Distribution of Foraminifera and Sediments from Twofold Bay, Eden, New South Wales

Distribution of Foraminifera and sediments from Twofold Bay, Eden, New South Wales.

Lyndsay Dean

A thesis submitted in partial fulfilment of the requirements for the degree of Bachelor of Science (Honours)

Research School of Earth Sciences The Australian National University October 2011

Natural cave in Edrom Bay, southern Twofold Bay, Eden.

Declaration

All research presented in this thesis is my own, except where otherwise acknowledged.

Lyndsay Dean October 27, 2011

Acknowledgements

First of all, I extend my heartfelt thanks and gratitude to my supervisor, Patrick De Deckker. This past year has been a steep learning curve, but your guidance and support have been unwavering. Thank you for the dinner parties, anecdotes, books, surprise presents and the many, many opportunities you’ve provided me. Your confidence in my abilities makes me have confidence in myself. I have no idea how a country girl ended up studying marine micropalaeontology, but it’s undoubtedly in no small part because of you. A massive thank you must also go to Jenny Robb, Arthur Robb, Syd Donaldson and everyone at the South Coast Marine Discovery Centre in Eden. Without your generous supply of time, boats and enthusiasm, this project would not have been possible. Thanks also to Sheree Epe, Ray Tynan, Matthew Proctor and Anne Felton for their supply of local knowledge and willingness to chase endless bits of information for me. Thank you also to Andrew Kiss (ADFA) for loan of the grab sampler, Shane Paxton for the wet splitter, and Graham Nash, Claire Thompson, Michael Ellwood and Steve Eggins for their generous supplies of equipment, support and suggestions throughout the year. Thanks also to the Australian Hydrographic Service for GeoTIFF maps, to Sam Eggins for his help in the field, to Joy McDermid and Maree Coldrick for being the office ladies from heaven, and to my lab support crew – Judith Shelly and Luna Brentegani – for being my encyclopaedias of lab techniques, foraminiferal identifications and general nerdy questions. To the amazing Matthew Lendrum and the extraordinary Kelly Strzepek, there are not enough baked goods in the world to show my appreciation for your proofing prowess. I don’t know why you willingly took the job on, but boy am I glad you did. Huge thanks also to my friends – particularly Anthony, Ellen, Gemma, Jamie, Kate, Leah, Tom, Pikey, and the entire Greenway gang – for your support, forced social contact, and earnest attempts to comprehend what the heck I was studying. Finally, to my family. Thank you for all the tech-support, for listening to my thesis neuroses, for unquestioningly lugging my laptop and plethora of books on every family trip this year, for giving me the time and space to write, and for constantly providing food, beverage and general distractions, regardless of whether I wanted them or not. You guys mean the world to me and I am so, so lucky to have you.

Abstract

The assessment of marine environmental conditions is a basic prerequisite for the development and implementation of viable coastal preservation and management plans. A detailed preliminary study of Twofold Bay (Eden, NSW) was conducted to gather baseline information on foraminiferal compositions and to determine local foraminiferal distribution patterns. This was performed in order to gain an understanding of the active processes within the bay, delineate areas subject to natural environmental stress, and identify pressures from anthropogenic activities. Sedimentological and foraminiferal analyses were conducted on 19 surficial sediment samples collected throughout the bay, and abiotic data gathered from representative bottom water at each sample station. To identify areas of similar traits statistical analyses were applied to each dataset. The results were then combined to provide an overview of the active processes and environmental condition of the bay. A total of 162 foraminiferal species, 156 benthonic and 6 planktonic, have been recognised in the open embayment of Twofold Bay. Fauna was found to be diverse and displayed a high degree of similarity. Four sedimentary associations were identified, with their distribution linked to a range of hydrographic effects including tidal and fluvial action, shoaling and strong intermittent oceanic currents. Three geographically restricted biotopes were identified: the Inner Bay Biotope, the Outer Bay Biotope, and the Harbour Biotope. Foraminiferal compositions in these biotopes were found to be correlated with substrate type and related to the degree of mixing and energy experienced at the sediment-water interface. Fluvio-marine mixing in the bay’s entrance, low sedimentation in the eastern part of the bay and low-energy environments in Eden Harbour strongly influenced faunal populations. Anthropogenic effects were minimal, concentrated in the marine ports of Snug Cove, Cattle Bay and Quarantine Bay. The northern region of Twofold Bay was identified as being subject to considerable environmental stress due to poor circulation and, as such, requires further investigation. This study successfully provides a baseline of foraminiferal compositions within Twofold Bay and explores ecologic controls over their distributions. A list of recommendations for future research is also provided.

iii

CONTENTS

Chapter 1: A Review and Thesis Objectives ...... 1 1.1 Preamble ...... 1 1.2 Introduction to foraminifera in the earth sciences ...... 1 1.2.1 Multivariate analysis in ecological studies ...... 4 1.3 Thesis objectives...... 5 1.4 Previous distribution studies of benthonic foraminifera in coastal environments.. 5 1.4.1 Biotopes of benthonic foraminifera ...... 7 1.5 Previous studies at Twofold Bay ...... 9

Chapter 2: Field Area – Twofold Bay...... 10 2.1 General...... 10 2.2 Industrial history...... 11 2.3 Climate...... 12 2.4 Hydrology ...... 13 2.5 Oceanographic setting...... 15 2.6 Sediments...... 17 2.7 Recent meteorological events ...... 18

Chapter 3: Study Methods and Rationale...... 19 3.1 Sample collection...... 19 3.2 Sample preparation and analysis...... 21 3.2.1 Abiotic parameters of water...... 21 3.2.2 Sediment ...... 22 3.2.3 Foraminifera...... 23 3.2.3.1 Foraminiferal density...... 24 3.2.3.2 Cluster analysis...... 25 3.2.3.3 Wall structure...... 25 3.2.3.4 Association scores...... 25 3.2.3.5 Species diversity...... 26 3.3 Rationale ...... 26

Chapter 4: Results ...... 29 4.1 Abiotic parameters...... 29 4.2 Sedimentological study...... 32 4.3 Foraminifera...... 33

Chapter 5: Discussion ...... 40 5.1 Abiotic facies...... 40 5.2 Sedimentary facies...... 41 5.2.1 Association A: Very Fine–Fine Sand ...... 42 5.2.2 Association B: Fine–Medium Sand ...... 43 5.2.3 Association C: Medium–Very Coarse Sand ...... 43 5.2.4 Association D: Medium Sand...... 44 5.3 General distribution of total foraminiferal populations ...... 45 5.4 Foraminiferal biotopes...... 52 5.4.1 Inner Bay Biotope – Elphidium advenum...... 54 5.4.2 Outer Bay Biotope – Lamellodiscorbus sp. 1 ...... 55

5.4.3 Harbour Biotope – Elphidium advenum var. 1 ...... 56

Chapter 6: Conclusions and Recommendations ...... 60 6.1 Conclusions...... 60 6.2 Recommendations for further research...... 63

References...... 67

APPENDICES ...... 81 Appendix 3.1...... 82 Appendix 4.1...... 83 Appendix 4.2...... 84 Appendix 4.3...... 86 Appendix 4.4...... 87 Appendix 4.5...... 88 Appendix 4.6...... CD-ROM Appendix 4.7...... CD-ROM Appendix 4.8...... CD-ROM

LIST OF TABLES

Table 1.1 ...... 3 Table 2.1 ...... 13 Table 2.2 ...... 15 Table 4.1 ...... 36 Table 4.2 ...... 38 Table 4.3 ...... 39 Table 4.4 ...... 39

LIST OF FIGURES

Figure 2.1...... 10 Figure 2.2...... 13 Figure 2.3...... 14 Figure 2.4...... 15 Figure 2.5...... 16 Figure 2.6...... 17 Figure 3.1...... 20 Figure 3.2...... 21 Figure 3.3...... 22 Figure 3.4...... 23 Figure 3.5...... 24 Figure 4.1...... 30 Figure 4.2...... 30 Figure 4.3...... 31 Figure 4.4...... 32 Figure 4.5...... 33 Figure 4.6...... 34 Figure 4.7...... 35 Figure 4.8...... 35 Figure 4.9...... 35 Figure 4.10...... 36 Figure 4.11...... 36 Figure 4.12...... 38 Figure 5.1...... 42 Figure 5.2...... 45 Figure 5.3...... 46 Figure 5.4...... 48 Figure 5.5...... 52 Figure 5.6...... 53 Figure 5.7...... 53 Figure 5.8...... 57

Chapter 1: A Review and Thesis Objectives

1.1 Preamble

Coastal areas have traditionally been places for human settlement and the development of industry; however, such activities often result in the degradation of nearshore ecosystems. With the burgeoning awareness of environmental health, ways to effectively detect pollution and monitor coastal environments over time are continuously being sought. Single-celled organisms called foraminifera are widely used as environmental bioindicators because individual species are known to inhabit defined niches and habitats. Foraminifera display higher environmental sensitivity than almost any other group of organisms, with distribution limited by factors such as substrate type, elevation relative to the sediment surface, nutrient flux, temperature, salinity and anthropogenic contamination (Scott, 1976; Alve, 1995; Yassini and Jones, 1995; Hayward et al., 1997b; Horton and Murray, 2007). Despite these well established controls, it is increasingly becoming evident that foraminifera do not follow ubiquitous distribution patterns and responses to environmental parameters vary both temporally and spatially (Albani, 1978; Murray, 2006). It is crucial to build up a localised knowledge of how foraminiferal populations are comprised and distributed before effective environmental monitoring can be undertaken. Foraminiferal studies on the south-east Australian coastline are sparse with large regions unexplored. Due to the economic significance of several areas on this coastline, it is extremely important to build up knowledge of how environmental factors affect foraminifera. Localised ecologic studies are crucial in determining the effects of environmental parameters on foraminiferal composition and distribution, for these are the baseline studies which will refine the application of benthonic foraminifera as bioindicators in Australian coastal environments.

1.2 Introduction to foraminifera in the earth sciences

Foraminifera are single celled organisms which live predominantly in the marine environment and constitute the most diverse group of shelled microfauna in modern

oceans (Sen Gupta and Machain-Castillo, 1993; Goldstein, 1999). Three main groups of foraminifera are recognized based on habitat: those that live freely in the water column (planktonic); those that live atop or attached to substrates (benthonic); and those that dwell within the sediment itself (infaunal benthonic). Nearly all foraminifera possess a test or shell that encompasses the organism and separates it from the surrounding water mass. These tests may be organic, agglutinated (constructed of foreign particles and cemented together by the organism), or comprised of calcium carbonate (CaCO3). Once foraminifera have completed gametogenesis or die, their empty tests accumulate on the seafloor. The number, diversity and preservation of these tests is widely utilized for a range of environmental studies within estuarine and nearshore settings (see reviews in Murray, 1991; and Haslett, 2002). Benthonic foraminifera possess several key characteristics which make them attractive environmental indicators: (1) they have short life spans and occupy specific niches, making them sensitive to environmental conditions; (2) they are commonly well preserved in sedimentary records; (3) they are widely distributed, yet relatively immobile; and (4) they live on and within the sediment, which receives and stores many pollutants. Consequently, foraminifera can be affected to a greater degree than plankton or nekton (Yanko et al., 1994). Furthermore, they are (5) taxonomically diverse and evolve quickly, (6) small, abundant and easily sampled, and therefore cost effective, and (7) stress- tolerant species are often among the last organisms to disappear completely from heavily contaminated sites (Schafer, 2000). Benthonic foraminifera have traditionally been used for stratigraphic control, hydrocarbon exploration and as palaeoecological indicators (Table 1.1); however, most of these applications employ isolated species. Total foraminiferal assemblages are largely utilized in studies assessing anthropogenically polluted environments (see reviews in Boltvskoy and Wright, 1976; Alve, 1995; Schafer, 2000). Such studies indicate that population diversity and density are lowest at areas of industrial waste discharge (Hornung et al., 1989; Alve, 1991). Conversely, foraminifera tend to respond positively to the presence of organic pollution with test size and population densities both increasing near the pollution source (Watkins, 1961; Schafer and Cole, 1974; Yanko et al., 1994). Geographical extent or type of pollutant can be determined by measures of population abundance and type of morphological abnormalities within total populations (see reviews in Boltvskoy and Wright, 1976; Alve, 1995; Schafer, 2000).

Table 1.1 – Examples of studies using foraminifera as palaeocological indicators.

The establishment of ubiquitous distribution trends is rare despite the utility of benthonic foraminifera in ecological studies. Local distribution of a species is determined by the parameter/parameters close to the limit of tolerance, however, these limitations vary both spatially and temporally (Murray, 2006). This variability is due to both the extreme heterogeneity among coastal environments, and the intercoupled nature of many oceanographic and hydrological parameters. The degree of flushing a system experiences, behaviour of currents, hydrography, water temperature, pH, natural salinity, level of oxygenation and hydrological inputs are all likely to influence the dispersal of individual foraminiferal species. Due to this fact, field studies are often supported by statistical analyses to clarify the response of foraminiferal assemblages to specific environmental factors.

1.2.1 Multivariate analysis in ecological studies

Cluster analysis (CA) is one of the most widely applied multivariate analytical techniques used to segregate entities (e.g. samples, species, physical measurements) into similar groups or clusters and to quantify the between-group relationships (Parker and Arnold, 1999). By quantitatively looking at physical and hydrological factors in unison, study sites can be defined by their similar traits. These patterns can then be used to analyse areal distribution of significant parameters, such as foraminiferal populations or substrate type, based on the relative abundance of characteristic features. CA programs represent these associations through the generation of dendrograms. The horizontal lines of the dendrogram track the individual entities in the clusters, while the linkage bars reflect the similarity of the clustered entities. The degree of similarity and number of clusters has been used to delineate sedimentary associations, biofacies (groups of organisms characterized by a common occurrence) and biotopes (groups of samples characterized by a common faunal association) (Johnson and Albani, 1973). It is important to remember that CA always produces clusters, regardless of whether any naturally occurring groupings are present in the data (Parker and Arnold, 1999). The recognition of ‘significant’ clusters within a dendrogram is largely a subjective process, most commonly defined based on relative spacing of linkages. Despite the subjective nature of clustering, CA serves to simplify or condense the complicated structure of multivariate data and makes exploration of relationships easier. Similarly, while the combination of field studies and statistical analyses allow correlation between distributions of species and environmental parameters, it must be recognised that these correlations are by nature circumstantial. Correlations can never be definitively proven, which accounts for why field studies from different areas can provide conflicting interpretations of species-parameter relationships (Murray, 2006). Despite these drawbacks, the combination of field studies and statistical treatment is useful in exploring patterns of foraminiferal distribution and clarifying their ecologic controls. This integrated approach serves not only to holistically characterise an environment but also to reveal relationships between benthic biota and environmental processes. These associations can then be used to inform environmental monitoring and palaeoenvironmental interpretation.

1.3 Thesis objectives

This study aims to: 1. characterize the composition and distribution of modern foraminiferal populations in Twofold Bay – a near-pristine open oceanic embayment on the south coast of New South Wales; 2. determine sedimentary associations, foraminiferal biotopes and biofacies of Twofold Bay through statistical correlation and cluster analyses; and 3. determine environmental controls on benthonic foraminiferal associations by comparing foraminiferal distributions against sedimentary characteristics and associated environmental conditions.

Reasons for this study are three-fold. Firstly, data on modern distributions are required to aid the interpretation of past environments; secondly, to provide baseline data for monitoring anthropogenic influence locally; and thirdly, to assist integrated environmental management of Twofold Bay. To achieve these aims, sedimentological and benthonic foraminiferal surveys were conducted, with abiotic data gathered to complement the findings. Statistical analysis was employed to define similar assemblages of both foraminiferal and physical datum, and to determine the principal components of each association. In this study, recent sediments and total foraminiferal populations were employed in order to explore an average of conditions over several years. This method offers an integrated view of the environmental processes and conditions within Twofold Bay. Measures of the chemical parameters of the water mass or analysis of seasonal shift in foraminiferal composition were not attempted as these are temporal factors that require constant monitoring over a time period outside the scope of this study.

1.4 Previous distribution studies of benthonic foraminifera in coastal environments

Foraminifera are exceptionally useful in coastal ecologic studies as no other group of organisms appears to have such clear environmental controls over population variation

(Albani, 1993). Studies in modern estuarine and nearshore environments have recognised significant correlations between zonal distribution and diversity and key environmental parameters such as salinity, dissolved oxygen, sediment grain size, subaerial exposure, organic carbon flux, heavy metals, or presence/absence of vegetative substrates (Scott and Medioli, 1980; Murray, 1991; Cann et al., 2002; Leorri and Cearreta, 2009). Scott and Medioli have demonstrated that foraminiferal distributions and diversities in modern marshes are a function of salinity (Scott, 1974; Scott, 1976; Scott et al., 1976; Scott et al., 1980; Smith et al., 1984). These studies demonstrate that estuaries are dominated by agglutinated morphologies and are characterised by a restricted fauna, especially in the upper reaches where salinities are lowest. Conversely, higher salinities generally favour calcareous morphologies and exhibit higher population diversities. Salinity, tidal exposure and presence of vegetation are also listed as factors which determine brackish-water faunal distributions in New Zealand (Hayward et al. 1996; 1999). However, in shallow-water (<100 m), normal marine environment assemblages are determined to be influenced by water depth, wave and current energy, bottom water oxygen concentrations and substrate type (Hayward et al., 1999). This is also supported by Hulme (1964), Hayward (1990, 1993) and Hayward et al. (1997a, 1997b). In Australia, foraminiferal distributions have been linked to depth (Michie, 1987), salinity and tidal mixing (Haslett, 2001; Wang and Chappell, 2001; Binnie and Cann, 2008; Nash et al., 2010), presence of vegetation (Cann et al., 2002; Nash et al., 2010), temperature and oxygen content (Quilty and Hosie, 2006; Nobes and Uthicke, 2008), bottom currents (Albani 1970), sedimentary environments (Palmeiri, 1976), or a combination of the above. However, the vast majority of distribution studies in Australia make no attempt to reconcile distribution with environmental controls and purely explore taxonomic occurrence, e.g. in Western Australia (McKenzie, 1962; Collins, 1981), Victoria (Apthorpe, 1980), South Australia (Cann and Gostin, 1985; Cann et al., 2000a, 2000b), and Queensland (Woodroffe et al. 2005). In south-eastern Australia there is a significant paucity in the description of foraminiferal distributions in addition to a shortage of ecologic studies. Distribution surveys along the south-eastern coast are geographically limited to the greater Sydney region, with notable studies published on Broken Bay (Albani and Johnson, 1975; Albani, 1978), Port Hacking (Albani, 1968) and Lake Illawarra (Yassini and Jones, 1989). In addition, numerous studies deal with eastern Australia in general rather than one specific location (Albani, 1974; Albani and Yassini, 1989). The main compilations

relevant to general distributions of Australasian foraminifera are those by Albani (1979), Yassini and Jones (1995) and Albani et al. (2001). However, while these works are important collations of recorded distributions, they do not serve to increase resolution of ecologic controls on benthonic foraminifera.

1.4.1 Biotopes of benthonic foraminifera

Numerous studies have shown that foraminifera can be effectively used to delineate ecological provinces (biotopes) (Johnson and Albani, 1973; Albani and Johnson, 1975; Schafer and Wagner, 1978; Michie, 1987). This method is based on a statistical approach using relative species abundance as the data base, with the level of linkage between stations which form a biotope expressing faunal similarity. Biotopes have been used to identify areas with significant ecological abnormality in coastal environments subject to industrial pressures. In Chaleur Bay, Canada, foraminiferal populations were noted to respond to effluent discharge, with the distribution of pollution-tolerant or -sensitive assemblages used to identify geographical extent of pollution (Schafer, 1973). Similarly, species distribution has also been correlated with tolerance of heavy metal contamination, with these biotopes used to determine the extent of contamination (Alve, 1991; Sharifi et al. 1991). Major efforts to integrate foraminiferal distributions with physical and biological parameters have been made by studies conducted overseas and within Australia (Yassini and Jones, 1995; Hayward et al., 1999). In the Lagoon of Venice, four distinct biotopes were identified, with depth, availability of nutrients and type of substrate proposed as factors which limited foraminiferal faunal distribution (Albani and Serandrei Barbero, 1982; Albani et al., 1984). In three separate estuaries in south-western Spain, three main foraminiferal assemblages were identified, with spatial distribution restricted by sedimentary environments, grain size classes and variations in salinity (Ruiz et al., 2005). Conversely, in New Caledonia, it was proposed that 30 foraminiferal species were mainly influenced by depth, with 18 influenced by the mud content of the sediment (Debenay, 1988). In Exmouth Gulf, Western Australia, Orpin et al. (1999) concluded that the distribution of individual foraminiferal species was governed by environmental parameters such as water depth, salinity and bottom-sediment conditions. Similar

correlations between salinity, water depth, foraminiferal diversity and foraminiferal population densities have been noted in studies from New South Wales (as discussed below). In Lake Illawarra, NSW, Yassini and Jones (1989) noted distinct separations between the lagoonal, inlet channel and intertidal sub-littoral marine faunal assemblages. Four foraminiferal assemblages were defined by CA, with lowest foraminiferal diversities recorded from the landward side of the lake and greatest diversities recorded from the region of marine-lagoonal mixing. CA dendrograms indicated that Ammonia beccarii and Cribrononion sydneyensis were distinctly linked to deeper water or silty substrates whereas textularids were associated with shallow sandy substrates. Similar geographic trends in foraminiferal diversity were reported from Broken Bay, NSW, where the marine dominated section of the estuary recorded a greater diversity and density of benthonic foraminifera than the up-stream section (Albani and Johnson, 1975; Albani, 1978). This pattern was attributed to the presence of variable fluvial conditions in the up-stream regions. Additionally, in the southern portion of Broken Bay, four biotopes were recognised from Pitt Water with depth, sediment type and hydrological parameters such as nutrient levels controlling their geographical extent (Johnson and Albani, 1973). Albani (1968) separated Port Hacking into six foraminiferal assemblages comprising three distinct faunal groups between the upper part of the estuary and its seaward extent. Faunal Group 1 (made up of Faunal Groups A and B) characterised the fluvial-dominated section and displayed a restricted fauna. Faunal Group 2 (Faunal Groups C and D) occupied the area between the fluvial and marine sections, while Faunal Group 3 (Faunal Groups E and F) characterised the marine dominated section of the estuary. A comparison of the faunal assemblages with variations in the physical environment showed a general correlation with salinity, depth and proximity to marine influences. Faunal diversities were again noted to be lowest at the most fluvial end of the estuary, and highest at the mouth. Increased NO3 concentration with increased distance from the estuary mouth was also cited as a control, while temperature and oxygen concentration were determined not to influence assemblage distribution (Albani, 1968). The distribution of foraminifera from the Clyde River Estuary and Bateman’s Bay are somewhat more complicated. Haslett (2007) recognised six foraminiferal assemblages, two of which are thought to represent geomorphological factors with the remaining assemblages reflecting downstream increases in salinity. Downstream shifts in

diversity were recorded with the low salinity, inner estuary assemblages dominated by Ammonia beccarii and the higher salinity (normal marine) assemblages of the bay dominated by Elphidium spp. Overall species diversities also increased with seaward progression, with the miliolid group only occurring in the outer parts of the estuary. As these studies indicate, foraminiferal distributions and their limiting factors vary widely. While controls such as salinity and substrate controls recur frequently, no uniform trend can be uncovered for south-eastern Australia. This high degree of variation implies that a national trend of ecological niches is unattainable, and highlights the importance of localised studies.

1.5 Previous studies at Twofold Bay

This study marks the first investigation into the composition, distribution and ecologic controls of foraminifera within the Eden area. Samples from Twofold Bay were collected by Iraj Yassini and appear as type specimens in works by Albani (1989) and Yassini and Jones (1995); however these have not formed part of any specific publication (B.G. Jones, 2011, pers. comm.). One study regarding the distribution of sediments in Twofold Bay has been carried out (Hudson, 1991); however it was restricted to geomorphic mapping and discussion of texture and composition of regional sediments. The study presented here is the first in-depth analysis of the distribution of living foraminifera of Twofold Bay, as well as the first study to define local biotopes and discuss their controls. This study addresses the paucity of information regarding ecologic niches of Australian coastal foraminifera and provides a baseline study for the Eden area.

Chapter 2: Field Area – Twofold Bay

2.1 General

Twofold Bay is a large (31 km2) open oceanic embayment at Eden on the southern New South Wales coast (37º05’S 149º54’S at the bay’s centre), approximately 380 km south of Sydney and 40 km north of the NSW-Victoria border (Fig. 2.1a). It is approximately 4.5 km wide and is comprised of a number of smaller bays (Calle Calle, Nullica, Quarantine, East Boyd; Fig 2.1a) which are separated by prominent rocky headlands (Lookout Point, Torarago Point). Population is concentrated on the north- eastern and south-western foreshores of the bay in the towns of Eden and Boydtown respectively. The primary port areas are on the northern and southern shores of Twofold Bay, in Snug Cove and East Boyd Bay, respectively (Fig. 2.1b). Due to the historical length and areal concentration of settlement, potential

a. b. Figure 2.1 – Location maps showing: a) Twofold Bay and surrounding townships in a regional setting; b) detailed map of the study site showing locations mentioned in the text: QBBR = Quarantine Bay Boat Ramp and Breakwater; MJEW = Main Jetty/Eden Wharf; MJFW = Mooring Jetty/Fishermans Wharf; EBW = Eden Breakwater; WWW = Harris-Daishowa Woodchip Wharf; RAN MPNAW = Royal Australian Navy Multi-purpose Navy Ammunition Wharf; green areas symbolise active mussel leases (after Pollard and Rankin, 2003; NSW DPI, 2005).

environmental pressures may exist in these areas. Other factors potentially influencing foraminiferal densities and distributions are pollution, marine infrastructure, regional climate, hydrological inputs, sediment distribution and oceanographic regimes.

2.2 Industrial history

Marine pollution is a potential concern due to the area’s long industrial history. Twofold Bay served as an export station for whale products, wattle bark, farm produce and livestock from 1828-1930, while several salmon canneries were operational on the bay’s foreshores from 1920-1999 (Bega Valley Council, 2001; Blaxell, 2008; Eden Community Access Centre Inc., 2009). Today, Eden is the largest commercial fishing port in NSW and is a principal export point for timber and timber products (Eden Community Access Centre Inc., 2009). The Harris Daishowa woodchip mill in East Boyd Bay exports, on average, 800,000 tonnes of woodchips and more than 60,000 tonnes of softwood each year (Eden Community Access Centre Inc., 2009). Due to these industries, Eden harbour’s main import commodities are petroleum products, with hydrocarbon contamination a potential issue in the main ports (Pollard and Rankin, 2003). The importance of local aquaculture requires ongoing awareness regarding ecosystem health. A large number of commercial fishing activities operate out of the bay such as ocean haul fisheries, beach seine nets, garfish nets, trawl fisheries, and dive, trap and line fisheries (M. Procter, 2011. pers. comm.). In addition, Twofold Bay is utilised by the mussel aquaculture industry for the production of blue mussels, Mytilus edulis. Since 1998 13.5 hectares of rafts and longlines has been situated between Oman and Cocora Points and an additional 2 ha of longline at the north western end of Torarago Point, Nullica Bay. In 2005, local mussel farming extended to 50ha with supplementary leases developed at Oman Point (4 ha) and Torarago Point (18 ha and 12 ha) (Fig. 2.1b; NSW DPI, 2005). While total revue figures are not available due to confidentiality reasons, the economic importance of local aquaculture necessitates the preservation of Twofold Bay through development and implementation of environmental monitoring programs. Foraminiferal populations may also be affected by the numerous physical structures which influence flow and restrict circulation around the bay (Fig. 2.1b). Snug Cove, Eden’s main port, is characterised by two large jetties (Main Jetty/Eden Wharf,

Mooring Jetty/Fishermans Wharf) and Eden Breakwater, a 100 meter blockade (Harris and O’Brien, 1998; Pollard and Rankin, 2003). To protect small vessel moorings, a second breakwater is situated in Quarantine Bay. While each structure is purposely built to increase harbour shelter, breakwaters affect natural circulation, artificially decrease wave energy, and reduce the effectiveness of flushing. These processes can often result in unintentionally stressful benthonic environments. Environmental stress can also result from the maintenance of physical structures, rather than from the structure itself. On Twofold Bay’s southern headland, two large wharves are situated at East Boyd Bay: The Royal Australian Navy’s Multi-Purpose Naval Ammunitioning Wharf and the Harris-Daishowa Woodchip Wharf. These structures are built on pylons and have no associated breakwater to impede water circulation; however frequent use necessitates periodic maintenance dredging. During construction of the ammunitioning wharf, dredging of its turning basin removed sediment approximately 0.75 m in depth, with maintenance dredging carried out as required (NSW Department of Urban Affairs and Planning, 2000). The repeat removal and redistribution of bottom sediment is highly likely to impact the composition and distribution of sediment-dwelling taxa in the immediate areas.

2.3 Climate

To identify potential seasonality in the relationship between foraminiferal populations and hydrological parameters an understanding of regional climate is necessary. Long-term climate statistics for Eden are unavailable, with the nearest meteorological stations being Green Cape Lighthouse (24.9 km to the south) and Merrimbula Airport (17.9 km to the north). While both locations show similar seasonal trends, Merimbula Airport is deemed to be most representative due to comparable aspect, topography and degree of protection against the Tasman Sea. Regional generalisations about atmospheric climate are therefore based records from Merrimbula Airport. The south coast region displays a typical maritime climate with mild winters, warm summers and year-round rainfall (Table 2.1; Fig. 2.2). Seasonal wind and rainfall patterns recorded at Merrimbula Airport display a weak tendency towards summer rainfall maximum and strong south-west winds during winter months. Summer months are characterized by warm, stable conditions with light to moderate sea breezes

Table 2.1 – Average wind speed direction and recorded at Merrimbula Airport over the period 1998-2011 (modified from the Commonwealth Bureau of Meteorology, 2011a; site no: 069147).

Figure 2.2 – Graphical climate statistics for Merrimbula Airport detailing mean monthly rainfall, and mean minimum (blue line) and maximum (red line) daily temperatures (modified from the Commonwealth Bureau of Meteorology, 2011a). associated with slow moving high pressure cells. During winter months the subtropical high pressure systems shift north, bringing the south coast under the influence of mid latitude cyclones (Linacre and Hobbs, 1977; Holton, 2004). Cold fronts become more frequent and generally cooler conditions prevail with a predominantly south to south- west airstream. Eden would also be affected by mid-latitude cyclones or East Coast Lows which develop in the Taman Sea at any time throughout the year. These low pressure systems would result in strong east to south-easterly onshore winds and prolonged periods of heavy rain which may cause heavy flooding (Commonwealth Bureau of Meteorology, 2007).

2.4 Hydrology

Foraminiferal distributions within Twofold Bay are also potentially affected by depth factors and extent of freshwater influence. Average water depth within Twofold Bay is 20 m, with a maximum of 40 m at the bay entrance (Fig. 2.3). Three main fluvial systems drain into the bay (Fig. 2.4; Table 2.2), as well as a number of estuaries (e.g. Towamba River, Nullica River and Fisheries Creek) and a coastal lake (Curalo Lagoon). Of these fluvial influences, the Towamba River is most dominant, typically recording

Figure 2.3 – Bathymetric map for Twofold Bay (from The Australian Hydrographic Service, 2011: map code AUS192). peak flows during February/March. Low flows are typical between November and January, however the Towamba River experiences generally high average flows all year- round (ANRA, 2009). Foraminiferal distribution would also be influenced by the degree of connection between an estuary and the open bay. Towamba River is a barrier estuary which maintains a permanent connection with the bay, while other barrier estuaries (e.g. Nullica River and Fisheries Creek), have semi-permanent connections (Hudson, 1991). Similarly, Curalo Lagoon lacks a permanent entrance and is only open to the bay after periods of heavy rainfall in the Pallestine catchment. The degree to which an estuary exchanges with the open marine setting greatly affects salinity gradients which, in turn, shape foraminiferal distributions. An intermittent connection would weaken salinity gradients and support brackish-water foraminiferal populations, while a permanent connection would cause the inverse. A permanent connection would also promote the open exchange and migration of fauna between estuarine and open marine settings.

Figure 2.4 – Boundaries for the (1) Towamba; (2) Nullica; and (3) Pallestine catchments which drain into Twofold Bay (after Hudson, 1991).

Table 2.2 – Catchment area and mean annual runoff data for the three major catchments affecting Twofold Bay as described in Fig. 2.4 (after Hudson, 1991).

2.5 Oceanographic setting

The oceanic climate of any region is important in understanding environments and energy conditions experienced at the bottom-sediment level. Typically, the NSW south coast is dominated by a high energy, deep-water wave environment and characterized by highly variable wind waves and persistent southerly swells (Thom and Hall, 1991). However, Twofold Bay is relatively quiet and sheltered due to shoaling, diffraction and refraction significantly reducing deep-water wave power (Hudson, 1991). The bay’s mouth is orientated away from the dominant south-east swell, further resulting in a progressive decrease of wave energy in a counter-clockwise direction around the bay

Figure 2.5 – Broadscale oceanographic processes off the NSW coast represented by sea-surface temperature (SST) satellite images and geostrophic velocities. The Eden study site is marked with a black dot; a) the East Australian Current warming inshore waters of lower NSW during November, 2010; b) an increased southerly reach of warm waters into the Twofold Shelf region during February, 2011; c) a slackening of the EAC, cooling of inshore waters and the development of a warm-water eddy during May, 2011; d) cool inshore waters in the Twofold Shelf region and warm- water eddy projecting into the Tasman Sea during August, 2011. Note the colour scale varies between images (from CSIRO Marine Remote Sensing, 2011).

(Hudson, 1991). Swells in the main harbour of Snug Cove are typically low, with turbulent conditions generally attributed to wind-driven currents and waves (Treloar, 2011). The small difference between the tides on the open coast and within the bay suggests weak tidal currents in the bay (Harris and O’Brien, 1998). In contrast, tidal currents are expected to play a significant role in the movement of sediment at the entrances of the Nullica, Towamba and Fisheries Creek estuaries. Flushing times in Twofold Bay are likely to vary due to differences in the strength of the currents penetrating the main entrance, as well as energy reduction as currents traverse the main bay and enter the smaller bays. This would result in different turnover rates depending on the geographical point and climatic conditions within the bay. However, by calculating the bay’s volume, Forteath (1996) estimated a flushing time of 5-8 days. Water circulation within Twofold Bay is predominantly clockwise, with prevailing north-east to south-east winds restricting water movement between the main headlands (Pollard and Rankin, 2003). Active mixing between the Twofold Bay water mass and the Pacific Ocean occurs only in the presence of extreme southerly winds (A. Felton, 2011, pers. comm.). Eden is also influenced by the East Australian Current (EAC) which brings warm tropical and subtropical water into the cooler temperate waters of NSW. The EAC flows

Figure 2.6 – Sediment distribution within Twofold Bay (after Hudson, 1991; from Yassini and Jones, 1995). strongest in summer, up to twice the strength of winter months (CSIRO Australia, 2001). During weak winter flows, temperate water masses flood the shelf leading to decreased sea surface temperatures (Fig. 2.5). Most of the EAC’s flow leaves the Australian coast somewhere off NSW to head east along the Tasman Front, resulting in decreased influence with southerly extent. By the time the EAC reaches the Twofold Shelf, it is estimated to influence inshore coastal waters only 10% of the time (Pollard et al., 1997). While mostly affected by temperate conditions, this region also experiences warm-core, anti-cyclonic eddies caused by the unstable mixing of warm EAC and cool temperate waters (Griffin et al., 2001).

2.6 Sediments

Seven types of surficial sediments from Twofold Bay are defined based on texture and composition (Hudson, 1991; Fig. 2.6). Mud and muddy sands occupy the deeper central portion of the bay mouth and are surrounded by coarse-grained quartz-carbonate sands containing shell detritus (Yassini and Jones, 1995). Sediments of the inner bay area

are predominantly fine quartz sands with a zone of coarse sand along the southern and outer parts of the bay. Increased mica and lithics of the inner bay sediments is likely to represent reworked fluvial detritus from major source rivers.

2.7 Recent meteorological events

It is important to note that this study took place during and immediately following one of the strongest La Niña events recorded since the late 1800s (Commonwealth Bureau of Meteorology, 2011b). Due to cooling in the Pacific following the 2009-10 El Niño cycle, La Niña conditions were established over Australia by July 2010. As a result, significantly higher than average rainfalls were experienced nationwide, with record high rainfall occurring across much of northern and eastern Australia resulting in widespread flooding between September 2010 and February 2011 (Commonwealth Bureau of Meteorology, 2011b). The townships of Towamba and Kiah (16 km and 8 km from Twofold Bay, respectively) experienced the worst flooding in almost 40 years over March 23-24, 2011, when nearly 600 mm of rain fell on the Towamba River catchment (Campbell, 2011; Stroud, 2011). This flood event resulted in mud plume discharge from Towamba River that filled Twofold Bay for several days (A. Felton, 2011, pers. comm.). Smaller, but still sizable flood events also occurred in both April and June, 2011. The strong La Niña conditions also resulted in warmer than usual sea-surface conditions experienced across the entire eastern seaboard. This enhanced southerly movement of the EAC resulted in abnormally warm waters off southern NSW (Figs. 2.5c, 2.5d). The prevalence of these abnormal conditions potentially exerted significant pressures on foraminiferal populations and their distributions. The implication of these events will only be identified through repeat studies.

Chapter 3: Study Methods and Rationale

3.1 Sample collection

A total of 46 samples were collected from 4 tributary estuaries and Twofold Bay over three separate days of sampling (12/13 February and 19 March 2011, respectively). At each site, three complementary sampling techniques were employed to collect representative living and sub-fossil microfauna and explore environmental conditions: foraminiferal and bottom sedimentary samples were gathered using either a 150 µm mesh hand net, or a grab sampler (as described below); samples were stained with Rose Bengal (rB) and fixed with ethanol; and temperature, dissolved oxygen concentration (DO), electrical conductivity (EC), pH and salinity were measured using an Orion 5-Star Portable Multiparameter Meter and a BRIX refractometer. Sample sites were based on ease of access, adequate distance of separation from adjacent sites, and presence of niches of interest such as weed beds, rock substrate and shell deposits. Eighteen samples were collected from the lower Towamba River between Kiah Inlet and Whelans Swamp against the gradient of freshwater (Fig. 3.1). Three samples were collected from each of Curalo Lagoon, Nullica River and Fisheries Creek. At these sample locations foraminiferal and bottom sediment samples were gathered using a 150 µm mesh hand net raked through the water column and vegetation (when present). Temperature, DO, EC, pH and salinity were measured in situ. Nineteen bottom-sediment samples were collected from sites distributed throughout Twofold Bay using a small grab sampler (Fig. 3.2). Water samples from the sediment/water interface were collected using a Niskin bottle, with temperature, DO, EC, pH and salinity measured ex situ. In Eden Harbour two distinct colour bands were noted from bottom sediment recovered at site TFB14. Two independent sedimentary and foraminiferal samples were taken from this site and are herein labelled as TFB14 (samples taken from the pale sandy layer) and TFB26 (samples from the dark organic layer). Abiotic data for this site remains referenced as TFB14. Excess sea water was carefully drained from each recovered sediment sample, and the samples preserved in ethanol. This ethanol treatment not only prevents microbial decay of foraminiferal tests, but also served to kill individuals that were alive at the time

Figure 3.1 – Google Earth image showing sample locations referenced in this study. From top: C = Curalo Lagoon (red markers), TFB = Twofold Bay (yellow markers), N = Nullica River (purple markers), T = Towamba River (green markers), F = Fisheries Creek (blue markers). For GPS coordinates of each site, refer to Appendix 3.1.

a. b. Figure 3.2 – Bottom sediment samples from Twofold Bay were gathered using a grab sampler. Attached to a rope, the (a) open grab sampler was deployed to the sea floor where it was (b) manually closed and brought back to the surface. Sediment held in the closed grab sampler was then spooned into labelled jars and used for foraminiferal and sedimentary grain size analysis. of collection. Following the method described by Bernhard (2000), bio-reactive rB stain was added to each sample in order to differentiate live and dead specimens. The rB stain absorbs onto proteins of cytoplasm-containing tests (i.e. those which were recently alive), turning them an obvious pinkish red colour.

3.2 Sample preparation and analysis

3.2.1 Abiotic parameters of water

Temperature, DO, EC and pH were measured in the field at each sample location using an Orion 5-Star Portable Multiparameter Meter. Salinity (according to the Practical Salinity Scale) was measured using a BRIX refractometer. Due to time constraints and boat size, water samples from the bay study site were analysed onshore at the completion of sampling. Depth at each sampling site in the bay was recorded using an on-board sonar system, and checked against the Australian Hydrographic Service bathymetric map.

Figure 3.3 – Grain size analysis was performed using the Malvern Mastersizer 2000. A combination of red and blue light sources and 52 light detectors allows this equipment to measure the angular dependence of light scattering of particles spanning a 0.2-2000 µm size range.

Using the statistical software package, ‘IBM SPSS Statistics 19’, similarity matrices were constructed for the abiotic data at each site using Ward’s matrix, Euclidian distance and z-score weighting (so that all parameters contributed equally). Sample associations were determined using dendrograms derived from these similarity matrices, with significant clusters identified by the level of linkage between the clusters. A discussion of single and complete linkage clustering can be found in Sneath and Sokal (1973). Temperature data from all Twofold Bay stations was excluded from this analysis due to the high likelihood of it being compromised by a lag between sampling and testing.

3.2.2 Sediment

A small portion of sediment (5-10 ml) was isolated for particle size analysis using a wet-dispersant Malvern Mastersizer 2000 (Fig. 3.3). This process uses laser diffraction to assess the angular dependence of light scattering, with Mie theory applied in order to theorize the scattering pattern. Grain size is subsequently estimated via comparison with standard scattering patterns, and the proportion of size fractions automatically estimated.

For increased accuracy of Mie theorisations the optical properties of both the particle and dispersant must be accurately identified. Water (refractive index 1.330) was used as a dispersant and quartz (refractive index 1.540, absorption quotient 0.01) was named as the predominant sample component. Background and measurement times were set to 12 seconds each. Each sample was homogenized by vigorous shaking and a fraction of sample was added to the dispersant until an obscuration range of 8-20% was reached. Results are averages of 5 measurements. Grain size distributions were determined using the D(0.1), D(0.5) and D(0.9) scores which indicate the size fraction through which 10%, 50% and 90% of the sample material passes. Similarity matrices were constructed for the average D-scores of each station using the aforementioned statistical technique. Sedimentary provinces were determined from the resulting dendrogram, and each association named for its characteristic grain size.

3.2.3 Foraminifera

Foraminiferal samples were initially divided into smaller, more workable fragments, using a wet-splitter (Fig. 3.4a). One half was stored in a large beaker, whilst the other was split again creating a representative quarter of the initial sample. This process was repeated, working with only one half, until a small portion of the original

8cm 6cm a b Figure 3.4 – Bulk foraminiferal samples were split into smaller, representative sub-samples by passing the entire collected sample through (a) a wet splitter. These sub-samples were then dried and split further using (b) a micro-splitter and other accessories shown in photo.

sample was retained. During each split, shells, shell fragments, pebbles, nematodes, brittle stars, or other biological or vegetal material were removed, labelled, and archived for identification. The split fraction was oven dried at ≤55oC and weighed using digital laboratory scales. The dry fraction was then wet-sieved over 100µm mesh with the retained sediment fraction oven dried at ≤55oC and weighed again. Dry sediment material >100 µm was further split using a micro- splitter (Fig. 3.4b) to create smaller non-biased Figure 3.5 – All foraminiferal work was performed using a stereoscopic sub-samples containing rB stained (living) and microscope and external light source, using 16x-32x magnifications. unstained (non-living) specimens. Micro-splits were weighed and all samples dry picked under a stereoscopic microscope (Fig. 3.5). Each sub-sample of rB stained material was picked for foraminifera, ostracods and other diagnostic biological components including sponge spicules, bryozoan fragments and echinoid spines. Where possible, 200-300 individual specimens were extracted from each sub-sample using a fine watercolour brush. Where possible, foraminifera were identified to species level, or otherwise genus, based on illustrations and descriptions in Albani (1968, 1970, 1978, 1979), Albani et al. (2001), Albani and Yassini (1989), Yassini and Jones (1989, 1995), Cann et al. (2002) and Hayward et al. (1997, 1999). Where variations within species were unable to be refined to subspecies level, genus was identified and these species noted as variants (‘var’). Data sheets were compiled for each sample noting the following characteristics: number of stained and unstained individuals, abrasion levels, and deformities. The following quantitative and statistical treatments were then carried out:

3.2.3.1 Foraminiferal density

For each site, the live:dead (stained:unstained) ratio was calculated. Foraminiferal density counts were standardized by calculating the number of identified foraminifera to a one gram mass of sediment. While this study is primarily concerned with benthonic

foraminifera, the planktonic:benthonic ratio was determined for each site in order to inform oceanic contributions to and depositional features of Twofold Bay .

3.2.3.2 Cluster analysis

As the number of individuals picked per station ranged between 100-300 benthonic specimens, the data are more comparable to species proportions than species densities. Higher counts would more accurately indicate the number of species present in Twofold Bay; however this study is concerned primarily with the most dominant species which would display larger populations regardless of sample size. Species abundance was determined by standardizing counts to proportions of sample totals with statistical analysis applied to this data. Unlike the statistical treatment of abiotic and sedimentary data, foraminiferal similarity matrices were constructed using unweighted data so that the most dominant species contributed greatest to the clustering. Sample associations were then determined from the dendrogram produced.

3.2.3.3 Wall structure

The geographic distribution of dominant wall structures was also determined. All modern foraminifera are classified as one of three groups (Textulariina, Miliolina and Rotaliina) which correspond with agglutinated, porcellaneous and hyaline test wall structures, respectively (Leoblich and Tappan, 1964). In order to determine the abundance of each form of test, data was standardized by converting counts of each form to proportions of sample totals. Data from each station was plotted on a ternary diagram.

3.2.3.4 Association scores

Characteristic foraminiferal species of each sample association were determined following Hayward et al. (1996). All species in each association were ranked using a value calculated to reflect their importance. This value is based on a combination of 5 criteria:

1. Dominance (Dom) – calculated by identifying the 5 most abundant species from each station in an association. The most abundant species is awarded a score of 10, the second most abundant a score of 8 and so on, with a taxon’s dominance determined by its mean score across all stations within that association. 2. Fidelity (Fid) – reflects the degree to which a species is restricted to an association and is expressed as the proportion of stations within the association in which the taxon occurs, minus the proportion of stations outside the association in which it occurs. 3. Abundance (Abund) – the mean abundance of a species within the association. 4. Relative Abundance (Rel) –the mean abundance of the taxon within the association minus its mean abundance throughout all stations. 5. Persistence (Pers) – the proportion of stations within the association in which the taxon occurs. The overall association score of a taxon is calculated using the formula: 4(0.3 Dom + 2 Fid + 0.11 Abund + 0.08 Rel + Pers) in order to give greater weight to criteria in the following decreasing order: Abundance, Relative Abundance, Dominance, Fidelity, Persistence (Hayward et al., 1996). The species with the highest association score is determined to characterise the association (maximum score 100).

3.2.3.5 Species diversity

The Fisher Alpha measure of species richness was calculated for each station within the bay. Based on Fisher et al. (1943), this score reflects biodiversity based on the number of individuals and species recorded from a sample. Values of α (diversity) can be determined from a base graph by plotting the number of species against the number of individuals in a sample (Murray, 1991: 319).

3.3 Rationale

There is no standardized approach to accurately estimate benthonic foraminiferal populations and distributions, despite a vast number of studies. There remains

considerable debate within this field and a number of issues must be acknowledged in order to rationalise the methodology of this study. For appropriate statistical analysis of population estimates, it is assumed that the wet splitting process created representative sub-samples with no directional bias in the machine. This was supported by initial testing of the wet-sampler which achieved almost identical weights of splits regardless of direction or speed of sample throughput. rB was chosen for this study due to its efficiency, inexpensiveness and easy employment in situations where large numbers of samples are acquired over a short time. However, accurately distinguishing live from dead populations is one of the most widely debated issues in benthonic foraminiferal studies (see review in Bernhard, 2000). Much of the controversy regards the phenomena of false positives (the staining of the organic lining of dead foraminifera; Walker et al., 1974); false negatives (the non-staining of recently alive foraminifera; Martin and Steinker, 1973); the difficulties in determining staining outcomes on opaque foraminifera (e.g. agglutinated or miliolid morphologies); and the difference between wet and dry samples in terms of ease of identification (Scott et al., 2001). Another issue is the uncertainty regarding stain times. Where the recommended time for effective staining ranges from hours to weeks, some others indicate rB can react weeks, months or even years after an individual’s death (Bernhard, 1988). While there are other more reliable live v. dead staining methods available (see review in Bernhard, 2000), their use in this project was seen as cost prohibitive and impractical due to the requirement of specialised equipment. Secondly, as a period of months had elapsed between the initial sampling and staining and subsequent laboratory processing and identification work, foraminiferal samples are assumed to be sufficiently stained. Furthermore, while Scott et al. (2001) recommend that rB-stained samples be examined in liquid suspension, this study chose a dry technique due to the ease of working with dry samples and to prevent breathing in alcohol vapours during analysis. Lastly, there is ongoing debate regarding the number of specimens required for a representative study. While 200-300 individual specimens is widely accepted to provide sufficient accuracy for most quantitative examinations (Phleger, 1960), some authors suggest a minimum of 300-400 individuals (Lowe and Walker, 1997) and others recommend 800-1000 (Albani and Johnson, 1975). Hayward et al. (1999) suggest that while extensive picks in qualitative studies are more likely to reveal rarer species, these are not significant in assessment of faunal associations. Hayward and Grenfell (1994)

suggest that 100 specimens are adequate to assesses faunal compositions used for identifying and map associations. This is because the numerically dominant taxa in each fauna will still influence statistical analysis regardless of sample size (Hayward et al., 1999). This study, therefore, aimed for between 200 and 300 individuals, and settled for counts above 100 where necessary.

Chapter 4: Results

While this study collected abiotic, sedimentary and foraminiferal data from Twofold Bay and a number of tributary estuaries, foraminiferal densities in Towamba River, Nullica Creek, Fisheries Creek and Curalo Lagoon were deemed too low to warrant further investigation. Results are therefore restricted to foraminiferal, sedimentary and abiotic characteristics within the open embayment of Twofold Bay.

4.1 Abiotic parameters

Abiotic parameters of Twofold Bay bottom-water are given in Fig. 4.1 and raw data in Appendix 4.1. Sampling depth varies around the bay with stations traversing the mouth of Twofold Bay (TFB10, TFB11, TFB12, TFB13) recording depths consistently greater than 18 m while the remainder of stations range between 7 and 11 m. EC falls within the normal marine range, however salinity remains consistent at unusually low levels (28-30 psu). pH is also constant throughout the bay (8.49-8.61 pH), with no drop recorded at any major port. Two measurements of dissolved oxygen (DO) indicate lowest levels in Edrom Bay (TFB9), with stations in the western and southern portions of Twofold Bay (TFB1, TFB2, TFB3, TFB3, TFB5, TFB6 and TFB7) displaying higher relative oxygen than stations situated in the eastern and western regions (TFB10, TFB11, TFB12, TFB13, TFB14, TFB26, TFB15, TFB16, TFB17 and TFB18). Temperature was shown to steadily increase regardless of station (Fig. 4.1) and is likely attributed to a lag between sampling and analysis. Temperature data is therefore taken to be erroneous and excluded. Cluster analysis indicates three associations which correlate with inner bay, open water, and harbour locations (Fig. 4.2). Analysis of means and standard deviations of each parameter in each association identifies depth and DO to be the association drivers. Association B is clearly defined by depth (20 m mean depth), and Association C records a greater DO (mean 86.3%; 6.45 mg/L) than the other two associations. All other parameters are relatively consistent throughout the clusters.

Figure 4.1 – Depth, temperature, pH, electrical conductivity (EC), salinity and dissolved oxygen (DO in mg/L and % saturation) plots for the Twofold Bay study site.

Figure 4.2 – Dendrograph for the abiotic parameters (DO, EC, salinity, pH, and depth) measured at each station within Twofold Bay. Clusters deemed as significant are coloured and labelled with letters corresponding to associations of similar hydrography: A = inner bay; B = outer bay; C = Eden Harbour. Parameters driving the clusters are given as mean scores ± standard error at 95% confidence and emboldened where significant (Appendix 4.2).

Figure 4.3 – Grain size distribution changes around Twofold Bay. Distributional ranges are determined by plotting particle size (as categorized by British sieve sizes; Appendix 4.4) on the horizontal axis against percentage abundance on the vertical axis.

4.2 Sedimentological study

Results of laser diffraction indicate a distinct sediment grain size change around Twofold Bay (Fig 4.3; Appendix 4.3). The western shoreline is characterized by a coarse silt–fine sand particle range (stations TFB18, TFB1, TFB2, TFB3, TFB4 and TFB5). Grain size increases to medium sands in the mouth of the Bay (TFB10, TFB11) while Snug Cove (TFB14, TFB26 and TFB15) displays an increasingly bimodal range of silts and fine sands. Two sites (TFB6 and TFB13) are dominated by coarse-very coarse sands with 78.83% and 67.64% of each respective sample falling in the 500-2000 µm size range. This trend is further indicated by graphing the distributional range as represented by the D(0.1), D(0.5) and D(0.9) distribution scores (Fig. 4.4). Stations in the western bay (TFB1, TFB2, TFB3, TFB4 and TFB5) display a small distributional range, with the majority of sediments being less than 200 µm. Stations from East Boyd Bay (TFB7, TFB8, TFB9, TFB10, and TFB11) have an increased range, with 40% of sediment falling in the medium-coarse size fraction. The bimodality of the Eden harbour sites (TFB14, TFB26 and TFB15) is demonstrated by the large range between the D(0.1) and D(0.9) scores, with the D(0.1) score indicating an increase in silt content. The D-score range at stations TFB12 and TFB18 is observed to be minimal compared to their surrounding samples.

Figure 4.4 – Trends in grain size regionality throughout Twofold Bay as indicated by the D(0.1), D(0.5) and D(0.9) distribution scores. Two or more data points for the same sample site indicate replicates. Plotting of stations TFB1, TFB2 and TFB3 has been repeated to highlight the sharp decline of coarse grains between TFB18 and the aforementioned stations.

Figure 4.5 – Dendrograph for the grain size distribution of stations within Twofold Bay. Clusters deemed as significant are coloured and labelled, with letter corresponding to sedimentary associations described in the text (A = Very Fine-Fine Sand; B = Fine-Medium Sand; C = Medium-Very Coarse Sand; D = Medium Sand).

Cluster analysis confirms the presence of four distinct sedimentary provinces (Fig. 4.5), with each association characterised by separate principal grain sizes (Fig. 4.6). TFB6 and TFB13 are deemed a separate association due to their very coarse fractions and late clustering. Stations TFB9, TFB12 and TFB15 are statistically correlated with the Bay’s western shore. There is no difference in clusters produced from standard or squared Euclidian distance, indicating a strong degree of relatedness.

4.3 Foraminifera

A total of 162 species, 156 benthonic and 6 planktonic, representing 28 families have been recognised in the open embayment of Twofold Bay (Appendix 4.5). Most of the species found are common to open estuaries and nearshore settings along the eastern coast of Australia (Albani, 1979; Yassini and Jones, 1995).

b. d. a. c.

Figure 4.6 – Sedimentary grain size distributions for each association determined by cluster analysis.

Figure 4.7 – Abundance of total foraminiferal assemblage as a number of individuals estimated per 1 gm of sediment.

Figure 4.8 – Stained (live) versus unstained (dead) Figure 4.9 – Abundance of planktonic and benthonic foraminiferids as a percentage of the benthonic forms as a percentage of total total assemblage. identified foraminiferal assemblage.

Most species are present in nearly all stations with only a small number of species recording a narrow range of occurrence. Ammonia beccarii, Quinqueloculina incisa and Rosalina bradyi are recorded at every station. Species occurring at almost all stations include Cibicides sp. 2, Glabratella australensis, Lamellodiscorbus sp. 1, Miliolinella australis, Q. anguina arenata, Q. seminula and Triloculina trigonula. All sites display a high proponent of the family Elphidiidae with Elphidium crispum, E. macellum, E. advenum var. 1, E. advenum var. 2, E. advenum var. 3 and Cribroelphidium poeyanum also displaying widespread occurrence. Benthonic foraminiferal abundance, as numbers of individuals per 1 gm of sediment, is unevenly distributed (Fig. 4.7). Maximum abundance is recorded at TFB2 and an abundance of less than 120 individuals per gram is found in stations TFB6, TFB7, TFB8, and TFB13. A higher percentage of stained (alive) than non-stained (dead) benthonic foraminifera is also recorded at stations TFB7, TFB8, TFB9, TFB10 and TFB26 (Fig. 4.8). Planktonic forms have a general abundance of less than 10% but are recorded as over 20% of the identified sample at stations TFB10 and TFB15 (Fig. 4.9).

Table 4.1 – Breakdown of deformities encountered (as a percentage of the total benthonic population) .

Station

Figure 4.10 – Ternary plot of the Twofold Bay marine assemblages. P = porcellaneous, Hy = hyaline, A = agglutinated.

Figure 4.11 – Abundance of wall structure types as percentage of total identified benthonic foraminiferal assemblage. Porcellaneous wall structures correspond with the Miliolina type, hyaline structures with the Rotaliina type and agglutinated wall structures with the Textulariina type.

Deformities among benthonic foraminifera were relatively infrequent with high abundances generally restricted to a small number of stations (Table 4.1; Appendix 4.6). The highest occurrence of partial dissolution is found at TFB4 (10%), with dissolution minimal elsewhere. Chamber distortion is only notably recorded from stations in the northern region (stations TFB14, TFB26, TFB16, TFB17 and TFB18) and is largely restricted to irregular chamber sizes in the Elphidiidae group. Almost all stations record broken foraminifera, with the highest occurrence noted at stations TFB6, TFB9 and TFB16 (each 5%). Discolouration is identified as the most common deformity with greatest abundance recorded in the entrance and Eden harbour (TFB11, TFB13, TFB14, TFB26, TFB15, TFB16, TFB17 and TFB18). These stations record an average discoloration rate of 20% with highest abundances found at TFB16 (30%). The average abundance for the remaining stations is 5%. The species most frequently found to be discoloured were Q. incisa and T. trigonula, with Q. seminula, A. beccarii and E. advenum var. 1 also regularly discoloured. Wall structure analysis reveals the dominance of hyaline forms with porcellaneous types sub-dominant throughout the bay (Fig. 4.10; Appendix 4.7). This is expected due to the widespread abundance of the Elphidiidae, Rotaliidae, Cibicididae and Glabratellidae families. Agglutinated structures only obtain >4% abundance in stations from the entrance, Snug Cove and Cattle Bay (TFB11, TFB13, TFB26, TFB15, TFB16, and TFB17), with maximum abundance is recorded at TFB16 (14%) (Fig. 4.11). While all stations fall within normal marine Fisher Alpha diversity values, species diversity is shown to vary around the bay (Table 4.2). Lowest diversity (α 8-9) is recorded in the western bay (station TFB3) while highest diversities (>α 20) are found independently at stations from the eastern and northern regions (TFB11, TFB26 and TFB18). Most stations in the bay’s entrance (TFB1, TFB10, TFB11, TFB13 and TFB14) record diversity scores of α 16 or above, indicating steadily high diversities. Foraminiferal similarity matrices constructed using standard and squared Euclidian distances resulted in slight variations between clustered associations (Fig. 4.12). Each analysis confirms the presence of three distinct foraminiferal provinces which correspond with western shore, open water, and harbour settings. Using the standard Euclidian distance Association A encompasses stations TFB12 and TFB18 (Fig. 4.12a), whilst the squared Euclidian distance links these stations late to Association B (Fig. 4.12b). Association scores were calculated for all benthonic foraminifera in each statistical association (Table 4.3, Table 4.4). Despite the variations in association

Table 4.2 – Graph of Fisher Alpha Diversity Indices for stations within Twofold Bay based on both total benthonic assemblage and stained benthonic assemblage. Note: Fisher Alpha Indices require a minimum of 100 individuals to estimate diversity. Asterisks denote stations where an informed estimate of diversity was made. Blanks indicate stations where 100 individuals were not recovered.

a. b. Figure 4.12 – Dendrographs for the distribution of total foraminiferal assemblages within Twofold Bay: (a) associations determined under standard Euclidian distancing, (b) associations determined under squared Euclidian distancing. Clusters deemed as significant are coloured and labelled with letters corresponding to foraminiferal associations described in the text.

composition, the characteristic taxa of associations A-C remains altered. Due to this feature combined with the late linkage under squared Euclidian distancing all discussions of foraminiferal distributions will herein refer to Fig. 4.12a.

). Note that despite the change despite that Note ). b analysis using Ward's method and analysis using Ward's of Euclidian distancing. distancing. of Euclidian erived for benthonic foraminiferal for erived ations, characteristic species remain characteristic ations, Squared Euclidian Distancing (Fig. 4.12 Distancing Squared Euclidian the associ in stations comprising Table 4.4 – Association scores d scores Association – 4.4 Table determined by cluster associations two methods unaltered between the

). a erived for benthonic foraminiferal for benthonic erived Table 4.3 – Association scores d scores Association – 4.3 Table using Ward's method and determined by cluster analysis associations Standard Euclidian Distancing(Fig. 4.12

Chapter 5: Discussion

Field observations are useful in exploring the spatial shift in a number of individual parameters; however, the use of these trends in establishing a broader environmental narrative is inherently subjective. Cluster analysis is able to characterise statistically similar provinces which field observations alone cannot. By exploring the totality of the physical and faunal data these analyses are able to recognise statistically significant regions of physico-chemical and faunal similarly. When integrated and assessed in unison, these analyses reveal significant ecological provinces with a resolution unachievable through standard methods. It is therefore only through a combination of field and statistical observations that an environment can be confidently characterised in a holistic and objective manner.

5.1 Abiotic facies

Abiotic conditions appear relatively homogenous in Twofold Bay, with the measured parameters showing little change between stations. Depth and DO are exceptions and serve to separate the bay’s three abiotic associations (Fig. 4.2.) Recorded bottom water salinities of Twofold Bay are notably lower than expected (28-30 psu). In principle, bottom waters commonly have salinities of 34-37 psu, while estuarine bottom waters experience strong fluctuations in salinity and oxygen (Potter et al., 2005). Due to the fact only bottom water samples were collected results are representative of this environment only, however two hypotheses are proposed. Reduced salinities may potentially be affecting the entire water mass due to El Niña conditions causing above average rainfall and riverine discharge into the bay prior to sampling. The 5-8 day flushing time estimated by Forteath (1996) could serve to temporarily dilute Twofold Bay and extensively reduce salinities throughout the entire water column. However, submarine groundwater discharge is a more plausible cause of decreased bottom water salinities. Shifts in the freshwater-saltwater interface are controlled by seasonal changes in water table elevation and lateral migration, with the interface regularly moving landward in winter and seaward in summer (Michael et al., 2005). The abnormally wet summer experienced in the Eden area would serve to recharge and lift the

water table, moving the freshwater-seawater interface into the coastal realm with the submarine groundwater discharged into nearshore areas. This would serve to reduce bottom water salinities, until the effect is minimised via dilution. A combination of the two hypotheses is likely, but separation would be hard without isotopic work of the groundwater reservoir. This is beyond the scope of this study. DO levels are highest in the inner bay and decrease in Eden Harbour and the entrance to Twofold Bay, with lowest levels of DO recorded at TFB9 (69.6%). This Edrom Bay location coincides with dense Zosteraceae seagrass beds, including Zostera muelleri and Heterozostera tasmanica species (Pollard and Rankin, 2003). These seagrass beds, located inshore of the Multi-Purpose Naval Ammunitioning Wharf, could result in localised depletion of oxygen due to decomposition of seagrass and subsequent increase in particulate organic material (Eyre and Ferguson, 2002). However, similar seagrass beds are also located in Quarantine Bay (station TFB1) and north East Boyd Bay (station TFB10), and no related drop in DO is recorded (Pollard and Rankin, 2003). Lack of circulation at TFB9 is potentially a factor due to its high degree of natural shelter, however the breakwater at Quarantine Bay would also serve to reduce circulation yet no trend is evident between the two locations. More information is required to explain the cause of this discrepancy. Interestingly, DO levels recorded within Snug Cove (stations TFB14, TFB26, TFB15 and TFB16) are comparative with levels recorded from the open water stations despite recovered sediment being black (7.5YR 2/1) and pungent, suggesting dysoxic conditions. This incongruity may be a result of inaccurate measurements or the sediment reflecting seasonal patterns of dysoxia. Further study of this location is needed to quantify potential stresses on benthonic foraminiferal populations.

5.2 Sedimentary facies

Provinces of sediment with similar grain size as determined through cluster analysis (Fig. 4.5) indicate a number of geographically restricted facies (Fig. 5.1). These facies are likely the result of bottom currents and water circulation within Twofold Bay and are named for the dominant substrate type.

Figure 5.1 – Geographical distribution of sedimentary associations derived from the cluster analysis dendrogram in Fig. 4.5.

5.2.1 Association A: Very Fine–Fine Sand

Stations TFB1, TFB2, TFB3, TFB4, TFB5, TFB9, TFB12 and TFB15. This association is characterised predominantly by a very fine to fine sand fraction, and a notable silty tail (Fig. 4.6a). The bulk of these stations are located in the inner bay, where the fine, mica-rich tailings of Towamba River are distributed by clockwise currents. Despite shallow depths, wave action is reduced due to diffraction resulting in a low to medium energy environment. TFB9 is included in this association due to its high proportion of very fine sands and comparatively large fine silt content. While this may be at odds because it is geographically removed, TFB9 is a logical inclusion as this station is relatively sheltered due to the Harris-Daishowa Woodchip Wharf, the prominent adjacent headlands and its

north-facing aspect. Local seagrass beds further act as a sediment trap, and are potentially responsible for the increased silt component (Pollard and Rankin, 2003). TFB15 records a finer sand component than its surrounding stations, and displays an almost equal distribution of all size ranges spanning from very fine silts to fine sands. This is likely due to natural sheltering within Snug Cove and reduced fluvial and oceanic current penetration. This study is largely in agreement with the Fine Sand (Quartz + Lithics + Mica) province designated by Hudson (1991). Exceptions are TFB9, which falls within the Fine Sand (Carbonates) province, and TFB12, which was characterised as Coarse Sand (Quartz + Carbonate). This could be due to insufficient resolution of provincial boundaries afforded by the small number of samples used in this study, or changes in hydrological characteristics between the two studies.

5.2.2 Association B: Fine–Medium Sand

Stations TFB14, TFB26 and TFB18. This association has a restricted geographical occurrence and is characterised by a varied grain-size range (Fig. 4.6b). This association displays an increased bimodality of grain size distribution, with the recurrence of very fine-medium silt components. A fine- grained fraction is largely expected due to the sheltered nature of this setting. The localised dominance of fine-medium sands could be attributed to the Eden Breakwater serving to impede longshore transport of sands, effecting their precipitation. Distributions of medium sand are less peaky than the Medium Sand association, and stations within this grouping also display an increased fine sand fraction, further indicating calmer conditions than experienced in Association D (as discussed below).

5.2.3 Association C: Medium–Very Coarse Sand

Stations TFB6 and TFB13. This association is recorded from two stations and is treated as a separate association due to its late linkage with Association D (Fig. 4.5) and its large distribution of medium to very coarse sands (Fig. 4.6c). Both stations were difficult to sample with

very little sample retrieved via grab sampling, indicating submarine rock outcropping of neighbouring material (Torarago Point and Lookout Point, respectively). The coarseness of this sample is potentially due to shoaling which removes finer substrates, leaving coarse sands atop bare outcrop. Interestingly, station TFB13 falls clearly within this association, while its neighbouring station, TFB12, records no coarse sand. This indicates a localised and restricted change in bottom-sediment environments and warrants further investigation. Again, this association is in agreement with conclusions drawn by Hudson (1991), with TFB6 falling within the Coarse Sand (Quartz + Lithics + Mica) province and TFB13 within the Coarse Sand (Quartz + Carbonate) province.

5.2.4 Association D: Medium Sand

Stations TFB7, TFB8, TFB10, TFB11, TFB16 and TFB17. This association is largely restricted to high energy sites in the deep entrance where active fluvio-marine mixing occurs, and to a lesser extent in Cattle Bay where energy levels are slightly higher than those encountered in Snug Cove. A comparatively reduced percentage of medium sands and slight bimodality is recorded from stations TFB16 and TFB17 (Fig. 4.6d), however the proportion of fine particles is not nearly as dominant as Association B. This increase of silt is again attributed to mixed energy conditions within Cattle Bay. The remainder of stations are likely affected by marine currents, tidal forcings and fluvial outputs. The highest abundance of medium sands as evidenced at TFB8 could be the effect of slightly stronger bottom water currents related to the outlet of Towamba River, or a relict of dredging for the Naval Ammunitioning Wharf. Station TFB7 has a reduced abundance of medium sands and an increased grain size distribution, likely due to slightly increased distance from the river mouth. In this instance, this association does not appear to correspond with any single province outlined by Hudson (1991). TFB7 and TFB8 fall within Hudson’s Coarse Sand (Quartz + Lithics + Mica) province, TFB10 and TFB11 within the Coarse Sand (Quartz + Carbonate) province, and TFB16 and TFB17 within the Fine Sand (Quartz + Lithics) province. This is likely due to the restricted sampling undertaken in this study. Increased

a. b.

Figure 5.2 – (a) Relative abundance of the three suborders of benthonic foraminifers in Twofold Bay as a percentage of total population; (b) Comparison of the three suborders as a number of species identified from the total benthonic population.

sampling would refine the geographical extent of the facies as well as better define its characteristics.

5.3 General distribution of total foraminiferal populations

The distribution of foraminifera is the result of two main environmental conditions: the normal habitat to which a species has adapted, and the effects of postmortem transport. An analysis of the abundance of individuals (population) and the variety of species (assemblages) makes it possible to recognize overall distribution trends of total populations as well as any slight changes in population composition in an environment. These variations suggest a persistent alteration in the environmental conditions experienced between locations, and are important indicators of different ecological regions. While analysis of total foraminiferal populations can often suggest various boundaries to distribution, integrated statistical analysis is necessary to effectively delineate ecological provinces (biotopes).

In Towamba River, Fisheries Creek, Nullica River and Curalo Lagoon foraminiferal populations were found to be insufficient for this study. Assemblage studies have therefore not been carried out on these locations due to inadequate counting statistics. Two explanations are proposed: that (a) foraminiferal populations are naturally

Figure 5.3 – A visual guide to the distribution of total benthonic foraminiferal populations, as a calculation of number per gram of sediment (from Fig. 4.7). Contouring is automatically generated and as a result, does not accurately represent some stations (e.g. TFB9). However, two regions of high population abundance (TFB2 and TFB12) are easily recognisable, as are zones of reduced abundance (TFB6, TFB7, TFB8, TFB13, TFB16). Low population densities off Whale Beach (Towamba River) are the result of high energy experienced at these sites, while decreased abundances in Eden Harbour are likely due to restricted circulation and periods of dysoxia. limited in these areas; and (b) the flood events of summer 2010-2011 served to flush the foraminiferal populations into Twofold Bay, with recruitment insufficient to be registered in the February 2011 sampling. These hypotheses are supported by the predominance of coarse sand in these samples, indicating regular high energy environments. In conjunction, sites near the outlets of major estuaries may be altered by these flood events. The sea floor may be covered with fluvial sediment which serves to bury the natural biocenose and alter the recorded foraminiferal populations. It is important to note that the following discussion represents a narrow window in time, and that the application of trends to all temporal and seasonal scales would be imprudent. Such trends can only be accurately determined through repeat studies. The benthonic foraminiferal fauna of Twofold Bay is shown diagrammatically in Fig. 5.2 where the number of individuals forming the three suborders are expressed both as a percentage of the total recorded population (3542 individuals), and as the number of species within this population. The species identified in this study are common along the eastern coast of Australia. Most species are present in nearly all stations, suggesting

marine conditions are widespread. All the sampling stations are interrelated and directly dependent upon exchange with marine waters. The distributional map of the total benthonic foraminifera, as a calculation of number per gram of sediment, (Fig. 5.3) reveals very low population densities on the southern side of Twofold Bay at stations TFB6, TFB7 and TFB8. This may be related to the total instability of the bottom sediment, due to active tidal reworking and the effect of shoaling on Torarago Point, or fluvial influences from Towamba River. The low abundance value shown by station TFB13 near Lookout Point suggests relatively high sediment mobility and active reworking. This is supported by coarse grain sediments, and may be caused by ocean current diffraction against submarine rock outcrops creating turbulent localised currents. The relatively high abundance of populations at stations TFB2, TFB9, TFB12 and TFB18 indicate a low sediment input with large collection of foraminiferal tests (Albani, 1993). By contrast, stations TFB4, TFB5, TFB11, and TFB17 reflect either a greater sediment input or elevated erosion, although neither to the extent of stations of lowest abundance. Interestingly, low abundance values at stations TFB1, TFB14, TFB26, TFB15 and TFB16 correspond with areas of high marine traffic (Quarantine Bay and Snug Cove). All stations are sheltered, with dominant fine grain sediment components, suggesting possible nutrient deficit or partial pollution effects due to restricted circulation. Alternatively, the abnormally high population density at stations TFB2, TFB9 and TFB12 could be the result of increased nutrient input. Studies indentify that increased nutrient supply stimulates the growth of benthonic foraminiferal communities, with local populations often several orders of magnitude greater than surrounding areas (Watkins, 1961; Bandy et al., 1965; Schafer and Cole, 1974; Mojtahid et al., 2008). However, studies also indicate that increased nutrient input allows foraminifera to reach their full growth potential and results in anomalously large test sizes and robust forms (Setty and Nigam, 1984; Yanko et al., 1994). Only one station (TFB9) recorded notably robust foraminifera, while TFB2 showed the inverse (small and delicate tests) and TFB12 displayed no apparent change in test size or thickness These inconsistencies suggest variations in the stability of nutrient levels. The reduced size of foraminifera and increased population density at TFB2 suggests an increased abundance of juvenile forms. This mixed population might reflect episodic nutrient pulses which promote foraminiferal reproduction (Loubere and Fariduddin,

Figure 5.4 – Substrate stability shown by the relationship between number of species identified from a station and total number of individuals estimated per 1 gm of sediment.

1999). Conversely, the large and robust forms recorded from TFB9 indicate permanently raised nutrient levels. This is likely attributed to the presence of dense seagrass beds at this location which produce organic carbon, particulate organic matter and cause a localised nutrient increase through seagrass frond decay (Zieman and Zieman, 1989; Eyre and Ferguson, 2002). In addition to a large food supply, seagrass beds also provide stability, shelter and anchor for a diverse biota (Posey, 1988; Sen Gupta, 1999). These seagrass beds and the numerous roles they play may be another explanation for the localised rise in population density and robustness of benthonic foraminifera in Edrom Bay. Stability of the benthonic environment can be assessed by comparing population abundance and population variability, as indicated by the number of species which form the population (Albani, 1993). Stable environments can support a diverse community structure whereas unstable environments tend to have fewer, more highly adapted species. The number of individuals expected in one gram of sediment has been replotted and related to the number of species identified from each Twofold Bay station (Fig. 5.4). As the amount of sediment analysed at each site was not constant, the number of species is expected to be slightly higher than the numbers used here. However, this is sufficient to give a graphic representation of stability. Stations TFB6 and TFB13 appear to be the least stable, which is to be expected from to their coarse sand fractions. Stations TFB7 and TFB8 reflect instabilities expected in an environment subject to constant wave and tidal activity. Similar instabilities are reflected in stations TFB1, TFB14, TFB26, TFB15 and TFB16 which appear to suggest a local increase in sediment mobility. However this is unsupported by sedimentary

analysis. Increased silt component suggests low energy environments with other unknown factors likely responsible for population instabilities. The distribution of planktonic foraminifera reflects the oceanic contribution and depositional features of the open embayment, and transportation into the near-shore by surface and tidal currents. Only adult planktonic individuals were found in this study, which suggests an incomplete life assemblage and implies significant transportation effects. Adult forms are significantly more robust than juveniles and are analogous to sand grains in transportation, being buffered by bottom-water currents and deposited in areas of varied energy flows (P. De Deckker, 2011, pers. comm.). Stations TFB10 and TFB15 record the highest number of planktonic individuals, indicating that conditions at these sites allow for the settling of their tests (Fig. 4.9). The predominance of planktonic species in the entrance region is expected with precipitation of tests likely due to increased ocean current diffraction around Jews Head. Increased abundance at station TFB15 can also be attributed to ocean current transportation, with marine forms being identified up to 80 km inland until a change in current energy initiates their precipitation (Wang and Chappell, 2001). Conversely, low abundance values at TFB7 and TFB8 reflect high energy flows while the near absence of planktonic foraminifera at TFB16 and TFB17 also suggest the existence of consistent stronger currents. As a measure of extant species the number of stained (live) individuals is compared with the number of unstained (dead) individuals (Fig. 4.8). Stations TFB1 and TFB15 record the greatest proportion of unstained tests, with 85-92% of their respective recorded assemblages nonreactive to staining. This is not surprising, as both stations are sheltered harbours where dead tests are likely to have been transported and deposited. This supports the low energy conditions identified for these sites through sedimentary analysis. Stations TFB6, TFB7 TFB8 and TFB9 record the greatest proportion of stained foraminifera, suggesting an overall favourable environment. However, this is misleading as low population densities and coarse grain sediments indicate high energy conditions. A high energy environment would remove dead tests through tidal and current forcings, leaving only those foraminifera who are alive and can sufficiently attach to the substrate. This would account for the high proportion of living foraminifera recovered from this site. As a result, information gathered from these live/dead techniques indicates the same conditions determined through sedimentary analysis and signify a robust measure of environmental characteristics.

Variety within the benthonic population is given through a comparison of the number of individuals with the number of species (Fig. 5.2). Of the three foraminiferal orders, the Rotaliina has the highest occurrence (71% of the total population; 24 common species) with Miliolina the next dominant (27% of total population; 8 commonly recurring species). However, the Textulariina behave atypically by first occurring in the open marine setting and recording highest abundances in Cattle Bay and Snug Cove. This is at odds with estuarine studies which show textularids to have higher abundance in areas of freshwater influence, with higher salinities generally favouring calcareous tests (Scott, 1974, 1976). Interestingly, a similar trend of higher abundance of agglutinated forms with increased marine influence was noted from Broken Bay (Albani, 1978). Further atypically, textularids in Twofold Bay first appear in areas of greatest depth. Traditionally, the textularid to rotalid ratio is used as indicator of palaeodepth – the former decreasing or increasing with falling or rising relative sea level, respectively, and the latter increasing or decreasing, respectively (Mason and Laws, 2011). However, this is unsupported in this study as textularids are completely absent from the western regions despite relatively invariant salinities. The restricted distribution of textularids to the entrance and Eden Harbour may therefore be related to other environmental factors such as comparative oxygen depletion or organic matter ‘build-up’ (Albani, 1978; Mojtahid et al., 2008). Further information is required to explain these distributions. Foraminiferal diversity is high throughout the entire study site according to the Fisher Alpha Diversity Index (Table 4.2). Fisher Alpha is used as an indication of environmental stress, with marine associations having higher mean diversity values (α 6.5-21) than brackish associations (α 1-3.5; Hayward et al., 1999). Diversities within Twofold Bay are lowest at station TFB3 (α 8-9) and TFB4 (α 10-11), with comparatively low diversities also recorded around the Towamba River mouth (TFB6, TFB7 and TFB8: each α 11-12). The low Alpha Index of TFB3 is a result of the restricted number of species identified from this station, despite the overall high population density. On the other hand, stations TFB6, TFB7 and TFB8 display a stable number of species, but with vastly reduced total populations suggesting. This indicates an environment restricted to a small number of species which have a higher tolerance of instable conditions. This restricted niche is supported by low stability scores (Fig. 5.8) and coarse grained sediment (Figs. 4.6c and 4.6d) which suggest high energy conditions. As expected, diversities increase at the entrance with increase marine influence, with the highest diversities recorded at stations TFB11, TFB26 and TFB18 (each >α 20).

While this is potentially influenced by post-mortem distribution and the increased proportion of unstained tests in these sites, diversities remain comparable when Fisher Alpha indexes are calculated using only the stained (live) benthonic population. This is true of all stations that recorded >100 stained individuals, with TFB2 the only exception. This indicates that the Fisher Alpha Index is an accurate indicator of species diversity, regardless of whether total or live populations are used. Despite Twofold Bay being subject to high amounts of marine traffic and several shore-based industries, deformities were relatively infrequent and no gross abnormal morphologies noted from any station (Table 4.1). Frequent small distortions of the test (over inflation of the last chamber and abnormal test periphery) were restricted to Snug Cove and Cattle Bay, and largely effected Elphidium advenum var. 1, Elphidium crispum and Cribroelphidium poeyanum. The largest record of chamber distortion was from station TFB17, however even this was restricted to less than 6% of the total population. While aberrant morphologies are widely used as pollution indicators, percentages of morphological abnormalities in areas of high contamination are vastly higher than experienced in this study (30% of total populations in Yanko et al., 1998; 16% in Romano et al., 2008). The small and infrequent aberrations recorded from Snug Cove and Cattle Bay are likely the result of natural stresses such as availability of food, natural fluxes in pH, or possible periodic dysoxic conditions (Murray, 1963; Le Cadre et al., 2003). Discolouration is frequent within stations TFB11, TFB13, TFB14, TFB26 and TFB15 and has the highest abundance at TFB16, suggesting organic build-up and eutrophication of bottom sediment in Eden Harbour. Similarly, the fluctuation of oxygen within Eden Harbour is suggested by the increased number of foraminifera displaying effects of pyrite or iron oxide staining. While still restricted to less than 6% of the total population, abundance of tests with pyrite infilling or pyretic grains on the external surface is highest within Eden Harbour (stations TFB14, TFB26 and TFB15). This indicates dysoxic bottom sediment, either as a constant condition or intermittently. Seasonality may be an important factor, with increased dysoxia likely over the summer. Repeat studies and down-core analysis would reveal the extent and frequency of such aberrations, and determine their seasonal and annual extent.

5.4 Foraminiferal biotopes

The environmental trends interpreted through analysis of sedimentary and foraminiferal data is strengthened through the production of biotopes and the identification of their principle components. These biotopes not only delineate statistically significant regions of similarity, but provide an overall view of dominant environmental characteristics. Three benthonic faunal associations were identified from a dendrogram produced by cluster analysis (Fig. 4.12a). These clusters, linked on the basis of faunal characteristics, are interpreted as a function of the physico-chemical parameters which may be responsible for the variation in faunal composition (Fig. 5.5). The biotopes are interpreted as reflecting more or less energetic environments, as indicated by average grain size distributions (Fig. 5.6). Based on these correlations, these biotopes indicate a mildly energetic nearshore environment, affected primarily by wave action, a highly energetic, large fluviomarine mixing zone, and a more or less energetic harbour environment affected primarily by tidal currents. The various biotopes are named according to the most conspicuous geographical unit that they occupy, with the terminology used being restricted to Twofold Bay.

Figure 5.5 – Foraminiferal biotopes of Twofold Bay and their relationship interpreted as a function of physico-chemical parameters.

Figure 5.6 – Average grain size distributions for the three Twofold Bay biotopes, recording the average percentage weighting of each grain size fraction.

Figure 5.7 – Geographical distribution of the three biotopes derived for Twofold Bay from the cluster analysis dendrogram in Fig. 4.12a.

Geographic distributions are shown in Fig. 5.7. Each biotope is identified both in terms of its characteristic species (as determined by the highest association score: Table 4.3) and its predominant assemblage (those species with the highest occurrence: Appendix 4.8). It must be remembered that the characterising specie/species is defined by the degree of variability associated with relative abundance and the degree to which that species is

restricted to one association. This accounts for why each association is defined by a separate species, while the predominant assemblage remains similar across all stations.

5.4.1 Inner Bay Biotope – Elphidium advenum

Stations TFB1, TFB2, TFB3, TFB4, TFB5, TFB12, TFB18. Depth: 7-11 m, pH: 8.49-8.67, EC: 54.6-54.9 mS/cm, DO: 74.1-80.6%, salinity: 28-30 psu. Sediment: very fine sand. Diversity: α range: 8-20; mean α: 13-14.

This biotope is dominant in Nullica Bay, and occupies the western landward part of Twofold Bay and a single station in the bay’s entrance. Its boundary with the Outer Bay Biotope on the southern shore corresponds with a distinct change in substrate. The Inner Bay Biotope is characterised by fine sands associated with fluvial currents and medium wave action with the exception of TFB12. This station forms its own subcluster indicating that it is influenced by different factors than its surrounding stations. A coarse silt-fine sand component would indicate reduced energy however this is at odds with its surrounding stations. This abnormality is inexplicable as no geomorphological difference between TFB12 and its surrounding sites is noted. This warrants further research. This biotope is characterised by Elphidium advenum, with Glabratella patelliformis and Rosalina bradyi also producing high association scores. The predominant foraminiferal assemblage consists of nine species:

Rosalina bradyi Elphidium advenum Ammonia beccarii Cribroelphidium poeyanum Glabratella patelliformis Quinqueloculina incisa Quinqueloculina seminula Glabratella australensis Nonionella auris

The high number of individuals (1717) indicates minimal sediment input, with the presence of seagrass beds in Quarantine Bay and a high individual to species ratio indicative of its stability. The presence of C. poeyanum, a small form generally correlated

with low energy environments, also indicates minimum sediment input and movement (Albani, 1993). The absence of textularids indicates freshwater influence in this biotope is not strong. When compared to the remaining biotopes reduced marine influence in the Inner Bay is also suggested by the increased abundance of Rosalina bradyi and Ammonia beccarii and the reduced record of Elphidium spp.. Previous studies have shown A. beccarii to be characteristic of less marine conditions, i.e. low-nutrient fluvial areas, with higher salinity (normal marine) assemblages generally dominated by Elphidium spp. (Albani and Johnson, 1975; Haslett, 2007). It must be noted that while most Elphidium species in this association have low abundances, E. advenum is the exception. The association of A. beccarii and a single Elphidium species is also recorded elsewhere in Australia, where an A. beccarii-E. advenum association is recorded from a muddy sand facies in Gippsland Lakes (Apthorpe, 1980). The same association is widely reported from sheltered sand flats and beaches in New Zealand, indicating certain Elphidium species may be more euryhaline than first thought (Hayward and Hollis, 1994).

5.4.2 Outer Bay Biotope – Lamellodiscorbus sp. 1

Stations TFB6, TFB7, TFB8, TFB9, TFB10, TFB11, TFB13. Depth: 18-23 m, pH: 8.57-8.67, EC: 54.6-54.9 mS/cm, DO: 76.5-79.5%, salinity: 28.5-30 psu. Sediment: medium to coarse sand, in places very coarse sand and slightly shelly. Diversity: α range: 11-20; mean α: 15-16.

This biotope dominates the eastern part of Twofold Bay, from Kiah Inlet to East Boyd Bay and the deep open entrance. A distinct change in sediment type marks its northern boundary with the Harbour Biotope. Characterised by medium to coarse sands, the Outer Bay Biotope is associated with the high energy environments characteristic of strong tidal and ocean currents. This association is characterised by Lameollodiscorbus sp. 1, with E. crispum sub-dominant. Textularids first appear in this biotope, with Textularia conica and T. sagittula atrata the most abundant forms. The Outer Bay Biotope is characterised by the following foraminiferal assemblage:

Lamellodiscorbus sp. 1 Quinqueloculina seminula Elphidium crispum Triloculina trigonula Quinqueloculina incisa Elphidium advenum var. 1 Rosalina bradyi Ammonia beccarii Glabratella australensis Cibicides sp. 2

Energy flow is often demonstrated by increased abundances of foraminifera with thicker, heavier tests (Albani, 1993). When compared to the Inner Bay Biotope, the increased flow energy of this biotope is suggested by the reduced abundance of C. poeyanum, the increased abundance of E. crispum, the presence of Q. seminula, T. trigonula and Miliolinella circularis, and the increased abundance of the miliolid group in general. Further indicative of an unstable environment with high levels of disturbance are species such as Lamellodiscorbus sp. 1 which are represented by very large individuals. Similar faunal associations were recorded in the Dropover and Channel Biotopes of Port Hacking (Albani, 1993), and in the High-Nutrient Fluviomarine Biotope of Broken Bay (Albani and Johnson, 1975). These regions were also associated with mixing between fluvial and open marine waters and increasingly active environments.

5.4.3 Harbour Biotope – Elphidium advenum var. 1

Stations TFB14, TFB26, TFB15, TFB16, TFB17. Depth: 6.9-10 m, pH: 8.49-8.6, EC: 54.7-54.9 mS/cm, DO: 85.2-87.2%, salinity: 28-30 psu. Sediment: medium sand with fine silts, in places very shelly and sometimes muddy. Diversity: α range: 14-20; mean α: 16-20.

This biotope is restricted to Snug Cove and Cattle Bay and is last recorded at Cocora Point. Sediment is characterised by medium sands, with large populations of disarticulated bivalves recovered from Snug Cove. The high distribution of silty sediment types suggest reduced water circulation in this area, with sands likely a result of infrequent storm events or longshore transport. This relative lack of throughflow is suggested to cause the high level of test staining, increase in test deformities, and change in wall structure composition.

a. b.

Figure 5.8 – Population abundance of the three foraminiferal types in each biotope as a percentage of total benthonic population. The Miliolina type corresponds with porcellaneous wall structures, Rotaliina with hyaline wall structure.

The Harbour Biotope shows a reduction of the rotalids, with the textularids recording their highest abundance (Fig. 5.8c). Trochammina inflata is the most dominant textularid (4% mean abundance), while Haplophragmoides, Reophax, and Textularia are also recorded. This biotope is characterised by the widespread species E. advenum var. 1, which records its greatest dominance in this biotope. The foraminiferal association is similar to the Outer Bay Biotope, from which it differs by the reduced abundance of Lamellodiscorbus sp. 1 and greater abundance of E. advenum var. 1. The dominant foraminiferal assemblage is composed of:

Elphidium advenum var. 1 Elphidium advenum var. 3 Quinqueloculina incisa Rosalina bradyi Elphidium advenum var. 2 Lamellodiscorbus sp. 1 Trochammina inflata Elphidium crispum

While substrate is deemed to define this biotope, an interesting incongruity between increased Elphidiidae (6 of the biotope’s top 10 most abundant species) and increased textularids (7% of the biotope’s total population) indicate that other factors are also potentially important. Higher Elphidiidae diversities could be taken to denote areas of increased salinity, possibly due to reduced fluvial mixing (Albani and Johnson, 1975). However, T. inflata has been recorded from estuarine inlets and harbours in New Zealand and suggests a slightly more brackish environment (Murray, 1991; Hayward and Hollis, 1994; Hayward et al., 1999). An Elphidium sp. and Trochammina sp. association has been identified from several athalassic (non-marine) saline lakes in Southern Australia which displayed seasonal variations in environmental conditions and foraminiferal populations (Cann and De Deckker, 1981). It was proposed that Elphidium sp. could survive the summer evaporative episodes where salinities and eutrophication increased, by possibly entering a dormant stage. This periodic dormancy when conditions become too harsh could account for the presence of morphological irregularities in Eden Harbour’s Elphidiidae, but also their restricted degree of deformity. Cann and De Deckker (1981) also proposed that the euryhaline Trochammina sp. dominated in organic rich sediments and could maintain continuing populations in the ephemeral saline lake cycles. It is therefore suggested that the high T. inflata abundance found in this study is representative of increased organic matter and low oxygen levels linked to the summer season, and recorded due to timing of sampling. The seemingly contradictory dominance of Elphidiidae and various textularids found in this biotope is proposed to represent strong seasonal effects in Eden Harbour. Repeat monitoring of the area, particularly concerning the presence of juvenile specimens, would help resolve potential mixed conditions and determine whether different groups reproduce and dominate under separate conditions.

In this study the overall examination of abiotic, sedimentary and foraminiferal data produces comparable indications of environmental characteristics. Field observations recognise the consistent spatial variability of a wide range of parameters such as depth, available oxygen, nutrient supply, sediment type, and effects of waves, tides and currents. While benthonic environments can be interpreted from these parameters alone, this study shows that cluster analysis is a useful addition. Cluster analysis is proven to be useful in determining the geographical extent of similar

parameters, and groups stations in a logical and meaningful manner. These statistical clusters can then be used to knowledgeably characterise the ecologic limits of foraminiferal distributions in a location-specific approach.

Chapter 6: Conclusions and Recommendations

6.1 Conclusions

In conclusion, this study demonstrates the effectiveness of CA in determining the geographical distribution of similar characteristics. CA of abiotic, sedimentary and foraminiferal data from Twofold Bay show consistent associations and indicate significant relationships between sites. In this study the integration of field observations and CA successfully characterises Twofold Bay in a holistic and objective manner, revealing statistically significant environmental processes which are taken to influence foraminiferal distributions.

Through statistical correlation this study recognises four sedimentary associations in Twofold Bay: 1. Very Fine-Fine Sand Association – occupies the western part of the bay and indicates medium energy environments largely linked to tidal and fluvial action. 2. Medium Sand Association – predominantly occupies the open entrance to the bay and indicates high energy environments linked to shoaling, tidal action and intermittent oceanic currents. 3. Medium-Very Coarse Sand Association – geographically restricted to two stations and potentially reflects submarine rock outcropping or strong localised currents. 4. Fine-Medium Sand Association – restricted to the northern bay and reflects mixed conditions. Bimodal substrates in Eden Harbour suggest mixed energy environments, while high silt components from Snug Cove and Cattle Bay indicate chiefly calm ports. The fine sand fraction found in this association is likely the result of longshore transport or storm events.

This study also successfully serves as a baseline study into the composition and distribution of modern foraminiferal populations. From the study of recent benthonic foraminifera in Twofold Bay and its tributaries, a number of conclusions can be drawn:

1. Faunal densities in Curalo Lagoon, Nullica River, Towamba River and Fisheries Creek are notably small. This is likely due to insufficient flushing and high summer temperatures due its shallow nature in the case of Curalo Lagoon, and high energy flows in the rivers. 2. The fauna of Twofold Bay is diverse and has a high degree of similarity, indicating an even distribution of water qualities throughout the study area. 3. No substantial evidence of local pollution has been detected through faunal depletion or morphological deformities. 4. The areas off Torarago Point (TFB6) and Lookout Point (TFB13) have poor and quite unstable populations due to very high energy conditions. 5. Seagrass beds play a significant role in the composition and stability of foraminiferal faunas in East Boyd Bay. 6. The fauna of the western landward extent of Twofold Bay generally display a rich and diverse community of generally small forms. This is parallel with relatively small sediment inputs and a relatively low energy environment, despite being situated near the opening of the Nullica River. 7. The fauna of the eastern areas display higher diversity, reflecting an increased marine influence. This area is also characterised by thicker, more robust forms, reflective of a high energy environment, likely linked to the effects of tidal action and open oceanic circulation. 8. The fauna of Eden Harbour display highest overall diversity and the highest abundance of textularids. Staining of tests is common in this environment with pyrite and iron oxide also identified, reflecting a low energy environment and poor exchange.

This study successfully uses CA to determine three foraminiferal biotopes from Twofold Bay. Through correlation of foraminiferal distributions and sedimentary characteristics and associated environmental conditions, substrate type and energy intensity are proposed as factors which limit foraminiferal geographical distribution and composition in Twofold Bay. The biotopes indentified in this study are concluded as such: 1. Inner Bay Biotope – occupies Nullica Bay and a single station in the entrance to Twofold Bay. This rich biotope is characterised by Elphidium advenum, with subdominant Glabratella patelliformis and Rosalina bradyi. Nine species

comprise the predominant foraminiferal assemblage and textularids are notably absent. Substrate is predominantly very fine-fine sand (63-250 µm). This biotope reflects a medium energy environment with minimum sediment input and largely tidal forcings. 2. Outer Bay Biotope – dominates the eastern part of Twofold Bay, from Kiah Inlet to East Boyd Bay and the deep open entrance. This biotope is characterised by Lameollodiscorbus sp. 1, with E. crispum sub-dominant. The fauna is dominated by forms with large and relatively thicker test and Textularids first appear in this biotope. Substrate is medium-coarse sands (250-1000 µm). This biotope reflects high energy environments characteristic of strong tidal and ocean currents. 3. Harbour Biotope – geographically restricted to Snug Cove and Cattle Bay. Sediment displays bimodality with very fine-coarse silts (2-63 µm) and medium sands (250-500 µm). Characterised by the widespread species E. advenum var. 1, this biotope records the reduced abundance of rotalids and displays the highest abundance of textularids. Staining and test deformation is most abundant in this association. This biotope reflects a low energy environment of reduced water circulation and potentially significant seasonal effects.

While field observations are useful in exploring a number of individual parameters, this study demonstrates that integrated use of statistical treatments is necessary to establish a broader environmental narrative. CA is effective in determining environmental similarities across a broad scale, and can be used to postulate environmental controls on benthonic foraminiferal associations. Localised ecologic studies such as this are necessary to determine relationships between foraminiferal composition and distribution and environmental parameters. Only through similar baseline studies can the application of benthonic foraminifera as bioindicators be refined for application in Australian coastal environments.

6.2 Recommendations for further research

This study was intended only as a preliminary investigation into the distribution of Foraminifera within Twofold Bay, and the factors that limit their geographic trends. The conclusions discussed above are inherently restricted to Twofold Bay, rather than representing ubiquitous characteristics. In order to determine the degree to which trends discovered here can be applied to other locations on the south-eastern coast of New South Wales, further studies are recommended. In order to refine which parameter is driving the correlations, it would be ideal to carry out a comparable study in an area of similar size, shape and marine influence, but one in which aquaculture and marine-based industry is limited and alterations to natural harbours are few. Such studies would provide important baseline information regarding pre-anthropogenic faunal composition and distribution. While anthropogenic activities have not been isolated by this study as a direct limiting factor, subsequent foraminiferal studies in Eden may help clarify the role that human activities, such as shipping, cleaning of vessels, infrastructure construction and dredging, have on benthonic populations. In order to assess anthropogenic stresses on Twofold Bay, an ideal site would be Snug Cove, the main port of Eden. A down-core study of Snug Cove would determine whether species distribution, densities and/or composition have changed in Eden Harbour over time. Analysis of sediment cores would help verify the possibility of mixed conditions within Eden Harbour and resolve the extent to which seasonal effects impact foraminiferal composition. Ideally, these down- core samples would pre-date anthropogenic use of the cove in order to explore the temporal extent of test staining and deformations. This may serve to reveal how long Eden Harbour has displayed dysoxic conditions, and whether conditions have been exacerbated or reduced since intensive human use of the area. While the estuarine data recovered in this study was not informative, the estuaries must be revisited if down-core studies are to be performed. The foraminiferal fauna of the main Twofold Bay tributaries (i.e. Nullica and Towamba Rivers) needs to be identified in order to determine the degree of exchange between the estuarine and marine environments. Such studies would be able to identify how much of the thanatocenose is actually transported. This information would have secondary application with down-core

analyses of flood events or determining the historical freshwater extent into Twofold Bay. Further investigation is desired in order to determine the fidelity of the abiotic data uncovered in this study. This study was conducted in an abnormal year and while there is always variation in abiotic parameters the data recovered by this study may not be reflective of average conditions. While possible explanations for the reduced salinity trend have been discussed, future studies could address this issue by analysing a number of CTD profiles of the water column. This would determine whether salinities are low throughout the entire column or only from one layer. Seasonal profiles would also be beneficial in order to determine whether the reduced salinity uncovered in this study is the result of seasonal freshening, or a year-round occurrence. Ideally, this study should be repeated in an average meteorological year, i.e. one that is under stable neutral conditions and not subject to abnormal rainfall or flooding events. In doing so, true species densities in Twofold Bay and its tributary estuaries may be revealed. Similarly, the ‘normal’ extent of the fluvial influence of Towamba River may be determined. Increased information regarding the bottom water currents of Twofold Bay would help shed light on levels of energy experienced by the benthonic environment. This is desirable in order to explore the extreme juxtaposition of neighbouring environments suggested by stations TFB5, TFB6, TFB7 and TFB11, TFB12, TFB13, and determines the regional extent of these high energy currents. The exploration of bottom-water currents on both a seasonal and temporal scale would help to ascertain their importance both in shaping live benthonic populations, and the distribution of dead tests. Several trace element studies on the foraminifera of Twofold Bay are also recommended. If significant salinity fluxes are found to be a common occurrence, their temporal extent can be explored using trace elements or isotopic analyses of tests in a down-core study. This would not only act as a potential proxy for groundwater discharge events, but also continental rainfall. Trace elements can also be employed in order to examine taphonomic processes, such as the degree to which the thanatocenose of Twofold Bay is the result of transportation or reworking. Such studies would serve to determine how accurately the foraminiferal fossil community reflects the living ecosystem. This study showed that aquatic vegetation is important in shaping foraminiferal populations, however the distribution of seagrass beds are subject to change. Seagrass meadows can change in response to light availability and changes in water movement due

to coastal developments or direct damage caused by fishing pressures, sand extraction and dredging (Butler, 1999). The distribution and temporal extent of seagrass beds could be explored through trace element or isotopic composition studies of foraminiferal tests and current seagrass beds, such as has occurred using the otolith chemistry of juvenile fish (Dorval, et al. 2005; Larkum et al., 2006). Such studies could isolate the occurrence and potential causes of mass vegetative die-off events and, in conjunction with down- core studies, could inform the role seagrasses play in the composition and health of benthonic foraminiferal populations. Seagrass studies could also better refine the relationship between vegetation and specific foraminiferal species. Broadly speaking, there are inherent difficulties with any foraminiferal study such as the one undertaken here. Scott et al. (2001) allude to the species identification problem which describes the near impossibility of identifying a single species in isolation, and the degree to which different micropalaeontologists either under or overestimate species diversity by ‘lumping’ or ‘splitting.’ While this study aimed to identify foraminifera accurately to species level across many samples, many encountered species could not be satisfactorily found in compendia of identified and illustrated species. Additionally, these compendia are based on Scanning Electron Microscope (SEM) micrographs which do not adequately reflect what is seen when using a light microscope. As a result, slight morphological differences noted between individuals were recorded in this study as separate morphological variants. Ideally, identification books should present both light and SEM micrographs of holotype forms and their accepted variants. Random order of identification is also recommended for any similar foraminiferal study in order to minimise problems of identification. While this method may make trend spotting between samples difficult, it may also serve to limit the ‘splitting’ factor, whereby slight differences in morphology are recorded as a separate species entirely. On the same issue, future research into coiling directions of benthonic foraminifera would resolve identification issues encountered by this study. Several miliolids were found to have comparable sinistral and dextral forms, despite being represented in the literature as being uniformly one or the other. As a result, such forms were noted in this study as ‘var. dextral’ or the reverse. No literature was found regarding changes in coiling direction of shallow, benthonic foraminifera. Future studies are therefore needed in order to clarify whether a benthonic species can have mixed directional coiling, or whether the variants uncovered in this study are truly separate

species. Genetic testing would easily resolve whether these forms represent different genetic groups. On a separate issue, this study originally intended to explore foraminiferal distributions in conjunction with similar ostracoda studies. However, an inverse pattern between these two groups was noted to occur in the study site as the sampled estuaries were ostracod rich and foraminifera poor, while the main bay displayed the opposite trend. This is of interest as normally these two microfaunal groups cohabitate and therefore should reflect similar conditions. The incongruity between distribution patterns and relative abundance of foraminifera and ostracods therefore warrants further research regarding the Twofold Bay area. These recommendations would greatly improve our ability to accurately identify foraminiferal compositions and explore the complex relationships between environmental conditions and foraminiferal distributions, not only Twofold Bay, but in a broader Australian context.

References:

Albani, A.D., 1968. Recent foraminiferida from Port Hacking, New South Wales. Contributions from the Foundation of Foraminiferal Research, 19(3): 85-119. Albani, A.D., 1970. A foraminiferal fauna from the eastern continental shelf of Australia. Contributions from the Foundation of Foraminiferal Research, 21(2): 71-80. Albani, A.D., 1974. New benthonic foraminiferida from Australian waters. Journal of Foraminiferal Research, 4(1): 33-39, 1 plate. Albani, A.D., 1978. Recent foraminifera of an estuarine environment in Broken Bay, New South Wales. Australian Journal of Marine and Freshwater Research, 29: 355-398. Albani, A.D., 1979. Recent shallow water foraminiferida from New South Wales. AMSA Handbook No. 3., Australian Marine Sciences Association: Sydney (57pp). Albani, A.D., 1989. Taxonomy and Distribution of Shallow-water Lagenid Foraminiferida from the South-eastern Coast of Australia, Australian Journal of Marine and Freshwater Research, 40: 369-401. Albani, A.D., 1993. Use of benthonic foraminifera to evaluate the environmental characteristics of Port Hacking: significance for coastal engineering works and pollution, pp.28-37, In McNally, G., Knight, M. and Smith, R. (Eds.) Collected Case Studies in Engineering Geology, Hydrology and Environmental Geology, Second Series. Geological Society of Australia: Springwood, N.S.W. Albani, A.D. and Johnson, K.R., 1975. Resolution of Foraminiferal biotopes in Broken Bay, NSW. Journal of the Geological Society of Australia, 22(4): 435-446. Albani, A.D. and Serandrei Barbero, R., 1982. A foraminiferal fauna from the Lagoon of Venice, Italy. Journal of Foraminiferal Research, 12: 234-241. Albani, A.D. and Yassini, I., 1989. Taxonomy and Distribution of Shallow-water Lagenid Foraminiferida from the South-eastern Coast of Australia. Australian Journal of Marine and Freshwater Research, 40: 369-401. Albani, A.D., Favero, V. and Serandrei Barbero, R., 1984. Benthonic foraminifera as indicators of intertidal environments. Geo-Marine Letters, 4: 43-47. Albani, A.D., Hayward, B.W., Grenfell, H.R. and Lombardo, R., 2001. Foraminifera from the South Pacific. Australian Biological Resources Study. CD-ROM – ISBN 0 7334 1835 X.

Alve, E., 1991. Benthic foraminifera in sediment cores reflecting heavy metal pollution in Sørfjord, western Norway. Journal of Foraminiferal Research, 21: 1-19. Alve, E., 1995. Benthic foraminiferal responses to estuarine pollution: a review. Journal of Foraminiferal Research, 25: 190-203. Anderson, O.R., Lee, J.J. and Faber, W.W., 1991. Collection, maintenance and culture methods for the study of living Foraminifera, pp. 225-358, In Lee, J.J. and Anderson, O.R. (Eds.) Biology of Foraminifera, Academic Press: London. Apthorpe, M., 1980. Foraminiferal distribution in the estuarine Gippsland Lakes system, Victoria. Royal Society of Victoria Proceedings, 91(1/2): 207-232. Australian Natural Resources Atlas, 2009. Basin & Surface Water Management Area: Towamba River, Department of Sustainability, Environment, Water, Population and Communities: Canberra, ACT, viewed 21 February, 2011. Bandy, O.L. and Arn, R.E., 1960. Concepts of foraminiferal paleoecology. AAPG Bulletin, 44(12): 1921-1932. Bandy, O.L., Ingle, J.C. and Resig, J.M., 1965. Foraminiferal trends, Hyperion outfall, California. Limnology and Oceanography, 9: 112-123. Bega Valley Shire Council, 2001. A bit of history. Bega Valley Shire Council, Bega, New South Wales, viewed 21 August, 2011. Bemis, B.E., Spero, H.J., Bijma, J. and Lea, D.W., 1998. Re-evaluation of the oxygen isotopic composition of planktonic foraminifera: Experimental results and revised paleotemperature equations. Paleoceanography, 13(2): 150-160. Berger, W.H., Killingley, J.S. and Vincent, E., 1978. Stable isotopes in deep-sea carbonates: box-core ERDC-92, West Equatorial Pacific. Oceanologica Acta, 1: 203-216. Bernhard, J.M., 1988. Postmortem vital staining in benthic foraminifera: duration and importance in population and distributional studies. Journal of Foraminiferal Research, 18: 143-146. Bernhard, J.M., 2000. Distinguishing live from dead foraminifera; methods review and proper applications. Micropalaeontology, 46(Supplement 1): 38-46.

Bignot, G., 1985. Elements of micropalaeontology: the microfossils – their geological and palaeobiological applications, Graham and Trotman Ltd: London (217pp). Binnie, M.N. and Cann, J.H., 2008. The distribution of foraminifera in Gulf St Vincent – indicators of sea-level change, pp.38-55, In Shepherd S.A., Bryars, S., Kirkegaard, I., Harbison, P. and Jennings, J.T. (Eds.) Natural History of Gulf St Vincent, Royal Society of South Australia Inc: Adelaide. Blaxell, G., 2008. Eden and Twofold Bay, Part 3: Promises Never Quite Realised. Afloat Magazine, November, viewed 21 August, 2011. Boersma, A., 1978. Foraminifera, pp.19-78, In Haq, B.U., Boersma, A. (Eds.) Introduction to Marine Micropaleontology, Elsevier: New York. Boltovskoy, E. and Wright, R., 1976. Recent Foraminifera. Dr. W. Junk Publishers: The Hague (515pp). Bowser, S.S., Alexander, S.P., Stockton, W.L. and DeLaca, T.E., 1992. Extracellular matrix augments mechanical properties of pseudopodia in the carnivorous foraminiferan Astrammina rara: role in prey capture. Journal of Protozoology, 39: 724-32. Boyle, E.A., 1981. Cadmium, zinc, copper and barium in foraminifera tests. Earth and Planetary Science Letters, 53: 11-35. Butler, A.J. 1999. Seagrass in Australia: Strategic Review and Development of an R&D Plan, FRDC Project 98/223. CSIRO Publishing: Collingwood, VIC (225pp). Campbell, I., 2011. NSW Election 2011: Local flood victims frustrated by red tape. Australian Broadcasting Commission. Cann J.H, Bourman, R.P. and Barnett, E.J., 2000a, Holocene foraminifera as indicators of relative estuarine-lagoonal and oceanic influences in estuarine sediments of the River Murray, South Australia. Quaternary Research, 53: 378-391. Cann, J.H., Belperio, A.P., Gostin, V.A. and Murray-Wallace, C.V., 1988. Sea-level history, 45,000 to 30,000 yr B.P., inferred from benthic foraminifera, Gulf St. Vincent, South Australia. Quaternary Research, 29(2): 153-175.

Cann J.H., Belperio, A.P. and Murray-Wallace, C.V., 2000b. Late Quaternary paleosealevels and paleoenvironemtns inferred from foraminifera, Northern Spencer Gulf, South Australia. Journal of Foraminiferal Research, 30(1): 29-53. Cann, J.H. and De Deckker, P., 1981. Fossil Quaternary and living foraminiferids from athalassic (non-marine) saline lakes, Southern Australia. Journal of Paleontology, 55(3): 660-670. Cann, J.H. and Gostin, V.A., 1985. Coastal sedimentary facies and foraminiferal biofacies of the St Kilda Formation at Port Grawler, South Australia. Transactions of the Royal Society of South Australia, 109: 121-142. Cann, J.H., Harvey, N., Barnett, E.J., Belperio, A.P. and Bourman, R.P., 2002. Foraminiferal biofacies eco-succession and Holocene sealevels, Port Pirie, South Australia. Marine Micropaleontology, 44: 31-55. Collins, A.C., 1981. Holocene Foraminifera from the Fitzroy River estuary, North West Australia. Memoirs of the National Museum of Victoria, 42: 1-6, 2 plates. Commonwealth Bureau of Meteorology, 2007. About East Coast Lows, Commonwealth of Australia, NSW Regional Office, viewed 24 August, 2011. Commonwealth Bureau of Meteorology, 2011a, Climate Statistics for Australian Locations: Summary statistics Merimbula Airport AWS, Commonwealth of Australia, viewed 15 February, 2011. Commonwealth Bureau of Meteorology, 2011b. ENSO Feature - The 2010-11 La Niña: Australia soaked by one of the strongest events on record, Commonwealth of Australia, viewed 21 August, 2011. <[http://www.bom.gov.au/climate/enso/feature/ENSO-feature.shtml> Commonwealth Scientific and Industrial Research Organisation, 2001. The East Australian Current. Information Sheet No. 37, CSIRO Australia. Commonwealth Scientific and Industrial Research Organisation, 2011. Marine Remote Sensing Ocean Currents, online database of sea surface temperature and atmospheric pressure images, CSIRO Australia, viewed 29 August, 2011. Debenay, J.P., 1988. Foraminifera larger than 0.5mm in the southwestern lagoon of New Caledonia: distribution related to abiotic properties. Journal of Foraminiferal Research, 18(2): 158-175.

Dorval, E., Jones, C.M., Hannigan, R. and van Montfrans, J., 2005. Can otolith chemistry be used for identifying essential seagrass habitats for juvenile spotted seatrout, Cynoscion nebulosus, in Chesapeake Bay? Marine and Freshwater Research, 56: 645-653. Eden Community Access Centre, 2009. A Brief History of Eden, Eden Community Access Centre Inc., Eden, New South Wales, viewed 21 August, 2011. Eggins, S.M., De Deckker P. and Marshall, J., 2003. Mg/Ca variation in planktonic foraminifera tests: implications for reconstructing palaeo-seawater temperature and habitat migration. Earth Planetary Science Letters, 212: 291-306. Eyre, B.D. and Ferguson, A.J.P., 2002. Comparison of carbon production and decomposition, benthic nutrient fluxes and denitrification in seagrass, phytoplankton, benthic microalgae- and macroalgae- dominated warm-temperate Australian lagoons. Marine Ecology Progress Series, 229: 43-59. Fisher, R.A., Corbet, A.S. and Williams, C.B., 1943. The relationship between the number of species and the number of individuals in a random sample of an animal population. Journal of Animal Ecology, 12(1): 42-58. Forteath, G.N.R., 1996. Environmental Impact Statement (incorporating Fauna Impact Statement) evaluating mussel aquaculture in Twofold Bay. Pacific Seafood Consulting Group Pty Ltd.: South Melbourne, Victoria. (171pp), cited in Harris, P. and O’Brien, P., 1998. Australian Ports Environmental Data and Risk Analysis. Phase 1: Literature Review. Report Prepared for Australian Quarantine Inspection Service (AQIS). Petroleum and Marine Division, Australian Geological Survey Organisation: Canberra (181pp). Goldstein, S.T., 1999. Foraminifera: a biological overview, pp. 37-56, In Sen Gupta, B.K. (Ed.) Modern Foraminifera. Kluwer Academic Publishers: Boston, M.A. Griffin, D., Cahill, M., Rathbone, C. and White, N., 2001. South-East Fishery Ocean Movies. CSIRO Marine Research, viewed 9 July, 2011. Harris, P. and O’Brien, P., 1998. Australian Ports Environmental Data and Risk Analysis. Phase 1: Literature Review. Report Prepared for Australian Quarantine Inspection Service (AQIS). Petroleum and Marine Division, Australian Geological Survey Organisation: Canberra, viewed 6 September, 2011.

Haslett, S.K., 2001. The palaeoenvironmental implications of the distribution of intertidal foraminifera in a tropical Australian estuary: a reconnaissance study. Australian Geographical Studies, 39(1): 67-74. Haslett, S.K. (Ed.), 2002, Quaternary Environmental Micropaleontology, Arnold: London. (340pp). Haslett, S.K., 2007. The distribution of foraminifera in surface sediments of the Clyde River Estuary and Bateman’s Bay (New South Wales, Australia). Revista Española de Micropaleontologia, 39(1-2): 63-70. Hayward, B.W., 1990. Foraminifera in nearshore sediments, Whale Island, Bay of Plenty. Tane, 32: 93-99. Hayward, B.W., 1993. Estuarine foraminifera, Helena Bay, Northland, New Zealand. Tane, 34: 79-88. Hayward, B.W. and Grenfell, H.R., 1994. Foraminiferal associations around northern Great Barrier Island, New Zealand. Records of Auckland Institute and Museum, 31: 231-237. Hayward, B.W., Grenfell, H., Cairns, G. and Smith, A., 1996. Environmental Controls on Benthic Foraminiferal and Thecamoebian Associations in a New Zealand Tidal Inlet. Journal of Foraminiferal Research, 26(2): 150-171. Hayward, B.W., Grenfell, H.R. and Reid, C., 1997a. Foraminiferal associations in Wanganui Bight and Queen Charlotte Sound, New Zealand. New Zealand Journal of Marine and Freshwater Research, 31: 337-365. Hayward, B.W., Grenfell, H.R., Reid, C.M. and Hayward, K.A., 1999. Recent New Zealand Shallow-water Benthic Foraminifera: Taxonomy, Ecologic Distribution, Biogeography, and Use in Palaeoenvironmental Assessment. Institute of Geological and Nuclear Sciences Monograph, 21, Institute of Geological and Nuclear Sciences Limited: Lower Hutt, New Zealand (264pp). Hayward, B.W. and Hollis, C.J., 1994. Brackish foraminifera in New Zealand: a taxonomic and ecologic review. Micropaleontology, 40(3): 185-222. Hayward, B.W., Hollis, C.J. and Grenfell, H.R., 1997b. Recent Elphidiidae (Foraminiferida) of the South-west Pacific and fossil Elphidiidae of New Zealand. Institute of Geological and Nuclear Sciences Monograph, 16, Institute of Geological and Nuclear Sciences Limited: Lower Hutt, New Zealand (166pp). Holton, J.R., 2004. An Introduction to Dynamic Meteorology, Fourth Edition. Elsevier Academic Press: London (535pp).

Hornung, H., Krom, M.D. and Cohen, Y., 1989. Trace metal distribution in sediments and benthic fauna of Haifa Bay. Estuarine, Coastal and Shelf Sciences, 29: 43-56. Horton, B.P. and Murray, J.W., 2007. The roles of elevation and salinity as primary controls on living foraminiferal distributions: Cowpen Marsh, Tees Estuary, UK. Marine Micropaleontology, 63: 169–186. Hudson, J.P., 1991. Late Quaternary Evolution of Twofold Bay, Southern New South Wales. A thesis submitted in fulfilment of the requirements for the degree of Master of Science. Coastal Studies Unit, Department of Geography, University of Sydney (unpublished). Hulme, S.G., 1964. Recent foraminifera from Manukau Harbour, Auckland, New Zealand. New Zealand Journal of Science, 7(3): 305-40. Johnson, K.R. and Albani, A.D., 1973. Biotopes of recent benthonic foraminifera in Pitt Water, Broken Bay, NSW (Australia). Palaeogeography, Palaeoclimatology, Palaeoecology, 14: 265-276. Larkum, A. W. D., Orth, R.J., Duarte, C.M., 2006. Seagrasses: biology, ecology, and conservation. Springer: Dordrecht, The Netherlands (691pp). Lawver, L.A. and Gahagan, L.M., 2003. Evolution of Cenozoic seaways in the circum- Antarctic region. Palaeogeography, Palaeoclimatology, Palaeoecology, 198(1- 2):11-37. Le Cadre, V., Debenay, J.P. and Lesourd, M., 2003. Low pH effects on Ammonia beccarii test deformation: implications for using test deformations as a pollution indicator. Journal of Foraminiferal Research, 33(1): 1-9. Lea, D.W., 1999. Trace elements in foraminiferal calcite, pp.259-277, In Sen Gupta, B.K. (Ed.) Modern Foraminifera, Kluwer Academic Press: Dordrecht, The Netherlands. Leoblich, A.R. Jr. and Tappan, H., 1964. Protista 2. Sarcodina, chiefly thecamoebians and Foraminiferida. 1:1-510, 2:511-590, In Moore, R.C. (Ed.) Treatise on Invertebrate Paleontology, Part C. Vols 1-2. Kansas University Press: Kansas. Leorri, E. and Cearreta, A., 2009. Quantitative assessment of the salinity gradient within the estuarine systems in the southern Bay of Biscay using benthic foraminifera. Continental Shelf Research, 29: 1226-1239. LeRoy, D.O., 1977. Economic Microbiostratigraphy, pp.212-233, In LeRoy, L.W., LeRoy, D.O., Raese, J.W. (Eds.) Subsurface Geology - Petroleum - Mining - Construction. Colorado School of Mines: Golden, Colorado.

Linacre, E. and Hobbs, J., 1977. The Australian Climatic Environment. Wiley and Sons: Brisbane. (354pp). Loubere, P. and Fariduddin, M., 1999. Benthic foraminifera and the flux of organic carbon to the seabed, pp.181-199, In Sen Gupta, B.K. (Ed.) Modern Foraminifera, Kluwer Academic Publishers: Dordrecht, The Netherlands. Lowe J.J. and Walker M.J.C., 1997. Reconstructing Quaternary Environments. Addison Wesley Longman: Harlow, England (446pp). Martin, R.E. and Steinker, D.C., 1973. Evaluation of techniques for recognition of living foraminifera. Compass of Sigma Gamma and Epsilon. 50(4): 26-30. Mason, P. and Laws, R.A., 2011. Benthic foraminiferal paleoecology of upper cretaceous strata, Dixon core, North Carolina, Geological Society of America Abstracts with Programs, 43(2): 72. McKenzie, K.G., 1962. A record of Foraminifera from Oyster Harbour, near Albany, Western Australia. Journal of the Royal Society of Western Australia, 45: 117- 132. Michael, H.A., Mulligan, A.E. and Harvey, C.F., 2005. Seasonal oscillations in water exchange between aquifers and the coastal ocean. Nature, 436(25): 1145-1148. Michie, M.G., 1987. Distribution of foraminifera in a macrotidal tropical estuary: Port Darwin, Northern Territory of Australia. Australian Journal of Marine and Freshwater Research, 38: 249-259. Mojtahid, M., Jorissen, F. and Pearson, T.H., 2008. Comparison of benthic foraminiferal and macrofaunal responses to organic pollution in the Firth of Clyde (Scotland). Marine Pollution Bulletin, 56: 42-76. Muerdter, D.R. and Kennett, J.P., 1984. Late Quaternary planktonic foraminiferal biostratrigraphy, Strait of Sicily, Mediterranean Sea, Marine Micropaleontology, 8(5): 339-359. Murray, J.W., 1963. Ecological experiments on Foraminiferida. Journal of the Marine Biological Association of the UK, 43: 621-42. Murray, J.W., 1991. Ecology and Palaeoecology of Benthic Foraminifera. Longman Scientific and Technical: Harlow, Essex (397pp). Murray, J.W., 1991. Palaeoecology, pp.303-312, In Murray, J.W. (Ed.) Ecology and Palaeoecology of Benthic Foraminifera, Longman Scientific and Technical: Harlow, Essex.

Murray, J.W., 2006. Ecology and Applications of Benthic Foraminifera. Cambridge University Press: Cambridge (426pp). Naish, T. and Kamp, P.J.J., 1997. Foraminiferal depth palaeoecology of late Pliocene shelf sequences and systems tracts, Wanganui Basin, New Zealand. Sedimentary Geology, 110: 237-255. Nash, G.J., Binnie, M.N. and Cann, J.H., 2010. Distribution of foraminifera and ostracods in the Onkaparinga Estuary, South Australia. Australian Journal of Earth Sciences, 57(7): 901-910. New South Wales Department of Primary Industries, 2005. Mussel Aquaculture in Twofold Bay: Stage 1 Review Report and Determination of Progression to Stage 2, prepared in accordance with the Mussel Aquaculture in Twofold Bay Determination dated 1998. Department of Primary Industries: Fishing and Aquaculture, viewed 6 September, 2011. New South Wales Department of Urban Affairs and Planning, 2000. Proposed Multi- purpose Wharf and Naval Ammunitioning Facility at Twofold Bay, NSW: Assessment Report, viewed 6 September, 2011. Nobes, K., and Uthicke, S., 2008. Benthic Foraminifera of the Great Barrier Reef: A guide to species potentially useful as Water Quality Indicators. Report to the Marine and Tropical Sciences Research Facility. Reef and Rainforest Research Centre Limited: Cairns (44pp). Orpin, A.R., Haig, D.W. and Woolfe, K.J., 1999. Sedimentary and foraminiferal facies in Exmouth Gulf, in arid tropical northwestern Australia. Australian Journal of Earth Sciences, 46: 607-621. Palmeiri, V., 1976. Recent and subrecent foraminifera from the Wynnum 1:25 000 sheet area, Moreton Bay, Queensland. Queensland Government Mining Journal, 77: 365-384. Palmer, M.R., 1985. Rare earth elements in foraminifera tests. Earth and Planetary Science Letters, 73(2-4): 285-298. Parker, W.C. and Arnold, A.J., 1999. Quantitative methods of data analysis in foraminiferal ecology, pp.71-89, In Sen Gupta, B.K. (Ed.) Modern Foraminifera, Kluwer Academic Publishers: Dordrecht, The Netherlands.

Phleger, F.B., 1960. Ecology and Distribution of Recent Foraminifera. John Hopkins Press: Baltimore (297pp). Pollard, D., Ortiz, E. and Pethebridge, R. 1997. New South Wales marine and coastal bioregionalisation study: Towards the development of a representative system of marine and estuarine protected areas for NSW. Report to Environment Australia, cited in Breen, D.A., Avery, R.P., and Otway N.M., 2005. Broadscale biodiversity assessment of marine protected areas in the Batemans Shelf and Twofold Shelf marine bioregions. Final report for the NSW Marine Parks Authority. Pollard, D.A. and Rankin, B.K., 2003. Port of Eden Introduced Marine Pest Species Study: Report to Coasts & Clean Seas Program. NSW Fisheries Final Report Series No. 46 (May). NSW Fisheries Office of Conservation: Cronulla (67pp) viewed 2 September, 2011. Posey, M.H., 1988. Community changes associated with the spread of an introduced seagrass, Zostera japonica. Ecology, 69: 974-83. Potter, P.E., Maynard, J.B. and Depetris, P.J., 2005. Mud and mudstones: introduction and overview, Springer: Berlin (297pp). Price, G.D., Sellwood, B.W., Corfield, R.M., Clarke, L. and Cartlidge, J.E., 1998. Isotopic evidence for palaeotemperatures and depth stratification of Middle Cretaceous planktonic foraminifera from the Pacific Ocean. Geological Magazine 135(2): 183–191. Quilty, P.G. and Hosie, G., 2006. Modern Foraminifera, Swan River Estuary, Western Australia: distribution and controlling factors. Journal of Foraminiferal Research, 36(4): 291-314. Romano, E., Bergamin, L., Finoia, M.G., Carboni, M.G., Ausili, A. and Gabellini, M., 2008. Industrial pollution at Bagnoli (Naples, Italy): Benthic foraminifera as a tool in integrated programs of environmental characterisation. Marine Pollution Bulletin, 56: 439-457. Ruiz, F., Gonzáles-Regalado, M.L., Pendón, J.G., Abad, M., Olías, M. and Muñoz, J.M., 2005. Correlation between foraminifera and sedimentary environments in recent estuaries of Southwestern Spain: Applications to Holocene reconstructions. Quaternary International, 140-141: 21-36.

Schafer, C.T., 1973. Distribution of foraminifera near pollution sources in Chaleur Bay, Water, Air, and Soil Pollution, 2(2): 219-233. Schafer, C.T., 2000. Monitoring nearshore marine environments using benthic foraminifera: some protocols and pitfalls. Micropaleontology, 46: 161-169. Schafer, C.T. and Cole, F.E., 1974. Distribution of benthonic foraminifera: their use in delimiting local nearshore environments, offshore geology of Canada, Eastern Canada. Geological Survey of Canada, Paper 74-30 (1): 103-108. Schafer, C.T. and Cole, F.E., 1974. Distribution of benthonic foraminifera: their use in delimiting local nearshore environments, offshore geology of Canada, Eastern Canada. Geological Survey of Canada, Paper 74-30 (1): 103-108. Schafer, C.T. and Wagner, F.J.E., 1978. Foraminifera-mollusc association in eastern Chaleur Bay. Canadian Journal of Earth Sciences, 15: 889-901. Schmiedl, G. and Mackensen, A., 1997. Late Quaternary paleoproductivity and deep water circulation in the eastern South Atlantic Ocean: Evidence from benthic foraminifera. Palaeogeography, Palaeoclimatology, Palaeoecology, 130(1-4): 43- 80. Scott, D.B., 1974. Recent benthonic foraminifera from Samish and Padilla Bays, Washington. Northwest Science, 48(4): 211-218. Scott, D.B., 1976. Brackish water foraminifera from Southern California and description of Polysaccammina ipohalina n. gen., n. sp.. Journal of Foraminiferal Research, 6(4): 312-321. Scott, D.B. and Medioli, F.S., 1980. Quantitative studies of marsh foraminiferal distributions in Nova Scotia: implications for sea level studies. Cushman Foundation for Foraminiferal Research Special Publication, 17: 1-47. Scott, D.B., Medioli, F.S. and Schafer, C.T., 2001. Monitoring in Coastal Environments Using Foraminifera and Thecamoebian Indicators, Cambridge University Press: Cambridge (177pp). Scott, D.B., Mudie, P.J. and Bradshaw, J.S., 1976. Benthonic foraminifera of three Southern California lagoons: ecology and recent stratigraphy. Journal of Foraminiferal Research, 6(1): 59-75. Scott, D.B., Schafer, C.T. and Medioli, F.S., 1980. Eastern Canadian estuarine foraminifera: a framework for comparison. Journal of Foraminiferal Research, 10(3): 205-234.

Sen Gupta, B.K., 1999. Foraminifera in marginal marine environments, pp.141-159, In Sen Gupta, B.K. (Ed.) Modern Foraminifera, Kluwer Academic Publishers: Dordrecht, The Netherlands. Sen Gupta, B.K. and Machain-Castillo, M.L., 1993. Benthic foraminifera in oxygen-poor habitats. Marine Micropaleontology, 20: 183-201. Setty, M.G. Anatha P. and Nigam, R., 1984. Benthic foraminifera as pollution indices in the marine environment of west coast of India. Rivista Italiana di Paleontologia E Stratigrafia, 89(3): 421-436. Shackleton, N.J. and Opdyke, N.D., 1973. Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28-238: oxygen isotope temperatures and ice volumes on a 105 and 106 year scale. Quaternary Research, 3: 39-55. Sharifi, A.R., Croudace, I.W. and Austin, R.L., 1991. Benthic foraminiferids as pollution indicators in Southampton Water, southern England, U.K.. Journal of Micropaleontology, 10(1): 109-113. Smith, D., Scott, D.B. and Medioli, F.S., 1984. Marsh foraminifera in the Bay of Fundy: modern distribution and application to sea-level determinations. Maritime Sediments and Atlantic Geology, 20: 127-142. Sneath, P.H.A. and Sokal, R.A., 1973. Numerical Taxonomy: The principles and practice of numerical classification. W.H. Freeman and Co; San Francisco (573pp). Stroud, A., 2011. After the flood. 24 March 2011, Eden Magnet Newspaper: Eden, NSW, viewed 2 September, 2011. The Australian Hydrographic Service, 2011. Seafarer GeoTIFF AUS Chart 192: Twofold Bay, Commonwealth of Australia, Licence No: 2976SG. Thom, B.G. and Hall, W., 1991.Behaviour of beach profiles during accretion and erosion dominated periods. Earth Surface Processes and Landforms, 16(2): 113–127. Travis, J.L. and Bowser, S.S., 1991. The motility of Foraminifera, pp.91-155, In Lee, J.J. and Anderson, O.R. (Eds.) Biology of Foraminifera, Academic Press: London. Treloar, P.D., 2011. Eden Harbour Wave Modelling: Prepared for Land and Property Management Authority Crown Lands NSW and Bega Valley Shire Council, LJ2916/R2685. Cardno (NSW/ACT) Pty Ltd: Gordon, NSW (97pp), viewed 18 August, 2011.

Van der Zwaan, G.J., Duijnstee, A.P., den Dulk, M., Ernst, S.R., Jannink, N.T. and Kouwenhoven, T.J., 1999. Benthic foraminifers: proxies or problems?: A review of paleocological concepts. Earth-Science Reviews, 46(1-4): 213-236. Waelbroeck, C., Labeyrie, L., Michel, E., Duplessy, J.C., McManus, J.F., Lambeck, K., Balbon, E. and Labracherie, M., 2002. Sea-level and deep water temperature changes derived from benthic foraminifera isotopic records. Quaternary Science Reviews, 21(1-3): 295-305. Walker, D.A., Linton, A.E. and Schafer, C.T., 1974. Sudan Black B: A superior stain to Rose Bengal for distinguishing living from nonliving foraminifera. Journal of Foraminiferal Research, 4: 205-215. Wang, P. and Chappell, J., 2001. Foraminifera as Holocene environmental indicators in the South Alligator River, Northern Australia. Quaternary International, 83- 85:47-62. Watkins, J.G., 1961. Foraminiferal ecology around the Orange County, California, ocean sewer outfall. Micropaleontology, 7(2): 199-206. Williams, C.B., 1964. Patterns in the Balance of Nature. Academic Press: London, (324pp). Woodroffe, S.A., Horton, B.P., Larcombe, P. and Whittaker, J.E., 2005. Intertidal mangrove foraminifera from the central Great Barrier Reef shelf, Australia: Implications for sea-level reconstruction. Journal of Foraminiferal Research, 35: 259-270. Yanko, V., Ahmad, M. and Kaminski, M., 1998. Morphological deformities of benthic foraminiferal tests in response to pollution by heavy metals: implications for pollution monitoring. Journal of Foraminiferal Research, 28(3): 177-200. Yanko, V., Kronfeld, J. and Flexer, A., 1994. Response of benthic foraminifera to various pollution sources: implications for pollution monitoring. Journal of Foraminiferal Research, 24(1): 1-17. Yassini, I. and Jones, B.G., 1989. Estuarine foraminiferal communities in Lake Illawarra, N.S.W. Proceedings of the Linnean Society of New South Wales, 110(3): 229-266. Yassini, I. and Jones, B.G., 1995. Foraminiferida and Ostracoda from Estuarine and Shelf Environments on the Southeastern Coast of Australia. University of Wollongong Press: Wollongong (484 pp).

Zachos, J.C., Pagani, M., Sloan, L., Thomas, E. and Billups, K., 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science, 292(5517): 686–693. Zieman, J.C. and Zieman, R.T., 1989. The ecology of seagrass meadows of the west coast of Florida: a community profile. U.S. Fish and Wildlife Service Biological Report, 85(7.25) (155pp).

APPENDICES

NOTE: Appendix prefix refers to chapter number, with suffix relating to order within that chapter.

e.g. Appendix 3.1 = Chapter 3, first appendix

Appendix 3.1

GPS co-ordinates of all sample locations referenced in this study. From top: T = Towamba River, C = Curalo Lagoon, N = Nullica River, F = Fisheries Creek, TFB = Twofold Bay.

Appendix 4.1

Raw data of the abiotic parameters gathered for each Twofold Bay (TFB) station, as discussed in the text.

Appendix 4.2

Drivers of abiotic clusters were determined by plotting the mean and standard error at 95% confidence for each measured parameter and determining where the difference between plotted values was greatest (depth [m], DO [mg/L], DO [%]).

Appendix 4.3

Raw data from the laser diffraction sedimentary analysis of all Twofold Bay (TFB) samples. Samples labelled ‘take 2’ indicate repeat analyses of their prefix station.

Appendix 4.4

Soil size fractions as referenced in the text, according to the standard British sieve size scale.

Appendix 4.5

Total foraminiferal species count data compiled from analysis of all Twofold Bay (TFB) samples.

Appendix 4.6

Foraminiferal identification and deformity data sheets from each Twofold Bay (TFB) station – see CD-ROM.

Appendix 4.7

Data sheets detailing breakdown of benthonic foraminiferal wall structure from each Twofold Bay (TFB) station – see CD-ROM.

Appendix 4.8

List of foraminiferal fauna from each biotope in order of highest abundance. Used to determine predominant assemblages – see CD-ROM.