs (EPAA s.(40)-(52)) LEPs (EPAA s.(53)-(72))*

T------Development 1 ~ Preliminary• research standards Developer seeks council for development I • (EPAA s.4(1)) '------'-_J..... consultation proposal

• Developer seeks independent Flora and fauna II , scientific input from survey* - researchers in the field*

8 part test comp• leted by developer or accredited , Local council to decide if an SIS ---+---~, consultant (EPAA s.5A; is required (EPAA s.77(3))* TSCAs.94)*

Species Impact• Statement (TSCA s.110) completed by an accredited person ....- I (TSCA s.113(1 ))*

Development• application ,_ submitted by developer , (EPAA s.78A) IV • Concurrence of NPWS if Evaluation of development ...-.._____,___ SIS shows a significant application (EPAA s.79C) ,.... effect (EPAA s.798(3)-(7))*

------!;------_'.______

Development conditions Development Application v .___ __y .. determined by NPWS with use application .,.No­ approved by ,.,___---+<~ e ! .... of up-to-date scientific withdrawn council I knowledge* I VI No Appeal to the Land Appeal Appeal Development and Environment dismissed upheld begins Court (EPAA s.97(1))*

Modification of development controls

Management plan with council and Monitoring during NPWS input, including monitoring and development (adaptive adaptive management and assessment management)* of development control success*

Figure 3.1: Flow chart showing an idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part). An italicised box refers to a step, which would be ideal but is not currently part of the decision-making process, as required under the legislation. Six parts have been identified: (I) formal production of development applications, (II) consultation with council and independent scientists, (III) the use of accreditation peer review systems, (IV) role of NPWS, (V) development conditions, and (VI) the Land and Environment Court. Chapter 3 Case studies 46

(39)), LEPs (EPAA s. (53)-(72) and REPs (EPAA s. (40)-(52)) and their expectations in relation to developments (e.g. retention of a percentage of native vegetation) (Figure 3.1, section II). Consultation with independent scientists (at the expense of the proponent) in an expert capacity would allow the incorporation of up-to-date knowledge in the development application process (e.g. ensure comprehensive SIS) (Figure 3.1, section II). Scientists should be accountable to an independent body of experts.

The present system makes no allowances for quality control or a feedback system to ensure that standards are met. This may partly explain to the frequent questioning of the independence and credibility of the science in development applications, and re-evaluation of information by conservationists and other opponents of developement. The incorporation of an accreditation and peer review system would allow independent input and appropriate use of scientific knowledge (e.g. extrapolation of theory) (Figure 3.1, section III). If an SIS had to be reviewed by independent scientific peers this would remove the pressure on a consultant to find in favour of the developer. A form of peer review would add strength to the conclusions of an eight part test and SIS by allowing an increase in scientific independence and assurance.

The National Parks and Wildlife Service (NPWS) can play an informal or formal role in the decision-making process. The formal role of concurrence by the NPWS is entrenched in the existing system (EPAA s. 79B(3)-(7)) (Figure 3.1, section IV). The concurrence of the Director-General of the NPWS is required when the development is on land where critical habitat exists or when the SIS shows that there is likely to be a significant effect on a threatened species. However, the NPWS can play an informed consultative role, and sometimes does so. The NPWS is not required to assess eight part tests or SISs (unless concurrence is required). However, it could play a beneficial role in aiding Council to assess not only private development but also Council's own development applications.

Development conditions and adaptive management processes with modification of developments and monitoring, can be powerful tools in making decisions where threatened species could be impacted (Figure 3.1, section V). Often the answers given by science are only relevant to specific time and space or even unavailable during the initial survey and Chapter 3 Case studies 47

decision-making process. The involvement of monitoring during development, with a provision for modifications of conditions incorporated in development consents, allows for increased accuracy in the detection of impacts upon threatened species, populations and communities. This form of decision-making recognises the unavailability of knowledge and the role of the precautionary principle and enforces the assessment of the success of development conditions. This situation is not currently part of the decision-making process and is not required under legislation.

The Land and Environment Court of New South Wales forms an essential part of the decision-making process (EPAA s. 97(1)) allowing for merit appeals and input by concerned third parties (Figure 3.1, section VI). The following close examination of cases dealt with by the Land and Environment Court demonstrates how different Councils have been using the current decision-making regime.

The purpose of this chapter is to outline an ideal decision-making process and use a number of case studies to test this framework in relation to existing practice. The following case studies (Figure 3.2) have been selected to show the shortcomings and strengths of the present decision-making process for threatened species conservation. Housing development at Aquatic Drive, Allambie Heights (section 3.2) demonstrates the present legal demands on science during the Species Impact Statement process and the onus of providing accurate information. Housing development at Green Road, Glenhaven (section 3.3) shows that proposed amelioration prior to the lodgement of a development application can facilitate development approval without adequate scientific input. Development controls which lack scientific approval are also highlighted in this case study in the use of 'bushland corridors' for pollinator movement with no knowledge of pollinators. Recognition of the importance of scientific study (in relation to seed banks) is highlighted in the case study of a proposed playing field at Carnarvon Drive, Frenches Forest (section 3.4). A proposed housing subdivision at Grosvenor Street, North Wahroonga (section 3.5) demonstrates the robustness and weaknesses of science (the existence and location of seed banks) and the essential role of the precautionary principle. Chapter 3 Case Studies 48

1~15 22 r5 ,~/µ 38 ' Broken Bay

* (2) Glenhaven ~> * (4) North Wahroonga r;;:.fj_-

~ -fc (3) FrencCorest tl ~ * (1) ~ambie Heights ~"ammw \~~

~~j~~.,._, ):"0Port Jackson Kilometres ~ ?J":l~~ ~~ ~} -----ii ~ d ~i"'a-'{;-~ ) 0 5 10 Sydney

Figure 3.2: The location of sites discussed in the case studies: (1) Aquatic Drive, Allambie Heights (T.glandulosa); (2) Green Road, Glenhaven (D. biflora); (3) Carnarvon Drive, Frenches Forest (T. glandulosa); and (4) Grosvenor Street, North Wahroonga (T. glandulosa, D. biflora). Chapter 3 Case studies 49

This chapter also provides the background to these case studies of Aquatic Drive, Allambie Heights and Green Road, Glenhaven that will later be discussed in reference to ecological data presented in Chapters 5, 6 and 7. An understanding of the present legal processes and how these differ in a number of situations also allows for a later discussion (Chapter 8), examining the interaction of science and law/policy and how this influences decision­ making processes and ultimately conservation of threatened species.

Methodology

Information collated for each of the case studies discussed below was obtained using the following methodology.

(1) Local council areas in which Tetretheca glandulosa and Darwinia biffora occurs were identified using species distribution data from NPWS databases. (2) The environmental officers at the identified Local Councils were then contacted and discussions held to obtain: (a) location of sites, using Local Council databases; (b) documentation for each of the sites (documents included SISs, eight part tests, development proposals, reports prepared by consultants, site history (including zoning and development plans), site maps and management plans). (3) Discussions were held with NPWS threatened species officers (Central Directorate), if they had been involved to determine the role played by NPWS in the development process (e.g. advice on the nature of the information needed in an SIS, concurrence if any, existence of recovery plans). (4) For the first case study (Aquatic Drive, Allambie Heights), a discussion was held with the developer (ARDEL Limited) to obtain some more detailed information on the proposed development of the site. (5) Sites were visited to confirm the locations of plant populations and describe the vegetation present. ( 6) Land and Environment Court judgements were obtained to describe the sequence of events in each of the case studies that involved litigation. Chapter 3 Case studies 50

3.2 Aquatic Drive, Allambie Heights (33°40'30" S 151°14'20" E)

This site was originally owned by the Spastic Centre, Australia, and was then sold to ARDEL Limited for $3 million in 1993 (Plate 3.1). Prior to the purchase of the land, the Local Environmental Plan (LEP) was amended in October 1992 to allow a spot rezoning from 'hospital and hostel purposes' to 'residential'. Spot rezoning allows for amendments to an IBP to allow a specific development to go ahead at a specific site (Farrier et al. 1999 p. 100). In other words there was no adequate planning at the strategic level.

The site is located in the local government area of W arringah Shire Council and forms the lower part of a rectangular portion of land fronting onto Aquatic Drive in the north (Figure 3.3). The land that fronts directly onto Aquatic Drive is owned by the Department of Housing. To the east are grounds used by the Spastic Centre for industrial and residential activities. Along the western and southern boundaries is a Crown road reserve, with the Pymble-Warringah water pipeline (1.8m high) further south in a Sydney Water Corporation reserve. Beyond this point is the Warringah War Memorial Park (the Manly Dam Reserve), the W arringah Aquatic Centre, a sports field, a carpark and "Eurobodalla" (a retirement/nursing home).

3.2.1 Vegetation attributes of the site

The site gently slopes from the north to the south, with lateritic gravel soil overlying Hawkesbury Sandstone, forming a sandy loam soil. Prior to development of the site, a number of different ecological communities (Table 3.1) were identified, many of which are Chapter 3 Case studies 51

Plate 3.1: Site at Aquatic Drive, Allarnbie Heights owned by ARDEL Limited.

Plate 3.2: Population of Tetratheca glandulosa at Aquatic Drive Allambie Heights (identified by pink flowers). Chapter 3 Case Studies 52 ///// .. ··////·/·/· /"/,, , .. /////////···/· / Aq~atic I>ri~~ / // / // .. ///// .. // ////////

Department of Housing Land Crown Road Reserve

./~:·:·· .//,/ .. · ,,,,./'"""'/' ... Transmission Line Easement / ""'~.. -""'' ... ;)' $' // ,,/ / ../~-:-~~~ ...... / )/ ... /' ,,,,.------·/' _...... ·" / f Spas ti c Centre Groun ds

ARDELLand A Duffy's Forest (Lis ted endangered B ecological community) Sandstone I I I Swamp 0 15 3 0 45 60 75m c Eucalyptus piperita, E sieberi, Corymbia Tetrath eca glandulosa -~ gummifera and Angophora ~ costata forest

D North Corymbia gummifera, Eucalyptus haemostoma and E. oblonga - woodland

\ Crown Road - Reserve

Figure 3.3: Approximate location of Tetratheca glandulosa (shaded) on the ARDEL site at Aquatic Drive, Allambie Heights (adapted from Smith & Smith 1996). The letters A, B, C and D represent different ecological communities on the site (see Table 3.1). Chapter 3 Case studies 53

Table 3.1: Ecological communities represented on the site found at Aquatic Drive, Allambie Heights (Letters A, B, C, and D refer to the location of the community on the site, see Figure 3.2). (Source: Ecohealth Services 1999)

Community Location Conservation Other dominant species status

A Duffy's Forest northern half of site listed endangered Eucalyptus sieberi L. Johnson, E. capitellata Smith, Corymbia gummifera (Sol. ex Gaertner) Hochr., Acacia myrtifolia (Smith Willd., Banksia ericifolia L.f., B. spinulosa Smith, Bossiaea obcordata (Vent.) Druce, Epacris pulchella Cav., Grevillea linearifolia (Cav.) Druce, Hakea sericea Schrader, Lambertia formosa Smith, Themeda australis (R.Br.) Stapf.

B. Sandstone swamp centre of site uncommon in Baeckea imbricata (Gaertner) Sydney Region Druce, Banksia ericifolia, Baumea and found in the rubiginosa (Spreng.) Boeck., Warringah Dillwynia floribunda Smith, Epacris Council area. obtusifolia, Hakea teretifolia (Salisb.) Britten, Lepidosperma filiforme Labill., Leptospermum squarrosum Gaertner, Restio fastigiatus R.Br., Sprengelia incarnata Smith

C. Eucalyptus piperita along creek and well represented Acacia longifolia (Andrews) Willd., Smith, E. sieberi, throughout the site and conserved Banksia ericifolia, Callicoma Corymbia gummifera locally and in the serratifolia, Ceratopetalum and Angophora costata State gummiferum, Dillwynia retorta (Gaertner) Britten (Wendi.) Druce, Grevillea forest linearifolia, Lomandra glauca (R.Br.) Ewart, Pultenaea daphnoidea Wendl.

D. Corymbia scattered well represented Acacia terminalis (Salisb.) Macbr, gummifera, Eucalyptus throughout the and conserved Banksia ericifolia, B. oblongfolia haemostoma Smith and southern half locally and in the Cav., B. serrata L.f., Dillwynia E. oblonga Blakely State retorta, Epacris microphylla R.Br., woodland Grevillea bu.xifolia (Smith) R.Br., G. speciosa (Knight) MacGillivary, Xanthorrhoea media R.Br. Chapter 3 Case studies 54

of local and regional significance (Ecohealth Services 1999) (Plate 3.2). The occurrence of ecological communities that are present have been a point of controversy and have been heavily debated by the developer, council and local green groups (Sydney Morning Herald, 17 June 1999, p. 6).

Tetratheca glandulosa, a plant species listed on Schedule 2 (vulnerable) of the TSCA is now known to occur along the mid-western boundary of the site, within the Duffy's Forest ecological community, in a population of over forty individuals. This population of T. glandulosa forms the southern most known population for this species distribution.

3.2.2 The development

The proposed development was divided into two sub-developments. The first development application was for a housing subdivision (creation of Torrens Title lots) consisting of 28 2 2 2 2 conventional lots (400m - 740m ) and 4 'super lots' with areas of 740m (lot 26), 1755m (lot 23), 1035m2 (lot 17) and 250m2 (lot 7). Five lots were allocated to Council for open space, natural drainage line, significant trees and a Water Quality Control Pond (WQCP). The second development application was for an access road via the Crown Road Reserve. The road is a 5.Sm wide road on the Crown road, past the Department of Housing Land and to the northwest corner of the site (Figure 3.3). This section of the development includes 175m of road within the Crown Road reserve.

The development application was submitted to Council without any prior consultation. Council assessed the merits of the development application and declined development consent. Following an appeal to the Land and Environment Court (ARDEL Limited v Warringah Council (No. 10606 of 1994 NSWLEC)), the developer modified the development and re-submitted the application to Council. Council assessed the development application a second time and consent was once again not given. Development consent was eventually obtained through the Land and Environment Court (Hassell Pty Ltd v Warringah Council (41NSWLEC1998)) subject to the implementation of conditions. Chapter 3 Case studies 55

3.2.3 The issues

The issues in this case study have been identified as: (1) the identification and presence of T. glandulosa on the site; (2) how an SIS can be assessed with inadequate ecological information; (3) the role of merit appeals in the Land and Environment Court (Council versus the developer); and, ( 4) the use of the development conditions imposed by the Court to implement a monitoring programme. These issues all have an impact upon the decision­ making process for assessing the development application for the ARDEL site (Figure 3.4).

Identification and presence of Tetratheca glandulosa

Tetratheca glandulosa is a cryptic species when not in flower, leading to difficulties in surveying for individuals and estimating population sizes. As will appear from the following discussions, this posed significant difficulties in relation to the assessment of development applications.

ARDEL Limited v Warringah Council (No. 10606of1994 NSWLEC)

The case involved a merit appeal (EPPA s. 97(1)) against the refusal of Council (18 November 1994) to give consent for the development application lodged on 13 April 1994, despite ARDEL amending their original development plans by reducing the number of housing lots (11 July 1994). There were three applicable planning provisions; Warringah Local Environmental Plan 1985 (LEP), Development Control Plan No. 19 - Allambie Heights (DCP 19) and State Environmental Planning Policy 1 No. 19: Bushland in Urban Areas (SEPP 19) .

1 SEPP 19 - Bushlands in Urban Areas states; protects and preserves bushland within certain areas, as part of the natural heritage or for recreational, educational and scientific purposes. The policy is designed to protect bushland in public open space zones and reservations, and to ensure that bush preservation is given a high priority when local environmental plans for urban development are prepared. Chapter 3 Case Studies 56

Planning instruments such as; Zoning (EPAA Strategic Planning------; SEPPs (EPAA s.(37)-(39)) Part 3) (EPAA s.5(a)(i)) REPs (EPAA s.(40)-(52)) LEPs (EPAA s.(53)-(72)) -T·------,--_,.. -:------Development standards Pre'liminary research f----+I __ Developer seeks counci (EPAA s.4(1)) for development consultation proposal

Developer seeks independen Flora and fauna II scientific input from survey* researchers in the field*

8 part test completed by Local council to decide if an SIS developer or consultant ---'"---!: is required (EPM s. 77(3))* (EPAA s.5A; TSCA s.94)* i ! s&s~~ ~~~~~~~~~:.:: !--===-:::: -- -···--Ill - -···· by an accredited person ---1---1 oredibi/ity by independent scientists* (TSCA s.113(1 ))* ,______

Development application ____ _, Consultation with NPW$ submitted by developer i (intormat PfbOlid}* (EPAA s.78A) IV

Evaluation of development ___,_ _, application (EPAA s.79C)

Application v No approved by >----Ye,--_...,., council

VI No Appeal to the Land ~~~-; Development and Environment begins Court (EPAA s.97(1))* ~~~=~ I j i ------~------J

Management plan with council and NPWS input, including monitoring and adaptive management and assessment of development control success*

Figure 3.4: Flow chart showing the decision-making process for housing sub-division at Aquatic Drive, Allambie Heights (Warringah Council) (Hassell Pty Ltd v Warringah Council). Overall chart represents idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part) presented in Figure 3 .1. An italicised box refers to a step, which would be ideal but is not currently part of the decision-making process, as required under the legislation. Shaded areas represent components that did not take place in this particular process. Six parts have been identified: (I) formal production of development applications, (II) consultation with council and independent scientists, (III) the use of accreditation peer review systems, (IV) role ofNPWS, (V) development conditions, and (VI) the Land and Environment Court. Chapter 3 Case studies 57

The issues highlighted by the Court involved breaches of these planning provisions (DCP 19 and SEPP 19), the Environmental Planning and Assessment Act 1979 (NSW) and the Endangered Fauna (Interim Protection) Act 1991 (NSW). The late lodgement of a Fauna Impact Statement (FIS)2 (this predated the TSCA) relating to the occurrence of the red-crowned toadlet was an important issue examined by the Court. This document should have been lodged with the original development application on 6th February 1995. The FIS was prepared by Ecotone Ecological Consultants and lodged later with Council. As the development may have a significant effect upon the toadlet population, the FIS needed to be lodged with the development application (EPAA s 77(3)(dl)). The red-crowned toadlet was listed under the National Parks and Wildlife Act 1974 (NSW), Schedule 12 as vulnerable and rare, with a distribution restricted to the Sydney Basin. The FIS stated that the development may result in the loss of the species to the site, but also concluded that, if some of the habitat were to be left intact, the species may migrate to that spot, with no basis for the prediction or monitoring to confirm this statement.

Other issues involved breaches of SEPP 19 (bushland and scenic quality), the provision of effective drainage, sedimentation and water control and lack of consideration of DCP 19. A report was constructed by Landscan Pty Ltd and submitted to Council in February 1995 covering these issues. This occurred during the court proceedings and the report contained mixed views.

The case concluded with Pearlman J. ruling in favour of the respondent (Warringah Council). The development would not proceed. This judgement was that ARDEL were attempting to maximise the site potential while neglecting the significant environmental constraints of the site. The presence of threatened species other than the red crowned toadlet at the site was not commented upon. It is to be noted that FIS only covered fauna not flora and therefore the issue of T. glandulosa was not raised.

2 A Faunal Impact Statement (FIS) is a mechanism triggered by listing of threatened species under the previous legislation, Endangered Fauna (Interim Protection) Act 1991 (NSW). See Chapter 1 for a history of threatened species legislation in N.S.W. Chapter 3 Case studies 58

Hassell Pty Ltd v Warringah Council (41NSWLEC1998)

The second case (in part a merit appeal and part judicial review) was the response to Council's refusal of a development consent application to use the Crown Road Reserve for access to the site. A new development application for the subdivision was submitted in July 1996 (reducing the number of houses), with a separate application for the use of the Crown Road Reserve. The development applications were considered concurrently by Council and refused on 22 July 1997. Since the previous court case, the Threatened Species Conservation Act 1995 (NSW) had come into force. A new suite of threatened species (including plant species) and communities believed to occur on the site had now to be considered in the development application, as Species Impact Statements (SIS) are not limited to fauna under the Threatened Species Conservation Act, whereas the previous Endangered Fauna (Interim Protection) Act 1991 (NSW) required an FIS to be prepared only in relation to listed fauna.

The issues did not differ significantly from the first court case and, from the Council's perspective, revolved around breaches of the Environmental Planning and Assessment Act 1979 (NSW), SEPP 19 and DCP 19. There were 21 issues concerning the subdivision application and a further 15 issues in relation to the Crown Road Reserve application. During the case, some of these issues emerged as being of considerable importance.

The first issue revolved around whether Species Impact Statements were required (EPAA s. SA) in relation to the listed threatened plants T. glandulosa, Eucalyptus capitellata, Gonocarpus salsoloides Reichb. ex Sprengel, Angophora crassifolia (G. Leach) L. Johnson & K. Hill, Pimelea curviflora R.Br., Prostanthera howelliae Blakely and the threatened vertebrate Pseudophryne australis (red crowned toadlet). Many of these species were identified in the original survey but not considered endangered until their listing and the enforcement of the TSCA. It was argued by the Council that the potential impact of the development upon threatened species Chapter 3 Case studies 59

present at the site had not been addressed (assuming the species occurred at the site). Council believed that there was going to be a significant effect on the threatened species and communities at the site (EPAA s. 90(1)(c3)). The Council also believed that the provisions of SEPP 19 and DCP 19 had not been addressed adequately.

Sheahan J. held that although the subdivision might have posed a threat (as assessed by the Courts) to the populations of T. glandulosa found on the site, the development had been re-designed to allow protection of between 80-90% of the largest population. He found that this was "quite acceptable when taken in the overall context of the incidence of the species elsewhere".

Sheahan J ruled that the development proposal for the subdivision could proceed (with a reduced number of lots and the protection of T. glandulosa), because the applicant had addressed all issues highlighted by the Council.

Species Impact Statements were made available to the Court. A survey team from CSIRO engaged by ARDEL Limited had completed SISs covering both flora and fauna species.

During the assessment of fauna at the site, there were two concerns over the presence of species (red crowned toadlet and broad headed snake). The site of the development was deemed to be relatively small (3.644ha), and it was decided by CSIRO that the densities of any species present would be low, with episodic (not continual) use of the habitat present. CSIRO decided that any field observations would require a period of time that would be impractical for an SIS (time and funding limitations). The potential presence of any fauna species was therefore determined through the use of literature and databases. The CSIRO also made a point (in their report) that determining the presence or absence of any rare species depends on the methods employed by the surveyor.

An SIS was also prepared for T. glandulosa, the only threatened plant species believed to possibly exist on the site (Table 3.2). The SIS was completed on the Chapter 3 Case studies 60

basis that the existence of the species on the site was doubtful, but the fact that it was adequately represented in conservation reserves in the Sydney region meant that, should any population exist on the site, maintaining it would not be critical to the conservation of the species. The legislative provisions requiring the SIS to explore feasible alternatives and to develop mitigation measures were largely dismissed as inoperable by the consultants on the basis that there was no known population on the site. As appears from the discussion above, however, by the time Sheahan J. made his decision, it was accepted knowledge that T. glandulosa did indeed exist on the site.

Hassell Pty Ltd v Warringah Shire Council (49 NSWLEC 2000)

The third case was in the form of another merit appeal from council decision to refuse consent to further subdivide some of the lots created in the original subdivision. At this point, the Court accepted the evidence of the applicant's consultant that he had originally misidentified the species on these particular lots as T. glandulosa, and this had been confirmed by the National Herbarium of NSW after it had examined samples.

Consultation with Council and with the National Parks and Wildlife Service

Although the DG of the National Parks and Wildlife Service (NPWS) was not required to give concurrence, the Service did provide advice to council and also expert witnesses for the Land and Environment Court cases. This was therefore an informal process of consultation rather than a defined procedure. The developer did not approach the Council or NPWS for advice or guidance on amelioration. Chapter 3 Case studies 61

Table 3.2: Summary of the SIS and outcomes for the population of Tetratheca glandulosa at Aquatic Drive, Allambie Heights.

Species Impact Statement (TSCA s. 110) Outcome

(1) a full description of activity proposed in the The action is summarised in section 3.2.2 of this chapter development application, including nature, extent, (summarised also in HASSELL PLAN 9208-10-4.6.96 location, timing and Layout of the proposed development prepared for Ardel Ltd.). or activity

(2a) a general description of the threatened species The only threatened plant species reported from the land known or Likely to be present in the area that is the is T. glandulosa (reported by one of the surveyors in subject of the action and in any area that is Likely to be 1995). affected by the action

(b) an assessment of which threatened species known or As per 2(a) above. likely to be present in the area are Likely to be affected by the action

(c) for each species likely to be affected, detail of its Species is well represented in Dharug NP, Marramarra Local, regional and State-wide conservation status, the NP, Muogamarra NR, Brisbane Waters NP, Garriga) NP, key threatening processes generally affecting it, its Lane Cove River NP and Yengo NP. These conservation habitat requirements and any recovery plan or threat areas cover the complete range of the species. Herbarium abatement plan applying to it samples show a broader range.

(d) an estimate of the local and regional abundance of Population sizes are unknown but thought to be small. It those species is difficult to estimate as the species is cryptic when not flowering. Region not defined in 8 part test. (e) a general description of the threatened species known As per (a) above or likely to be present in the area that is the subject of the action and in any area that is likely to be affected by the action

(j) a description of type, Location, size and condition of T. glandulosa was not located on the site (by CSIRO) and the habitat and details of the distribution and condition its presence is doubtful. It is possible that the species was of similar habitats in the region mis-identified as T. ericifolia. This population was considered to be non-critical given that the species is conserved in 8 reserves.

(g) a full assessment of the likely effect of the action on There is not likely to be any impact as the possible those species, including, ifpossible, the quantitative effect existence of the species at the site is doubtful, therefore of local populations in the cumulative effect in the region the habitat was considered to be of no importance to the species survival. (h) a description of any feasible alternatives to the action Given the absence of a known population on the site, in the manner justifying the carrying out of the action in feasible alternatives to the action cannot be developed. the manner proposed, having regard to biophysical, economic, social and ESD principles

(i) a full description and justification of the measures Given the absence of a known population on the site, proposed to mitigate any adverse effect of the action on mitigation actions cannot be developed. the threatened species

(j) a list of any approvals that must be obtained under No other approvals are known to be necessary any other Act or law before the action may be lawfully carried out, including details of conditions of any existing approvals that are relevant to the species or populations Chapter 3 Case studies 62

Monitoring Programme

No monitoring programme has been implemented, as none was required by the conditions. However, in accordance with the outcomes of Hassell Pty Ltd v Warringah Council (41 NSWLEC 1998) a bushland management plan has been 3 completed . This will be implemented following completion of the development. The management plan aims to: • conserve the remaining bushland on the site • minimise the impact of the subdivision • optimise remaining bushland as a wildlife corridor • promote restoration and regeneration • manage public access and fire hazard reduction .

Specifically, the bushland management plan aims to ensure the conservation of rare and threatened species through management actions, with responsibility being allocated to Warringah Council. Actions include; (i) rare and threatened species to be used within the regeneration programme, (ii) monitoring the success of re­ introducing such species ( eg. plant survival and seed germination), (iii) community awareness of threatened species on site. It is important to note that the management plan does not allow for any provision for adaptive management or a mechanism to score the success of the present management plan ( eg. long term plant survival and continual seedling recruitment).

3.2.3 Analysis

The Ardel development is a complex case study containing many elements of scientific uncertainty. The site has an extensive development history revolving around the role of scientific knowledge and ecological studies. Initially, application of the legislation required

3 "The restoration of the bushland can be made subject to a Bushland Management Plan to be submitted to Council prior to the release of the subdivision plan. I would be prepared to impose an appropriate condition regarding a Bushland Management Plan, as basically agreed between the parties." (NSWLEC 10427of1997 p12). Chapter 3 Case studies 63

assessment of whether the proposed development would have a significant effect on a population of red-crowned toadlet, through a Faunal Impact Statement. It was determined that there would be a significant impact and that the development should not go ahead. This was prior to the implementation of the Threatened Species Conservation Act 1995 (NSW). The second court case focused upon the apparent existence of T. glandulosa on the site (when was discovered by council before they rejected the development application) and whether an SIS was needed. This demonstrates a complicated role for existing knowledge and scientific uncertainty. T. glandulosa does exist at the site. However, at the time the development application was submitted, ARDEL Ltd. believed that the species did not occur at the site. This know ledge was based upon a survey of T. glandulosa by CSIRO. However, T. glandulosa is cryptic (see section 4.2.1.) and therefore is difficult to detect unless in flower (survey work needs to occur during the flowering season). An SIS was therefore submitted for T. glandulosa, despite the developer's belief that the species did not occur at the site, but the SIS was completed as if the population did not exist. The outcome was inevitably one of no significant effect, because of the phantom population. There are three possible decision-making pathways: (1) the use of the precautionary principle to reject the development application, as the existence of the population of T. glandulosa is uncertain and there is considerable scientific uncertainty; (2) the development could proceed on the basis that all decision-making is completed as if T. glandulosa did exist at the site; and, (3) the development approval could be delayed until further and more in depth surveying could take place (Stein J requested from the developer to complete further research following the court case Leatch v National Parks and Wildlife Service (1993)). However, none of these pathways were taken during the decision-making process, resulting in conflicting assumptions about scientific knowledge and an expensive pathway for both the developer and Local Council for the obtainment of development consent. A number of ideal steps (Figure 3.4) may have removed much of the confusion and inadequate science, including council consultation and independent science (Figure 3.4, section II), a peer review of the SIS (Figure 3.4, section III) and the use of up-to-date science to determine appropriate development conditions (Figure 3.4, section V). Chapter 3 Case studies 64

3.4.4 Outcomes

The population of T. glandulosa had been partially cleared in January 2000 as allowed under the terms of the consent. Ten individuals of the remaining plants were smothered when debris was placed on the population (Plate 3.3 and 3.4). The undergrowth of the remnant bushland in which the population of T. glandulosa occurs is thick and moist, with dominant species including Hakea sericea, H. terretifolia and Bauera sp. With the increase of undergrowth shrubs, the remaining T. glandulosa may be out-competed for light, nutrients and space. There is no formal protection for the remaining plants, as the implications of isolation and other processes associated with development (including an increase in fire) are unknown. It is also thought that T. glandulosa may be a clonal species. This would impact upon the population dynamics and the perceived numbers of individuals present. If the species is clonal, wrong information could have led to inappropriate decision­ making and management outcomes.

3.3 Green Road, Glenhaven (33°42'S 150°157'E)

This site4 is currently (1999-2000) under construction for a housing subdivision (Plate 3.4). It occurs in the local government area of Baulkham Hills Shire Council. The site is zoned Residential 2(b), Residential 2(d) and Open Space 6(a) under the Local Environmental Plan for the Shire (Figure 3.5). The site is located to the south of Green Road and Cattai Creek, which winds down the eastern side. To the north of the site is Glenhaven Road, a major road for local traffic, while the southern border is formed by Poole Road.

3 .3 .1 Vegetation attributes of the site

Mixes of skeletal and lateritic soils are found at the site, givmg rise to Eucalyptus haemostoma/Corymbia gummifera and E. globoidea Blakely woodland. This community occurs at a number of sites. Other common species including Acacia terminalis,

4 In future chapters this site is referred to as Glenhaven. Chapter 3 Case studies 65

Plate 3.3: The site at Aquatic Drive, Allambie Heights was cleared for development. Population of Tetratheca glandulosa occurs in vegetation on the left of the photograph.

Plate 3.4: During clearing of the site at Aquatic drive, Allambie Heights, between 10-20% of the population of Tetratheca glandulosa was destroyed. A cage used during an exclusion experiment can be seen in the fore ground of the photograph. Chapter 3 Case Studies 66

Glenhaven Road

Open Space

North Residential

Corridor r SCALE 1:4000

Residential Darwinia biffora

Figure 3.5: Approximate location of Darwinia bif/,ora on the site at Green Road, Kellyville (adapted from the Draft Local Environment Plan 1991). Chapter 3 Case studies 67

Banksia ericifolia, B. oblongfolia, B. serrata, Dillwynia retorta, Epacris microphylla, Grevillea buxifolia, G. speciosa and Xanthorrhoea media. A population of approximately 80 individuals of D. biflora occurs in the western section of the site.

3.3.2 The issues

The development application was approved not long after the introduction of the Threatened Species Conservation Act 1995 (NSW). There are three central issues to the granting of development consent at Glenhaven: (1) the eight part test; (2) the developer seeking amelioration with local council before the writing of the eight part test; and, (3) the use of bushland corridors.

Eight Part Test and Species Impact Statement

As a result of discussions with Council, amelioration measures were proposed in the development application, which resulted in an eight part test (Table 3.3) indicating that there was unlikely to be a significant effect. It was therefore concluded that an SIS was not needed with the development application and NPWS concurrence was not required. Through discussion with the Council, development controls were agreed upon despite the fact that the development had the potential to have a significant impact upon D. biflora.

National Parks and Wildlife Service concurrence and council consultation

The populations of D. biflora were identified during the flora and fauna studies undertaken in support of the development application (i.e. before the completion of the eight part test). Therefore, an eight part test was required. The eight part test concluded that there was not likely to be a significant effect on the population of D. biflora. Therefore no SIS was required and concurrence from the National Parks and Wildlife Service (NPWS) was not sought. Consultation with local council occurred before the writing and lodgement of the development application and eight part test. The amelioration process (a negotiation process where impacts from a development Table 3.3: Summary of the Eight Part Test and outcomes for the population of Darwinia biffora at Green Road, Glenhaven. 9 -§ Eight part test Outcome Development controls & (..J.) (a) in the case of a threatened species, whether the life A viable population can be maintained by placing appropriate 1. drainage construction cycle of the species is likely to be disrupted such that a conditions on the development. 2. revegetation with D. bi/fora ~ "1 viable local population of the species is likely to be placed 3. fencing around population ~ at risk of extinction 4. fire management ~ ~ (b) in the case of an endangered population, whether the n/a n/a ~· life cycle of the species that constitutes the endangered population is likely to be disrupted such that the viability of the population is likely to be significantly compromised

( c) in relation to the regional distribution of the habitat of Approximately 1% of D. bi/fora habitat would be lost (regional none a threatened species, population or ecological community, habitat). Suitable habitat occurs sporadically in the local government whether a significant area of known habitat is to be area, mainly on private land (eg. 5000 individuals growing on 2 ha, modified or removed lkm S.E. of the site). Through interpretation of vegetation and soil landscape maps, the site in relation to the regional distribution of T. glandulosa, the area is not significant.

(d) whether an area of known habitat is likely to become No, a bushland corridor was designed allow access for insects for cross Corridor of habitat retained isolated from currently interconnecting or proximate areas pollination. of habitat for a threatened species, population or ecological community

( e) whether critical habitat will be affected n/a n/a

(:t) whether a threatened species, population or ecological Assumed to be adequately conserved in reserve systems. This none community, or their habitats, are adequately represented assumption is based upon the species' ROTAP codes and the size of in conservation reserves )or other similar protected areas) populations existing in reserves. in the region

(g) whether the action proposed is of a class of action Subdivision results in habitat removal which is generally a threatening none recognised as a threatening process process but, it was not listed in schedule 3 of the TSCA.

(h) whether any threatened species or ecological S.W. limit of distribution in and around site. none community at the limit of its known distribution 0\ 00 Chapter 3 Case studies 69

are minimised) mainly concerned the boundaries of the open space that was set aside as part of the development proposal. Council negotiated the boundary to include a "trade-off' of land. This is a narrow strip of land, lOm along the edge of the open space in exchange for the section of land that includes most of the D. biflora populations.

Development Controls

Four conditions were attached to the development consent. These conditions included: (i) increase in the area of open space allowed for a conservation zone to protect the population of D. biflora; (ii) suitable fencing around the population of D. biflora (Plate 3.5); (iii) during development, a diversion for drainage constructed around the conservation zone to allow for a reduction in the potential change to the microenvironment of the area such as an increase of water to the area; (iv) importantly, Council insisted upon the "leaving of' a bushland corridor between the population of D. biflora and surrounding bushland (Plate 3.6).

Monitoring Programme

At present no monitoring plan has been implemented. There are plans by the developer (not required as a condition of development) to write and implement a management plan when the housing subdivision has been completed.

3.3.3 Analysis

Extensive consultation with local council avoided many potential issues involving threatened species conservation in regards to the population of D. biflora at the site (Figure 3.6). Even though there was extensive consultation development controls (e.g. a bushland corridor) without the backing of ecological knowledge were implemented (see 1.1.3 for complete discussion involving the use of corridors for conservation). Further consultation with NPWS may have seen different development control measures evaluated and a robust monitoring programme implemented. Chapter 3 Case studies 70

Plate 3.5 (above): A fence was placed around an area where a population of Darwinia biflora occurs at Green Road, Glenhaven

Plate 3.6 (left): A bushland corridor was included in the development controls for Green Road, Glenhaven. Chapter 3 case Studies 71

Planning instruments such as; Zoning (EPAA Strategic Planning -----1 SEPPs (EPAA s.(37)-(39)) Part 3) (EPAA s.5(a)(i)) REPs (EPAA s.(40)-(52)) LEPs (EPAA s.(53)-(72)) ------Development Pral;m;nary raseamh r·------.. ------standards for development .______.___ Developer seeks council (EPAA s.4(1)) consultation proposal

Developer seeks independen Flora and fauna II ----~---< scientific input from survey* researchers in the field*

8 part test completed by Local council to decide if an SIS developer or consultant -____._;..---< is required (EPAA s.77(3))* (EPAA s.5A; TSCA s.94)*

Species Impact Statement Peer reviewed for scientific 111 (TSCA s.110) completed ----+--1 credibility by independent by an accredited person scientists* (TSCA s.113(1))* .. ______'------'J ______

Development application submitted by developer _____._---< Consultation with NPWS (informal process)* (EPAA s. 78A) IV Concurrence of NPWS if Evaluation of development _ ___,___., application (EPAA s.79C) SIS shows a s~lficant effeot {EPM s.798(3)-{7))*

Application v No approved by council

Development begins

Figure 3.6: Flow chart showing the decision-making process for a subdivision at Green Road, Glenhaven. Overall Chart represents an idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part) presented in Figure 3 .1. An italicised box refers to a step, which would be ideal but is not cl.UTently part of the decision-making process, as required under the legislation. Shaded areas represent components that did not take place in this particular application process. Six parts have been identified: (I) formal production of development applications, (II) consultation with council and independent scientists, (III) the use of accreditation peer review systems, (IV) role of NPWS, (V) development conditions, and (VI) the Land and Environment Court. Chapter 3 Case studies 72

An amelioration measure of a bushland 'habitat corridor' is mentioned in the eight part test conducted in this case in regard to the population of D. biflora, as a means of addressing part (d) of the test (Table 3.2) which states; (d) whether an area of known habitat is likely to become isolated from currently interconnecting or proximate areas of habitat from a threatened species, population or ecological community. It was assumed that, even though the habitat in the corridor is unsuitable for D. biflora itself, insects associated with the species would use the corridor to move between the adjacent bushland and the D. biflora population to be isolated. It was assumed that insects are needed to aid in cross-pollination of D. biflora, and that D. biflora needs insect pollinators to produce viable seeds. It has been noted (see draft Recovery Plan) that insects rarely visit D. biflora flowers (see Table 5.3). However, there was no mention of how this amelioration measure was devised or what body of research was used to justify it. The use of a bushland corridor for conservation stresses the importance of independent scientific input and review especially when the debate has divided the scientific community (Figure 3.6, sections II and III) as well as the crucial need for up-to-date scientific knowledge (Figure 3.6, section V).

3.4 Camarvon Drive, Frenches Forest

The development of a playing field at Carnarvon Drive was proposed by Warringah Council (i.e. a development by the Local Council itself) for an area zoned open space, because of power lines running over the site (Plate 3. 7).

The proposed site for the development fronts on to Carnarvon Drive at Frenches Forest, with Wakehurst Parkway (a major freeway) running along the back of the site (Figure 3.7). Privately owned housing surrounds the site. The site contains three populations of T. glandulosa, one population of 40 individuals and two populations each with a single individual. These populations are believed to be the southern limit of the species. However the surrounding habitat is considered by Council to be ideal for this species. Q ~ ~ Carnarvon Drive ""! v..i

~ ~

~ ~

~·

Playing Field

? Re•tor•tion Area'---. 0L---

Tetratheca glandulosa Existing Vegetation

-.....} Figure 3.7: Approximate location of Tetratheca glandulosa at Camarvon Drive Frenches Forest (Adapted from development plan). (.;.) Chapter 3 Case studies 74

3.4.1 Vegetation attributes of the site

The native vegetation community occurring at this site is characterised by Broad-leaved Scribbly Gum Eucalyptus haemostoma - Red Bloodwood Corymbia gummifera - Narrow­ leaved stringbark E. oblonga s.1. Low Open Woodland (Smith and Smith 1998) (Plate 3.8). The site is relatively weed free, with a number of dominant species. The canopy is dominated by a shrub layer consisting of Banksia ericifolia, Kunzea ambigua Smith (Druce), Leptospermum squarrosum, L. trinervium (Smith) J. Thompson and Dillwynia retorta (Smith and Smith 1998). The ground cover consists largely of Lepyrodia scariosa R.Br., Dampiera stricta (Smith) R.Br., Micranthium ericoides Desf., Cyathochaeta diandra (R. Br.) Ness, Epacris microphylla and Hemigenia purpurea R.Br. (Smith and Smith 1998). This vegetation community is widespread in Warringah Shire and is conserved within Ku­ Ring-Gai NP and Garigal NP.

3.4.2 The development

The development was a proposed neighbourhood playing field (90m in length and 70m wide) to accommodate junior rugby league, rugby union and soccer, and senior touch football, with an associated carpark ( 42m in length and 17.Sm wide). The carpark and road access would be located on areas that are currently largely cleared. Areas of proposed landscape restoration with indigenous species form part of the development proposal, as well as extensive retention of the existing vegetation to provide screening for the proposed field from Wakehurst Parkway.

3.4.3 The Issues

The adequate completion of an eight part test and the role and attitude of the NPWS are the two main issues in this case study. The lack of scientific certainty plays a role in the decision-making process of this case study (Figure 3.8). Chapter 3 Case studies 75

PlateJ.7: The site at Carnarvon Drive, Frenches Forest was zoned open space due to the powerlines that crossed the site.

Plate 3.8: Camarvon Drive, Frenches Forest Chapter 3 Case Studies 76

Planning instruments such as; Zoning (EPAA Strategic Planning ______, SEPPs (EPAA s.(37)-(39)) Part 3) (EPAA s.5(a}(i)) REPs (EPAA s.(40)-(52)) LEPs (EPAA s.(53)-(72))

Development Preliminary research standards Developer seeks council for development (EPAA s.4(1)) 1-----t--- consultation proposal

Developer seeks independent Flora and fauna II soientific input from survey* researohers in the field*

8 part test completed by Local council to decide if an SIS developer or consultant ______, is required (EPM s.77(3))* (EPAA s.5A; TSCA s.94)* f------·------Species Impact Statement i Peer reviewed for soientific ie----+i---1 111 (TSCA s.110) completed d?·/ll"'1 b l (Mrt111ndent by an accredited person · ore ~ 'T IV n - !. .solentiltl"' (TSCAs.113(1))* 1 ~------) Development application Consultation with NPWS submitted by developer l (informal process)* (EPAA s.7BA) IV

Concurrence of NPWS if Evaluation of development ie----+---i SIS shows a significant application (EPAA s.79C) effect (EPAA s.798(3)-(7))*

Development Application v application No approved by r----Yes---r. withdrawn council

VI No

------' •

Figure 3.8: Flow chart showing the decision-making process for a playing field at Carnarvon Drive, Frenches Forest (Warringah Council). Overall chart represents an idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part) presented in Figure 3.1. An italicised box refers to a step, which would be ideal but is not currently part of the decision-making process, as required under the legislation. Shaded areas represent components that did not take place in this particular application process. Six parts have been identified: (I) formal production of development applications, (II) consultation with council and independent scientists, (III) the use of accreditation peer review systems, (IV) role of NPWS, (V) development conditions, and (VI) the Land and Environment Court. Chapter 3 Case studies 77

Species Impact Statement

In conjunction with a botanical survey, an environmental consultancy company 5 completed an SIS . The SIS concluded that the proposed development was likely to have a significant impact upon the large population of T. glandulosa present on the site. It was suggested that, to minimise this impact (by retaining the complete population), the development be moved lOm to the east. On this advice, Council modified their development plans. Consequently, the consultant concluded that there was not likely to be a significant impact on T. glandulosa by the modified proposal.

National Parks and Wildlife Service concurrence and Council consultation

The development was a Council development and National Parks and Wildlife Service (NPWS) concurrence was sought following the completion of an eight part test that showed a significant effect. The development application and all relevant information were lodged with NPWS for assessment. NPWS recommended an experimental bum with monitoring up to 2 years afterwards. NPWS felt that, without the experimental bum, council would not have completed the eight part test (s 94(2)(b) TSCA), as they would have neglected the potential soil seed bank. Local council could have argued that there was no available information or extrapolated information from other sites (such as the case study at Grosvenor Street, North Wahroonga discussed in section 3.5). It is important that the local council wanted to operate within sound ecological practices and knowledge and follow guidance from the NPWS. NPWS assessed whether the development was going to disrupt the life cycle and viability of the populations. It was determined that the populations would not be significantly compromised. Regionally the population may be important, due to the potential soil seed bank. NPWS believed that the population would not be isolated, as a connecting strip of native vegetation to the north and east of the largest population would be left intact.

5 P & J Smith Ecological Consultancy Chapter 3 Case studies 78

Development Controls/Conditions of Consent

Development controls were not suggested in the early planning for the development. However, the inherent problems of playing fields such as trampling, weed invasion, litter and nutrient run off were noted for further discussion.

Monitoring Programme

As the development application was withdrawn, no monitoring programme was established. Council are still (2000) considering an experimental burn and monitoring of seedling emergence, to aid in the overall management of threatened species in the local government area.

3.4.3 Analysis

Ecological knowledge played a dominant role in the decision-making process for this case study. The concurrence role of the NPWS provides an interesting scenario in relation to consideration of all aspects of the life cycle in regards to the eight part test. NPWS would only be involved informally at this stage. T. glandulosa produces a soil seed bank, with germination believed to be triggered by fire events (Benson 1985). The existing knowledge does not help determine where the seed bank is stored, but an experimental burn may answer the question. The maintenance of seed banks is important for new recruitment (Auld 1995; Enright et al. 1997) and is accumulated between fire events (Meney et al. 1994). Certain fire regimes can affect the seed supply at a site. The soil-stored seed bank is insulated by the soil and hard seed coat, often withstanding high temperatures (Campbell et al. 1994) and allowing for a flush of germination afterwards (O'Connell et al. 1979; Bradstock and Bedward 1992; Bradstock and Auld 1995). Seed banks are an important part of the life cycle of a plant and allow for the next generation of recruitments following a disturbance (Auld 1986; Morgan 1995). With this existing knowledge it was assumed by NPWS that a seed bank could be present and a playing field at the site would disrupt potential recruitment. The early incorporation of existing scientific knowledge in the decision making process allowed for appropriate decisions. This argument could also have Chapter 3 Case studies 79

been used within the ARDEL case study. Like this site there was no clear evidence where and if a seed bank existed. This is a situation where adaptive management and strong experimental design could have played a role. Through the use of NPWS, a concurrence role with NPWS and available scientific knowledge, Council was able to make a well­ informed decision not to allow the development to proceed. The reality was that NPWS did not give concurrence; therefore the decision did not lie in Council's hands in any case.

3.5 Grosvenor Street, North Wahroonga (33°42'S, 151°07'30"E)

The site of the proposed development is owned by Chanrich Properties Pty Ltd and Hi­ Return Investments Pty Ltd, and occurs in the local government area of Ku-Ring-Gai Municipal Council. The development application was withdrawn after an unsuccessful appeal to the Land and Environment Court. The site (occurring on the Wahroonga Plateau) covers the block numbers 153 to 165 in Grosvenor Street, North Wahroonga (Plates 3.9 & 3.10). The site terminates approximately 200m from Ku-Ring-Gai Chase National Park.

The site has had a varied history, involving rezoning and change of ownership. Previous decisions made in regard to development on the site were taken into consideration when evaluating new development applications. In 1883, a trig station (Cook's) was established on vacant Crown land and in 1894 a reserve was established for the trig station. A reserve for 'public purposes' was created over the surrounding Crown land in 1958. It was during the 1960s that the Crown extended Grosvenor Street to its present position and subdivided the site for residential purposes and a future location for a school. Over the following decade, urban development moved closer to the site and various tracks and roads in and around the reserve became evident. The site was officially zoned 'special use - school', except for the actual trig station, in April 1974. By 1975, housing had been erected on Holt Avenue and Barton Crescent (the surrounding streets). Investigations into the possible future use of the site were conducted in 1982 and the 'school site' was deemed surplus to requirements of the Department of Education. In 1985, Landcom surveyed the site and approached council to consider rezoning the area to medium density housing. Rezoning was Chapter 3 Case studies 80

Plate 3.9: Grosvenor Street, North Wahroonga.

Plate 3.10: Grosvenor Street, North Wahroonga. Chapter 3 Case studies 81

approved to 2(h)6 in the local environmental plan. In 1988 the site was placed on sale by tender by Landcom, at which time Council approved a development application for a 7 subdivision of 13 dual occupancy dwellings .

The Department of School Education sold the site to Chanrich Properties Pty Ltd in 8 October 1995 (for an amount in excess of $2 million ). Chanrich Properties Pty Ltd was told by Council later in the year that they were opposed 'in principle' to any commercial or residential development of the site. Council prepared a Housing Strategy for the area, which further compounded this objection. The Housing Strategy did not allow for development of the site. A development application was submitted and rejected by council in 1996, the same year the Development Control Plan 29 (DCP) and Local Environment Plan 164 (IBP) 9 were drafted . The Council changed its direction for residential development of the site after rezoning the land with focus on medium density residential development as opposed to high-density residential development.

3 .5 .1 Vegetation attributes of the site

The site has a mixture of Hawkesbury Sandstone and Wianamatta shale giving rise to low woodland. Species dominant in the canopy include Eucalyptus leuhmanniana F. Muell., Angophora bakeri C. Hall and Corymbia gummifera. The understorey possesses considerable diversity, largely from the families Proteaceae, Fabaceae and Rutaceae.

Large populations of D. biffora are located beside the track below Barton Crescent and near the end of Grosvenor Street. Small populations of T. glandulosa were found throughout the site by ecologists working for the council.

6 The zone 2(h) allowed for housing development to occur with consent.

7 Restrictions on subdivisions in the local government area had been in place since 1972.

8 Amount disclosed during the hearing of Peter William Lean v. Ku-Ring-Gai Council No. 10457of1996 and 40265 of 1996. 9 Information disclosed during the hearing of Peter William Lean v. Ku-Ring-Gai Council No. 10457of1996 and 40265of1996. DCP 29 was later replaced by DCP 48- Medium density residential development. Chapter 3 Case studies 82

3.5.2 The Development

The proposed development for the site consisted of a 32 lot housing subdivision and access road. Chanrich Properties Pty Limited later revised the development application to 19 lots, access road and buffer zone.

3 .5 .3 The Issues

Three issues were identified in connection with this case study: (1) changes in the zoning of the site; and (2) the use of scientific knowledge to determine the existence and distribution of potential seed banks (Peter William Lean v. Ku-Ring-Gai Council No. 10457 of 1996 and 40265 of 1996). These issues have played an important role in the decision-making process in relation to the site (Figure 3.9). An SIS was prepared in this case but reference was made to it in the Courts decision, 10 discussed below •

Zoning

Landcom (after surveying site in 1985) approached the local council to rezone the site from "special use (a) school" to "residential (high density, 2c)". The local council in 1986 rezoned the area instead as "medium density residential 2(h)".

Peter William Lean v. Ku-Ring-Gai Council No. 10457 of 1996 and 40265 of 1996

An appeal was lodged with the Land and Environment Court after Ku-Ring-Gai Municipal Council did not grant development consent for the housing subdivision, 11 despite rezoning the land to allow such a development • Consent from local council was not granted, as the development was in direct conflict with the aims of the

10 The existence of the SIS was referred to in the later decision in Lean v Ku Ring Gai Council (1997) NSWLEC152

11 The attitude and politics of the local council had changed during the period of re-zoning to when the development application was lodged. 83 Planning instruments such as; Zoning (EPAA Strategic Planning ,..._ _ _ _ _, SEPPs (EPAA s.(37)-(39)) Part 3) (EPAA s.5(a)(i)) REPs (EPAA s.(40)-(52)) LEPs (EPAA s.(53)-(72)) T------1 Development i standards Preliminary research I for development r----+--- Developer seeks council (EPAA s.4(1)) consultation proposal I I Developer seeks independen Flora and fauna I II scientific input from survey* researchers in the field*

8 part test completed by Local council to decide if an SIS developer or consultant _ __._---1 is required (EPAA s. 77(3))* (EPAA s.5A; TSCA s.94)*

5&~~~ ~~~~~!~~~:.;~• r ·:;:=:= ------111 ______by an accredited person ---+---1 credibility by independent scientists* (TSCA s.113(1))* .. ______

Development application submitted by developer ______, Consultation with NPWS (informal process)* (EPAA s.78A) IV O()nourrence otiNPWS if Evaluation of development_--+_ $ ~ • ai;nificant application (EPAA s.79C) effttll~M£.V.SaCJ)-(7))'

Application v No approved by council

VI No

Appeal to the Land Appeal and Environment dismissed Court (EPAA s.97(1))*

Figure 3.9: Flow chart showing the decision-making process for a housing sub-division at Grosvenor St, North Wahroonga (Ku-Ring-Gai Municipal Council). Overall chart represents an idealised decision-making process based upon existing NSW law from a scientific point of view (*denotes when scientific knowledge or seeking scientific information plays and important part) presented in Figure 3 .1. An italicised box refers to a step, which would be ideal but is not currently part of the decision-making process, as required under the legislation. Shaded areas represent components that did not take place in this particular decision-making process. Six parts have been identified: (I) formal production of development applications, (II) consultation with council and independent scientists, (III) the use of accreditation peer review systems, (IV) role of NPWS, (V) development conditions, and (VI) the Land and Environment Court. Chapter 3 Case studies 84

Development Control Plan 29 and the Local Environmental Plan 164, which were in 12 a draft form for the site • The case discussed a number of merit issues, which were addressed as possible development constraints. A number of conditions were agreed upon but a number of conditions remained in dispute. The potential impacts on threatened species and visual impacts of the development were still in dispute.

Both D. biffora and T. glandulosa were observed (by experts for Council and the Developer) in large populations, with D. bi/fora occurring towards the edge of the site and T. glandulosa throughout the site. It is believed by many ecologists that T. glandulosa is well conserved within the Sydney area (see other case studies). The size of the potential soil stored seed banks and their location caused some debate during the course of the case. The local Council and the developer both agreed on the following points; (a) germination of the seed banks is linked with intensity and frequency of fire, (b) frequent fire regimes have been demonstrated to destroy the communities in which these species occur, and, (c) a combination of urbanisation (basic earthworks) and reduction of fire at the site will also potentially destroy the existing communities in which the two threatened plant species occur.

The Council believed that the seed banks could be distributed through the western and northern parts of the site, but not the south side, because a fire during 1994 that burnt the southern side did not produce seedlings. The fire histories of the site and the surrounding areas of Ku-Ring-Gai Chase National Park (KCNP) are critical in assessing the potential importance of seed banks and their relative location. The NPWS believed that the site's seed bank was likely to be substantially larger than potential seed banks in KCNP. Therefore these populations should be conserved as a precaution against any catastrophic losses at other known sites nearby, even though these are in a National Park.

12 Council, with full knowledge that the site was in the process of being sold to a developer and a preliminary development application had been approved, drafted both DCP 29 and LEP 164. LEPs, REPs (Regional Environmental Plans) and SEPPs (State Environmental Planning Policies) are all Environmental Planning Instruments (EPis). LEPs are developed by Council but are made by the Minister and are legally binding (even in a draft stage). The Council also develops DCPs. . They are not binding but act in an advisory manner, and provide the detail for an LEP. Chapter 3 Case studies 85

Sheahan J. did not regard the presence of D. biflora and T. glandulosa as sufficient grounds for rejecting the subdivision proposal. His decision was based upon the following (a) there was too much uncertainty regarding whether the seed bank would germinate following fire, (b) the two species were well represented and conserved in the Sydney area, therefore the loss of individuals from this site and the seed banks would not impact upon the species as a whole, and, ( c) both species could be successfully transplanted from nurseries into the finished landscape.

The visual impact of the development was causing the largest amount of concern. The development would have obstructed the view of the national park of already existing dwellings. It was this aspect of the case on which Sheahan J. based his a refusal of the development application.

3.5.3. Analysis

In this case, there was a contrast between existing knowledge and the uncertainty of ecological knowledge. Expert witnesses on soil-stored seed banks were used to discuss the ecology during the course of the hearing. A number of pieces of evidence (such as the location of the individual plants) were agreed upon by both parties while the existence of the soil seed bank was the central issue of scientific uncertainty. All interested parties do not deny the importance of the potential soil seed bank. Accumulation of seed banks allows for species survival following disturbance, either biotic or abiotic (Meney et al. 1994; Morgan 1995). However, it has not been established whether a seed bank is present for T. glandulosa and D. biflora at the site. Auld argued, in his evidence, that a seed bank was present in the northern and western sections of the site, rather than the eastern/southern sections, which was burnt in 1994 (no germination was detected). It is not always possible to determine the location of soil seed banks, as they do not always mirror the location of existing plants (Wang 1997).

To form a complete picture of the seed bank, a rigorous sampling regime would be required (Enright et al. 1997; Wang 1997; Tozar 1998), which is not always possible. Often it is not Chapter 3 Case studies 86

always possible to extrapolate from one part of a site to another in numbers and position of the seed bank. Chapter 4 Species biology and site descriptions 87

C h a p t e r 4 S pe c ie s B io l o g y a n d S it e D escriptions

4.1 Introduction

By August 2000 (the start of this part of my study), over 766 species, 17 populations and 28 ecological communities had been listed on the Schedules 1 and 2 of the Threatened Species Conservation Act 1995 (NSW). With the continued development of urban sprawl in the outer suburbs of Sydney, some of these listed species have now been considered in development applications. As part of the process of considering a listed species in development applications, eight part tests and Species Impact Statements (Environment Planning and Assessment Act 1979 (NSW)) have been completed.

Decisions about whether consent should be granted for a development have been plagued by controversy over the information contained in the eight part tests and SISs. In many circumstances, the extent of scientific uncertainty has been ignored, as explored in Chapter 1. Information on a number of listed threatened species was collated at the beginning of this study to identify species with appropriate characteristics for detailed study (see Table 4.1). The current scientific knowledge, status of recovery planning and development applications for eighteen species were assessed.

Several criteria were used to select species suitable for use as case studies, based on discussions with councils, NPWS and surveying the literature and available draft and completed recovery plans. These selection criteria were; (1) species occurs in the Sydney basin, (2) species is listed in the schedules of the TSCA, (3) little scientific knowledge about the species, (4) species is an issue in a development application, (5) species occurs in a number of local government areas with different opinions on development, (6) potential for development and conservation outcomes to vary, (7) data were sufficently scarce that the precautionary principle should be implemented during the development assessment process, and (8) the species occurs both within and outside conservation areas (such as National Parks). wkflpter 4 Species biology and site descriptions 88 ^ l e 4 1 : This is a summary of characteristics of some threatened plant species in the Sydney and South Coast regions. A number of parameters were considered when choosing study species (* also protected by Commonwealth legislation). Data correct as of August

> « ■

Distribution Listing on Recovery Scientific knowledge Development examples LGA Outcome of development Precautionary principle used Outcome relevant to Occurrence in reserves Is the species adequately conserved? the TSC plan involved example conservation Act Mountain Schedule 2, No Basic demography and Industrial warehouse at 42a Liverpool 8 part test and SIS completed Council asserts that a Requirement to consult Occurs at 124 sites in 15 LGA but No. Especially in main range and is continually Lagoon to Vulnerable* pollinators known. Jedda Road, Prestons with consultation with council precautionary approach was NPWS. only 4 sites within Scheyville N.P. threatened by land clearing for urban R Br. Menai, Bradwell & NPWS used. development and farming activities. to Woodford Waterfall to Mt. Schedule 2, No Basic fire ecology known. Unknown Unknown Unknown Unknown Unknown Water Catchment Areas Adequately conserved in Illawarra in > DC Keira Vulnerable Pollinators and seed dispersal Catchments. Not known outside this area. not known. -p'afcnaea Western part of Schedule 1, No Pollination and fire regimes are housing subdivision St Marys Unknown Unknown Unknown Unknown No. The swamp habitat in which this species is p fV iflo ra DC. the Cumberland part 1, not known. found has been affected by farming activities Plain Endangered - 0 v y n i a Cumberland Schedule 2, No Basic fire regime known. Unknown Unknown Unknown Unknown N/A Blue Mountains NP, Windsor No. Very localised species making it Downs NR, Yengo NP vulnerable to urban developments. tetiufolia D C. Plain Vulnerable Pollinators and seed dispersal not known. ~p^Slanlhera Somersby Schedule 1, Yes Assumption that it is a resprout, Somersby Industrial Estate Gosford SIS was requested and has not Council believes that a Unknown 2 populations (10 and 110 Not believed to be. Extensive surveying east junoms B.J. part 1, pollinator unknown, seeds into 17 lot subdivision with been received yet. NPWS precautionary approach was individuals) in Brisbane Waters side of Somersby Plateau and Mangrove concurrence under the EPAA if used. NP. Nine other populations have Mountain has occurred. Con"- Endangered difficult to germinate - little some clearing of the work done on seed bank approx. 70 individual developed. Negotiations with been found on the Somersby population. council since 1993. plateau on land-zoned industrial. 'lie fiu Bomaderry Schedule 1, Yes 1 population only discovered in Proposed road and bridge Shoalhaven Nomination for critical habitat No; development is held up due If development goes ahead, None No fjgeuerlenii Creek part 1, 1984 (6 geographically discrete development. being assessed. to local community efforts to the population will be j A. Armstrong Endangered groups). Basic ecology and retain the piece of bushland as a destroyed. * genetics known. whole habitat. N/A Population fenced to reduce N/A Zierw Pambula area Schedule 1, No 1 population west of Pambula None Pambula N/A N/A buxijugum J.D. part 1, (120 plants). Basic ecology not grazing from wallabies and feral Briggs & J.A. Endangered known goats. No sites in reserves. * Armstrong Ziena Kiama Schedule 1, No Basic ecology known (work by Extension of the extraction Kiama Rail Services Australia required NPWS rejected rezoning None predicted. Budderoo NP, Killalea SRA Not known. It is generally thought that the granulata (F. part 1, Kevin Mills & Associates). area for Bombo quarry. to conduct further investigations application on the basis that the species is not threatened by urban Muell.) Bcnth. Endangered (incl. genetics). population of 20 would be lost. development. Shows the use of a precautionary approach at the planning level. Allocasuarina Sydney Harbour Schedule 1, Yes Small population of 2 females. None Sydney area Recovery of single population a N/A N/A Sydney Harbour NP No. Concerns due to population size 2 females portuensis NP part 1, Basic demography known, main priority with a further 59 planted (survived) in 1994 L.A.S. Johnson Endangered emphasis on plantings Epacris Blackheath Schedule 1, Yes Potential pollinators identified None yet, occurs on Sydney Blue Free hold land may be brought N/A Acquisition of land and Blue Mountains NP No. A very rare and local endemic species in hamiltonii part 1, & germination rates determined Water Catchment area, M ountains by NPWS. management of all danger from trampling and erosion due to Maiden & Endangered in labs. Fire & pollinator regime BMCC reserves, vacant populations. Population at bushwalking. Betche unknown. crown land and free hold Neates Glenn has been blocks fenced to stop bushwalkers interfering and causing degradation. Pterostylis Yallah, Albion Schedule 1, Drafted Basic demographic, fire. N/A N/A N/A N/A N/A gibbosa R.Br. Park, Cummbere part 1, Pollinator and recruitment S.F., Hunter Endangered information. Seed bank and fruit Valley set is unknown. Unknown Blue Mountains NP, Council Unknown balium Higher Blue Schedule 1, Drafted Taxonomic information is Unknown Unknown Unknown Unknown reserves at Bonnie Doon, Shipley ^°hnoides Mountains part 1, known and has been to infer Plateau and Nellies Glen Minn. Endangered potential insect pollination and a annual seed bank. Basic ecology is unknown. Gardens constructed around Garigal NP, Ku-Ring-Gai Chase p'^WV/ea caylei Terry Hills to Schedule 1, Yes - one Extensive demography work Comer of Forest Way and Warringah Possible SIS No. 21 known sites with only 4 of these natural population. NP Belrose, Duffy’s part 1, at a completed by NPWS. Genetics Oates Place - residential occurring in a NP (2 in Garigal NP and 2 Ku- Forest to Endangered national being assessed at present. area being developed. Ring Gai Chase NP). Ingleside * level N/A Part of a population in Budderoo A total of 2000 plants in 4 disjunct locations J S t f f e a Carrington Falls Schedule 1, Currently Seed bank study. None Robertson N/A N/A NP part 1, being reserved within Budderoo NP. WHQris^ Johnson Endangered written '^XtCiillivray * N?A Berowra Waters Nature Reserve, A number of populations exist on land Homsby Heights Schedule 1, Yes Comprehensive study of N/A N/A N/A N/A *o'i,°onia Ku-Ring-Gai Chase NP to Mt Colah part 1, genetics managed by state and local governments and sub,'s R Br- private land. 'p . m axim a Endangered * ~7)>v Much of the population is Ku-Ring-Gai Chase NP, Lane Thought to be. biff ^in ia North western to Schedule 2, Currently Very patchy and mainly Housing subdivision, Green Baulkham Consultation with Council and No to be removed and small Cove NP, Marramarra NP (C heel) Northern Vulnerable being assumptions Road, Glenhaven Hills development controls forming remnant is to be fenced in. •Briggs suburbs of written part of the eight part test. Possible interruption of Sydney natural regimes. Small isolated population or Marramarra NP, Berowra Valley Thought to be. I gl^Qtheca Northern Schedule 2, Currently Basic ecology based upon Housing subdivision, Warringah Development application reject No approximately 40 NR, Ku-Ring-Gai NP, Garrigal J V dulosa suburbs of Vulnerable being studies on T. ericifolia and T Aquatic Drive, Allambie by Council, appeal to the Land individuals. NP. Sydney written shiressii (Benson 1985). Heights. and Environment Court . >). Glenrock SRA, Awabakal NR, Thought to be. J& Q th eca Lake Macquarie Schedule 2, No Hunter Regional Athletics Hunter Munmorah SRA N£a Sm ith to Bulahdelah Vulnerable Facility Chapter 4 Species biology and site descriptions 89

Tetratheca glandulosa and Darwinia biflora were considered suitable species for studying how the TSCA was applied in the development approval process and also how well it has delivered a satisfactory conservation outcome. These two species have been considered in a number of development applications since 1995 with various outcomes. Recovery plans have not yet been completed for either species, highlighting the fact that basic knowledge of the biology of each species and their responses to specific threats were not readily available at the time decisions on development approval were being made. This lack of information makes the process of preparing and assessing 8 part tests and SIS especially difficult.

The following sections of this chapter, outline, for T. glandulosa and D. biflora, a taxonomic description, the recovery plan status and the state of knowledge of ecological attributes. Finally, site descriptions and location are given for study sites that have not been discussed as case studies in Chapter 3.

4.2 Tetratheca glandulosa Smith

Tetratheca glandulosa is in the family Tremandraceae. Species in this Family are typically small, perennial heath-like shrubs, within three genera distributed worldwide. Within the genus Tetratheca, there are twenty five species, all endemic to Australia (Harden 1992).

4.2.1 Taxonomic Description

Tetratheca glandulosa (Plate 4.1) is a small, resprouting, spreading shrub (10 to 50 cm) restricted to ridgetops on Sydney's North Shore and north of the Hawkesbury River1 (Thompson, 1976; Harden 1992; Maryott-Brown & Wilks 1993) (Figure 4.1). Tetratheca glandulosa is strongly associated with particular soil (Thompson 1976), hence producing a restricted and fragmented distribution. The soil type in the area is predominantly

1 The original distribution of this species has changed significantly since European settlement. The continual growth of housing on the ridgetops of the Sydney Basin has greatly restricted the distribution that used to range as far west as the present suburb of Auburn in 1887. Present range is between 33°09', 151°00' to 33°51', 151°13'. Chapter 4 Species biology and site descriptions 90

Plate 4.1: Tetratheca glandulosa is a small spreading shrub growing to between 10 to 50cm (plant can be identified by the mass of pink flowers).

Plate 4.2: Tetratheca glandulosa produces solitary flowers that are deep lilac to pink in colour. Chapter 4 Species biology and site descriptions 91

0 10 20km

Figure 4.1: Approximate known distribution of Tetratheca glandulosa c:::::::> . Also represented is the urban and bushland - interface.

(Sources: map from N.S.W. NPWS and distribution from Harden 1992). Chapter 4 Species biology and site descriptions 92

Hawkesbury Sandstone with shale caps (soil landscapes: Lucas Heights, Lambert, Faulconbridge) (Douglas 1998). It is at this soil transition zone that T. glandulosa occurs, producing scattered discontinuous populations (Douglas 1998). The habitat tends to be either rocky or sandy heath, or low woodland (Harden 1992). More specifically, T. glandulosa is often associated with Banksia ericifolia scrub and Eucalyptus haemostoma/ E. racemosa/E. sparsifolia woodland (Douglas 1998). Similar to other species of Tetratheca, T. glandulosa requires semi-shade and apparent low disturbance levels (e.g. removal of adult plants by fire) for establishment and growth (Maryott-Brown & Wilks 1993). It appears to favour some forms of human-caused disturbances (such as slashing along fire trails).

An individual plant of T. glandulosa will often consist of many stems that appear just above the woody rootstock (Maryott-Brown & Wilks 1993). The leaf characteristics can vary among individuals, with their position on the stem being either alternate or opposite. Tetratheca glandulosa receives its species name from the stiff, gland-tipped hairs that occur on the revolute margins of the leaves, giving a toothed appearance and easily distinguishing the species from other Tetratheca species. These glandular hairs continue along the pedicel and sepals (Harden 1992; Maryott & Wilks 1993).

Since 1995, this species has frequently been detected in proposed construction sites (Douglas 1998) within the Hornsby, Ku-Ring-Gai and Warringah Shires. There have been disputes about the appropriate survey time (month of the year) used for determining the presence and population sizes for T. glandulosa (Douglas 1998). This species is cryptic2 except when observed during the flowering period between July and November (Maryott­ Brown & Wilks 1993). Plants produce deep lilac, pink, solitary flowers3 (Plate 4.2). The fruit is also easily distinguished from others in this genus by the persisting sepals (Thompson 1976) (Plate 4.3 (a) & (b)).

2 The word cryptic is used here to mean that T. glandulosa when not in flower gives the appearance of a grass species and blends with its surrounds.

3 Occasionally flowers are pale pink to white in colour (pers obs.). Chapter 4 Species biology and site descriptions 93

Plate 4.3 (a): The persisting sepals distinguish Fruit produced by Tetratheca glandulosa. Open fruit are also shown on the right hand side of the photograph.

Plate 4.3 (b): A developed Tetratheca glandulosa fruit surrounded by persisting sepals (Objective 40x). Chapter 4 Species biology and site descriptions 94

4.2.2. Recovery Plan Status

A number of ecological attributes should be known about a species to enable the preparation of a recovery plan, in accordance with the Threatened Species Conservation Act 1995 (NSW). However, many of these attributes are unknown for T. glandulosa (Table 4.2) and this poses a problem when trying to determine the threats to the survival of the species. At present the recovery plan has not been written for T. glandulosa.

4.3 Danvinia biflora (Cheel) B. Briggs

Darwinia biflora belongs to the family Myrtaceae. Worldwide, this family consists of 170 genera and 3000 species, with 70 of these genera endemic to Australia. The genus Darwinia is endemic to Australia and almost entirely restricted to New South Wales (Briggs 1962).

4.3.1 Taxonomic Description

4 Darwinia biflora is an obligate seeder . It has an erect or a spreading habit (Plate 4.4). With distinct edaphic5 requirements, D. biflora is characteristically found on Wianamatta shale and laterite areas in which other species of Darwinia are absent (Briggs 1962). Darwinia biflora grows on heath or woodland on sandstone with shale-capped ridges in the northern suburbs of Sydney (Hawkesbury River to Port Jackson) (Figure 4.2) (Harden 1991). Most of these ridge tops have been subjected to residential development and populations have consequently been reduced, causing a disjunct distribution. Darwinia biflora is easily distinguished for other heathland Darwinia species by the pink colour on the stems of the new growth (pers. obs.).

4 Obligate seeding is the trait in which the adult is killed and recruitment occurs solely from a seed bank stored either in the soil or in woody fruits in the canopy (Atwell et al. 1999).

5 An edaphic requirement is a requirement relating to the physical or chemical composition of the soil present in a particular area. Chapter 4 Species biology and site descriptions 95

Table 4.2: Known ecological attributes of and threats to Tetratheca glandulosa, and gaps in knowledge.

Ecological attributes Ecological attributes that are known Consequences (if that should be known known) relating to this attribute

Growth form Low, spreading shrub, 10-15 cm tall with multiple stems NIA (Harden 1992).

Vegetative spread Thought to show some degree of clonality, exact amount Potential threat that some unknown. so-called "populations" might be single clones.

Longevity of an adult Not known but thought to be 6 - 10 years (based on Unknown Benson's (1985) study of T. ericifolia and T. shiressii).

Flowering phenology Individuals flower between July and November. Fire regime may affect flowering, and thus seed production and population size.

Fruit/seed maturity rate and Unknown The lack of information may longevity (seed bank result in inappropriate profile) decisions during the assessment of development applications (e.g. isolation of a population).

Dispersal mechanisms Unknown The seed bank profile is not (ability to recolonise) known at this stage; therefore the importance of dispersal mechanisms is unable to be determined.

Establishment pre- and post- Juveniles are uncommon (10% of population) (Douglas, Unknown disturbance 1998)

Growth (time to reach Can produce seed 3-4 years after establishment A high frequency of maturity) (assumption based on Benson's (1985) work on T. disturbance (e.g. fire) may ericifolia, T. shiressii). not allow for setting of sufficient seed.

Response to aspects of the Woody rootstock present may resprout (assumption by This species is often found fire regime Benson's (1985) work on T. ericifolia, T. shiressii). It is near recent disturbances, also generally assumed (with no evidence) that hot fires in hence at may be late summer or autumn would control the woody disadvantaged if fire is overstorey, therefore making clearings in the vegetation excluded. However for recruitment to occur. individual aspects of the fire regime (wrong season, too frequent) may be contributing to the decline of this species.

Interaction with other Unknown Small and isolated organisms: populations may lead to * pollination inbreeding depression (low * seed predation seed viability resulting). * herbivory Chapter 4 Species biology and site descriptions 96

Plate 4.4: Darwinia bi.flora grows in an erect or spreading habit (plant is inside the circle).

Plate 4.5: Flowers produced by Darwinia bi.flora are small, green m colour and surrounded by red bracteoles. Chapter 4 Species biology and site descriptions 97

0 10 20km

Figure 4.2: Approximate known distribution of Darwinia biflora 0 . The urban and bushland - interface also shown.

(Sources: map from N.S.W. NPWS and distribution from Harden 1991) Chapter 4 Species biology and site descriptions 98

This species obtains its name from its small flowers (Plate 4.5), which occur in pairs. Flowers are inconspicuous, green and surrounded by red bracteoles that are almost as long as the floral tube (Briggs 1962; Harden 1991). This gives the appearance of two different types of flowers, the distinguishing feature for this species. The bracteoles persist until flowering and then fall off to reveal a floral tube (5 to 8mm in length) and a style (10 to 14mm long). Flowers of D. bifl.ora are produced mainly in autumn but flowering does occur intermittently throughout the year, with fruits developing mostly between May and August (Briggs 1962; Harden 1991). The fruits are indehiscent and only slightly larger than the flowers, usually producing one seed each (Harden 1991).

4.3.2 Recovery Plan Status

As for Tetratheca glandulosa, preparation of a recovery plan depends on knowledge of a range of ecological attributes. Most of these are unknown for Darwinia bifl.ora (Table 4.3). How these ecological attributes contribute to the survival of D. biflora is unknown. The recovery plan for D. bifl.ora is in the process of being written.

4.4 Site Descriptions

Each of the six sites chosen for study (Figure 4.3) represents a population conserved either in a National Park or in a Council Reserve (Table 4.4). The habitats found at the sites are a range of woodlands, except at the Cobah Track, Marramarra NP where the vegetation is tall open forest. The dominant species at each are similar, representing the families M yrtaceae, Protecaeae and Fabaceae.

All populations of Tetratheca glandulosa and Darwinia bifl.ora in this study occurred along the edges of fire trails and tracks. Populations tended to disappear beyond 10 metres from the edge of the track. Chapter 4 Species biology and site descriptions 99

Table 4.3: Known and unknown ecological attributes of and threats to Darwinia biflora.

Ecological attributes that Ecological attributes that are known Consequences (if known) should be known relating to this attribute

Growth form Erect or spreading shrub to 80 cm high (Harden NIA 1991)

Vegetative spread Unknown Unknown

Longevity of an adult Unknown Unknown

Flowering phenology Flowering in autumn (Harden 1992). Unknown

Fruit/seed maturity rate and Fruit production May to August. Production is May affect the density of longevity (seed bank variable and low when individual is less than 5 the seed bank, causing profile) years old. Soil seed bank present, reaches its many seeds not to peak 10 years post-fire. Seed viability high germinate. (99% ). Seed bank model (Auld et al. 1993) needs confirmation

Dispersal mechanisms Unknown Unknown (ability to recolonise)

Establishment pre- and No establishment of seedlings without The number of individuals post-disturbance disturbance due to competition of shrub, herb at a site is variable and and grass cover (Auld et al. 1993). should not be used in assessing site significance.

Growth (time to reach 18 months following germination produces Inappropriate disturbance maturity) flowers and fruits (few reach maturity) (Auld et regimes causing al. 1993). elimination of population from a site.

Response to aspects of the Fire is required to break dormancy. Plants are Frequent fires deplete seed fire regime obligate seeders with high recruitment bank (<10 years). following fire. Effect of season of fire not known. Management by fire is recommended (draft Recovery Plan).

Interaction with other Rarely visited by insects (pers obs). However Out-crossing appears to organisms: bisexual, self-pollination thought to occur due occur rarely. This may * pollination to the pattern of floral development (draft cause inbreeding and the * seed predation Recovery Plan). potential to reduce seed * herbivory viability. Chapter 4 Species biology and site descriptions 100

N fi

Marramarra NP*

BobbinHea4 Murra Trail 1t * ClifTOval \ ·~ ~==- . 1~ ~1-~~~ ~. ,. ~ ~ortJackson Kilometres :.r ~~-;--\ ~~ .df I :

Figure 4.3: Location of sites that are conserved in National Parks or Council Reserves (1) Cobah Track, Marramarra NP (2) Long Track, KCNP (3) Bobbin Head, KCNP ( 4) Murra Trail, KCNP (5) Tumbledown Dick Hill, Terry Hills and (6) Cliff Oval, North Wahroonga. Table 4.4: Descriptions, population size and management options for sites that are conserved in National Parks or Council Reserves. 9 Site Species Number of Recent Vegetation Dominant species Management options/instruments ~ ~ individuals disturbances type ""'! Cobah Track, T. glandulosa >100 burned tall open forest Corymbia gummifera, C. eximia, Priority given for the protection of this species ~ Marramarra NP January 1998 Eucalyptus haemastoma, Grevillea through implementation of the recovery plan ~ ~ (33° 35' 40"S buxifolia, G. sericea, Hibbertia sp., and encouragement of research (Plan of ("') 151 o 03' 35"E) Banksia serrata, Fabaceae species Management for Marramarra NP 1998). ~· \.)" (Plate 4.6) s· Long Track, T. glandulosa > 100 no known tall woodland Angophora costata, Corymbia Priority given to this species and protection is KCNP disturbances gummifera, Banksia ericifolia, B. offered through the maintenance of natural g (33° 39' 20"S . serrata, Hakea sericea, Grevillea processes (KCNP Draft Plan of Management ;::i ~ 151° 11' OO"E) buxifolia, G. speciosa and Fabaceae 1999). ~ C'l (Plate 4.7) _ u---- · ·- species ~· Bobbin Head D. biflora >100 burned woodland Angophora costata, Corymbia Priority given to this species and protection is ~ Track, KCNP December gummifera, Boronia ledifolia, B. offered through the maintenance of natural C'l ("') (33° 41' 26"S 1990 floribunda, Comesperma ericinum, processes (KCNP Draft Plan of Management ""'! >B' 151° 09' OO"E) Calytrix tetragona, Acacia 1999)...... (Plate 4.8) suaveolens g· Murra Trail, D. biflora > 100 burned woodland Angophora costata, Corymbia Priority given to this species and protection is C'l KCNP December gummifera, Acacia longifolia, A. offered through the maintenance of natural (33° 41' 40"S 1990 suaveolens, Hakea sericea, Banksia processes (KCNP Draft Plan of Management 151° 08' 33"E) serrata, Grevillea sericea 1999). (Plate 4.9) Tumbledown T. glandulosa >100 no known low open Corymbia gummifera, Grevilleq Management plan yet to be written and Dick Hill, Terry disturbances woodland sericea, G. speciosa, G. buxifolia, implemented Hills G. caylei, Acacia longifolia, Hakea (33° 41' 17"S sericea, Patersonia sericea, 151°14' OO"E) Labertia formosa (Plate 4.10) Cliff Oval, Nth Both species T.glandulosa burned 1999 remnant Eucalyptus haemastoma, Corymbia A recent experimental burn but no specific Wahroonga 1, D.biflora 3 (Plate 4.12) woodland gummifera, Hakea teretifolia, management plan. (33° 42' 18"S Persoonia Levis and Melalueca sp., 151° 07' 5l"E) Leucopogon sp. (Plate 4.11} ,.... 0,.... Chapter 4 Species biology and site descriptions 102

Plate 4.6 (left): Cobah Track, Marramarra National Park.

Plate 4.7 (below): Long Track, Ku-Ring-Gai Chase National Park. Chapter 4 Species biology and site descriptions 103

Plate 4.8: Bobbin Head Track, Ku-Ring-Gai Chase National Park.

Plate 4.9: Murra Trail, Ku-Ring-Gai Chase National Park Chapter 4 Species biology and site descriptions 104

Plate 4.10: Tumbledown Dick Hill, Terry Hills Chapter 4 Species biology and site descriptions 105

Plate 4.11 (above): Cliff Oval, North Wahroonga.

Plate 4.12 (left): Cliff Oval, North Wahroonga following management burn in 1999. Chapter 5 Pollination biology 106

CHAPTER 5 POLLINATION BIOLOGY OF Tetratheca glandulosa AND Darwinia bijlora

5.1 Introduction

Tetratheca glandulosa and Darwinia biffora provide interesting examples of the importance of pollination biology in the conservation of species in the Sydney Basin. Both species are listed as vulnerable and are under threat from the expansion of the urban fringe. Before the threats of fragmentation and appropriate management options can be assessed, knowledge of the pollination biology for both species is required. Tetratheca glandulosa displays an interesting pollination mechanism, through sonication.

The mating systems of Australian hermaphroditic plant species have attracted many studies over the years (Table 5.1). Mating systems have been defined as the "preferred" mating system, "potential" mating system or the "realised" mating system (Richardson et al. 2000). These authors defined preferred mating system (which may be relatively constant in space and time) as the biological constraint set on mating by the plants breeding system. Thus, an obligatory self-pollinated species would be constrained to the highest level of inbreeding. A species with a mixed mating system could produce any level of outcrossing from all selfed seed to all outcrossed seed, depending on the nature of pollination. The realised mating system is defined as the mating system that is actually happening within any given population at a particular time. It can be determined by the use of genetic markers. The realised mating system can vary not only among related species in similar conditions, within the same population at different times and in different populations that are geographically separated (Richardson et al. 2000). It is readily accepted that the main determinants of mating systems include pollinator abundance and movement (pollen flow), pollen presence and quality (compatibility), flower and fruit densities, successful fruit and seed development and germinability of seed. Changes to any one of these components can cause variation in mating systems (Goldingay and Carthew 1998). Chapter 5 Pollination biology 107

Table 5.1: Selected studies between 1986-2000 (taken from a variety of sources), showing a variety of preferred mating systems found in Australian flora.

Author Species Breeding/mating What was measured? s stem Whelan & Goldingay Banksia paludosa Out crossing Pollen supplementation, and (1986) Banksia spinulosa seed set

Gross (1990) Dillwynia hispida Obligate Stigma receptivity, pollen age, Dillwynia uncinata outcrossing pollen supplementation and fruit Pultenaea densifolia set Vaughton (1990,1992) Banksia spinulosa Out crossing Inflorence abundance, fruit set, pollen removal and pollinator visits

Burbidge & Jam es Stylidium species Out crossing Seed abortion (1991)

Gross (1993) Melastoma affine Self compatible Stigma receptivity, pollinator but not visits, genotyping autogmous

Paton (1993) Callistemon rugulosus Out crossing Fruit set

Sampson et al. (1994) Banksia brownii Baxter Mixed mating Allele frequency ex R.Br.

Horskins & Turner Eucalyptus costata unknown Nectar measurements, pollen (1999) load and stigmatic receptivity

Cunningham (2000a) Acacia brachybotrya Out crossing Pollen counts, pollen Eremophila glabra supplementation, pollinator visitation

Krauss (2000) Persoonia mollis Out crossing Fruit and flower densities and genotyping of individuals in populations

Richardson et al. (2000) Grevillea mucronulata Out crossing Pollinator visitation and Grevillea sphacelata alloyzme work to genotype seed

The breeding system (i.e. preferred mating system) of plants can range from complete self­ compatibility (e.g. Banksia spinulosa var. neoanglica, Grevillea macleayana) to obligate out-crossing (Banksia ericifolia). Self-pollination can arise in three different ways (i) autogamy (pollen from the anthers moves to the stigmatic surface within the same flower), (ii) geitonogamy (pollen from a flower moves to another flower within the same plant) and (iii) cleistogamy (selfing is enforced, as the flowers never open) (Crawley 1997). Plant species that have out-crossing breeding systems can be either obligate out-crossers (must Chapter 5 Pollination biology 108

have pollen from other plants) or a mixed breeding system (a combination of self­ compatibility and out-crossing perhaps with outcrossing preferred). Studies examining the breeding system of plants have typically been pollen supplementation and pollinator exclusion experiments, measuring either the presence of pollen tubes in flowers or the level of fruit/seed set (Goldingay et al. 1991; Carthew 1993; Sedgley & Harbard 1993; Krauss 1994) following experimental hand pollination. Seed measurements may also vary depending on where the pollen originated (e.g. seed may be smaller and weight less if the ovule is the result of self-fertilisation c.f. out crossed seed) (Vaughton 1988) (e.g. bee pollinators can reduce the potential to disperse pollen due to the frequency and intensity with which a bee grooms (Larson & Barrett 1999a)). It is thought that smaller seeds have a decreased chance of survival compared to larger, out-crossed seed (Marshall 1986).

Whether a species has the ability to form clones will affect the mating system within a population, depending on the breeding system. Clonal spread can be achieved by either the production of asexual propagules (vegetative spread) or setting asexually produced seed (Ellstrand & Roose 1987). Populations of a clonal species are characterized by; (i) infrequent seedling recruitment (even though some species produce large amount of seeds) (Pomon & Escaravage 1999; Pomon et al. 2000), (ii) low and localised genetic diversity (Persson & Gustavsson 2001) and (iii) confinement to the site with favourable conditions (Okland 1995; Eriksson 1996).

Conservation biology and threatened species legislation operate with a maJor goal of preserving the evolutionary potential of a species by maintaining genetic diversity (Travis et al. 1996; ANZECC 2001) hence, pollination ecology is fundamental to conservation biology. While habitat and demographic factors play an important role in managing for threatened species, as defined in management plans or recovery plans, genetic considerations are also important for plant conservation in both the short term and the long term (Frankel et al. 1995; Young & Brown 1996). A major, current concern of conservation biology is how many genetic individuals are present at any one site (Esselman et al. 1999). What constitutes an individual plant is not a straightforward concept, especially if a species is clonal (Escaravage et al. 1998). If a threatened plant species displays clonality, there can Chapter 5 Pollination biology 109

be a number of implications for its conservation. These include (i) inaccurate assessment of observation of the true conservation status (i.e. overestimation of the numbers of genetically distinct individuals, which may lead to inadequate conservation, and a consequent loss of genetic variation), and (ii) the use of general principles for reserve design or ex-situ collections which have been formulated for non-clonal plants (i.e. restriction of pollen flow to within clones may lead to lower fruit set in a self-incompatible species or for inbreeding in a self-compatible species) (Sydes & Peakall 1998). Hence, an understanding of what constitutes an individual clone within a population, the number of genetic individuals that occur and their spatial distribution are often crucial questions that need to be asked when examining the evolutionary potential of a species (Escaravage et al. 1998).

A large number of plant species are now found in restricted, small populations and geographically isolated habitat remnants due to the processes of habitat destruction and fragmentation (Young et al. 1999). Alterations to the breeding system of a plant (e.g. resulting from habitat fragmentation) can ultimately impact upon seedling recruitment. Habitat fragmentation has been associated with disruptions to pollinator services in agricultural environments (Allen-Wardell et al. 1998; Kearns et al. 1998). When pollinator services break down, a number of changes may occur, leading to altered breeding system (i) removal of the pollinator from the system (Mustajarvi et al. 2001), (ii) lower pollinator visitation rate (Ferdy et al. 1999; Walther-Hellwig & Frankl 2000), (iii) reduced fruit production (Cunningham 2000b), (iv) reduce ovule-to-seed ratio (Baker et al. 2000), (v) increase movement of self pollen (Karron et al. 1995) and (vi) a decrease in seed fitness (germination) (Brown & Kephart 1999; Buza et al. 2000). To comprehend these changes, it is important to have an understanding of the breeding system, which will also help with formulating conservation management plans and making appropriate decisions in regard to urban development. Chapter 5 Pollination biology 110

5.2 Aims

The breeding system and pollination ecology of threatened plants clearly have implications for: (i) accurate assessment of threats; (ii) assessment of the effective mitigation actions; and, (iii) predicting impacts of developments. Little is known of either the mating systems or the pollination ecology of Tetratheca glandulosa or Darwinia biflora. Recovery teams in N.S.W. (see Chapter 2) have identified this knowledge as important to ensure effective conservation and recovery of these species (see Chapter 4). Despite this, decisions have been made on development proposals for lands known to support individuals of these species. In this chapter, I research details of the breeding systems for T. glandulosa and D. biflora and then assess the implications for conservation and the effectiveness of the TSCA. To achieve this, the following four specific aims were established: (1) to determine the observed mating system for Tetratheca glandulosa and Darwinia biflora at a number of sites; (2) to determine whether pollen parentage (natural, cross, self) has an effect on seed weight and length for T. glandulosa and D. biflora; (3) to determine the identity and conservation status of potential pollinators for T. glandulosa and D. biflora; and, (4) to determine if T. glandulosa is a clonal species.

5.3 Materials and Methods

5.3.1 Determination of the mating system

To determine the preferred mating system of a species pollen supplementation and pollinator exclusion experiments are used (Kearns & Inouye 1993). Seed set is measured as the outcome of such experiments.

A number of mature plants (10-65 individuals per treatment) of T. glandulosa and D. biflora were selected at each site, reflecting numbers of individuals available within Chapter 5 Pollination biology 111

naturally occurring populations (Table 5.2). Using a randomised cluster (block) design (Krebs 1989), individuals were allocated to one of three treatments within each block: (i) bagged1 and self-pollinated; (ii) bagged and cross-pollinated (xenogamy); and (iii) unbagged and not manipulated (natural pollination).

Table 5.2: The number of individual plants used at each site for Tetratheca glandulosa and Darwinia biffora for pollen supplementation and pollinator exclusion experiments.

Species Site Number of individuals per treatment Tetratheca glandulosa Marramarra NP 30 KCNP 30 Tumbledown Dick Hill 20 ARDEL 10 Cliff Oval 1* Darwinia biflora Murra Trail KCNP 65 Bobbin Head Track, KCNP 65 Glenhaven 30 Cliff Oval 1

*At Cliff Oval there was only one individual of Tetratheca glandulosa. All treatments were performed on the one individual.

Tetratheca glandulosa and Darwinia biflora produce a flush of flowering at the beginning of winter (July and June respectively). Due to the different pollinator system of each species, two different methods for pollen treatments were used. Flowers in treatments (i) and (ii) were bagged while still in bud to eliminate animal pollinators. Flowers in treatment (iii) were left exposed to natural pollination but bagged once fruit began to mature, to prevent any seeds set from escaping. Pollen donors for the outcross treatment in Darwinia biflora were arbitrarily chosen from a mixture of plants at least 20m away (other end of site). Pollen donors for Tetratheca glandulosa came from a different site (see Table 5.3), because this species has the potential to be clonal (section 5.4.4) and, without prior genetic studies, it was impossible to know whether all plants at a single site were more than a single distinct individual. Pollen was removed from the pollen presenter of Darwinia

1 Bags were constructed of bridal tulle (1 x 1 mm) cut into 16 cm square pieces. A drawstring was threaded around the edge. The square of tulle was placed over a flower and drawstring pulled tight, thus excluding pollinators. The flowers of both species are delicate and heavier material would have crushed flower parts. Chapter 5 Pollination biology 112

biflora using the tip of a scalpel blade and then placed on the stigmatic surface of receptive flowers. The removal of pollen from Tetratheca glandulosa involved vibrating the anthers with a tuning fork ( 440 hz) with a piece of sticky tape was held above the flower to receive the pollen (refer to Appendix 5.1 for more information on sonication pollination). The pollen released as a cloud, which adhered to the piece of tape. Clumps of pollen were then removed from the tape using tweezers and placed on the stigmatic surface of recipient flowers.

Table 5.3: Table showing where donor pollen for Tetratheca glandulosa came for outcross-pollination experiments at each site. Sites further away were chosen as donor sites

Site of cross-pollination Pollen donor site

Marramarra NP Tumbledown Dick Hill

KCNP Marramarra NP

Tumbledown Dick Hill KCNP

ARD EL Tumbledown Dick Hill

Cliff Oval KCNP

Seeds were collected from the bags in December for Tetratheca glandulosa and in January 2 for Darwinia biflora. Seeds for each treatment within a site were scored for seed predation , 3 4 abortion , and viability •

The effect of treatment on flowers successfully setting fruit and the fruit viability were assessed using a three-factor chi-squared test (using SPSS). Data were collected over two

2 Where a seed had been partially destroyed or showed other signs of invasion (e.g. holes in the seed coat) it was scored as eaten.

3 Fruits that failed to develop were scored as aborted.

4 A seed was scored as viable if it germinated or when the seed coat was removed the seed was white and hard and tested positive with tetrazolium. Chapter 5 Pollination biology 113

years and analyses were conducted independently for each year because, I was confident that data collected for each year was independent from each other (i.e. different, randomly selected set of plants each year).

5.3.2 Seed fitness (weight and length)

For each species, 10 seeds from each treatment (one seed per plant) (section 5.3.1) were collected in 2000, were measured in length, and weighed. Length was measured to the nearest millimeter using vernier calipers, and seeds were weighed to 10-3g using an electronic balance.

Data were analysed (individually for seed length and weight) using a two factor ANOVA (JMP, Version 3.0.2. 1989-94). The two factors were site (fixed factor) and treatment, also a fixed factor. All statistical testing was performed at an a-level of 0.05.

5.3.3 Pollinator identification

Insect traps were set up in early, middle and late flowering season in 1999 and 2000. The traps consisted of a microscope slide with a layer of "Bird-off™" applied to one side (Plate 5 .1 & Plate 5 .2; A. York pers corns). The microscope slide was then attached to a 15mm fold back clip wired to a bamboo pole. Traps were positioned to face individual flowers and left for one week. When traps were collected, a clean microscope slide was placed over the used slide to avoid contamination during transportation back to the lab. Microscope slides were then examined under a light microscope for the presence of pollen5 on either the insect or slide. Microscope slides were then soaked in "Baby-oil™" for 48 hours to dissolve the "Bird-off™". Insects were removed and stored in 70% ethanol for identification.

5 Pollen was verified at belonging to either Tetratheca glandulosa or Darwinia bi/fora by using a previously created pollen library (see appendix 5.2) of these two species and other species found at the sites. Chapter 5 Pollination biology 114

Plate 5.1: A 'sticky' (insect) trap used to capture potential pollinators at an individual of Tetratheca glandulosa.

Plate 5.2: A microscope slide covered in 'bird-off and attached to a bamboo pole to form an insect trap to capture potential pollinators at an individual of Tetratheca glandulosa. Chapter 5 Pollination biology 115

Twenty traps were placed at Marramarra N.P., Tumbledown Dick Hill, Glenhaven and at each of the three sites in KCNP. Only five traps were placed at ARDEL and four at Cliff Oval.

5.3.4 Test for clonality in Tetratheca glandulosa

Amplified fragment length polymorphisms (AFLPs) were chosen as a genetic marker to test for clonality in Tetratheca glandulosa. This marker system was chosen, as it is a PCR based system that uses targeted amplification as opposed to random searching of the polymorphism chain (Krauss & Peakall 1998).

DNA was extracted from fresh stem material usmg the CTAB extraction procedure described in Doyle & Doyle (1987), using liquid nitrogen to powder the tissue. DNA samples were stored in 50µ1 of TE at -20 C. Ten individuals were sampled from populations at ARDEL, KCNP and Marramarra N.P. and one individual at Cliff Oval.

A four-step procedure was used to produce AFLPs: (1) Restriction digest of genomic DNA. 16.6µ1 of DNA was digested with 8.4µ1 of master mix (1.4µ1 RNAse, 5µ1 5x reaction buffer, 2µ1 ECOR l/Mse I) and incubated for 2 hours at 37 Cina water bath and then incubated for a further 15 min at 70 Cina heat block. The sample was then placed on ice. (2) Ligation of adapters. A mix of 20.16µ1 of adapter ligation solution and 0.84µ1 of T4 DNA ligase was added to the digested DNA and incubated at 20 C for 2 hours and then diluted with 1:10 TE buffer. (3) Pre-amplification reactions. 5µ1 of diluted template DNA was added to 37µ1 of pre-amp primer mix, 5µ1 of lOx PCR buffer, 3µ1 25mM MgCh and 0.5µ1 of Taq DNA polymerase in a small eppendorf tube. A PCR was preformed for 20 cycles, each cycle consisting of 30s at 94 C, 60s at 56 C and 60s at 72 C. The resulting solution was further diluted with 1 :50 TE buffer. (4) Selective AFLP Amplification. 2.5µ1 of pre-amp DNA was combined with 0.05µ1 Taq DNA polymerase, lµl lOx PCR buffer, 3.35µ1 MilliQ water, 0.6µ1 25mM MgCh, Chapter 5 Pollination biology 116

0.25µ1 EcoRI-primer and 2.25µ1 MseI-primer (Table 5.4). The PCR reaction (touchdown) began with one cycle of 30s at 94 C, 30s at 65 C and 60s at 72 C. Then subsequent cycles, the annealing temperature was reduced in ldeg steps from 65 C to 56 C, followed by 23 cycles at 56 C. Table 5.4: Primer pairs that were tested. Asterisk (*) indicates the pair that was eventually used for analysis.

Fluorescently labeled primers Unlabelled primers EcoRIACC MseI-CAC EcoRI-ACT MseI-CAC * EcoRI-AGG MseI-CAC EcoRIACA MseI-CAT EcoRIACC MseI-CAT EcoRIAGG MseI-CAT

Fragments were visualised by gel electrophoresis (5% acrylamide gels), using the ABI Prism 377 Automated Genetic Analysis System (AGAS). This system uses a laser, which migrates down the gel detecting uniquely coloured fluorescent dye. Details of how the system operates are discussed in Krauss & Peakall (1998).

Fourteen profiles (outputs from the AGAS) were examined for repeatability, by extracting DNA from the same individual plant on two different days. Profiles that were not repeatable were removed from the overall data set. A total of seven profiles of the set where deemed to have 100% repeatability.

Problems with 'noise' and un-repeatability with some runs posed a number of problems. In order to minimize these problems, a number of criteria were applied, to select which profiles to analyse. (1) The overall quality of the profile was assessed and the profile was excluded if peaks were not readily identified relative to background 'noise'; (2) For each profile, only the clean part of the trace was identified (i.e. the range of fragment sizes with clean peaks); (3) Bands that were at a frequency of less than a 100 units where removed from profiles; and, Chapter 5 Pollination biology 117

(4) Ten loci were removed to, as they were not repeatable. These loci were 108, 126, 152, 153, 172, 184, 241, 246, 271, and 387.

A complete examination of the data set was conducted by comparing each individual (or profile) with every other individual (or profile). Comparison involved examining bands, which were both present and absent in individuals. Initially, similarity was calculated using the following equation:

n n- L(xi - yJ2 1 Similarity = ----'i--"='------n

Where n = number of polymorphisms

Xi = binary trace result for locus i on individual 1 Yi = binary trace result for locus i on individual 2

Genetic distances were calculated (Peakall & Smouse 2001) and plotted using Principal Coordinates Analysis (PCA). PCA plots the relationship between distance matrix elements based on their first two principal coordinates (Peakall & Smouse 2001). Data were analysed using an Analysis of Molecular Variance (AMOVA) (Peakall & Smouse 2001).

5.4 Results

5.4.1 Determination of the mating system

Percentage offlowers setting viable fruit

Successful fruit set indicated that the glue on the sticky tape did not affect pollen viability (Figure 5.1). The percentage of flowers setting fruit varied significantly among sites and treatments for Tetratheca glandulosa in 1999 but not in 2000 (Figure 5.1 (a), (b)). In general, there were few or no fruits set from self-pollination at any sites in either year, while cross-pollination (by hand) produced between 30% Chapter 5 Pollination biology 118

(a) Fruit set 1999 (b) Fruit set 2000 LOO - 100 -

] r<. R: '·. ~- ~ 80 - 80 - ' ..-.:::. ··... .s ~:: ··. ., ·... ·,, ~ ·•. ··, - ', ··.. "'

(c) Percentage viable 1999 (d) Percentage viable 2000 100- 100 -

80 - 80- ., ., ., ·· ... ·,,' ·· ...... , ··. 60- '-, '· "· ... ··...... '·, 0 ··... , "''"·.... 40- -~' ~~ ., '· '·'',...... ',. > 20- "' 0 - p.; i:: ~ c;; p.; r:= ~ c;; ~ ~ '1.l > '1.l i 0 ~ > u 'O 0 i 0 0 ro..... ~ Q) ro ~ 'O ti:::; !:s I::: ~ Q) a ::c ~ ro 0 ::c ~ 0 ~ s sro s E-<= ..... ~ ~ a ~ ~ Site

Figure 5.1: The percentage of flowers setting fruit for Tetratheca glandulosa in 1999 (a) and 2000 (b) and the percentage of the fruit which was viable in 1999 ( c) and 2000 ( d) at five different sites where populations are different sizes. The site at Cliff Oval consists of one individual. Fruit was set following hand pollination of flowers(• open-pollination, D self-pollination, and rn cross-pollination). Chapter 5 Pollination biology 119

and 85% fruit set, depending on the site and year (Figure 5.1 (c), (d)). Un­ manipulated flowers (open-pollination) at the two smaller disturbed sites (ARD EL and Cliff Oval) failed to produce fruits in either year.

Fruit production varied among sites and treatments for Darwinia biflora in 1999 (Figure 5.2 (a)) but not significantly in 2000 (Figure 5.3 (b)). The percentage of fruit set was significantly different between sites in both years. In general open­ pollination set the highest amount of fruit (66%-80%). Fruit was not produced, regardless of treatment for the site at Cliff Oval.

While percentage of fruit set was high, viability was low for Darwinia biflora (Figure 5.2 (c), (d)). The most viable fruit was set at the site at Bobbin Head in 1999 following the open-pollination treatment (84% ). Fruit viability was significantly different between sites in this year (1999). In 2000, there was no significant difference detected, with fruit viability between 24% (self-pollination) and 65% (cross-pollination and self-pollination).

Seed development

Seed production in Tetratheca glandulosa was classified into four different categories (aborted seed, seed eaten by insects before dispersal, viable whole seed and unviable whole seed) (Figure 5.3). Seed was more likely to be aborted in the self-pollination treatment (93.75% -100%). Pre-dispersal predation was observed in both years and most sites but at a higher rate in 2000 (getting to 26% ). A large amount of viable seeds were observed across site and years compared to non-viable seeds (e.g. Marramarra 50% of seed viable c.f. 27.5% of seed not viable in 1999).

Aborted seed was observed across sites, years and treatments ( <7 5 %) in Darwinia biflora (Figure 5.4). Pre-dispersal seed predation destroyed between 5%-30% of seed across sites, years and treatments. Seed viability was not always greater than Chapter 5 Pollination biology 120

(a) Fruit set 1999 (b) Fruit set 2000 100 100 -

:!:::! :::J 80 ...... 80 - al c E Q) 60 ..."' "'Q) 31: 0 40 - 40 ....;;:::: 0 Q) al ~ 20 ..c Q) ...(.) Q) Q. 0 '@ .,, c: ... E!:: "' "> 0 ::c:" ..c:"' "' .!3 !t:l ~ .D .D " 0 ::E 0 6 .c

(c) Percentage viable 1999 ( d) Percentage viable 2000

100 100

80 80 ·:;- ~ GI 60 jj 60 m > Ci GI 40 40 f GI u lii 20 20 a.

0 .,, 0 '"""' c: O! .,, ... E!:: "' :;" 0 '""~ "' "> ::c:" -"' :i::" " 0 ~ .!3 !t:l "' .D "' ~ !t:l .D " 0 ~ :E" ::E"' 0 6 .D 0 ::E 0 6 p:i Site .c

Figure 5.2: The percentage of flowers setting fruit for Darwinia bifl.ora in (a) 1999 and (b) 2000 and the percentage of the fruit which was viable in (c) 1999 and (d) 2000 at four different sites where populations are different sizes. Treatments were (•-open pollination,o-self pollination, and~ -cross pollination). Cliff Oval was burned in 1999. Chapter 5 Pollination biology 121

(a) Marramarra NP Onen (b) Marramarra NP Self ( c) Marramarra NP Cross (60 seeds) ( 60 seeds) 100 (60 seeds) (60 seeds) (60 seeds) (60 seeds) 100 100 ...... ••••••••••••u•••• ...... 80 ~ggggg~H~g. 80 80

60 60 60

40 40 40

20 20 20

1999 2000

(d) KCNP Open (e) KCNP Self (f) KCNP Cross

(60 seeds) (60 seeds) (60 seeds) (60 seeds) (60 seeds) 100 100 ...... 100 ...... ~!~gg;gg;fr ;r .::::::::::::::::: ...... •••••••••••o••••• ...... -:::::::::::::::::: ::::::::::::::::::...... 80 ...... 80 ...... 80 ··················......

60 60 60 ......

= 40 40 40 -=~ t 20 20 ~ 20 "O Q> Q> 0 0 I'll "' c..c ~ Q> 1:)1;) ...... = (g) Tumbledown (h) Tumbledown Cross 5 (40 seeds) Cj 100 ...... 100 ...... ~ ~ = 80 80 =Q> ~ 60 60

40 40

20 20

1999 2000

Figure 5.3: The mean percentage of seed (averaged (i) ARDEL Cross over plants for Tetratheca glandulosa that were (40 seeds) (60 seeds) 100 aborted following pollination (•), eaten by insects ···················...... ················· before dispersal(~ , viable whole seed (D) and seed i~ i~ii~~~i ~iiii!ii: 80 that was whole but not viable (ID). Seeds were produced following exclusion experiments (open, 60 self and cross pollination) at Marramarra N.P., KCNP, Tumbledown and ARDEL in 1999 and 2000. 40

Seed was not set at Tumbledown in the self­ 20 pollination treatment or at ARDEL in the self- or open-pollination treatment. 0 Chapter 5 Pollination biology 122

(a) Murra Trail Open (b) Murra Trail Self (c) Murra Trail Cross

(130seeds) (130 seeds) (130 seeds) (130 seeds) (130seeds) (130 seeds) 100 100 100 ·-·-·····-·-·····...... -···· ...... ~HHH~HHH~~~~: 80 80 80 :;;;;;;;;;:;;;;;;;

60 60 60

40 40 40

20 20 20

( d) Bobbin Head Open (e) Bobbin Head Self (t) Bobbin Head Cross

(130 seeds) (130seeds) (130 seeds) (130seeds) (130 seeds) (130 seeds) 100 100 100 ...... ······ ...... ::::::::::::::;::...... 80 ...... 80 :::::::::::::::::: 80 ...... , ......

60 60 60

40 40 40

20 20 20

0 0

(g) Glenhaven Open (h) Glenhaven Self (i) Glenhaven Cross

(60 seeds) (60 seeds) 100 (60 seeds) (60 seeds) ...... 100 ...... (60 seeds) (SO seeds) ...... 100 ...... ••••••••••••••••u ...... :::::::::::::::::· ...... 80 ...... ···············--· ...... ~················· 80

60 60 60

40 40

20 20

Year

Figure 5.4: The mean percentage of seed (averaged over plants) for Darwinia biflora that were aborted following pollination (•), eaten by insects before dispersal (lS:I), viable whole seed (0) and seed that was whole but not viable (!1 ). Seeds were produced following exclusion experiments (open-, self- and cross-pollination) at Murra Trail, Bobbin Head and Glenhaven in 1999 and 2000. Chapter 5 Pollination biology 123

seed non- viability and seed viability was greatest in the self-pollination treatment at approximately 45% of the seed crop.

5.4.2 Seed weight and length

The mean weight (gm) of individual seeds of Tetratheca glandulosa varied little across sites and treatments (Figure 5.5 (a)) with no significant difference detected between sites or treatment groups (Table 5.5). The mean length (mm) of the same individual seeds also varied little across sites (Figure 5.5 (c)). Generally, selfed seeds were smaller for T. glandulosa (where they were available). A significant difference occurred between the combination of site x treatment (p<0.05, F(2)=0.017) (Table 5.6).

Table 5.5: Two-factor ANOVA for seed weight for three sites across two treatments (open­ and cross-pollination) for Tetratheca glandulosa (significance p<0.05). Data were untransformed (* denotes a significant effect). Cliff Oval, ARDEL and self-pollination treatment were not included in the analyses.

Source of variation SS df MS F p Site 0.00000343 2 0.0000017 1.61 0.210 Treatment 0.00000202 1 0.0000202 1.89 0.175 Site x Treatment 0.00000543 2 0.0000027 2.54 0.088 Residual 0.00006858 59 0.0000012

Table 5.6: Two-factor ANOVA for seed length at three different sites across two treatments (open- and cross-pollination) for Tetratheca glandulosa (significance p<0.05). Data were untransformed (* denotes a significance effect). Cliff Oval, ARDEL and self­ pollination treatment were not included in the analyses.

Source of variation SS df MS F p Site 1.23 2 0.62 1.65 0.200 Treatment 0.42 1 0.42 1.12 0.295 Site x Treatment 3.23 2 1.62 4.34 0.017* Residual 20.10 54 0.37

The mean weights and sizes of individual seeds of Darwinia biflora did not differ significantly between sites or treatments (Figure 5.5 (b), (d), Table 5.7 & 5.8). Seed weight of D. biflora was measured between 0.003±0.000lgm and 0.004±0.000lgm, while seed length was measured between 4.1±0.000lmm and 3.5±0.05mm. Chapter 5 Pollination biology 124

(a) T. glandulosa (b) D. biflora 0.004 0.005

T ~ ~ ~ ~ iii 0.003 0.004 ,,:I ·:; '6 .5 T T T 0.003 0 0.002 T T T -E1::1Ill .g: 0.002 .c- Ill DI 1 0.001 c 0.001 ca GI :Ii 0 ;..l .,.; ~ 0 z "' ~ :i u 6'" ';j 'O 5 Ol ~ j E!:: "' 0 .c ~ lE :r:" ~ ~ "'t:: .!3 lE 8 8 0 .c ""'5 ~ ::E" .g 6 0 "'~ ::E !!!

(c) T. glandulosa ( d) D. biflora 5- 5

iii T 4 :I 'tJ T T :c> T .E 3 T 0 e 111 E -o 2 ~: 2 .ca, Ill c ..!! c ca GI :IE 0 .,.; 0 :z: ~ '§ 'O Ol ] '"0 " > "'t:: ~ ii:: E-< ~"' ~" 0 .c -" ..8 8 ""u "'t:: lE .. :0"' " ..t:: ~ ::E" .c 6" 0 ~ iB Site

Figure 5.5: The mean weights (gm) of individual seeds of (a) Tetratheca glandulosa and (b) Darwinia biflora and the mean length (mm) of an individual seed for (c) T. glandulosa and ( d) D biflora where seeds have been collected from flowers that have been hand pollinated (• open-pollination, D self-pollination, and~ cross-pollination). Asterisks represent significant differences. Error bars represent standard error. Chapter 5 Pollination biology 125

Table 5.7: Two-factor ANOVA for seed weight at four different sites across three treatments (open-, self- and cross-pollination) for Darwinia biffora (significance p<0.05). Data were untransformed (*denotes a significance effect).

Source of variation SS df MS F p Site 0.00000349 2 0.000001745 2.27 0.110 Treatment 0.00000056 2 0.000000280 0.36 0.698 Site x Treatment 0.00000798 4 0.000001995 2.59 0.043 Residual 0.00006230 81 0.000000769

Table 5.8: Two-factor ANOVA for seed length at four different sites across three treatments (open-, self- and cross-pollination) for Darwinia bif/,ora (significance p<0.05). Data were untransformed (*denotes a significance effect).

Source of variation SS df MS F p Site 0.47 2 0.23 0.70 0.501 Treatment 0.87 2 0.43 1.30 0.280 Site x Treatment 2.07 4 0.52 1.54 0.197 Residual 27.10 81 0.33

5 .4.3 Pollinator identification

Four insects were caught in the traps across all sites. All individuals came from the order Diptera with 2 individuals from the family Asilida (Murra Trail and KCNP) and 2 individuals from the family Muscidae collected at Tumbledown. These results indicate that the technique was not very successful in capturing potential pollinators for Tetratheca glandulosa and Darwinia biflora.

5.4.4 Test for clonality in Tetratheca glandulosa

Once profiles had been examined in accordance to the criteria set down in section 5.3.4, repeatability was 100% using similarity data. A PCA graph was plotted; using genetic distances showed that individual repeats were identical (Figure 5.6). The AMOVA showed that most genetic variation was going to occur within populations (83%) as opposed to among populations (17% ). Chapter 5 Pollination biology 126

I • K5aK5b

• K6b K6c I K4a K4b • K9a K9b

I I K8a • K3a K3b K8b •• A12a AJ2b

AxisX

Figure 5.6: A PCA plot for repeat AFLP profiles for Tetratheca glandulosa. Repeats come from two different sites KCNP (K) and ARDEL (A). And only some individuals (indicated by the number) were repeated. Each dot represents a plant.

K2 ·~ K3

• Kl

AxisX

Figure 5.7: A PCA plot showing AFLP profile data for Tetratheca glandulosa for four populations; ARDEL (A), Cliff Oval (C), KCNP (K) and Marramarra NP (M). The individuals at each site is identified by a number. Each dot represents a plant. It should be noted that plants that are close together have similar genotypes. Sites are grouped together. Chapter 5 Pollination biology 127

Genetic distance was calculated and while 93% of diversity occurs within populations, individuals within a population still clumped together when plotted on a PCA graph (Figure 5.7). The population at ARDEL showed high levels of similarity (81-97%) compared to other sites (e.g. KCNP 58-92%, Marramarra 70-94%) (Figure 5.9). If T. glandulosa were indeed a clonal species then we would expect the similarity index to be 1.

5.5 Discussion and Conclusions

Tetratheca glandulosa appears to have a breeding system that may be partially self­ compatible ([10% fruit set across sites) with a strong preference for outcrossing. This mating system occurs commonly in Australian flora across many plant families (Vaughton & Ramsey 1991; Krauss 1994; Vaughton 1996; Kalinganire et al. 2000). The breeding system of a particular species may not be a constant trait between populations (Vaughton 1988; Hermanutz et al. 1998), which may explain why some individuals set fruit from self pollen in some sites. Darwinia biflora exhibits a self compatible breeding system. It is not unusual for a member of the Myrtaceae family to show a self compatible breeding system. Self-pollination and self compatibility has been reported in a number of Darwinia species (Beardsell et al. 1993).

Fruit set6 in T. glandulosa varied among sites in the open-pollination treatments, with no fruit set in small, disturbed sites. No clear pattern was displayed in fruit set for D. biflora. Seed abortion was high in both T. glandulosa and D. biflora, regardless of treatment group. A number of studies have shown that many hermaphroditic plants show a low fruiting success (Ayre & Whelan 1989; Day et al. 1997) irrespective of different pollination treatments (Vaughton 1988).

6 By October, flowers of D. biflora have begun to lose their bracts and change colour (yellow), with a pronounced swelling at the ovule (pers. obs.). Compared with T. glandulosa while the fruit is growing the petals become limp and the flower gives a closed appearance. The developing fruit is white/green in colour with the style still attached (pers. obs.). Chapter 5 Pollination biology 128

(a) ARDEL (b) KCNP 10 - 10 -

8 - 8 - ·"'..; ~ -; ::I ·;;::"O 6 - 6 - :a .!:! ...0 4 - .. 4 - ,.Q 6 z::I 2 - 2 -

0 - I ~ .,.., •r, •r, ....., .,.., ~, Q"( r-: ""! 0 ~ C• 0 ~ 0 0 a 0

Similarity index between genotypic pairs Similarity index between genotypic pairs

(c) Marramarra N.P. 10 -

.;"' 8 - g,, -; ::I "O :~ 6 - "O .....!:! 0 .. 4 - -a z::I 2 -

Similarity index between genotypic pairs

Figure 5.8: Frequency distribution of the similarity index calculated for pairs of individual plants for three different populations (ARDEL, KCNP and Marramarra N.P.) for Tetratheca glandulosa. Chapter 5 Pollination biology 129

Many genera of Myrtaceae have protandrous flowers with a separate male and female phase. Stylar extensions occur commonly throughout the family and coincide with the onset of stigma receptivity (recorded in Darwinia species, Leptospermum and Baeckea species). Even though stylar extensions promote out crossing, it does not prevent self pollen from occurring on the stigmatic surface. Tetratheca glandulosa produces pink flowers, which are larger than the green flowers produced by D. biflora. Large flowers are often associated with the promotion of out crossing as they are usually showy and attract pollinators while small flowers are often seen to promote selfing (Harrison et al. 1999).

Seed predation can further reduce the availability of viable seeds (Vaughton 1988; Auld 1991; Auld & Denham 1999; Brown & Whelan 1999). Pre-dispersal seed predation has the potential to impact upon the seed crop each year for T. glandulosa (23%) and D. biflora (30% ). Insects are major predators of seeds in many terrestrial plant communities (Robertson et al. 1990) and can influence seed viability before or after seed dispersal. This is achieved by decreasing the number of seeds that mature, ultimately decreasing the number of recruits. For both T. glandulosa and D. biflora, pre-dispersal seed predation fluctuated through time across sites. In other studies, pre-dispersal seed predation has been shown to be heterogeneous through time and space (Auld 1986).

The bagging of flowers for experiments has previously been found to lead to lower seed weight and length (Krauss 1994). However, in this experiment there were no significant differences between treatments or sites for either T. glandulosa or D. biflora. Other studies have shown that lower seed weight and length indicate less fit seeds, leading to a competitive disadvantage during recruitment (Harper 1977; Marshall 1986; Primack 1987; Johnston 1992; Brundet & Mundt 2000). However, the relationship between seed size and seedling fitness may not be linear and yields a far more complex relationship within and between species (Marshall 1986). Neither seed length nor weight varied significantly among treatments in D. biflora. In contrast, the four selfed seeds of T. glandulosa were much smaller than seed produced in the open-pollination and cross-pollination treatments (see Figure 5.5). Chapter 5 Pollination biology 130

It has been thought that Tetratheca glandulosa might form clonal populations. Infrequent seedling recruitment and seed production is a common feature of clonal populations (Pomon & Escaravage 1999). Field observations have noted vegetative spread associated with some individuals (pers. obs.). Data collected using AFLP markers showed a high level of genetic diversity within populations, a pattern that is common in clonal species (Escaravage et al. 1998; Pornon & Escaravage 1999; Pornon et al. 2000), and can be explained by microsite heterogeneity (Pornon & Escaravage 1999). Frequency of similarity between individuals within populations was high and shows a similar pattern to that found in the clonal species Rhododendron ferrugineum (Pornon et al. 2000). How clonal Tetratheca glandulosa is cannot be shown from these data. The use of AFLP markers was successful in determining individuals, the inability to determine clonality may be an issue of sampling on a fine scale for individual plant material. Due to the high level of similarity present in some populations (ARDEL and KCNP) presence of clonal individuals should not be dismissed from recovery management options.

Aizen & Feinsinger (1994) recorded that small habitat fragments were species-poor in potential native pollinators. I caught very few insects in traps; in particular none were recorded at the sites at ARDEL, Cliff Oval or Glenhaven. It is unlikely that the individuals that were caught are potential pollinators, as none of the individuals are classified in any of the recognized groups of buzz pollinators (see Appendix 5.1) (e.g. teddy bear beesAmegilla (Asaropoda) bombiformis, blue banded bees Amegilla (Zongamegilla) cingulata and carpenter bees Lestis sp. (Hogendoorn et al. 2000)). Species that rely on buzz pollination usually disperse pollen from their anthers as a response to vibrations and frequency­ dependent dispersal of pollen is often taken as evidence to support the presence of a buzz pollinated system (Harrison et al. 1999). Chapter 6 Pollination Success 131

CHAPTER 6 POLLINATION SUCCESS OF Tetratheca glandulosa AND Darwinia biflora. Association with fragmentation

6.1 Introduction

Fragmentation of habitats is assumed to have a detrimental impact upon many differ·ent types of organisms (Soule et al. 1992; Houland et al. 1999; Scariot 1999; Kearns 2001). There can be many forms of impacts, such as changes to gene flow (e.g. seed dispersal), increased invasion of exotic organisms, increased potential of diseases, edge effects, and changes in ecological processes such as pollination (Andren & Angelstam 1988; Rathcke & Jules 1993; Kearns & Inouye 1997; Nour et al. 1997; Gilfedder & Kirkpatrick 1998; Steffan-Dewenter & Tscharntke 1999). Often the conservation of threatened plants must be attempted when individuals are confined to small, locally restricted populations in a fragmented landscape (Wolf et al. 2000). Fragmentation is viewed as one of the most common causes of endangerment amongst plant species (Coates et al. 1999).

Modifications to habitats by humans have resulted in what is becoming known as the "pollinator crisis scenario" (Buchmann & Nabhan 1996, Kearns et al. 1998; Williams et al. 2001). This scenario involves either the loss of specialised pollinator guilds (Cane & Tepedino 2001) or the disruption of the loose mutalistic relationships between pollinator and plant and it is based on the premise that habitat destruction reduces pollinator abundance, ultimately leading to lower fruit and seed set (Cunningham 2000a; Cane & Tepedino 2001).

Metastudies show that in 62% of natural populations studied, fruit and seed set are limited by insufficient pollen at some time in their life history (Thomson 2001). However, theory suggests that natural systems should evolve time to a point of equilibrium where pollination and maternal resources are balanced (Thomson 2001). So, when severe pollination deficits are detected, it may indicate that the pollinator service has been disrupted (Thomson 2001). It has commonly been reported that habitat fragmentation is deleterious to populations of Chapter 6 Pollination Success 132

pollinating insects, despite some data (e.g. Rathcke & Jules 1993; Buchmann & Nabhan 1996; Matheson et al. 1996; Murica 1996 and Renner 1996) suggesting that some pollinator species may proliferate in small fragments (Cane 2001). However, several studies have shown that fragmentation in agricultural systems greatly influences pollen deposition, seed production and fecundity, suggesting that rare plants would be more susceptible, ultimately affecting seed germination and seedling recruitment (Karron 1987; Allen-Wardell et al. 1998; Brown & Kephart 1999, Cunningham 2000a; Vaknin et al. 2000).

In general, pollinators are under threat from (1) habitat alteration (agriculture, grazing and habitat fragmentation), (2) introduction of non-endemic pollinators (intentional or accidental) and (3) pesticides (Keams & Inouye 1997). The loss or reduction in pollinator guilds or numbers of pollinators in a system can result in such impacts as smaller pollen loads on individual pollinators (Vaughton 1991) thus reduced pollen competition in styles (Kearns & Inouye 1997) and fruit set or seed set (Paton & Turner 1985), or seed quality, alteration to the mating system, and less vigorous offspring. It is often hard to measure the impacts (e.g. decreased seed set) of decreased pollinator movement, due to habitat fragmentation and destruction, because confounding factors are often involved, such as changes in microclimate and predation levels (Steffan-Dewenter & Tscharntke 1999).

The transfer of inappropriate pollen (e.g. pollen from the wrong species or conspecific pollen) can result in low reproductive success or reduced viability of the resulting progeny (Shore & Barrett 1984; Levin 1984; Paton & Turner 1985; Ramsey 1988; Carthew 1993; Brown & Kephart 1999). The transfer of such pollen is dependent upon the effectiveness of pollinators in terms of the number, timing and duration of visits to individual flowers, their effectiveness and ability to pick up and deposit pollen (Vaughton 1991; Carthew 1993), and their constancy to a particular plant species. These activities may vary temporally, spatially or among individuals of a population. The movement and abundance of pollinators respond to yearly and seasonal changes in floral densities (Vaughton 1990). Pollinators are attracted to individual flowers by odour, colour or even morphological changes that may be associated with opening flowers, or the readiness of the stigmatic surface (Collins & Spice Chapter 6 Pollination Success 133

1986) and typically are rewarded with nectar or pollen (Ferdy et al. 1998). In some pollination systems, such as 'buzz' pollination seen in genera such as Tetratheca and Solanum, pollen is the only reward/attractant for pollinators; it is therefore more difficult to attract a large variety of pollinators. Species such as these spread pollen availability out over a long time and space. To achieve this, pollen needs to be dispersed over time, via a pollen drying process that allows for the release of pollen in many bursts (King & Buchmann 1996). In the case of Tetratheca species, this is achieved via apical pores in the anthers.

The management and conservation of pollinators and their systems is a new and almost untested feature of restoration and recovery of plant species throughout the world (Kearns & Inouye 1997; Kearns et al. 1998; Neal 1998). The re-establishment or enhancing of plant­ pollinator interactions will take conservation past just protection and management (Neal 1998) and into ensuring the viability of populations, in particular the viability of populations of rare plants. These often loose mutalistic relationships can be affected by change to any part of a vulnerable plant-pollinator partnership, causing alteration to pollinator services (Paton 1997).

In conservation management, the use of corridors to mitigate many of the outcomes of fragmentation has become widely accepted, (Bentley & Catterall 1997) (see section 1.1.4) despite the lack of evidence that bushland corridors will generally aid in the local survival of a species. Habitat corridors have found their way in to the management decisions of local governments within N.S.W. including one example of a corridor explicitly to mitigate pollinator loss (e.g. the site at Glenhaven).

Tetratheca glandulosa and Darwinia biffora both occur on ridge tops within the Sydney basin. Consequently, both species exist in vegetation that has been greatly fragmented by urban development. In light of the available literature, it can be hypothesised that small and isolated populations of both T. glandulosa and D. biflora will experience lower reproduction success, as the plant-pollinator relationship may have been disrupted. One Chapter 6 Pollination Success 134

isolated population of D. biflora was connected to nearby populations via a habitat corridor specifically identified as mitigation in the eight part test attached to the development application submitted to council. If the pollination corridor was successful in mitigation, then I predict that it would exhibit similar reproductive success to a range of large populations in national parks.

6.2 Aims

The aims of this part of my study were to investigate the impacts of urban development upon pollination regimes in a number of populations of Tetratheca glandulosa and Darwinia biflora, by examining pollination success. Specifically, I investigated: (1) the pattern of flowering densities (flowers per plant) and the proportion of flowers developing into viable fruit with reference to the size of the population and the presence or absence of a pollinator corridor, and; (2) the frequency and movements of potential pollinators amongst individuals within a population.

6.3 Materials and Methods

6.3.1 Flowering and fruiting and densities

At each site, populations of Tetratheca glandulosa and Darwinia biflora were marked with 1 stainless steel tags at the beginning of the flowering period in July 1999 . The numbers of flowers on individual plants were counted and recorded. The process was repeated in July 2000 for the following flowering season. Plants were revisited in the October of the respective flowering years and the number of fruits developed were counted and recorded. The mean flowering density (flowers per plant) and standard error were calculated for each

1 The site at Cliff Oval was burned in 1999. The adult individuals of D. biflora appeared burned while T glandulosa. was not included in the fire area. Chapter 6 Pollination Success 135

year for flowering densities. Fruit/flower ratios were calculated for each plant. Arithmetic mean and standard error were then calculated for each year for fruit/flower ratio.

The data collected for flower densities were analysed using a two-factor ANOVA using JMP (version 3.0.2, 1989-94) to test for differences between years, sites and their interaction. In this experimental design, years was a random factor and site a fixed factor. Data for fruit/flower ratios were transformed using an arcsine transformation to approximate a normal distribution. When an ANOV A revealed a significant difference, a Tukey Kramer HSD was preformed (JMP, version 3.0.2, 1989-94) to ascertain where the significant difference existed. All statistical testing was performed with a set at 0.05.

6.3.2 Pollinator behaviour and pollen removal

Pollinator visits

Pollinator visits were observed by monitoring individual plants (or a collection of neighbouring individuals) for a period of 15 minutes at a time, from within one metre. The number of insects and length of visit to flowers was recorded. Observations were carried out at different times of the day over the flowering seasons in 1999 and 2000, to maximise the chance of recording pollinators. Data were pooled across a site and the mean number of visits per flower per plant per one hour of observation(± standard errors) was calculated for each site during 1999 and 2000. The data were transformed to match an approximate normal distribution using an arcsine transformation. A two-factor ANOVA was performed to analyse the data (JMP, version 3.0.2, 1989-94), using year as a random factor and sites as a fixed factor. Where a significant difference occurred, a Tukey Kramer HSD test (JMP, version 3.0.2, 1989-94) was used to determine where the difference occurred. Chapter 6 Pollination Success 136

Pollen removal

Pollen removal night versus day was measured in 2000 only. While flowers were still in bud, individuals were enclosed with bags constructed of bridal tulle2 (1 x 1 mm) cut into 16 cm x 16cm square pieces. A drawstring was threaded around the edge, which was pulled tight around a branchlet, excluding pollinators from the enclosed flowers. The flowers completed development inside the bag. Bags were placed over forty flowers at each of Marramarra N.P., Tumbledown Dick Hill, Glenhaven and at the three sites within KCNP, over twenty flowers at ARDEL, but only over two flowers of Tetratheca glandulosa and six flowers of Darwinia biflora at Cliff Oval. The numbers of flowers that were bagged were limited by the population size present at each site. Flowers were allocated to one of two groups (exposed in morning or exposed in the afternoon). Each group was exposed to natural pollinators at the appropriate time and left un-observed for one hour. Pollen removal was detected visually by observing disruption to the oily drop of pollen on the end of the anther for Darwinia biflora (Figure 6.1) and by harvesting anthers for Tetratheca glandulosa and examining them under a dissecting microscope for pollen removal.

Data collected in the morning and afternoon were pooled, because of small sample sizes and the percentage of flowers that had pollen removed was calculated. Data were analysed using a chi square contingency test with a set at 0.05.

2 The flowers of both Tetratheca glandulosa and Darwinia biflora are delicate and would collapse under a heavy material. Chapter 6 Pollination Success 137

pollen drop anthers

style

flower

bracteole

Adapted from Harden 1991

Figure 6.1: Floral structure of Darwinia biflora, showing positions of the oily pollen drop attached to the anthers.

6.4 Results

6.4.1 Flowering and fruit densities

Flower intensities

The mean number of flowers produced for a ramet of Tetratheca glandulosa varied between five different sites in the years 1999 and 2000 (Figure 6.2). Individuals at Marramarra N.P. produced the most flowers (16.29±2.19, 1999; 14.01±1.69, 2000), while individuals at Tumbledown produced the fewest (5.36±0.72, 1999; 7.12±1.11, 2000). There was a significant difference between Chapter 6 Pollination Success 138

20 a b b ab b T Cll i... ~ 15 ~ Q ....=- = Q = i... :s! ~ j> 10 .ca ·--g ==·- i... = ~ = Q. ~ 5 ~

0 ~ J:L. c;J z 0~ > :i u "d 0 "' ~ 0 it:: ~ :=lu !ii E-<~ ~ ::?J Site Figure 6.2: The mean number of flowers per plant for Tetratheca glandulosa at five sites. Flowers were counted in 1999 (II) and 2000 4:J). Sites that are significantly different are indicated with either an a orb (Tukey-Kramer HSD, p< 0.05). Error bars represent standard error.

25

20 ~ ~ Q =-.... = 15 Q = i... :s! ~ j> .c;a ~.e 10 =i... =~ =Q. ~ 5 ~

0 ~ "d c;J ~ ii3 ~ ::i:: ~ c t:: c:l ii !'8 "' :B ., ,Q 0 ~ 0 6 ~ Site

Figure 6.3: The mean number of flowers per plant for Darwinia biflora at four sites. Flowers were counted in two years 1999 (II) and 2000 4:J). Error bars represent standard error. Chapter 6 Pollination Success 139

sites in flower production (Fc3)=0.000, p<0.05) but no significant difference between years (F(i)=0.516, p>0.05), as shown in Table 6.1. The population at Marramarra N.P. showed a significant difference (Tukey-Kramer HSD test, p<0.05) from other sites and the other sites did not differ significantly from each other.

Mean flowering per individual of Darwinia biflora was slightly greater in 2000 compared to 1999 across all four sites (Figure 6.3) but this difference was statistically significant (Table 6.2). The slight differences between sites were also not statistically significant.

Table 6.1: Summary of a two-factor ANOVA for flowering densities at five sites across two years (1999 and 2000) for Tetratheca glandulosa. Data were untransformed(* denotes a significant difference at a=0.05). source of variation SS df MS F p Year 77.2 1 77.2 0.43 0.516 Site 5187.2 3 1729.1 9.48 0.000* Site x Year 552.0 3 184.0 1.01 0.389 Residual 69842.2 383 182.4

Table 6.2: Summary of a two-factor ANOV A for flowering densities at four sites across two years (1999 and 2000) for Darwinia biflora. Data were untransformed. source of variation SS df MS F p Year 98.0 1 98.0 0.49 0.491 Site 879.6 3 293.2 1.42 0.237 Site x Year 24.1 3 21.1 0.01 0.989 Residual 28812.8 285 206.4

Fruit:flower ratio

The proportion of flowers of T. glandulosa setting fruit varied significantly among sites (Figure 6.4). In 1999, it was highest at KCNP (0.559±0.035), but it was highest at Tumbledown in 2000 (0.516±0.041). In both years, fruit set at Marramarra N.P. (0.265±0.035, 1999; 0.387±0.040, 2000) was lower than that at KCNP and Tumbledown. No fruit was produced at either ARDEL or Cliff Oval. There was no Chapter 6 Pollination success 140

"' 0.75 a b b cu '"'~ 0 ....= T 0 =;::: 0.5 ~.5 0 OIJ s:i.=Q ... '"'s:i.~ .... 0.25 ="' =cu :a * * 0 ~ -.; ~ "'z "'z " 0 > "C~ 0 "'t:: ~

Figure 6.4: Mean proportion of flowers setting fruit per plant at five different sites over two years, 1999 (II) and 2000 (D) for Tetratheca glandulosa. Different letters above the bars indicate sites that are significantly different (Tukey-Kramer HSD, P<0.05). ARDEL and Cliff Oval were not included in the analysis. Asterisk indicates that no fruit was produced. Error bars represent standard error.

"'cu 0.75 '"'~ c0 T ....0 =:::! ~.5 0.5 0 OIJ §'.a i=..'"' ;::cu ="' 0.25 :acu= * 0 -.; ] ~ il .... ::c: ~ 0 § .!3 !!:I ~ 5 :E .g 6 0 ~ Site

Figure 6.5: Mean proportion of flowers setting fruit per plant at four different sites over two years, 1999 •)and 2000 (D) for Darwinia bi/fora. Cliff Oval was not included in the analysis. An asterisk indicates that no fruit production occurred. Error bars represent standard error. Chapter 6 Pollination Success 141

significant difference between years (F(i)=0.301, p<0.05,). There was, however, a significant interaction between year and site (F(z)=0.000, p<0.05), indicating that the nature of the differences between years depended on the site.

The proportion of flowers of D. biffora setting fruit did not vary significantly among sites or years (Figure 6.5). At Murra Trail and Bobbin Head, the mean proportion of flowers of D. biffora setting fruit was higher in 1999 (0.592±0.041, Murra Trail; 0.528±0.039, Bobbin Head) than 2000 (0.428±0.050, Murra Trail; 0.443±0.030, Bobbin Head), while at Glenhaven, fruit production did not differ greatly between years. Fruit was not produced at Cliff Oval. There were no significant differences detected (Table 6.4).

Table 6.3: Summary of a two-factor ANOVA for the mean proportion of flowers setting fruit per plant for two years (1999 and 2000) for Tetratheca glandulosa at three sites. Cliff Oval and ARDEL were not included in the analyses. Data were arcsine transformed (* denotes a significant difference at a=0.05).

source of variation SS d.f MS F p Year 0.07 1 0.07 1.07 0.301 Site 1.39 2 0.70 11.40 0.000* Site x year 1.33 2 0.67 10.81 0.000* Residual 20.80 339 0.06

Table 6.4: Summary of a two-factor ANOVA for the mean proportion of flowers setting fruit per plant for two years (1999 and 2000) for Darwinia biflora at three sites. Cliff Oval was not included in the analyses. Data were arcsine transformed.

source of variation SS d.f MS F p Year 0.14 1 0.14 2.75 0.099 Site 0.02 2 0.01 0.17 0.841 Site x year 0.35 2 0.18 3.49 0.318 Residual 13.94 280 0.05 Chapter 6 Pollination Success 142

6.4.2 Insect visits and pollen removal

Visits

Visits to flowers of Tetratheca glandulosa varied between sites (Figure 6.6) over two years of observations. At each site, more visits were recorded in 2000, with Tumbledown recording the greatest number of visits (0.674±0.281). Differences of visitation between sites was significant (F(2)=0.000, p<0.05) while difference between years (1999 and 2000) was not significant (F(l)=0.871, p<0.05) (Table 6.5). Marramarra N.P. was significantly different from other sites (Tukey-Kramer HSD). Cliff Oval and ARDEL were not included in analysis as plants at ARDEL and Cliff Oval received no visits from insects.

Visits by insects to flowers of Darwinia biflora were greater in 2000 than 1999 across all sites (Figure 6.7). The greatest number of visits in 1999 was recorded at Glenhaven (0.292±0.024) and in 2000 at Murra Trail (0. 756±0.099), while flowers at Cliff Oval received no visits in either year. A significant difference was detected between years (Fci)=0.001, p<0.05) and sites (Fcz)=0.024, p<0.05) (Table 6.6). Hence, a significant interaction was detected between different sites and years

Table 6.5: Summary of a two-factor ANOVA for the mean number of visits by insects per flower per plant per hour of observation for Tetratheca glandulosa over two years (1999, 2000). Cliff Oval and ARDEL were not included in the analysis. Data were arcsine transformed(* denotes a significant difference at a=0.05). source of variation SS d.f MS F p Year 0.0006 1 0.0006 0.03 0.871 Site 0.59 2 0.30 13.71 0.000* Site x year 0.02 2 0.01 0.42 0.660 Residual 1.06 49 0.02 Chapter 6 Pollination success 143

a b b ""'QI ~ =0 0.75 T ""'QI Q.""' 1 ~o"'= -~ ..= 0.5 ...... ""'QI 0 Q. i...- ,QQI =cu s c. 0.25 =""'=QI =cu Q. QI ~ 0 * * ;....i OJ ~ ~ ~ ~ > :i u 0 0 "' ~ ~ :D""" ~ !l.:l s ~ 0 ~ E-< l: Site

Figure 6.6: Mean number of visits by insects per flower per plant for an hour of observation for populations of Tetratheca glandulosa at five sites, over two years, 1999.) and 2000 ( 0 ). Letters above the graph indicate sites that were significantly different (Tukey-Kramer HSD, P<0.005). The sites at ARDEL and Cliff Oval was not included in the analysis. An asterisk indicates that no visits from pollinators were recorded. Error bars represent standard error.

a b b

""'QI ~ 0.75 T =0 1 QI Q.""'""' ;::"'= Q .~..c= 0.5 ...... ""'QI 0 Q. ""' .... ,QQI =cu sc. 0.25 =""'=QI =cu Q. QI * ~ 0 '@ ""~ "> ?: :i:: " 0 "'t: .!3 '§"' !l.:l .D .D 0 l:" 0 6 ~ Site

Figure 6.7: Mean number of visits by insects per flower per plant for an hour of observation for populations of Darwinia biffora at four sites, over two years, 1999(•) and 2000 ( o ). Letters above graph indicate sites that were significantly different (Tukey­ Kramer HSD, P<0.005). Cliff Oval was not included in the analysis. An asterisk indicates that no visits from pollinators were recorded. Error bars represent standard errors. Chapter 6 Pollination Success 144

Table 6.6: Summary of a two-factor ANOV A for the mean number of visits by insects per flower per plant per hour of observation for Darwinia biflora over two years (1999, 2000). Cliff Oval was omitted from the analysis. Data were arcsine transformed (* denotes a significant difference at a=0.05).

source of variation SS d.f MS F p Year 0.2 1 0.2 14.9 0.001 * Site 0.1 2 0.05 4.4 0.024* Site x year 0.1 2 0.05 6.2 0.007* Residual 0.2 23 0.01

Pollen removal

Pollen removal from anthers of Tetratheca glandulosa was low across sites (Figure 6.8) reaching a maximum of 30% of flowers with pollen disturbed at Marramarra

N.P. Pollen removal was not independent of site (S24=8.625; p<0.05). No pollen was removed from flowers at either ARDEL or Cliff Oval.

The removal of pollen from the anthers of Darwinia biflora was high at three sites (Figure 6.9) varying between 60% and 70%. No pollen was removed from individuals at Cliff Oval and the extent of pollen removal was not independent of

2 site (s 3=13.77; p<0.05).

6.5 Discussion and Conclusions

All populations of Tetratheca glandulosa produced flowers, with a significantly higher number of flowers per plant a Marramarra N.P. However, plants at ARDEL and Cliff Oval did not produce any fruit and plants at Marramarra N.P. had a significantly lower fruit: flower ratio than other populations. Marramarra N.P. also showed a significantly lower frequency of visits from potential pollinators, while no potential pollinators were detected at ARDEL and Cliff Oval. The finding of no pollen removal at these sites supported the lack of pollinators at ARDEL and Cliff Oval. Chapter 6 Pollination success 145

100

.... 0: ..c: ';;al 75 Ill"" 0.... :r: a 0 Ill ="" 50 ....0 Ill= 111- ....~o c. (40) Ill=-o 0: (40) ~ ..c: 25 Ill =... * * (20) (1) 0 0.. ;..i .,.; z ~ Ul ~ ;i u 0 ..: 1l 0 ~ :E ~ !10 8 0 ! f-" i Site

Figure 6.8: Pollen removal from anthers of Tetratheca glandulosa at five different sites during a one hour period in 2000. Plant sample size is indicated in brackets above data bar. An asterisk indicates that no pollen removal was recorded.

100

.... (40) 0: (40) ..c: 75 .... (40) ~ al Ill .... :r: 0 o a ~~ 50 0 = IllOI)= Ill ....0: c.0 =-ofl 0: 1il ..c: 25 =... * (3)

0 ~ -0 " ';;! ~ ::c:""' "> 0 "' .!3 ~"' ~ .J;;, ::;: .J;;, ~ al a Site

Figure 6.9: Pollen removal from anthers of Darwinia biflora at four different sites during a one hour period in 2000. Plant sample size is indicated in brackets above data bar. An asterisk indicates that no pollen removal was recorded. Chapter 6 Pollination Success 146

There was no significant difference between sites in flowering of Darwinia biflora. Cliff Oval plants did not produce any fruit but, apart from this, there were no significant differences between the other populations in fruit production. There were no potential pollinators recorded at Cliff Oval and pollinator visits at other populations varied between sites and years. There were significant differences in pollen removal among sites with no pollen removal being recorded from Cliff Oval.

Profuse flowering has been associated with attracting pollinators and the potential for pollen limitation of fruit set (Vaughton 1991, Eckert 2000, Tomimatsu & Ohara 2002), indirectly leading to greater reproductive success (Carthew 1993; Harder & Barrett 1995; Snow et al. 1996; Worley et al. 2000). For Tetratheca glandulosa, individuals at Marramarra N.P. produced the most flowers in 1999 and 2000 but set the lowest amount of fruit. This suggests that there was pollen limitation (i.e. limited number of pollinators given the number of flowers available) taking place with an increased number of flowers. Tetratheca glandulosa plants at KCNP and Tumbledown had significantly fewer flowers per an individual than those at Marramarra N.P., but they had a significantly higher proportion of fruit. Hence, other factors (e.g. pollinator visitation) may influence fruit set. Plants of T. glandulosa at ARDEL (small isolated population) and Cliff Oval (population of one individual) produced no fruit in either year, despite profuse flowering in both years.

A number of studies have shown that fruit set is lower in plants occurring in fragmented habitats compared to pristine sites (Kearns & Inouye 1997; Steffan-Dewenter & Tscharntke 1999; Cunningham 2000a; Mataumura & Washitani 2000; Gross 2001; Ghazoul & McLeish 2001 ). Tetratheca glandulos plants at Tumbledown, which is a large, roadside, remnant population, had a high fruit set (proportion of flowers becoming fruits) even compared to Marramarra N.P., which is a large conserved population. Insect pollinators may be able to navigate across close-by patches, thus fragmentation will have no association with pollination success (Feinsinger et al. 1987). I also observed that, at Marramarri N.P., individuals of T. glandulosa growing in more open areas tended to set Chapter 6 Pollination Success 147

higher numbers of fruit compared to individuals at the same site growing in more shaded positions.

Darwinia biflora showed a different pattern of flower and fruit set across sites. There was no significant difference across sites of D. biflora with respect to flowering, generally with more flowers produced in 2000. This may be explained by weather patterns. Variations in flowering patterns due to seasonal conditions are well documented (e.g. Law et al. 2000). The only site where D. biflora did not set fruit was Cliff Oval. This population is small and isolated and was subjected to a hazard reduction bum late in 1999, destroying any immature fruit. The individuals did flower again in 2000, despite having produced little new foliage. These flowers did not produce any fruit, which may have been the result of resource limitation within the plants. It is interesting to note that there were no significant differences in flowering and fruit set for D. biflora between the large conserved sites within national parks (Murra Trail and Bobbin Head) and the site at Glenhaven (isolated site joined to surrounding bushland). Mass flowering has been shown to increase self­ pollination (Ramsey &Vaughton 2000; Eckert 2000; Schmidt-Adam et al. 2000), but for a species that is self-compatible (Chapter 5), fruit set would still occur following selfing.

Throughout the literature, there have been a number of concerns that human alteration to the environment (habitat fragmentation and destruction) has led to the collapse of pollinator services, ultimately affecting plant processes and cascading into wider implications (Cox et al. 1991; Bond 1994; Aizen & Feinsinger 1994; Saville et al. 1997; Kearns et al. 1998; Morgan 1999; Steffan-Dewenter & Tschamtke 1999; Johnson & Steiner 2000). ARDEL and Cliff Oval populations had no visits from pollinators and no reproductive success, which may be the result of the collapse of a pollinator-plant mutalistic relationship. However, in this study, individuals of Tetratheca glandulosa at Tumbledown (large roadside remnant) recorded the greatest mean number of insect visits in both 1999 and 2000 compared to sites in national parks. Therefore perhaps large sites connected to general bushland can sustain pollinator populations in general and can permit inter-population Chapter 6 Pollination Success 148

pollen flow. Pollinator movements are an important factor for fitness and pollination success in sexually reproducing plants (Emms & Arnold 2000).

I predict that, T. glandulosa will eventually become extinct at the ARDEL site unless pollinators are returned. The large number of insect visits to flowers of T. glandulosa at Tumbledown may explain the high level of fruit set there. In comparison, the population of Darwinia biflora at Glenhaven was not different to large conserved populations in National Parks. Both species at Cliff Oval had no insect visits in either year; this may be a direct result of the hazard reduction burn that took place in 1999, eliminating plant species with large, colourful floral displays. Nectar and pollen are common rewards offered by flowering plants in return for pollinator services (Goulson et al. 1998; Harder 1998; Robertson et al. 1999).

The quantification of pollinator effectiveness is a central issue to pollination ecology and poses many difficulties in predicting reproductive success and fitness from pollination data. These problems arise from the identification of pollinators responsible, and the fact that the most commonly observed flower visitors are often given credit for effective pollination (Fishbein & Venable 1996). Apis mellifera was the only recorded visitor to both T. glandulosa and D. biflora. Whereas A. mellifera has the potential to remove pollen and deposit this pollen on the stigma of D. biflora, this is not the case for T. glandulosa. Tetratheca glandulosa is buzz pollinated and it is well documented that A. mellifera is unable to pollinate plants with this pollination system (Moco & Pinherio 1999; Goldblatt et al. 2000; Thorp 2000) and are capable of stealing previously deposited pollen, thus reducing seed set (Vaughton 1996; Gross & Mackay 1998). Therefore, the presence of A. mellifera on flowers of T. glandulosa should be interpreted with caution. There are many native bee species, which are capable of buzz pollination (Gross & Mackay 1998). In South Australia, native bees (Homa/ictus species) have been observed on Tetratheca pilosa. Homa/ictus megastigmus occurs on the eastern seaboard of Australia and has been identified as a possible pollinator for Tetratheca juncea (Bartier et al. 2001). Tetratheca juncea, like T. glandulosa, appears to receive very few visits from native bees. Homa/ictus Chapter 6 Pollination Success 149

megastigmus forms vertical nests in the soil in embankments, road verges, up-turned tree root masses and riverbank cliffs (Bartier et al. 2001). If indeed this species is also a potential pollinator for T. glandulosa, its absence at sites in particular ARDEL, may be attributed to the lack of nesting sites available. This hypothesis would need to be examined further.

Pollen removal was high for both species compared to other species (e.g. Grevillea macleayana - unpublished data collected by F. Beynon). Although plants at Marramarra N.P. received the lowest visitation rate by potential pollinators, they had the largest pollen removal, with almost a similar amount being removed from Tumbledown, which recorded the greatest visitation rate of all sites. Pollen removal from T. glandulosa at ARDEL was non-existent and correlates directly with the lack of potential pollinators recorded visiting flowers. It is also important to note that the population of T. glandulosa at ARDEL had dense foliage from other species growing over the individuals, almost overshading T. glandulosa for light and space. It may have been difficult for potential pollinators to visit the flowers and transfer and deposit pollen with thick foliage shielding T. glandulosa flowers (Lippok et al. 2000). Removal of pollen from D. bi/fora was similar across sites at Murra Trail, Bobbin Head and Glenhaven. This suggests that visits from native pollinators are very rare but highly effective. It can therefore be assumed that pollination success is associated with fragmentation (Bruna & Kress 2002) through influences of pollinator movement in some species, such as T. glandulosa. Chapter 7 Seed Banks 150

CHAPTER 7 THE VIABLE SEED STORE FOR Tetratheca glandulosa AND Darwinia biflora

(THE HIDDEN SCIENCE)

7 .1 Introduction

The capacity for plants to develop seeds in large quantities over both temporal and spatial scales is well established in the literature (Cabin et al. 2000) In many species around the world, seed production results in the development of a dormant seed bank (Thompson & Grime 1979; Wang 1997). A seed bank allows a species to survive the sort of disturbances that may eliminate established plants (Morgan 1995) and it can therefore be critical for the maintenance of plant community diversity (Wisheu & Keddy 1991). Within the Sydney region, 89% of species that have seed banks store their seeds within the soil (e.g. Grevillea caleyi, Kunzea ambigua and Persoonia pinifolia), while the remainder store their seed banks in capsules and woody fruits in the canopy (e.g. Hakea sericea, Petrophile sessilis and Banksia serrata) (Vaughton 1998, Auld et al. 2000). Although both types of seed banks have attracted many studies within the Sydney region, especially in relation to the impact of fire (e.g. Auld 1986; Bradstock & Bedward 1992; Whelan & York 1998; Brown & Whelan 1999), there is still little information on seed bank dynamics for most species.

Production of viable seeds in a population is directly related to population size (Pavlik et al. 1993; Auld 1995): fewer seeds are typically produced in smaller populations. This effect can be exacerbated for species that occur in fragmented habitats (Morgan 1999; Knapp et al. 2001). There are several possible reasons for a decline in reproductive success with declining population size. These include disrupted pollination services (Hendrix & Kyhl 2000) and inbreeding depression (Ferdy et al. 2001) leading to fewer viable seeds per plant in small, isolated populations. There have been few studies on seed banks in relation to the impact of inadequate pollination services (e.g. Jennersten 1988; Aizen & Feinsinger 1994; Burd 1994; Pavlik et al. 1993) and I found none specifically examining the effect of fragmentation due to urban development upon pollination services and seed banks. Chapter 7 Seed Banks 151

The issue of stored seed banks has arisen in a number of cases relating to impacts or proposed development on threatened species, in particular, T. glandulosa (see section 3.4 and 3.5). It is extremely difficult to study seed bank dynamics directly, especially in rare or threatened species. In order to predict and mitigate impacts of developments, models must be constructed from those parts of the life cycle that can be readily measured (Lamont & Van Leeuwen 1998; Wiser et al. 1998; Bevill et al. 1999; Burgman et al. 2001; Cox & Engstrom 2001; Milton 2001). One approach is to examine the seed bank and resulting seedling population by measuring a number of parameters related to seed input and persistence into the seed bank. Annual or seasonal input of seeds (Morgan 1995) can be estimated by measuring flower production, how many flowers produce fruit (Fr/Fl ratio), and calculating the conversion of ovules to viable seed (0/S ratio) (Gross 1981; Lee & Bazzaz 1982; Hirose & Kachi 1986; Weins et al. 1987; Pavlik et al. 1993). Seed longevity (the ability for the seed to persist through time) can be estimated by measuring at losses due to seed decay, pathogen attack on seeds and seed predation following dispersal (Pierce & Cowling 1991; Gunter 1994, Morgan 1995). Pre-dispersal seed predation can also reduce the available seed bank (Tozar 1999) but this is hardly ever quantified (Auld & O'Connell 1991). Finally, an understanding of germination rates, germination cues and the size and dynamics of the already existing seed bank can help in predicting population outcomes (Harper 1977, Tozar 1998) in relation to potential impacts from urban development.

With a good model of the dynamics of seed banks, plant population dynamics can be predicted for various management regimes of disturbances, thus aiding the decision-making process for urban development.

7.2 Aims

This chapter describes the results of studies on the seed banks of Tetratheca glandulosa and Darwinia biflora, and examines how these might be altered through human disturbance of pollination regimes. In particular, the aims of these studies were to determine:

(1) the sizes of the soil seed banks; Chapter 7 Seed Banks 152

(2) the cues for seed germination; (3) the levels of seed dormancy exhibited by seeds comprising the new cohort entering the seed bank; and (4) variation in germination between sites and pollination treatments.

I then combine these results with the data presented in Chapters five, six to model the viable seed bank for 1999 and 2000.

7.3 Materials and Methods

7.3.1 Seeds of Tetratheca glandulosa found in the soil

Seeds that were present in the soil were sampled only for Tetratheca glandulosa as studies had already been conducted for Darwinia biflora (see Auld et al. 1993; Auld 2000).

An estimation of the seed bank for Tetratheca glandulosa (in 2000) contained in the soil was determined by sieving soil and counting the seed found. Ten core samples (25 x 25x lOcm deep) were taken from underneath each of ten randomly selected plants at each of four sites: Marramarra N.P., KCNP, Tumbledown and ARDEL (Figure 7.1). Only two cores (one per plant) were taken from Cliff Oval because there were only two recorded individuals1 at the site. A number of studies have shown that most seeds dispersed in the soil are found in the top lOcm (Auld 1986) and therefore, it was decided to only examine this layer for T. glandulosa seeds. Soil was taken back to the lab and air-dried, sifted initially through a coarse sieve (Smm x Smm) and then through a finer sieve (lmm x lmm), and finally sorted by hand for removal of seed. Seeds were easily distinguished from Tetratheca ericifolia also found at the site which where larger and rounder in size.

Statistical analysis was completed using a one factor ANOVA to test for variation among sites (JMP, Version 3.02, 1989-94). I predicted that smaller, more isolated sites would have

1 Initially only one individual was thought to occur at Cliff Oval but later a second plant was found. Chapter 7 Seed Banks 153

~ 00 0 0 00 Bill co ~00 Oo Ill 0 Will C11m: · . ·. ·.· . Plants '\\.\.

!Ooml Core sample ....______,, ~ 25crn--~ I

5mm sieve

2mm sieve I e~---'Soi/tray

Figure 7.1: Soil cores (ten) were taken from sites and then sieved twice before being sorted by hand in a soil tray for seeds. · Chapter 7 Seed Banks 154 lower seed densities. Site was tested as a fixed factor. All statistical testing was performed at a= 0.05.

7.3.2 Estimation of annual inputs to the seed banks

The inputs to the viable seed stores (Vs) for Tetratheca glandulosa and Darwinia biflora were estimated from the product of five different measures: mean number of flowers per plant (Nf), mean proportion of fruit produced per plant (Nfr), mean number of seeds per fruit (Ns), mean proportion of seeds escaping pre-dispersal seed predation (Ps) and the mean proportion of seeds that were viable (Pv) (Figure 7.2):

Data were obtained from the studies described in Chapter 5 and 6.

The numbers of flowers (Nf) were counted on a sample of individuals at each site during peak flowering times (section 6.4.1) during 1999 and 2000. The proportion of these flowers setting fruit (section 6.4.1) was also counted (Nfr). Mature fruits were harvested early the year following the flowering season (section 5.4.1) and seeds were counted (Ns) (section 5.4.1). Seeds collected were examined for signs (e.g. holes in seed coat or destruction of seed) of pre-dispersal seed predation (section 5.4.1). The proportion not affected by pre­

dispersal seed predation was calculated (P 8). Seed viability was determined by germination/dormancy trials (sections 7.4.3 and 7.4.4) and, for the seeds that failed to germinate, the 'cut test' (Pv)· A 'cut test' allows for the examination of the embryo, which is scored as viable (white and hard) or unviable (grey and mushy) (Kearns and Inoyue 1998). Errors was calculated by using the following formula:

2 2 2 7' ( error ( error error n .L ota 1 error = 1J + 2J + K + ( J ' total 1 total 2 total n Chapter 7 Seed Banks 155

flowers Flowers produced

fruits Percentage of flowers that developed fruits

seeds Percentage of fruits that developed through to maturity (containing seed)

intact damaged Percentage of seed that were not destroyed seeds seeds by pre-dispersal predation

Percentage of intact seed that were viable viable -[J-·, unviable seeds ' seeds

:-..-::::

Figure 7.2: The input to the viable seed store for any plant species can be estimated by using five different parameters (flower development, fruit development, seed development, impact of pre-dispersal seed predation and seed viability). The proportion that each parameter contributes to the next stage of the seed bank development is represented by (D) and each time a section of potential seeds are removed from the resulting seed bank is represented by (c ). Chapter 7 Seed Banks 156

7.3.3 Seed germination

Once the seeds had been obtained from the fruits, germination trials were used to test for viability. The germination cues for Tetratheca glandulosa are unknown but studies on other species in the genus have shown a positive germination response to smoke (e.g. Payne 1998). A number of studies (e.g. Auld et al. 1993; Auld 2000) have shown a positive germination response to heat in Darwinia biflora.

The seeds of T. glandulosa have a small elaiosome-like appendage (Plate 7.1). This appendage is known as a chalazal. It is cream coloured in Tetratheca glandulosa and functions like an elaiosome. Walls of the epidermal cells are rich in fatty substances (Boesewinkel 1999). The function is unknown but it is presumably associated with seed dispersal by ants (Boesewinkel 1999). As there is a possibility that removal of the elaiosome by ants could result in scarification of the seeds, or could otherwise affect germination, I decided to remove the elaiosome on half the seeds using the tip if a scalpel.

An hierarchical set of treatments for the seeds was established (Figure 7.3). Initially, seeds where placed on a bed of moist cotton wool and filter paper in a petri dish (Plate 7.2). Each petri dish was coated with a standard fungicide (Fungarid TM) to reduce fungal attack due to the moist conditions. The petri dish was left in a laboratory for up to eight weeks to allow viable seeds to germinate (in natural light and temperature). Over the eight week period, the petri dishes were checked every day to keep the microenvironment moist and to control fungus. If the moisture content of the petri dishes dropped, water was added. If a fungus attack occurred then the filter paper was replaced, seed gently wiped and re-treated with fungicide. Once germination began, individuals (seeds that had begun to show the root radicle) were counted and removed.

After eight weeks, those seeds that had not germinated where treated with smoke. Treatment took place using a smoke machine, with smoke being created from leaf litter and a small content of fresh leaf material from the representative vegetation from each site. Seeds were then placed back into the petri dishes and germination was monitored, as Chapter 7 Pollination Success 157

germination (within 8 ~ seeds not treated p weeks)

•• •ir

no germination seed viable

•r

seeds treated with smoke germination (within 8 ~ for 2 hours .. weeks)

•r •ir

no germination seed viable

•r

seeds heat treated for 1O ~ germination (within 8 min at 100°c ... weeks)

1r •r

no germination seed viable

...

~ germination (within 8 scarification of seed coat p weeks)

,., ...

no germination seed viable

...

seed not viable

Figure 7.3: This hierarchical methodology was established for treatment of seeds was used to test the effects of several different potential germination cues with a limited number of seeds. Chapter 7 Seed Banks 158

Plate 7.1: Tetratheca glandulosa seeds have a chalazal, which 1s an elaiosome-like appendage indicated by the arrow.

Plate 7.2: Seeds were germinated in petri dishes on cotton wool and filter paper. Chapter 7 Seed Banks 159 already described, for a further eight weeks. Any individuals that showed signs of germination were removed. Seeds that still had not germinated were then treated with heat.

A sample of soil from field sites was collected and air-dried. The soil was placed into ceramic containers and heated in a muffle furnace at lOOQC, seeds were then added and the container returned to the muffle furnace for 10 minutes at lOOQC (Auld et al. 1991). Seeds were removed from the soil, allowed to cool and placed back into the petri dishes and germination was monitored for a further eight weeks. Any individuals that showed signs of germination were removed.

Finally, the seed coats of seeds that still had not germinated were scarified by a small nick in the middle of the coat with a razor blade. Seeds were placed into petri dishes and germination was monitored for eight weeks. Individuals that showed signs of germination were scored as and those seeds, which did not germinate, were scored as unviable.

Data were expressed as percentage of seeds that germinated for each pollination treatment after eight weeks. Data were transformed using the angular transformation ( arcsin .../ proportion germination) and analysed using a two-factor ANOVA (JMP, version 3.0.2, 1989-94). Site and treatment were treated as fixed factors. Any significant differences that occurred where further analysed by a Tukey-Kramer HSD test (JMP, version 3.0.2, 1989- 94) to determine which treatments differed from each other.

7.3.4 Seed dormancy

Seeds were collected following the pollinator exclusion experiments conducted in 2000 (see section 5.3.1). For each site, seeds were pooled across the pollination treatments and then allocated to one of two seed treatments (scarified, unscarified). After treatments, seeds were placed on filter paper over moist cotton wool in a petri dish and left for eight weeks in a laboratory, to allow viable seeds to germinate. Over the eight-week period the petri dishes where checked every day in an attempt to keep them moist and free of fungal attack. If the moisture content of the petri dished dropped, water was added. If a fungus started to appear, Chapter 7 Seed Banks 160 the filter paper was replaced, seed gently wiped and treated with fungicide. Individuals that germinated (i.e. seeds that had begun to show their radicle) were counted and removed.

Data were analysed using a two factor ANOVA (JMP, Version 3.0.2, 1989-94) to test for variation among sites and for effect of scarification. Site and treatment were fixed factors. All statistical testing was performed at an a of 0.05.

7.4 Results

7.4.1 Seeds of Tetratheca glandulosa found in the soil

Very few seeds of Tetratheca glandulosa were found in the soil (Figure 7.4) at any site. The highest numbers of seeds per core were found at Marramarra N.P. and KCNP (0.3 seeds/core), while only two seeds were found at ARDEL across all samples. No seeds were found at Tumbledown or Cliff Oval. There were no significant differences between sites (s24=2.7297,p<0.05). None of the seeds found was viable.

1 - ~r.. 0 (,) r.. ~ Q. 0.75 - "O"' ~ ~ ...."' 0.5 - 0 r.. ~ ,.Q I 1 8= 0.25 - = l l * * =~ 0 - ~ ~ i:: ..... c;; ~ ~ > :i ~ 0 u "O 0 ~ Q) "'t:: :0 ~ !tl 8"' 8 0 "'t:: ~ "' ~ Site Figure 7.4: The mean number of seeds of Tetratheca glandulosa found per core at five different sites in 2000. Asterisks indicate sites were no seeds were found. Error bars represent the standard error. Chapter 7 Seed Banks 161

7.4.2 Estimation of annual inputs to the seed banks

Estimating inputs to the viable seed banks from separate flowering, fruit set and seed viability measures showed that substantial annual inputs to the seed banks are expected for both T. glandulosa and D. biflora (Figure 7.5 and 7.6). For T. glandulosa, most viable seed was calculated for Marramarra N.P. in both years (3.9±0.22, 1999; 3.4±0.27, 2000) and none for the two most disturbed and isolated sites (ARDEL and Cliff Oval). The estimated seed bank for Marramarra N.P. was significantly different from the seed bank of Tumbledown and KCNP (Tukey-Kramer, a=0.05). The estimated seed bank was highest at Murra Trail (5.5±0.85, 1999; 5.0±0.71, 2000) and the least contribution at Glenhaven in 1999 (0.63±0.513). There was no seed bank estimated for Cliff Oval due to no seed being set. There was no significant variation across sites or years.

7.4.3 Seed germination

The percentage of T etratheca glandulosa seed germinating was greater for seeds that resulted from the cross-pollination compared with selfing, across all sites (Figure 7.7). The highest percentages of seed germinating in both the open-pollination treatment (70%) and the cross-pollination treatment (75%) were at Tumbledown. At Marramarra N.P. seeds from the self-pollination treatment showed a low percent germination (25% ). There were no statistically significant differences between sites (p<0.05, F (z)=0.082) or treatments (p<0.05, F (l)=0.460) (Table 7.1).

For Darwinia biflora the greatest percentage of seed germinating was recorded in the cross­ pollination treatment at Glenhaven (75%) (Figure 7.8). Seed germination within the self­ pollination treatment was the highest -at Glenhaven (52.77%), while germination of seeds from the open-pollination treatment was highest at Bobbin Head ( 43. 75 %). Significant differences occurred between sites (p<0.05, F(2)=0.005) as well as between treatment groups (p<0.05, F(2)=0.029) (Table 7.2). A significant interaction between the two factors also Chapter 7 Seed bank 162

8 -

* *

~ ~ ....l t 0 0 I::"' ~ "O ,.!,! ~ '.E 8"' "8 0 I::"' ~ ::E"' Site Figure 7.5: The estimated viable seed production per plant for two different years 1999 (•) and 2000 (o) for Tetratheca glandulosa at five different sites. There was no seed set at either the ARDEL of Cliff Oval (indicated by an asterisk).

8 -

6- T

T ..1

*

"O i::

~ ..<:::~ .!3 ~ ~ Site . G Figure 7.6: The estimated viable seed production per plant for two different years 1999(•) and 2000 ( o) for Darwinia biflora at four different sites. The plants at Cliff Oval did not set any seed (indicated by an asterisk) and was burned in 1999. Chapter 7 Pollination Success 163

100 -

* • *

..~ : '.' e*A ....

i:i..: c ....:i t;; ~ ~ ril > :i 0 u "O 0 ~ Q) "'t:: ::0 ~ ~ 8 0 8"' :s "'t:: F-< 2:"' Site Figure 7.7: The percentage of seed germinating of Tetratheca glandulosa at five sites where populations are different sizes. The site at Cliff Oval consisted of one individual. Seeds were collected following pollination experiments (•-open pollination, o-self pollination, and ~ - cross pollination). Asterisks indicate that no seed was produced in the self-pollination treatment. Circles indicate no seed was produced in the open-pollination treatment. Triangles indicate that no seed was produced in the cross-pollination treatment.

100 -

Cl> Cl> Cl> ,...- -c =:::> -"'.. ~ (f)

"O c t;; ·ca Q) -.... Q) "' ~ E-- ::c: ..c: c c c "'t:: Q) !B :s :8 6 0 2: cc0 Site Figure 7.8: The percentage of seed germinating of Darwinia biflora at four sites where populations are different sizes. Seeds were collected following pollination experiments(•• open pollination, o-self pollination, and ~ - cross pollination). No seeds were produced at Cliff Oval. Significant differences were found between sites, treatments and site x treatment interaction. Chapter 7 Seed bank 164

1 1 0.75 - .~....= ....= ai... 0.5 - Q.I tlJl

=Q.I ~ 0.25 -

* *

~ c:: ....) ~ ~ :i 0 '"O "" ~

1 -

T 0.75 -

Q ':C= ....= ai... 0.5 - Q.I tlJl

=Q.I ~ 0.25 -

* 0 - '"O c:: ';;j -·01....

occurred (p<0.05, Fc4)<0.001). There was a significant difference between the percentages of seed germinating from Glenhaven compared to seed germinating at Murra Trail (Tukey­ Kramer HSD).

Table 7.1: Summary of a two-factor ANOVA for seed germination three sites across two treatments (open and cross pollination) for Tetratheca glandulosa. Cliff Oval, ARD EL and the self pollination treatment were not included in the analysis. Data were transformed using arcsine. Significance was determined at a=0.05.

Source of variation SS d.f MS F p Site 0.63 2 0.31 11.18 0.082 Treatment 0.03 1 0.03 0.61 0.460 Site x Treatment 0.06 2 0.03 0.14 0.872 Residual 26.05 127 0.21

Table 7.2: Summary of a two-factor ANOVA for seed germination at three sites across three treatments (open, self, cross pollination) for Darwinia bifl.ora. Cliff Oval was not included in the analysis. Data were transformed using arcsine and asterisks show a significant difference at a=0.05.

Source of variation SS d.f MS F p Site 1.67 2 0.84 5.384 0.005* Treatment 1.11 2 0.55 3.557 0.029* Site x Treatment 6.22 4 1.56 10.015 0.000* Residual 75.77 488 0.16

7.4.4 Seed dormancy

Scarification increased the germination of Tetratheca glandulosa seed, across all sites (Figure 7.9). The highest germinating proportion following scarification occurred at Tumbledown (0.8±0.09), while the highest germination rate for the unscarified seed coat group was recorded at Marramarra N.P. (0.1±0.06). There was a significant difference in mean germination between treatments (p<0.05, Fci)<0.001) (Table 7.3), but not among sites. Chapter 7 Seed Banks 166

Scarification also increased germination in Darwinia biflora (Figure 7 .10). The highest germination was recorded from seeds collected at Bobbin Head (0.8±0.07) following scarification and for the treatment group of unscarified seeds (0.5±0.09). A significant difference in mean germination occurred between treatments (p<0.05 , F(l)<0.001) (Table 7.4), but not among sites.

Table 7.3 Two-factor ANOVA comparing percentage germination (after arcsin transformation) at four sites for two treatments (scarified and unscarified) for Tetratheca glandulosa. Cliff Oval was not included in the analysis. Asterisk indicates a significant difference at a=0.05.

Source of variation SS df MS F p Site 0.34 2 0.17 1.14 0.324 Scarification 11.86 1 11.87 79.18 0.000* Site x Scarification 0.43 2 0.21 1.42 0.246 Residual 18.44 123 0.15

Table 7.4 Two-factor ANOVA comparing percentage germination at three sites for two treatments (scarified and unscarified) for Darwinia biflora. Cliff oval was not included in the analysis. Data were arcsine transformed. Asterisk indicates a significant difference at a=0.05.

Source of variation SS df MS F p Site 0.53 2 0.27 1.25 0.288 Scarification 6.05 1 6.05 28.43 0.000* Site x Scarification 0.13 2 0.07 0.31 0.731 Residual 37.03 174 0.21

7 .4 Discussion and Conclusions

This study confirmed the presence of a soil seed bank for both T. glandulosa and D. biflora and for other species in these genera (Payne 1998; Auld et al. 2000). The seed bank was not sampled directly for D. biflora in this study (as it has already been sampled by Auld et al 2000). The absence of a seed bank for T. glandulosa at sampled sites may be explained through seed dispersal, seed predation and seed decay. Dispersal of seeds by ants is well documented throughout Australia and has been viewed as an important factor in seed bank Chapter 7 Seed Banks 167 dynamics (Harrington & Driver 1995; Yates et al. 1995; Andersen & Morrison 1998; Auld & Denham 1999).

Many plant species occurring within the fire-prone vegetation of Australia show seed dormancy (Auld 1986; Auld & O'Connell 1991; Edwards & Whelan 1995; Ellery & Chapman 2000; Kenny 2000; Morris 2000; Read et al. 2000), producing persistent seed banks. These are crucial for maintaining populations, especially for threatened species that are obligate seeders (Aparicio & Guisande 1997) (i.e. adult plants die as a result of disturbance and population recovery is solely from stored seeds). Dormancy restricts germination to favourable establishment conditions, such as following fire (Letnic et al. 2000). One mechanism of seed dormancy is "hard" seeds that require scarification of the seed coat before germination can occur (Auld 1986; Gunter, 1994; Aparicio & Guisande 1997), or some other disturbance to the seed coat (e.g. heat or smoke) (Gilmour et al. 2000; Kenny 2000, Letnic et al. 2000). Both T. glandulosa and D. biflora showed higher levels of germination following scarification of the seed coat. However, Auld et al. (2000) found clear evidence, in a single cohort of seeds of D. biflora, that dormancy was short-lived. After two years, 7-22% of seeds were still dormant. This is unusual, because obligate seeders are entirely dependent on seedling recruitment for survival (e.g. Acacia suaveolens, Banksia ericifolia). However, some other species within the Sydney region have shown similar short-lived seed bank dynamics (e.g. Persoonia pinifolia, Angophora hispida). Tetratheca glandulosa may have a similar seed bank dynamic, with high seed decay and short-lived dormancy. However, T. glandulosa can resprout and is thought to be clonal which may compensate for high seed decay and low seed production.

Individual plants of T. glandulosa contributed very few seeds to the viable seed bank over two years, even in the most productive National Park sites (Marramarra N.P. and KCNP), suggesting that seed production is naturally low. If seed longevity in the soil is limited, as suggested for D. biflora, effective conservation will require maintenance of adult populations, good pollination and seed set and appropriate conditions for recruitment. These two species may therefore be particularly sensitive to the impact of fragmentation from urban development. Chapter 7 Seed Banks 168

There was no viable seed set at ARDEL or Cliff Oval for T. glandulosa or at Cliff Oval for D. bifl.ora. The ARDEL site was recently fragmented and it is difficult to determine if the action of fragmentation has caused the lack of seed set or if this was due to conditions unique to the site (e.g. water and mineral limitation, predation of reproductive floral parts, lack of pollinators) (Pavlik et al. 1993). The contribution of seeds by individuals to the seed bank for D. bifl.ora was higher than for T. glandulosa but not compared to other Darwinia species (e.g. Darwinia diminuta B. Briggs, Darwinia glaucophylla B. Briggs and Darwinia procera B. Briggs all produce larger seed banks than D. biflora (Auld et al. 1993)). Therefore neither T. glandulosa nor D. bifl.ora can depend on a long-lived seed bank after adult plants have gone, such as seen in many Acacia species (Auld 1986; Auld & O'Connell 1991), especially when seed contribution to the seed bank can vary between years (e.g. D. bifl.ora at Glenhaven).

Very few studies have looked for a correlation between habitat fragmentation, preferred mating system of a particular species and seed germination (Fischer & Matthies 1997; Nason & Hamrick 1987). When a population is fragmented it is generally reduced in size. This reduction in population size may impact upon the fitness of any offspring and ultimately decreasing seed germination accordingly (Menges 1990; Morgan 1999).

Populations of T. glandulosa at Tumbledown and ARDEL showed a high percentage seed germination from the cross-pollination treatment, however this was not significantly different from the populations in national parks. Even through T. glandulosa did not naturally set any seed at ARDEL, a similar pattern was seen across the other sites in respect to the cross-pollination treatment. These patterns seen in seed germination of T. glandulosa are not the predicted patterns seen across large conserved population and small isolated populations in other species (Buza et al. 2000; Cunningham 2000a, Eisto et al. 2000).

The population of D. biffora at Glenhaven showed a higher percentage of seed germination in both the self-pollination and cross-pollination treatments compared to the other populations examined. It should be noted that seed germination within the open treatment at Chapter 7 Seed Banks 169

Glenhaven was substantially lower than the same treatment at other sites. The population at Glenhaven is connected to surrounding vegetation by a bushland corridor that was designed to enhance pollinator movement. Corridors have attracted polarised views in the literature and they are commonly viewed as being species-specific (Harris & Scheck 1991; Haddad 1999;Wolff et al. 1997). This study shows that apparently the corridor is not working to enhance pollinator movement (at least using seed germination as an indicator). Darwinia biflora appears to also have a mixed mating system and when cross pollen is not available, self-pollination seems to be adequate to sustain the years seed crop. A number of studies (e.g. Denham & Whelan 2000) have shown that seed germination is higher in treatment groups, which represent the preferred mating system. However, the interpretation of the data should be treated with caution, as the percentage seed set did not differ greatly between treatments interpretation is difficult. Chapter 8 General Discussion 170

CHAPTER 8 GENERAL DISCUSSION

8.1 Introduction

With continuing urban sprawl (in particular around the outer suburbs of Sydney) and the introduction of the Threatened Species Conservation Act 1995 (NSW), there is increasing focus on controlling biodiversity losses. Large impacts on biodiversity have already occurred in Australia mainly through land clearing in the agricultural landscape (ANZECC 2001 ). Increased habitat fragmentation in the urban environment will inevitably lead to an increase in the numbers of threatened species and a reduction in biodiversity.

The TSCA in NSW was established to help halt the ongoing erosion of biodiversity, specifically by offering protection to threatened species, populations and ecological communities within the planning system. The introduction of the TSCA was based on a belief that legislation may and will deliver conservation. However, ecological knowledge is only one input into the decision-making process, ecological knowledge is incomplete and dynamic, and development can still go ahead regardless. This highlights the important question of whether this legislation actually can produce effective conservation outcomes.

The TSCA is embedded in the planning process via links with Part 4 of the EPAA (see Chapter 2), thus providing a potentially powerful mechanism for controlling biodiversity losses within the decision-making processes. The TSCA is made up of four principal components: listing threatened species, eight part tests, species impact statements, and recovery planning. The listing of a species as 'endangered' or 'vulnerable' triggers the eight part test, and potentially, an SIS, when there is a development proposal. Listing also triggers the completion of a species recovery plan. The listing process is, then, the backbone of the TSCA. While species can be listed as endangered, vulnerable or extinct, populations and ecological communities can only be listed as endangered. Critical habitat and key threatening processes can also be listed. This thesis concentrates on the interaction between the science of ecology and legislation and policy at the level of species conservation. Chapter 8 General Discussion 171

The eight part test (TSCA s.94, EP AA s. SA) and the SIS (TSCA s. 110), as discussed in Chapter 2, form a package that is triggered once a species is listed and a development proposal is made. This package demands consideration of the long-term viability of a species (i.e. its evolutionary potential) during development planning. Recovery planning (TSCA Part 4) is also essential for the conservation of threatened species. Recovery plans are put in place with the aim of removing a species from its listing on the schedules of the TSCA by ensuring its long-term viability and evolutionary potential in the wild.

Adequate assessment of each of the individual components of the TSCAJEP AA requires a degree of ecological knowledge (Chapter 2). Behind each component, there are several assumptions about the status of scientific knowledge, both about individual species and about ecological processes. Problems can arise, for example, from trying to determine if a population is viable for an eight part test where there is limited ecological understanding or data. What happens if there are limits in knowledge, or it is altogether absent? There is ample evidence that knowledge is then assumed (Chapter 3), or that the lack of specific knowledge is defacto, an indication that there is not likely to be a significant impact of a development.

8.2 Summary of thesis

In this thesis, I have addressed the issue of how science and legislation are interacting with policy and management to produce decisions, using two case study species: Tetratheca glandulosa and Darwinia biflora. These species are both listed as 'vulnerable' on the TSCA. I used a number of approaches and applied them to a set of specific questions. Brief summaries of the outcomes for each question are presented in Table 8.1. The outcomes of the legislative analysis and ecological outcomes are summarised in the following paragraphs. Chapter 8 General Discussion 172

Table 8.1: Summary of the specific questions in the study and a brief summation of the answers and/or outcomes obtained.

Specific Question Approach Outcome/answer Can the TSCA conserve biodiversity Examination of the words used in the TSCA currently provides inadequate and how has the wording of the act TSCA and correlating these with biodiversity conservation. Suggested influenced decisions (from councils, development applications and the changes include: developers, NPWS and the Land and outcomes of the decisions made by 1. listing uncommon species and Environment Court) on development consent authorities. species with an unknown applications for sites in which there distribution; are populations of threatened flora 2. peer review system for scientific species? (Chapters 2 and 3) credibility of SIS; and, 3. a more prominent role in legislation for recovery plans.

What is the role and understanding of An examination of the TSCA, other 1. Local councils rely on ecological scientific knowledge and processes in policies and decisions dealing with knowledge in 8 part tests and SIS. policy development for species species conservation in relation to: 2. Incorporating precautionary conservation? (Chapters 3 and 8) • site specificity, the importance of principle and scientific uncertainty remnant vegetation and reserves; and, into decision-making and • dealing with lack of scientific establishing pathways for decision­ knowledge (precautionary principle). making to occur.

What is the pollination biology of T. An examination of: 1. T. glandulosa has a partially self­ glandulosa and D. biflora? (Chapter 1. the breeding system of the compatible breeding system, while D. 5) two species, using biflora is self-compatible. flower:fruit ratios, 2. Seed abortion was high for both 2. the effect of pollen species and pre-dispersal seed parentage on seed quality predation has the potential to have an 3. identification of pollinators impact. 4. clonal spread of T. 3. No pollinators positively glandulosa identified. 4. T. glandulosa is not 100% clonal

What is the impact of urban The examination of natural flowering 1. Flower production differed development (through habitat and fruiting densities, pollinator between sites for T. glandulosa but fragmentation) upon pollination movement and seed germination in not for D. biflora. regimes? (Chapter 6) relation to population size and 2. Flower:fruit ratio differed between connectivity. sites for T. glandulosa but there were no significant differences for D. biflora. 3. Pollinator visits were significantly different between sites for both species. 4. Pollen removal was low and dependant on site for T. glandulosa c.f. D. biflora.

What is the seed bank profile An examination of the level of seed 1.Very few T. glandulosa seeds (recruitment potential and success) dormancy for each study species and found in the soil. for the two study species and how is the size of the soil seed bank, which 2. Substantial seed inputs expected this altered through human is then contrasted, with models of the for both species. disturbances? (Chapter 7) potential seed bank each year. 3. Seed germination highest in cross­ pollination treatment for both species. 4. Scarification increased germination in both species. Chapter 8 General Discussion 173

Local councils rely on the ecological knowledge presented in an eight part test or SIS to aid their decisions in assessing development applications (see Section 2.3.2). It is widely accepted that it is impossible to know everything about a species at present. However, it is difficult for local councils to assess whether an action will impact upon a threatened species if very little is known about its life cycle (an important feature in an eight part test and SIS). This legislation gives no guidance as to the approach that should be taken by a council when there is no ecological knowledge available at all. Local councils at this point can take a number of approaches. The local council can reject the development application by invoking the precautionary principle, or demand further investigations focusing upon factors that keep a species, population or community evolutionarily viable. Local Councils have no set pattern of how they deal with this situation and the response can vary between Local Council areas and cases.

A number of ecological studies were carried out using Tetratheca glandulosa and Darwinia biflora. The first suite of experiments was designed to obtain some details about breeding systems of the two chosen species, for later assessment of the implications for conservation and the effectiveness of the TSCA (Chapter 5). Tetratheca glandulosa and D. biflora both have a breeding system with a strong preference for outcrossing. This was determined by measuring seed set following bagging experiments and further supported by seed weight and length. This means that a loss of potential native pollinators, particularly in the case of T. glandulosa, would have an immediate impact on populations by reducing seed set.

The second set of ecological studies investigated the impacts of urban development upon pollination regimes of T. glandulosa and D. biflora (Chapter 6). Specifically, the pattern of flowering densities and the frequency and movement of potential pollinators were examined. While flowers were produced at ARDEL and Cliff Oval for T. glandulosa, no fruit was produced or pollinator visits recorded. These findings were supported by a complete lack of pollen removal from flowers in small, disturbed, isolated sites while pollen removal and seed set was greater in sites that were larger reserves. This confirms that urban development will affect the pollination regime of T. glandulosa. Chapter 8 General Discussion 174

Finally, the seed bank was studied to determine recruitment success and how this might be altered through human disturbance (Chapter 7). In particular, the size of the soil seed bank was estimated, the cue of seed germination determined, and germination and level of dormancy quantified. A soil seed bank was confirmed for both T. glandulosa and D. biffora. However, not many seeds were found to exist at sites of T. glandulosa. Both species showed a low dormancy in their seeds, with germination triggered by heat and/or smoke. ARDEL and Cliff Oval showed no viable seed set or soil seed bank. The results suggest that neither species would be resilient in the face of habitat disturbance nor fragmentation as seed longevity may be limited in the soil.

Together, these studies revealed significant deficiencies in the state of ecological knowledge about these species. Given that the legislation has been applied in planning for development in areas where these species occur there is a critical question of how scientific uncertainty is currently dealt with in the decision-making context.

8.3 What scientific uncertainty exists and how is it dealt with?

Two case studies were examined in detail for Tetratheca glandulosa and Darwinia biffora. These illustrate the existence of scientific uncertainty and two different ways in which councils in decision-making processes treat it.

8.3.1 Tetratheca glandulosa (Aquatic Drive, Allambie Heights)

The housing subdivision at Aquatic Drive, Allambie Heights is owned by ARDEL Limited (see section 3.2), and there were two central issues in the decision-making process; (1) did T. glandulosa exist on the site and (2) was the outcome of the SIS appropriate? Initially it was disputed whether T. glandulosa was growing on the site. Later it was thought that individuals had been mis-identified as Tetratheca ericifolia. The species is cryptic and can only be properly identified when in flower. It has since been confirmed that T. glandulosa is indeed occurring at the site in a population consisting of approximately 40 individuals. Chapter 8 General Discussion 175

Species Impact Statements, which are applied to determine the significance of an activity or development on a species, as an aid to decision-making (Mamoumey 2000), have attracted criticism from councils, developers and ecologists (see Section 2.3.3). An SIS requires detailed ecological knowledge, not only about the threatened species but also how this species will respond to the proposed activity. In the case of T. glandulosa at the ARDEL site, an SIS was completed and lodged as part of the development application despite the developers' belief that T. glandulosa did not occur on the site (see Section 3.2.3 for an explanation of why). Scientific data are rarely collected for a specific SIS and information is often extrapolated from other sites, congeneric species, different genera and even different families (Table 3.2). I reassessed the outcomes of the original SIS after gathering data on breeding systems, clonality and pollination ecology of T. glandulosa (Table 8.2).

This reassessment of the SIS for the housing development at Aquatic Drive allows an increased understanding (for local council) of how the development is likely to affect the population of T. glandulosa present. Even though the population of T. glandulosa is not setting fruit (Section 6.4.1) or being visited by potential pollinators (Section 6.4.2), seed was found in the soil. It is predicted that the seed bank of T. glandulosa is short lived (Section 7.4.1), and therefore, as seed was found, fruit set and pollinator visitation must have been occurring quite recently in time. In my view, had this knowledge been available prior to consideration of the development application, it may have been more appropriate for a different design of the boundaries of the development to retain the whole population and more surrounding bushland.

8.3.2 Darwinia biflora (Green Road, Glenhaven)

Green Road is the site of a housing development in Baulkham Hills Shire Council. An eight part test for Darwinia biflora was included with the development application. The eight part test was conducted in light of an amelioration process that had been agreed between the developer and the local council (for a summary of the amelioration process see Section Table 8.2: A revised summary of the SIS and outcomes for the population of Tetratheca glandulosa at Aquatic Drive , Allambie Height s, after the collection of data in the current study (see chapters 5, 6 and 7). 9 -§ Outcome in original SIS submitted with Species Impact Statement (TSCA s. 110) Evaluation following the gathering of scientific data ...,~ development application Oo (1) a full description of activity proposed in the The action is summarised in section 3.2.2 of this The action is summarized adequately, however, the development later C) development application, including nature, exten~ chapter (summarised also in HASSELL PLAN 9208- changed after controls are put into place by local council. A final location, timing and layout of the proposed 10-4.6.96 prepared for Ardel Ltd.). summary is stated in section 3.2.2 of this thesis. ~ development or activity ~ \:::) (2a) a general description of the threatened species The only threatened plant species reported from the No adequate description is given. c;· (") known or likely to be present in the area that is the land is Tetratheca glandulosa, reported by Clements ;::: 1995. subject of the action and in any area that is likely to ~ be affected by the action ~·

(b) an assessment of which threatened species As per (2a) above. This is not adequate, as no assessment of the threatened species is known or likely to be present in the area are likely to included. An assessment should include site-specific data such as; the be affected by the action population of T. glandulosa is flowering however, no seed was set and flowers were not visited by any potential pollinators or had pollen removed. Seed was found in the soil suggesting that seed must have been set in recent flowering episodes. Thus isolation is likely to affect the viability of the population.

( c) for each species likely to be affected, detail of its Species is well represented in Dharug NP, The species is restricted to ridge tops (shale on sandstone) and is more local, regional and State-wide conservation status, Marramarra NP, Muogamarra NR, Brisbane Waters than likely threatened by clearing of native vegetation, making

the key threatening processes generally affecting i~ NP, Garriga) NP, Lane Cove River NP and Yengo populations that occur in reserves very important. However, "well its habitat requirements and any recovery plan or NP. These conservation areas cover the complete represented" is not defined and it is unknown what conditions the threat abatement plan applying to it range of the species. Herbarium samples show a populations are in. Also, each population may represent part of the broader range. genetic di versity of the species.

(d) an estimate of the local and regional abundance Population sizes are unknown but thought to be The regional and local abundances are not described. Surveying has of those species small. It is difficult to estimate as the species is shown some large populations present in National Parks and on cryptic when not flowering roadside verges (Tumbledown Dick Hill). It is also thought that the species does have the potential to be clonal, which could further lower the effective population size.

( e) a general description of the threatened species As per (a) above As per (a) above. known or likely to be present in the area that is the subject of the action and in any area that is likely to be affected by the action

(f) a description of type, location, size and condition T. glandulosa was not located on the site (by CSIRO) A population of 40 individuals occurs along the mid-western boundary of the habitat and details of the distribution and and its presence is doubtful. It is possible that the of the site within the Duffy"s Forest ecological community. It has been condition of similar habitats in the region species was mis-identified as T. ericifolia. This thought the population is non-critical but the role the site plays with population is non-critical given that the species is other sites is unknown in relation to gene flow. The site is the most conserved in 8 reserves. southern population for the species.

(g) a full assessment of the likely effect of the action There is not likely to be any impact as the possible The impact of the action on the population has been assessed without on those species, including, if possible, the existence of the species at the site is doubtful, the use of scientific knowledge. The action may impact upon the quantitative effect of local populations in the therefore the habitat is of no importance to the removal of potential pollinator nesting sites and the isolation and cumulative effect in the region species survival. destruction of potential habitat. The cumulative effect in the region is unknown as the complete biology of the species is unknown.

(h) a description of any feasible alternatives to the Given the absence of a known population on the site, As the SIS was completed with the assumption that the population was action in the manner justifying the carrying out of feasible alternatives to the action cannot be not present, it is therefore impossible to assess any possible alternatives the action in the manner proposed, having regard to developed. to the proposed action. As soon ass. (2a) confirms presence of the biophysical, economic, social and ESD principles species, this section becomes inadequate.

(i) a full description and justification of the Given the absence of a known population on the site, As the SIS was completed with the assumption that the population was measures proposed to mitigate any adverse effect of mitigation actions cannot be developed. not present, it is therefore impossible to assess any possible mitigation. the action on the threatened species As soon as s. (2a) confirms presence of the species, this section becomes inadequate.

(j) a list of any approvals that must be obtained No other approvals are known to be necessary No other approvals are known to be necessary. under any other Act or law before the action may be lawfully carried ou~including details of conditions of any existing approvals that are relevant to the species or populations

...... -...) °' Chapter 8 General Discussion 177

3.3.2). Eight part tests are used to determine if there is likely to be an impact on a listed threatened species, thus determining if an SIS is required for submission with a development application (Section 2.3.2).

I re-evaluated the original eight part test for Green Road, Glenhaven after gathering data on flowering, pollination and fruit set (Table 8.3). I conclude that the additional data collected in this study would not have dramatically altered the outcome of the eight part test. In this study, I did not assess whether the bushland corridor, proposed as amelioration, was used by pollinators to access otherwise isolated D. biffora plants and thence move pollen between sites. Corridors have received a lot of attention in the scientific literature (Section 1.1.3) and their effectiveness has been found to be species-specific. I consider that the incorporation of a bushland corridor in the proposed development, as amelioration in response to the likelihood that the known habitat of the species was likely to become isolated (EPAA s SA (d)), was premature. While such a corridor might appear, to council officers and the developers, to facilitate movement of pollinators, there is no evidence to support this claim.

8.4 How could the legislation be improved?

Conservation may be approached in one of two ways, reactive or proactive. Reactive conservation is a response to an individual event or set of events (e.g. a development proposal that would result in habitat removal and isolation of a population, thus providing the "ambulance at the bottom of the cliff' (Craig 1997)). Proactive conservation, on the other hand, refers to attempts to conserve a species before numbers of individuals or populations decline drastically.

There are two different senses in which we can understand proactive/reactive conservation. In one sense, all the provisions in the Act are reactive because they cannot be used until a species is listed as threatened (and therefore in need of an ambulance). On another Table 8.3: A revised summary of the eight-part test and outcomes for the population of Darwinia bifiora, at Green Road Glenhaven, after scientific data has been collected. Q {5 ~ Eight part test Response to the original eight part test Development controls as Evaluation following the gathering of scientific data .... stated in the original eight Oo part test c;:i (a) in the case of a threatened species, A viable population can be maintained by placing I. drainage construction The population appears to be viable (however the ~ whether the life cycle of the species is likely appropriate conditions on the development. 2. revegetation with D. biflora genetics is unknown and inbreeding depression may be "' to be disrupted such that a viable local 3. fencing around population occurring) and the population is thought to be able to be ~ population the species is likely to placed 4. fire management maintained through appropriate development controls. b of t:;· at risk of extinction <"\ ;:: (b) in the case of an endangered population, n/a n/a The eight part test is not concerned with an endangered ~ whether the life cycle of the species that population. ~· constitutes the endangered population is likely to be disrupted such that the viability of the population is likely to be significantly compromised

( c) in relation to the regional distribution of Approximately 1% of D. biflora would be lost. none Even though only 1% of D. biflora would be lost, the the habitat of a threatened species, Suitable habitat occurs sporadically in the local accumulated regional impacts of development on the population or ecological community, government area, main I y on private land ( eg. species are not discussed. whether a significant area of known habitat 5000 individuals growing on 2 ha, lkm S.E. of is to be modified or removed the site). Through interpretation of vegetation and soil landscape maps, the site in relation to the regional distribution of T. glandulosa, the area is not significant.

( d) whether an area of known habitat is No, a bushland corridor was designed allow Corridor of habitat retained The use of the corridor of pollinators was not examined. likely to become isolated from currently access for insects for cross pollination. However, the use of a bushland corridor allowed the interconnecting or proximate areas of habitat of population to remain connected in a larger habitat for a threatened species, population mosaic of habitat. or ecological community

(e) whether critical habitat will be affected n/a n/a There was no critical habitat declared before the completion of the eight part test.

(f) whether a threatened species, population Assumed to be adequately conserved in reserve none The species is thought to be well conserved with large or ecological community, or their habitats, systems. This assumption is based upon the populations occurring in NPs (e.g. KCNP, Lane Cove NP are adequately represented in conservation species' ROTAP codes and the size of and Marramarra NP). reserves (or other similar protected areas) populations existing in reserves. in the region

(g) whether the action proposed is of a class Subdivision results in habitat removal which is none At the time of development it was recognized that habitat of action recognised as a threatening generally a threatening process but, it was not clearing was a threat but was not listed on the schedules process listed in schedule 3 of the TSCA. of the TSCA.

(h) whether any threatened species or S.W. limit of distribution in and around site. none It is agreed that the SW limit of D. biflora occurs in and ecological community at the limit of its around the site. known distribution ,_. -...) 00 Chapter 8 General Discussion 179

dimension, however, recovery planning is proactive insofar as the government takes the initiative, rather than waiting for a developer to put in an application (which is reactive in this sense). The integration of these two different approaches to conservation gives potential strength to the TSCA.

I argue that, within the TSCNEP AA, the proactive elements are the key to delivering long­ term conservation outcomes. These need to deal with scientific uncertainty, adaptive management and the precautionary principle. I consider that the better use of recovery planning in the decision-making process is fundamental to improving the legislation. In the following section, I explain the recovery planning process, how the precautionary principle can be more effectively incorporated in to decision-making, and how adaptive management (in response to activities) and improved pathways for decision-making can take place in response to threatened species.

8.4.1 Recovery Planning

Due to the large number of recovery plans required (over 700) and a very long procedure for preparation and implementation (which may take up to a number of years), some priority needs to be determined (TSCA s. 58) (Wilson 1997). Nationally endangered and threatened species are given first priority (TSCA s. 58(2)). The Director-General, in consultation with the Scientific Committee, prioritises the other endangered and threatened species listed using the following criteria (TSCA s. 58(1)): (1) likelihood of extinction (TSCA s. 58(3)(a))(e.g. Bengtsson 1989; Dennis et al. 1991; Mann 1991; Gill & Bradstock 1995; Keith 1996); (2) likelihood of recovery (TSCA s. 58(3)(b)); (3) whether it maybe 'keystone species' (TSCA s. 58(3)(c)) (e.g. Fowler et al. 1989; Mills et al. 1993); and, (4) whether it maybe 'indicator species' (TSCA s. 58(3)(d) (e.g. Griffith et al. 1995; Vasseur et al. 1998).

Local council participation in the recovery planning process is voluntary but active involvement often occurs. Once approved, recovery plans have to be taken into account by Chapter 8 General Discussion 180

a Council but do not automatically override LEPs (TSCA s. 69(1)-(3)). The TSCA contains a provision (TSCA s. 69(2)) that allows for the exercise of discretion by a local council during development consent and, to this extent does not require it to comply with the plan. A recovery plan can apply to any type of land ownership, but private landholders do not have to consent to a recovery plan or comply with any actions described within the plan (Kelly 1996). Recovery plans are enforceable through development conditions (Farrier 2002).

Legislating for compulsory recovery plans for all listed threatened species has been viewed as a powerful tool for conservation. Recovery plans are likely to be more objective, as they are not dictated by a particular development. However, just over fifty recovery plans had been approved by October 2002 (Appendix 8.1). Ultimately, a recovery plan leads to an increase in knowledge about the threatened species. It can be argued that this mechanism is most likely to protect biodiversity because sound scientific knowledge is brought to bear on the protection of the basic building blocks, species. However, this potential power of the recovery plan is not utilised. Whereas listing, eight part tests, and SISs operate as a unit, recovery plans operate independently but, once completed, they do need to be considered as part of the SIS process. Interestingly, much of the information typically contained in recovery plans is what is required in a comprehensive SIS. Why is it that recovery plans are usually the last mechanism activated?

Recovery plans have the potential to be a powerful conservation instrument. However, with the requirement for consideration of all social and economic factors, they are nowhere near as forceful as they might be (Brunton 1997; Kelly 1996). Therefore, the greatest strength of a recovery plan lies with developments on public land. For example, Council can choose to integrate a recovery plan with their plans of management for community land (Brebach 1997; Kelly 1996; Lim 1996). It is important to understand that the TSCA places the onus on local councils to take part in conservation of the environment. Chapter 8 General Discussion 181

Research and further survey work are the main requirements in the recovery plans that have been written to date. In the present recovery plans, fourteen species occur wholly within managed reserves (nature reserves, national parks etc) (e.g. Egernia margaretae, Burramys parvus, Allocasuarina portuensis and Wollemia nobilis). Six species (Neobatrachus pictus, Cercartetus concinums, Angiopteris evecta, Zieria formosa, Zieria buxijugum and Zieria parrisae) occur only on private lands, while the other species occur on a mix of land tenures.

It is important to note that only three recovery plans specifically mention urban/residential development as a threat to the decline of the species, and only one plan mentions including habitat protection for a listed species in planning documents (Grevillea wilkinsonii). However, a number of recovery plans do refer to population and habitat protection as well as identifying populations of the highest conservation value (e.g. Genoplesium plumosum ). The identification of critical habitat has always been a difficult issue, with only two areas nominated so far, both on public lands. A number of recovery plans do take an important step towards identifying potential critical habitat for the species (e.g. Persoonia mollis subsp. maxima, Elecocharis teraquetra). A number of threats have been identified that are the result of human activity such as power boating (population of Little Penguins, Eudyptula minor), chemical and insecticide use (Litoria castanea, Litoria piperata, Antechniomys laniger), snow making and global warming (Burramys parvus) and rubbish dumping (Grevillea caleyi). However, recovery actions do not necessarily address these precise problems.

It is also important to note, at this stage, that recovery plans do refer to potential socio­ economic impacts. This section of recovery plans typically states that these impacts are unknown but expected to be minimal, if any (e.g. "There are minimal social or economic costs resulting from the protection of this species" (Zieria adenophora) and "Any adverse social and economic consequences of conserving this species are either unknown or insignificant" (Hakea pulvinifera). So, do recovery plans have the ability to halt development that may impact upon a species? Chapter 8 General Discussion 182

Using the draft recovery plan for Prostanthera junonis as an example, it is possible to examine whether implementing a recovery plan could possibly halt the granting of development consent. There are nine known populations of P. junonis existing on land under a variety of tenures and zonings. Under section 8.2.3 of the draft recovery plan, development is classified as a threatening process. What would happen if a development application were lodged to allow industrial development on private land, resulting in the destruction of 40% of the population of P. junonis? Approval for development can be obtained under TSCA s 69(2) because the body responsible for making the decision on development consent (e.g. local council) has a statutory discretion to exercise (Table 8.4). Even when development is inconsistent with a recovery plan, approval can be obtained and the NPWS only needs to be notified of departures from the plan. The Director-General has no power over the decision on whether it is appropriate to depart from a recovery plan, as the Minister (s. 73(3) or even the Premier (s. 73(4)) can make the final decision. Therefore, identification of development as a threatening process in a recovery plan is only a factor to be considered in the decision-making process.

I argue that recovery planning should take a stronger proactive role. This could be achieved in two distinct ways. Before an approval body can assess a SIS, the recovery plan for that particular threatened species should be written and the information considered. However with pressure for Local Councils to streamline the development approval process, at present this may not be realistic. Strengthening the recovery planing process requires recovery plans to be integrated into Local Environmental Plans (LEPs ). At present, LEPs do not have to take into consideration any recommendations in a recovery plan, unless critical habitat is declared. Efficient plans at the local and regional scale should take into consideration specially designated reserve or sensitive areas identified by a recovery plan. These areas could also include areas of potential habitat for threatened species. The NPWS has been slow to produce recovery plans because of the expense. An option would be to require developers to contribute financially to the making of recovery plans, if sites they own Q {5 Table 8.4: Implementation of recovery plans are covered in Division 2, sections 69(1) - 73 (5) of the Threatened Species Conservation Act 1995 (NSW). ...~ Section of the TSCA Note: oublic authorities include local councils Outcomes Oo C) 69 Ministers and public authorities to implement recovery plans A development application should be refused if development is listed as a ;:~ (1) Ministers and public authorities (including the Director-General) are to take any threatening process as it would be inconsistent with the recovery plan. ~ appropriate action available to them to implement those measures included in a recovery ~ plan for which they are responsible and must not take decisions that are inconsistent with b the provisions of a recovery plan. o;;· ~ (2) If the implementation of a recovery plan affects a statutory discretion of a Minister or However, a development application may be granted permission by the ~ ~· public authority, this section does not operate to exclude the discretion, but the Minister discretion of the Minister or another public authority. or authority must take the recovery plan into account.

(3) This section does not operate to require or authorise any action by a Minister or If a development is approved by a local government plan, then a Minister of public authority that is inconsistent with any statutory or other legal obligation of the public authority may exercise discretion to approve a development application. Minister or public authority

71 Notification of proposed departures from recovery plan A development application must not be approved if its outcomes are (1) A public authority must not exercise a function in a manner that is inconsistent with inconsistent with the recovery plan. the implementation of measures included in a recovery plan unless:

(a) in the case of a public authority other than the Director-Genera/ - it has given written Development approval may be granted with notification to the Director­ notice of the proposed exercise of the function to the Director-General, General. or

(b) in the case of the Director-General - the Director-General has given written notice of Development approval may be granted with notification to the Minister. the proposed exercise of the function to the Minister.

(2) The Director-General must comply with any directions given by the Minister The Director-General must give approval if the Minister gives development concerning a proposed departure from a recovery plan. permission (in particular under Part 5 of the EPAA).

(3) This section does not apply in relation to anything authorised to be done by or under If there is a threat to life or property, a development (in particular an activity the Bush Fires Act 1949 or the State Emergency and Rescue Management Act 1989 that is such as hazard reduction fires), permission is granted. reasonably necessary in order to avoid A threat to life or property.

72 Consultations with Director-general concerning proposed departures The Director-General must decide if the departure from the recovery plan is (1) The Director-General must, on receiving notice of a proposed departure from a acceptable, however, the discretion of the Minister is final. recovery plan from a public authority, determine whether exercise of the function in the manner proposed is acceptable or whether it is likely to jeopardise the effective implementation of the plan

(2) If the Director-General considers that the departure is acceptable, the Director­ NIA General must notify the public authority accordingly.

(3) If the Director-General considers that the departure is unacceptable because it is A consultation process can take place however, has little point once a Minister likely to jeopardise the effective implementation of the recovery plan, the Director-general or public authority have made a discretionary decision. must consult with the public authority in an endeavour to resolve the matter by modification of the action proposed or by other mutually acceptable means.

73 reference of proposed departures to ministers and Premier The respective Minsters have the final decision (1) A matter that has not been resolved after consultation between the Director-General and the public authority concerned must be referred by the parties to their respective Ministers.

(2) In the case of a council, the reference is to the Minister administrating the Local NIA Government Act 1993 unless the matter relates, in whole or in part, to the exercise of functions under the Environmental planning and assessment Act 1979. In that event, the reference is to be to the Minister administrating the environmental planning and assessment act 1979.

(3) The Ministers, on receiving a reference, are to consult in an endeavour to resolve the Minister may consult with each other. matter by means that the Ministers consider to be appropriate.

(4) If the Ministers are unable to resolve the matter after consultation, it is to be refereed If a decision cannot be reached the Premier may take the final decision. to the Premier for resolution.

(5) A public authority (including the Director-General) must give effect to any decision of, The Premier has the final decision regardless of discretionary matters. or directions made or given by, the Premier on the matter and is, despite the requirements ,_.. of any other Act or law. empowered to comply with a'D'. such decision or directions. ~ Chapter 8 General Discussion 184

and/or wish to develop contain that species. Developers would then have a stake in the appropriate protection of listed threatened species.

8.2.4 Incorporating the precautionary principle into decision-making

In a broad sense or in populist notions, the term 'science' gives a sense of systematised knowledge (Stratford et al. 2000). Newtonian physics forms the basis of the western view of science, ultimately forming a mechanistic view of the natural world by breaking it down into basic building blocks, which can be studied in controlled experiments. However, objective science is often weighed up against people, policies and politics (Holling et al. 1998).

The place of ecological knowledge and investigation has always been tenuous within legislation and policy (Dovers et al. 1996). A knowledge of the science of ecology is necessary both to provide information in conducting an assessment of the likely impact of a development or of amelioration, and to decide where appropriate information or studies are lacking, in the process of assessing the adequacy of 8-part-tests and SISs (Dovers et al. 1996). The ecology-policy relationship is difficult, because environmental problems are hard to define, study and resolve due to features such as complexity, uncertainty, irreversibility and a wide range of spatial and temporal scales. These features clash with the concepts that underpin legislation, policy and management (Dovers et al. 1996), such as certainty.

It is important to recognise that ecology is a relatively young science compared to other established sciences (Dovers et al. 1996). Ecology has an incomplete knowledge base and is still gripped by debates in fundamental knowledge (Dovers et al. 1996). Ecosystems (or any other natural system) are complex and constantly changing. Thus, they have an inherent unknowability and unpredictability (Holling et al. 1998) leading to uncertainty. It must be remembered that knowledge is not absolute (Stratford et al. 2000) and the need to make a Chapter 8 General Discussion 185

decision, can stretch knowledge (and therefore predictive capacity) to its limits in policy and management (Dovers et al. 1996).

A possible scenario is illustrated in the case study of Carnarvon Drive, Frenches Forest (see Section 3.4). Tetratheca glandulosa produces a soil seed bank, and its existence or the impact of development on the seed bank was unknown. One response to scientific uncertainty is to adopt of the precautionary principle. Therefore, it could be argued that the only possible response to this scenario in legislation, policy and management is to apply an approach which: (1) maximises the gathering and application of current knowledge; (2) applies precaution in some form; and, (3) incorporates adaptive management.

8.4.3. Adaptive Management

Local councils can incorporate the precautionary principle into their decision-making pathways (Figure 8.1 & Figure 8.2) dealing with the assessment of development applications when threatened species are a concern. When an impact is detected, it does not mean that the development automatically needs to be halted. In some cases, adaptive management may be a more appropriate pathway (providing the threatened species is not destroyed on the site), which would allow for the gathering of ecological data in relation to the impact of development on the threatened species present.

Adaptive management acknowledges the uncertainty in the effect of applying a particular policy (Young 1999), which arises from unpredictable interactions between people and ecosystems (Berkes & Folke 1989) and has the goal of ensuring · that the path of development remains as reversible as possible (Young 1999). Such an approach allows management and policy to change as scientific knowledge is gathered, reducing uncertainty (Young 1999). Adaptive management has been described as a "reserved rationality" approach, which would seem to be intuitive when there is a unmeasurable risk but severe consequences, such as irreversible loss or damage (Young 1999; Whelan 2003). This approach to management differs from traditional management because of the inclusion of a ...... _'111*'11....,..,...."'- 9 .,.._..,.(TSCA..110(1)11 ~

~ o.o;m---- Clo -.:..~::=(~)- [ -·-w.1-opod•»ioiy ~~=r..: .. :.:::..:~..=\:;? -....io~-~ - .. -~)1 ti

~·

l_ _ _l.::J - r ~,..--:,_ I ______· ~- ~

_:::::=,,~::."' ?-~~::=:::~------·------~ °'"'""'o11e1om... p ... 11>t1-to ...... - ... (n;ic>.•''tll2\l

'"9w.:::=i.1"!:',,.,.I- -·------tpocioo(TSCA.o..••OQ110)f7

1------i ··--"'-~"'-:::::;:....,.,,.:::.::d ...~...

~~=.:::::.~------·------~ -~o..llotn(O!J

°"""""9-~-­ -°'"'"{TSC;Ao..11012)!1)7 .....~"' "'"'""- _....,,._..,,.-..,_ ..,p1oca1~- ­ tu'n-oll~lr>U..!.,.a.ol "'"-(TSCA,o. 110(2)(Q)7 §;]

~~~::=~·------;----;======~------i

....-aan~o.llQ(2)(11)) No~~...- i I .I--··-·-· -U

~=-~~ .:.-:-1 ~..,.-- .... ~ .....-~--v.­ .. t~

~~ ~~-I 1 l---- - ~.-1---,--_~-----~1

Figure 8.1: Generic pathway for decision-making when completing and assessing Species Impact Statements, including when scientific evidence .... is required. The precautionary principle should be implemented when caution and scientific uncertainty prevail (indicated with a* ), leading to a ~ rejection of the development application. There are also opportunities for adaptive management strategies to be incorporated in the decision­ making process. Chapter 8 General Discussion 187

Is the life cycle of the species or population going to be interrupted rendering the population unviable Is there the possibility SIS must be (changes to any parameters such as r-Yes No---. for amelioration? reproduction and survival to completed reproductive age could cause a loss in viability (EPAA s. 5A (a)(b)) Yes

Is there scientific N~ SIS must be +r----Yes------1 evidence to support the completed amelioration outcome? Survey regional distribution of habitat, and determine how much Is there the possibility SIS must be habitat is to be destroyed during the ~ Yes No for amelioration? completed stated action. Is this a significant amount (EPAA s.5A(c))? Yes I No Is there scientific SIS must be .>----Yes---___, evidence to support the No completed amelioration outcome?

Is the habitat for the threatened Is there the possibility SIS must be species likely to become isolated r-Yes No for amelioration? completed (EPAA s.5A(d))?

Yes I Is there scientific No SIS must be 1-----Yes- - --_, evidence to support the No*. completed amelioration outcome? Is the area listed as critical habitat (EPAA s.5A(e))? Yes SIS must be No completed

Is the threatened• species adequately represented in SIS must be -Yes .... conservation reserves (EPAA completed s.5A(f))? I No

Is the action a threatening process SIS must be Yes-. (EPAA s.5A(g))? completed

No

What is the limit of the distribution SIS must be of the threatened species (EPAA Yes completed s.5A(h))?

No

Therefore, is there going to be an SIS must be Yes-. impact upon the threatened completed species?

No Figure 8.2: Generic pathway for decision-making when y completing and assessing eight part tests. The Submit eight part test with precautionary principle should be implemented when development application to local caution and scientific uncertainty prevail (indicated council with a*) leading to an SIS. Chapter 8 General Discussion 188

feedback loop, which allows for learning and the progressive accumulation of knowledge and for systematic experimentation to become entrenched in an overall management scheme (Berkes & Folke 1989).

Adaptive management is, however inappropriate in many circumstances, particularly where development is concerned, because the impacts of the particular proposal on threatened species are irreversible.

8.4.4 Improved pathways for decision-making in Local Councils

There is currently no process set out in legislation to guide the evaluation of eight part tests and SISs and the process can be influenced both by the limits of available information and by the desired outcomes for both local council and the developer. There are guidelines provided to consultants from NPWS.

Species Impact Statements could, however, provide an opportunity for an adequate response to scientific uncertainty, by invoking the precautionary principle and also requiring incorporation of adaptive management strategies into development conditions (Figure 8.3). An amelioration process via consultation with Local Council and appropriate experts (e.g. ecologists) can be built into the SIS process to allow for smoother and shorter decision­ making periods with more informed outcomes (see Section 3.3.2). However, the developer should be required to supply scientific evidence that such measures are likely to benefit the threatened species population in the long term. When this is not possible, the precautionary principle could be implemented in one of two ways: amelioration could be devised on the basis of possible impacts and could err on the side of conservative strategies; the development application could be rejected until the scientific evidence is available.

The amelioration process used in the completion of eight part tests (see Section 3.3.2, Green Road, Glenhaven) could be successfully extended to include scientific evidence (Figure 8.1) to support claims or statements about significant impacts. Alternatively the Chapter 8 General Discussion 189

Is T. glandulosa present Is it the flowering No No, SIS required on the site? (site survey) season? Yes

No

Yes Resurvey in appropriate time

Are potential pollinators Are there potential nesting No-. Population unlikely to be viable present at the site? No sites for pollinators

Yes SIS required with an examination of ------1 why pollinators may not be present.

I Further managment options should be I ! examined I I Are potential pollinators I :I I visiting T. glandulosa? I ~------·

Yes------..i Is pollen being removed from 1----..i No flowers?

Yes

Is fruit being set? No

Yes Estimation 'of seed bank through modeling for long term management of population

Figure 8.3: A flow chart demonstrating potential paths for deciding the importance of a population of Tetratheca glandulosa within the decision-making process of development application. It is suggested that once the process is passed the dotted line all development applications should be either rejected or enter an amelioration process. Chapter 8 General Discussion 190

eight part test could include an assessment of the nature of the evidence that the amelioration will be successful or that the devising of the amelioration could incorporate scientific study before the 8-part-test is completed.

Developers and local councils need to recognise that, in a majority of situations, properly designed surveys or experiments may be required to assess the viability and importance of a population, the potential impact of a proposed development, and the likely success of proposed amelioration actions. Assessing the viability of a population will slow down the process of decision-making but will allow for conservation and informed decision-making and reduce the likelihood of concluding that there will be no significant effect when, in fact, there will be one.

The importance of gathering appropriate ecological data before the decision-making process commences is highlighted in the ARDEL case study in relation to Tetratheca glandulosa populations. My research highlighted the importance of pollinators and their role in maintaining the viability of populations of T. glandulosa, providing a potential pathway for deciding the importance of an individual population (Figure 8.2). There are a number of questions that should be asked by consultants or developers preparing an eight part test or SIS for T. glandulosa. These are: (1) Is the surveying taking place during the flowering season? (2) Were potential pollinators present at the site? (3) Are the potential pollinators visiting flowers? ( 4) Is pollen being removed? (5) Is fruit being set? Hence pollinators play an important role in conservation decision-making (Donaldson et al. 2002)

It is apparent that the TSCA needs to change to accommodate the state of scientific knowledge, theory and practices. A more appropriate model for the TSCA would be as shown in Figure 8.4. Such a model is based on the incorporation of the precautionary principle and an active role for scientific knowledge. First, my model incorporates a greater emphasis on basic survey work for distribution and abundance of species and ecological assessment, which aids in the change of burden of proof (e.g. from proving that a species is threatened and should therefore be listed to proving a species is not threatened and can be Chapter 8 General Discussion 191

rightfully left off a list). Second, my model demands an integration of this more realistic listing process with the more proactive tools of (i) recovery planning, (ii) LEPs, and (iii) regional planning, as well as the reactive process of assessing development applications. Third, I include the use of a peer review system to allow the overall system to be scrutinised and maintain integrity.

precautionary listing of species

general schedule

listing process threatened incorporating the 8 ecological assessment species of species part test

is the species in no reality common? present schedule

Yes

remove from naturally rare species recovery planning schedule process

listed on different ,___ _ LEPs and regional schedules planning

development ------< development application rejected submitted

SIS and DA assessed SIS

development approved peer review of SIS Figure 8.4: Proposed alternative model for the Threatened Species Conservation Act 1995 (NSW). References 192

REFERENCES

Abbott, I. (2000). Improving the conservation of threatened and rare mammal species through translocation to islands: Case study Western Australia. Biological Conservation 92: 335-343.

Abenspergtraun, M., Smith, G.T., Arnold, G.W. and Steven, D.E. (1996). The effects of habitat fragmentation and livestock grazing on animal communities in remnants of Gimlet Eucalyptus salvbris woodland in the Western Australian Wheatbelt. Journal ofApplied Ecology 33: 1281-1301.

Adam, P., Auld, T., Benson, D., Catling, P., Dickman, C., Fleming, M., Gunning, R., Hutchings, P., Kemp, D. and Shields, J. (1997). The 10 lords of the universe respond to Lim. Pacific Conservation Biology 3: 319-321.

Aizen, M.A. and Feinsinger, P. (1994). Habitat fragmentation, native insect pollinators and feral honey bees in Argentine "Chaco serrano". Ecological Applications 4: 378-392.

Allan, P. (1995). Parliamentary debates. Legislative Assembly. HANSARD September 1995.

Allan-Wardell, G., Bernhardt, P., Bitner, R., Burquez, A., Buchmann, S., Cane, J., Cox, P.A., Dalton, V., Feinsinger, P., Ingram, M., Inouye, D., Jones, C.E., Kennedy, K., Kevan, P., Koopowitz, H., Medellin, R., Medellian-Morales, S., Nabhan, G.P., Pavlik, B., Tepedino V., Torchio, P., and Walker, S. (1998). The potential consequences of pollinators decline on the conservation of biodiversity and stability of food crop yields. Conservation Biology 12: 8-17.

Andersen, A.N. and Morrison, S.C. (1998). Myrmecochory in Australia's seasonal tropics: effects of disturbance on distance dispersal. Australian Journal of Ecology 23: 483-491.

Anderson, V.J. , Hodgkinson, K.C. and Grice, A.C. 1996. The influence of recent grazing pressure and landscape position on grass recruitment in a semi-arid woodland of eastern Australia. Rangeland Journal 18: 3-9.

Andreassen, H.P., Halle, S. and Ankerims, R. (1996). Optimal width of movement corridors for root voles: not too narrow and too wide. Journal ofApplied Ecology 33: 63-70.

Andren, H. and Angelstam, P. (1988). Elevated predation rates as an edge effect in habitat islands: experimental evidence. Ecology 69: 544-547

Angermeier, P.L. and Karr, J.R. (1994). Biological integrity versus biological diversity as policy directives. BioScience 44: 690-697.

ANZECC (Australian and New Zealand Environment and Conservation Council) (2001). Review of the National Strategy for the Conservation of Australia's Biological Diversity. Environment Australia, Canberra, Australia. References 193

Aparicio, A. and Guisanda, R. (1997). Replenishment of the endangered Echinospartum algibicum (Genisteae, Fabaceae). Biological Conservation 81: 267-273.

Atwell, B.J., Kriedemann, P.E. and Turnbull, C.G.N. (1999). Plants in Action: Adaptation in Nature, Performance in Cultivation. McMillan Education, South Melbourne, Australia.

Auld, T.D (1986). Dormancy and viability m Acacia suaveolens (Sm.) Willd. Australian Journal of Botany 34: 463-4 72.

Auld, T.D. (1991). Patterns of predispersal seed predators in the Fabaceae of the Sydney region, south-eastern Australia. Australian Journal of Zoology 39: 519-528.

Auld, T.D. (1995). Burning Grevilleas, ants, rats and wallabies. CALM Science Supplement 4: 159-164.

Auld, T.D. and Denham, A.J. (1999). The role of ants and mammals in dispersal and post-dispersal seed predation of the shrubs Grevillea (Proteaceae ). Plant Ecology 144: 201-213.

Auld, T.D. and O'Connell, M.A. (1991). Predicting patterns of post-fire germination in 35 eastern Australian Fabaceae. Australian Journal of Ecology 16: 53-70.

Auld, T.D., Bradstock, R.A. and Keith, D.A. (1991). Germination or rare plants in relation to fire. World Wide Fund for Nature Australia, Final report project P154, Hurstville NPWS.

Auld, T.D., Bradstock, R., Keith, D. (1993). Fire as a threat to populations of rare plants. Australian National Parks and Wildlife Service, Endangered Species Program, Endangered species project no. 31, Hurstville, N.S.W.

Auld, T.D., Keith, D.A., and Bradstock, R.A. (2000). Patterns in longevity of soil seedbanks in fire-prone communities of south-eastern Australia. Australian Journal of Botany 48: 539-548.

Australian Network for Plant Conservation (1998). Guidelines for the Translocation of Threatened Plants in Australia. ANPC, Canberra.

Ayre, D.J. and Whelan, R.J. (1989). Factors controlling fruit set in hermaphroditic plants: studies with the Australian Proteaceae. Trends in Ecology and Evolution 4: 267-272.

Bailey, P. (1991a). Forest Plan a threat to rare fauna: scientists. Sydney Morning Herald 22nd August 1991.

Bailey, P. (199lb). Estate listing hope for forests. Sydney Morning Herald. 27th August 1991. References 194

Bailey, P. (1991c). Government tramples wildlife laws. Sydney Morning Herald 12 October 1991.

Baker, A.M., Barrett, S.C.H. and Thompson, J.D. (2000). Variation of pollen limitation in the early flowering Mediterranean geophyte Narcissuss assoanua (Amaryllidaceae ). Oecologia 124: 529-535.

Baker, J. and Whelan, R. (1994). Ground parrots and fire at Barren Grounds, New South Wales: A long-term study and an assessment of management implications. Emu 94: 300-304.

Bartier, F.V., Gross, C.L., Mulligan, D.R., Bellairs, S.M. and Bowen, D. (2001). Understanding the biology and ecology of vulnerable plant species - a case study with Tetratheca juncea occurring over coal leases, ACARP Project C8012. A report prepared for Australian Coal Research. June 2001.

Barton, C. (1998). The status of the precautionary principle in Australia: its emergence in legislation and as a common law doctrine. Harvard Environmental Law Review 22: 509-558.

Beardsell, D.V., O'Brien, S.P., Williams, E.G., Knox, R.B. and Calder, D.M. (1993). Reproductive biology of Australian Myrtaceae. Australian Journal of Botany 41: 511-526.

Beattie, M. (1996). Biodiversity policy and ecosystem management. In: Sharpe, W.J. (ed). Biodiversity and the Law. Island Press, Washington DC, pp. 11-15.

Beder, S. (1993). The Nature of Sustainable Development, Scribe publications, Australia.

Beier, P. and Noss, R.F. (1998). Do habitat corridors provide connectivity? Conservation Biology 12: 1241-1252.

Bengtsson, J. (1989). Interspecific competition increases local extinction rate m a metapopulation system. Nature 340: 713-715.

Benson, D.H. (1985). Maturation periods for fire sensitive shrub species in Hawkesbury sandstone vegetation. Cunninghamia 1: 339-349.

Bentley, J.M. and Catteral, C.P. (1997). The use of bushland corridors and linear remnants by birds in southeastern Queensland Australia. Conservation Biology 11: 1173-1189.

Bentley, J.M., Catterall, C.P. and Smith, G.C. (2000). Effects of fragmentation of araucarian vine forests on small mammal communities. Conservation Biology 14: 1075-1087.

Eerkes, F. and Folke, C. (1998). Linking social and ecological systems for resilience and sustainability. In: Eerkes, F. and Folke, C. (eds). Linking Social and References 195

Ecological Systems: Management Practices and Social Mechanisms for Building Resilience, Cambridge University Press, Cambridge, pp. 1-26.

Bevill, R.L., Louda, S.M. and Stanforth, L.M. (1999). Protection from natural enemies in managing rare plant species. Conservation Biology 13: 1323-1331

Blair, R.B. (1999). Birds and butterflies along an urban gradient: surrogate taxa for assessing biodiversity. Ecological Applications 9: 164-170.

Blarney, R., Rolfe, J., Bennett, J. and Morrison, M. (2000). Valuing remnant vegetation in central Queensland using choice modeling. Australian Journal of Agricultural and Resource Economics 44: 439-456.

Blay, S.K.N. and Piotrowicz, R.W. (1993). Biodiversity and conservation in the twenty­ first century: a critique of the Earth Summit 1992. Environmental and Planning Law Journal 10: 450-469.

Bodansky, D. (1991). Scientific uncertainty and the precautionary principle. Environment 33: 4-5.

Boesewinkel, F.D. (1999). Ovules and seeds of Tremandraceae. Australian Journal of Botany 47: 769-781.

Bolger, D.T., Suarez, A.V., Crooks, K.R., Morrison, S.A. and Case, T.J. (2000). Arthropods in urban habitat fragments in southern California: area, age and edge effects. Ecological Applications 10: 1230-1248

Bond, W.J. (1994). Do mutualisms matter - assessing the impact of pollinators and disperser disruption on plant extinction. Philosophical Transactions of the Royal Society ofLondon - Series B: Biological Sciences. 344: 83-90.

Bonyhady, T. (1993). Places Worth Keeping: Conversationalists, Politics and Law. Allen & Unwin; Sydney.

Bowers, J. (1999). Policy instruments for the conservation of remnant vegetation on private land. Biological Conservation 87: 327-339.

Bradsen, J. (1992). Biodiversity legislation: species, vegetation, habitat. Environmental and Planning Law Journal 9: 175-180.

Bradstock, R.A. and Auld, T.D. (1995). Soil temperatures during experimental bushfires in relation to fire intensity: consequences for legume generation and fire management in south-eastern Australia.Journal of Applied Ecology 32: 76-84.

Bradstock, R.A. and Bedward, M. (1992). Simulation of the effect of season of fire on post-fire seedling emergence of two Banksia species based on long-term rainfall records. Australian Journal ofBotany 40: 75-88.

Brebach K. (1997). The threatened species conservation act, how to nominate an endangered species for protection. National Parks Journal 40: 18-19. References 196

Briggs, B.G. (1962). The New South Wales species of Darwinia. Contributions from the N.S. W National Herbarium 3: 129-150.

Brown, C.L. and Whelan, R.J. (1999). Seasonal occurrence of fire and availability of germinable seeds in Hakea sericea and Petrophile sessilis. Journal of Ecology 87: 932-941.

Brown, E. and Kephart, S. (1999). Variability in pollen load: implications for reproduction and seedling vigor in a rare plant, Silene douglasii var. oraria. International Journal of Plant Sciences 160: 1145-1152.

Brundet, J. and Mundt, C. (2000). Combined effects of disease and competition on plant fitness. Canadian Journal of Botany 78: 646-654.

Brunner, R.D. and Clark, T.W. (1997). A practice-based approach to ecosystem management. Conservation Biology 11: 48-58.

Brunton, N. (1996). The nature of recent environmental law reforms in New South Wales. Environmental and Planning Law Journal 13: 71-81.

Bruna, E.M. and Kress, W.J. (2002). Habitat fragmentation and the demographic structure of an Amazonian understory herb (Heliconia acuminata). Conservation Biology 16: 1256-1266.

Buchmann, S.L. and Cane, J.H. (1989). Bees assess pollen returns while sonicating Solanum flowers. Oecologia 81: 289-294.

Buchmann S.L. and Nabhan G.P. (1996). The Forgotten Pollinators. Island Press, Washington, DC.

Burbidge, A.H. and James, S.H. (1991). Postzygotic seed abortion in the genetic system of Stylidium (Angiospermae: Stylidiaceae) Journal of Heredity 82: 319-328.

Burd, M. (1994). A probabilistic analysis of pollinator behaviour and seed production in Lobellia deckenii. Ecology 75: 1635-1646.

Burel, F., Baudry, J., Butet, A., Clergeau, P., Delettre, Y., Lecoeur, D., Dubs, f., Morvan, N., Paillat, G., Petit, S., Thenail, C., Brunel, E., and Lefeuvre, J.C. (1998). Comparative biodiversity along a gradient of agricultural landscapes. Acta Oecologia 19: 47-60.

Burgman, M.A., Keith, D.A., Rohlf, F.J. and Todd, C.R. (1999). Probabilistic classification rules for setting conservation priorities. Biological Conservation 89: 227-231.

Burgman, M.A., Possingham, H.P., Lynch, A.J .J ., Keith, D.A., McCarthy, M.A., Hopper, S.D., Drury, W.L., Passioura, J.A. and Devries, R.J. (2001). A method for setting the size of plant conservation target areas. Conservation Biology 15: 603-616. References 197

Buza, L., Young, A. and Thrall, P. (2000). Genetic erosion, inbreeding and reduced fitness in fragmented populations of the endangered tetraploid pea Swainsona recta. Biological Conservation 93: 177-186.

Cabin, R.J., Marshall, D.L. and Mitchell, R.J. (2000). The demographic role of soil seed banks. II. Investigations of the fate of experimental seeds of the desert mustard Lesquerella fendleri. Journal ofEcology 88: 293-302.

Calver, M.C., Bradley, J.S., and Wright, I.W. (1999a). Towards scientific contributions in applying the precautionary principle: an example from southwestern Australia. Pacific Conservation Biology 5: 63-72.

Calver, M.C., Bradley, J.S., and Wright, I.W. (1999b). Environmental management: the precautionary principle and null hypothesis. Pacific Conservation Biology 5: 78- 82.

Campbell, G.S., Jungbauer, J.D., Bidlake, W.R. and Hungerford, R.D. (1994). Predicting the effects of temperature on soil thermal conductivity. Soil Science 158: 307-313

Cane, J.H. (2001). Habitat fragmentation and native bees: a premature verdict? Conservation Ecology 5:3. [online] URL: http://www.consecol.org/vol5/issl/art3

Cane, J.H. and Tepedino, V.J. (2001). Causes and extent of declines among Native North American invertebrate pollinators: detection, evidence and consequences. Conservation Ecology 5:1. [online] URL: http:(/www.consecol.org/vol5/issl/artl

Carson, R. (1962). Silent Spring. Penguin Books: U.S.A.

Carthew, S.M. (1993). Patterns of flowering and fruit production in a natural population of Banksia spinulosa. Australian Journal of Botany 41: 465-480.

Catterall, C.P., Kingston, M.B., Park, K. and Sewell, S. (1998). Deforestation, urbanisation and seasonality interacting effects on a regional bird assemblage. Biological Conservation 84:65-81.

Chalmers, A.F. (1991 ). What is this Thing Called Science? An Assessment of the Nature and Status of Science and its Methods. University of Queensland Press, Queensland.

Christie, E. (1991). The role of law and science in the resolution of disputes over factual evidence. Environmental Planning Law Journal 8: 200-210.

Christie, E. (1993). The eternal triangle: the biodiversity convention, endangered species legislation and the precautionary principle. Environmental and Planning Law Journal 10: 470-485. References 198

Clements, A. (1983). Suburban development and resultant changes in the vegetation of the bushland of the northern Sydney region. Australian Journal of Ecology 8: 307-319.

Clergeau, P., Savard, J.P.L., Mennechez, G. and Falardeau, G. (1998). Bird abundance and diversity along an urban-rural gradient: a comparative study between two cities on different continents. Condor 100: 413-425.

Coates, D.J. and Atkins, K.A. (2001). Priority setting and the conservation of Western Australia's diverse and highly endemic flora. Biological Conservation 97: 251- 263.

Coates, F., Kirkpatrick, J.B., and Minchin, P.R. (1999). Towards an explanation of the causes of the rarity of two Tasmanian Spyridium species. Australian Journal of Ecology 24: 11-17.

Cohen, I. (1997). Green Fire. Angus & Robertson, Sydney: Australia.

Collins, B.G. and Spice, J. (1986). Honeyeaters and the pollination biology of Banksia prionotes (Proteaceae ). Australian Journal ofBotany 34: 175-185.

Compton, J.E. and Boone, J.D. (2000). Long term impacts of agriculture on soil carbon and nitrogen in New England forest. Ecology 81: 2314-2330.

Cox, P.A., Elmqvist, T., Pierson, E.D. and Rainey, E.D. (1991). Flying foxes as strong pollinators in South Pacific Island ecosystems: a conservation hypothesis. Conservation Biology 5: 448-454.

Cox, J. and Engstrom, R.T. (2001). Influence of the spatial pattern of conserved lands on the persistence of a large population of red-cockaded woodpeckers. Biological Conservation 100: 137-150.

Craig, J.L. (1997). Managing bird populations: for whom and at what cost? Pacific Conservation Biology 3: 172-182.

Crawley, M.J. (1997). Plant Ecology. Blackwell Science, London.

Cunningham, S.A. (2000a). Depressed pollination in habitat fragments causes low fruit set. Proceedings of the Royal Society ofLondon, Series B 267: 1149-1152.

Cunningham, S.A. (2000b ). Effects of habitat fragmentation on the reproductive ecology of four plant species in mallee woodland. Conservation Biology 14: 758- 768.

Dawson, F. (1996). The significance of property rights for biodiversity conservation in the northern territory. The Australasian Journal of Natural Resources Law and Policy 3: 179-238. References 199

Day, D.A., Collins, B.G. and Rees, R.G. (1997). Reproductive biology of the rare and endangered Banksia brownii Baxter ex R. Br. (Proteaceae). Australian Journal of Ecology 22: 307-315.

Denham, A.J. and Whelan, R.J. (2000). Reproductive ecology and breeding system of Lomatia silaifolia (Proteaceae) following a fire. Australian Journal of Botany 48: 261-269.

Dennis, B., Munholland, P.L. and Scott, J.M. (1991). Estimation and extinction parameters for endangered species. Ecological Monographs 61: 115-143.

Deville, A. and Harding, R. (1997). Applying the Precautionary Principle. The Federation Press, Sydney.

Dixon, N. (1994). Protection of endangered species - How will Australia cope? Environmental and Planning Law Journal 11: 6-30.

Donaldson, J., Nanni, I., Zachariades, C. and Kemper, J. (2002). Effects of ahbitat fragmentation on pollinator diversity and plant reproductive success in Renosterveld Shrublands of South Africa. Conservation Biology 16: 1267-1276.

Douglas, S. (1998). Report on Tetratheca glandulosa for National Parks Wildlife Service. ESP Ecological Surveys and Planning.

Dovers, S. (1994). Australian environmental history: introduction, review and principles. In: Dovers, S. ( ed) Australian Environmental History: Essays and Cases. Oxford University Press, Melbourne.

Dovers, S.R., Norton, T.W., and Handmer, J.W. (1996). Uncertainty, ecology, sustainability and policy. Biodiversity and Conservation 5: 1143-1167.

Doyle, J.J. and Doyle, J.L. (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15.

Dunning, J.B., Borgella, R., Clements, K. and Meffe, G.K. (1995). Patch isolation, corridor effects, and colonization by a resident sparrow in a managed pine woodland. Conservation Biology 9: 542-550.

Earl, J.M. and Jones, C.E. (1996). The need for a new approach to grazing management: is cell grazing the answer? Rangeland Journal 18: 327-350

Eckert, C.G. (2000). Contributions of autogamy and geitonogamy to self-fertilization in a mass-flowering, clonal plant. Ecology 81: 532-542.

EcoHealth Services (1999). Bushland Management Plan for the subdivision of the ARDEL property (lot 2854) Aquatic drive, Allambie Heights including connecting road reserve. Unpublished report for ARDEL Pty. Ltd. References 200

Edwards, W. and Whelan, R.J. (1995). The size distribution and germination requirements of the soil-stored seed-bank of Grevillea barklyana (Proteaceae). Australian Journal of Ecology 20: 548-555.

Eisto, A.K., Kuiyunen, M., Lammi, A., Saari, V., Suhonen, S. and Tikka, P.M. (2000). Population persistence and offspring fitness in the rare bellflower Campanula cervicaria in relation to population size and habitat quality. Conservation Biology 14: 1413-1421.

Ellery, A.J. and Chapman, R. (2000). Embryo and seed coat factors produce seed dormancy in capeweed (Arctotheca calendula). Australian Journal ofAgricultural Research 51: 849-854.

Ellstrand, N.C. and Roose, M.L. (1987). Patterns of genotypic diversity in clonal plant species. American Journal of Botany 74: 123-131.

Emms, S.K. and Arnold, M.L. (2000). Site-to-site differences in pollinator visitation patterns in a Louisiana iris hybrid zone. Oikos 91: 568-578.

Enright, N.J., Goldblum, D., Ata, P. and Ashton D.H. (1997). The independent effects of heat, smoke and ash on emergence of seedlings from the soil seed bank of a healthy Eucalyptus woodland in Grampians (Gariwerd) National Park, western Victoria. Australian Journal of Ecology 22: 81-88.

Eriksson, 0. (1996). Regional dynamics of plants - a review of evidence for remnant, source-sink and metapopulations. Oikos 77: 248-258.

Escaravage, N., Questiau, S., Pornon, A., Doche, B. and Taberlet, P. (1998). Clonal diversity in a rhododendron Ferrugineum L. (Ericaceae) population inferred from AFLP markers. Molecular Ecology 7: 975-982.

Esselman, E.J., Jiangqiang, J, Crawford, D.J., Windus, J.L. and Wolfe, A.D. (1999). Clonal diversity in the rare Calamagrostis porteri spp insperata (Poaceae ): comparative results for allozymes and random amplified polymorphic DNA (RAPD) and intersimple sequence repeat (ISSR) markers. Molecular Ecology 8: 443-451.

Fallding, M., Kelly, A.H.H., Bateson, P., and Donovan, I. (2001). Biodiversity Planning Guide for NSW Local Government. NSW National Parks and Wildlife Service.

Farrier, D. (1996). Implementing the in-situ conservation provisions of the United Nations Convention on Biological Diversity in Australia: Questioning the role of national parks. The Australasian Journal of Natural Resources Law and Policy 3: 1-24

Farrier, D. (1999). Factoring biodiversity conservation into decision-making processes: the role of the precautionary principle. In Harding, R. and Fisher, E. (eds). Perspectives on the Precautionary Principle. The Federation Press, Sydney. Pp 99-1222. References 201

Farrier, D. (2002). Fragmented law in fragmented landscapes: the slow evolution of integrated natural resource management legislation in NSW. Environment and Planning Law Journal 19: 89-108

Farrier, D. and Tucker, L. (1998). Beyond a walk in the park: the impact of international nature conservation law on private land in Australia. Melbourne University Law Review 22: 565-270.

Farrier, D., Lyster R., and Pearson L. (1999). The Environmental Law Handbook. Redfern Legal Centre Publishing: Australia.

Farrier, D., Whelan. R. and Brown, C. (in press). Addressing scientific uncertainty in local government decision-making processes. Environment and Planning Law· Journal

Feinsinger, P., Beach, J.H., Linhart, Y.B., Busby, W.H. and Murray, K.G. (1987). Disturbance, pollinator predictability, and pollination success among Costa Rican cloud forest plants. Ecology 68: 1294-1305.

Ferdy, J.B., Austerlitz, F., Moret, J., Gouyon, P.H. and Godelle, B. (1999). Pollinator­ induced density dependence in deceptive species. Oikos 87: 549-560.

Ferdy, J., Gouyon, P., Moret, J. and Godelle, B. (1998). Pollinator behaviour and deceptive pollination: learning process and floral evolution. The American Naturalist 152: 696-703.

Ferdy, J.B., Loriot, S., Sandmeier, M., LeFranc, M. and Raquin, C. (2001). Inbreeding depression in a rare deceptive orchid. Canadian Journal of Botany 79: 1181-1188.

Fischer, M. and Matthies, D. (1997). Mating structure and inbredding and outbreeding depression in the rare plant Gentianella germanica (Gentianaceae). America Journal ofBotany 84: 1685-1692.

Fishbein, M. and Venable, D.L. (1996). Diversity and temporal change in the effective pollinators ofAsclepias tuberosa. Ecology 77: 1061-1073.

Forney, K.A. and Gipin, M.E. (1989). Spatial structure and population extinction: a study with Drosophila flies. Conservation Biology 3: 45-51.

Fowler, H.G., Pagani, M.I., Da Silva, O.A., Forti, L.C., Da Silva, V.P. and De Vasconcelos, H.L. (1989). Pest is a pest is a pest. The dilemma of neotropical leaf-cutting ants. Keystone taxa of natural ecosystems. Environmental Management 13: 671-675.

Frangos, J. (1999). Environmental science and the law. Environmental and Planning Law Journal 16: 175-181.

Franklin, J.P. 1(993). Preserving biodiversity: species, ecosystems, or landscapes? Ecological Applications 3: 202-205. References 202

Frankle, O.H., Brown, A.H.D., and Burdon, J.J. (1995). The Conservation Plant Biodiversity. Cambridge, University Press, Cambridge.

Frawley, K. (1994). Evolving vision: environmental management and nature conservation. In: Dovers, S. (ed) Australian environmental history: essays and cases Oxford University Press, Melbourne.

Gaston, K.J. (1996). What is biodiversity? In: Gaston, K.J. (ed) Biodiversity, A Biology ofNumbers and Difference, Blackwell Science, London, pp. 1-12.

Ghazoul, J. and McLeish, M. (2001). Reproductive ecology of tropical forests in logged and fragmented habitats in Thailand and Costa Rica. Plant Ecology 153: 335-345.

Gilfedder, L. and Kirkpatrick, J.B. (1998). Factors influencing the integrity of remnant bushland in subhumid Tasmania. Biological Conservation 84: 89-96.

Gill, A.M. and Bradstock, R. (1995). Extinction of biota by fires. In: Bradstock, R.A., Auld, T.D., Kingsford, R.T., Lunney, D. and Sivertsen, D.P. (eds) Conserving biodiversity: threats and solutions, Surrey Beatty & Sons, Chipping Norton, pp. 309-322.

Gill, A.M. and Williams, J.E. (1996). Fire regimes and biodiversity: the effects of fragmentation of southeastern Australian eucalypt forests by urbanisation, agriculture and pine plantations. Forest Ecology and Management 85: 261-278.

Gilmour, C.A., Crowden, R.K. and Koutoulis, A. (2000). Heat shock, smoke and darkness: partner cues in promoting seed germination in Epacris tasmanica (Epacridaceae ). Australian Journal ofBotany 48: 603-309.

Goldblatt, P., Bernhardt, P. and Manning, J.C. (2000). Adaptive radiation of pollination mechanisms in Ixia (Iridaceae: Crocoideae ). Annals of the Missouri botanical Garden 87: 564-577.

Goldingay, R.L. and Carthew, S.M. (1998). Breeding and mating systems of Australian Proteaceae. Australian Journal of Botany 46:421-437.

Goldingay, R.L., Schibeci, S.M. and Walker, B.A. (1991). Breeding system and pollination levels of Banksia ericifolia. Australian Journal of Botany 36: 365-372.

Goulson, D., Stout, J.C., Rawson, S.A. and Allen, J.A. (1998). Floral display size in comfrey Symphytum officinale L. (Boraginaceae ): relationaships with visitation by three bumblebee species and subsequent seed set. Oecologia 113: 502-508.

Griffith, M.B., Perry, S.A. and Perry, W.A. (1995). Macroinvertebrate communities in head water streams affected by acidic precipitation in the central Appalachians. Journal of Environmental Quality 24: 233-238.

Gross, K.L. (1981). Predictions of fate from rosette size in in four 'biennial' plant species: Verbascum thapsus, Oenothera biennis, Daucus carota and Tragopogon dubius. Oecologia 48: 209-213 References 203

Gross, C.L. (1990). The breeding systems of three co-occurring legumes: Dillwynia hispida, D. uncinata and Pultenaea densifolia (Legumunisae: Papilionoideae) Australian Journal of Botany 38: 207-215.

Gross, C.L. (1993). The breeding system and pollinators of Melastoma affine (Melastomataceae); a pioneer shrub in tropical Australia. Biotropica 25: 468-474.

Gross, C.L. (2001). The effects of introduced honey bees on native bee visitation and fruit set in Dillwynia juniperina (Fabaceae) in a fragmented ecosystem. Biological Conservation 102: 89-95.

Gross, C.L. and Mackay, D. (1998). Honeybees reduce fitness in the pioneer shrub Melastome affine (Melastomataceae). Biological Conservation 86: 169-178.

Grumbine, R.E. (1994). What is ecosystem management? Conservation Biology 8:27-38

Grumbine, R.E. (1997). Reflections on "what is ecosystem management?" Conservation Biology 11: 41-47.

Gullett, W. (1997). Environmental protection and the "precautionary principle": a response to scientific uncertainty in environmental management. Environment and Planning Law Journal 14: 52-69.

Gunter, A. (1994). Seed bank dynamics - longevity, viability and predation of seeds of serotinous plants in the central Namib Desert. Journal of Arid Environments 28: 195-205.

Haddad, N.M. (1999). Corridor and distance effects on interpatch movements: a landscape experiment with butterflies. Ecological Applications 9: 612-622.

Haddad, N. (2000). Corridor length and patch colonization by butterfly, Junonia coenia. Conservation Biology 14: 738-745.

Haddad, N.M. and Baum, K.A. (1999). An experimental test of corridor effects on butterfly densities. Ecological Applications 9: 623-633.

Haddad, N.M., Rosenberg, D.K. and Noon, B.R. (2000). On experimentation and the study of corridors: response to Beier and Noss. Conservation Biology 14: 1543- 1545.

Harden, G.J. (1991). Flora of New South Wales, Volume 2. New South Wales University Press, Kensington: Australia.

Harden, G.J. (1992). Flora of New South Wales, Volume 3. New South Wales University Press, Kensington: Australia.

Harder, L.D. (1998). Pollen-size comparisons among animal pollinated angiosperms with different pollination characteristics. Biological Journal of the Linnean Society 64: 513-525. References 204

Harder, L.D. and Barclay, M.R. (1994) The functional significance of poricidal anthers and buzz pollination: controlled pollen removal from Dodecatheon. Functional Ecology 8: 509-517.

Harder, L.D. and Barrett, S.C.H. (1995). Mating costs of large floral displays m hermaphrodite plants. Nature 373: 512-515.

Harding, E. and Fisher, R. (1999). The precautionary principle: towards a deliberative, transdisciplinary problem-solving process. In: Harding, E. and Fisher, R. (eds) Perspectives on the Precautionary Principle, The Federation Press, N.S. W. pp. 290-298

Hardy, P.B. and Dennis, R.L.H. (1999). The impact of urban development on butterflies within a city region. Biodiversity and Conservation 8: 1261-1279.

Harper, J.L. (1977). The Population Biology ofPlants. London: Academic Press.

Harrington, G.N. and Driver, M.A. (1995). The effect of fire and ants on the seed-bank of a shrub in a semi-arid grassland. Australian Journal ofEcology 20: 538-547.

Harris, L.B. and Scheck, J. (1991). From implications to applications: the dispersal corridor principle applied to the conservation of biological diversity. In Saunders, D.A. and Hobbs, R.J. (eds). The role of corridors. Surrey Beatty, Chipping Norton, NSW: Australia.

Harrison, J., Moller, M. and Cronk, Q.C.B. (1999). Evolution and development of floral diversity in Streptocarpus and Saintpaulia. Annals of Botany 84: 49-60

Hartcher, C.P. (1993). Parliamentary Debates, Legislative Assembly, Hansard, 20 May 1993, p. 2482.

Hendrix, S.D. and Kyhl, J.F. (2000). Population size and reproduction in Phlox pilosa. Conservation Biology 14:304-313.

Hermanutz, L., Innes, D., Denham, A. and Whelan, R. (1998). Very low fruit-flower ratios in Grevillea (Proteaceae) are independent of breeding system. Australian Journal ofBotany 46: 465-478.

Hirose T. and Kachi N. (1986). Graphical analysis of the life history evolution of Oenothera glazioviana. Oecologia 68: 490

Hobbs, R.J. (1992) The role of corridors in conservation: solution or bandwagon? Trends in Ecology and Evolution 7: 389-392.

Hogendoorn, K., Steen, Z. and Schwarz, M.P. (2000). Native Australian carpenter bees as a potential alternative to introducing bumble bees for tomato pollination in greenhouses. Journal ofApicultural Research 39: 67-74. References 205

Hogbin, P.M. and Peakall R. (1999). Evaluation of the contribution of genetic research to the management of the endangered plant Zieria prostrata. Conservation Biology 13: 514-522.

Hogbin, P.M., Ayre, D.J. and Whelan, R.J. (1998). Genetic variation and reproductive success on road verge populations of the rare shrub Grevillea barklyana (Proteaceae). Heredity 80: 180-186.

Holling, C.S. (1996). Resilience of ecosystems; local surprise and global change. In: Clark, W.C. and Munn, R.E. (eds) Sustainable development of the biosphere, Cambridge University Press, Cambridge. Pp. 292 - 317.

Holling, C.S., Berkes, F. and Folke, C. (1998). Science, sustainability and resource, management. In: Berkes, F and Folke, C. (eds) Linking Social and Ecological Systems: Management Practices and Social Mechanisms for Building Resilience, Cambridge University Press, Cambridge. pp. 342-362.

Holms, S. and Harris, J. (1999). Precautionary principle stifles discovery. Nature 400: 398.

Hoskins, K. and Turner, V.B. (1999). Resource use and foraging patterns of honeybees, Apis mellifera, and native insects on flowers of eucalyptus costata. Australian Journal of Ecology 24: 221-227.

Houland, N., Andreassan, H.P. and Ims, R.A. (1999). Foraging behaviours of the root vole Microtus oeconomus in fragmented habitats. Oecologia 121: 236-244.

Jennersten, 0. (1988). Pollination in Dianthusdeltoides caryophyllaceae effects of habitat fragmentation on visitation and seed set. Conservation Biology 2 359-366

Johnson, S.D. and Steiner, K.E. (2000). Generalisation versus specialisation in plant pollination systems. Trends in Ecology and Evolution 15 140-143.

Johnston, M.O. (1992). Effects of cross and self fertilization on progeny fitness in Lobelia cardinalis and L. siphilitica. Evolution 46: 688-702.

Kalinganire, A., Harwood, C.E., Slee, M.U. and Simons, A.J. (2000). Floral structure, stigma receptivity and pollen viability in relation to protandry and self - incompatibility in silky oak (Grevillea robusta A. Cunn.). Annals of Botany 86: 133-148.

Karron, J.D., Tucker, R., Thumser, N.N. and Reinartz, J.A. (1995). Comparison of pollinator flight movements and gene dispersal patterns in Mimulus ringens. Heredity 75: 612-617.

Kearns, C.A. (2001). North American dipteran pollinators: assessing their value and conservation status. Conservation ecology 5 :5 [online] URL: http://www.consecol.org/volS/issl/artS References 206

Kearns, C.A. and Inouye, D.W. (1993). Techniques for Pollination Biologists. University Press of Colorado, Boulder, Co.

Kearns, C.A. and Inouye, D.W. (1997). Pollinators, flowering plants and conservation biology. BioScience 47: 297-306.

Kearns, C.A., Inouye, D.W. and Waser N.M. (1998). Endangered mutualisms: the conservation of plant-pollinator interactions. Annual Review of Ecology and Systematics 29: 83-112.

Kenny, B.J. (2000). Influences of multiple fire-related germination cues on three Sydney Grevillea (Proteaceae) species. Austral Ecology 6: 664-669.

Keith, D. (1996). Fire driven extinction of plant populations: a synthesis of theory and review of evidence from Australian vegetation. Proceedings of the Linnean SocietyofNSWll6: 37-78.

Kelly, A. H. H. (1994). Protecting endangered fauna in NSW: informed habitat destruction. The Australasian Journal of Natural Resources Law and Policy 1: 69- 105.

Kelly, A.H.H. (1995). Land use planning in New South Wales: its evolution and current trends. Asia Pacific Law Review 4: 69-90.

Kelly, A. H. H. (1996). New mandates for protecting threatened species, populations and ecological communities at the local government level in New South Wales. Local Government Law Journal 2:78-94.

Kelly, A.H. and Farrier, D. (1996). Local government and biodiversity conservation in New South Wales. Environmental and Planning Law Journal 13: 374-389.

Kelly, D., Ladley, J.J., Robertson, A.W. and Norton, D.A. (2000). Limited forest fragmentation improves reproduction in the declining New Zealand mistletoe Peraxilla tetrapetala (Loranthaceae ). In: Genetics, Demography and Viability of Fragmented Populations edited by Young A.G and Clarke, G.M. published by Cambridge University Press, Cambridge.

King, M.J. and Buchmann, S.L. (1996). Sonication dispensing of pollen of Solanum laciniatum flowers. Functional Ecology 10: 449-456.

King, M.J. and Ferguson, A.M. (1994). Vibratory collection of Actinidia deliciosa (kiwifruit) pollen. Annals ofBotany 74: 479-482.

King, S.A. and Buckney, R.T. (2000). Urbanization and exotic plants m northern Sydney streams. Austral Ecology 25: 455-461.

Kronfrost, M.R. and Fleming, T.H. (2001). Lack of genetic differentiation among widely spread subpopulations of a butterfly with home range behaviour. Heredity 86: 243-250. References 207

Knapp, E.E., Goedde, M.A. and Rice, K.J. (2001). Pollen-limited reproduction in blue oak - implications for wind pollination in fragmented populations. Oecologia 128: 48-55

Knudsen, J.T. and Olesen, J.M. (1993). Buzz-pollination and patterns in sexual traits in north European Pyrolaceae. American Journal of Botany 80: 900-913.

Krauss, S.L. (1994). Restricted gene flow within the morphologically complex species Persoonia mollis (Proteaceae): contrasting evidence from the mating system and pollen dispersal. Heredity 73: 142-154.

Krauss, S.L. (2000). Patterns of mating in Persoonia mollis (Proteaceae) revealed by an analysis of paternity using AFLP: implications for conservation. Australian Journal ofBotany 48: 349-356.

Krauss, S.L. and Peakall, R. (1998). An evaluation of the AFLP fingerprinting technique for the analysis of paternity in natural populations of Persoonia mollis (Proteaceae ). Australian Journal of Botany 46: 533-546.

Krebs, C.J. (1989). Ecological Methods. Harper Collins Publishers, Inc.

Lambert, M.J. and Turner, J. (1987). Suburban development and change in vegetation nutritional status. Australian Journal ofEcology 12: 193-196.

Lamont, B.B. and van Leeuwen, S.J. (1988). Seed production and mortality in a rare Banksia species. Journal ofApplied Ecology 25: 551-559.

Larson, B.M.H. and Barrett, S.C.H. (1999a). The ecology of pollen limitation in buzz­ pollinated Rhexia virginica (Melastomataceae). Journal of Ecology 87: 371-381.

Larson, B.M.H. and Barrett, S.C.H. (1999b ). The pollination ecology of buzz-pollinated Rhexia virginica (meiastomataceae).AmericanJournalof Botany 86: 502-511.

Law, B., Mackowski, C., Schaer, L. and Tweedie, T. (2000). Flowering phenology of myrtaceous trees and their relation to climatic, environmental and disturbance variables in northern New South Wales.Austral Ecology 25: 160-178.

Lee, T.D. and Bazzaz, P.A. (1982). Regulation of fruit maturation pattern in an annual legume, Cassia fasciculate. Ecology 63: 1374-1388.

Leishman, M.R. (1990). Suburban development and resultant changes in the phosphorus status of soils in the area of Ku-Ring-Gai, Sydney. Proceedings of the Linnean Society ofN.S. W. 112: 15-25.

Letnic, M., Dickman, C.R. and McNaught, G. (2000) Bet-hedging and germination in the Australian arid zone shrub Acacia ligulata. Austral Ecology 25: 368-3 7 4.

Levin, D.A. (1984). Inbreeding depression and proximity-dependent crossing success in Phlox drummondii. Evolution 38: 116-127. References 208

Lim, L. (1996). NSW Threatened Species Conservation Act 1995: implications for planners. Australian Planner 33: 172-174.

Lim, L. (1997). The 10 lords of the universe - New South Wales TSC Act's scientific committee. Pacific Conservation Biology 3: 4-12.

Lindenmayer, D.B., Cunningham, R.B. and Donnelly, C.F. (1993). The conservation of arboreal marsupials in the montane ash forests of the central highlands of Victoria, south-eastern Australia. IV. The presence and abundance or arboreal marsupials in retained linear habitats (wildlife corridors) within logged forests. Biological Conservation 66: 207-221.

Lippok, B., Gardine, A.A., Williamson, P.S. and Renner, S.S. (2000). Pollination by flies, bees and beetles of Nuphar ozarkana and N. advena (Nymphaeaceae ). American Journal of Botany 87: 898-902.

Lovich, J.E. and Bainbridge, D. (1999). Anthropogenic degradation of the southern California desert ecosystem and prospects for natural recovery and restoration. Environmental Management 24: 309-326.

Lunney, D., dawson, L. and Law, B. (1998). Is the biodiversity tail wagging the zoologcial dog? Royal Zoological Society of New South Wales, Sydney.

Lunney, D., Matthews, A., Moon C. and Ferrier, S. (2000). Incorporating habitat mapping into practical koala conservation on private lands. Conservation Biology 14: 669-680.

Lunt, ID (1997). Effects of long-term vegetation management on remnant grassy forests and anthropogenic native grasslands in south-eastern Australia. Biological Conservation 81: 287-297.

Lunt, I.D. and Morgan, J.W. (1999). Vegetation changes after 10 years of grazing exclusion and intermittent burning in a Themeda triandra (Poaceae) grassland reserve in south-eastern Australia. Australian Journal of Botany 47: 537-552.

Lyons, K.G. and Schwartz, M.W. (2001). Rare species loss alters ecosystem function - invasion resistance. Ecology Letters 4: 358-365.

MacGarvin, M. (1999). The precautionary principle, science and policy. In: Harding, R. and Fisher, E. Perspectives on the Precautionary Principle. The Federation Press, Sydney. Pp 225-239.

Major, R.E., Gowing, G. and Kendal, C.E. (1996). Nest predation in Australian urban environments ad the role of the pied currawong, Strepera graculina. Australian Journal of Ecology 21:399-409.

Mamouney, L. (2000). Should local government be responsible for biodiversity management? A critical review of local government's ability to manage biodiversity loss in NSW through the development process. Environmental and Planning Law Journal 17: 138-150. References 209

Mann, C.C. (1991). Extinction. Are ecologists crying wolf? Science 253: 736-738.

Mann, C.C. and Plummer, M.L. (1995). Are wildlife corridors the right path? Science 270: 1428-1430.

Marshall, D.L. (1986). Effect of seed size on seedling success in three species of Sesbania (Fabaceae ). American Journal of Botany 73: 457-464.

Martin, T.G. and Catterall C.P. (2001). Do fragmented coastal heathlands have habitat value to birds in eastern Australia. Wild/ife research 28: 17-31.

Maryott-Brown, K. and Wilks, D. (1993). Rare and endangered plants of Yengo National Park and adjacent area. Report for National Parks and Wildlife Service.

Mataumura, C. and Washitani, I. (2000). Effects of population size and pollinator limitation on seed set of Primula sieboldii populations in a fragmented landscape. Ecological Research 15: 307-322.

Matheson, A., Buchmann, S.L., O'Toole, C., and Williams, l.H. (eds) (1996). The conservation of bees. Academic Press, Harcourt Brace, London.

McNeely, J.A. (1992). The sinking ark: pollution and the worldwide loss of biodiversity. Biological Conservation l: 2-18.

Meney, K.A., Nielssen, G.M. and Dixon, K.W. (1994). Seed bank patterns in Restionaceae and Epacridaceae after wildfire in Kwongan in south-western Australia. Journal of Vegetation Science 5: 5-12.

Menges, E.S. (1990). Seed germination percentage increases with population size in a fragmented prairie species. In: Ehenfold, D. (ed) Readings from Conservation Biology. Blackwell Science, U.S.A.

Meurling, R. (1999). Look before you leap - the precautionary principle. Local Government Law Journal 5: 27-41

Mills, L.S., Soule, M.E. and Doak, D.F. (1993). The keystone species concept in ecology and conservation. BioScience 43: 219-224.

Milton, D.A. (2001). Assessing susceptibility to fishing of populations of rare trawl by catch: sea snakes caught by Australia's Northern prawn fishery. Biological Conservation 101: 281-290.

Mokany, K. and Adam, P. (2000). The biogeographical attributes of the threatened flora of New South Wales. Cunninghamia 6: 873-892.

Moore, P.D. (1996). The buzz about pollination. Nature 384: 27. References 210

Morgan, J.W. (1995). Ecological studies of the endangered Rutidosis leptorrhynchoides. I seed production, soil seed bank dynamics, population density and their effects on recruitment. Australian Journal of Botany 43: 1-11.

Morgan, J.W. (1999). Effects of population size on seed production and germinability in an endangered fragmented grassland plant. Conservation Biology 13: 266-273.

Moro, M.C.D. and Pinherio, M.C.B. (1999). Pollination ecology of Swartzia apetala Raddi var. apetala (Leguminosae: Papilionoideae ). Brazilian Archives of Biology and Technology 42: 415-423.

Morton, S.R., Smith, D.M.S., Friedel, M.H., Griffin, G.F. and Pickup, G. (1995). The stewardship of arid Australia: ecology and landscape management. Journal of Environmental Management 43: 195.

Morris, E.C. (2000). Germination response of seven east Australian Grevillea species (Proteaceae) to smoke, heat exposure and scarification. Australian Journal of Botany 48: 179-189.

Murcia, C. (1996). Forest fragmentation and the pollination of neotropical plants. In Schelhas, J. and Greenberg, R. (eds) Forest patches in tropical landscapes. Island Press, Washington, DC. Pp19-36.

Mustajarvi, K., Siikamaki, P., Rytkonen, S. and Lammi, A. (2001). Consequences of plant population size and density for plant-pollinator interactions and plant performance. Journal ofEcology 89: 80-87.

Nason, J.D. and Hanrick, J.L. (1997). Reproductive and genetic consequences of forest fragmentation: two case studies of neotropical canopy trees. Journal of Heredity 88: 264-276.

Neal, P.R. (1998). Pollinator restoration. Trends in Ecology and Evolution 13: 132-133.

Niemela, J. (1999). Ecology and urban planning. Biodiversity and Conservation 8: 119- 131.

Norton, D.A. and Reid, N. (1997). Lessons in ecosystem management from management of threatened and pest Loranthaceous mistletoes in New Zealand and Australia. Conservation Biology 11: 759-769.

Noss, R.F. (1987). Corridors in real landscapes: a reply to Simberloff and Cox. Conservation Biology 1: 159-164.

Nour, N., Matthysen, E. and Dhondt, A.A. (1997). Effects of habitat fragmentation on foraging behaviour of tits and related species: does niche space vary in relation to size and degree of isolation of forest fragments? Ecography 20: 281-286.

O'Connell, A.M., Grove, T.S., and Dimmack, G.M. (1979). The effects of a high intensity fire on nutrient cycles in jarrah forest. Australian Journal of Ecology 4: 331-337. References 211

O'Dwyer, C. and Attiwill, P.M. (1999). A comparative study of habitats of the Golden sun moth Synemon plana Walker (Lepidoptera: castniidae): implications for restoration. Biological Conservation 89: 131-141.

Okland, R.H. (1995). Population biology of the clonal moss Hylocomium splendens in Norwegian boreal spruce forests. 1. Demography. Journal of Ecology 83: 697- 712.

Paton, D.C. (1993). Honeybees in the Australian environment: does Apis mellifera disrupt or benefit the Australian biota. BioScience 43: 95-103.

Paton, D.C. (1997). Honey bees Apis mellifera and the disruption of plant-pollinator systems in Australia. The Victorian Naturalist 114: 23-29.

Paton, D.C. and Turner, V. (1985). Pollination of Banksia ericifolia Smith: birds, mammals and insects as pollen vectors. Australian Journal of Botany 33: 271- 286.

Pavlik, B.M., Ferguson, N. and Nelson, M. (1993). Assessing limitations on the growth of endangered plant populations. II. Seed production and seed bank dynamics of Erysimum capitatum ssp. angustatum and Oenothera deltoids ssp. howellii. Biological Conservation 65: 267-278.

Payne, R. (1998). Lake Macquaire Tetratheca juncea conservation management plan. Prepared for NPWS, BHP PTY Ltd and Lake Macquaire City Council by ES & M. Interim report.

Peakall, R. and Smouse, P.E. (2001). GenAIEx VS: Genetic Analysis in Excel. Population genetic software for teaching and research. Australian National University, Canberra, Australia. http://www.anu.edu.au/BOZO/GenAlEx

Persson, H.A. and Gustavsson, B.A. (2001). The extent of clonality and genetic diversity in lingonberry (Vaccinium vitis-idaea L.) revealed by RAPDs and leaf shape analysis. Molecular Ecology 10: 1385-1397.

Pierce, S.M. and Cowling, R.M. (1991). Dynamics of soil-stored seed banks of six shrubs in fire-prone dune fynbos. Journal ofEcology 79: 731-747.

Pierre, J.P. (1999). Reintroduction of the South Island saddleback (Philesturnus carunculatus carunculatus): dispersal, social organisation and survival. Biological Conservation 89: 153-159.

Pornon, A. and Escaravage, N. (1999) Genotypic structure in clonal Rhododendron ferrugineum L. (Ericaceae) populations: origin and maintenance. Plant Ecology 141: 145-150.

Pornon A, Escaravage N, Thomas P and Taberlet P. (2000). Dynamics of genotypic structure in clonal Rhododendron ferrugineum (Ericaceae) populations. Molecular Ecology 9: 1099-1111. References 212

Powledge, F. (1998). Biodiversity at the crossroads. BioScience 48: 347-353.

Pressey, B. (1995). Conservation reserves in N.S.W.: Crown jewels or leftovers. Search 26: 47-81.

Prest, J. (1995). Licensed to kill: endangered fauna licensing under the national parks and wildlife act 1974 (NSW) between 1991-1995. Published by the Australian Center for Environmental Law. ANU: A.C.T.

Priddle, D. and Carlile, N. (2001). A trial translocation of Gould's petrel (Pterodroma leucoptera leucoptera). Emu 101: 79-88.

Primack, R.B. (1987). Relationships among flowers, fruits and seeds. Annual Review of Ecology and Systematics 18: 409-430.

Ramsey, M.W. (1988). Differences in pollinator effectiveness of birds and insects visiting Banksia menziesii (Proteaceae ). Oecologia 76: 119-124.

Ramsey, M. and Vaughton, G. (2000). Pollen quality limits seed set in Burchardia umbellate (Colchicaceae). America Journal of Botany 87: 845-852.

Randell, A. (1991). The value ofbiodiversity.AMB/O 20: 64-67.

Rathcke B.J. and Jules, E. S. (1993). Habitat fragmentation and plant pollinator interactions. Current Science 65: 273-277.

Read, T.R., Bellairs, S.M., Mulligan, D.R. and Lamb, D. (2000). Smoke and heat effects on soil seed bank germination for the re-establishment of a native forest community in New South Wales. Austral Ecology 25: 48-57.

Recher, H.F. (1997). Conservation priorities: myths and realities. Pacific Conservation Biology 3: 81.

Recher, H.F. (1999). The precautionary principle. Pacific Conservation Biology 5: 81.

Renner, S.S. (1996). Effects of habitat fragmentation on plant pollinator interactions in the tropics. In: Newbery, D.M., Prins, H.T. and Brown, N.D. (eds). Dynamics of Tropical Communities. Blackwell Scientific, Cambridge: UK. Pp 339-361.

Richards, S.A., Possingham, H.P. and Tizard, J. (1999). Optimal fire management for monitoring community diversity. Ecological Applications 9: 880-892.

Richard-Hansen, C., Vie, J.C. and de Thoisy B. (2000). Translocation of red howler monkeys (Alouatta seniculus) in French Guiana. Biological Conservation 93: 247- 253.

Richardson, M.B.G., Ayre, D.J. and Whelan, R.J. (2000). Pollinator behaviour, mate choice and the realised mating systems of Grevillea mucronulata and Grevillea sphacelata. Australian Journal ofBotany 48: 357-366. References 213

Risser, P. G. (1995). Biodiversity and ecosystem function. Conservation Biology 9: 742-746.

Robertson, A.W., Mountjoy, C., Faulkner, B.E., Roberts, M.V. and MacNair, M.R. (1999). Bumble bee selection of Mimulus guttatus flowers: the effects of pollen quality and reward depletion. Ecology 80: 2594-2606.

Rose, S. (1997). Influences of suburban edges on invasion of Pittosporum undulatum into the bushland of northern Sydney, Australia. Australian Journal of Ecology 22: 89-99.

Rose, S. and Fairweather, P.G. (1997). Changes in floristic composition of urban bushland invaded by Pittosporum undulatum in northern Sydney, Australia. Australian Journal ofBotany 45: 123-149.

Rosenberg, D.K., Noon, B.R. and Meslow, E.C. (1997). Biological corridors: form, function, and efficacy. BioScience 47: 677-687.

ROTAP (1995). Rare or threatened Australian plants. Briggs, J.D. and Leigh, J.H. (eds). CSIRO: Australia.

Sampson, J.F., Collins, B.G. and Coates, D.J. (1994). Mixed mating in Banksia brownii Baxter ex R. Br. (Proteaceae). Australian Journal of Botany 42: 103-111.

Santillo, D., Stringer, R.L., Johnston, P.A. and Tickner, J. (1998). The precautionary principle: protecting against failures of scientific metod and risk assessment. Marine Pollution Bullentin 36: 939-950

Saville, N.M., Dramstad, W.E., Fry, G.L.A. and Corbet, S.A. (1997). Bumblebee movement in a fragmented agricultural landscape. Agriculture, Ecosystems and Environment 61: 145-154.

Scariot, A. (1999). Forest fragmentation effects on palm diversity in central Amazonia. Journal of Ecology 87: 66-76.

Schmidt-Adam, G., Young, A.G. and Murray, B.G. (2000). Low outcrossing rates and shift in pollinators in New Zealand pohutukawa (Metrosideros excelsa; M yrtaceae ). America Journal of Botany 87: 1265-1271.

Sedgley, M. and Harbard, J. (1993). Pollen storage and breeding system in relation to controlled pollination of four species of Acacia (Leguminosae: Mimosoideae). Australian Journal ofBotany 41: 601-609.

Shachak, M., Sachs, M. and Moshe I. (1998). Ecosystem management of desertified shrublands in Israel. Ecosystems 1: 475-485.

Shore, J.S. and Barrett, S.C.H. (1984). The effect of pollination intensity and incompatiable pollen on seed set in Turnera ulmifolia (Turneraceae ). Canadian Journal of Botany 62: 1298-1303. References 214

Short, J. and Smith, A. (1994). Mammal decline and recovery in Australia. Journal of Mammalogy 75: 288-297.

Short, J. and Turner, B. (2000). Reintroduction of the burrowing bettong Bettongia lesueur (Marsupialia: Potoridae) to mainland Australia. Biological Conservation 96: 185-196.

Simberloff, D. and Cox, J. (1987). Consequences and costs of conservation corridors. Conservation Biology 1: 63-71.

Simberloff, D., Farr, J.A., Cox, J. and Mehlman, D.W. (1992). Movement corridors: conservation bargains or poor investments. Conservation Biology 6: 493-504.

Smith, J. (1997). Skinning cats, putting tigers in tanks and bringing up baby: a critique of the Threatened Species Conservation Act 1995 (NSW). Environmental and Planning Law Journal 14: 17-37.

Smith, P.J., Pressey, R.L. and Smith, J.E. (1994). Birds of particular conservation concern in the western division of New South Wales. Biological Conservation 69: 315-338.

Smith and Smith (1998). Interim report on the vegetation survey of Warringah bushland. For Warringah Council.

Snow, A.A., Spira, T.P., Simpson, R. and Klips, R.A. (1996). The ecology of geitonogamous pollination. In Lloyd, D.G. and Barrett, S.C.H. (eds). Floral biology. Studies on floral evolution in animal-pollinated plants. Chapman and hall, New York, New York, U.S.A.

Soule, M.E., Alberts, A.C. and Bolger, D.T. (1992). The effects of habitat fragmentation on chaparral plants and vertebrates. OIKOS 63: 39-47.

Sperling, K. (1999). If caution really mattered. Environmental and Planning Law Journal 16: 425-440.

Stanley, T.R. (1995). Ecosystem management and the arrogance of humanism. Conservation Biology 9: 255-262.

State of the Parks (2001). An overview of the conservation values of NSW and their management within the parks. NSW National Parks and Wildlife Services, Hurstville, NSW: Australia.

Steffan-Dewenter, I. and Tscharntke, T. (1999) Effects of habitat isolation on pollinator communities and seed set. Oecologia 121: 432-440.

Stein, P.L. (2000). Are decision-makers too cautious with the precautionary principle? Environmental and Planning Law Journal 17: 3-23. References 215

Steininger, M.K., Tucker, C.J., Townshend, J.R.G., Killeen, T.J., Desch, A., Bell, V. and Ersts, P. (2001). Tropical deforestation in the Bolivian Amazon. Evironmental Conservation 28: 127-134.

Stewart, A (1992). Environmental evidence: a conflict of paradigms. In: Lunney, D. (ed). Zoology in Court. Royal Zoological Society of NSW, Mosman: Australia. Pp 59-69.

Stratford, E., Mazur, N., Lunney, D. and Bennett, D. (2000). Managing the koala problem: interdisciplinary perspectives. Conservation Biology 14: 610-618.

Suarez, A.V., Bolger, D.T. and Case, T.J. (1998). Effects of fragmentation and invasion on native ant communities in coastal southern California. Ecology 79: 2041-2056.

Swenson, J.J. and Franklin, J. (2000). The effects of future urban development on habitat fragmentation in the Santa Monica Mountains. Landscape Ecology 15: 713-730.

Sydes, M.A. and Peakall, R. (1988). extensive clonality in the endangered shrub Haloragodendron lucasii (Haloragaceae) revealed by allozymes and RAPDs. Molecular Ecology 7: 87-93.

Tabarelli, M., Mantovani, W. and Peres, C.A. (1999). Effects of habitat fragmentation on plant guild structure in the montane Atlantic forest of southeastern Brazil. Biological Conservation 91: 119-127.

Thompson, J. (1976). A revision of the genus Tetratheca (Tremandraceae). Telopea 1: 139-215

Thompson, K. and Grime, J.P. (1979). Seasonal vanat10n in the seed banks of herbaceous species in ten contrasting habitats. Journal of Ecology 67: 893-921.

Thomson, J.D. (2001). Using pollination deficits to infer pollinator declines: can theory guide us? Conservation Ecology 5:6. [online] URL: http://www.consecol.org/vol5/issl/art6

Thorp, R.W. (2000). The collection of pollen by bees. Plant Systematics and Evolution 222: 211-223.

Tomimatsu, H. and Ohara, M. (2002). Effects of forest fragmentation on seed production of the understory herb Trillium camschatcense. Conservation Biology 16: 1277-1285.

Tozer, M.G. (1998). Distribution of the soil seed bank and influence of fire on seedling emergence in Acacia saligna growing on the central coast of New South Wales. Australian Journal ofBotany 46: 743-755.

Travis, S.E., Maschinski, J. and Keim, P. (1996). An analysis of genetic variation in Astragalus cremnophylax var. cremnophylax, a critically endangered plant, using AFLP markers. Molecular Ecology 5: 735-745. References 216

Vakin, Y., Gan-mor, S., Bechar, A., Ronen, B. and Eisikowitch, D. (2000). The role of electrostatic forces in pollination. Plant Systematics and Evolution 222: 133-142.

Vasseur, L., Fortin, M.J. and Cyr, J. (1998). Clover and cress as indicator species of impacts from limed sewage sludge and landfill waste water land application. Science of the Total Environment 217: 231-239.

Vaughton, G. (1988). Pollination and seed set of Banksia spinulosa: evidence for autogamy. Australian Journal of Botany 36: 633-642.

Vaughton, G. (1990). Seasonal variation in honeyeater foraging behaviour, inflorescence abundance and fruit set in Banksia spinulosa (Proteaceae ). Australian Journal of Ecology 15: 109-116.

Vaughton, G. (1991). Variation between years in pollen and nutrient limitation of fruit­ set in Banksia spinulosa. Journal ofEcology 78: 389-400.

Vaughton, G. (1992). Effectiveness of nectarivorous birds and honey bees as pollinators of Banksia spinulosa (Proteaceae ). Australian Journal ofEcology 17: 43-50.

Vaughton, G. (1996). Pollination disruption by European honey bees in the Australian bird-pollinated shrub Grevillea barklyana (Proteaceae ). Plant Systematics and Evolution 200: 89-100.

Vaughton, G. (1998). Soil seed bank dynamics in the rare obligate seedling shrub, Grevillea barklyana (Protecaeae ). Australian Journal ofEcology 23: 375-384.

Vaughton, G. and Ramsey, M. (1991). Floral biology and inefficient pollen removal in Banksia spinulosa var. neoanglica (Proteaceae). Australian Journal of Botany 39: 167-177.

Walther-Hellwig, K. and Frankl, R. (2000). Foraging habitats and foraging distances of bumblebees, Bombus spp. (hym., apidae), in an agricultural landscape. Journal of Applied Entomology 124: 299-306.

Wang, L. (1997). The soil seed bank and understorey regeneration in Eucalyptus regnans forest, Victoria. Australian Journal ofEcology 22: 404-411.

Weins, D., Calvin, C.L., Wilson, C.A., Darren, C.I., Frank, D., and Seavey, S.R. (1987). Reproductive success, spontaneous embryo abortion, and genetic load in flowering plants. Oecologia 71: 501-509.

Weiss, S.B. (1999). Cars, cows and checkerspot butterflies: nitrogen deposition and management of nutrient poor grasslands for a threatened species. Conservation Biology 13: 1476-1486.

West, G.J. (1991). Parliamentary Debates, Legislative Assembly, Hansard, 12 December 1991, p. 6647. References 217

Western, D. (1992). The biodiversity crisis: a challenge for biology. OIKOS 63: 29-38.

Whelan, R.J. (2003). Adaptive management: What does it mean and how can it be used in fire management? In: Halse, S. (ed.) Bushfires: Managing the Risk. Nature Conservation Council, Sydney.

Whelan, R.J. and Goldingay, R.L. (1986). Do pollinators influence seed-set in Banksia paludosa Sm. and Banksia spinulosa R. Br.? Australian Journal of Ecology 11: 181-186.

Whelan, R.J. and York, J. (1998). Does post-fire germination of Hakea sericea and Petrophile sessilis differ between spring and autumn burning? Australian Journal ofBotany 16: 367-376.

Williams, N.M., Minckley, R.L. and Silveira, F.A. (2001). Variation in native bee faunas and its implications for detecting community changes. Conservation Ecology 5: 7 [online] URL: http:l/www.consecol.org/vol5/issl/art7.

Wilson, E.O (1988). Biodiversity. National Academy Press, Washington DC: USA.

Wilson, J.D., Evans, J., Browne, S.J. and King, J.R. (1997). Territory distribution and breeding success of skylarks Alauda arvensis on organic and intensive farmland in southern England. Journal ofApplied Ecology 34: 1462-1478.

Wiser, S.K., Peet, R.K. and White, P.S. (1998). Prediction of rare-plant occurrence: A southern Appalachian example. Ecological Applications 8: 909-920.

Wisheu, LC. and Keddy, P.A. (1991). Seed banks of a rare wetland plant community: distribution patterns and effects of human-induced disturbances. Journal of Vegetation Science 2: 181-188.

Wolff, A.T., Harrison, S.P. and Hamrick, J.L. (2000). Influence of habitat patchiness on genetic diversity and spatial structure of a serpentine endemic plant. Conservation Biology 14: 454-463.

Woolfe, B. (1992). Corkill v Forestry Commission of New South Wales. Environmental and Planning Law Journal 9: 1-3.

Woolley, A. and Kirkpatrick, J.B. (1999). Factors related to condition and rare and threatened species occurrence in lowland humid basalt remnants in northern Tasmania. Biological Conservation 87: 131-142.

Worley, A.C., Baker, A.M., Thompson, J.D. and Barrett, S.C.H. (2000). Floral display in Narcissus: variation in flower size and number at the species, population and individual levels. International Journal ofPlant Sciences 161: 69-79.

Yates, CJ., Taplin, R., Hobbs, R.J. and Bell, R.W. (1995). Factors limiting the recruitment of Eucalyptus salmonophloia in remnant woodlands 2. post-dispersal seed predation and soil seed reserves. Australian Journal of Botany 43: 145-155. References 218

Young, A.G. and Brown, A.H.D. (1996). Comparative analysis of the mating system of the rare woodland shrub Daviesia suaveolens and its common congener D. mimosoides. Heredity 80: 374-381.

Young, A., Boyle, T. and Brown, T. (1996). The population genetic consequences of habitat fragmentation for plants. Trends in Ecology and Evolution 11: 413-418.

Young, A.G., Brown, A.H.D. and Zich, P.A. (1999). Genetic structure of fragmented populations of the endangered daisy Rutidosis leptprrhynchoides. Conservation Biology 13: 256-265.

Young, M.D. (1999). The precautionary principle as a key element of ecologically sustainable development. In: Harding, R. and Fisher, E. (eds). Perspectives on the Precautionary Principle. The Federation Press NSW: Australia. Appendix 2.1 History of the TSCA 219

APPENDIX 2.1 HISTORY OF THE THREATENED SPECIES CONSERVATION ACT 1995 (NSW)

Chaelundi State Forest is 7000 ha of old growth tallowwood Eucalyptus microcorys forest (Nth Washpool) containing one of the highest densities of arboreal marsupials recorded in Australia as well as a number of threatened species (Cohen 1997, Bailey 1991b & c). In March 1990, the North East Forest Alliance (NEFA) constructed a blockade to restrain the Commission from roading and logging the forest (Cohen 1997). An injunction to restrain the Commission from further activity was issued by the Land and Environment Court. The Commission was ordered by Justice Cripps (4 July 1990) to stop all work and prepare an Environmental Impact Statement (EIS) for the area. Hans Drielsma, the then Commissioner for Forests, declared the EIS to be the most detailed and thorough yet, prepared for the Commission (Bonyhady 1993). The EIS was placed on public display in November 1990. A number of weaknesses were exposed in the EIS particularly regarding the population size and abundance of threatened species within the forest. Despite the flaws in the EIS, logging of Chaelundi resumed in March 1991(Bonyhady1993; Woolfe 1992).

The 25 May 1991 saw the election of the Coalition government with the National Party to look after land-oriented ministries. Despite four months of monitoring by conservationists, the Commission decided to extend the road further into the State Forest. The forest was officially closed to the public, causing the blockade to be intensified (and the police to move, in complete view of the media) (Cohen 1997). This formed an integral part of the overall strategy of the conservation movement to bring to public and political awareness the plight of Australian threatened species. Simultaneously, Tim Robertson (a Sydney barrister) and John Corkill (a political activist for the NEFA) entered the Land and Environment Court, asserting that the Commission's logging activities were illegal.

In a landmark case, Corkill v Forestry Commission of New South Wales [71 LGRA 116 (1990)], the "applicant sought an interlocutory injunction to restrain further logging" in compartment 695 (an area of 395 ha) in Chaelundi State Forest. Historically, progressive logging had been occurring since 1982, under a non-site-specific EIS written in 1980. During late 1989, all logging activities were suspended, to enable consideration of the Appendix 2.1 History of the TSCA 220

alleged Aboriginal significance of the area and possible designation of a wilderness area (under the Wilderness Act 1987 (NSW)) (Woolfe 1992, 71 LGRA 115). A government announcement (September 1990) allowing logging activities to continue led to the involvement of the environmental movement and the court case. The court decided not only that the Commission was unlawful in its approval process but that the original assessment ignored many possible ecological impacts ( eg. soil degradation and erosion, threatened species), therefore the Commission was in breach of Part 51 of the EPAA. All logging activities were ordered to cease by Justice Stein (25 September 1991).

The Commission, concerned that the industry would be affected (Bailey 1991), filed an appeal, (Forestry Commission of New South Wales v Corkill (1991) 73 LGRA 247), which was unsuccessful. During the appeal process, consideration was given to whether any fauna would be "taken and/or killed" (s 98-99 NPWA). The Commission stated that it was perfectly acceptable for any fauna to be killed as it was an unintended but a necessary consequence of logging activities proposed under the Forestry Act 1916 (NSW)2. It was ruled that the Commission was in breach of s 98-99 of the National Parks and Wildlife Act 1979 (NSW) (Prest 1995). During this case, 'taking' was defined as including any significant modification to the habitat of a threatened species (Farrier 1993; Woolfe 1992). The case highlights the importance of understanding the application of the NPWA to the Forestry Act 1916 (NSW).

The Minister for the Environment (Tim Moore), sympathetic to the Commission, formulated a regulation, Fauna Protection Regulation 1991, which effectively overturned the Court's decision (Woolfe 1992; Cohen 1997). It was this regulation that led to a Parliamentary attack on the Premier and the resignation of Terry Metherall. Metherall who

1 EP AA, Part 5, is concerned with developments that do not require consent under Part 4 or other relevant Environmental Planning Instruments (EPis). Such developments include developments by public bodies or private development, which requires an approval other than development consent. When a significant impact is detected, through the environmental assessment process, then concurrence is required from NPWS or the environmental minister (EPAA s 76 & s 112).

2 s ll(l)a of the Forestry Act 1916 (NSW) allows for logging in a controlled and managed manner in State Forests "in such a manner as best serves the public interests ..... ". It is also under this Act that the Commission was allowed to grant licences for their own activities. Appendix 2.1 History of the TSCA 221

had grown disillusioned with the liberals environmental ethics and policies (Woolfe 1992;), provided the crucial vote for the opposition that overturned Moore's regulation (Cohen 1997).

The Endangered Fauna (Interim Protection) Act 1991 (NSW) (EFIPA) was presented to Parliament by the Opposition. The EFIPA was the emergency legislation put into place to give time for comprehensive threatened species legislation could be formulated. The EFIPA was built on two entrenched regulatory systems: (1) licensing under the NPWA, and (2) the assessment of proposed developments using the EPAA (Kelly 1994). The limitations of the EFIPA were obvious because only fauna was protected and only 'endangered fauna' were assessed during development planning, using Fauna Impact Assessment (FIS) (Farrier 1993).

The Threatened Species Conservation Act 1995 (NSW) (TSCA) can in to effect on the 1st January 1996. The TSCA is focuses not only on fauna but also flora through a number of mechanisms and is embedded into planning legislation, thus raise the consideration that must be afforded to threatened species. Appendix 2.2 Recovery Plan Contents 222

APPENDIX 2.2 CONTENTS OF RECOVERY PLANS

From the Threatened Species Conservation Act 1995 (NSW) s. 59 Contents of recovery plans A recovery plan must: (a) identify the threatened species, population or ecological community to which it applies, and (b) identify any critical habitat declared in relation to the threatened species, population or ecological community, and (c) identify an threatening process or processes threatening the threatened species, population or ecological community, and ( d) identify methods by which adverse social and economic consequences of the making of the plan can be minimised, and ( e) state what must be done to ensure the recovery of the threatened species, population or ecological community, and (f) state what must be done to protect the critical habitat (if) identified in the plan, and (g) state, with reference of this ACT: (i) the way in which those objects are to be implemented or promoted for the benefit of the threatened species, population or ecological community, and (ii) the method by which progress towards achieving those objects is to be assessed, and (h) identify the persons or public authorities who are responsible for the implementation of the measures included in the plan, and (i) state the date by which the recovery plan should be subject to review by the Director-General. Appendix 5.1 Sonication Pollination 223

APPENDIX 5.1 SONICATION (BUZZ) POLLINATION

Sonication or buzz pollination has raised a large amount of curiosity from researchers in the field. Buzz pollination obtains its name from the sound the bees make while actively vibrating flowers to extract pollen (Harder & Barclay 1994; Larson and Barrett 1999b). The mechanisms requires bees to land on the flowers and vibrate their indirect flight muscles (by uncoupling their wing movements from the actual flight mechanism) with the wings folded at a high frequency sonicating the anthers (Knudsen & Olesen 1993; Harder & Barclay 1994; Larson & Barrett 1999b; Moore 1996). Pollen streams from the anthers (Larson and Barrett 1999b) through the use of poricidal dehiscence (i.e. an opening by apical pores, valves or flaps) (Larson and Barrett 1999a Knudsen & Olesen 1993), ready for harvesting. Often pollen (present often in large amounts) is the sole reward from flowers that rely on buzz pollination (i.e. Solanum species and Tetratheca species) (Buchmann & Cane 1989).

John H. Barrett first recorded buzz pollination in 1959 while observing low frequency buzzing sounds coming from foraging bees in the highland forests in New Guinea. In the last 40 years, sonication pollination has been observed in nearly all climatic zones around the world (Barth 1985). Coleoptera, Diptera and Hymenoptera have all since been identified as having the ability to participate in buzz pollination. Individuals have been identified as either social or solitary species and generalists or specialists (Buchmann & Cane 1989). However, only a sub-set of species from the three Orders is able to efficiently remove pollen through buzzing (Larson & Barrett 1999a).

The use of apical/poricidal dehiscence allows for directive pollen loading (Harder and Barclay 1994), however there are many disadvantages to releasing pollen in one visit. Species that rely on sonication pollination have evolved to maximise the benefits of this form of pollination (Larson & Barrett 1999a). To ensure success pollination, pollen should be released over a number of visits and this is done in two ways. Firstly bees species operate at a frequency of 400Hz or less while the flowers are more than often tuned to a higher frequency, thus ensuring only a small amount of pollen is released each visit (Moore Appendix 5.1 Sonication Pollination 224

1996). The gradual drying of pollen is also viewed as regulating release (King & Ferguson 1994), increasing dispersal potential over a longer period of time (King and Buchmann 1996). With these forms of restrictive pollen release, a species may become vulnerable to pollen limitation, which is a critical factor in influencing individual plant fertility (Larson & Barrett 1999a). Appendix 5.2: Pollen Library 225

APPENDIX 5.2: POLLEN LIBRARY

Plate A: Pollen from Tetratheca glandulosa (indicated by the arrow). Pollen is circular in shape.

Plate B: Pollen from Darwinia biflora (indicated by the arrow). Pollen is triangular in shape. Chapter 1 General Introduction General 1 Chapter Table 1.3: History of wildlife conservation legislation specific to New South Wales (Sourced from the legislation). c • mm 9 "3 S ! a. #o as 3 ‘ *-c •4-» 4-1 ■— n 3 on a a> o o 0) a> cs ca © 0) a © a u c CJ O 03 C s (B - I SJ •— c !/l 60 x ZJ 3 c*>

d d cd cd M M J _Q r S 5 s o •s u ) X »_r 73 ‘o 4H to E d T3 cd O £ D cd 00 3 0 c o .2 C o o cd o a o CO -a > o ° S - S g o s co ° u g • i8 3 2 § S E O cd o O k § 0

o °

& 2 C <+H Os 3 5/3 ■3 d CO 3 — o o o ij > g CL G cr ZJ 3 h d cocd O £

'o I to 0 0 o h_ o o V- r 3

* 12 <4-1 S3 <2 .2 S 15 ■S u -a • - « « •« U D ° J= 2 ■a ^ t II — H bjO 00 O t-H & 5P S o s 73 . a o & 8 ^ c 3 .3 g o u 5 53 3

^ o ^ .O < -C .2 *C .co X C *X« s £ ‘o S < t U X £ 73 2 Os ■4—> -2 5 5 -H E 0JD cd G CO a o O o c 3 y: co E £ C <*> 00 CD o Q- o o u< S § S’ ° & B h 2

o

U GO t

X ^ 'X co • !-< .S Id o IS o .5 73 — > 3 cd ^ C3 o -a -*-> cd - j c o o 3 o c' C to oo cd o ^ > c 3 00 ° O cd c c o o d co cd co 'o ~

co CL o

1 )

O C 'O •— ^dX) S ■s I I & ® a 3 c •a ^ C C ^ d ■° 5 ^ < 03 O o CO If « o C O 5 co .52 o co ° O c cd O o a s ^ is a; 2 73 Cd ^ •a 73 Z, oo Os o O o « t; 2 o 2 - ; t u g o « y H o o CL o o o cd c X 3 v— o O

«4H X .C M ^ O CO .§ & o •= o co O o t-H - C CO O 00 . X w. (U CL cd C CX g c E ^3 3 -C O O cd° B.5 u, o coco

5 2 *c #o *o #c ■^3 -*—> a o o c cd > 3 dcd cd o cd Q c 00 c oo ■— 00 c CL —c •—

*3 St 'S s -e .s -*—» 2 c o C CO y o c E cd >» Cu O CL c 00 ^ S c “ O O cd CL 3 t_ CO cd 00 r3 X c C 00 o. o

co jd 1 a ' ’C X 73 73 O o o 3 c 00 u- o un

Wildflowers and Under the guidance of the A fine of £5 ( 1st offence), £10 (2 nd Not specified. Time periods for Minister may issue a Native Plants Governor for the first time offence) or £20 (3 rd or subsequent protection can be licence to pick for Protection Act 1927 plants could be listed. offence) for picking or selling a specified. scientific purposes. listed species. 3 ' 3 C T3 *-* o .a § CO O cdo O S o - -*-* r- •- c o o Q 00 o s:

x : _ . . _ : x 73 -O T jd X) .22 .2 •*-* T3 Os 3 oo ) X O g o -g S 0) O a ^ ! 2 g E 3 *

j I T3 2 u c — .23 73 73 O OJO O ) X L _ a O ^ 2 X a -5 CL CO d o cd o o cd JS c O § (D -

o

M X 00 o cd S E co u. p O W c O y: rd c o h jd 73 X -*-* - D O C cd CL CL J-H o O > .to is l-l c O S c .S f O cfc J o x *o m C o cd o u co Cu t3 , < j3 z o o — 2 11 2 P ^ P o ^ 1 § Q O x cd 2 O O o Cd (D o "fc f " c aj ^ cd a g g ■*—» *> O o — COcd 52 o o .2 T3 cd o C ) cd 4) L co CL 3 _ & § o ^ 2 £ & ° ^ ’■O

P

Z h _o cdC/D ^ U < 1 < | z o ? ■o H o O x: o *- cd a o § « O CO o o T3 O CL CL U c c 2 .S I J =d 1— “ - 0 “ £ o O -o o . X d cd CO cd C .. ^ P

CO «* 2 ^3 _ l/J4_> ft o 0JD c c u s.

co CL o O r_ 2

32 'O T3 co p o Z o K o SI c < & - CO H-1 M .CO 73 .CO 73 < 73 x 73 w Cu < H G 3 c t xi i* o S • S > ^ •S o S u "o 6 .2 *—< O 2 «* ob o cd § C o o o O ^ o o c o I u c 3 c J—. o cd u- V CO J & SJ P co" 2P O o cd o o u o g- s — C | I u C cd CO . Os g ^ I V. § n h 2 2 ^ fli ^ -g *43 t < <4-1 X 2 .8 2 2 C/5 X X *s X X O z rr- 4-> -2 -2 -H -*—» U O . u E o o o > o o o c C o > 5 5 CL> h h 3

c a o O 73 3= H 3 .CO C/5 U X < d P id 3 'O X H go i o g ^ -= 3 73 73 2 O o CO ^ o ‘S c c w »- o Cd 3 , > 73 o O U o C 8 S O 1) cd c E CO o o o O O o C o u- cd C co o t-H E o t-M o 00 oj « O O- CL cd o C C u g g> £ 3 J £

7T

£ < _c M 73 73 .CO X 5 — -*-* -*-» 3 c o 73 o o c 5 o O > o o O o CL E o c c 1 - 0 '_ 8 8 c — ^ h h ‘E u

O c ^ -S g PU■g .o 73 ."3 to cd CL 2 cd > c cd cd co UJ § CL Q G co O cd o O E c . X E 3

— X _c td o CO o u 'o N Q U c o 0 X 00 CL CO 3 . X ) 0 o *-H V—

3 ~ 73 <4H 73 X C > d cd cd — CL .a . a S o u « ■o S c C 4) 5 o c o O CO co >% o X 1 T 73 X *C 13 u J= -*—* o u CJ o £ E C o o o E o o cd o 3

x "3 , •4—> Q s o l-l o O’ c 3 -is o d) O O O - I i o o o c

-2 14 general of NPWS. appendix 8.1 227

Gre\'illea kennedyana (Dame spider-flower) September, 2000 vulnerable ar north-western NSW 70% or recorded individuals occur in Sturt NP, remaining fire, prolonged and intensive grazing monitoring of populations and evaluation of browsing impacts, rescarc. individuals (3-4 populations) occur on leasehold land on requirements, Community awareness

Crevillea obtusiflora subsp. Obtusiflora and October, 2001 endangered Rylstone Shire Council for subsp obtusiflora 2 populations occur in Clandulla Slate unrcstrictcd vehicular access, tire, grading and slashing lor roads, and weeds habitat mgml (liaise with landholders, install significant roadside w v u m utual area), survey lor further i r------:—:------— -----—— ------—:------r ~ ------fecunda Forests, 1 population on private land and 1 population on land response to fire, pathogens and genetic variation) and community awareness y ,urU*“ P o ta tio n s , momtonng, research 9secd b.ology, fruit producbon, pollinators, iwned by Rylstone shire coundl; for subsp. Fecunda 2 populations occur on land owned by private landholders, 1 populations within the gardens of stone NP and 2 populations occur on land owned by Rylstone shire coundl

Domestic stock grazing, weed infestation, competition from native species, clearing, fire and flooding. Grevillea wilkinsonii (Tumul Grcvillca) October, 2001 endangered Goobarragandra River, Tumul 7 sites (80% of population) occur on private freehold land and 2 pcnod.c evaluation of populalion aac, treads and thrcals. a*vc mgml i » ! « = on p n v a tc land, secure ,„„e lc m l proleaion mgm, „f |and silcsi ^introduction of Ike sites on crown land species, include habilai protertion in planning documents, com m unity awareness endangered 1 population at lake keppit slate Park (crown land) browsing, trampling, invasion, disease and Hakea pulvinifera June, 2000 Namoi River, Gunncdah fire survey all potential habitat for further populations, monitoring, cx-suu ----- ^ %Jllon and genetic diversity, and pine control l.eionema lachnaeoides October, 2001 endangered Upper Blue Mountains 8 populations with 1 occurring in Blue Mountains National Park, 3 up-siopc disturbances, weed invasion, inappropriate fire regimes and site access sue reassessment and ongomg momlonng. fire management, de.ctoplU-,,1 J-.scssmeot mintaisc prolccUon „„ privatc lan(is, assessment for critical habilai in Blue Mountains City coundl recreation Reserve, 2 on private land and 2 on crown Recreation reserve

Persoonis m ollis subsp. maxima June, 2000 endangered Hornsby Hcights-Mt Colah-my include, Bcrowra valley regional Park, KCNP, Vacant crown Land habitat loss through development of Binya Close (vacant crown land) and habitat degradation (water quality and nutrient hatiilat management, surveying for further populations, research (rccnmmcnl and survivorship, growth rale, reproductive maturity), cx-situ conservation (seed) and possible listing of Ku-ring-gai area (owned by Landcom), private landholders and regional open space levels in catchments, weed invasion, rubbish dumping, infrastructure maintenance and recrcatiqn) critical habitat (owned by DUAP) Prostanlhera junonis (Somcrsby Minlbush) January, 2001 endangered Somcrsby plateau 9 populations in 2 LGAs with 5 occurring on private land, 1 on a level of current understanding, vegetation clearing, development, adjacent development, fire control activities (fire trails), threat and habitat management (including consideration of dcvclopmcn applications, rczonmg and assessment o f all activates), surveys for further populations, rescardi into seed crown reserve, 1 on crown land, 3 within Brisbane waters NP and 2 habitat degradation 9unrestricted access), inappropriate fire regimes and weed invasion ecology, population dynamics and response to fire, potential listing of critical habitat, ex-situ conservation programmes (storing genetic material) and community awareness. on land owned by Gosford City Coundl

Pterostylis gibbosa (Illawarra Grccnliood pending endangered 3 sites in the Illawarra, 1 site at All populations arc found on private land habitat loss, habitat degradation (grazing, inappropriate fire regimes, weed invasion and collecting from the wild) threat and habitat mgmt, further surveying for populations, research and monitoring and community awareness Ordiid) finalisation Milbrodale and 1 site near Nowra Pterostylis sp. 15 (Botany bar Bearded October, 2001 endangered Botany Bay national Park, 1 population within the national park inappropriate fire regimes, habitat degradation related to unrcstrictcd access, unauthorised collection, insecure tenure of habitat and threat mgmt, survey for populations, research (biology and ecology) and cx-situ conservation Greenhood ordiid) Kumell peninsula land where potential habitat exists W ollemia nobilis (Wollcmi Fine) September, 1998 endangered Wollcni national Park All known populations occur within the national park restricted distribution, low seed set, slow maturation , unauthorised seed collection, fire events. Pathogens and development of an access strategy, community awareness, fire plan lor the NP, best catachmcnt practices, monitoring, research into age structure and fire response, mycological studies, unauthorised visits from the public genetics, further surveys and cx-situ collections (genetic material, commercialisation strategy, re-introduction in original sites)

Zieria adenophora (Araluen Zieria) December, 2001 endangered Araluen 1 population (56 individuals) on crown lease hold land (DLWC) trampling, soil disturbances, and removal of assodatcd vegetation by goats and wallabies control known threats, obtain appropriate protection and mgmt of lan

Zieria lasiocaulis pending endangered W illi Willi NP National Park inappropriate disturbance regimes through road and track construction, maintenance and development of potential protect and maintain wild populations, increase sdcntific knowledge, protect any new populations, liaise with land managers, look in cx situ conservation, assess needed for critical recreational fadlitics, hybridisation habitat, increase community awareness. Zieria prostrata April, 1999 endangered Coffs Harbour Occurs on only 4 headlands within Moonec Beach nature Reserve weeds, habitat degradation by pedestrians, p

"lost" threatened flora of south-eastern NSW October, 2001 all spcdcs are south-eastern NSW only vague location details unknown obtain appropriate listing for spcdcs likely to be cxtinct, relocate those spcdcs present in NSW listed as

threatened Alpine flora December, 2001 endangered Kosduszko National Park all spcdcs occur entirely within national park trampling by walkers, mitigation of tramplin »effects, resort development, larval damage evaluation of population distribution and sizes, quantification of threats