Spatial Patterns of Lepidoptera in the Eucalypt Woodlands of the Sydney Basin, New South Wales, Australia

University of Wollongong Theses Collection University of Wollongong Theses Collection

University of Wollongong Year 

Spatial patterns of Lepidoptera in the eucalypt woodlands of the Sydney Basin, New South Wales, Australia

Lachlan Ashby University of Wollongong

Ashby, Lachlan, Spatial patterns of Lepidoptera in the eucalypt woodlands of the Sydney Basin, New South Wales, Australia, MSc-Res thesis, Department of Biological Sciences, University of Wollongong, 2008. http://ro.uow.edu.au/theses/93

This paper is posted at Research Online. http://ro.uow.edu.au/theses/93

SPATIAL PATTERNS OF LEPIDOPTERA IN THE EUCALYPT WOODLANDS OF THE SYDNEY BASIN, NEW SOUTH WALES, AUSTRALIA.

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

MASTER OF SCIENCE - RESEARCH

from

UNIVERSITY OF WOLLONGONG

LACHLAN ASHBY

Department of Biological Sciences 2008 CERTIFICATION

I, Lachlan Ashby, declare that this thesis, submitted in fulfilment of the requirement for the award of Master of Science, in the Department of Biological Sciences, University of Wollongong, is wholly my own work unless otherwise referenced or acknowledged. This document has not been submitted for qualifications at any other academic institution.

Lachlan Ashby

July 2008

Table of Contents

List of Tables, Figures and Plates...... vi

ABSTRACT ...... xi

ACKNOWLEDGEMENTS...... xii

CHAPTER 1: INTRODUCTION ...... 1

1.1 Community Ecology Of Invertebrates ...... 1

1.1.1 Spatial Distribution of Invertebrates...... 2

1.2 Community Ecology of Lepidoptera And Indicators Of Ecosystem Function ...... 7

1.2.1 Spatial Patterns of Lepidoptera...... 10

1.3 Aims of the Investigation ...... 15

CHAPTER 2: SITE DESCRIPTION AND METHODS ...... 17

2.1 Site Description...... 17

2.2 Trap Design...... 26

2.3 Sorting and Identification ...... 29

2.4 Methods of Analyses...... 29

CHAPTER 3: RESULTS...... 33

iii

3.1 Community Composition ...... 34

3.1.1 Abundance and Species Richness...... 34

3.1.2 Family Richness...... 38

3.1.3 Distribution of Rarity ...... 40

3.1.4 The Relationship Between Species Distribution and Abundance...... 42

3.1.5 The Relationship Between Family Distribution and Abundance...... 43

3.1.6 Turnover Between Assemblages ...... 44

3.1.7 Relationship Between Assemblage Turnover and Rarity...... 47

3.2 Multivariate Analysis ...... 48

3.2.1 Ordinal Assemblage Structure...... 48

3.2.2 National Parks...... 52

3.2.3 Sites ...... 53

3.2.4 Trap Locations ...... 55

CHAPTER 4: DISCUSSION...... 56

4.1 Richness and Abundance Patterns of Lepidoptera ...... 56

4.2 Assemblage Structure Over a Hierarchy of Spatial Scales...... 58

4.3 Ecology Of Moth Spatial Patterns ...... 60

4.4 Patterns At Different Taxonomic Levels ...... 62

4.5 Rarity ...... 64

4.6 Urbanisation and its Potential Effect on Moth Distributions...... 65

4.7 Conclusion ...... 67

4.8 Avenues for Future Study...... 68

REFERENCES...... 70

APPENDIX A – TAXA INDEX ...... 93

List of Tables, Figures and Plates

Figure 2.1: Average annual rainfall for Sydney and the surrounding environs (millimetres per year). From Benson and Howell (2000)...... 19

Figure 2.2: Diagrammatic representation of the spatial hierarchy incorporated within the experimental design of the study. At the largest sampling scale is the national park, the intermediate scale is the site and smallest scale is the trap location...... 21

Table 2.1: Description for each of the nine sites used to collect Lepidoptera specimens outlining site name, grid reference, national park for each site and sample collection date...... 22

Plate 1: Typical eucalypt woodland vegetation structure of the Sydney Basin in which sample collection of Lepidoptera took place...... 25

Figure 3.1: The total number of species of Lepidoptera which were identified within each family over all sampling occasions at all locations...... 33

Figure 3.2: The accumulation of total number of species of Lepidoptera for all trap locations of the study. The plotted relationship suggests that the species richness within the dataset may not be entirely representative of the community as a whole...... 34

Figure 3.3: The accumulation of species of Lepidoptera for trap locations within each of the national parks. The plotted relationship suggests that the species richness within the dataset may be representative of the community contained within the Blue Mountains National Park community, but not necessarily for the other two national parks...... 35

Figure 3.4: Mean abundance of Lepidoptera caught at each of the sites (±S.D.). There were three sites within each

national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring-gai National Park...... 36

Table 3.1: Summary of nested ANOVA for the mean abundance of Lepidoptera caught per trap at each site, with three sites nested within each of three national parks. df= degrees of freedom, SS = sum of squares, F ratio is the test statistic with its associated probability (P)...... 36

Figure 3.5: Mean number of species of Lepidoptera caught at each of the sites (±S.D.). There were three sites within each national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring-gai National Park...... 37

Table 3.2: Summary of nested ANOVA for the mean number of species of Lepidoptera caught per trap at each sites, with three sites nested within each of three national parks. df= degrees of freedom, SS = sum of squares, F ratio = test statistic with its associated probability (P)...... 37

Figure 3.6: The accumulation of families of Lepidoptera for all the trap locations across the landscape, with the plotted relationship suggesting that family richness within the dataset may be representative of the community as a whole...... 38

Figure 3.7: The accumulation of families of Lepidoptera for trap locations within each of the national parks. The plotted relationship suggest that the family richness within the dataset may be representative of the communities sampled...... 39

Figure 3.8: Mean number of families of Lepidoptera caught at each of the sites (±S.D.) . There were three sites within each national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring- gai National Park...... 39

Table 3.3: Summary of nested ANOVA for the mean number of families of Lepidoptera caught per trap at each sites, with three sites nested within each of three national

vii

parks. df= degrees of freedom, SS = sum of squares, F ratio = test statistic with its associated probability (P)...... 40

Figure 3.9: Species richness of Lepidoptera for three national parks in relation to the percentage of occupied traps...... 41

Table 3.4: Number of species trapped at the nine sites (meso- spatial scale) within 3 national parks (Blue Mts, Ku- ring-gai and Royal) as a function of the percentage of occupied traps...... 41

Figure 3.10: The relationship between abundance and distribution for the majority of the 228 species of Lepidoptera trapped at three national parks. Three outliers from the line of best fit are identified...... 42

Figure 3.11: The relationship between abundance and distribution for the 25 families of Lepidoptera trapped at three national parks...... 44

Table 3.5: The percentage of species of Lepidoptera shared between each national park. The base sampling unit (top row) is measured against the comparison sampling unit and indicates that the proportion of species shared between each of the national parks are relatively similar to one another...... 45

Table 3.6: The average percentage of species of Lepidoptera shared between each site (+ SD). The base sampling unit (top row) is measured against the comparison sampling unit and indicates that the proportion of species shared between each of the sites are relatively similar to one another. Blue italicised figures indicate within national park comparison. Red underlined figures indicate between national park comparison...... 45

Table 3.7: The percentage of families of Lepidoptera shared between each national park. The base sampling unit (top row) is measured against the comparison sampling unit and indicates that the proportion of families shared between each of the national parks are relatively

viii

similar to one another...... 46

Table 3.8: The average percentage of families of Lepidoptera shared between each site (+ SD). The base sampling unit (top row) is measured against the comparison sampling unit. The base sampling unit (top row) is measured against the comparison sampling unit and indicates that the proportion of families shared between each of the sites are relatively similar to one another. Blue italicised figures indicate within national park comparison. Red underlined figures indicate between national park comparison...... 46

Table 3.9: The percentage of species of Lepidoptera shared between each site for families occurring in more than 20% traps. The base sampling unit (top row) is measured against the comparison sampling unit...... 47

Figure 3.12: Two dimensional MDS plot of species of Lepidoptera for the 51 locations in nine sites across three national parks. The grey dashed line indicates congregations of sampling points from more than one site from different national parks. A blue dashed line indicate congregations of sampling points of a single site...... 49

Figure 3.13: Two dimensional MDS plot of families of Lepidoptera for the 51 locations in nine sites across the three national parks. A blue dashed line indicate congregations of sampling points which do not overlap other sites...... 50

Table 3.10: Number of pair-wise site comparisons where there was an observable assemblage difference of Lepidoptera at 1% significance level indicating the number of sites that contain shared assemblages...... 51

Table 3.11: The average contribution to Bray Curtis indices of similarity within national parks of the four most abundant species of Lepidoptera...... 52

Table 3.12: The average contribution to Bray Curtis indices of

similarity within national parks of the five most abundant families of Lepidoptera...... 53

Table 3.13: The average contribution to Bray Curtis indices of similarity within sites of the two most abundant species of Lepidoptera...... 54

Table 3.14: The average contribution to Bray Curtis indices of similarity within sites of the eight most abundant families of Lepidoptera...... 55

ABSTRACT

The patterns of spatial distribution and abundance were investigated for moth assemblages in the eucalypt woodlands of the Sydney Basin. A total of 228 species of Lepidoptera, distributed among 25 families, were recorded from three national parks located on the perimeter of the Sydney metropolitan region.

From within the eucalypt woodland habitat of the Sydney Basin, the study investigated the spatial variation of night-flying Lepidoptera present at several different scales of observation, from the trap level through to across the landscape. Assemblages varied with spatial scale, with uniformity occurring across the landscape as a whole, however becoming patchy at finer spatial scales. Multivariate and turnover analysis indicated that although heterogeneity of abundance and richness may vary significantly depending on spatial scale, sites and national parks contained their own unique suite of species in comparison to one another.

The structure of the assemblages of moths in the eucalypt woodlands of the Sydney Basin can vary, and is dependant on the level of spatial scale of observation. Further study needs to be conducted at a range of temporal scales to ascertain the presence of patterns in the Lepidoptera communities in the Sydney region in order to contribute to the development of suitable conservation strategies in the Sydney Basin.

ACKNOWLEDGEMENTS

I firstly would like to thank my supervisor, Kris French for her patience and guidance throughout this whole project.

A special thanks goes to Esther, my partner, for her support, encouragement and possession of an understanding that knew no bounds.

Thankyou Dave Britton, at the Australian Museum, who generously donated his time in helping me with Lepidoptera identification techniques.

Also, thankyou to the New South Wales National Parks and Wildlife service for allowing me to sample within their parks.

xii CHAPTER 1: Introduction

CHAPTER 1: INTRODUCTION

1.1 COMMUNITY ECOLOGY OF INVERTEBRATES

Invertebrates offer opportunities for the measurement of impact of disturbances, that in some circumstances, will outweigh the technical difficulties (Kitching et al. 2000). Technical difficulties that may be incurred when using invertebrates to measure impact are often associated with the high diversity levels which are regularly observed during insect collection (Landau et al., 1999). However, it is this attribute of being highly diverse, and their quick response to environmental change, which give inventories of arthropods the potential to be good indicators of change (Oliver and Beattie, 1993) and habitat diversity (Landau et al., 1999). Insects, in particular, are ubiquitous, abundant and diverse in terrestrial environments and may therefore make ideal indicators of change. They therefore should be a major element in conservation planning and natural resource management (Fisher, 1998). For example, failure to incorporate knowledge of community patterns such as the distribution of species richness across a range of scales may obscure a potentially alarming problem of conservation biology and extant reserves may not be able to support a significant or unique proportion of the local or regional species diversity (Summerville et al., 2001).

Community patterns are a product of the population dynamics of their component species (Strong et al., 1984). Investigating patterns of species distribution and abundance can provide an insight to the mechanisms that structure these assemblages (Begon et al., 1996).

1 CHAPTER 1: Introduction

Patterns in ecological communities depend on a complex interplay between large- and local-scale processes. Factors which may affect large-scale variation include environmental stress, dispersal and productivity with ecological processes which potentially operate on a local-scale including predation and competition. Determining which factors underlie spatial variation in community structure, and placing this knowledge within a predictive framework, is a central issue in community ecology (Menge and Olson, 1990).

1.1.1 Spatial Distribution of Invertebrates

The causes for observed spatial distributions of invertebrates within eucalypt woodlands are varied. For instance, habitat fragmentation is likely to cause a decrease in species richness, creating changes in spatial patterns of abundance, patterns of endemism, and ecosystem structure (Kitching et al., 2000). Fragmentation causes a disruption in the spatial distribution of resources (Van Dyck and Matthysen, 1999), and affects biological diversity because it can influence processes from individual behaviour through to population dynamics and ecosystem fluxes (Robinson et al., 1992). Habitat fragmentation, disturbance and pollution have all been shown to have major effects on insect physiology, population dynamics, species richness and diversity (Cappuccino and Kareiva, 1985) and different components of an ecosystem can respond in different ways to habitat fragmentation (Robinson et al., 1992). The survival of populations in remnant habitat patches is dependant on the nature and condition of the patch, as well as on the size, population density and dispersal

2 CHAPTER 1: Introduction ability of the species (Frankie and Ehler, 1978). Due to the relatively intensive urbanisation of the south-eastern Australia, in particular in and around Sydney, the eucalypt woodlands of the Sydney Basin can be considered a fragmented landscape. With extensive ecosystems of undisturbed areas being progressively being cleared for human use (Kitching et al., 2000) to accommodate expanding cities such as Sydney, fragmented landscapes are becoming increasingly important for conservation of biodiversity and the understanding of ecological processes at work in these landscapes is essential (Saunders et al., 1991). Therefore fragmentation due to urbanisation is a potentially significant consideration when investigating the spatial variation of abundance and richness in the eucalypt woodlands around Sydney.

Patterns of spatial variation and abundance can potentially be an important consideration in the design of nature reserves and the conservation of biological diversity (Brown et al., 1995) with variation in species richness at the landscape scale a critical component in conservation planning and natural resource management (Coddington et al., 1996). Estimating species richness is inherently a scale-dependent process (Summerville et al., 2001) where Currie (1991) suggested that the richness of a species on a local scale varies with factors such as productivity, disturbance and habitat diversity. The observed species richness can also be significantly influenced by the number of samples taken, the sample resolution and to a certain extent geographic breadth (Summerville et al., 2001). Therefore, often detailed characterisation of insect biodiversity concentrates on determining local species richness which is often accomplished

3 CHAPTER 1: Introduction through a series of site-specific surveys for selected taxa, which is then frequently published as a species checklist and less commonly used to test or challenge ecological hypotheses of insect biodiversity (Summerville et al., 2001).

Quantitative measures of species and individual abundance provide standardised comparative values for evaluation of various habitats, communities and ecosystems (Hammond and Miller, 1998). Positive correlations have been found between maximum density, average density over the area where the species occurs and the number of sites inhabited within the local region and the area of the geographic range (Brown, 1984). The geographical distribution of species abundance is as a rule heterogeneous, with the highest abundances occurring near the centres of the species’ range and the lowest at the margins (Hengeveld and Haeck, 1982). Southwood et al. (1982) found arthropod abundance increased with species richness and Gaston and Lawton (1988a) attempted to provide a contribution to the understanding of distribution and abundance by analysing patterns of insect populations, which included moths, through the exploration of the relationships between local population abundance, local population variability and regional distribution. It was found that widespread species were generally locally abundant, and have populations that fluctuate more than scarce species which are geographically restricted. Hengeveld and Haeck (1982) suggested from the results of their investigation that local frequencies can be partly explained by the position of sampling areas within the geographical range of a species and partly by some biological property

4 CHAPTER 1: Introduction of the species, such as feeding habits.

There seems little prospect of employing detailed knowledge of the dynamics of each and every species in the community with the aim of understanding ecological patterns. With the interplay of local and regional dynamics further compounding this problem, an alternative is to search for broad patterns in local and regional dynamics of species (Gaston and Lawton, 1988). As a result, there has been a paradigm shift away from a site-based, species-centred philosophy toward a broader “eco-regional” approach (Summerville et al., 2001). The study of structure of assemblages over different spatial scales has frequently been ignored by field workers and theoreticians, despite its potential importance. Increases in the types and amounts of data needed for such field studies and the need for complicated mathematical theory are possible reasons for avoidance (Brown, 1995).

The relative importance of factors regulating community structure varies with spatial scale (McCreadie and Adler, 1998). Although few studies have compared spatial patterns at different scales for a large number of species over a wide range of habitats (Underwood and Chapman, 1996), a positive relationship between species’ population size at a single site and their regional distribution (measured as the number of sites occupied) has been demonstrated for a number of animal groups (Gaston, 1988). This observed correlation could be explained on the basis that widespread species are more flexible in their use of resources (Brown, 1984) and/or that generalists are able to

5 CHAPTER 1: Introduction use more resources than specialists (McNaughton and Wolf, 1970).

Broad-scale spatial patterns in insect diversity have been explored in some studies, such as Kerr et al. (1998). However, a greater integration between local and regional patterns of insect diversity might be achieved by linking site-specific patterns of insect richness reported to regional patterns in range geometry, habitat heterogeneity or biogeographic history (Summerville et al., 2001). Barriers to this integration include limited understanding of how spatial resolution and extent contributed to the accumulation of species in a local inventory (Hammond, 1994).

Despite few investigators systematically studying variation in population density over the geographic range of species (Brown, 1984), the field distribution of insect herbivores in temperate woodlands has been documented, which have then been related to several features of their host-plants. In particular, it has been found that leaf nutrients and defence characteristics play a major role in determining herbivore food choice, abundance and distribution (Basset, 1990). Variance in arthropod spatial distribution could also be attributed to leaf age characteristics, arthropod aggregation patterns, arthropod activity and distance from tree and distance to tree trunk (Basset, 1990).

In light of the relative patchy nature of research aimed at identifying and understanding patterns within invertebrate ecological communities, this study will explore the community ecology of night-

6 CHAPTER 1: Introduction flying Lepidoptera by focusing on patterns and distributions over a hierarchy of spatial scales in the eucalypt woodlands of the Sydney Basin.

1.2 COMMUNITY ECOLOGY OF LEPIDOPTERA AND INDICATORS OF

ECOSYSTEM FUNCTION

It is often proposed that indicator taxa are an efficient method for identifying conservation priorities. The diurnally active butterflies are often a popular option (Ricketts et al., 2002). From a conservation perspective, Lepidoptera have been advocated as potential indicators of environmental impact (Daily and Ehrlich, 1995) as they have a relatively well known invertebrate taxonomy, can be considered as a useful group for biogeographical and conservation research (Ricketts et al., 2001) and can yield large sample sizes containing many species (Landau et al., 1999). Lepidopteran species are relatively host- specific (Janzen, 1988) and have the potential to serve as indicators of native plant diversity (Luff and Woiwood, 1995). Although lepidopteran studies have usually involved butterflies and/or day- flying moths (Kitching et al., 2000) which are not necessarily easily trapped (Young, 1997), the majority of lepidopteran species are nocturnal (Janzen, 1988).

Night-active Lepidoptera species have a number of attributes that could make them good ecological indicators. The majority of nocturnal Lepidoptera families are readily attracted to light-traps, which allows standardised ecological measures. Moths are specious

7 CHAPTER 1: Introduction and sufficiently diverse to offer potentially powerful discrimination in detecting impacts on an ecosystem (Kitching et al., 2000) and moths are for the most part taxonomically well known (Kitching et al., 2000). Lepidoptera fall into two groups according to whether adults have functional mouth parts or not. Adults that feed mainly consume carbohydrates during adult life, while non-feeding adults depend entirely on resources derived from the larval period (Tammaru and Haukioja, 1996). The structure of the mouth is used to help identify and differentiate the most ‘primitive’ groups of moths and the pattern of openings to the exterior of the female genitalia is generally used to separate the other main subgroups (Young, 1997). Other important features used in the classification of Lepidoptera include the wing venation, antennae structure and genitalia (Dickens, 1974).

Moths play a vital role in the ecosystem (Common, 1990). They convert plant biomass into animal biomass, which inturn support first order carnivores such as arthropod predators, parasitoids, amphibians, reptiles, birds, bats and other small mammals (Pyle et al., 1981). Lepidoptera are important as herbivores, pollinators and prey (Barlow and Woiwod, 1989) and can serve as flagship taxa in biodiversity studies (Butler et al., 1999). Lepidoptera can significantly influence nutrient cycles, plant population dynamics and predator-prey population dynamics (Hammond and Miller, 1998) and their relationship between plant species and the number of lepidopteran species that they host has been documented by numerous studies (Hammond and Miller, 1998). Moths and their larvae depend almost entirely on plants, where larvae act as plant defoliators, leaf miners,

8 CHAPTER 1: Introduction stem borers, as well as feeders on flowers, fruits and seeds (Hollway, 1989). In addition, leaf water, which usually limits nitrogen assimilation efficiency and nitrogen accumulation rate in chewing insects, is a factor of some importance in the spatial distribution of temperate herbivores (Scriber, 1977).

Ricketts et al. (2002) examined whether diversity of butterflies, as popular indicator taxon, would be a useful surrogate measure of diversity of moths. The study found however that there was no correlation in diversity patterns. Where butterflies tended to be restricted to a single habitat type (meadows), moths samples yielded equivalent diversity levels among different habitat types which include forests of aspen, conifer and also meadows. Gaden et al. (1993) specifically looked at assemblages across three different habitat types and found that samples of Microlepidoptera (and Pyraloidea) from the three different forest types in Borneo had very few species in common. The diversity was found to be very high in the lowland forests, where as, much lower in the habitats found at the edge of mangroves. Just eight families of accounted for 90% of the species in samples collected from all three forest types.

It therefore has been established that moths play an important ecological role, are taxonomically relatively well known, can be captured via ecological quantifiable and standardised methods and are under-exploited for the purposes of ecological research. As a result, moths potentially provide an opportunity to explore spatial community structure with the possibility of allowing a greater understanding of

9 CHAPTER 1: Introduction overall ecosystem function.

1.2.1 Spatial Patterns of Lepidoptera

Both physical and biological factors can cause differences within and among communities (Menge and Olson, 1990). Ricketts et al. (2001) established that suites of moth species were spatially restricted to either meadow, aspen and conifer habitats while many other species of moths were widespread. This suggests that the widespread species were either associated with widespread plant resources (such as a broad distribution of the host plant), or depend on a single habitat but move frequently among patches. In a similar vein, Gaston (1988) found that the most abundant species were the most variable and occurred across the greatest number of sites. By analysing data to determine whether similar patterns are apparent in a group of closely related species drawn from different communities, Gaston (1988) was able to relate within and cross-community processes. The data showed positive correlations between local population abundance, local population variability and regional distribution. Relationships with physical attributes, such as body size, were much less robust and were sensitive to factors such as the range of sizes used. Small species tended to have the most variable populations and also were the most widely distributed.

Communities of Lepidoptera can be a rich and heterogeneous fauna, even within the one habitat type. Summerville et al. (2001) found that considerable heterogeneity appeared to exist among moth

10 CHAPTER 1: Introduction communities at both local and regional scales. A greater number of moth species were recorded when multiple sites and multiple years were examined. In addition species accumulation curves did not saturate for any inventory, regardless of spatial extent. Although sampling extent was partly attributed to local and regional moth heterogeneity within the community, forest stands in the same landscape may contain their own unique suite of rare species.

McGeoch and Chown (1997) found differences between moth assemblages at sites with different habitat variables. However, in spite of differing habitat qualities, there was no strong match between any single habitat quality variable and the moth assemblage. The habitat qualities included roadway proximity (and traffic volume), size of habitat patch, and vegetation density and structure. Similarly, by examining the relationship between species abundance, variability, life-history and food characteristics for moth species, Leps et al. (1998) also found that changes in moth population size not being driven by changes in single standalone factors such as food plants.

A number of studies have focused on anthropological practices and their effect on moth communities. McGeoch and Chown (1997) looked at the impact of local habitat quality parameters on moth assemblages in an urban setting in South Africa and found that a combination of habitat variables such as habitat patch size, vegetation structure and density provided the best match with the assemblage ordination. The authors considered this was probably due to a synergistic effect on the moth assemblage as a result of the numerous

11 CHAPTER 1: Introduction forms of disturbance in urban areas. This highlights the multiplicity of factors that determine a moth assemblage pattern. Ricketts et al. (2001) found that different agricultural regimes appear to offer similar habitat elements and thus may not differ substantially in their capacities to support moth populations. Therefore, factors affecting moth community composition are more likely to be a result of a combination factors rather a single reason.

When searching and quantifying possible factors which may affect moth ecosystem composition, it is important to note sequences of temporal fluctuations. Cook and Graham (1996) made collections from different five-year periods over a 24 year period in southern England which were very different from one another, both in terms of numbers of individuals present and the taxonomic composition. Temporal fluctuation therefore should be taken into account when considering moth assemblages to make assessments of habitat patches and reserves. Cook and Graham (1996) considered that temporal fluctuations were important for programs designed to monitor the effect of man-made environmental changes (for example, urbanisation) on fauna. By looking at the evenness and species numbers in moth populations, they illustrated a number of temporal patterns in Lepidoptera communities. Numbers of species were higher in samples from the woodland in comparison to the grasslands, and higher numbers of species were found earlier, rather than later in the season. Although notable fluctuations were observed, the difference in patterns between catches in grasslands and in woodlands suggest that the communities in these two habitat types in close proximity are

12 CHAPTER 1: Introduction different. However, it could be argued that the different species have different favoured flight paths. Therefore, when investigating moth assemblages, it is important to bear in mind the many different facets that influence on the structure of the community being observed.

McGeoch and Chown (1997a) also found positive relationships between species distribution and abundance. The species composition of the Lepidoptera community examined were predominantly similar between localities although the localities were spatially distant. The common and moderately common species within the community were highly concordant, and rare species were considered to be comparatively diffusive (rare at some localities and not at others). The authors of the investigation considered a couple of possible explanations for differences of spatial concordance between rare and common species. Species could attain their highest abundances near the centre of their ranges, with lower abundances near to the range margins. Therefore, as rare species may have a more fragmented geographic range, compared to a more common species, a higher variability would be detected in their abundances compared to a more continuously distributed common species. Another possible reason for a difference in spatial concordance may be due to rare populations of species behaving within a meta-population, whereby their populations are continually becoming extinct and being recolonised by founder populations.

Therefore in light of the discussion provided within this chapter, this study has aimed to distinguish spatial patterns of night-flying moths in

13 CHAPTER 1: Introduction the large tracts of eucalyptus woodlands which are available in a number of national parks which surround the City of Sydney.

14 CHAPTER 1: Introduction

1.3 AIMS OF THE INVESTIGATION

Although moths appear to have the potential for being useful indicators of ecological process, there is a scarcity of studies which consider the patterns of distribution and abundance of moths in the Sydney region. With the City of Sydney itself being centrally nested within three significant national parks, and thus acting as a possible invertebrate dispersal barrier between each of these relatively large tracts of essentially natural vegetation, this study looked at the spatial patterns of night-flying Lepidoptera within and between each of the national parks.

An investigation into moth distributional patterns is an important step in understanding the ecological processes within the eucalypt woodlands of the Sydney Basin. Therefore, the aims of the investigation were:

• To determine what ecological patterns are present through the characterisation of spatial distributions of night-flying Lepidoptera; and

• To examine any observed spatial patterns of night-flying Lepidoptera in terms of species and family distribution.

These two aims were addressed by examining:

• Assemblages of night-flying Lepidoptera in three significant national parks within the Sydney Basin over a hierarchy of

15 CHAPTER 1: Introduction

scales; and

• If community characterisation of night-flying Lepidoptera at the family level correlates with community characterisation at the species level to ascertain the presence of differences assemblages in relation to taxonomic resolution.

16 CHAPTER 2: Site Description and Methods

CHAPTER 2: SITE DESCRIPTION AND METHODS

2.1 SITE DESCRIPTION

The Sydney Basin extends from Batemans Bay (35.72°S; 150.19°E) to the south and Nelson Bay (32.71°S; 152.16°E) to the north, and almost as far west as Mudgee (32.60°S; 149.60°E) (Bureau of Meteorology, 2007). The City of Sydney is nested within three national parks which offer tracts of undisturbed eucalypt woodlands which border onto urbanised fringes. To the west is Blue Mountains National Park, to the south is Royal National Park and Ku-Ring-Gai Chase National Park is located along the northerly borders of metropolitan Sydney.

The Triassic sediments of the Sydney Basin has its centre in the Cumberland Lowlands, which consists of plains and gently undulating low hills on the youngest of the Triassic rocks, the Wianamatta Group (Chapman and Murphy, 1989). Sandstone and shale, on which most of Sydney sits, are responsible for the distribution patterns of most of Sydney’s native vegetation. The shale landscapes form on Wianamatta Shale, while the sandstone landscapes are formed from Hawkesbury sandstone and the underlying Narrabeen Group sandstones (Benson and Howell, 2000).

Variation in Sydney’s soils are a result of slope and local climate. As well as the plain and low hills of the Cumberland Lowland, the landscapes of the area also contain a topography of steep hills, long narrow ridges (occasionally capped with remnant Wianamatta Shale),

17 CHAPTER 2: Site Description and Methods and deep rocky valleys (Benson and Howell, 1990). Local climatic conditions vary according to topographical features and distance from the coast with temperature extremes becoming more pronounced, and an increase in the incidence of frost, for the inland areas. There is an approximate decrease in the rainfall gradient for the inland areas (Benson and Howell, 1990), however the lower reaches of the Blue Mountains have a precipitation comparable to the coastal areas (Figure 2.1).

The eucalypt woodlands found in Sydney Basin vary structurally as a result of factors including local aspect, soil, drainage and time since last fire. Woodlands are the main type of vegetation found on the Hawkesbury Sandstone landscapes. Woodland also occurs on shales, although the trees tend to be taller and straighter with a richer understorey of grasses than the woodlands on sandstone (Benson and Howell, 2000). Dominant tree species of the Hawkesbury sandstone woodlands include Eucalyptus gummifera, E. haemastoma, and E. sieberi with the shrub layer including Lambertia formosa, Banksia spinulosa and B. serrata (McKenna, 1998).

18 CHAPTER 2: Site Description and Methods

Figure 2.1: Average annual rainfall for Sydney and the surrounding environs (millimetres per year). From Benson and Howell (2000).

19 CHAPTER 2: Site Description and Methods

The three national parks (Figure 2.1) are all sandstone woodlands and are large tracts of natural vegetation with negligible anthropogenic development.

The hierarchical sampling design within this investigation comprised of three spatial levels (Figure 2.2). At the highest spatial level, patterns across national parks were compared, where each of the three “parks” (Blue Mountains, Ku-Ring-Gai and Royal) were separated by the largest (or macro-) distance of at least 30 kilometres. Within each of the national parks, there were three sampling “sites”, nine in total for the investigation. Sites were spatially separated from one another by a medium (or meso-) scale, which was a minimum distance of seven kilometres. The smallest (or micro-) sampling scale occurred within each of the sites where there were five to six sampling “trap locations”. Five to six trap locations were placed in a linear transect at 20 metres intervals. Details of the traps used in this study are provided below and Table 2.1 provides specific details for each site.

20 CHAPTER 2: Site Description and Methods

National Park

Trap Location(s) Site(s) Transect (1 to 6) (1 to 3)

20m

100m

CHAPTER 2: Study Site Description

Table 2.1: Description for each of the nine sites used to collect Lepidoptera specimens outlining site name, grid reference, national park for each site and sample collection date.

Grid Sample Site National Park Site Description Reference Collection Date

S 33.6684 K1 January 2003 Duffy’s Track, located at the end of Booralie Road, to the west of Terry Hills. E 151.1805 S 33.5966 K2 Ku-Ring-Gai Chase NP February 2004 America Track, off West Head Road on the Lambert Peninsular. E 151.2754 S 33.6644 K3 February 2004 At the end of Chilten Road, to the south-west of Church Point. E 151.2673 S 34.0866 Beyond the locked national park gates at the end of Loftus Street, to the west R1 January 2003 E 151.1438 of Bundeena. S 34.1486 R2 Royal NP February 2004 Couranga Track, off McKell Avenue, Fosters Flat. E 151.0294 S 34.0825 R3 February 2004 Winnifred Falls Track, off Warumbul Road to the south-east of Audley. E 151.0710 S 33.6595 B1 January 2003 At the end of Booker Road, to the north of Hawkesbury Heights. E 150.6426 S 33.7318 B2 Blue Mountains NP March 2004 At the end of Elizabeth Street, to the west of Mount Riverview. E 150.6271 S 33.6636 Beyond the locked national park gates along Grose Road north of B3 March 2004 E 150.5499 Faulconbridge.

22 CHAPTER 2: Site Description and Methods

The transects contained within each of the sites were placed along single-track fire trails or open walking trails which were between two and five metres wide. The topography of the immediate surrounds of the sites was flat and each transect located between approximately 500 to 750 metres from the outer edge of the eucalypt woodlands. All localities had an altitude of less than 500 metres above sea level and a local average rainfall was approximately 1000mm to 1400mm per annum.

Two inherent sources of error confound all invertebrate sampling procedures; variation in the application of the procedure itself and variation in the distribution and abundance outside the measured parameters of the investigation, for example temporal fluctuations. Sources of error due to temporal variations in invertebrate numbers can be controlled by taking samples at consecutive time intervals and restricting samples to specific weather conditions, such as temperature, cloud cover, wind speed, incidence or rain. Assuming that sampling procedures can be standardised, the major sources of error in invertebrate sampling can be reduced significantly (Majer et al., 1990). Therefore, to account for possible fluctuations in temporal variations, general environmental variables were kept uniform. This was done by ensuring that on trap nights there was no rain or strong winds were present and cloud cover was minimal throughout the evening. In addition, uniformity in the structure of habitat was ensured across all sites through a qualitative assessment of each site, where consideration was given to woodland canopy cover, tree species, tree height, understorey height and density and time since last

23 CHAPTER 2: Site Description and Methods fire. Figure 2.3 illustrates the general woodland structure for the sites used in this investigation.

24 CHAPTER 2: Site Description and Methods

Plate 1: Typical eucalypt woodland vegetation structure of the Sydney Basin in which sample collection of Lepidoptera took place.

25 CHAPTER 2: Site Description and Methods

2.2 TRAP DESIGN

Adult macro-Lepidoptera can be sampled with bait traps, light-traps, Malaise traps, pheromone traps, sticky traps, sweep nets or window traps (Butler et al., 1999). Among these methods, the most widely used and most productive way to catch night-flying moths has been light-traps (Baker, 1985). Light-traps are considered the standard technique for sampling nocturnal Lepidoptera (Summerville et al., 2001) as they can be surveyed readily in large numbers and allow a relatively efficient estimation of geographic patterns diversity and abundance (Ricketts et al., 2001).

The light trapping method has proved consistently successful in capturing large numbers and a great variety of species of night-flying Lepidoptera (Butler et al., 1999) and was the adopted method for this investigation to compare the relative differences amongst moth assemblages in three national parks in the Sydney Basin. As the study consisted of a comparison of relative differences in Lepidoptera assemblages, variables and potential biases were uniform throughout the study. Variables that may potentially bias the richness and abundance a light trap catches include temperature, moon-phase, air movements, precipitation, vegetation, natural population fluctuations, time of year, trap design and positioning (Butler et al. 1999), some species are predominantly diurnal and/or while others are not phototactic and thus may not be attracted and sampled by the light traps. All these variable were considered for this investigation.

26 CHAPTER 2: Site Description and Methods

Six standard light-traps were used, which were switched on at dusk and left on for a period of 7.5 hours. The trap consisted of an eight watt ultraviolet (UV) fluorescent tube (or “black light”) powered by a portable twelve volt battery. The fluorescent tube was flanked by three clear Perspex wings. The night-flying Lepidoptera attracted to UV light were collected in a funnel at the base of the UV light/Perspex wings. The funnel which lead into the pail, contained a household insecticide. The light-traps used for this study is illustrated in Figure 2.4.

27 CHAPTER 2: Site Description and Methods

Fluorescent Power Lead to UV Tube 12V Battery

27cm Clear Perspex Wing

25cm

Fine plastic meshing

25cm

Figure 2.4: Light trap used for sampling night-flying Lepidoptera. Lepidoptera are attracted to the fluorescent UV light, which is flanked by clear Perspex wings. The Perspex wings capture the Lepidoptera by funnelling the specimens into the enclosed pail located beneath the UV light.

28 CHAPTER 2: Site Description and Methods

2.3 SORTING AND IDENTIFICATION

Gross sorting took place where Lepidoptera samples were separated from other invertebrates collected in each of the traps. Individuals where the specimen body length was less than three millimetres were not included. Lepidoptera specimens were pinned and identified to family using the formal keys by Nielsen et al. (1996). To further ensure correct identification was carried out, advice and instruction regarding the identification of Lepidoptera was also gained from the collection manager (D. Britton pers. comm.) at the Australian Museum in Sydney, as well as comparing specimens with the collection at the Australian Museum. Identification was taken to morphospecies based on differences in morphological characteristics. A reference collection was established containing the different morphospecies from each family.

2.4 METHODS OF ANALYSES

Mean species abundance and mean richness for family and species were calculated for 51 trap locations (micro-scale) across nine sites (meso-scale) over three national parks (macro-scale). Analysis was conducted at the family taxa level, as well as the species level, to allow analysis at two different taxonomic levels. Data transformation prior to calculation of the means was not required for data standardisation. To detect if any differences existed across the three scales being examined, abundance and species richness were analysed using a nested Analysis of Variance (ANOVA) of sites within national park.

29 CHAPTER 2: Site Description and Methods

Species-area curves were created to assess the respresentativeness of sampling of the moth assemblages. The total species richness of each trap location, site and national park was then compared to determine if any variation was evident at these scales. If all the sites had the same assemblage structure, then the species and richness of the trap location would be the same as that for the sites. However, it was predicted that there would be difference in assemblage structure between the sites. This principle was also used to examine variation in species composition of sites within a national park.

To examine the relationship between distribution of the species and families sampled and their abundance, the number of trap locations that each species was observed at was plotted against the mean number of those species in the traps. The hypothesis was that distribution would be positively correlated with abundance.

To describe the community structure and establish an understanding of the spatial distribution of Lepidoptera in the eucalypt woodlands of the Sydney Basin, rarity was taken into consideration. The distribution of uncommon families and species was examined to determine if any trends were discernable across the range of scales incorporated into the study. To examine the distribution of uncommon species and families, richness was plotted against the percentage of traps a particular species or family was present in.

To determine the extent that species and families changed from one assemblage to the next, the percentage of the total richness that was shared by each assemblage with each other was calculated. This was

30 CHAPTER 2: Site Description and Methods done both at the landscape level (i.e. national park) and the intermediate level (i.e. sites) to test the hypothesis that similarity in terms of the number of shared species (or families) would be greater for assemblages within the same national park compared to assemblages from different national parks.

Multivariate analysis with PRIMER (Clarke, 1993) was used to examine the differences of assemblages within and among national parks and sites at both the family and species taxonomic level.

To analyse for broad-scale differences in species composition between national parks and between the sites, a two-way nested Analysis of Similarity (ANOSIM) with sites nested as replicates in national parks at both the species and family taxonomic levels was carried out. Two- dimensional Multidimensional Scaling (MDS) plots were also constructed for both species and family to examine the relationship between the sites from the national parks.

Similarity Percentage breakdowns (SIMPER) was used to determine which species and families were contributing the most to similarity, or dissimilarity, at both the national park and site spatial scales. A stepwise algorithm, BIO-ENV (BVSTEP), was then used to determine sub-sets of species and families that were having significant influences on the overall relationships and patterns within the community.

Finally, a two-way crossed ANOSIM2 (no replication) where trap locations were nested within sites, was conducted to assess differences

31 CHAPTER 2: Site Description and Methods in assemblages at both the species and family taxonomic levels for the smallest spatial scale incorporated within this study.

32 CHAPTER 3: Results

CHAPTER 3: RESULTS

A total of 4935 moths were collected from 51 light-traps from three national parks during late summer and early autumn. This represented 228 morphospecies (hereafter referred to as species) distributed among 25 families. The most speciose family was Pyralidae which contained 43 species (Figure 4.1). This was followed by Geometridae with 36 species and Oecophoridae with 26 species.

40 s e i 35 c e p s

30 f s e o i c r e p e 25 S b

. o m N u 20 N

e e

e e e e e e e e e e e a e a e e e e e e e e a a r e a a a a a a a a e a a a a a a a a a a a a d d a a d d d d d d d d d d d d d i d i d d t i i i i i i d i i i d i i d i i d i l d i l i d i i i i i i t l i i t l i c o d d r e t u r o r p t h e s r i i n n h t a t n o e m a c d n r r o r g e c r u s h t r o c i i c o e r e t m h L t y y r h u y y n e o h e t u e d o i T g t p a A l n t l y h h m p s Tortridae P p y o C E h c a o m N p e o h t o A P T T o P Z p o o S u r c e o i T m G n S e e s s E N G t o o a O P p L C Y FFamailmy ily

Figure 3.1: The total number of species of Lepidoptera which were identified within each family over all sampling occasions at all locations.

33 CHAPTER 3: Results

3.1 COMMUNITY COMPOSITION

3.1.1 Abundance and Species Richness

For abundance, 26 of the 228 species collected (11.4%) accounted for 75.0% of the total individuals. Geometridae sp. 32 (Oenochroma vinaria) was the most abundant species (11.0%; 595 individuals). Other species that were also abundant included Oecophoridae sp. 27 (8.9%; 449 individuals), Arctiidae sp. 16 (5.6%; 281 individuals) Cossidae sp. 21 (4.7%; 238 individuaAlsll) and Cosmopteridae sp. 11 (4.7%; 238 individuals).

250

200 s s e i c e e i p S c f 150 o

r e

e All b p m u 100 N S

. o 50 N

0 1 5 9 13 17 21 25 29 33 37 41 45 49 Traps Figure 3.2: The accumulation of total numbTer roaf specsies of Lepidoptera for all trap locations of the study. The plotted relationship suggests that the species richness within the dataset may not be entirely representative of the community as a whole.

The accumulation of species across the entire sampling landscape is illustrated in Figure 3.2. The nature of the curve suggests that the diversity of species within the dataset was not likely to be entirely representative of the species richness present across the sampling

34 Species Area Curve CHAPTER 3: Results

region at the time the investigation was carried out.

160 K

140 R

120

s B e i c

e 100 Ku-Ring-Gai p s e S i f c Royal o e

r 80 p e S b

Blue Mountains . m o u

N 60 N

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Traps

Traps Figure 3.3: The accumulation of species of Lepidoptera for trap locations within each of the national parks. The plotted relationship suggests that the species richness within the dataset may be representative of the community contained within the Blue Mountains National Park community, but not necessarily for the other two national parks.

The accumulation of species at each of the national parks level is illustrated in Figure 3.3. Although it is likely that the frequently occurring or common species have been sampled adequately, the curves for Ku-Ring-Gai Chase and Royal National parks suggest that sampling may not be thoroughly representative and the diversity of species present in the dataset, particularly for the rarer species, has not been entirely captured. However, the flattening of the curve of the Blue Mountains National park indicates that the dataset for this park maybe close to being representative.

The mean abundance of Lepidoptera caught per trap for each of the three national parks ranged from 74.9 ±44.7 (s.d.) at the Blue

35 CHAPTER 3: Results

Mountains National Park to 113.7 ±80.4 at the Royal National Park. The abundance at Ku-Ring-Gai Chase National Park was intermediate (100.2 ±50.5) (Figure 3.4). There were significant differences in species abundance at the site level but not at a national park level (Table 3.1).M ean Species Abundance

300

250 e t i e c S n r a 200 d e n p u b e A c n n 150 a a e d M n u

b 100 A n a e 50 M

0 B1 B2 B3 R1 R2 R3 K1 K2 K3 Sites Site Figure 3.4: Mean abundance of Lepidoptera caught at each of the sites (±S.D.). There were three sites within each national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring-gai National Park.

Source df SS F Ratio P

National Park 2 9597.124 2.320 0.1108

Site (National Park) 6 91047.03 7.336 <0.0001

Error 42 86882.90

Total 50 190613.18

36 CHAPTER 3: Results

The average abundance caught per trap for each of the three national parks ranged from 33.1 ±12.7 at Royal National Park to 28.3 ±10.2 at the Blue Mountains National Park. The average abundance caught per trap at Ku-Ring-Gai Chase National Park was 32.2 ±12.8 (Figure 3.5). Similar to the abundance data, there were significant differences in species richness at the site level but not at a national park level (Table 3.2). Mean # Spp caught

s 50 e i e c t i e S p

40 S r e f o P

r s e 30 e b i c m e u p

N 20 S n a e 10 M

0 B1 B2 B3 R1 R2 R3 K1 K2 K3 SitSeiste Figure 3.5: Mean number of species of Lepidoptera caught at each of the sites (±S.D.). There were three sites within each national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring-gai National Park.

Source df SS F Ratio P

National Park 2 141.877 0.806 0.4533

Site (National Park) 6 3244.713 6.146 <0.0001

Error 42 3695.60

Total 50 7154.59

The species richness and abundance observed in this dataset suggests most of the variation among the assemblages is at the site level, with

37 CHAPTER 3: Results the moth assemblages appearing to lack any significant variance across the landscape.

3.1.2 Family Richness

Of the 25 observed families, three families accounted for 51.1% of the total individuals collected. Geometridae was the most abundant family (22.6%; 1115 individuals) followed by Pyralidae (16.5%; 816 individuals), then Oecophoridae (12.0%; 592 individuals). The curve’s relatively flattened nature suggests that the diversity of families is probably a representative dataset of the family richness present across the sampled landscape (Figure 3.6). Furthermore, when individual parks are considered, each park has a representative dataset (Figure 3.7) All

25 s e i l i 20 m a F f 15 o Ser r e b ies

m 10 u 1 N 5

0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 Traps

38 CHAPTER 3: Results

s e

i 15 K l i

m Ku-Ring-Gai a R F

f Royal o

10 B r

e Blue Mountains b m u

N 5

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Traps Figure 3.7: The accumulation of families of Lepidoptera for trap locations within each of the national parks. The plotted relationship suggest that the family richness within the dataset may be repFreasenmtativiel oyf t hRe coimcmhunnitiees sasmspled.

16.00

14.00 s e i l i e t i

m 12.00 a S F

r f o e

10.00 r p e

b y l m i 8.00 u m N a F

6.00 n a e 4.00 M

2.00

0.00 B1 B2 B3 R1 Sites R2 R3 K1 K2 K3 Site Figure 3.8: Mean number of families of Lepidoptera caught at each of the sites (±S.D.) . There were three sites within each national park. B = Blue Mountains National Park, R = Royal National Park, K = Ku-ring-gai National Park.

There were significant differences (p = 0.0012) in family richness at the site level but not at a national park level (Table 3.3, Figure 3.8).

39 CHAPTER 3: Results

National parks had on average 11.4 families, with a standard deviation ranging from ±2.1 to ±2.6.

Table 3.3: Summary of nested ANOVA for the mean number of families of Lepidoptera caught per trap at each sites, with three sites nested within each of three national parks. df= degrees of freedom, SS = sum of squares, F ratio = test statistic with its associated probability (P).

Source df SS F Ratio P

National Park 2 0.51943 0.0708 0.9318

Site (National Park) 6 100.41863 4.5625 0.0012

Error 42 154.067

Total 50 256.157

This significant difference in family richness suggests that most of the variation among the assemblages is occurring at the site level, with the assemblages at the family level appearing to be homogeneous between all three national parks.

3.1.3 Distribution of Rarity

To examine the distribution of uncommon species and determine if any trends were discernable across the range of scales, species richness was plotted against the percentage of traps (or sampling unit) in which a particular species was present (Figure 3.9, Table 3.4).

40 By National Park CHAPTER 3: Results 250 200

180 200 160 Blue Mts Blue Mountains NP 140

150 e Royal i

c 120 Royal NP e . p BLUE MTS p S

s f

. 100 ROYAL

o Ku-Ring-Gai

r Ku-Ring-Gai NP N 100 e KU-RING-GAI b

No. Species 80 m u N 60 50 40 20 0 0 0-5 6-10 11-150%-20% 21%-40% 164-12%0-60% 61%-80% 81%-100% No. Traps Perc%en otaf gtrea posf tsrpa pasr ea p srpeesceinets was found in

Figure 3.9: Species richness of Lepidoptera for three national parks in relation to the percentage of occupied traps.

Table 3.4: Number of species trapped at the nine sites (meso-spatial scale) within 3 national parks (Blue Mts, Ku-ring-gai and Royal) as a function of the percentage of occupied traps.

Blue Mts Royal Ku-Ring-Gai Percentage of Traps B1 B2 B3 R1 R2 R3 K1 K2 K3

0%-20% 198 196 187 190 166 203 169 182 196

21%-40% 12 14 12 9 28 11 23 20 12

41%-60% 6 7 12 11 4 5 13 9 6

61%-80% 7 7 9 6 10 5 7 4 4

81%-100% 5 4 8 12 20 4 16 13 10

At the national park or site level, a large number of species sampled in this study had a distribution locally restricted to a limited number of sampling units (1158 individuals, representing 23% of the abundance, comprised of 187 species in less than 20% of the traps). This

41 CHAPTER 3: Results

observation however may be partially attributable to the species area curves indicating species (including rare ones) may not have been completely represented in the sampling.

3.1.4 The Relationship Between Species Distribution and Abundance

To test the hypothesis that the abundance and distribution of a species are correlated, the number of traps in which a species was caught was plotted against the mean number of that species in those traps. Figure 3.10 showed that a positive correlation existed between distribution and abundance (r = 0.77, p < 0.001, n = 227).

12 Geometridae sp. 32 p p a a r r T T 10 Cossidae sp. 21 / / s l s a l u a d i u v i d 8 i d n v I i Pyralidae sp. 32 n d a e n

I 6

M e g a r

e 4 v A 2

0 0 10 20 30 40 50 60

Number Traps Occupied Figure 3.10: The relationship Nbeutwmeenb eabru Tndraanpces a nOd cdcisutripbuietiodn for the majority of the 228 species of Lepidoptera trapped at three national parks. Three outliers from the line of best fit are identified.

42 CHAPTER 3: Results

Arctiidae sp. 16 was atypical in that, although it was abundant, it was only found in eight traps and 279 of the 281 individuals (99.1%) were caught at one site (Royal National Park). This may have been a result of the traps intersecting a swarm of the moths on the particular trapping evening. Therefore Arctiidae sp. 16 was excluded from the dataset used to create Figure 3.10. Similarly, Pyralidae sp. 32 was abundant and occurred in a relatively limited number of traps. This species was however not excluded. Cossidae sp. 21, and particularly Geometridae sp. 32, were observed to occur in a majority of traps, however the relative abundance of these two species were noted to be particularly high, suggesting swarming behaviour.

3.1.5 The Relationship Between Family Distribution and Abundance

Similarly, a positive correlation was also observed between the number of traps in which families were caught and the mean number of that family in those traps (r = 0.80, p < 0.001, n = 25) (Figure 3.11).

43 CHAPTER 3: Results

s p

a Geometridae r p 20 T / a s l r d a u T e d

i i

f 15 v i p d o n

I u r n c a e

e 10 c b M O m

u 5 N 0 0.0 1.0 2.0 3.0 4.0 5.0 Number Traps Occupied Mean Individuals/Trap Figure 3.11: The relationship between abundance and distribution for the 25 families of Lepidoptera trapped at three national parks.

3.1.6 Turnover Between Assemblages

There is likely to be a relatively high proportion of any given assemblage (at national park or site) that will not be present in any adjacent or subsequent sampling unit given the large percentage of species that only occur in a few traps (Section 3.1.3). Consequently, to examine the amount that the species composition changes from one sampling unit to another, that is the turnover, the percentage of each species of each sample unit (for national park and site) that is shared with each other sample unit was calculated at the species and family level.

44 CHAPTER 3: Results

Base Sampling Unit

Blue Mts Royal Ku-Ring-Gai t n ni o U s Blue Mts x 63.3 57.9 ri ng i pa pl m Royal 66.9 x 64.5 o m C Sa Ku-Ring-Gai 63.3 66.7 x

The percentage of species shared between the three national parks is illustrated in Table 3.5 and shows that between 42% and 33% of the species of any one national park are restricted to that national park. The average percentage of species restricted to each of the nine sites was between 43% and 51%, as demonstrated in Table 3.6.

Base Sampling Unit ng i pl

m Blue Mts Royal Ku-Ring-Gai

Sa t n ni Blue Mts 50.8 (±8.9) 49.6 (±5.4) 48.8 (±9.0) o U s ri

pa Royal 54.8 (±12.5) 56.1 (±13.5) 54.4 (±10.6) m o

C Ku-Ring-Gai 54.3 (±10.7) 55.5 (±9.3) 57.1 (±12.8)

Generally speaking, turnover between sites and national parks is comparable to one another, however when turnovers are compared at the landscape level (base sampling unit) they do not differ markedly. On the other hand, turnovers of the base sampling unit differ

45 CHAPTER 3: Results significantly from one another at the site level.

Base Sampling Unit

Blue Mts Royal Ku-Ring-Gai t n ni o U s Blue Mts x 81.8 86.4 ri ng i pa pl m Royal 85.7 x 86.4 o m C Sa Ku-Ring-Gai 90.5 86.4 x

The percentage of families shared between the three national parks is illustrated in Table 3.7 and shows that only 9-18% of the families were restricted to a particular national park.

Base Sampling Unit ng i pl

m Blue Mts Royal Ku-Ring-Gai

Sa t n ni Blue Mts 86.3(±14.3) 83.9(±12.0) 79.4(±10.3) o U s ri

pa Royal 87.4(±11.7) 86.6(±13.7) 83.3(±8.1) m o

C Ku-Ring-Gai 85.1(±9.2) 86.0(±10.2) 85.2(±6.8)

The average percentage of families shared between the nine sites (Table 3.8), shows that between 13-21% of the families are restricted to any one site. The Blue Mountains National Park displayed a greater variation in the average family turnover between sites within

46 CHAPTER 3: Results the national park as opposed to turnover between a site from the Blue Mountains National Park and a site from another national park. This is also the case for Royal National Park. Ku-Ring-Gai National Park however had a lower variation in the average turnover between sites within the Ku-Ring-Gai National Park than between a site in Royal or Blue Mountains National Parks.

Base Sampling Unit

B1 B2 B3 R1 R2 R3 K1 K2 K3

B1 x 0.79 0.79 0.82 0.73 0.79 0.78 0.78 0.77

t B2 0.81 x 0.85 0.76 0.87 0.79 0.78 0.78 0.85 ni U B3 0.81 0.85 x 0.82 0.87 0.85 0.78 0.81 0.85 ng i pl

m R1 0.88 0.79 0.85 x 0.80 0.85 0.86 0.86 0.82 Sa n

o R2 0.69 0.79 0.79 0.71 x 0.74 0.73 0.73 0.77 s ri

pa R3 0.97 0.94 1.00 0.97 0.97 x 0.95 0.95 0.95 m o

C K1 0.91 0.88 0.88 0.94 0.90 0.90 x 0.92 0.90

K2 0.91 0.88 0.91 0.94 0.90 0.90 0.92 x 0.90

K3 0.94 1.00 1.00 0.94 1.00 0.95 0.95 0.95 x

The percentage of families which occurred in more than 20% of the traps which are shared between each of the nine sites shows that between zero and 27% of the species of any one sites it lost as one moves to one of the other sites.

3.1.7 Relationship Between Assemblage Turnover and Rarity

To assess the effect of the less common species on the assemblage

47 CHAPTER 3: Results turnover at the local level, rare species (singletons and doubletons) were excluded and the data analysed in a single factor ANOVA (site).

There was a significant difference at the local level (F8,63 = 4.68, P = 0.0002).

The 41 common species, which occurred in greater than 20% of traps, accounted for 77% of the total specimens (3777 individuals). Species that occurred in 20% or less of traps were excluded in data analysed in a single factor ANOVA (site), where no significant difference at the local level (F8,63 = 0.238, P = 0.9821).

3.2 MULTIVARIATE ANALYSIS

3.2.1 Ordinal Assemblage Structure

Ordinations did not identify a distinctive aggregation of sites into their respective national parks; all national parks overlapped (Figure 3.12). The grey dashed line indicates aggregations of traps from two sites of different national parks, while the blue dashed line indicates sites which have unique aggregations with no overlap from another site. Therefore, this ordination indicates that no national park has a particularly unique assemblage of species, however a number of sites, (irrespective of national park) may contain an assemblage of species that is unique relative to other sites.

48 CHAPTER 3: Results

Blue Mountains B1 = B2 = B3 = Royal R1 = R2 = R3 =

K u-Ring-Gai K1 = K2 = K3 =

49 CHAPTER 3: Results

Blue Mountains B1 = B2 = B3 = Royal R1 = R2 = R3 = K u-Ring-Gai K1 = K2 = K3 =

Fi gure 3.13: Two dimensional MDS plot of families of Lepidoptera for the 51 locations in nine sites across the three national parks. A blue dashed line indicate congregations of sampling points which do not overlap other sites.

The grey dashed line (which indicates areas of aggregations of traps within sites from different national parks), and the blue dashed line (which indicates unique aggregation and no overlap from another site) illustrates some aggregation of trap locations into their respective sites according to family assemblage (Figure 3.13), but not into their respective national parks. This aggregation of sites however is not as strong as in the species ordination (Figure 3.12).

The ordination of assemblage at a more coarse ecological survey level supports Figure 3.12 in that the family assemblages are somewhat distinguishable from site to site, but not necessarily from national park

50 CHAPTER 3: Results to national park.

Differences in the assemblages were not evident when the three different national parks were compared using ANOSIM (R=-0.14, p>0.01) for species. However at the site level, the ANOSIM detected significant differences in the species assemblages between the sites (R=0.62, p<0.01). Pair-wise tests were conducted where all but six of the 36 site pairs had observable differences from one another at 1% significance level.

Blue Mts Royal Ku-Ring-Gai

Blue Mts 3 4 2

Royal 3 4 4

Ku-Ring-Gai 4 3 3

No national park contained a complete suite of three sites which shared similar assemblages with one another, as illustrated in Table 3.10. However Blue Mountains and Ku-Ring-Gai National Parks each contained a pair of sites with similar assemblages, but all sites from within Royal National Park differed from one another.

As was observed at the species level, differences in the family assemblages were not evident when the three national parks were compared using ANOSIM (two-way nested, trap locations in sites, sites in national park, R=-0.11, p>0.01). The ANOSIM detected significant differences in the family assemblages between the sites (R=0.46, p<0.01). However, pair-wise tests detected only 15 of the 36

51 CHAPTER 3: Results site pairs with observable differences from one another at 1% significance level, indicating that family assemblage patterns were relatively similar to species.

3.2.2 National Parks

To determine which species contributed to similarity across each the three national parks, species contributing to greater than 10% of similarity within any one national park was determined with SIMPER (Table 3.11), which identifies species that are the most influential within the dataset.

Table 3.11: The average contribution to Bray Curtis indices of similarity within national parks of the four most abundant species of Lepidoptera.

Blue Mts Royal Ku-Ring-Gai

Geometridae sp. 32 22% 12% 23%

Oecophoridae sp. 27 14% 12% 13%

Cossidae sp. 21 13% 5% 10%

Cosmopteridae sp. 11 4% 11% 7%

A number of species were found to occur uniformly across the landscape, however these were patchy at the localised level. Four of the five most abundant species were also found to contribute to a considerable proportion of the similarity within all three of the national parks a (total of 40% to 53%). Therefore abundant species had a significant influence on the homogeneity at the landscape level. Arctiidae sp. 16, although abundant, was atypical in that it was only found in eight traps, where 99.1% of individuals were caught at one site. Abundant species, in particular Geometridae sp. 32, contributed

52 CHAPTER 3: Results in similar proportions to within national park similarity for both the Blue Mountains and Ku-Ring-Gai. However, Royal National Park appeared to have a slightly different species set causing within national park similarity.

To determine which families contributed to national park similarity, families that contributed to greater than 10% of the similarity within any one national park was determined using SIMPER. A summary is provided in Table 3.12.

Table 3.12: The average contribution to Bray Curtis indices of similarity within national parks of the five most abundant families of Lepidoptera.

Blue Mts Royal Ku-Ring-Gai

Geometridae 26% 21% 32%

Pyralidae 18% 21% 17%

Oecophoridae 16% 13% 14%

Cossidae 13% 7% 9%

Gelechiidae 5% 10% 6%

The four most abundant families also contributed to the most similarity within each of the national parks, in particular, Geometridae was responsible for a noticeable proportion of similarity within each of the national parks.

3.2.3 Sites

To determine the species which contributed to dissimilarity between each of the nine sites, species with greater than 10% of the dissimilarity between any of the sites was determined with SIMPER (Table 3.13).

53 CHAPTER 3: Results

Table 3.13: The average contribution to Bray Curtis indices of similarity within sites of the two most abundant species of Lepidoptera.

Blue Mts Royal Ku-Ring-Gai Av.

B1 B2 B3 R1 R2 R3 K1 K2 K3 -

Arctiidae sp. 16 11.7 11.3 7.4 11.3 21.1 3.4 10.8 9.9 8.1 10.6

Cosmopteridae sp. 11 4.2 4.0 3.1 3.5 18.9 6.7 3.0 5.1 3.9 5.8

BVSTEP detected twenty species having a significant influence on the overall sample relationships and patterns (ρ = 0.95). These were: - Anthelidae sp. 11; - Arctiidae sp. 16 and Utetheisa pulchelloides; - Cosmopteridae sp. 11; - Cossidae sp. 8 and sp. 21; - Gelechiidae sp. 6 and sp. 10; - Geometridae sp. 30, sp. 31 & sp. 32; - Noctuidae sp. 5; - Oecophoridae sp. 27; - Plutellidae sp. 2; - Pyralidae sp. 35 and sp. 39; - Thyridoidae sp. 1; - Yponomeutidae sp. 6, sp. 7 and sp. 8.

To determine the families which contributed to the dissimilarity between the sites, families that contributed to greater than 10% of the dissimilarity between any of the sites was determined with SIMPER, which is summarised in Table 3.14.

54 CHAPTER 3: Results

Table 3.14: The average contribution to Bray Curtis indices of similarity within sites of the eight most abundant families of Lepidoptera

Blue Mts Royal Ku-Ring-Gai Av.

B1 B2 B3 R1 R2 R3 K1 K2 K3 -

Arctiidae 16.2 12.1 7.7 12.6 29.4 12.4 9.2 15.6 17.3 14.7

Cosmopteridae 7.6 7.1 5.7 6.3 4.1 9.0 5.5 7.5 5.8 6.5

Cossidae 7.6 6.5 7.7 7.7 5.8 7.8 7.9 5.4 7.0 7.0

Gelechiidae 8.4 8.4 6.6 6.9 7.1 8.0 8.1 7.6 8.2 7.7

Geometridae 20.4 18.7 23.7 19.7 9.4 20.0 17.3 35.4 22.1 20.7

Oecophoridae 10.1 9.1 9.6 10.5 8.4 9.7 6.8 7.8 12.4 9.4

Pyralidae 12.1 12.3 12.7 14.5 13.9 14.0 10.4 12.2 14.1 12.9

Thyridoidae 8.6 11.3 5.5 7.8 9.5 5.9 10.3 5.3 5.3 7.7

BVSTEP detected a subset of six families bearing a significant influence on sample relationships and patterns (ρ = 0.96), these being: - Geometridae; - Pyralidae; - Oecophoridae; - Yponomeutidae; - Cosmopteridae; and - Arctiidae.

3.2.4 Trap Locations

Differences in the assemblages were not evident across the trap locations when compared using ANOSIM2 (two-way crossed with no replication, trap locations in sites, R=-0.05, p>0.01) for species. Furthermore, differences in families at each site were not evident when the trap locations were compared using a two-way crossed ANOSIM with no replication (R=-0.02, p>0.01).

55 CHAPTER 4: Discussion

CHAPTER 4: DISCUSSION

Spatial extent is likely to have a significant effect on insect species richness estimates and community composition through the distance- decay phenomenon, but few studies have considered its effects across a hierarchy of scales (Summerville et al., 2001). As a result, the primary aim of this study was to examine the patterns of moth assemblages over a hierarchy of spatial scales in the eucalypt woodlands of the Sydney Basin.

4.1 RICHNESS AND ABUNDANCE PATTERNS OF LEPIDOPTERA

The moths within this study showed a pattern of distribution that reflected results from other studies investigating the relationship between abundance and distribution. It was found that the distribution of species or families of moths were correlated to their distribution in the eucalypt woodlands of the Sydney Basin. Gaston (1988) found a significant link between local population abundance, local population variability and regional distribution, with the most abundant species being the most variable and occurring at the greatest number of sites. Despite this well observed correlation between abundance and distribution, the reasons for the relationship are not necessarily understood. Brown (1984) suggested that species which are locally more abundant might also be more widespread because they have “broader niches” in that they are more flexible in their use of resources.

All three national parks used in this study contained large tracts of relatively undisturbed eucalypt woodlands, which were separated from

56 CHAPTER 4: Discussion one another by a minimum distance of approximately 30 kilometres. The intervening distance between each of the national parks was dominated by a potentially inhospitable urban matrix which provided a landscape that was structurally different to the large tracts of eucalypt woodlands contained within the national parks. By investigating the assemblage of moths at three national parks which were separated by Sydney’s metropolitan region, it was found that variation in moth communities in the eucalypt woodlands of the Sydney Basin depended on spatial scale, where abundance, species richness and family richness did not differ markedly over the landscape as a whole, but was highly variable at smaller spatial scales. The general uniformity over the landscape, indicates that comparable levels of abundance and richness of night-flying Lepidoptera were contained within each of the three national parks and that each national park is a good representative of the regional richness of moths.

At a finer spatial scale the study compared multiple sites contained within each of the national parks (or meso-scale). Sites were separated by a minimum distance of approximately seven kilometres and were separated by either urban or eucalypt woodland matrixes. The species area curves at this scale did not show any indications of flattening suggesting that it had not captured the full complement of species during the sampling period. At this fine spatial level the richness and abundance of night-flying Lepidoptera was generally more patchy and heterogeneous than at the landscape scale. Richness and abundance in this study consisted of a high variation in faunal assemblage at the meso-scale of observation. Several authors have

57 CHAPTER 4: Discussion found that Lepidoptera in other forest types are a rich and heterogeneous fauna (tropics: Solis and Pogue 1999, Pogue 1999, Robinson and Tuck 1993 and Chey et al. 1997 temperate North America: Wagner et al. 1995, Hammond and Miller 1998, Butler and Strazanac 2000, Summerville et al. 2001).

4.2 ASSEMBLAGE STRUCTURE OVER A HIERARCHY OF SPATIAL

SCALES

Greatest variability at the meso-scale suggests processes occurring over this spatial scale are most important in determining moth assemblages. While two dimensional MDS ordination plot for both species and families indicated a certain level of aggregation at the site level, no particularly observable aggregation of sites within their respective national parks occurred. This indicates that observable changes in the spatial structure in Lepidoptera assemblages in the eucalypt woodlands of the Sydney Basin is predominantly occurring at the site level rather than the landscape or national park level. At the smallest spatial scales of tens of metres (trap level), there were no differences in assemblage structure, similar to what was observed at the landscape. Two species, Geometridae sp. 32 and Cosmopteridae sp. 11, were significantly homogenous in their distribution across the landscape, but significantly patchy in their distribution at a more localised level. On the other hand, another two species, Oecophoridae sp. 27 and Cossidae sp. 21 were significantly uniform across the landscape but not necessarily patchy in their distribution at a more localised spatial level, as they did not contribute to any significant amount of dissimilarity between the sites. In addition, Arctiidae sp.

58 CHAPTER 4: Discussion 16 did not contribute significantly in landscape similarity, however was responsible for a significant amount of dissimilarity at a more localised spatial level.

The BVSTEP analysis indicated that only a small proportion (0.09) of the species in this investigation had a significant influence on the overall patterns and relationships within the dataset as a whole. The majority of the species had a clumped distribution that were unique over a spatial scale of kilometres. This clumped and patchy distribution was not necessarily identifiable when the landscape was looked at a broad scale of observation. Brown et al. (1995) suggested that clumped patterns of distribution may reflect spatial variation in the environment, as was found by McCreadie and Alder (1997) where the majority of species distributions (across ecoregions) were clumped, i.e. they were most frequent in one (or two adjacent) areas and rare elsewhere. Where highly diversified interaction assemblages of varying specificity, such as many mutualistic (e.g. plant – pollinator) and antagonistic (e.g. plant – herbivore) networks maybe present, evolutionary processes and dynamics yielding complex patterns can be difficult to decipher. However, it is possible that most interactions between given sets of animals and plants can be depicted as ordinations of the interacting entities that position them on a continuum of patterns. Such a continuum may reflect coevolutionary dynamics varying between sequential specialisation, by way of coevolutionary vortexes within diverse assemblages (Lewinsohn et al., 2006).

Therefore, more localised factors and processes, for example plant

59 CHAPTER 4: Discussion community types and site connectedness with other sites of similar habitat type or fire regimes, which occur over a scale of kilometres, maybe related to the patchiness in richness, abundance and assemblages of night-flying moth species. On the other hand, factors and processes that maybe related to the more uniformity of richness and abundance at the landscape level maybe include those that occur at the landscape level, such as urbanisation.

4.3 ECOLOGY OF MOTH SPATIAL PATTERNS

This study has shown that differences in assemblage structure can be dependant on spatial scale of the observation. This raises the question, what can be the ecological reasons for such patterns be? An obvious and potentially important ecological characteristic when considering moth communities is the relationship between plant species and the Lepidoptera species that they host. It has been suggested by Gaston (1988) that there is a high degree of similarity between processes determining patterns of dynamics both within and between communities of plant-feeding insects. Although plant-animal interaction assemblages present a striking diversity of patterns, they become even more baffling when various researchers portray them with relatively distinct representations (Lewinsohn et al. 2006). However, despite the complexity involved in determining plant- animal relationships, the relationship of Lepidoptera with plant species has been well documented in numerous studies. Hammond and Miller (1998) suggest that as species richness in plants change during succession, the species richness in Lepidoptera may exhibit a coincident change. Usher and Keiller (1998) found that the single

60 CHAPTER 4: Discussion most important factor for determining moth species richness was plant species richness of the woodlands in which their study was conducted. A similar relationship was also found by Southwood et al. (1979) where plant diversity best described the diversity of Heteroptera and Coleoptera in fields in southern England. However, direct correlations between habitat variables, such as specific plant species, and moth communities are not necessarily always clear. Leps et al. (1998) for example looked at the relationship between moth species abundance and variability, and life-history and food supply characteristics and it was found that moth population size was positively correlated with food plant(s) abundance only in monophagous species with a correlation close to zero for population sizes (characterised by mean annual catch) for oligo- and polyphagous species (species which rely on more than one food supply type). Similarly, McGeoch and Chown (1997) found only a weak correlation with habitat and moth assemblage structure and therefore, changes in overall moth population size were not necessarily driven by changes in their food plants, but also maybe influenced by the situation in the surrounding landscape.

Therefore, despite the relatively diverse array of research which attempts to search for a definitive reason, or a combination of reasons, there is still no real clear indicator as to what determines Lepidoptera assemblage structure. Although determining actual reasons that may affect the spatial distributions of Lepidoptera appears to be a somewhat difficult issue to address, taxon- and species-specific factors are likely to dominate causes of species richness, such as resource specificity and availability and species interactions rather

61 CHAPTER 4: Discussion than “macro variables”, such as latitude, potential evapotranspiration (i.e. energy available), climate stability and rainfall, which do not vary so widely on the smaller scales (Ricketts et al., 2002). This study took place within such “macro variables”, where variation in species richness within the landscape maybe occurring with changes in resource specificity and availability as well as species interaction. Factors that contribute to the structure of the moth assemblages may therefore possibly operate at, or be confined to, the localised spatial scales that the observed changes to moth assemblage structures occur.

4.4 PATTERNS AT DIFFERENT TAXONOMIC LEVELS

Distinct and definable patterns with moth were assemblages studied over a hierarchy of spatial scales and such patterns and observations can be used to determine suitable forest management practices. However, to ensure invertebrates are included into assessments, simpler approaches are needed (Andersen et al., 2002) and one method is to identify specimens only to higher taxonomic ranks (Gaston and Williams, 1993). Sorting to a higher taxonomic level, such as family, allows easier and more rapid identification and such simplified approaches can help us assess the needs of invertebrates when managing forests (Brennan et al., 2006). If one taxonomic level predicts another, it can be potentially inferred that the response patterns for the different taxonomic levels are similar. Cardoso et al., (2004) showed that richness at genus and family level predicts species richness and Brennan et al., (2006) found that for assemblage composition, significant congruency existed between taxonomic ranks for spiders. Similarly, it was also found that there was similarity in

62 CHAPTER 4: Discussion the ecological patterns observed at the species and family levels within the data for this investigation.

There are potentially several additional benefits to using a higher taxonomic level, other than efficiency in specimen identification. Curves plotting species or family richness for increasing numbers are likely to approach an asymptote more rapidly at the higher taxonomic level (Brennan et al., 2006) as was observed in this study. Also, using higher taxonomic ranks may be less “noisy” than species level data when examining the effects of major disturbances, although at very high ranks patterns will be lost (Somerfield and Clarke, 1995). Noise within species taxonomic level data may result in small differences in environmental variables and interspecific interactions which may cause species replacements within a family, but not a change in the distribution and/or abundance of that family (Warwick, 1988).

Therefore, differing levels of taxa for Lepidoptera in the eucalypt woodlands of the Sydney Basin were compared. Similar patterns across different levels of taxa were identified, however these patterns tended to be less distinguishable at the family level than the finer taxonomic level of species. For example, richness was found to differ significantly between sites for both species and family, however the significance was greater for species than family. Therefore, if a dataset were to be analysed to the family level only, there is the potential for less pronounced patterns within the community structure to be missed, which otherwise may be identified at the species level. However, flattening of the area curves for both at the landscape level and local level for family, which suggests that the data set was

63 CHAPTER 4: Discussion probably representative of the families that were present at the time of sampling, is in contrast to what was observed at the species taxonomic level of classification. The species area curves showed no signs of flattening. Although differences occurred depending on taxonomic level for the area curves, the ANOVA of family richness data suggested a similar pattern was occurring at this more coarse taxonomic level as what was occurring for species, where there was a relatively distinct homogeneity of assemblage composition across the landscape level, but a patchiness and lack of uniformity at a smaller spatial scale. Therefore, whether observed at family or species, similar patterns appear to be generally occurring over differing spatial scales.

4.5 RARITY

Rare species have historically received much conservation attention because of their apparent vulnerability to extinction (Gaston, 1994). Rare species are of particular interest to conservation biologists (Gaston et al., 1995) and have often been regarded as potentially useful indicators of habitat quality and changes in the environment. Ecologically, rare species tend to have small ranges, small populations or both (Brown 1995). Within this study there was a significant proportion of species which were restricted to a spatially limited range. Although the position and role of rare species in the structure and functioning of communities and ecosystems are poorly known (McGeoch and Chown, 1997a), Summerville et al. (2001) found that rarity was a more important factor in determining community heterogeneity than coarse taxonomic affiliation. In this study, species

64 CHAPTER 4: Discussion found not to be common also had a significant influence on the general heterogeneity of the lepidopteran community in Sydney’s Eucalypt Woodlands. This pattern of heterogeneity maybe influenced by both the ecological and sampling characteristics of the species. It was found that there were 187 species, or 82% of the species sampled, occurring in less than 20% of the traps. These relatively uncommon species accounted for only 23% of total abundance within the dataset, but were responsible for a significant proportion of species assemblage turnover observed between the sites. Bearing in mind that sampling error can bias the estimation of species turnover among sites (Summerville et al. 2001), it was found that the actual variation in turnover was significantly different between the sites. Species which occurred as singletons or doubletons (where only one or two specimens were found in the entire investigation) were a significant cause in the variation in assemblage turnover between the nine sites of this study. However, no variation in assemblage turnover was detected between the sites when just common species was examined (those species which occurred in more than 20% of the traps throughout the study). Similar to the findings of this investigations Summerville et al. (2001) found that nature reserves of similar size and floristic composition, supported equally rich moth communities, but species turnover among reserves suggested that floristically similar habitats supported different species assemblages.

4.6 URBANISATION AND ITS POTENTIAL EFFECT ON MOTH

DISTRIBUTIONS

Urbanisation is a phenomenon that may potentially have a complex

65 CHAPTER 4: Discussion interplay with the structure of ecological communities. Urbanisation is a significant feature of the Sydney Basin landscape which differentiates each of the three national parks from one another. This urban sprawl provides a potentially inhospitable environment for moth communities found in the surrounding eucalypt woodlands due to a lack of essential habitat features. Anthropological practices, such as urbanisation, and also industrialisation and agriculture, can result in population sizes of plant and animal species becoming smaller, their distribution ranges divided and constricted and certain species locally and geographically extinct (Wilcox and Murphy, 1985). Urbanisation may be associated with changes habitat quality such as the invasion and planting of alien flora, the increasing isolation of similarly vegetated areas and increasing disturbance in and around the habitat fragments (McIntyre and Barrett, 1992). This process of fragmentation can have profound effects on biological diversity. Possible consequences may include diminishing species richness, abundance patterns change, patterns of endemism are disrupted, and ecosystem structural properties are modified (Kitching et al., 2000). An understanding of the impact of fragmentation on a range of taxonomically defined assemblages maybe a necessary precursor to the development of both monitoring programs and conservation prescriptions. Such information allows assessment of the severity of fragmentation, and appraisal of whether management intervention are successful (Kitching, 1994).

Although habitat requirements of invertebrates are sometimes not met by nature reserves, in many cases urban habitat patches supporting indigenous vegetation have been shown to be suitable habitats for

66 CHAPTER 4: Discussion insects because they meet the necessary requirements of the species concerned (Kirby, 1992). Although the sites incorporated with this study may not necessarily be considered as urban patches of habitat, the sites were on the fringes of woodland which bordered onto significant area of urbanisation.

This study found that identifiable patterns in moth assemblages were occurring over a spatial scale in the order of magnitude of kilometres with many species displaying a clumping in their distribution and despite the extensive urbanisation of the Sydney Basin, the national parks appear to have equivalent and consistent moth communities with little apparent evidence of an impact from urbanisation. In an exploration of the impact of urbanisation (and habitat quality) and moth-assemblage structure McGeoch and Chown (1997) did not find a strong match between any single habitat variable, or combination thereof. It was suggested however, that habitat variables in a combination provided the best match with the assemblage ordination.

4.7 CONCLUSION

Therefore, although the presence and abundance of a species in a particular location maybe related to factors such as food sources, the complex interplay of environmental factors will also need to be considered. Environmental factors that could possibly be considered might occur on a relatively local scale, for example unique variations in local topography. However, broad-scale factors such as urbanisation may also play a role as well. If the landscape is examined at a broad level, such as comparing one national park to another, Lepidoptera assemblages do not appear to significantly differ

67 CHAPTER 4: Discussion spatially and this facet of the assemblage structure appears to be relatively homogeneous. However, if the landscape is broken down into study sites within the national parks, this higher resolution of observation of the lepidopteran assemblages shows that there are areas containing a unique level of richness and abundance with their own suite of unique species which can be differentiated from another area within the same landscape type.

It has been shown in this study that certain ecological measures of the Lepidoptera community can vary significantly depending on the scale of observation. In addition, although Lepidoptera play an important role within plant and community organisation (Gange and Brown 1989), there is little known in regards to the variation of moth assemblages across a range of scales. This investigation has therefore demonstrated that although there may not always be an appropriate degree of appreciation of a spatial hierarchies within in a landscape, scale can have a significant effect on the interpretation a community and therefore should be taken into account as a potentially important element of any conservation effort.

4.8 AVENUES FOR FUTURE STUDY

This study looked to answer a number of questions by investigating patterns of lepidopteran assemblages over a hierarchy of spatial scales in the eucalypt woodlands of the Sydney Basin. The study of temporal patterns would also potentially provide insight to the assemblage structure of moths in the Sydney Basin eucalypt woodlands. For example, Cook and Graham (1996) through observing temporal fluctuations, demonstrated that measuring

68 CHAPTER 4: Discussion variation of a community through time is an important facet to the greater understanding of any ecological system. This study only considered assemblages at one point over time and replication of time is important to consider in the design of future studies that may follow a similar vein to this one.

Three spatial scales of hierarchy were incorporated into this experiment. A greater number of scales across space could possibly be included into future investigations. This would potentially allow a comparison Lepidoptera assemblage structure across a variety of types of ecosystems, as opposed to just one, as was the case for this study. An increase in the number of spatial scales within the investigation would also enable a greater level of ‘resolution’ to differentiate where spatial scale becomes a significant factor in lepidopteran assemblages.

Specific species life history and environmental variables with consideration of how such variables might interact and play a significant ecological role spatially are also potentially important factors that could be considered when attempting to explain observed patterns in the assemblage structure of the ecological communities of Lepidoptera.

69 References

REFERENCES

Abbott, I., Van Heurck, P., Burbidge, T., and Williams, M. (1992). Arthropod fauna of Jarrah (Eucalyptus mariginata) foliage in Mediterranean forest of Western Australia; spatial and temporal variation in abundance, biomass, guild structure and species composition. Australian Journal of Ecology 17, 263–274.

Abbott, I., Van Heurck, P., Burbidge, T., and Williams, M. (1993). Damage caused by insects and fungi to eucalypt foliage: spatial and temporal patterns in Mediterranean forest of Western Australia. Forest Ecology and Management 58, 85–110.

Andersen, A.N., Benjamin, D., Hoffmann, B.D., Müller, and W.J., Griffiths, A.D. (2002). Using ants as bioindicators in land management: simplifying assessment of ant community responses. Journal of Applied Ecology 39 (1), 8–17.

Australian Bureau of Meteorology. Climate statistics for Australian locations. Retrieved 16 April 2007 from http://www.bom.gov.au/climate/averages/tables/ca_nsw_names.shtml

Baker, R.R. (1985). Moths: population estimates, light-traps and migration. In: ‘Case Studies In Population Biology’. (Ed. L.M. Cook) pp. 188 – 211. (Manchester University Press: Manchester, UK).

Baker, R.R., and Sadvoy, Y. (1978). The distance and nature of the light trap response to moths. Nature 276, 818–821.

70 References

Barlow, H.S. and Woiwod, I.P. (1989). Moth diversity of a tropical forest in Peninsular Malaysia. Journal of Tropical Ecology 5, 37–50.

Basset, Y. (1991). Influence of leaf traits on the spatial distribution of insect herbivores associated with an over-storey rainforest tree. Oecologia 90, 388–393.

Beccaloni, G.W., and Gaston, K.J. (1998). Predicting the species richness of neotropical forest butterflies: Ithomiinae (Lepidoptera: Nymphalidae) as indicators. Biological Conservation 71, 77–86.

Brennan, K.E.C., Ashby, L., Majer, J.D., Moir, M.L., and Koch, J.M. (2006). Simplifying assessment of forest management practices for invertebrates: How effective are higher taxon and habitat surrogates for spiders following prescribed burning? Forest Ecology and Management 231, 138–154.

Brown, J.H. (1995). ‘Macroecology’. (University of Chicago Press: Chicago, USA).

Butler, L., and Strazanac, J. (2000). Occurrence of Lepidoptera on selected host trees in two central Appalachian national forests. Annals of the Entomological Society of America 93, 500–511.

Benson, D.H. (1992). The natural vegetation of the Penrith 1:100,000 map sheet. Cunninghamia 2, 541 – 596.

Benson, D., and Howell, J. (2003). Cumberland Plain Woodland

71 References ecology then and now: interpretations and implications from the work of Robert Brown and others. Cunninghamia 7(4), 631–650.

Benson, J.S. (1999). ‘Setting the Scene: The Native Vegetation of New South Wales. Background paper No 1. Native Vegetation Advisory Council.’ (Department of Land and Water Conservation: Sydney, Australia).

Benson, D., and Howell, J. (2000). ‘Sydney’s Bushland: more than meets the eye.’ (Royal Botanic Gardens: Sydney, Australia)

Benson, D., Howell, J. and McDougall, L. (1996). ‘A Guide to Natural Vegetation in the Hawkesbury-Nepean Catchment.’ (Royal Botanic Gardens: Sydney, Australia).

Benson, J. (1999). ‘Setting the scene: the native vegetation of New South Wales.’ (Native Vegetation Advisory Council of New South Wales: Parramatta, Australia).

Benson, D., Howell, J., and McDougall, L. (1996). 'Native plant species in the Hawkesbury-Nepean catchment.' (Royal Botanic Gardens: Sydney, Australia).

Benson, J.S. (1998). ‘The structure and composition of pre-European native vegetation in south-eastern Australia.’ Proceedings of Caring for the Land conference (Environmental Defenders Office: Sydney, Australia).

72 References

Benson, D., and Howell, J. (1990). ‘Taken for Granted; The Bushland of Sydney and its Suburbs.’ (Kangaroo Press: Sydney, Australia).

Brown, J.H., (1984). On the relationship between abundance and distribution of species. American Naturalist 124, 255–279.

Brown, J.H., Mehiman, D.W., and Stevens, G.C. (1995). Spatial variation in abundance. Ecology 76, 2028–2043.

Burkey, T.V. (1989). Extinction in nature reserves: the effect of fragmentation and the importance of migration between reserve fragments. Oikos 55, 75–81.

Butler, L., Kondo, C., Barrows, E.M., and Townsend, E.C. (1999). Effects of weather conditions and trap types on sampling for richness and abundance of forest macrolepidoptera. Environmental Entomology 28(5), 795–811.

Cappuccino, N., and Kareiva, P. (1985). Coping with a capricious environment: a population study of a rare pierid butterfly. Ecology 66, 152–61.

Cardoso, P., Silva, I., de Oliveira, N.G., and Serrano, A.R.M. (2004). Higher taxa surrogates of spider (Araneae) diversity and their efficiency in conservation. Biological Conservation 117, 453–459.

Carroll, S.S., and Pearson, D.L. (1998). Spatial modelling of butterfly species richness using tiger beetles (Cicindelidae) as bio-indicator

73 References taxon. Ecological Applications 8, 531–543.

Chapman, G.A., Murphy, C.L. (1989) ‘Soil Landscapes of the Sydney 1:100 000 Sheet.’ (Department of Land and Water Conservation: Sydney, Australia).

Chey, V.K., Holloway, J.D., and Speight, M.R. (1997). Diversity of moths in forest plantations and natural forests of Sabah. Bulletin in Entomological Research 87, 371–385.

Clarke, K.R. (1993). Non-parametric multivariate analyses of changes in community structure. Australian Journal Ecology 18, 117 – 43.

Clarke, S., and French, K. (2005). Germination response to heat and smoke of 22 Poaceae species from Cumberland Plain Woodlands. Australian Journal of Botany 53, 445–454.

Coddington, J.A., Young, L.H., and Coyle, F.A. (1996). Estimating spider species richness in a southern Appalachian cove hardwood forest. Journal of Arachnology 24, 123–128.

Common, I.B.F. (1990). ‘Moths of Australia.’ (Melbourne University Press: Melbourne, Australia)

Connor, E.F., and McCoy E.D. (1979). The statistics and biology of the species-area relationship. American Naturalist 113, 791 – 833.

Cook, L.M., and Graham, C.S. (1996). Evenness and species number in some moth populations. Biological Journal of the Linnean Society

74 References

58(1), 75–84.

Daily, G.C., and Ehrlich, P.R. (1995). Preservation of biodiversity in small rainforest patches: rapid evaluations using butterfly trapping. Biodiversity and Conservation 44, 35–55.

Dempster, J.P. (1983). The natural control of populations of butterflies and moths. Biological reviews of the Cambridge Philosophical Society 58, 461–481.

Dempster, J.P. (1989). Fragmentation, isolation and mobility of insect populations. In: ‘Conservation of insects and their habitats’. (Ed. N.M. Collins and J.A. Thomas) pp. 143–53. (Academic: London, UK.) den Boer PJ. 1990. The survival value of dispersal in terrestrial arthropods. Biological Conservation 54: 175-92.

Department of Environment and Conservation (NSW) (n.d.) The Bioregions of New South Wales – their biodiversity, conservation and history – Chapter 15 The Sydney Basin Bioregion. Retrieved 17 April 2006 from http://www.nationalparks.nsw.gov.au/PDFs/sydney_basin_text.pdf

Dickens, M. (1974). ‘The World of Moths.’ (Osprey: Berkshire. UK).

Douglas, S. (1997). Part three. Native vegetation communities of the Sydney metropolitan region. In: ‘Green Web-Sydney. A vegetation management plan for the Sydney region’. (Sydney Regional

75 References

Organisations of Councils: Sydney, Australia).

Ehrlich, P.R. (1992). Population biology of checkerspot butterflies and the preservation of global biodiversity. Oikos 63, 6–12.

Erhardt, A., and Thomas, J.A. (1991). Lepidoptera as indicators of change in the semi-natural grasslands of lowland and upland Europe. In: ‘The conservation of insects and their habitats’. (Ed. N.M. Collins and J.A. Thomas) pp. 213–236. (Academic Press: London, UK).

Fairly, A., and Moore, P. (1995). ‘Native Plants of the Sydney District: An identification Guide.’ (Kangaroo Press, Sydney, Australia).

Faith, D.P., and Walker, P.A. (1996). How do indicator groups provide information about the relative biodiversity of different sets of areas? On hotspots, complementary and pattern-based approaches. Biodiversity Letters 3, 18–25.

Fauth, J.E., Bernardo, J., Camara, M., Resetarits, W.J., Van Buskirk, J., and McCollum, S.A. (1996). Simplifying the jargon of community ecology: A conceptual approach. American Naturalist 147, 282–286.

Field, J.G., Clarke, K.R. and Warwick, R.M. (1982). A practical strategy for analysing ultispecies distribution patterns. Marine Ecology. Progressive Serials 8, 37–52.

Fisher, R.A., Corbet, A.S., and Williams, C.B. (1943). The relation

76 References between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology 12, 42–58.

Fisher, B.L. (1998). Insect behaviour and ecology in conservation planning: Preserving functional species interactions. Annals of the Entomological Society of America 91, 155–158.

Fraser, S.M., and Lawton, J.H. (1994). Host range expansion by British moths onto introduced conifers. Ecological Entomology 19(2), 127–137.

French, K., Callaghan, B., and Hill, S. (2000). Classifying endangered vegetation communities: A case study of Cumberland Plain Woodlands. Pacific Conservation Biology 6, 120–129.

Gange, A.C., and Brown V.K., (1989). Insect herbivory affects size and variability in plant populations. Oikos 56, 351–356.

Gaston, K.J. (1988). Patterns in the local and regional dynamics of moth populations. Oikos 53, 49–57.

Gaston, K.J., and Lawton, J.H. (1988). Patterns in body size, population dynamics and regional distribution of backen herbivores. American Naturalist 132, 662–680.

Gaston, K.J., Lawton, J.H. (1988)a. Patterns in the distribution and abundance of insect populations. Nature 331, 709–712.

77 References

Gaston, K.J., and Lawton, J.H. (1986). Temporal variability of abundance and the distribution of species. Oikos 47, 309–314.

Gaston, K.J., and Williams, P.H. (1993). Mapping the world’s species – the higher taxon approach. Biodiversity Letters 1, 2–8.

Gaston, K.J. (1994). Patterns in geographical ranges of species. Biological Reviews 65, 105–129.

Gaston, K., Blackburn, T., and Loder, N. (1995). Which species are described first? The case of the North American butterflies. Biodiversity Conservation 4, 119–127.

Gibb, H., and Hochuli, D.F. (2002). Habitat fragmentation in an urban environment: large and small fragments support different arthropod assemblages. Biological Conservation 106, 91–100.

Haldane, J.B.S. (1953). Animal populations and their regulation. New Biology 15, 9–24.

Hammond, P.C., and Miller, J.C. (1998). Comparison of the biodiversity of Lepidoptera within three forested ecosystems. Annals of the Entomological Society of America 91(3), 323–328.

Hanski, I. (1987). Cross-correlation in population dynamics and the slope of spatial variance-mean regressions. Oikos 50, 148–151.

Hengeveld, R. and Haeck, J. (1982). The distribution of abundance. Journal of Biogeography 9, 303–316.

78 References

Hill, C. J., Gillison, A. N., and Jones, R. E. (1992). The spatial distribution of rain forest butterflies at three sites in North Queensland. Australian Journal of Tropical Ecology 8, 37–46.

Hill, S., and French, K. (2003). The response of the soil seedbank in Cumberland Plain Woodlands to fire. Austral Ecology 28, 14–22.

Hill, S., and French, K. (2004). The effect of fire and grazing on the regeneration of shrub and eucalypt species in Cumberland Plain Woodland. Australian Journal of Botany 52, 23–29.

Hochuli, D.F., Gibb, H., Burrows, S.E., and Christie, F.J. (2003). Ecology of Sydney's urban fragments: has fragmentation taken the sting out of insect herbivory? In: ‘Urban Wildlife: More than meets the Eye’. (Ed. D. Lunney, and S. Burgin) (Royal Zoological Society of NSW: Sydney, Australia).

Holl, K.D. (1996). The effect of coal surface mine reclaimation on diurnal lepidopteran conservation. Journal of Applied Ecology 33, 225–236.

Holloway, J.D. (1989). Moths In: ‘Tropical rain forest ecosystems. Biogeographical and ecological studies.’ (Ed. H. Lieth, and M. J. A. Werger). pp. 437–453 (Elsevier: Amsterdam, Netherlands).

Intachat, J., Holloway, J.D., and Staines, H. (2001). Effects of weather and phenology on the abundance and diversity of geometroid moths in a natural Malaysian tropical rain forest. Journal of Tropical Ecology

79 References

17(Part 3), 411–429.

Janzen, D.H. (1988). Ecological characterization of a Costa Rican dry forest caterpillar fauna. Biotropica 20, 120–135.

Kerr, J.T.R., Vincent, R. and Currie, D. (1998) Lepidopteran richness patterns in North America. Ecosciences 5, 448–453.

Kirby, P. (1992). ‘Habitat Management for Invertebrates: a Practical Handbook.’ (Royal Society for the Protection of Birds: London, United Kingdom).

Kitching, R.L., (1994). Biodiversity – political responsibility and agendas for research and conservation. Pacific Conservation Biology 1, 279–283.

Kitching, I.J. 1996. Identifying complementary areas for conservation in Thailand: an example using owls, hawkmoths and tiger beetles. Biodiversity and Conservation 5, 84 –858.

Kitching, R.L., Orr, A.G., Thailib, L., Mitchell, H., Hopkins, M.S., and Graham, A.W. (2000). Moth assemblages as indicators of environmental quality in remnant of upland Australian rain forest. Journal of Applied Ecology 37, 284–297.

Landau, D., Prowell, D., and Carlton, C.E. (1999). Intensive versus long-term sampling to assess Lepidopteran diversity in a southern mixed mesophytic forest. Annals of the Entomological Society of

80 References

America 92(3), 435–441.

Lewinsohn, T. M., Prado, P.I., Jordano, P., Bascompte, J., and Olesen, J. M. (2006). Structure in plant–animal interaction assemblages. Oikos 113, 174 – 184.

Leps, J., Spitzer, K., and Jaros, J. (1998). Food plants, species, composition and variability of the moth community in undisturbed forest. Oikos 81, 538–548.

Levin, S.A. (1992). The problem of pattern and scale in ecology. Ecology 73, 1943–1967.

Lowman. M.D. 1985. Temporal and spatial variability in insect grazing of the canopies of five Australian rainforest tree species. Australian Journal of Ecology 10, 7–24

MacArthur, R.H., (1960). On the relative abundance of species. American Naturalist 94, 25–36.

Magurran, A.E. (1985). The diversity of macrolepidoptera in two contrasting woodland habitats at Banagher, Northern Ireland. Proceedings of the Royal Irish Academy 85b, 121–32.

Majer, J.D., Recher, H.F., Perriman, W.S., and Achuthan, N. (1990). Spatial variation of invertebrate abundance the canopies of two Australian eucalyptus forests. Studies in Avian Biology 13, 65–72.

Major, R.E., Christie, F.J., Gowing, G., Cassis, G. and Reid, C.A.M.

81 References

(2003). The effects of habitat configuration on arboreal insects in fragmented woodlands of south-eastern Australia. Biological Conservation 113, 35–48.

Margules, C.R., and Stein, J.L. (1989). Patterns in the distribution of species and the selection of nature reserves: An example from Eucalyptus forests in south-eastern New South Wales. Biological Conservation 50, 219–238.

May, R.M., (1984). An overview: Real and apparent patterns in community structure. In: ‘Ecological communities: conceptual issues and the evidence.’ (Ed. D.R. Strong, D. Simberloff, L.G. Abele, A.B. Thistle). pp. 3–16. (Princeton University Press: Princeton, USA).

McArdle, B.H., and Gaston, K.J. (1992). Comparing population variables. Oikos 64, 610–612.

McArdle, B.H., Gaston, K.J., and Lawton, J.H., (1990). Variation in the size of the animal populations, patterns, problems and artefacts. Journal of Animal Ecology 59, 439–454.

McCreadie, J.W., and Adler, P.H. (1995). Community structure of larval black flies (Diptera: Simuliidae) from the Avalon Peninsular, Newfoundland. Annals of the Entomological Society of America 88, 51–57.

McCreadie, J.W., and Adler, P.H. (1998). Scale, time, space and predictability: species distributions of preimaginal black flies

82 References

(Diptera: Simuliidae). Oecologia 114, 79–92.

McDonnell, M.J., and Pickett, S.T.A. (1990). Ecosystem structure and function along urban-rural gradients: an unexploited opportunity for ecology. Ecology 71, 1232–1237.

McGeoch, M.A., and Chown, S.L. (1997). Impact Of Urbanization On A Gall-Inhabiting Lepidoptera Assemblage - The Importance Of Reserves In Urban Areas. Biodiversity and Conservation 6(7), 979– 993.

McGeoch, M.A., and Chown, S.L. (1997)a. The spatial variability of rare and common species in a gall-inhabiting Lepidoptera community. Ecography 20, 123–131.

McIntyre, S. and Barrett, G.W. (1992). Habitat variegation, an alternative to fragmentation. Conservation Biology 6, 146–147.

McKenna, D.J. 1998. Ant Assemblages of the Hawkesbury Sandstone Woodlands. Honours thesis. (Department of Biological Sciences, University of Wollongong, Australia).

McNaughton, S.J., and Wolf, L.L. (1970). Dominance of the niche in ecological systems. Science 167, 131–139.

Menge, B.A., and Olson, A.M. (1990). Role of scale of environmental factors in regulation of community structure. Trends in Ecological Evolution 5, 52–57.

83 References

Mills, K., and Jakeman, J. (1995). Rainforests of the Illawarra District. (Coachwood Publishing: Sydney, Australia).

Nag, A., and Nath, P. (1991). Effect of moon light and lunar periodicity on the light trap catches of cutworm Agrotis ipsilon moths. Journal of Applied Entomology 111, 358–360.

Naumann, I.D. (1994). Systematic and applied Entomology. (Melbourne University Press: Melbourne, Australia).

Nekola, J.C., and White, P.S. (1999). The distance decay of similarity in biogeography and ecology. Journal of Biogeography 26, 867–878.

Nielsen, E.S., Edwards, E.D., Rangsi, T.V. (1996). ‘Monographs on Australian Lepidoptera: Volume 4. Checklist of the Lepidoptera of Australia.’ (CSIRO: Canberra, Australia)

NSW National Parks and Wildlife Service. (1997). ‘Urban Bushland Biodiversity Survey of Western Sydney.’ (National Parks and Wildlife Service: Sydney, Australia).

NSW National Parks and Wildlife Service. (2000). ‘Royal National Park, Heathcote National Park And Garawarra State Recreation Area Plan Of Management.’

NSW National Parks and Wildlife Service. (2004). ‘Sydney Basin – Climate.’ Retrieved January 24th, 2006, from: http://www.nationalparks.nsw.gov.au/npws.nsf/Content/Sydney+Basi

84 References n+-+climate

Oliver, I., and Beattie, A.J., (1993). A possible method for the rapid assessment of biodiversity. Conservation Biology 3, 563–568.

Oliver, I. and Beattie, A.J. (1998). Spatial fidelity of plant, vertebrate and invertebrate assemblages in multiple-use forest in eastern Australia. Conservation Biology 12(2), 822–835.

Pearson, D.L., and Carroll, S.S., (1998). Global Patterns of species richness: spatial models for conservation planning using bio-indicators and precipitation data. Conservation Biology 12, 809–821.

Pearson, and D.L., Carroll S.S. (1999). The influence of spatial scale on cross-taxon congruence patterns and predictions accuracy of species richness. Journal of Biogeography 26, 1079–1090.

Pimm, S.L., and Redfeam, A. (1988). The variability of population densities. Nature (London). 334, 613–614.

Pogue, M.G. (1999). Preliminary estimates of Lepidoptera diversity from specific sites in the neo-tropics using complementarity and species richness indicators. Journal of the Lepidoptera Society. 53: 65-71.

Pollard, E. (1984). Fluctuations in the abundance of butterflies. Ecological Entomology 9, 179–188.

Pollard, E. (1991). Synchrony of population fluctuations: the

85 References dominant influence of widespread factors on local butterfly populations. Oikos 60, 7-10.

Pyle, R.M., Bentzien, M., and Opler, P. (1981). Insect conservation Annual Records of Entomology. 26, 233–258.

Reader, T., and Hochuli, D.F. (2003). Understanding gregariousness in a larval lepidopteran: the roles of host plant, predation and microclimate. Ecological Entomology 28, 729–737.

Recher, H.F., Hutchings, P.A. and Rosen, S. (1993). The biota of the Hawkesbury-Nepean catchment: reconstruction and restoration. Australian Zoologist 29, 3–41.

Ricketts, T.H., Daily, G.C., Ehrlich, P.R. (2002). Does butterfly diversity predict moth diversity? Testing a popular indicator taxon at local scales. Biological Conservation 103(3), 361–370.

Ricketts, T.H., Daily, G.C., Ehrlich, P.R., and Fay J.P. (2001). Countryside biogeography of moths in a fragmented landscape: Biodiversity in native and agricultural habitats. Conservation Biology 15(2), 378–388.

Ricklefs, R.E., and O’Rourke, K. (1975). Aspect diversity in moths: a temperate-tropical comparison. Evolution 29, 313–324.

Robinson, G.S., and Tuck, K.R. (1993). Diversity and faunistics of small moths (Microlepidoptera) in Bornean rainforest. Ecological

86 References

Entomology 18, 385–393.

Robinson, G.R., Holt, R.D., Gaines, M.S., Hamburg, S.P., Johnson, M.L., Fitch, H.S., and Martinko, E.A. (1992). Diverse and contrasting effects of habitat fragmentation. Science 257, 524–526.

Root, R.B., and Cappuccino, N. (1992). Patterns in population change and the organisation of the insect community associated with Goldenrod. Ecological Entomology. 62(3), 393–420.

Saunders, D.A., Hobbs, R.J., and Margules, C.R. (1991). Biological consequences of ecosystem fragmentation: a review. Conservation Biology 5, 18–32.

Schulze, C.H., Linsenmair, K.E., and Fiedler, K. (2001). Understorey versus canopy: patterns of vertical stratification and diversity among Lepidoptera in a Bornean rainforest. Plant Ecology 53(1–2), 133–152.

Slater, P., and Slater, P. (1974). ‘Australian Moths and Butterflies.’ pp. 3–16 (Rigby: Sydney, Australia)

Smart, S.M., Firbank, L.G., Bunce, R.G.H., Watkins, J.W. (2000). Larvae and farmland birds. Journal of Applied Ecology 37, 398–414.

Solis, M.A., and Pogue, M.G. (1999). Lepidopteran biodiversity: patterns and estimators. American Entomology 45, 206–212.

Somerfield, P.J., and Clarke, K.R. (1995). Taxonomic levels, in marine community studies, revisited. Marine Ecology Progress Series

87 References

127, 113–119.

Southwood, T.R.E., Brown, V.K. and Reader, P.M. (1979). The relationship of plant and insect diversities in succession. Biological Journal of the Linnean Society 12, 327–348.

Southwood, T.R.E., Moran, V.C., and Kennedy, C.E.J. (1982). The richness, abundance and biomass of the arthropod communities on trees. Journal of Animal Ecology 51:, 635–649.

Spitzer, K., and Leps, J. (1988). Determinants of temporal variation in moth abundance. Oikos 53, 31–36.

Stanek, V.J. (1977). ‘Encyclopaedia of Butterflies and moths.’ (Octopus: London, UK).

Strong, D.R., Lawton, J.H., and Southwood. T.R.E. (1984). ‘Insects on plants: community patterns and mechanisms.’ (Blackwell: Oxford, UK).

Summerville, K.S., Metzler, E.H., and Crist, T.O. (2001). Diversity of Lepidoptera in Ohio forests at local and regional scales: How heterogeneous is the fauna? Annals of the Entomological Society of America. 94(4), 583–591.

Sutton, S. L. (1989). The spatial distribution of flying insects. In: ‘Tropical rain forest ecosystems. Biogeographical and ecological studies.’ (Ed. H. Lieth, and M. J. A. Werger). pp. 427–436 (Elsevier:

88 References

Amsterdam, Netherlands).

Sutton, S.L. (1983). The spatial distribution of flying insects in tropical rain forests. In: (eds), Tropical rain forest: ecology and management. (Ed. S.L. Sutton, T.C. Whitmore, and A.C. Chadwick). pp 77–91. (Blackwell: Oxford, UK).

Tammaru, T. and Haukioja, E. (1996). Capital breeders and income breeders among Lepidoptera – consequences to population dynamics. Oikos 77, 561–564.

Taylor, L.R. (1984). Assessing and interpreting the spatial distributions of insect populations. Annual Review of Entomology 29, 321–357

Taylor, L.R., and Woiwood, I.P. (1982). Relationships between inter- and intra- specific spatial and temporal variance/mean population parameters. Journal of Animal Ecology 51, 879–906

Taylor, L.R., Woiwood, I.P., and Perry. J.N. (1980). Variance and the large scale spatial stability of aphids, moths and birds. Journal of Animal Ecology 49, 831–854.

Taylor, L.R., Kempton, R.A. and Woiwod, I.P. (1976). Diversity statistics and the log-series model. Journal of Animal Ecology 45, 255–271.

Thomas, A.W., and Thomas, G.M. (1994). Sampling strategies for

89 References estimating moth species diversity using light trap in north-eastern softwood forest. Journal of the Lepidopteran Society. 48, 85–105.

Thomas, C.D. (1991). Spatial and temporal variability in a butterfly population. Oecologia Berlin. 87, 577–580.

Thomas, C.D., and Harrison, S. (1992). Spatial Dynamics of a Patchily Distributed Butterfly Species. Journal of Animal Ecology 61(2), 437–446.

Thomas, C.D., Thomas, J.A., and Warren, M.S. (1992). Distributions of occupied and vacant butterfly habitats in a fragmented landscape. Journal of Animal Ecology 62, 472–81.

Tilman, D. (1994.) Competition and biodiversity in spatially structured habitats. Ecology 75, 2–16.

Underwood, A.J., and Chapman, M.G. (1998). A method for analysing spatial scales of variation in composition assemblages. Oecologia 117, 570–578.

Underwood, A.J., and Chapman, M.G. (1996). Scales of spatial patterns of distribution of intertidal invertebrates. Oecologia 107, 212– 224.

Usher, M.B., and Keiller, S.W.J. (1998). The macrolepidoptera of farm woodlands: Determinants of diversity and community structure. Biodiversity and Conservation 7(6), 725–748.

90 References

Van Dyke, H., and Matthysen, E. (1999). Habitat fragmentation and insect flight: A changing ‘design’ in a changing landscape? Trends in Ecology and Evolution 14, 172–174.

Wagner, D.L., Peacock, J.W., Carter, J.L., and Talley, S. 1995. Spring caterpillar fauna of oak and blueberry in a Virginia deciduous forest. Annals of the Entomological Society of America. 88, 416–426

Warwick, R.M. (1988). Analysis of community attributes of the macrobenthos of Frierfjord/Langesundfjord at taxonomic levels higher than species. Marine Ecology Progress Series 46, 167–170.

Weaver, J.C. (1995). Indicator species and scale of observation. Conservation Biology 9, 939–942.

Wilcox, B.A., and Murphy, D.D., 1985. Conservation strategy: the effects of fragmentation on extinction. American Naturalist 125, 879– 87.

Wilcox, B.A., Murphy, D.D., Erhlich, P.R, and Austin, G.T. (1986). Insular biogeography of the montane butterfly faunas in the Great Basin: comparison with birds and mammals. Oecologia 69, 188–194.

Wilson, R.J., Thomas, C.D., Fox, R., Roy, D.B., and Kunin, W.E. (2004). Spatial Patterns in species distributions reveal biodiversity change. Nature 432, 393.

Woinarski, J.C.Z., and Culten, J.M. (1984). Distribution of

91 References invertebrates on foliage in forests of south-eastern Australia. Australian Journal of Ecology 9, 207–32.

Woiwod, I.P., and Stewart JA. (1990). Butterflies and moths - migration in the agricultural environment. In: ‘Species Dispersal in Agricultural Habitats.’ (Ed. R.G.H. Bunce and D.C. Howard) pp. 189 – 202. (Bellhaven Press London, UK).

Young, M. (1997). ‘The Natural History of Moths.’ (T and A.D. Poyser Ltd: London, UK).

Zandt, H.S. (1994). A comparison of 3 sampling techniques to estimate the population size of caterpillars in tree. Oecologia 9(3), 399–406.

Zar, J.H. (1984). ‘Biostatisical Analysis.’ (Prentice–Hall: Englewood Cliffs, New Jersey, USA).

92 Appendix A

APPENDIX A – TAXA INDEX

Family a Taxonomically identified b Morpho-species c

Anthela replete, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Anthelidae Anthela nicothoe 12, 13, 14, 15, 16

Utetheisa pulchelloides, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Arctiidae Asura lydia 12, 13, 14, 15, 16, F5, Amata 1

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Cosmopteridae - 12

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Cossidae - 12, 13, 14, 15, 16, 17, 18, 19, 20, 21

Ethmidae - 1, 2

Eupterotidae - 1, 2, 3, 4

Gelechiidae - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 Chlorocoma dichloraria, Phrataria replicataria, 1, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, Hemichloeis exoterica, 14, 15,16, 17, 18, 19, 20, 21, Geometridae Onychopsis lutosaria, 22, 23, 24, 25, 26, 27, 28, 29, Thailaina clara, 30, 31, 32 Phallaria ophusaria,

Lactara - 1

Lasiocampidae - 1

Noctuidae Hedymiges aridoxa 1, 2, 3, 4, 5

Notodontidae - 1, 2, 3, 4

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Oecophoridae - 12, 13, 14, 15,16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27

Plutellidae - 1, 2

Psychidae - 1, 2

Pterophoridae - 1

Hednota bivettella, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, Pyralidae Uresiphita ornthopteralis 12, 13, 14, 15,16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27,

Appendix A

Family a Taxonomically identified b Morpho-species c 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41

Sphingoidea Theretra laterelli -

Saturniidae Opodiphthera eucalypti -

Thyrididae - 1, 2, 3

Tineidae - 1, 2, 3, 4

Tortridae - 1, 2

Thyridoidae - 1

Yponomeutidae - 1, 2, 3, 4, 5, 6, 7, 8

Zygenidae - 1 a Lepidoptera families identified to exist within the data set. b Taxonomically identifiable species within each family of Lepidoptera within the data set. c Species which were not taxonomically identified for each family were allocated a morphospecies number. Each number represents an identification label for a separate morpho-species within the corresponding family (e.g. Tortridae is represented by two morpho-species, Tortridae 1 and Tortridae 2, however Saturniidae is represented by Opodiphthera eucalypti and no morpho-species).