Biology and ecology of deployed shellfish habitats in the Swan-Canning Estuary
Submitted by Charles J. Maus, BSc.
This thesis is presented for the degree of Bachelor of Science Honours, Murdoch University, 2020
Declaration
I declare that this thesis is my own account of my research and contains as its main content work which has not previously been submitted for a degree at any tertiary education institution.
Charles J. Maus
ii Abstract
Large extents of shellfish reefs have become degraded around the world as a result of anthropogenic activities to the point where such reefs are functionally extinct in some regions. Due to the ecosystem services provided by these biogenic habitats, there has recently been a concerted effort to restore shellfish reefs, particularly in Australia. While oysters have traditionally been used as a candidate species, the
Mediterranean Mussel Mytilus galloprovincialis is gaining popularity and provides a similar suite of ecosystem services to oysters. As part of a pilot program, shellfish habitats, each comprising translocated M. galloprovincialis seeded onto 100 wooden stakes, were deployed at three sites in Melville Water in the Swan-Canning Estuary (Western Australia). The aims of this study where to; 1) investigate the mortality, body condition and growth of the translocated M. galloprovincialis; 2) compare the characteristics of fish fauna at the shellfish reef habitats and nearby unstructured (control) habitats and 3) determine the benthic macroinvertebrate and tunicate species associated with the shellfish habitat.
The mortality of M. galloprovincialis was high at all three sites. This was attributed to poor environmental conditions in offshore waters of Melville Water, compounded by stress associated with translocation, their spawning activity, and fouling by ascidians. Seasonally adjusted von Bertalanffy growth models best explained the growth of M. galloprovincialis and growth was rapid, with individuals attaining ~50 mm within their first year. This is likely due to the high phytoplankton availability in the Swan-Canning Estuary.
The shellfish habitats harboured a significantly different fish faunal composition compared to nearby unstructured habitats (sandy areas), with many species observed only at shellfish sites or in greater densities. The increased abundances of zoobenthivores on the shellfish habitats suggest they are utilising the invertebrate prey communities associated with the structure as a food source. The invertebrate
iii community varied spatially among the three sites and over time. A suite of non-native ascidians rapidly colonised the stakes along with the mussels, which, in turn, supported many small crustaceans.
Given the importance of shellfish restoration globally, and the aim to undertake large-scale projects to provide such habitats in south-western Australian estuaries, the results of this study will increase the understanding of the biology of M. galloprovincialis and help elucidate how faunal communities respond and utilise shellfish habitats. The results of this pilot study will assist in the planning of future mussel reef restoration projects, in particularly those under development in southern Australia.
iv Table of contents Abstract ...... iii Acknowledgements ...... vi Chapter 1: General introduction ...... 1 1.1 Shellfish reefs ...... 1 1.2 Biology of Mytilus galloprovincialis...... 4 1.3 Estuaries ...... 6 1.4 Rationale and aims ...... 8 Chapter 2: General material and methods ...... 10 2.1 Study site ...... 10 2.2 Site deployment ...... 12 2.3 Sampling regime ...... 14 2.4 Fish communities ...... 15 2.5 Invertebrate communities ...... 18 Chapter 3: Biology of translocated Mytilus galloprovincialis deployed on shellfish habitats in the Swan-Canning Estuary...... 20 3.1 Abstract ...... 20 3.2. Introduction ...... 21 3.3 Materials and methods ...... 22 3.3.1 Environmental data and laboratory methods ...... 22 3.3.2 Statistical analysis ...... 23 3.4 Results ...... 25 3.4.1 Environmental parameters ...... 25 3.4.2 Mortality ...... 27 3.4.3 Body condition ...... 29 3.4.4 Growth ...... 30 3.5 Discussion ...... 34 3.5.1 Mortality ...... 34 3.5.2 Body condition ...... 36 3.5.3 Growth ...... 36 Chapter 4: Faunal community associated with deployed shellfish habitats in the Swan-Canning Estuary ...... 38 4.1. Abstract ...... 38 4.2 Introduction ...... 39 4.3. Materials and methods ...... 41 4.3.1 Statistical analysis ...... 42 4.3.2 Invertebrate communities ...... 46 4.4. Results ...... 47 4.4.1. Environmental conditions ...... 47 4.4.2 Fish communities ...... 48 4.4.3. Invertebrate communities ...... 61 4.5 Discussion ...... 79 4.5.1 Fish communities ...... 79 4.5.2 Large invertebrate fauna ...... 83 4.5.3. Small invertebrate fauna ...... 84 4.6 Appendices ...... 86 Chapter 5: General conclusion ...... 91 5.1 Critical review ...... 91 5.2 Future directions ...... 95 References ...... 97
v Acknowledgements
Firstly, thank you to my supervisors, Dr James Tweedley and Dr Alan Cottingham for their ongoing support, feedback, and mentorship throughout the duration of this project. I would also like to thank you both for providing me the opportunity to work on a wide range of projects in many different locations, from other shellfish projects to fish and invertebrate sampling, over the last two years. It is amazing how much extra one can learn by simply being in the field!
To Sorcha Cronin-O'Reilly, thank you for introducing me to the basics of invertebrate identification and support throughout the writing stages of this thesis. Thank you to Dr Thea Linke for critiquing my references and Jesse Newton for reading over my work and providing feedback. To the rest of the fish group for creating a great work environment, in particular Kurt Krispyn, it is always better to have someone work late with, whether it be in the lab or on the Swan, I look forward to hauling nets again soon!
Thank you to all the volunteers who assisted in the creation of the shellfish habitats, over three consecutive weekends; Murdoch University Dive Club, Oz Fish Perth chapter, Marine Men’s Shed and friends. A special thankyou to Rob Hatton for his assistance during sampling, particularly during the first six months. Gratitude is expressed to the funding body, Recfishwest, for proving the opportunity to undertake this project through the Recreational Fishing Initiative Fund.
To my partner Branda, thank you for your ongoing support throughout this process and making sure I eat proper food every so often! You have been truly amazing; I cannot thank you enough. Lastly to friends and family who have supported me throughout this journey, thank you. Finally, to my family for their support throughout the duration of my studies.
vi Chapter 1: General introduction
1.1 Shellfish reefs
Shellfish reefs are complex, multi-dimensional structures comprising dense aggregations of living bivalves (i.e. oysters and mussels), dead shells and accumulated sediments (Coen et al., 2007). These reefs provide a suite of ecosystem goods and services, such as water filtration and, due to their elevation from the substrate, can also provide defence against coastal erosion in a similar way to breakwaters, groynes or seawalls as a living shoreline (Buschbaum et al., 2009; Scyphers et al., 2011; Reguero et al., 2018). Moreover, as filter feeders, bivalves consume phytoplankton and bind particulate matter from the overlying water column to the benthos, with mucus forming pseudofaeces. These actions improve water quality by decreasing turbidity and increasing light penetration, which can have secondary benefits on submerged aquatic vegetation, such as seagrasses (Grabowski et al., 2012; Sharma et al., 2016). Furthermore, the filter-feeding activities of bivalves facilitate the removal of nitrogen, which is advantageous given the general increase in eutrophication in estuaries and coastal waters around the world (Newell et al., 2005; Justic et al., 2009).
The structural complexity of shellfish reefs provides refuge and feeding opportunities for a wide range of fauna (Christianen et al., 2017). These biogenic structures often harbour a more diverse and abundant range of fish and decapod crustaceans than adjacent unstructured habitats, i.e. sandy and muddy substrates (Meyer & Townsend, 2000; Peterson et al., 2003; Tolley & Volety, 2005; Humphries & Peyre, 2015). This is likely attributed to the range of smaller invertebrate species (e.g. other molluscs, polychaetes and crustaceans) that utilise the reefs and thus provide a food source for higher trophic levels (Rodney & Paynter, 2006). The provision of food and habitat provided by shellfish reefs has also been linked to increases in the production of both recreational and commercial fisheries. (Tolley & Volety, 2005; Rodney & Paynter, 2006; Borthagaray & Carranza, 2007). Through the delivery of the
1 ecosystem services described above, it is not surprising that shellfish reefs provide significant economic benefits, with Grabowski et al. (2012) conservatively valuing restored oyster reefs at between US $5,500 and 99,000 ha y-1.
Shellfish reefs are in severe decline with an extensive review by Beck et al. (2011) concluding that > 85% of oyster reefs have been lost globally, to the point where they are now considered to be functionally extinct in some regions, including Australia
(Fig. 1.1). Similar losses of shellfish reefs have been documented for other reef-forming bivalves. For example, destructive fishing practices and a changing climate have resulted in a ~95% decline in the abundance of the Blue Mussel Mytilus edulis in the Wadden Sea (Herlyn & Millat, 2000) and a ~50% decline in the California Mussel Mytilus californianus in southern California (Smith et al., 2006).
Shellfish populations in the Southern Hemisphere have succumbed to the same fate as their Northern Hemisphere counterparts. In the Hauraki Gulf in New Zealand, only 0.64 km2 of the 1,200 km2 of the Green-lipped Mussel Perna canaliculus reefs remain, a loss of > 99% (Wilcox et al., 2018). Similarly, < 10% of Australian shellfish reefs remain intact (Beck et al., 2011; Ford & Hamer, 2016; Gillies et al., 2018, 2020).
Fig. 1.1. The global condition of oyster reefs in bays and ecoregions. The condition ratings of good, fair, poor and functionally extinct are based on the percentage of the current to historical abundance of oyster reefs remaining: < 50% lost (good), 50 – 89% lost (fair), 90 – 99% lost (poor) and > 99% lost (functionally extinct). Figure taken from Beck et al. (2011).
2 A range of factors have caused these declines. Harvesting, and particularly overexploitation, for example, simultaneously removes bivalves and the structural habitat they would provide. As juvenile shellfish settle onto the shells of their conspecifics, the removal of this habitat consequently reduces the amount of suitable substrate available and limits recruitment. Furthermore, the remaining fragmented reef structure becomes increasingly vulnerable to physical damage from wave energy, sedimentation and disease (Beck et al., 2011; Gillies et al., 2018, 2020; Veiga et al., 2020).
Increasing catchment urbanisation, acting separately or synergistically with overexploitation, decreases water quality, which is exacerbated by declines in shellfish abundance, which, if present, would remove nitrogen and phytoplankton. Thus, the decline in water quality reduces the abundance of shellfish and ultimately decreases the resilience of the reefs. This chain of events has occurred in Chesapeake Bay (USA), which has suffered extensive losses of the Eastern Oyster Crassostrea virginica (Rothschild et al., 1991; Beck et al., 2011).
The importance of habitat restoration is being recognised on a global scale. The United Nations has declared 2021-2030 the “decade of restoration” and aims to restore 3.5 million km2 of degraded ecosystems, which is an area almost the size of India
(Waltham et al., 2020). The end goal is to reverse biodiversity loss, restore functionality and resilience to climate and anthropogenic change (Coleman et al., 2020; Waltham et al., 2020). Over recent decades there has been a concerted effort to restore shellfish reefs and the ecosystem goods and services they provide. Successful restoration projects have occurred in Chesapeake Bay, Pamlico Sound and the Gulf of Mexico (all USA), Strangford Lough (Northern Ireland), Limfjord (Denmark), the Yangtze Estuary (China) and Port Phillip Bay in Australia (Schulte et al., 2010; Beck et al., 2011; La Peyre et al., 2014; Ly et al., 2016; Gillies et al., 2017).
3 In areas where shellfish reefs have been lost, through destructive fishing and dredging practices, their substrate has also been removed, inhibiting natural recovery. Therefore, shell clutch or limestone rubble is used to increase the area available for bivalve settlement and also reduces the effects of smothering by suspended or mobile bottom sediments (La Peyre et al., 2014; Gillies et al., 2017; Branigan et al., 2020). The deployed substrate is then seeded with spat (oysters or mussels), commonly produced in local aquaculture facilities (Brumbaugh & Coen, 2009; Geraldi et al., 2013).
Typically, oysters have been used as a candidate species for restoration. However, other reef-forming bivalves, such as mussels, are rapidly becoming recognised as providing the same suite of ecosystem goods and services as oysters and their use as a candidate species for restoration is increasingly becoming recognised (Gillies et al., 2017; Wilcox et al., 2018; Schotanus et al., 2020). The mytilid, Perna canalicus, has been used to restore lost shellfish reefs in New Zealand (Wilcox et al., 2018) and the Blue Mussel, Mytilus edulis, has been trialled in restoration projects the Oosterschelde Estuary in the Netherlands (Schotanus et al., 2020). Additionally, Mytilus galloprovincialis was used in a multispecies approach to restoring lost shellfish reefs in Port Phillip Bay (Gillies et al., 2017, 2018). This last species is also currently being trialled as a candidate species for large-scale restoration projects in the Peel-Harvey and Swan- Canning estuaries in south-western Australia (A. Cottingham, unpublished data).
1.2 Biology of Mytilus galloprovincialis
Mytilus galloprovincialis Lamarck, 1819, is one of three closely related taxa of smooth shelled marine mussel in the M. edulis complex, which also include M. edulis and M. trossolus. These bivalves are widely distributed in temperate coastal and estuarine waters in both the Northern and Southern hemispheres (McDonald et al., 1991).
4 Genetic studies have identified three distinct groups within the M. galloprovincialis species, namely the north-eastern Atlantic group, Mediterranean group, (which includes populations from the eastern and western North Pacific coastline) and an Australasian group (Daguin & Borsa, 2000). Although these groups are thought to be the result of ancient hybridisation (McDonald et al., 1991; Borsa et al., 2007), human-mediated transport has seen a spatial shift in distribution to new countries, such as South Africa (Grant & Cherry, 1985; Branch & Steffani, 2004). Within Australia, M. galloprovincialis is found in coastal and estuarine waters from New South Wales along the southern coast (including Tasmania) and northwards to Fremantle in Western Australia (Hope et al., 1977; Gardner, 2004; Gardner & Westfall, 2012; Dias et al., 2014; Ford & Hamer, 2016; Gillies et al., 2018). Australian populations of M. galloprovincialis typically contain both native southern and introduced northern hemisphere haplotypes (Dias et al., 2014).
Mytilus galloprovincialis are prolific in intertidal and subtidal waters where they form dense aggregations (reefs) by attaching to had substrates and their conspecifics using byssal threads. By modifying the complexity and functioning of the substrate they occupy, mussels are considered ecosystem engineers (Borthagaray & Carranza, 2007; Buschbaum et al., 2009). Mussel reefs provide a similar suite of ecosystem goods and services to those of oyster reefs, including shoreline stabilisation, the provision of complex habitat and prey communities for higher-level consumers, e.g. fish (Buschbaum et al., 2009). Their effectiveness in filter-feeding is illustrated in the Oosterschelde Estuary, where it is estimated that the entire volume of water is filtered in < 4 days by M. edulis and the cockle Cerastoderma edule (Smaal et al., 1986; Dame et al., 1991; Dame & Prins, 1997). Given this filtering capacity, it is unsurprising that this system was the only estuary in a review of 131 from around the world that had a net negative annual phytoplankton production (Cloern et al., 2014).
5 Growth of M. galloprovincialis is largely determined by environmental factors, such as wave energy, food availability and temperature (Anestis et al., 2007; Bergström & Lindegarth, 2016). For example, individuals in aggregations on exposed shorelines developed thicker, more robust shells and thus do not reach the same length as those in calmer waters (Seed, 1969). Seasonal variations in growth and body condition (i.e. weight at a standardised size) are well documented, being greatest during summer when food availability is high, but declining during the winter months, as gametogenesis depletes energy reserves and phytoplankton abundance is lower (Peharda et al., 2007; Celik et al., 2009; Azpeitia et al., 2018). Mytilus galloprovincialis exhibits signs of stress when water temperatures exceed 24°C for extended periods of time. Increased mortality is often a result and has been attributed to oxidative stress and reduced ability to assimilate food (Anestis et al., 2007).
Individuals of this species attain sexual maturity around one year old. Mytilid mussels have an extended spawning period, commencing during winter and continuing into the summer months. However, the spawning event coinciding with spring (i.e. austral September) is considered the major spawning event, contributing to most of the recruitment (Ceccherelli & Rossi, 1984; Cáceres-Martínez & Figueras, 1998). Fecundity exhibits high interannual variability and is attributed to phytoplankton availability (Thompson, 1979; Smart et al., 2020).
1.3 Estuaries
Estuaries are transitional systems located at the interface of marine and freshwater environments (McLusky & Elliott, 2007; Tweedley et al., 2016c). The competing influences of tidal water exchange and freshwater discharge results in dynamic physio-chemical conditions and the allochthonous input of nutrients from a range of sources (Elliott & Whitfield, 2011). These features, together with
6 autochthonous primary production, make estuaries among the most productive of all aquatic ecosystems (Schelske & Odum, 1961; Elliott & Whitfield, 2011). They also contain a wide variety of habitats, including seagrass meadows, salt marshes, mangrove forests and shellfish reefs, which provide food and shelter for a wide range of fish, invertebrate and avian fauna. For example, juvenile fish in estuaries achieve a higher rate of growth and thus also survival than those in adjacent marine environments
(Sogard, 1992; Vasconcelos et al., 2010; Veale et al., 2016). Moreover, the fish species that inhabit estuaries for all or part of their life contribute > 75% of the commercial fish catch in Australia and, in some regions, up to 90% of all recreational angling catch (Creighton et al., 2015).
In addition to providing habitat and resources for fauna, estuaries have also played a crucial role in human settlement (Wilson, 2002). They are attractive to humans as they provide a source for food and water, fertile catchments for agriculture, sheltered harbours and sites for settlement, industry, and trade (Potter et al., 2015b). Indeed, most human settlement has occurred within 100 km of the coastline, with coastal populations contributing three times the global average and seven of ten of the world’s largest megacities are located on or close to estuaries (Barragán & de Andrés, 2015; Selig et al., 2019). With an ever-increasing global population and intensification of agriculture and industrial practices in the Anthropocene, estuaries are subjective to increasing pressures, such as habitat loss, hydrological modification and human, agricultural and industrial waste (Kennish, 2002). Excess nutrients, particularly in systems with limited tidal water exchange and those with clear (low turbidity) and warm waters can lead to large microalgal and macroalgal algal blooms and hypoxia (Potter et al., 2016; Tweedley et al., 2016b). These deleterious events have been shown to have major effects on fish, invertebrate and shellfish populations, resulting in their movement away from affected areas and, in some cases, mass mortality events (Whitfield et al., 2006; Taylor & Whitfield, 2010; Hallett et al., 2018).
7 1.4 Rationale and aims
Temperate estuaries are regarded as the most degraded aquatic ecosystems (Jackson et al., 2001), with those in microtidal regions (tidal range < 2 m), such as south- western Australia, being particularly susceptible to the effects of anthropogenic perturbations and climate change (Warwick et al., 2018). Eutrophication combined with a Mediterranean climate and extended water residence time can result in large blooms of microalgae (Monbet, 1992; Adams et al., 2020). For example, the Swan-Canning Estuary (the study site in this thesis) was found to have among the greatest annual phytoplankton production in a global review of 131 estuaries (Cloern et al., 2014). Although there are limited naturally occurring benthic shellfish reefs in this estuary (Cottingham, unpublished data), shellfish reefs have been suggested by Warwick et al. (2018) as a way of alleviating some of the deleterious stressors on southern Australian microtidal estuaries. This was based on the situation in microtidal fjordic estuaries in Denmark, where high densities of filter-feeding macrobenthos, such as Mytilus spp. and the ascidians Ciona spp., greatly reduce phytoplankton concentrations and ameliorated the effects of eutrophication and associated blooms of phytoplankton (Conley et al., 2000).
In light of the above, the overall aim of this study was to trial the seeding of translocated M. galloprovincialis onto wooden stakes as a method of creating habitat and prey communities for fish in the Swan-Canning Estuary. Specifically, this involved.
i) Determining the mortality, body condition and growth of translocated M. galloprovincialis among several reef sites. ii) Comparing the characteristics of the fish fauna on the deployed shellfish habitats and nearby unstructured habitats. iii) Determining which benthic invertebrate and colonising tunicate species
were associated with these structures.
8 Mytilus galloprovincialis was selected as a candidate species for seeding onto the stakes as it is naturally abundant on jetty pylons and boat moorings in the Swan-Canning Estuary. Given the importance of shellfish habitats and the aim to undertake large-scale restoration projects in southern Australia, this study will increase the understanding of the biology of the candidate species as well as how fauna (fish and invertebrates) utilise this habitat in Melville Water of the Swan-Canning Estuary. The results of this pilot study are currently being used to inform a largescale mussel reef restoration project in this estuary which is funded by The Nature Conservancy. This thesis began with a General Introduction (Chapter 1), which provided context on shellfish reefs, their importance, degradation and restoration, mytild biology and estuaries, and is followed by a short General Materials and Methods (Chapter 2) describing the Swan-Canning Estuary, the selection and deployment of the stakes and sampling methodology. The three aims above are then addressed in two data chapters detailing the results of an investigation of the biology of M. galloprovincialis at each of the three sites (Chapter 3) and their associated fish and invertebrate communities (Chapter 4). The key findings of these chapters are summarised, and their implications discussed with recommendations for future research in a General Discussion (Chapter 5).
9 Chapter 2: General material and methods
2.1 Study site
The Swan-Canning Estuary (31⁰ 57’ S and 116⁰ 04’ E), located in south-western Australia, is ~ 50 km long, has a surface area of ~ 55 km2 and remains permanently open to the Indian Ocean (Valesini et al., 2014). The estuary comprises a narrow entrance channel that opens into two basins (Melville and Perth waters) and the tidal portions of the Swan and Canning rivers (Fig. 2.1). Although most of the estuary is shallow, i.e. < 2 m in depth, it reaches a maximum depth of ~ 20 m in the entrance channel, with parts of Melville Water reaching ~ 17 m deep (Hogan-West et al., 2019). Sediment composition differs markedly between the shallow marginal shoals (< 2 m deep) and the deeper waters (> 2 m deep). Nearshore substrates contain a greater proportion of medium sand than the offshore sediments, which comprise higher percentages of particulate organic matter and finer inorganic particles (i.e. silt and fine sand; Tweedley et al., 2017a). The seagrasses Zostera sp. and Heterozostera sp. are present in the entrance channel, while Halophila ovalis forms extensive beds in the shallow parts of Melville Water (Valesini et al., 2009; Department of Water, 2014).
South-western Australia experiences a Mediterranean climate (Hallett et al., 2018), with warm, dry summers and cool, wet winters with ~ 80 % of the annual rainfall occurring between June and September (Hodgkin & Hesp, 1998). This highly seasonal rainfall pattern, combined with the microtidal tidal regime of the Swan-Canning Estuary (tidal range < 2 m), results in marked seasonal variations in water physical-chemical conditions (Tweedley et al., 2016b). Salinities are typically stable and relatively high throughout much of the estuary during the austral summer (December to February), but during winter, may vary markedly along the estuary following substantial freshwater discharge (Crisp et al., 2018), leading to marked stratification of the water column and hypoxia (Tweedley et al., 2016b).
10 The Swan-Canning Estuary and its catchment have been heavily modified since European settlement and contains the City of Perth, the capital of Western Australia (~ 2 million people). Such modifications include substantial land clearing for urbanisation and agriculture, hardening of shorelines, blasting, and dredging of the mouth of the estuary to permanently connect it to the ocean and allow the construction of Fremantle Port. As such, the system was categorised as “extensively modified” in an audit of the condition of ~ 1,000 estuaries across Australia (Tweedley et al., 2017b). Agricultural and urban run-off results in the input of large quantities of nutrients (i.e. nitrogen and phosphorus) into the estuary. When combined with sunny conditions in summer and a long water residence times, due to the highly seasonal rainfall, this can result in extensive phytoplankton blooms. Some of the algal blooms such as those of the dinoflagellate Karlodinium veneficum, can be toxic (Place et al., 2012; Müller et al., 2019) and have resulted in the movement of fish away from affected areas and, in some cases, mass fish deaths (Hallett et al., 2018).
11
Fig. 2.1. Maps showing (a) the location of the Swan-Canning Estuary in south-western Australia and (b) the sites in the large central basin (Melville Water) where the three mussel reefs were deployed and their control sites. The location at which water quality measurements are collected weekly by the Department of Biodiversity, Conservation and Attractions is also provided as are those sites that were also considered as places where mussel reefs could be situated. Image retrieved and edited from Poh et al. (2018).
2.2 Site deployment
All activities in this study were undertaken in Melville Water, the large central basin of the Swan-Canning Estuary. This region is ~ 10 km long and up to 4 km wide and attains a maximum depth of 17 m. Salinities in Melville Water are subject to seasonal fluctuations, but to a lesser extent than those areas further upstream (Crisp et al., 2018; Poh et al., 2019). For example, during the warm, dry summer months, when rainfall and thus freshwater discharge is minimal, salinities at the surface and bottom of the water column are around that of full-strength seawater (~ 35 ppt). However, during winter, salinities in the surface waters decline to < 20 ppt, but values in the bottom waters typically exceed 31 ppt. Thus, Melville Water is subjected to some stratification, which is typically highest in the months following the greatest rainfall, although the intensity of stratification varies among years (Crisp et al., 2018; Poh et al., 2019).
12 Melville Water was selected as a suitable area to deploy the shellfish habitats as its deeper waters (> 2 m) are largely devoid of complex three-dimensional benthic structure and seagrass meadows and are typified by sandy and silty substrate. Initially, seven locations were proposed as potential shellfish habitat deployment sites (Fig. 2.1). However, as the Swan-Canning Estuary is utilised by a wide range of stakeholders, each proposed site had to be sufficiently deep so that the top of each stake (when fixed into the substrate) was > 3 m below the lowest astronomical tide, to avoid contact to vessels and/or recreational users in the area. In addition, sites that experience high levels of recreational activity (i.e. Point Walter and Matilda Bay) or those located near shorebird nesting habitats (i.e. Pelican point; Valesini et al., 2018) were deemed unsuitable. Three suitable sites (R1, R2 and R3; Fig. 2.1), each ~ 4 m deep, were approved by the Department of Biodiversity, Conservation and Attractions (DBCA) and Department of Transport (DoT).
Individuals of M. galloprovincialis were collected from mooring ropes in the lower reaches of the Swan-Canning Estuary and their external epibionts removed by a combination of aerial exposure and manual agitation. A total of 3 litres of mussels were then seeded onto the top 1 m of 300 jarrah (Eucalyptus marginata) stakes that were 1.5 m tall and 25 x 25 mm wide (Fig. 2.2). At each of the three sites in late March and early April 2019, 100 stakes were hammered ~ 50 cm deep into the substrate approximately
1 m apart, in a grid pattern by SCUBA divers (Fig. 2.3). Each reef covered an area of ~100 m2. The DBCA undertake daytime water quality monitoring at 40 sites throughout the Swan-Canning Estuary on a weekly basis. Measurements of salinity (ppt), water temperature (°C) and dissolved oxygen concentration (mgL-1) at 50 cm intervals through the entire water column were obtained from the closest site to the three mussel reefs, i.e. Heathcote (Fig. 2.1), between April 2019 and July 2020 using the online Water
13 Information Reporting system (Department of Water and Environmental Regulation, 2020).
Fig. 2.2. Photographs of (a) Mytilus galloprovincialis seeded onto the top 1 m of 1.5 m tall jarrah stakes, fitted with a long cable tie to facilitate future retrieval and (b) seeded stakes tethered in the shallow waters of Melville Water awaiting deployment at one of the three mussel reef sites.
Fig. 2.3. Underwater photographs showing a shellfish habitat site (a) prior to and (b) after deployment. Note the additional vertical structure provided by the stakes seeded with Mytilus galloprovincialis
2.3 Sampling regime
During seeding, three randomly selected stakes were taken to the laboratory for processing. Three more stakes were then removed monthly from each reef by snorkelling between May 2019 and February 2020 (excluding October 2019 due to logistical reasons). Each stake was placed in a large plastic bag and sealed using duct tape to retain any fauna that fell off the stake and prevent ‘cross-contamination’ of fauna between samples. In the laboratory, each M. galloprovincialis was removed to
14 allow estimation of the initial number of mussels per stake, as well as their body mass and total shell lengths.
2.4 Fish communities
Video monitoring of fish assemblages at shellfish habitat sites and adjacent unstructured control sites was conducted using Remote Underwater Video (RUVs) systems. RUVs were chosen over baited systems (BRUVs; Florisson et al., 2018) as the spread of the bait plume has the potential to attract fish from outside of the reef area and thus those fish that are not using the shellfish habitat directly. Moreover, the absence of bait removes the potentially confounding effect of variable bait plumes depending on the strength and direction of the tide (Taylor et al., 2013). Each RUV comprised a camera (GoPro Hero 2018; GoPro, Inc. California, USA) attached to a weighted rectangle steel frame (34 cm in height and 31 cm x 31 cm in length and width). GoPros are the most used camera of BRUV/RUV deployments in shallow waters. All footage was recorded at a resolution of 1920 x 1080 pixels at 60 frames s-1.
RUVs were deployed at each of the shellfish habitat sites (subsequently referred to as ‘reefs’) and their adjacent control site located 0.5 – 1 km away (Fig. 2.1) in each month between October 2018 and June 2020, except for January, March and October 2019. As three reefs were deployed and permutation-based statistics require a minimum of four replicates (Clarke & Gorley, 2015) at the lowest level of replication (in this case Habitat Type [Reef vs Control] in a given sampling occasion), on each sampling occasion one of the three reef and control sites were chosen at random and a second deployment was conducted the following day. Note that this does not affect the design of the experiment as no attempt was made to compare the fish fauna between reef sites or between control sites, i.e. Reef 1 vs Reef 2. Instead, the focus was placed on elucidating any differences between reef and control sites (Chapter 4). Note that RUV videos were collected prior to the deployment of the reefs between October 2018 and
15 February 2019, to determine whether there were any underlying differences in the characteristics of the fish fauna and then for 15 months post-deployment (April 2019 to June 2020). On each sampling occasion, RUVs were positioned at an angle where their field of view faced the reef directly (Fig. 2.4), as camera angle has been shown to influence the fish fauna recorded using underwater video methods on artificially engineered habitats (Tweedley et al., 2016a). Each RUV was deployed for a minimum of one hour.
The relative abundance of each fish and large crustacean species (e.g. Blue Swimmer Crabs, Portunus armatus, which are a key recreationally targeted species (Obregón et al., 2019)) were determined by counting the maximum number of individuals (MaxN) of each species in each frame. This is the most commonly used method of enumeration and prevents the overestimation of species through multiple counting (Whitmarsh et al., 2017). The MaxN value of each species was recorded in five- minute intervals as per Florisson et al. (2018), commencing from the moment the RUV frame hit the benthos and continuing for 60 minutes. Taxa were identified to their lowest taxonomic rank, which was predominantly to species level. However, some taxa (i.e. atherinids and monacanthids) were only identified to family level, as accurate species determination often requires detailed morphological examination under a microscope, which is not possible from video footage.
Each of the taxa recorded at the reef and control sites during the study was allocated to a number of functional guilds following the rationale given by Elliott et al. (2007) and Potter et al. (2015a). Three categories of guild were employed, namely (i) Estuarine usage (ii) Feeding mode and (iii) Habitat. Estuarine usage reflects extent to which species use the estuarine environment and for which parts of their lifecycle. Species belonging to six of the fourteen guilds were recorded. Marine straggler: species that spawn at sea and typically enter estuaries sporadically and in low numbers. Marine estuarine opportunist: also spawn at sea but regularly enter estuaries in large numbers
16 and particularly as juveniles. Solely estuarine: species were individuals complete their life cycles solely within the estuary. Estuarine & marine: species that can complete their life cycles within the estuary and in the marine environment. Estuarine & freshwater: species that can complete their life cycles within the estuary and in freshwaters. Semi- anadromous: species that undertake an anadromous migration, but whose spawning run extends only as far upstream as the upper estuary.
Feeding guilds were employed to represent the diet of each species and comprised six guilds. Piscivore: feed primarily on fish. Zooplanktivore: consume small crustaceans located in the water column. Zoobenthivore: predate on small organisms which live on or immediately above the benthos. Omnivore: diet includes a combination of organisms and plants. Detritivore: feed on decaying organic material and associated organisms. Opportunist: consume a wide variety of prey, the components of which change with availability.
Habitat guilds reflect the relative size of each species and their position within the water column. Five of such guilds were employed. Pelagic: medium to large species that occupy the middle and surface parts of the water column. Small pelagic: Small- bodied, usually schooling species, present in the middle and surface parts of the water column. Benthopelagic: species found throughout the water column. Demersal: species that live and feed on or close to the bottom of the water column. Small benthic: small- bodied species living on or in the substrate.
17
Fig. 2.4. Photographs extracted from remote underwater videos collected from the shellfish habitats deployed in the Swan-Canning Estuary. Fish observed in the photographs include (a) Western gobbleguts, Ostorhinchus rueppellii, (b) Fan-bellied Leatherjacket, Monacanthus chinensis (c) Blue swimmer crab, Portunus armatus (d) Eight lined Trumpeter, Helotes octolineatus (e) Atherinidae spp. (f) and the Western Australian Seahorse, Hippocampus angustus.
2.5 Invertebrate communities
Immediately on returning to the laboratory, individual mussel stakes were removed from their sealed bags and placed on a large plastic tray. Material on the top third of the stakes (~300 cm2) was removed using a paint scraper. Portions of the resultant invertebrates were then wet sieved through a stack of two sieves (2,000 and 500 µm) to separate larger invertebrates (e.g. barnacles and ascidians) from the smaller macroinvertebrates (e.g. amphipods and polychaetes). Samples from each sieve mesh
18 size were stored in a 70% ethanol solution. A total of three stakes were collected from each of the three mussel reefs on each monthly sampling occasion.
Under a dissecting microscope, invertebrates were separated from any material and sediment retained from the sieving and subsequently identified to their lowest taxonomic level. The rationale behind the collection of these samples was to describe how the invertebrate communities on the stakes changed over time due to colonisation and seasonal changes. Due to time constraints, only samples collected every second month were processed, i.e. the initial sample in May, then July, September, and December 2019. In the case of the invertebrates retained on the 2,000 µm sieve the blotted wet weight of each species in each sample was measured to the nearest 0.1 g.
19 Chapter 3: Biology of translocated Mytilus galloprovincialis deployed on shellfish habitats in the Swan-Canning Estuary
3.1 Abstract
Mytilus galloprovincialis is increasingly being used as a candidate species for shellfish reef restoration projects in Australia, including projects in the Swan-Canning
Estuary (Western Australia). This study investigated the biological characteristics of translocated M. galloprovincialis, which were collected from surface buoys in the entrance channel of the estuary and translocated to three reef sites in Melville Water.
Translocated mussels were sampled monthly from each of the three sites to determine their mortality, body condition and growth between April 2019 and February 2020. Results showed that translocated M. galloprovincialis mortality was high at all three sites, ranging from 2.48 y-1 to 3.06 y-1. Mortality was attributed mainly to the underlying poor water quality in Melville Water, compounded by stress associated with translocation, spawning activity, and competition with colonising ascidians. The body condition of M. galloprovincialis declined after translocation, attributed to spawning activity, which was consistent with the presence of recruits and increases in body condition in the months following the lowest values. Seasonally adjusted von Bertalanffy growth models best explained the growth of M. galloprovincialis. Growth was rapid, with individuals attaining a shell length of ~ 50 mm within their first year. The results of this study are the first of its kind to investigate the biology of M. galloprovincialis in the Swan-Canning Estuary. Given that this species is the candidate for future restoration projects in this estuary, the results are currently being used to inform the planning of a much larger-scale project.
20 3.2. Introduction
Organisms with indeterminate growth, such as mussels, allocate energy strategically into growth, maintenance and reproduction (Heino & Kaitala, 1999). Because such allocation strategies have implications for optimal lifetime fecundity, understanding these trade-offs has been a significant goal for evolutionary biologists and geneticists (Roff, 1992; Heino & Kaitala, 1999; Prieto et al., 2020). For many organisms, maintenance has priority (Jokela & Mutikainen, 1995), which is reflected in body condition (weight at a standardised size and those individuals with greater body condition tending to grow fastest (Lloret et al., 2014; Cottingham et al., 2018). In turn, faster growth improves survival and thus increases the probability of an individual surviving to a size at which allocating energy into reproduction optimises its lifetime fitness (Lorenzen & Enberg, 2002).
For many bivalves, such as Mytilus galloprovincialis, growth is not only influenced by the availability of food and thus the acquisition of energy, but is also a genetically determined trait (Prieto et al., 2020). Consequently, growth between populations can vary markedly, even when the individuals are exposed to identical environmental conditions (Prieto et al., 2018). As well as the very different growing environment in the Swan-Canning Estuary compared to macrotidal estuaries present in other parts of its geographical distribution, it is also relevant that the M. galloprovincialis population in the Swan-Canning contains unique genetics which are likely to comprise a combination of both native southern and northern hemisphere haplotypes (Dias et al., 2014).
As a filter feeder, when occurring in large numbers, M. galloprovincialis provides similar ecosystem services to oyster reefs but has several advantages. For example, mussels attain far greater densities than oysters, with individuals having similar filtering capacity (Riisgård et al., 2014). Thus, proportionally a mussel reef has a far greater filtration capacity than a similar habitat comprised of oysters. Furthermore, historically,
21 oysters built vertical reefs that accrued over hundreds of years and so it would take a similar amount of time to rebuild these structures. In contrast, there are several approaches that can be used to create, almost immediately, three-dimensional mussel reef structures. While such structures have typically been deployed as tendrils suspended under rafts, which are mobile (Galimany et al., 2017), this study was the first to attempt to establish a permanent vertical reef using stakes seeded with individuals of M. galloprovincialis.
The aim of this chapter is to compare the mortality, growth, and body condition of M. galloprovincialis among three different reef sites in the Swan-Canning Estuary, which were deployed in April 2019 and subsequently monitored monthly for ~ 1 year. The results of this chapter have informed other shellfish reef ‘restoration’ efforts in this estuary, with a multi-million-dollar reef set to be deployed by The Nature Conservancy in mid-2021.
3.3 Materials and methods 3.3.1 Laboratory methods
Following the deployment of the shellfish habitats, three stakes were removed from each of the three sites (subsequently referred to as reefs) each month and transported to the laboratory for processing. The above ground portion of each stake was divided into three sections (i.e. top, middle, bottom) and the number of M. galloprovincialis alive on each stake section were enumerated. Dead individuals were not counted. The shell length of each M. galloprovincialis was measured to the nearest 0.5 mm using callipers (Fig. 3.1a). Dry weight (excluding the shell) was determined by removing all tissue and drying in an oven at 60 °C for 72 h or until a constant dry weight was reached (Fig. 3.1b,c; Davenport & Chen, 1987). The assumption was made that the individuals of M. galloprovincialis were of the same age, recruited from the winter
22 spawning event of 2018, and were thus ~ 0.7 years old at the time of translocation onto the reefs.
Fig. 3.1. Photographs showing the (a) measurement of M. galloprovincialis, (b) the amount of tissue removed from each mussel and (c) tissue being dried to a constant weigh in a drying oven.
3.3.2 Statistical analysis 3.3.2.1 Mortality The number of alive mussels was recorded for each stake section in each month of the study period. As likelihood ratio tests demonstrated mortality did not differ statistically among sections of the stake (i.e. top, middle and bottom; P = 0.277), the data for each stake section were pooled. Pairwise likelihood ratio tests were then employed to determine whether the mortality of M. galloprovincialis differed among the three reefs.
3.3.2.2 Body condition The resultant dry tissue mass was subjected to Analysis of Covariance (ANCOVA) using the Statistical Package for the Social Sciences (IBM SPSS version 24.0) to determine whether differences in body condition differed among the stake sections, reef or month. For this analysis, stake, reef, and month were fixed factors, dry mass the independent variable and shell length the covariate. A Bonferroni adjustment was employed to reduce the impact of type 1 error during multiple means-testing.
23 3.3.2.3 Growth Although von Bertalanffy growth models have been used extensively to describe the growth of mussels throughout their life, due to the limited age range in the current study, i.e. ~1 year, three models were initially fitted; a two-parameter linear model, 푇퐿 = 푎푡 + 푏; where 푇퐿 is the total shell length (mm) and 푡 is the mussels age (years-1). The parameters 푎 and 푏 where calculated using the maximum likelihood approach. The
−푘(푡−푡0) three-parameter von Bertalanffy model, 퐿푡 = 퐿∞[1 − 푒 ], where 퐿∞ is the
-1 asymptotic length of the mussel, 푘 is the growth rate (years ) and 푡0 is the time where the mussel would have zero length, and a five parameter seasonally adjusted von
−푘(푡−푡0)+푆(푡)−푆(푡0) Bertalanffy growth model, 퐿푡 = 퐿∞[1 − 푒 ], where 푆(푡) = 퐶푘 퐶푘 ( ) 푆푖푛휋(푡 − 푡 ) and 푆(푡 ) = ( ) 푆푖푛휋(푡 − 푡 ). Seasonal variation in growth is 2휋 푠 0 2휋 0 푠 denoted by the parameter 퐶. When 퐶 = 0 there is no seasonal variation. Additionally, if 퐶 is > 1 or <1 the growth of the organism shall be negative during the season with non- growth. The parameter 푡푠 denotes the time between the growth oscillation and 푡 = 0.
To determine the most appropriate model, the data for all stake sections and reefs were pooled. With increasing levels of complexity, the Akaike Information Criterion (AIC) was used to determine the best fitting model, i.e. the model with the smallest AIC value, calculated as 퐴퐼퐶 = 2푛푝 − 2퐿퐿, where 푛푝 is the number of 푛 parameters and 퐿퐿 is the log-likelihood; 퐿퐿 = [ln(2휋) + ln(휎̂) + 1], where 푛 is the 2 number of mussels, 푙푛 is the natural logarithm, 휋 is the constant pi and 휎̂ is: (퐿 −퐿̂ )2 ∑푛 푗 푗 . Here the observed and expected lengths of the 푗th mussel are denoted by 푗=1 푛
퐿푗 and 퐿̂푗, respectively. As the AIC values demonstrated that the seasonally adjusted von Bertalanffy growth model was the most appropriate model, this model was then fitted separately, and collectively, to the total lengths and ages of M. galloprovincialis for all three stake sections (top, middle, bottom) and compared using pairwise likelihood ratio test
(Cerrato, 1990). As the hypothesis that the growth of M. galloprovincialis for all three stake sections can be explained by one growth curve was accepted at 5% significance (α
24 = 0.05), the data for each stake section were pooled. Pairwise likelihood ratio tests were employed to test whether the growth of M. galloprovincialis differed between reefs.
3.4 Results 3.4.1 Environmental parameters
While mean monthly salinities in bottom waters did not vary considerably from
36 ppt, those of surface waters decreased precipitously from that value during July 2019 and attained their minimum of 20 ppt during September 2019 (Fig. 3.2a). Mean monthly bottom water salinities did not return to full strength seawater (36 ppt) until January 2020 and remained at that level for the duration of the study.
Between April and July 2019, mean monthly surface water dissolved oxygen concentrations recorded during daylight rose from ~ 7.0 to 9.0 mgL-1 before declining in subsequent months to a minimum of 6.0 mgL-1 in February 2020. Concentrations varied markedly between surface and bottom waters from July 2019 to December 2020. In contrast, concentrations were ~ 6.5 mgL-1 in bottom waters between April and June 2019, before declining to < 4 mgL-1 in during July 2019 to ~ 4 mgL-1 and remained at this level until December 2019 before increasing to 5 mgL-1 for the remainder of the study period. Dissolved oxygen was at its lowest during September 2019 (~ 3 mgL-1) and displayed minimal variability. Bottom water dissolved oxygen returned to ~ 6.5 mgL-1 again during January and February 2020 (Fig. 3.2b).
Surface water temperature followed a seasonal trend, decreasing from 21 to 15 ˚C between April and June 2019, before increasing gradually to 25 ˚C in December 2019 incrementally increasing into spring and summer and maintaining its maximum ~24 from December 2019 until the culmination of the trial at the end of February 2020. Bottom waters followed a similar trend albeit slightly warmer during July and August
2019 and cooler between September 2019 and February 2020 (Fig. 3.2c).
25 (a) Surface waters 40 Bottom waters
35
30
25 Salinity (ppt) Salinity
20
15
Jul 19 Jul
Jan 20 Jan
Jun 19
Oct 19 Oct
Apr 19
Sep 19
Feb 20
Dec 19 Dec
Aug 19 Aug Nov 19 Nov May 19 May
(b)
(c)
Fig. 3.2. Mean monthly (a) salinity (ppt), (b) dissolved oxygen concentrations (mgL-1) and (c) water temperature (°C), showing surface waters (grey) and bottom waters (black) in Melville Water of the Swan- Canning Estuary between April 2019 and February 2020. Data obtained from the Water Information Reporting system (Department of Water and Environmental Regulation, 2020).
26 3.4.2 Mortality
The mortality rate was greatest at Reef 1 (3.06 y-1), which was only slightly greater than that at Reef 2 (2.96 y-1), and lowest at Reef 3 (2.48 y-1; Fig. 3.3). Pairwise comparisons between reefs detected significant differences between Reef 1 and Reef 3 (P = 0.039) and between Reef 2 and Reef 3 (P = 0.028; Fig. 3.4). There was no significant difference between Reef 1 and Reef 2 (P = 0.440).
27 (a)
(b) 140 Reef 2
120
100
80
60 Number Alive Number 40
20
0
0.6 0.8 1.0 1.2 1.4 1.6 1.8
(c)
140 Reef 3 120
100
80
60 Number alive Number 40
20
0
0.6 0.8 1.0 1.2 1.4 1.6 1.8 Age (years)
Fig. 3.3. Mortality curves fitted to the monthly number of alive M. galloprovincialis present on each stake at reefs 1, 2 and 3 (a, b and c, respectively) in the Swan-Canning Estuary from April 2019 to February 2020.
28
Fig. 3.4. Comparisons of the mortality curves fitted to the numbers of alive M. galloprovincialis at three sites in the Swan-Canning Estuary between April 2019 and February 2020.
3.4.3 Body condition
In April 2020, i.e. prior to the translocation of the mussels to the reef sites, the average dry weight of tissue, for mussels of a standardised shell length of 61 mm, was 0.97 g (Fig. 3.5). During the ensuing months, condition declined, reaching a minimum in
July and August (2019) of between 0.40 and 0.42 g. In the following months, however, condition improved, with the mean dry mass in February 2020 being greatest at Reef 2 (0.98 g) and least at Reef 1 (0.57 g; Fig. 3.5).
29 1.1 Reef 1 1.0 Reef 2 Reef 3 0.9
0.8
0.7
0.6 Dry mass (g) mass Dry
0.5
0.4
0.3
Jul 19
Jun 19
Apr 19
Feb 20
Jan 20
Sep 19
Dec 19
Aug 19
Nov 19 May 19 Month
Fig. 3.5. Monthly mean dry mass of M. galloprovincialis meat standardised for shell length of 61 mm, at the three sites in the Swan-Canning Estuary between April 2019 and February 2020.
3.4.4 Growth
The modal 5 mm length classes of M. galloprovincialis when the stakes were seeded and deployed in April was 50-55 mm (Fig. 3.6). At Reef 1, modal length class progressed to 55-60 in May and 60-65 mm in June, where it remained until December. At Reef 2, modal length classes typically increased each month attaining 70-75 mm by
November. Reef 3 followed a similar trend, with the modal length attaining 65-75 mm by November. A second cohort was evident at Reef 2 in August and Reef 3 in September, when they were ~ 10 mm in length. There was also evidence that mussel was recruited onto Reef 1, but this was not detected until February 2020, at which time they were ~ 4 mm in length. There was also a high mortality of the translocated mussels at Reef 3 in December (Fig. 3.6).
30 40 Apr 20 0 40 May 20 0 40 Jun 20 0 40 20 Jul 0 40 Aug 20 0 40 Sep 20
Contribution (%) Contribution 0 40 Nov 20 0 40 100 Dec 20 0 40 Jan 20 0 40 20 Feb 0 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
Length (mm) Fig. 3.6. Monthly length-frequency histograms of translocated Mytilus galloprovincialis at the three reef sites (i.e. 1 = ◼, 2 = ◼, 3 = ◼) in the Swan-Canning Estuary between April 2019 and February 2020. Dashed lines are presented at 60 mm for reference.
Although the seasonally adjusted von Bertalanffy growth model was shown to provide the better fit of three models to the age and length data for the translocated M. galloprovincialis, the fit was poor for Reef 1 and Reef 3, as demonstrated by the r2 = 0.43 and 0.20, respectively, but provided a good fit to the data from Reef 2 (r2 = 0.73)
(Table 3.1; Fig. 3.7). Growth trajectories of each reef were all significantly different from each other (P = < 0.001 for all) and is demonstrated in Fig. 3.8.
31 (a) 90
80
70
60
50
40
Length (mm) Length 30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
(b) 90
80
70
60
50
40
Length (mm) Length 30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
(c)
Fig. 3.7. Seasonally adjusted von Bertalanffy growth curve fitted to the total shell lengths and age of Mytilus galloprovincialis from reefs 1, 2 and 3 (a, b and c, respectively) in the Swan-Canning Estuary between April 2019 and February 2021.
32 90 Reef 1 80 Reef 2 Reef 3 70
60
50
40
Length (mm) Length 30
20
10
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Age (years)
Fig. 3.8. Comparisons of the seasonal von Bertalanffy growth models fitted to length at age data for Mytilus galloprovincialis at three reef sites in the Swan-Canning Estuary between April 2019 and February 2020.
Table 3.1. Seasonally adjusted von Bertalanffy growth curve parameters and their 95% confidence limits for translocated M. galloprovincialis from 3 reef sites in the Swan-Canning Estuary from April 2019 – -1 February 2020. 퐿∞ is the asymptotic length (mm); 푘 the growth coefficient (years ); 푡0 the hypothetical age (years-1) where the mussel has zero length, C the amplitude of seasonal oscillation; Ts the time between T = 0 and the start of the growth oscillation; r2 the coefficient of determination and n the number of mussels.
2 L∞ k C t0 ts r n Reef 1 Estimate 1084.37 0.02 0.08 -1.32 0.07 0.20 796 Lower -83125.97 -1.97 0.05 -4.64 -0.03 Upper 85294.70 2.02 0.11 1.99 0.17 Reef 2 Estimate 76.16 1.89 0.05 0.01 0.23 0.73 1184 Lower 70.43 1.35 0.04 -0.06 0.15 upper 81.88 2.43 0.06 0.09 0.30 Reef 3 Estimate 75.80 1.90 0.01 0.09 -0.28 0.43 998 lower 70.10 1.30 - 0.01 -0.01 -0.64 upper 81.40 2.50 0.03 0.18 0.08
33 3.5 Discussion
This study was the first of its kind globally to document the biology of translocated M. galloprovincialis onto fixed vertical structures (i.e. wooden stakes). The efficacy of using such structures for future shellfish reef restoration/enhancement efforts was assessed by measuring the mortality, body condition and growth of ~ 2,800 translocated M. galloprovincialis over a year at three sites in the Swan-Canning Estuary.
The results of this trial have already been employed to inform the site selection and methodology of a 50-hectare mussel reef being deployed in this estuary by The Nature Conservancy which will be completed by mid-2021.
3.5.1 Mortality
While few studies have provided estimates of mortality of M. galloprovincialis, the maximum life span of this species is thought to be around four years (Okaniwa et al., 2010) and thus the mortality rate, based on maximum age and the equation relating
−0.916 -1 mortality to longevity by Then et al. (2015), i.e. 푀 = 4.899푡푚푎푥 , is ~1.38 y . This value is considerably less than those for the translocated M. galloprovincialis in the Swan- Canning Estuary, which ranged from 2.48 to 3.06 y-1. Although the exact factors responsible for the higher mortality rate recorded during this study cannot be determined from the current data, there are several potential reasons why this may have occurred.
The bottom waters of Melville Water were subjected to prolonged periods of low oxygen, with daytime values being consistently below 4 mgL-1 between July and October 2019. As, due to respiration by phytoplankton and benthic metabolism, oxygen concentrations decline during the night (Rose et al., 2019; 2020), minimum values were almost certainly lower than those recorded during the daytime by the DBCA. It is likely that some of the early mortality may have been attributed to the stress related to the translocation, particularly as mussels were translocated from surface buoys, held out of
34 water for several hours while being seeded onto stakes and then deployed into waters of ~ 4 m depth. Mortality related to the translocation, e.g. due to increased metabolic stress, has likewise been demonstrated in other species of Mytilus (Yanick et al., 2003; Kovačić et al., 2018).
Suspended sediment was visually observed during the sampling of seeded stakes at each of the three sites, however, the extent of which was more pronounced at reefs
1 and 2 than at Reef 3. Sedimentation has been shown to affect the filtration rate of M. galloprovincialis (Pinheiro et al., 2019). For example, sand particles can cause abrasions to the gill tissue and, over prolonged periods of exposure, can result in reduced fitness and mortality of the mussel (Zardi et al., 2008).
As M. galloprovincialis typically attains maturity after one year (Okaniwa et al., 2010) it is likely that spawning occurred during the winter months. Spawning has also been linked to mass mortality events (Tremblay et al., 1998; Myrand et al., 2000). For example, in the southern Gulf of St. Lawrence, high mortality occurs in M. edulis following peak spawning events as a result of low glycogen levels and stressful environmental conditions (Myrand et al., 2000).
Lastly, the effect of the fouling ascidian communities cannot be disregarded as a contributing factor to some of the mortality exhibited at the three sites. Ascidians, belonging to several species, rapidly colonised the mussel stakes in large numbers (see Chapter 4). Once established, these sessile organisms may reduce water flow and thus the availability of food and dissolved oxygen (Lodeiros & Himmelman, 1996; Taylor et al., 1997; Su et al., 2007; Lins & Rocha, 2020). Furthermore, they contribute a considerable amount of weight to the mussels, potentially dislodging them from their byssal attachment on the stake. This theory is supported by the fact that dislodged mussels fouled with ascidians were observed on the substrate during stake retrieval (C. Maus; personal observation). The new space was, rapidly colonised by ascidian
35 recruits (Chapter 4), furthering the competition for space, food, and oxygen. Given the already low dissolved oxygen concentrations in the bottom waters of Melville Water, the compounding effect of reduced oxygen availability as a result of the ascidians may have also contributed to the mortality of the mussels at each of the three sites.
3.5.2 Body condition
The body condition of M. galloprovincialis decreased instantly from translocation, particularly at Reef 1. It is highly relevant that the translocation of the mussels occurred just prior to the winter spawning season at which time energy would be expected to be allocated into reproduction. Such seasonal fluctuations in the body condition of M. galloprovincialis are well documented and attributed to primarily to weight reduction attributed to gamete development and spawning (Çelik et al., 2012;
Karayücel et al., 2015). Similar declines in body condition during the winter months have been observed in other marine bivalves, such as the New Zealand Green Lipped mussel, Perna canaliculus (Hickman & Illingworth, 1980). The conclusion that the reduction in body condition was related to spawning is consistent with the presence of new recruits around August and that body condition increased during the ensuing months.
3.5.3 Growth
The seasonally adjusted von Bertalanffy growth model best fit the length at age data for the translocated mussels. While it provided an adequate fit for Reef 2, the fit was not so good for reefs 1 and 3 as shown by their coefficient of determination. This is likely the result of restrictive age data, in that this study only was able to collect data over an 11-month period. The seasonal growth parameter C returned low values for all three reefs. These low values suggest that, while seasonal growth was evident in the translocated M. galloprovincialis, demonstrated by providing a better fit than the
36 traditional von Bertalanffy growth model, season did not play a major role in growth patterns of the translocated M. galloprovincialis. Seasonal growth patterns are common in shellfish due to their relationship with food availability and temperature. For example, the growth of longlined cultured M. galloprovincialis in the Bay of Biscay, in the Eastern Atlantic Ocean declining during the winter months (Azpeitia et al., 2016, 2018). Interestingly, the extent and magnitude of these seasonal changes in water temperature are the same as experienced in the Swan-Canning Estuary.
The growth of M. galloprovincialis in the Swan-Canning Estuary was greater than that reported in several other studies, in which mussels attained shell lengths of less than 50 mm by the end of their first year of life (Ceccherelli & Rossi, 1984; Abada- Boudjema & Dauvin, 1995; Villalba, 2005; Lök et al., 2007; Okaniwa et al., 2010). For example, on the Algerian coast, Abada-Boudjema & Dauvin (1995) found
M. galloprovincialis to only attain shell lengths ~16-28 mm, whereas in north-west Spain and Tokyo Bay, Japan this species reached 30-40 mm at the end of their first year of life (Villalba, 2005; Okaniwa et al., 2010). Thus, in terms of growth, M. galloprovincialis in the Swan-Canning Estuary performed better than that recoreded for this species elsewhere. It is thus relevant that this system was identified as having the second highest annual phytoplankton production of the 131 estuaries studied by Cloern et al. (2014), implying that adundance of their phytoplankton prey are plentiful.
37 Chapter 4: Faunal community associated with deployed shellfish habitats in the Swan-Canning Estuary
4.1. Abstract
Shellfish habitats provide a range of ecosystem services, including their ability to increase fisheries production. This study aimed to determine whether the deployment of novel vertical shellfish habitats (reefs) seeded with translocated mussels in the Swan- Canning Estuary led to changes in the fish and macroinvertebrate communities. Remote underwater video was used, monthly between April 2019 – June 2020, to survey the fish communities present on the reefs and nearby unvegetated substrate. Fish faunal composition differed between the reefs and unvegetated sites, with an increased number of species observed on the reefs. Temporal variations in faunal composition reflected the seasonal ways in which fish utilise estuaries, i.e. an increase in diversity in summer due to the presence of stable salinities that favour marine species. There was an increased number of zoobenthic feeding species on the reefs than unvegetated control sites (16 vs 10), implying that the reefs provided invertebrate prey for these fish to feed on. Large invertebrate communities on the reefs, which comprised mainly ascidians and barnacles, differed between reef sites and diverged over time following the loss of mussels. Smaller invertebrate communities associated with the reef, typically small crustaceans such as isopods and amphipods, varied among reef sites and over time. Temporal trends in this fauna was attributed to the underlying poor water quality during the winter months (e.g. stratification and low dissolved oxygen concentrations) and rapid proliferation occurred in December when water quality improved. The results from this pilot study will increase the understanding of how fish and invertebrate species utilise structurally complex vertical shellfish habitats in temperate Australian estuaries, where different types of shellfish habitats are currently being constructed.
38 4.2 Introduction
Located at the interface of both freshwater and marine ecosystems, estuaries are dynamic ecosystem that harbours a unique faunal community (McLusky & Elliott, 2004). Due to changes in the extent of their contrasting influences, e.g. freshwater and tidal flow, there are marked environmental gradients present in estuaries, the magnitudes of which differ over a range of temporal scales (e.g. tidal and seasonal). The characteristics of faunal communities often change along the gradients, particularly in response to salinity (Whitfield et al., 2012) and undergo seasonal changes (Elliott et al., 2007).
Faunal composition in estuaries has been shown to differ markedly between habitats. This is particularly the case when comparing structured habitats, such as seagrass meadows, mangroves and shellfish reefs, and those with limited structural complexity, e.g. bare sand (Humphries & Peyre, 2015; Whitfield, 2017; Gilby et al., 2018). For example, Bloomfield & Gillanders (2005) sampled fish and epifaunal invertebrates in seagrass, mangroves, saltmarshes and unvegetated habitats in the Barker Inlet-Port River Estuary in South Australia. Although a total of 44 species were collected, only one, an atherinid, was found in all habitats. Given the marked influence structural habitats have on faunal composition, the addition of new habitat would be expected to attract different communities, based on habitat isolation and species-specific traits, such as dietary and habitat preference. For example, small low-relief artificial reefs in three south-eastern Australian estuaries harboured a distinct and more speciose fish assemblages than those that resided on adjacent natural rocky-reef and bare sand habitats (Folpp et al., 2013).
Shellfish reefs generally harbour an increased abundance and diversity of fish species than unvegetated sites (Tolley & Volety, 2005; Humphries & Peyre, 2015).
Following the restoration of oyster reefs in the Noosa River Estuary in Queensland, the new habitat harboured 1.4 times the fish abundance present in unvegetated sites (Gilby
39 et al., 2019). The authors attributed this to the increased feeding opportunities and refuge these habitats provided. Thus, complex habitats with high structural complexity and vertical relief result in increased abundance and diversity among estuarine fish communities, compared to featureless habitats (Gilby et al., 2018).
Following the provision of new or restored habitat, there is a pattern of colonisation and succession (Chapman, 2002; Moseman et al., 2004). The rate at which these processes occur in estuaries appears to vary among systems and is largely dependent on complex interactions such as the amount of pre-existing habitat, which act as sources of natural recruitment (Pizzolon et al., 2008; Lowry et al., 2014). Initial recruitment to new habitat appears to be rapid, exceeding that of natural habitats, however, is followed by a decline in abundance several years post-restoration (Ziegler et al., 2018) as the system reaches its ideal carrying capacity (Choi, 2004).
As with fish, the benthic macroinvertebrate communities present in estuaries vary markedly from those in coastal waters (Platell & Potter, 1996; Tweedley et al., 2015). Within estuaries, they vary both spatially and temporally, often a result of environmental changes, such as salinity, dissolved oxygen concentration and sediment composition (Tweedley et al., 2016b). Depth also plays a role in benthic macroinvertebrate community structure, with more species found in shallower waters associated with macrophytes and shell-debris than in deeper, muddy habitats (Platell & Potter, 1996).
The complex habitat of shellfish reefs provides food resources and refuge for invertebrates compared to bare sand habitats, with oysters reefs being associated with a five-fold increase in abundance of benthic macroinvertebrates (McLeod et al., 2020). Moreover, the introduction of hard structures into estuarine waters provides a unique colonising opportunity for epibiont fauna such as tunicates (Coleman & Connell, 2001),
40 which may not be as prolific in estuarine seascapes largely devoid of hard structure (Tweedley et al. 2014). Reproductive strategy assists in the colonisation of new environments and habitats. R strategist species focus primarily on the production of large amounts of juveniles, fast growth and so are able to colonise newly available habitat faster and proliferate in abundance compared to K strategist species (Parry, 1981). Furthermore, ascidians, which often colonise hard substrates, are hermaphroditic and predominantly reproduce through sexual cross-fertilisation and the production of free-swimming larvae (Rius et al., 2010). Additionally, those that are non-native to Australia are often able to reproduce throughout the year aiding in their proliferation (Bourque et al., 2007; Pineda et al., 2013).
This chapter investigates the impact of novel vertical shellfish habitats (referred to as mussel reefs) on the fish and invertebrate fauna in the Melville Water region of the Swan-Canning Estuary in south-western Australia. The aims of this chapter are two-fold.
1) Compare the characteristics of the fish fauna on the mussel reefs and nearby unstructured habitats. 2) Determine which benthic invertebrate and colonising tunicate species were associated with these structures.
4.3. Materials and methods
Details of the design, construction and deployment of the mussel reefs are given in Chapter 2, together with a map of their location and those of the adjacent control sites. The methodology for the sampling of the fish and invertebrate communities is also provided in Chapter 2, however, is summarised here to provide context for the description of data analyses. In brief, Remote Underwater Video (RUV) systems were used to survey the fish fauna (which includes a recreationally important decapod crustacean) of the mussel reefs and nearby control sites over 14 months between April
41 2019 to June 2020. Note that sampling of the fish communities was conducted prior (October – December 2018 and February 2019) to the deployment of the reefs to determine whether there were any pre-existing differences in fish communities between proposed reef and control sites. A total of 144 RUV videos were analysed; 32 prior to the deployment of the mussel reefs and 112 afterwards, with 56 obtained from the mussel reefs and 56 from the nearby control sites.
The invertebrates directly attached to the stakes and their associated epibiont community were collected from three stakes from each mussel reef on each of four sampling occasions, i.e. directly after seeding (May) and in July, September and December 2019 (36 samples). Each sample was then wet-sieved into two sizes (> 2000 and 500-2000 µm) and the data for each size class analysed separately as previous studies have detected significant differences in faunal composition with size mesh size
(e.g. Schlacher & Wooldridge, 1996; Somerfield et al., 2018). These faunas are referred to as ‘large’ and ‘small’, respectively. Note that the invertebrate fauna includes several ascidian species (see below), but for convivence is still referred to as an invertebrate community.
4.3.1 Statistical analysis
All statistical analyses for fish and invertebrate communities were performed using the PRIMER v7 multivariate statistics software package (Clarke & Gorley, 2015) with the PERMANOVA+ add-on (Anderson et al., 2008).
4.3.1.1 Fish communities: preliminary analyses To assess whether a 60-minute soak time for the RUVs was enough to accurately represent the characteristics of the fish fauna on the mussel reefs and control sites, preliminary analyses were undertaken. The MaxN of each species in each five-minute
42 interval of each of the 144 videos were subjected to the DIVERSE routine to calculate the number of species, total MaxN (i.e. the sum of the MaxN values for individual species) and Simpson’s Index (Simpson, 1949). The resultant values for each of the three univariate variables were then averaged to provide a single value for each five-minute interval at each habitat (i.e. mussel reef and control site) before and after the deployment of the reefs to give four habitat and deployment combinations and thus remove any potentially confounding temporal influences (i.e. by pooling months). Increases in the mean values for each of the three univariate variables with time were plotted as accumulation curves.
Changes in species composition over time were also investigated. Firstly, the MaxN values of each species in each five-minute interval for each of the four habitat and deployment combinations were firstly dispersion-weighted (Clarke et al., 2006) to down-weight the abundances of heavily-schooling species, such as Atherinidae spp. and Ostorhinchus rueppellii, whose numbers are erratic over replicates, relative to those species that return more consistent values, e.g. Anoplocapros amygdaloides (Veale et al., 2014; Potter et al., 2016). These dispersion-weighted data were then square-root transformed to balance the contribution of relatively abundant species, compared to those with lower MaxN values (Clarke et al., 2014). The transformed data for each five- minute interval were then averaged across the replicates for each habitat and deployment combination and used to construct a Bray-Curtis resemblance matrix. This matrix was subjected to hierarchical agglomerative clustering (CLUSTER; Clarke et al., 2014) to determine the time intervals that were ≥ 95 % similar in terms of their species composition and used to construct a non-metric Multi-Dimensional Scaling (nMDS) ordination plot (Clarke, 1993).
Finally, the transformed MaxN data for each time interval in each habitat and deployment combination were used to construct a shade-plot (Clarke et al., 2014). The shade plot is a visualisation of the averaged data matrix, where a white space for a
43 species demonstrates that it was not recorded, while the depth and colour of shading, ranging from light grey shades through the spectrum to black, represents increasing values for the abundance of that species in that time interval. The averaged samples (x axis of the plot) are ordered from lowest to the highest time interval for each reef. Species (y axis of the plot) are ordered to optimise the seriation statistic ρ by non- parametrically correlating their resemblances to the distance structure of a linear sequence and constrained by a cluster dendrogram (Clarke et al., 2014).
4.3.1.2 Fish communities: univariate analyses The MaxN of each species in each replicate RUV sample (60 minutes) was used to construct a data matrix that was subjected to DIVERSE routine to calculate the number of species, total MaxN and Simpson’s Index. Prior to any analyses, the data were examined using a Draftsman plot, to visually assess the extent to which the distribution of values for each variable were skewed and, if so, the type of transformation required to ameliorate this effect. This showed the number of species and Simpson’s Index did not require transformation, but that total MaxN needed to be fourth-rooted.
Each of the three univariate variables before and after the deployment of the mussel reef was used to construct a Euclidean distance matrix. In turn, the six matrices
(e.g. the number of species before the deployment) was subjected to a two-way univariate Permutational Multivariate Analysis of Variance (PERMANOVA; Anderson et al., 2008) test to determine whether they varied significantly across Habitat (two levels; mussel reef and control) and Time (two levels; pre-deployment, i.e. October-December 2018 and February 2019 and post-deployment, i.e. the 14 months between April 2019 and June 2020). In this and all other analyses, the null hypothesis of no significant differences among a priori groups was rejected if the significance level was P ≤ 0.05. The percentage contribution of the mean squares to each term in the model was also calculated as a way of indicating the relative importance of each term in the model (Crisp
44 et al., 2018). The main source of any significant differences was investigated by examining plots of the means, back transformed where necessary, with their associated 95% confidence limits.
4.3.1.3 Fish communities: multivariate analyses The data matrix of the MaxN abundance of each species in each replicate sample was divided, to separate those samples collected at each site before and after the deployment of the mussel reefs. The data in each matrix were dispersion-weighted, square-root transformed and used to construct a Bray-Curtis resemblance matrix. These were separately subjected to the same two-way PERMANOVA design above. As in the case of both the pre- and post-deployment data, Habitat only has two levels, no post- hoc testing was required. However, if a significant difference with Time, either as a main effect or the Time × Habitat interactions then post-hoc testing was conducted using one- way Analysis of Similarity (ANOSIM; Clarke & Green, 1988). ANOSIM was preferred at this stage of the analyses as, unlike PERMANOVA, it is fully non-parametric (and thus more robust), and the test statistic (R) provides a universal measure of group separation (Lek et al., 2011). R statistic values typically range from 1, when all samples within each group are more like each other than to any of the samples from other groups, down to 0, when the average similarity among and within groups do not differ (Clarke & Gorley,
2015).
Trends in the fish faunal community data were displayed visually using a range of multi-dimensional scaling (MDS) approaches. When a significant difference was detected for Habitat (pre- and post-deployment) and Time (pre-deployment only) the Bray-Curtis resemblance matrix derived from the replicate samples was subjected to the Bootstrap Averages Routine (Clarke and Gorley, 2015) to bootstrap those samples in m- dimensional metric multi-dimensional scaling (mMDS) space. The averages of repeated bootstrap samples (bootstrapped averages) for each group of samples were used to
45 construct an mMDS ordination plot. Superimposed on each plot was a point representing the group average (i.e. the average of the bootstrapped averages) and, for two-dimensional plots, the associated, smoothed and marginally bias-corrected 95% bootstrap region, in which 95% of the bootstrapped averages fall. As the Time main effect in the post-deployment contained 14 levels, patterns in these data were not able to be visually distinguished using the bootstrapped mMDS approach. Thus, the Bray-
Curtis resemblance matrix was used to construct a centroid nMDS ordination plot. This involves producing a distance-among-centroids matrix, which calculates the averages between each pair of samples in ‘Bray-Curtis space’ and uses those distances to produce an nMDS plot (Lek et al., 2011).
Shade plots were used to illustrate trends in the dispersion-weighted and transformed abundance of each species across habitats and months separately for the pre- and post-deployment data. Species (y-axis) were seriated based on their Bray-Curtis similarities and placed in optimum serial order.
4.3.1.4 Invertebrate communities As in the case of the large invertebrates, both a blotted wet-weight and numerical abundance were calculated. A preliminary analysis was used to determine whether the trends in the two data sets (biomass and abundance) were statistically similar (Wu et al., 2019). The dispersion-weighted and square-root transformed average abundance of each species and the loge (x+0.01) average biomass of each species in each 2,000 µm sample, were used to construct separate Bray-Curtis resemblance matrices. Pairwise combinations of the matrices were subjected to separate RELATE tests to determine if they were significantly correlated (Clarke and Gorley, 2015). The value for the test statistic (ρ) reflects the strength of the correlation, with ρ values ranging from
~ 0 (little correlation) to ~ 1 (near-perfect correlation). The results demonstrated that there was a strong and significant correlation between the abundance and biomass of
46 the large invertebrates (P = 0.001; ρ = 0.867) and so only the abundance data were used in subsequent analyses.
The univariate and multivariate analyses for the large and small invertebrates followed essentially the same approach for the fish communities with two differences. Firstly, that characteristics in the fauna, i.e. the number of species, number of individuals, Simpson’s Index and community composition, were only investigated after the deployment of the mussel reefs. Secondly, the focus of the two-way PERMANOVA design was altered to detect the potential influence of Month (four levels, i.e. May, July, September and December 2019) and Site (three levels, i.e. Reef 1, 2 and 3).
4.4. Results 4.4.1. Environmental conditions
Although described in detail in Chapter 3, the environmental conditions at the time of sampling are briefly summarised here to provide context for any changes in faunal composition. Salinity at the bottom of the water column remained relatively stable at between 30-35 ppt. In contrast, salinities at the surface were 36 ppt during the summer but decline markedly during the winter months to ~ 20 ppt, resulting in vertical stratification of the water column. Dissolved oxygen concentrations at the bottom of the water column remained below 4 mgL-1 for an extended period during the winter and early spring (July – October 2019). Water temperature followed a seasonal trend, reaching its lowest value of 15 °C during July 2019 and increasing sequentially until December 2019 (25 °C). High loads of suspended sediment were observed in the water column upon retrieval of stakes (C. Maus; Personal observation), which occurred between July and September, when the water column was stratified, and dissolved oxygen concentrations declined (Fig. 3.2b).
47
4.4.2 Fish communities 4.4.2.1 Preliminary analyses Visual inspection of the accumulation curves for the number of species, total MaxN and Simpson’s diversity for each of the four habitat and deployment combinations indicated that the values all reached an asymptote prior to the 60-minute mark and that there was an almost complete overlap of the 95% confidence limits (Appendix 1, 2). For fish faunal composition, the extent of Bray-Curtis similarity decreased as time increased in each of the four combinations, and the composition at 60 minutes was ≥ 95% similar to those at earlier time intervals ranging from as little as 35 minutes in the pre-control data to as much as 50 minutes in the post-control (Appendix 3a). This is shown visually on the nMDS plot where, after these times, the symbols representing the points essentially converge (Appendix 3b). The consistency in the faunal assemblage prior to 60 minutes is further illustrated on the shade plot, where the MaxN abundance of each species ‘stabilise’ prior to the 60-minute mark in the video (Appendix 4). Based on these results, it can be concluded that 60 minutes was a sufficient length of time to deploy a RUV to provide a representative sample of the fish fauna in the Swan-Canning Estuary.
4.4.2.2. Description of the fish fauna A total MaxN of 6,946 individuals, representing 32 fish species from 23 families, and one crustacean species (Portunus armatus) were recorded on RUVs at the mussel reefs and control sites over the entire duration of this study (Table 4.1). Overall, the five most abundant species contributed ~95% to the total MaxN, namely the Western Gobbleguts Ostorhinchus rueppellii (34.7%), Atherinidae spp. (31.1%), Eight-lined Trumpeter Helotes (formerly Pelates) octolineatus (17.1%), Australian Anchovy Engraulis australis (7.2%) and Weeping Toadfish Torquigener pleurogramma (4.7%). Prior to the deployment of the mussel reefs (n = 32), 10 species were recorded with an average total MaxN of 12. After the deployment of the mussel reefs (n = 56) a larger number of species and total MaxN were recorded, i.e. control site = 17 and 51.1,
48 respectively and mussel reef = 31 and 65.1, respectively (Table 4.1). Only two species found on the control site were not detected on the mussel reefs (E. australis and the Mulloway Argyrosomus japonicus), whereas 16 species were unique to the mussel reef habitat including three recreational targeted species in King George Whiting Sillaginodes punctatus, Tarwhine Rhabdosargus sarba and Tailor Pomatomus saltatrix, albeit none were particularly abundant. The most notable difference in the post-deployment fauna between the two habitats was the much greater abundance of O. rueppellii at the mussel reefs than control sites (i.e. 30.6 vs 10.0; Table 4.1). Prior to the deployment of the mussel reefs, atherinids (Atherinidae spp.) were most abundant taxa (37.2%) followed by O. rueppellii and H. octolineatus (both 26.4%), which together constituted 90% of all individuals recorded (Table 4.1). After the deployment of the mussel reefs, the control sites were still dominated by the same three species (i.e. 40.1, 19.5 and 14.1%, respectively). Notable contributions were also made by E. australis (17.5%) and T. pleurogramma (5.4%). After deployment, although 31 species were detected on the mussel reefs, the fauna was dominated by same three species, i.e. O. rueppellii (46.9%), Atherinidae spp. (23.8%) and H. octolineatus (18.7%; Table 4.1). Based on the frequency of occurrence, the most observed species prior to reef deployment were O. rueppellii and Yellow Spotted Sand Goby Favonigobius punctatus, which were seen on 48 and 44% of videos, respectively (Table 4.2). Post-deployment, four species were observed in ≥ 25% of the videos on each of the two habitats, with O. rueppellii and T. pleurogramma being the most frequently occurring, albeit more so at the reef than control sites (Table 4.2). The same was true for the Bridled Goby, Arenigobius bifrenatus (39.7 vs 23.2%) and H. octolineatus (32.8 vs 23.2%). Among the species only recorded at the mussel reefs, the Western Smooth Boxfish Anoplocapros amydaloides and Western Australian Seahorse Hippocampus angustus were common occurring in 17.2 and 8.6% of the RUV videos, respectively.
49 Table 4.1. Mean MaxN (X), standard deviation (SD), percentage contribution to MaxN (%) and ranking (R) for each species recorded on RUVs at all sites before shellfish reef deployment (Pre; October-December 2018 and February 2019) and the control and reef sites after the deployment of the reefs (April 2019 – June 2020) in in the Swan-Canning Estuary. The Estuarine usage guild (EUG), habitat guild (HG) and feeding guild (FG) and overall abundance of each species are provided. Semi anadromous (SA), Estuarine & Freshwater (E&F), Solely Estuarine (E), Estuarine & Marine (E&F), Marine estuarine-opportunists (MEO) and Marine straggler (MS); Pelagic (P), Small pelagic (SP), Benthopelagic (BP), Demersal (D) and Small benthic (SB); Piscivore (PV), Zooplanktivore (ZP), Zoobenthivore (ZB), Omnivore (OV), Detritivore (DV) and opportunistic (OP). Taxon ranked by overall abundance, species shades in grey contributed > 5% to the total MaxN and those in bold targeted by recreational fishers. * denotes a crustacean species.
Pre-deployment Control site (post) Mussel reef (post) Species Family EUG HG FG Overall x X SD % R X SD %% R x X SD % R Ostorhinchus rueppellii Apogonidae E&M BP ZB 2413 3.1 5.7 26.4 2 10.0 19.8 19.5 2 30.6 67.8 46.9 1 Atherinidae spp. Atherinidae E SP ZB 2160 4.4 13.9 37.2 1 20.5 56.2 40.1 1 15.6 69.0 23.8 2 Helotes octolineatus Terapontidae MEO BP OV 1191 3.1 14.7 26.4 2 7.2 21.3 14.1 4 12.2 24.2 18.7 3 Engraulis australis Engraulidae E&M SP ZP 500 8.9 66.2 17.5 3 Torquigener pleurogramma Tetraodontidae MEO BP OP 329 0.2 0.5 1.4 5 2.8 4.1 5.4 5 2.9 4.9 4.5 4 Gerres subfasciatus Gerriedae MEO BP ZB 67 0.1 0.7 0.3 10 1.0 5.4 1.6 5 Arenigobius bifrenatus Gobiidae E&M SB ZB 44 0.1 0.3 0.7 7 0.3 0.6 0.6 6 0.4 0.6 0.7 6 Sillago burrus Sillaginidae MEO D ZB 38 0.1 0.3 0.7 7 0.2 0.7 0.5 8 0.4 1.8 0.6 7 Portunus armatus* Portunidae MEO SB ZB 30 0.3 0.1 0.7 7 0.3 0.5 0.5 8 0.2 0.5 0.3 9 Favonigobius punctatus Gobiidae E SB ZB 28 0.6 0.7 5.1 4 0.1 0.3 0.2 13 0.1 0.4 0.2 13 Myliobatis tenuicaudatus Myliobatidae MS D ZB 24 0.2 0.4 1.4 5 0.3 0.5 0.6 6 0.1 0.2 0.1 16 Nematalosa vlaminghi Clupeidae SA BP DV 23 0.1 1.1 0.3 10 0.3 2.0 0.4 8 Pentapodus vitta Nemipteridae MS D OV 21 0.1 0.6 0.3 10 0.2 0.8 0.3 9 Acanthopagrus butcheri Sparidae E BP OP 13 <0.1 0.1 <0.1 15 0.2 1.2 0.3 9 Favonigobius lateralis Gobiidae E&M SB ZB 10 0.1 0.3 0.2 13 0.1 0.3 0.1 16 Anoplocapros amygdaloides Aracanidae MS D ZB 10 0.2 0.4 0.3 9 Sillaginodes punctatus Sillaginidae MEO D ZB 9 0.2 0.6 0.2 13 Pseudogobius olorum Gobiidae E&F SB OV 6 <0.1 0.2 0.3 10 0.1 0.5 0.1 16 Hippocampus angustus Syngnathidae MS D ZP 6 0.1 0.4 0.2 13 Rhabdosargus sarba Sparidae MEO BP ZB 5 0.1 0.5 0.1 16 Monacanthus chinensis Monacanthidae MS D OV 4 0.1 0.3 0.1 16 Pomatomus saltatrix Pomatomidae MEO P PV 3 0.1 0.2 0.1 16 Acanthaluteres vittiger Monacanthidae MS D OV 2 <0.1 0.3 0.1 16 Eubalichthys mosaicus Monacanthidae MS D OV 2 <0.1 0.2 0.1 16 Amniabata caudavittata Terapontidae E BP OP 2 <0.1 0.2 0.1 16 Callogobius depressus Gobiidae MS SB ZB 1 <0.1 0.1 <0.1 25 Monacanthidae spp. Monacanthidae MS D OV 1 <0.1 0.1 <0.1 25 Argyrosomus japonicus Sciaenidae MEO BP PV 1 <0.1 0.1 <0.1 15 Cnidoglanis macrocephalus Plotosidae E&M D ZB 1 <0.1 0.1 <0.1 25 Platycephalus westraliae Platycephalidae E D PV 1 <0.1 0.1 <0.1 25 Pseudorhombus jenynsii Paralichthyidae MEO D ZB 1 <0.1 0.1 <0.1 15 <0.1 0.1 <0.1 25 Gymnapistes marmoratus Tetrarogidae MEO D ZB 1 <0.1 0.1 <0.1 25 Parablennius postoculomaculatus Bleniidae MS D ZB 1 <0.1 0.1 <0.1 25 Total number of species 32 10 17 31 Mean total MaxN 6946 12 51.1 65.1 Number of samples 144 32 56 56
50 Table 4.2. Frequency of occurrence (%O) of taxa and rankings by occurrence (R) for each species recorded prior to the deployment of mussel reefs (pre-deployment) as well as control and mussel reef sites afterwards. Species occurring in ≥ 25% of videos are shaded grey. Species ranked in accordance with Table 4.1.
Pre-deployment Control site Mussel reef Species %O R X %O R X %O R Ostorhinchus rueppellii 48.0 1 76.8 1 84.5 1 Torquigener pleurogramma 12.0 4 58.9 2 60.3 2 Helotes octolineatus 8.0 6 23.2 5 32.8 4 Myliobatis tenuicaudatus 16.0 3 30.4 3 5.2 16 Arenigobius bifrenatus 8.0 6 23.2 5 39.7 3 Favonigobius punctatus 44.0 2 10.7 9 10.3 7 Callogobius depressus 1.7 23 Sillago burrus 8.0 6 12.5 8 10.3 7 Sillaginoides punctata 6.9 14 Atherinidae spp. 12.0 4 17.9 7 8.6 10 Favonigobius lateralis 7.1 10 8.6 10 Pseudogobius olorum 4.0 10 3.4 19 Pentapodus vitta 5.4 11 10.3 7 Monacanthidae spp. 1.7 23 Acanthaluteres vittiger 1.7 23 Monacanthus chinensis 6.9 14 Anoplocapros amydaloides 17.2 6 Eubalichthys mosaicus 3.4 19 Gerres subfasciatus 5.4 11 8.6 10 Acanthopagrus butcheri 1.8 13 5.2 16 Rhabdosargus sarba 3.4 19 Nematalosa vlaminghi 1.8 13 1.7 23 Amniatoba caudavittata 3.4 19 Pomatomus saltatrix 5.2 16 Cnidoglanis macrocephalus 1.7 23 Platycephalus westraliae 1.7 23 Hippocampus angustus 8.6 10 Gymnapistes marmoratus 1.7 23 Parablennius postoculomaculatus 1.7 23 Portunas armatus 8.0 6 25.0 4 20.7 5 Argyrosomus japonicus 1.8 13 Engraulis australis 1.8 13 Pseudorhombus jenynsii 1.8 13 Number of samples 32 56 56
Several species were more frequently recorded on control sites than the reefs, including the Southern Eagle Ray Myliobatis tenuicaudatus and atherinids (30.4 vs 5.2% and 17.9 vs 8.6%, respectively; Table 4.2). The recreationally important Blue Swimmer
Crab Portunus armatus, Trumpeter Whiting Sillago burrus and F. punctatus were equally frequent in both habitats.
51 4.4.2.3. Pre-mussel reef deployment Two-way univariate PERMANOVA tests comparing the number of species, total MaxN and Simpson’s Index among the four times or two habitats did not detect a significant difference with either main effect or the Habitat × Time interaction (P = 0.145 to 0.987; Table 4.3; Appendix 5).
Table 4.3. Mean squares (MS) and percentage contribution of mean squares (%MS), pseudo-F ratios (pF) and significance level (P) derived from two-way PERMANOVA tests on the (a) number of species, (b) total MaxN and (c) Simpson’s Index of fish fauna of mussel reefs and control sites before (Pre) and after (Post) reef deployment. Significant results (P = < 0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
Pre deployment Post deployment (a) No. of species df MS %MS pF P df MS %MS pF P Time 3 1.86 28.98 1.03 0.389 13 25.85 54.96 11.22 0.001 Habitat 1 2.67 41.48 1.47 0.229 1 15.64 33.25 6.79 0.011 Time x Habitat 3 0.09 1.32 0.05 0.987 13 3.24 6.88 1.40 0.178 Residual 17 1.81 28.21 86 2.30 4.90 (b) Total MaxN Time 3 1.31 35.04 1.88 0.181 13 5.06 57.68 6.10 0.001 Habitat 1 1.50 40.23 2.16 0.159 1 2.30 26.20 2.77 0.118 Time × Habitat 3 0.23 6.09 0.33 0.826 13 0.58 6.66 0.70 0.740 Residual 17 0.69 18.64 86 0.83 9.46 (c) Simpson’s Index Time 3 0.22 50.40 2.03 0.145 13 0.29 52.01 5.37 0.001 Habitat 1 0.07 17.07 0.69 0.404 1 0.14 24.82 2.56 0.121 Time × Habitat 3 0.03 7.72 0.31 0.819 13 0.08 13.49 1.39 0.196 Residual 17 0.11 24.81 86 0.05 9.68
Faunal composition was shown not to differ significantly with Habitat, either as a main effect or in the Time × Habitat interaction term, although a difference over time was detected (Table 4.4). This is illustrated on the mMDS plots derived from bootstrap averages, where the 95% region estimates overlap substantially for Habitat, but the points for each month are predominantly discrete (Fig. 4.1). Monthly changes were driven by the shift in the abundance of several species. For example, the fauna in both habitats was fairly depauperate in October and heavily dominated by F. punctatus, the abundance of which decreased in subsequent months (Fig. 4.2). While in these months, similar abundances of O. rueppellii, F. punctatus and Myliobatis tenuicaudatus were recorded, the crustacean P. armatus was only recorded in November as was
52 H. octolineatus in December and Atherinidae spp. and Sillago burrus in February (Fig. 4.2).
Table 4.4. Mean squares (MS), percentage contribution of mean squares to the total mean squares (%MS), pseudo-F ratios (pF) and significant levels (P) derived from a two-way PERMANOVA test on the fish faunal composition at the mussel reef and control sites prior to reef deployment. Significant results (P = ≤ 0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
df MS %MS pF P Time 3 7701.9 49.54 2.34 0.013 Habitat 1 3202.4 20.6 0.97 0.436 Time × Habitat 3 1349 8.68 0.41 0.975 Residual 17 3293.4 21.18
2D Stress: 0.16 Habitat Reef Control 20 av:Reef
av:Control 2
S 0
D M
-20
-40 -60 -40 -20 0 20 40 MDS1 3D Stress: 0.06 Month Oct Nov Dec Feb av:Oct av:Nov av:Dec av:Feb
Fig. 4.1. Two and three dimensional mMDS plots constructed from bootstrapped averages for (a) habitat and (b) month, calculated from a Bray-Curtis resemblance matrix of transformed abundance data collected prior to the deployment of the mussel reef. Group averages (black squares) are provided, as are 95% region estimates fitted to the bootstrap averages in (a).
53
8
8
8
8
9
9 8
2 8 Habitat
1
1
1
1
1
1
1
1
v
v
c
c
t
t
b
b
c
c o
o Reef
e
e
e
e
O
O
N
N
D
D F F Control 1 Favonigobius punctatus
Ostorhinchus ruepellii
Myliobatis tenuicaudatus 0 Portunas armatus
Arenigobius bifrenatus
Helotes octolineatus
Torquigener pleurogramma
Atherinidae spp.
Sillago burrus
Pseudogobius olorum
Fig. 4.2. Shade plot constructed using transformed and averaged MaxN abundances of each species recorded at the mussel reef and control sites prior to the deployment of mussel reefs. Shading intensity is proportional to abundance.
4.4.2.4. Post-mussel reef deployment Guild composition In terms of the ways in which fish use estuaries, ~ 53 and 63 % of the species found on the control and mussel reef sites, respectively, belonged to the marine category (i.e. marine estuarine-opportunists and marine stragglers; Table 4.5a). The mussel reef harboured two more species from the former guild and eight from the latter, while the numbers of estuarine species (solely estuarine and estuarine & marine) were similar between the two habitats. When based on the number of individuals, there was a ~ 60 % increase in the number of marine estuarine-opportunists and 72 % increase in the number of marine stragglers at reef sites. Although, in the case of the latter guild, this only comprised 43 fish, thus exemplifying the definition of a straggler, i.e. occurring in low abundances. Despite representing 33% of species, marine stragglers only accounted for 1.1% of total MaxN at mussel reef sites (Table 4.5a). There was a ~ 75 % increase in the number of individuals of estuarine & marine species at reef than control
54 sites despite, both habitats containing four species. Control sites did, however, contain a larger number of individuals from solely estuarine species than the mussel reefs, which, due to the lower number of fish they harboured overall, markedly increase the contribution of this guild (i.e. 40.3 vs 24.4; Table 4.5a). Semi anadromous and estuarine & freshwater species made minimal contributions to the number of species or individuals at either site.
Based on feeding guilds, most species recorded at both sites were zoobenthivores; representing 10 of the 17 species at the control site (59%) and 16 of the 31 on the mussel reefs (53%; Table 4.5b). Omnivorous and opportunist species made the next largest contribution, and particularly the former guild at the mussel reefs with seven species, comprising 23% of the total. Combined, zoobenthivores and omnivores accounted for 94.6% of the total number of individuals at reef habitats and 76.8% at control sites and thus dominated the fish faunal community. Despite being relatively specious in both habitats, only ~5% of the individuals were opportunists. Most of the species recorded belonged to the benthopelagic, demersal and small benthic habitat guilds (Table 4.3b). Such species constituted 15 of the 17 (88%) species at the control site and 29 of the 31 species (94%) at the mussel reefs. Among these benthopelagic species were the most speciose 7 (41%) at the control site as were demersal species on the reefs (14; 47%). Differences between the two habitats were largely due to the far greater numbers of demersal species on the mussel reefs (14 vs 4). Although the small pelagic guild was represented by two species on the control site, the schooling nature of these fish, resulted in their individuals comprising 58% of all fish recorded. Benthopelagic species were also abundant (40% of all individuals). The single small pelagic species also made a disproportionally high contribution to the number of individuals on the mussel reefs (24%), but the fauna was dominated by benthopelagic fish (73%; Table 4.5c). Finally, although speciose, demersal fish comprised only 2% of the total number of fish recorded.
55 Table 4.5. Number of species (Sp) and individuals (N) of each of the (a) estuarine usage, (b) feeding and (c) habitat guilds, together with their percentage contribution (%), in RUV samples collected from the mussel reef and control sites in the Swan-Canning Estuary between April 2019 and June 2020. Codes for guilds given in Table 4.1.
Control site Mussel reef Sp %Sp N %N Sp %Sp N %N (a) Estuarine usage Semi anadromous 1 5.9 8 0.3 1 3.3 15 0.4
Estuarine & freshwater 1 3.3 5 0.1 Solely estuarine 3 17.6 1155 40.3 5 16.7 924 24.4 Estuarine & marine 4 23.5 1080 37.7 4 13.3 1,808 47.7 Marine estuarine opportunist 7 41.2 597 20.8 9 30.0 992 26.2 Marine straggler 2 11.8 25 0.9 10 33.3 43 1.1 (b) Feeding Piscivore 1 5.9 1 0 2 6.7 4 0.1 Zooplanktivore 1 5.9 500 17.5 1 3.3 6 0.2 Zoobenthivore 10 58.8 1,788 62.4 16 53.3 2,842 75 Omnivore 2 11.8 412 14.4 7 23.3 736 19.4 Detritivore 1 5.9 8 0.3 1 3.3 15 0.4 Opportunistic 2 11.8 156 5.4 3 10 184 4.9 (c) Habitat
Pelagic 1 3.3 3 0.1 Small pelagic 2 11.8 1,648 57.5 1 3.3 902 23.8 Benthopelagic 7 41.2 1,136 39.7 8 26.7 2,748 72.6 Demersal 4 23.5 39 1.4 14 46.7 77 2.0 Small benthic 4 23.5 42 1.5 6 20 57 1.5
4.4.2.5 Univariate diversity and abundance indices The number of species was shown by two-way PERMANOVA to differ significantly with Time and Habitat (P = 0.001 and 0.011, respectively), but their interaction was not significant (Table 4.3a). Based on the percentage mean squares, the time main effect was more influential than Habitat (54.96 vs 33.25 %, respectively). The mean number of species underwent a sinusoidal trend, declining from ~ 4 in April 2019 to a minimum of 0.5 in June 2019 (Fig. 4.3a). After which values increased essentially sequentially to ~ 6 between January and April 2020, before declining to ~ 2 in May and June of that year. Mussel reef habitats harboured a greater mean number of species (3.8) than control sites (3; Fig. 4.3b). Total MaxN differed significantly only with Time (P = 0.001; Table 4.3b), with values between January and April 2020 (~ 110 to 175) being larger than those in most other months and particularly between May and July 2019 (< 15) and June 2020 (9; Fig.
56 4.3c). Values for Simpson’s Index also differed only with Time (P = 0.001; Table 4.3c). The temporal trend was like that of the number of species; decreasing from April and May 2019, to a minimum in July and August 2019, before increasing through to June 2020 (Fig. 4.3d).
(a) (b)
10 4.5 s
s 8 4.0
e e
i i
c c
e e
p p s
s 6 3.5
f f
o
o
r r e
e 4 3.0
b
b
m
m
u
u N 2 N 2.5
0 2.0
l
f
9
9
9
9
9
9
9
9
0
0
0
0
0
0
o
e
1
1
1
1
1
1
1
1
2
2
2
2
2
2
r
e
t
l
r
r
r
c
y
v
y
g
n
p
n
b
n
R
n
u
a
p
p
a
a
e
o
u
a
e
e
u
u
J
o
J
J
J
A
F
A
S
D
A
N
M
M
C M Month Habitat (c) (d) 400 1.0
300 x
e
d
n
I
N
s
x n
a 200 0.5
o
s
M
p
m i
100 S
0 0
9 9
9 9
9 9
9 9
9 9
9 9
9 9
9 9
0 0
0 0
0 0
0 0
0 0
0 0
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
2 2
2 2
2 2
2 2
2 2
2 2
l l
r r
r r
r r
c c
y y
v v
y y
g g
n n
p p
n n
b b
n n
u u
a a
p p
p p
a a
a a
e e
o o
u u
a a
e e
e e
u u
u u
J J
J J
J J
J J
A A
F F
A A
S S
D D
A A
N N
M M
M M
M M Month Month Fig. 4.3. Mean number of species among (a) months and (b) habitats and (c) total MaxN and (d) Simpson’s Index among months. Error bars represent ±95% confidence limits.
Fish faunal composition
Following the deployment of the mussel reefs in March 2020, fish faunal composition differed significantly with Time and between Habitats, but the Time × Habitat interaction was not significant (Table 4.6). Both main effects contributed similar amounts to the proportion of mean squares (Table 4.6). On the mMDS plot derived from bootstrap averages for habitat, the 95% region estimates remain discrete, with the points for the mussel reef lying on the left of the plot clearly separated from those of the control site (Fig. 4.4). The changes in faunal composition are likely attributed to the wider range of species found at the reef sites (Fig. 4.5). Examples include,
57 A. amydaloides, H. angustus and a range of leatherjackets, i.e. Monacanthus chinensis, Acanthaluteres vittiger, Eubalichthys mosaicus and Monacanthidae spp.. Although O. rueppellii, A. bifrenatus and H. octolineatus were found on both habitats, their abundances were all greater on the mussel reef sites. Several species such as M. tenuicaudatus, Atherinidae spp. and E. australis were more abundant at the control site (Fig. 4.5).
Post-hoc analysis using one-way ANOSIM determined that 54 of the 91 (59%) pairwise comparisons between months were significant. Among these, all except four involved months between May and September 2019 and May and June 2020 (Table 4.7). When the fish faunal community data were visualised on a centroid nMDS plot, the order of points formed a spiral (Fig. 4.6). There was a clockwise serial progression in samples between April 2019 to April 2020 before heading upwards and then to the left in May and June 2020. This pattern resulted in a broad group of seven months located in the bottom right of the plot, containing samples from April, November, December, January, February and March (i.e. ‘summer’ months) and the widely spaced samples from other months that were significantly different (cf. Table4.7; Fig. 4.6). These latter samples tended to be those from the ‘winter’ months. A core group of three species, i.e. O. rueppellii, A. bifrenatus and T. pleurogramma were found in all months and usually in relatively large abundances
(Fig. 4.7). These species comprised the majority of the depauperate fauna observed between May and September 2019. In November and December, additional species were recorded such as M. tenuicaudatus, F. punctatus, A. amydaloides, P. armatus and H. octolineatus. The last two species persisted for the next four months (January to April) together with P. saltatrix, A. butcheri, Sillago burrus and Gerres subfasciatus. These species were not recorded in May and June 2020 and were replaced with M. chinensis,
H. angustus and S. punctata.
58 Table 4.6. Mean squares (MS), percentage contribution of mean squares to the total mean squares (%MS), pseudo-F ratios (pF) and significant levels (P) derived from a two-way PERMANOVA test on the fish faunal composition at the mussel reef and control sites post reef deployment. Significant results (P = ≤ 0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
df MS %MS pF P Time 13 8302.8 37.08 3.46 0.001 Habitat 1 9026.4 40.32 3.76 0.001 Time × Habitat 13 2660.7 11.88 1.11 0.242 Residual 86 2398.7 10.71
2D Stress: 0.17 Habitat Reef 10 Control av:Reef
av:Control
2 S
D 0 M
-10
-20 -10 0 10 20 MDS1 Fig. 4.4. Two dimensional mMDS plot constructed from bootstrapped averages for habitat, calculated from a Bray-Curtis resemblance matrix of transformed abundance data collected after deployment of the mussel reef. Group averages (black squares) and 95% region estimates fitted to the bootstrap averages are provided.
0.7 l
o
r t
0.6 f
n
e e
0.5 o
R C Argyrosomus japonicus 0.4 Engraulis australis Pseudorhombus jenynsii 0.3 Torquigener pleurogramma Ostorhinchus ruepellii 0.2 Arenigobius bifrenatus Helotes octolineatus 0.1 Portunas armatus Myliobatis tenuicaudatus 0 Atherinidae spp. Sillago burrus Favonigobius punctatus Favonigobius lateralis Pentapodus vitta Gerres subfasciatus Anoplocapros amydaloides Hippocampus angustus Sillaginoides punctata Monacanthus chinensis Acanthopagrus butcheri Pomatomus saltatrix Nematalosa vlaminghi Eubalichthys mosaicus Amniatoba caudavittata Rhabdosargus sarba Pseudogobius olorum Acanthaluteres vittiger Cnidoglanis macrocephalus Gymnapistes marmoratus Parablennius postoculomaculatus Platycephalus australis Monacanthidae spp. Callogobius depressus
Fig. 4.5. Shade plot constructed using transformed and averaged MaxN abundances of each species recorded at the mussel reef and control sites after the deployment of mussel reefs. Shading intensity is proportional to abundance.
59 Table 4.7. R statistic and significance level (P) values derived from a one-way ANOSIM test on the composition of the fish fauna recorded on mussel reef and control sites combined among months. R values are shaded grey if the pairwise compassion were significant (P = ≤ 0.05).
2019 2020 Month Apr May Jun Jul Aug Sep Nov Dec Jan Feb Mar Apr May May 0.046 Jun 0.333 -0.135 Jul 0.759 0.497 0.359 Aug 0.130 0.188 0.193 0.112 Sep 0.442 0.385 0.288 0.023 0.035 Nov -0.137 0.114 0.274 0.547 0.253 0.495 Dec 0.111 0.354 0.568 0.714 0.198 0.558 -0.150 Jan -0.065 0.245 0.536 0.714 0.167 0.681 0.024 0.052 Feb 0.019 0.208 0.453 0.786 0.417 0.692 0.189 0.328 -0.109 Mar 0.169 0.354 0.562 0.839 0.455 0.521 0.140 0.306 -0.044 -0.059 Apr 0.259 0.448 0.552 0.875 0.573 0.768 0.083 0.339 0.120 -0.021 0.007 May 0.269 0.107 0.328 0.729 0.354 0.771 0.227 0.495 0.474 0.432 0.686 0.615 Jun 0.324 0.076 0.188 0.797 0.464 0.794 0.196 0.625 0.484 0.438 0.643 0.563 -0.016
2D Stress: 0.07
May 20 Jun 20
May 19 Jun 19 Jul 19
Sep 19 Apr 19 Apr 20 Feb 20 Nov 19
Mar 20 Jan 20
Aug 19
Dec 19
Fig. 4.6. Two-dimensional centroid nMDS plot derived from a distance-among-centroids matrix constructed using the abundance data in each sample collected after deployment of the mussel reef.
60 9
2 0
0
9
9
9
9
9
0
0
9
0
0
1
2
2
1
9
1
1
1
1
2
2
1
2
2
r
y
1
y
v
r
g
r
c
p
b
a
n
l
a
a
n
n
o
e
p
u
p
e
e
u
u
a
u
A
M
J
J
A
S
N
D
J
F
M
A
M J Acanthaluteres vittiger Gymnapistes marmoratus 1 Parablennius postoculomaculatus Hippocampus angustus Sillaginoides punctata Eubalichthys mosaicus Monacanthus chinensis Anoplocapros amydaloides Favonigobius punctatus 0 Myliobatis tenuicaudatus Torquigener pleurogramma Ostorhinchus ruepellii Arenigobius bifrenatus Portunas armatus Helotes octolineatus Atherinidae spp. Sillago burrus Pentapodus vitta Favonigobius lateralis Gerres subfasciatus Amniatoba caudavittata Nematalosa vlaminghi Rhabdosargus sarba Engraulis australis Callogobius depressus Cnidoglanis macrocephalus Acanthopagrus butcheri Pomatomus saltatrix Monacanthidae spp. Pseudogobius olorum Pseudorhombus jenynsii Platycephalus australis Argyrosomus japonicus
Fig. 4.7. Shade plot constructed using transformed and averaged MaxN abundances of each species recorded at the mussel reef and control sites combined in each month after the deployment of mussel reefs. Shading intensity is proportional to abundance.
4.4.3. Invertebrate communities 4.4.3.1 Large invertebrates: faunal description A total of 16,363 invertebrates > 2,000 µm in size from 16 taxa and five classes were recorded from mussel reefs over four sampling occasions between May and December 2019 (Table 4.8). The Ascidiacea and Bivalvia were the most speciose classes with six and five species, respectively, and together contributed 68.8 % of all species and
41.1 % of total number of invertebrates. The Hexanauplia, which comprised only Amphibalanus variegatus, represented 54.1 % of the abundance. The three representatives of the Malacrustaca, were the only mobile species and contributed < 1 % to the total number of invertebrates. At the species level, five contributed to ~ 97 % of all individuals, namely the barnacle A. variegatus (56.33 %), the ascidians Ascidiella aspersa (18.67 %), Ciona intestinalis (7.13 %) and Mogulidae (3.1 5%) and the bivalve
Arcuatula senhousia (11.5 %). The remaining 11 species made very minor contributions.
61 Table 4.8. Total number of individuals (Total), mean (X 300 cm-2), standard deviation (SD) and percentage contribution (%C) of each large invertebrate species (i.e. those retained on a 2,000 µm sieve). Samples obtained from three mussel reefs samples over four months (May, July, September and December 2019). Species ranked in order of their total abundance and those contributing ≥ 5% in abundance are shaded in grey. The class to which each species belongs is also provided.
Species Class Total X SD %C Amphibalanus variegatus Hexanauplia 9,218 256.06 241.03 56.33 Ascidiella aspersa Ascidiacea 3,055 84.86 51.55 18.67 Arcuatula senhousia Bivalvia 1,882 52.28 183.73 11.50 Ciona intestinalis Ascidiacea 1,166 32.39 54.04 7.13 Mogulidae sp. Ascidiacea 516 14.33 45.77 3.15 Fluviolanatus subtortus Bivalvia 183 5.08 12.26 1.12 Sphaeromatidae sp. Malacostraca 130 3.61 21.36 0.79 Styela plicata Ascidiacea 129 3.58 6.41 0.79 Xenostrobus securis Bivalvia 61 1.69 4.52 0.37 Ostreidae sp. Bivalvia 8 0.22 0.48 0.05 Porifera sp. Porifera 4 0.11 0.39 0.02 Didemnum perlucidum Ascidiacea 3 0.08 0.36 0.02 Botrylloides sp. Ascidiacea 3 0.08 0.28 0.02 Scaeochlamys livida Bivalvia 2 0.06 0.23 0.01 Pilumnidae sp. Malacostraca 2 0.06 0.23 0.01 Naxia sp. Malacostraca 1 0.03 0.16 0.01
4.4.3.2. Large invertebrates: diversity and abundance The mean number of species differed significantly with Month both as a main effect and in the Month × Site interaction (Table 4.9a), with the former term being more influential (%MS = 48 %). There was no significant difference with Site (P = 0.082). The mean number of species increased sequentially over the four months from ~ 3.8 in May to ~ 6 in December (Fig. 4.8a). This increase over time was not consistent at all sites
(e.g. Reef 3 in December), thereby explaining the interaction (Fig. 4.8b). The total number of individuals differed with Month and Site, with the Month effect explaining over three times that of Site (69 vs 22 %, respectively; Table 4.9b). The faunal community was most numerous in May 2019 (709 individuals 300 cm-2), but declined to a minimum of 164 in September, before increasing to 494 in December (Fig. 4.8c). Among the three reefs, densities were greater at Reef 2 than either Reef 1 or Reef 3, i.e. 619 vs 399 and 344 individuals 300 cm-2, respectively (Fig. 4.8d).
As with the number of species, values of Simpson’s Index differed significantly with Month and Month × Site the interaction, with 80% of the variability explained by
62 the main effect (Table 4.9c). Simpson’s Index values was greatest in December (0.59) and least in May (0.26), with intermediate values recorded in July and September (Fig. 4.8e). There was some inter-reef variability within a month, which likely caused the significant interaction (Fig. 4.8f).
Table 4.9. Mean squares (MS) and percentage contribution of mean squares (%MS), pseudo-F ratios (pF) and significance level (P) derived from two-way PERMANOVA tests on the (a) number of species (b) number of individuals and (c) Simpson’s Index of the large invertebrate fauna of the mussel reefs. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
(a) Number of species df MS %MS pF P Month 3 8.62 48.82 6.21 0.003 Site 2 3.69 20.92 2.66 0.082 Month × Site 6 3.95 22.39 2.85 0.024 Residual 24 1.39 7.87
(b) Number of individuals Month 3 3.43 68.97 25.11 0.001 Site 2 1.09 21.95 7.99 0.003 Month × Site 6 0.32 6.33 2.31 0.088 Residual 24 0.14 2.75
(c) Simpson’s Index Month 3 0.18 81.45 19.55 0.001 Site 2 0.01 2.66 0.64 0.531 Month × Site 6 0.03 11.72 2.81 0.032 Residual 24 0.01 4.17
63 (a) (b)
7 15
s
6 e
i
s
c
e i
e 10
c
p
e
s
p
5 f
s
o
f
r
o
e
r b
e 4 5
m
b
u
m
N u
N 3
0
l
l
l
c
c
c
y
y
y
p
p
p
u
u
u
a
a
a
e
e
e
e
e
e
J
J
J
S
S
S
D
D
D
1
2
3
M
M
M
1
2
3
1
2
3
f f
2 f
1
2
3
f
f
f
f
f
f
e
e
e
l
f
f
f
c
y
e
e
e
p
e
e
e
e
e
e
u
e
e
e
a
e
e
e
e
e
e
e
e
J
R
R
R
e
e
e
S
R
R
R
R
R
R
D
M
R
R R Month Site x Month
(c) (d)
1000 1000
s
l
s l
a 800 a
u 800
u
d
i
d
v
i
i
v i
d 600
d
n
i
600 n
f
i
o
f
r
o
e
r 400
b
e b
m 400
u
m N
u 200 N
200
1
2
3
f f
f 0
e
e
e
l
c
y
e
e
e
p
u
a
e
R
R
R
e
J
S D Site M Month
(e) (f) 0.8 1.5
0.7
x e
x 1.0 d
e 0.6
n
I
d
n
n
I
o n
0.5 s
o
p
s m
p 0.5
i
S m
i 0.4 S
0.3
0
l l l
c c c
y y y
p p p
u u u
a a a
e e e
e e e
J J J
S S S
D D D
1 2 3
M M M
1 2 3
1 2 3
f f
0.2 f
1 2 3
f f f
f f f
e e e
l
f f f
c
y
p
e e e
e e e
e e e
u
e e e
a
e
e
e e e
J
e e e
R R R
e e e
S
R R R
D
R R R
M
R R R Month Site x Month Fig. 4.8. Mean number of species among (a) months and (b) site and month combinations, number of individuals among (c) sites and (d) months and Simpson’s Index among (e) months (f) site and month combinations. Error bars represent ±95% confidence limits.
4.4.3.3. Large invertebrates: faunal composition Composition of large invertebrates associated with the three mussel reefs was shown by PERMANOVA to differ significantly with Month, Site, and the Month × Site interaction was also significant (all P = 0.001; Table 10). Month explained the greatest proportion of the variation (50.6 %), followed by Site (29.3 %) and lastly the one-way interaction (16.5 %).
64 Post-hoc testing using one-way ANOSIM showed that across all reefs (i.e. the Month main effect) there a significant in composition between each pair of months (Table 4.11f). These differences are emphasised by the clear separation between the 95% bootstrapped regions for each month (Fig. 9a). Aside from A. variegatus, A. aspersa and to a lesser extent, C. intestinalis, which were abundant in each month, there was a serial gain or loss of species (Fig. 9b). Species such as Scaeochlamys livida were only recorded in May, and others like A. variegatus, A. aspersa were most abundant in this season. By July, Ostreidae sp. and S. livida had disappeared and been replaced by Styela plicata and Fluviolanatus subtortus. While most of these species remained in October their densities increased in December together with the addition of species such as Sphaeromatidae sp. and Botrylloides sp. (Fig. 9b). Across all seasons (i.e. the Reef main effect) there was no difference in composition between Reef 1 and Reef 2, but that both reefs harboured a different large invertebrate fauna from Reef 3 (Table 4.11a). This is illustrated by the mMDS plot where the 95% bootstrapped regions for reefs 1 and 2 overlap substantially, but that of Reef 3 is discrete (Fig. 4.10a). While A. variegatus and A. aspersa dominated the fauna at all reefs, species such as Ostreidae sp., Sphaeromatidae sp. and Porifera sp. were only recorded at Reef 3 (Fig. 4.10b). The lack of a difference between reefs 1 and 2 was also due to similar abundances of C. intestinalis and the shared presence species such as of
A. senhousia and Mogulidae sp., albeit in difference densities. The significance of the Month × Site interaction effect indicates that the monthly trends in faunal composition are not consistent amongst sites and vice versa. This is shown on the centroid nMDS plot where the points for May collapse indicating their similarity, before dispersion with the trajectory of reefs 1 and 2 following a similar pattern and different to that of Reef 3 (Fig. 10a). The consistency of the fauna in May is due to the depauperate nature of the community, predominantly comprising A. variegatus, A. aspersa and C. intestinalis. Whereas, as the months progress, species
65 like F. subtortus and A. senhousia and Mogulidae sp. are present in increasing densities at reefs 1 and 2 and the same is true for S. plicata, Sphaeromatidae sp. and Botrylloides sp. at Reef 3 (Fig. 10b).
Table 4.10. Mean squares (MS), percentage contribution of mean squares to the total mean squares (%MS), pseudo-F ratios (pF) and significant levels (P) derived from a two-way PERMANOVA tests on the composition of the large invertebrate fauna of the mussel reefs. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
df MS %MS pF P Month 3 4004 50.62 13.9 0.001 Site 2 2314.6 29.26 8.03 0.001 Month × Site 6 1302.7 16.47 4.52 0.001 Residual 24 288.17 3.64
Table 4.11. R statistic and/or significance level (P) values derived from one-way ANOSIM tests on the composition of the large invertebrate fauna of the mussel reefs (a) overall and in each month (b-e) and among (f) month and (g-i) at each site. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
(a) P = 0.010; Global R = 0.106 (f) P = 0.001; Global R = 0.437 Reef 1 Reef 2 May Jul Sep Reef 2 0.007 Jul 0.450 Reef 3 0.220 0.113 Sep 0.513 0.167 Dec 0.694 0.539 0.308 (b) P = 0.886; Global R = -0.226 Reef 1 Reef 2 (g) P = 0.001; Global R = 0.914 Reef 2 n/a May Jul Sep Reef 3 n/a n/a Jul 0.704 Sep 1.000 0.963 (c) P = 0.004; Global R = 0.588 Dec 1.000 1.000 1.000 Reef 1 Reef 2 Reef 2 0.333 (h) P = 0.001; Global R = 0.867 Reef 3 0.889 0.667 May Jul Sep Jul 0.556 (d) P = 0.021; Global R = 0.704 Sep 0.630 0.852 Reef 1 Reef 2 Dec 1.000 1.000 1.000 Reef 2 1.000 Reef 3 0.926 -0.037 (i) P = 0.005; Global R = 0.537 May Jul Sep (e) P = 0.004; Global R = 0.613 Jul 0.704 Reef 1 Reef 2 Sep 0.519 0.407 Reef 2 0.519 Dec 0.667 0.667 0.333 Reef 3 0.815 0.852
66 (a) 2D Stress: 0.09 Month May 20 Jul Sep Dec av:May av:Jul av:Sep
2 av:Dec S
D 0 M
-20
-40 -20 0 20 40
MDS1
y
c
p a 5 l
(b) e
e
u
M
J
S D 4 Scaeochlamys livida Didemnum perlucidum 3 Pilumnidae sp 2 Ostreidae sp 1 Xenostrobis securis
Ciona intestinalis 0 Amphibalanus variegatus
Ascidiella aspersa
Arcuatula senhousia
Styela plicata Mulgulidae sp
Fluviolanatus subtortus
Sphaeromatidae sp
Botrylloides sp
Porifera spp Naxia sp
Fig. 4.9. (a) Two dimensional mMDS plot constructed from bootstrapped averages for month, calculated from a Bray-Curtis resemblance matrix of transformed abundances of large invertebrates. Group averages (black squares) and 95% region estimates fitted to the bootstrap averages are provided. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded in each month. Shading intensity is proportional to abundance.
67 (a) 2D Stress: 0.11 Site Reef 1 Reef 2 Reef 3 av:Reef 1 av:Reef 2 av:Reef 3
0
2
S
D M
-20
-20 0 20
MDS1
1
2
3
4
f f
(b) f
e
e
e
e
e
e
R
R R 3 Pilumnidae sp Amphibalanus variegatus 2 Ascidiella aspersa Ciona intestinalis 1 Arcuatula senhousia
0 Mulgulidae sp Fluviolanatus subtortus Styela plicata Xenostrobis securis Ostreidae sp Botrylloides sp Sphaeromatidae sp Porifera spp Didemnum perlucidum Scaeochlamys livida Naxia sp
Fig. 4.10. (a) Two dimensional mMDS plot constructed from bootstrapped averages for site, calculated from a Bray-Curtis resemblance matrix of transformed abundances of large invertebrates. Group averages (black squares) and 95% region estimates fitted to the bootstrap averages are provided. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded at each site. Shading intensity is proportional to abundance.
68 (a) 2D Stress: 0.07 SiteMonth Reef 1May Reef 2 Reef 1Jul Reef 1Sep Reef 1Dec Reef 1 Reef 1 Reef 2May Reef 2Jul Reef 2Sep Reef 2 Reef 2Dec Reef 3May Reef 1 Reef 3 Reef 3Jul Reef 2 Reef 3Sep Reef 3Dec
Reef 1 Reef 23
Reef 3 Reef 3
1 2 3 1 2 3 1 2 3 1 2 3
(b) 6 Month
f f f f f f f f f f f f
e e e e e e e e e e e e
e e e e e e e e e e e
e May
R R R R R R R R R R R 5 R Jul Sep 4 Pilumnidae sp Dec 3 Naxia sp
2 Sphaeromatidae sp Botrylloides sp 1 Porifera spp 0 Styela plicata
Amphibalanus variegatus
Ascidiella aspersa
Fluviolanatus subtortus
Arcuatula senhousia
Mulgulidae sp
Ciona intestinalis
Xenostrobis securis
Ostreidae sp
Didemnum perlucidum Scaeochlamys livida
Fig. 4.11. (a) Two-dimensional centroid nMDS plot derived from a distance-among-centroids matrix constructed using the abundance data in each sample collected after deployment of the mussel reef. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded in month and site combination. Shading intensity is proportional to abundance.
4.4.3.4 Small invertebrates: faunal description A total of 19,908 small invertebrates were collected from three reefs over the four months, comprising 33 species from eight classes (Table 4.12). The most specious classes were the Polychaeta (14) and Malacostraca (11), which together included ~ 76 %
69 of all species. Although containing the largest number of species, polychaetes only accounted for 1 % of the total number of individuals. Instead, individuals of the Malacostraca dominated 88.2 %, followed by those of the bivalves 6.5 %. Eight species each contributed ≥ 5 % of the total abundance, and together comprised 82.4 % of all invertebrates, all belonging to the Malacostraca except the bivalve A. senhousia. Among the top ranked species, the isopod Janiridae sp. (24.31 %) and amphipods Corophium minor (16.49 %) and Ericthonius sp.1 (12.39 %) were particularly abundant and together accounted for ~ 53 % of the total abundance (Table 4. 12).
70 Table 4.12. Total number of individuals (Total), mean (X 300 cm-2), standard deviation (SD) and percentage contribution (%C) of each small invertebrate species (i.e. those retained on a 500 µm sieve). Samples obtained from three mussel reefs samples over four months (May, July, September, and December 2019). Species ranked in order of their total abundance and those contributing ≥ 5% in abundance are shaded in grey. The class to which each species belongs is also provided.
Species Class Overall X SD %C Janiridae sp. Malacostraca 4,839 134.42 504.63 24.31 Corophium minor Malacostraca 3,282 91.17 337.02 16.49 Ericthonius sp. Malacostraca 2,467 68.53 139.72 12.39 Arcuatula senhousia Bivalvia 1,231 34.19 105.12 6.18 Amphipoda Malacostraca 1,229 34.14 85.45 6.17 Melita sp. Malacostraca 1,189 33.03 64.95 5.97 Paracorophium excavatum Malacostraca 1,139 31.64 80.84 5.72 Barnardomelita matilda Malacostraca 1,033 28.69 38.15 5.19 Tanaidacea spp. Malacostraca 857 23.81 72.16 4.30 Sphaeromatidae sp. Malacostraca 762 21.17 40.70 3.83 Caprella spp Malacostraca 727 20.77 34.05 3.65 Ostracoda sp. Ostracoda 504 14.00 67.23 2.53 Copepoda spp. Hexanauplia 242 6.72 20.85 1.22 Ophiuroidea spp. Ophiuroidea 75 2.08 12.33 0.38 Sabellidae sp.1 Polychaeta 61 1.69 3.75 0.31 Fluviolanatus subtortus Bivalvia 56 1.56 5.27 0.28 Amphipoda Malacostraca 35 0.97 4.59 0.18 Capitella capitata Polychaeta 23 0.64 1.53 0.12 Sabellidae sp.2 Polychaeta 23 0.64 3.31 0.12 Cirriformia spp. Polychaeta 21 0.58 1.28 0.11 Polynoidea spp. Polychaeta 19 0.53 1.04 0.10 Polychaete sp.1 Polychaeta 17 0.47 1.52 0.09 Simplisetia spp. Polychaeta 17 0.47 1.14 0.09 Lumbrinereidae spp. Polychaeta 12 0.33 1.03 0.06 Prionospio cirrifera Polychaeta 11 0.31 0.78 0.06 Syllidae sp.2 Polychaeta 10 0.28 1.17 0.05 Dorvilleidae spp. Polychaeta 9 0.25 0.79 0.05 Mytilus galloprovincialis Bivalvia 8 0.22 1.00 0.04 Syllidae sp.1 Polychaeta 4 0.11 0.39 0.02 Oenonidae spp. Polychaeta 3 0.08 0.49 0.02 Polychaete sp.2 Polychaeta 1 0.03 0.16 0.01 Cnidaria spp. Cnidaria 1 0.03 0.16 0.01 Platyhelminthes spp. Platyhelminthes 1 0.03 0.16 0.01
4.4.3.5. Small invertebrates: diversity and abundance Two-way PERMANOVA detected a significant difference in the number of species with Month, but not Site or the one-way interaction (Table 4.13a). The mean number of species was relatively consistent in May, July, and September (7.2 to 7.9), however, increased markedly during December (11.7; Fig. 4.12a). The number of individuals was found to differ with Month and Site, with the former main effect contributing
71 significantly more to the proportion of mean squares (i.e. 67.3 vs 29.8 %; Table 4.13b). Values were relatively similar during May and July (i.e. 182 and 244 individuals 300 cm-2) before declining in September (45) and increasing dramatically in December (1,740; Fig. 4.12b). Differences between sites were less conspicuous, with Reef 1 having a lower mean number of individuals (167 individuals 300 cm-2), than either Reef 2 or Reef 3 (i.e. 697 and 795 respectively; Fig. 4.12c). Values of Simpson’s Index were found not to differ significantly with any term in the PERMANOVA model (Table 4.13c).
Table 4.13. Mean squares (MS) and percentage contribution of mean squares (%MS), pseudo-F ratios (pF) and significance level (P) derived from two-way PERMANOVA tests on the (a) number of species (b) number of individuals and (c) Simpson’s Index of the small invertebrate fauna of the mussel reefs. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
(a) Number of species
df MS %MS pF P Month 3 38 65.5 11.3 0.001 Site 2 10.5 18.1 3.13 0.059 Month × Site 6 6.12 10.5 1.82 0.134 Residual 24 3.36 5.79
(b) Number of individuals Month 3 20.5 67.3 47.5 0.001 Site 2 9.1 29.8 21 0.001 Month × Site 6 0.45 1.47 1.04 0.466 Residual 24 0.43 1.42
(c) Simpson’s Index Month 3 0.01 15.9 0.8 0.462 Site 2 0.05 53.6 2.67 0.075 Month × Site 6 0.01 10.5 0.52 0.792 Residual 24 0.02 20
72 (a) (b)
14 3000
s 2500
12 l
a
s
e
u
i
d c
i 2000
e
v i
p 10
d
s
n
f
i
o 1500
f
r
o
e
8 r
b e
b 1000
m
u
m N 6 u N 500
4 0
c c
y
y
y
y
p
p
l
l
a
a
e e
e
e
u
u
S
S
J
J
D D
M M Month Month (c)
2000
s
l a
u 1500
d
i
v
i
d
n i
1000
f
o
r
e b
m 500
u N
0
1
2
3
f
f
f
e
e
e
e
e
e
R
R R Site
Fig. 4.12. Mean number of species among (a) months and number of individuals among (b) months and (c) sites. Error bars represent ±95% confidence limits.
4.4.3.6 Small invertebrates: faunal composition The composition of the small invertebrate fauna differed significantly between Month and Site and the Month × Site interaction was also significant (Table 4.14). Month constituted the highest proportion of the mean squares, accounting for twice as much as Site and three times that of the one-way interaction.
Pairwise comparisons using ANOSIM, determined that the faunal composition in each of the four months were each unique (Table 4.15f), as evidenced by their distinct separation on the bootstrapped mMDS plot (Fig. 13a). Differences between May and June were largely due to differences in the abundances of a few crustacean species, i.e. C. minor, Paracorophium excavatum, Caprella spp. and Ericthonius sp.1, all of which declined precipitously in abundance in September, before increasing in December
(Fig. 4.13b). Other species whose abundance increase markedly in this month include Sphaeromatidae sp. Janiridae sp., A. senhousia, Melita sp1, Amphipoda spp. and
73 Ostracoda spp. Each mussel reef was also found to harbour a significantly different small invertebrate fauna (Table 4.15a), with the point for each and their associated 95% bootstrapped regions on the mMDS plot all being discrete (Fig. 4.14a). In the case of Reef 2, this site contained larger densities of C. minor, Caprella spp. and A. senhousia, whereas Reef 3 harboured relatively high abundances of Melita sp.1, Janiridae sp. and Tanaidacea spp. (Fig. 4.14b). In contrast the lowest densities of most species except
B. matilda, were found at Reef 1.
The significant but relatively uninfluential Month × Site interaction is visualised in a centroid nMDS plot (Fig. 4.15a) and explored in detail in a suit of one-way ANOSIM tests (Table 4.15). On this ordination the point for each month at each site have a different trajectory. This lack of consistency is highlighted in the associated shade plot, were species such as A. senhousia were found at reefs 1 and 2 during the first three month, but only at reefs 1 and 3 in December (Fig. 4.15b). Moreover, much of the pattern of differences between sites and months are driven by inconsistent changes in the abundance of a few key species, e.g. C. minor, P. excavatum, Barnardomelita matilda and Ericthonius sp.1.
Table 4.14. Mean squares (MS), percentage contribution of mean squares to the total mean squares (%MS), pseudo-F ratios (pF) and significant levels (P) derived from a two-way PERMANOVA tests on the composition of the small invertebrate fauna of the mussel reefs. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
df MS %MS pF P Month 3 7285.3 46.34 7.12 0.001 Site 2 4918.4 31.28 4.81 0.001 Month × Site 6 2494.9 15.87 2.44 0.001 Residual 24 1022.7 6.51
74 Table 4.15. R statistic and/or significance level (P) values derived from one-way ANOSIM tests on the composition of the small invertebrate fauna of the mussel (a) overall and in each month (b-e) and among (f) month and (g-i) at each site. Significant results (P = <0.05) are highlighted in bold, and those significant and with a %MS ≥ 25 are shaded grey.
(a) P = 0.004; Global R = 0.137 (f) P = 0.001; Global R = 0.445 Reef 1 Reef 2 May Jul Sep Reef 2 0.113 Jul 0.255 Reef 3 0.109 0.21 Sep 0.407 0.299 Dec 0.811 0.447 0.059 (b) P = 0.014; Global R = 0.350 Reef 1 Reef 2 (g) P = 0.001; Global R = 0.836 Reef 2 0.630 May Jul Sep Reef 3 0.074 0.222 Jul 0.741 Sep 0.481 0.852 (c) P = 0.004; Global R = 0.588 Dec 1.000 1.000 1.000 Reef 1 Reef 2 Reef 2 0.667 (h) P = 0.001; Global R = 0.759 Reef 3 0.815 0.444 May Jul Sep Jul 0.296 (d) P = 0.004; Global R = 0.506 Sep 1.000 0.815 Reef 1 Reef 2 Dec 0.778 0.852 0.963 Reef 2 0.481 Reef 3 0.778 0.630 (i) P = 0.006; Global R = 0.568 May Jul Sep (e) P = 0.004; Global R = 0.753 Jul -0.185 Reef 1 Reef 2 Sep 0.407 0.370 Reef 2 0.333 Dec 0.926 0.926 0.741 Reef 3 1.000 0.593
75 40 (a) 2D Stress: 0.2 Month May July Sep Dec 20 av:May av:July av:Sep
av:Dec
2 S
D 0 M
-20
-40 -20 0 20 40
MDS1 y
(b) c
y
p a
3 l
e
e
u
M
J
S D Tanaidacea spp. Amphipoda 2 Amphipoda Corophium minor Paracorophium excavatum Caprella spp Ericthonius sp.1 1 Melita sp.1 Barnardomelita matilda Copepoda spp. Polychaete sp.1 0 Polychaete sp.2 Capitella capitata Sabellidae sp.1 Sabellidae sp.2 Lumbrinereidae spp. Polynoidea spp. Simplisetia spp. Oenonidae spp. Cirriformia spp. Dorvilleidae spp. Syllidae sp.1 Syllidae sp.2 Prionospio cirrifera Sphaeromatidae sp. Janiridae sp. Arcuatula senhousia Fluviolanatus subtortus Mytilus galloprovincialis Ostracoda sp.1 Ophiuroidea spp. Cnidaria spp. Platyhelminthes spp.
Fig. 4.13. (a) Two dimensional mMDS plot constructed from bootstrapped averages for month, calculated from a Bray-Curtis resemblance matrix of transformed abundances of large invertebrates. Group averages (black squares) and 95% region estimates fitted to the bootstrap averages are provided. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded in each month. Shading intensity is proportional to abundance.
76 (a) 2D Stress: 0.16 Site Reef 1 Reef 2 Reef 3 20 av:Reef 1 av:Reef 2
av:Reef 3
2 S
D 0 M
-20
-40 -20 0 20 40
MDS1
1 2
(b) 3
3
f
f
f
e
e
e
e
e
e
R
R R Ericthonius sp.1 2 Barnardomelita matilda Corophium minor Caprella spp Paracorophium excavatum Melita sp.1 1 Sphaeromatidae sp. Tanaidacea spp. Arcuatula senhousia Amphipoda Janiridae sp. 0 Copepoda spp. Ostracoda sp.1 Cirriformia spp. Polynoidea spp. Sabellidae sp.1 Fluviolanatus subtortus Capitella capitata Mytilus galloprovincialis Dorvilleidae spp. Simplisetia spp. Prionospio cirrifera Lumbrinereidae spp. Polychaete sp.1 Syllidae sp.2 Oenonidae spp. Amphipoda Sabellidae sp.2 Syllidae sp.1 Ophiuroidea spp. Polychaete sp.2 Platyhelminthes spp. Cnidaria spp.
Fig. 4.14. (a) Two dimensional mMDS plot constructed from bootstrapped averages for site, calculated from a Bray-Curtis resemblance matrix of transformed abundances of large invertebrates. Group averages (black squares) and 95% region estimates fitted to the bootstrap averages are provided. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded at each site. Shading intensity is proportional to abundance.
77 (a) 2D Stress: 0.1 SiteMonth Reef 1 Reef 1May Reef 2May Reef 3May Reef 1July Reef 2July Reef 3July Reef 3 Reef 1Sep Reef 2Sep Reef 3Sep Reef 1Dec Reef 3 Reef 1 Reef 2Dec Reef 3Dec Reef 1 Reef 1 Reef 3
Reef 2
Reef 2 Reef 2 Reef 3
Reef 2
1
2
3
1
2
3
1
2
3
1 2
(b) 3
7 Month
f
f
f
f
f
f
f
f
f
f
f
f
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e
e e
6 e May
R
R
R
R
R
R
R
R
R
R
R R 5 July Cnidaria spp. 4 Sabellidae sp.2 Sep Oenonidae spp. 3 Lumbrinereidae spp. Dec Syllidae sp.1 Amphipoda 2 Polychaete sp.1 Sabellidae sp.1 1 Fluviolanatus subtortus Copepoda spp. 0 Caprella spp Arcuatula senhousia Corophium minor Barnardomelita matilda Ericthonius sp.1 Paracorophium excavatum Sphaeromatidae sp. Melita sp.1 Amphipoda Tanaidacea spp. Simplisetia spp. Janiridae sp. Ostracoda sp.1 Polynoidea spp. Cirriformia spp. Capitella capitata Prionospio cirrifera Syllidae sp.2 Ophiuroidea spp. Dorvilleidae spp. Polychaete sp.2 Mytilus galloprovincialis Platyhelminthes spp.
Fig. 4.15. (a) Two-dimensional centroid nMDS plot derived from a distance-among-centroids matrix constructed using the abundance data in each sample collected after deployment of the mussel reef. (b) Shade plot constructed using transformed and averaged abundances of each large invertebrate species recorded in month and site combination. Shading intensity is proportional to abundance.
78 4.5 Discussion
This study determined the characteristics of the fish and invertebrate faunas on novel vertical shellfish habitats (i.e. wooden stakes seeded with translocated M. galloprovincialis) deployed in the Melville Water region of the Swan-Canning Estuary. Remote Underwater Video (RUV) systems were used to monitor the fish communities associated with the mussel reefs and nearby unstructured habitats before and after their deployment. A total of 6,946 fish (including the portunid P. armatus) from 32 species were identified from the 144 RUV videos, with O. rueppellii, Atherinidae spp. and H. octolineatus being the most abundant. The communities of large (> 2,000 µm) and small (500 - 2,000 µm) invertebrates from the top sections of the stakes were sampled from each of the three reef sites in May, July, September, and December 2019. These samples yielded > 35,000 invertebrates from > 40 species, with ascidians and barnacles dominating the large and crustaceans the small invertebrates, respectively.
4.5.1 Fish communities 4.5.1.1 Pre-mussel reef deployment There were no significant differences any of the measures of fish faunal community structure between the mussel reefs sites prior to the deployment of the stakes and the control sites. Thus, the was no pre-existing underlying difference between the two types of habitat. This is not surprising as most spatial differences in estuarine fish communities, including in the Swan-Canning Estuary, are related to changes in physio-chemical conditions, normally salinity, or the provision of habitat, usually submerged aquatic vegetation (e.g. Valesini et al. 2009; Hogan-west et al. 2019; Poh et al. 2019). In current study control sites were only 0.5 – 1 km away from the mussel reef sites and thus not sufficiently far away to experience a different suite of environmental conditions and comprised only unstructured, silty substrate (Department of Water and Environmental Regulation, 2020).
79 Although the mean number of species, total MaxN and Simpson’s Index remained similar over the four months before reef deployment, there was a shift in faunal composition. Monthly changes in the composition of the fauna in the Swan- Canning Estuary have been detected previously and are usually related the increase salinities facilitating the spread of marine associated species further upstream (e.g. Hoeksema & Potter, 2006; Poh et al. 2019).
4.5.1.2 Post-mussel reef deployment A significantly greater mean number of species were observed on the mussel reef than control sites. Comparisons between structurally complex shellfish reef sites vs unvegetated controls undertaken in other parts of the world have yielded similar results, which are explained by the provision of shelter and additional prey sources (La Peyre et al., 2014; McLeod et al., 2014; Humphries & Peyre, 2015; Christianen et al., 2017; Gilby et al., 2020). Almost all species found exclusively at reef sites, namely the boxfish A. amygdaloides, the seahorse H. angustus and several species leatherjacket, (e.g. Monacanthus chinensis, Acanthaluteres vittiger and Eubalichthys mosaicus), have a preference for complex structures, explaining their non-existence at unvegetated sites (Jenkins & Wheatley, 1998; Hellyer et al., 2011; Florisson et al. 2018). These species also have strong marine affinities and are not typically recorded in south-western Australian estuaries (Potter et al., 1990; Valesini et al., 2014).
The lack of significant difference in mean total MaxN between the mussel reef and control sites may have been attributed to the increased abundance of heavily schooling species such as Atherinidae spp. and the anchovy E. australis, at the control sites. These two taxa accounted for large proportion of the number of individuals (30%), despite their frequency of occurrence being relatively low (i.e. 18 and 2 %, respectively).
Zoobenthivores and omnivores were the dominant feeding guilds at reef and control sites accounting for ~ 97 % of individuals and ~ 77 % of species at reef sites and
80 ~ 77 % of individuals and ~ 71 % of species at control sites. The pronounced contribution made by these groups could be reflective of the urbanisation of the Swan-Canning Estuary and, to a lesser extent, the provision of the shellfish habitat. Studies in estuaries on the eastern coast of Australia have demonstrated the zoobenthivores and omnivores are typically more abundant in highly urbanised estuaries, attributed to the decreased number of predators (i.e. piscivores) and increased feeding opportunities through artificial habitats e.g. hardened shorelines and jetties (Chapman & Blockley, 2009; Bishop et al., 2017; Henderson et al., 2020). This is observation is supported by the essential absence of piscivores (i.e. represented by three species P. saltatrix, A. japonicus and Platycephalus westraliae and a total MaxN of five) recorded on either the reef or control sites throughout the 14 months of study. Moreover, the link between degradation and urbanisation is reinforced by the fact that the multimetric index of health for the Swan-Canning Estuary uses the number and proportion of trophic specialists, e.g. piscivores, as a positive metric (Hallett et al., 2012, 2019). Unsurprisingly, zoobenthivores were more specious at reef sites. This is likely due to the increased availably of invertebrate prey communities these vertical structures provide, and zoobenthivorous fish species have been shown to be associated with complex structures that provide vertical relief (Gilby et al., 2018; Aslam et al., 2020).
Although the abundance of most species was greater on the reefs and they harboured some distinct fauna, there were several species that were mainly found at the control site. The most notable of which was the Eagle Ray M. tenuicaudatus. Although not abundant this species was recorded in > 30% of RUV videos from the control site, but only occurred on reef sites when the field of view of the RUV extended beyond the reef, indicating they do not use the structure. This result is due to this species feeding on sand flats by pumping water from their mouth to liquify sediment, creating pits, from which they primarily consume infauna, particularly bivalves (Hines et al., 1997). Thus, the rays would not directly benefit from the abundance of mussels on a
81 vertical structure as their feeding method would not allow ingestion. Moreover, while M. tenuicaudatus has been found on the large (3 m3) concrete artificial reefs in Geographe Bay, their abundances were greater when the camera was facing away from the reef modules (Tweedley et al., 2016a).
Contrary to expectations, abundances of the recreationally iconic Black Bream A. butcheri were relatively low. This was unexpected, as its association with structural complexity in estuaries has been well documented (Hindell et al., 2008) and it is known to feed on mussels (Sarre et al., 2000; Poh et al., 2018). Moreover, the congeneric Yellowfin Bream Acanthopagrus australis was found to be more abundant on deployed oyster habitat in the Noosa River (Queensland) than on unvegetated substrate (Gilby et al., 2019). While A. butcheri does occur in Melville Water (Poh et al., 2018), its relatively low abundance probably reflects the preference this species has for areas further upstream that experience lower salinities (Cottingham et al., 2018).
The number of species, total MaxN, Simpson’s Index and faunal community differed among months. Seasonal changes in fish fauna are a well-established characteristic of permanently open microtidal estuaries globally (Hoeksema & Potter, 2006; Harrison & Whitfield, 2006, 2008; Potter et al. 2016). During the summer and autumn months, when freshwater discharge is low and salinities are high, there is an increase in the number fish species with marine affinities (Whitfield, 1999; Elliott et al., 2007). Whereas, the opposite is true in winter, when seasonal rainfall results in freshwater flow and a subsequent decline in salinity. This does cause a reduction in the abundance and occurrence of marine species, which are not able to osmoregulate as effectively as estuarine species (Potter & Hyndes, 1999). While this effect has been reported in netting studies, it is worth noting that visibility was reduced during these months too, which may have exacerbated the temporal trends by obscuring some fish, particularly those that do not approach the camera and/or bait.
82 4.5.2 Large invertebrate communities
A total of 16,363 large invertebrates (> 2,000 µm) were recorded on the mussel reefs, comprising 16 taxa from five classes. Of the species, those of barnacles and ascidians were the most abundant. The number of species, individuals, Simpson’s Index and faunal composition all differed significantly between months. The successive increases in mean species richness observed, likely reflect the colonisation of the stakes over time by more species and the mortality of M. galloprovincialis which occurred throughout the study (Chapter 3). The reduction in the number of mussels would create space for colonisation by other fauna, thus increasing diversity.
Temporal changes in the number of individuals and composition reflect patterns in the abundance of individual species. The ascidian C. intestinalis was one of most abundant species during winter (May-September), but declined in December. This may be attributed to the biology of this species, as its filtration capacity decreases rapidly in temperatures > 21 °C (Petersen & Riisgard, 1992). Although this may not directly contribute to mortality, it would reduce the organism’s fitness and facilitate it being outcompeted. This was evident as the presence of the colonial ascidian S. plicata during December. While this species was not abundant, the individuals present were often large and robust. Styela plicata is also more tolerant to warmer waters and prefers higher salinities, which could explain its appearance in December and not in the cooler months when C. intestinalis were more abundant (Pineda et al., 2013).
Significant differences in the composition of large invertebrates were recorded between sites, with Reef 3 being different from both 1 and 2 which were not significantly different from each other. A possible explanation for the similarity in faunal composition between reefs 1 and 2 may be their closer proximity to other prominent hard structures, i.e. the South Perth Yacht Club. It is well established that vessels act as a transport vector for sessile organisms (Acosta & Forrest, 2009; Veiga et al., 2020) and therefore could facilitate the relocation of taxa to the seeded mussel stakes.
83 Differences between the reefs were largely driven by greater abundances of the non-native bivalve A. senhousia at reefs 1 and 2. The presence of much finer, silty sediments around Como (reefs 1 and 2) compared to Heathcote (Reef 3) suggests that water movement is reduced at reefs 1 and 2, allowing the deposition of fine particles (C. Maus; personal observation; Tweedley et al., 2017a). This is relevant as A. senhousia is associated with fine, muddy sediment and reduced water movement. Moreover, a drop- camera survey of > 2,000 sites in the Swan-Canning Estuary by Cottingham et al. (in prep) found similar results, with A. senhousia being abundant in sediments in the upper parts of Melville Water, surrounding reefs 1 and 2, but only sporadically around Reef 3.
Although not contributing much to changes in faunal composition, it is important to note the impact of the non-native solidary ascidian A. aspersa and barnacle A. variegatus, which colonised the stakes rapidly upon their deployment. Both these species were recorded in high abundances throughout months and on all reefs. The rapid colonisation of the stakes, particularly by A. aspersa may be attributed to its biology. This species can live for 18 months and adapt their reproductive biology from gonochoristic to a hermaphrodite to allow spawning to occur throughout the year (Lynch et al., 2016). Furthermore, A. aspersa is well suited to urbanised and highly eutrophic estuaries and deeper slow-moving waters (Mastrototaro et al., 2008). These biological traits likely explain the lack of temporal change in the abundance of this non- native ascidian.
4.5.3. Small invertebrate communities
A total of ~ 19,000 macroinvertebrates (500 – 2,000 µm) from 33 species were recorded as epifauna associated with the mussel reefs. While polychaetes and crustaceans dominated the fauna, the species associated with the vertical mussel reefs were distinct from those that typify the infauna of estuaries in south-western Australia (cf. Table 4.12; Wildsmith et al., 2011; Tweedley et al., 2012, 2014). Polychaetes are
84 normally among the most abundant taxa, but only accounted for 1% of individuals in the current study. This is likely due to the lack of habitat available for infaunal species (e.g. Capitella capitata), albeit the hard substrate would facilitate colonisation by the erpulid, Ficopomatus enigmaticus, that does occur in the estuary (Wildsmith et al., 2011; Tweedley et al., 2016b). Polychaete abundance has been negatively correlated with high densities of ascidians in suspended mussel culture (Lutz-Collins et al., 2009) and so the large numbers of ascidians present on the stakes may play an ancillary role in the limited number of polychaetes.
Some amphipod species common in benthic habitats were, however, recorded relatively high abundances i.e. Corophium minor and Paracorophium excavatum. These species are epifaunal and would thus be able to exploit the complex vertical habitats provided by the stakes and their associated fauna. It is possible that the bio deposits produced by M. galloprovincialis on the stake may have created a ‘pseudo-benthic’ environment allowing some species normally associated with sediment to utilise the habitat such as burrowing amphipod C. minor.
Faunal composition differed significantly between sites and months, the most conspicuous trend among which was a reduction in abundance of fauna in September and proliferation of species and individuals in December. This was likely caused by the extended period of low oxygen (< 4 mgL-1) deeper waters of Melville Water throughout the winter months. Hypoxia in the upper reaches of this estuary led to a reduction in the number of species and individuals and, as happened in the current study, particularly of crustacean species (Tweedley et al., 2016b). In that study it is thought that mobile species, such as amphipods, undertook a vertical migration into the better oxygenated surface waters. This phenomenon may have occurred here to allow the rapid colonisation observed in December.
85 4.6 Appendices Mussel reef (Pre deployment) Control site (Pre deployment) (a) (b) 4.5 4.5 4.0 4.0
s 3.5 s 3.5
e e
i i
c c
e 3.0 e 3.0
p p
s s
2.5 2.5
f f
o o
r 2.0 r 2.0
e e
b 1.5 b 1.5
m m
u 1.0 u 1.0
N N 0.5 0.5
0 0
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6 Minutes Minutes
(c) (d) 100 100
80 80
N
N
x x
a 60 60
a
M
M
l
n
a
a t
t 40 40
o
o
T T 20 20
0 0
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6 Minutes Minutes
(e) (f) 0.6 0.6
0.5 0.5
x x
e e
d 0.4 d 0.4
n n
I I
s s
' '
n 0.3 n 0.3
o o
s s
p 0.2 p 0.2
m m
i i
S S 0.1 0.1
0 0
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6 Mins Minutes
Appendix 1. Accumulation curves for the mean (a,b) number of species, (c,d) total MaxN and (e,f) Simpson’s Diversity Index for the mussel reef (a, c, e) and control site (b, d, f), respectively prior to the deployment of the reef. Error bars represent ±95% confidence limits.
86 Mussel reef (Post deployment) Control site (Post deployment) (a) (b) 4.5 4.5
4.0 4.0 s
3.5 s 3.5
e
e
i
i
c
c e
3.0 e 3.0
p
p
s
s
2.5 2.5
f
f
o
o
r
2.0 r 2.0
e
e b
1.5 b 1.5
m
m u
1.0 u 1.0
N N 0.5 0.5
0 0
0 5 0 5 0 5 0 5 0 5 0
0
5
0
5
0
5
0
5
0
5
0
5
5
1 1 2 2 3 3 4 4 5 5 6
1
1
2
2
3
3
4
4
5
5 6 Minutes Minutes
(c) (d) 100 100
80 80
N N
x 60 x 60
a a
M M
l l
a a
t 40 t 40
o o
T T
20 20
0 0
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6 Minutes Minutes
(e) (f) 0.6 0.6
0.5 0.5
x x
e e
d d 0.4
n n
I I
0.4
s s
' '
n n 0.3
o o
s 0.3 s
p p 0.2
m m
i i
S S 0.2 0.1
0.1 0
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
0 0
5 5
1 1
1 1
2 2
2 2
3 3
3 3
4 4
4 4
5 5
5 5
6 6 Minutes Minutes
Appendix 2. Accumulation curves for the mean (a,b) number of species, (c,d) total MaxN and (e,f) Simpson’s Diversity Index for the mussel reef (a, c, e) and control site (b, d, f), respectively after the deployment of the reef. Error bars represent ±95% confidence limits.
87 (a) 0 Habitat PreReef PreControl PostReef PostControl 20
40
y
t
i
r
a
l
i
m
i S 60
80
100
0
5
0
5
0
5
0
5
0
5
0
0
5
0
5
0
5
0
5
0
5
0
0
5
0
5
0
5
0
5
0
5
0
0
5
0
5
0
5
0
5
0
0
5
5
5
5
5
1
1
2
2
3
3
4
4
5
5
6
1
1
2
2
3
3
4
4
5
5
6
1
1
2
2
3
3
4
4
5
5
6
6
5
5
4
4
3
3
2
2
1 1 Samples (b) 2D Stress: 0.05 Habitat PreReef 10 15 5 PreControl PostReef 25 PostControl 20 35 30 4045505560 10 15 20 5 4045505560 25 3035
15 2530354045 10 20 505560 5
10 30354045505560 5 152025
Appendix 3. (a) Cluster dendrogram and (b) nMDS ordination plot derived from a Bray-Curtis resemblance matrix, constructed from the dispersion-weighted and square-root transformed and averaged MaxN abundances of each species in five-minute intervals, captured from RUV footage on the mussel reef and control site before and after the deployment of the reef. The dashed horizontal line in (a) indicates a Bray-Curtis similarity of 95%.
88
PostControl
PostReef
PreControl
PreReef Habitat
6 0
5 5
5 0
4 5
4 0
3 5
3 0
2 5
2 0
1 5
1 0
5
6 0
5 5
5 0
4 5
4 0
3 5
3 0
2 5
2 0
1 5
1 0
5
6 0
5 5
5 0
4 5
4 0
3 5
3 0
2 5
2 0
1 5
1 0
5
6 0
5 5
5 0
4 5
4 0
3 5
3 0
2 5
2 0
1 5
1 0
5
Sillagoburrus
Atherinidaespp.
Pentapodusvitta
Portunasarmatus
Engraulisaustralis
Gerres subfasciatusGerres
Monacanthidaespp.
Helotesoctolineatus
Pomatomussaltatrix
Rhabdosargussarba
Ostorhinchusruepellii
Favonigobiuslateralis
Acanthaluteresvittiger
Sillaginoidespunctata
Pseudogobiusolorum
Nematalosavlaminghi
Arenigobiusbifrenatus
Platycephalusaustralis
Callogobiusdepressus
Eubalichthysmosaicus
Amniatobacaudavittata
Monacanthuschinensis
Argyrosomusjaponicus
Acanthopagrusbutcheri
Hippocampusangustus
Favonigobiuspunctatus
Myliobatistenuicaudatus
Pseudorhombusjenynsii
Gymnapistesmarmoratus
Anoplocaprosamydaloides
Torquigener pleurogrammaTorquigener
Cnidoglanismacrocephalus
Parablenniuspostoculomaculatus
0
0.1
0.2
0.3
0.4
0.5 0.6
Appendix 4. Shade plot constructed from the dispersion-weighted and square-root transformed and averaged MaxN abundances of each species in five-minute intervals, captured from RUV footage on the mussel reef and control site before and after the deployment of the reef. Darker shading indicates a higher abundance of a species.
89 (a) (b) 3.0 30
s 2.5
e
i
c e
2.0 N 20
p
x
s
a
f
M
o 1.5
l
r
a
t
e o
b 1.0 10
T m
u 0.5 N
0 0
l l
f
f
o o
e
e
r r
e
e
t t
n n
R
R
o o
C C Habitat Habitat (c)
0.6
x
e d
n 0.4
I
s
'
n
o s
p 0.2
m
i S
0
l
f
o
e
r
e
t
n
R
o C Habitat Appendix 5. Mean (a) number of species (b) total MaxN and (c) Simpson’s Diversity Index on sites the mussel reef and control site prior to the deployment of the reef. Error bars represent ±95% confidence limits.
90 Chapter 5: General conclusion
5.1 Critical review
Due to increasing recognition of the ecosystem goods and services shellfish reefs provide and the fact that much of this type of habitat has been lost to anthropogenic influences (fishing and urbanisation), restoration efforts are increasing in Australia and around the world (Beck et al., 2011; Gillies et al., 2018, 2020). While most restoration efforts typically involve the provision of low relief habitat, e.g. limestone and/or recycled shells (Geraldi et al., 2013; Branigan et al., 2020), this project provided a novel approach to creating shellfish habitat in an area of an estuary that is devoid of complex structure. Vertical shellfish reefs were constructed using 100 wooden stakes (~ 1 m vertical profile) each seeded with 3 L of translocated individuals of the Mediterranean Mussel Mytilus galloprovincialis. Three of such reefs were deployed in the Melville Water region of the
Swan-Canning Estuary (south-western Australia) in March 2019 and monitored until July 2020. The aims of the study were to: 1) investigate the biology (i.e. mortality, body condition and growth) of the translocated M. galloprovincialis at each of the three newly constructed reef sites; 2) compare the characteristics of the fish fauna on the mussel reefs and nearby unstructured habitats; and 3) determine which benthic invertebrates (including ascidians) were associated with these structures.
The first component of this study (Chapter3) investigated the biology of the translocated M. galloprovincialis and showed that mortality throughout the three sites was considerably higher than that of expected natural mortality given the maximum age of four-years recorded elsewhere. This high mortality was attributed to poor environmental conditions (e.g. low dissolved oxygen) and sedimentation, which was further compounded by translocation-induced stress and the fouling of the stakes by ascidians. Furthermore, the translocation occurred in the lead up to the spawning season, at which time of the year the body condition of most Mytilus species decreases. The conclusion that the decline in body condition was attributable to spawning was
91 consistent with the presence of recruits in the months following the lowest body mass means, coupled with the increase in body condition in the months following. The growth of M. galloprovincialis exhibited some seasonal variability, which is common in mytilid mussels. The presence of high phytoplankton biomass, however, did result in some individuals on the reefs attaining total shell lengths of ~50 mm by the end of their first year of life, which was greater than that recorded in other waterways.
Focus was next placed on investigating whether the fish communities at the mussel reefs differed from those at nearby unstructured (sandy) habitats (Chapter 4). A greater number and abundance of fish were recorded at the reef sites compared to control sites, with most of the species found only on the reef having marine affinities and a presence for rocky reef habitat (e.g. monacanthids, aracanids and syngathids). Furthermore, there was an increase in zoobenthivore fish (e.g. Ostorhinchus rueppellii) at the reef sites, suggesting that the reef structures may provide increased food resources, by providing a habitat for small benthic invertebrates. Macroinvertebrate communities differed both spatially and temporally and a high abundance of class Malacrustaca dominated the macroinvertebrate abundance. Polychaetes were specious over the study, however, accounted for very little of the overall abundance at all three reef sites.
The sustainability and success of any project to create shellfish habitat ultimately relies on adequate survival and recruitment of the candidate species (i.e. M. galloprovincialis). Thus, while the mortality of translocated mussels was higher than natural mortality rates, it was encouraging that recruitment did occur, particularly at reefs 2 and 3. Given that few of the originally deployed mussels would survive to reach a second spawning season, i.e. reach two years of age, annual recruitment is essential for the mussels to retain their ‘real estate’ among the reef and not be smothered with ascidians. Lack of such natural recruitment has posed problems for shellfish reef projects elsewhere. For example, the translocation and deployment of Green-lipped Mussels
92 Perna canaliculus in New Zealand resulted in their overall mortality after two years due to a lack of recruitment (Wilcox et al., 2018). Although this may be counteracted by the continual seeding of reefs, albeit at additional economic costs.
Mytilid mussels typically recruit back to the hard substrate that the aggregations provide (Ceccherelli & Rossi, 1984; Newell et al., 2010). Given the dominance of ascidians in colonising hard substrates in the Swan-Canning Estuary (Chapter 4), this may become problematic, given the soft tunics do not provide a suitable settling substrate for reef-forming bivalves. In fact, a recently deployed pilot limestone rubble reef in Melville Water, as part of a large program to create shellfish reef habitat, has shown similar results with the colonisation of similar ascidian species to the limestone (Fig. 5.1). The large numbers of ascidians present on the stakes in the current study is attributable, in part, to the low stocking density of M. galloprovincialis used. This was, however, unavoidable given that the mussels were hand (rather than machine) seeded and the limited supply of mussels available at the time. Future studies may investigate whether increasing the stocking density of bivalves plays some role in preventing or reducing the colonisation of ascidians.
Fig. 5.1. Ascidians (predominantly Ascidiella aspersa) colonising deployed limestone rubble for future shellfish reef restoration project in Melville Water in the Swan-Canning Estuary (C. Maus., 2020).
It is important to consider the implications of a shift from a mytilid-dominated structure to an ascidian-dominated structure. Both are filter feeders and thus provide the same ecosystem service in the removal of phytoplankton and sediment from the water column. However, given that all the ascidians colonising the stakes in the current
93 project were non-native species, consideration must be given to the reefs likely aiding their dispersal. It should be noted all except one of the ascidian species recorded here are known to occur in Western Australian waters (Wells et al. 2009). Many fish living in the Swan-Canning Estuary, particularly the recreationally important Black Bream (A. butcheri), are known to feed on mussels (Sarre et al., 2000; Poh et al., 2018). Thus, the mussel provides a direct and, through the provision of habitat, indirect food source.
While ascidians would provide some habitat for invertebrates, a study of a mussel aquaculture operation in Atlantic Canada showed that colonisation of the mussel lines decreased the abundance of particular invertebrates (Lutz-Collins et al., 2009). Furthermore, ascidians have few natural predators particularly among the fish community (Kincaid & Rivera, 2020) and so provide no direct nutritional benefit to fish. This could be due to their low nutritional content compared to other benthic invertebrates (Gabriele et al., 1997). Ascidians, like sponges, are known to be unpalatable to fish through a combination of physical and/or chemical mechanisms (Teo & Ryland, 1994; Tarjuelo et al., 2002). For example, some species are acidic (pH ~ 1). Thus, these species would likely only be predated on by specialised feeders (Maschette et al., 2020), which are not known to occur in the Swan-Canning Estuary.
The placement of shellfish reefs has been shown to influence the abundance of fish which inhabit and utilise them. For example, oyster reefs in the Noosa River Estuary (Australia) placed in close proximity to mangroves and seagrass meadows, contained less fish than those corresponding habitats in more isolated areas, presumably as fish can utilise multiple nursery habitats (Gilby et al., 2020). On the other hand, when placed into areas without structurally complex habitats (i.e. Melville Water), fish utilise the reef quickly and in high numbers due to the lack of alternative habitat in the area, i.e. deeper waters in other parts of the estuary. Given this the extent of utilisation by fish, particularly those targeted by recreational fishers, the success of artificially deployed shellfish habitats is likely influenced by the surrounding ‘estuary scape’.
94 5.2 Future directions
Given global efforts to restore and enhance estuaries and coastal waters with shellfish habitats, and the aim to undertake large-scale projects in south-western Australia, this study has increased the knowledge not only of the biology of translocated M. galloprovincialis, but also the ecology and succession of organisms associated with such structures. The results of this pilot study are currently being used to inform a multi- million-dollar shellfish reef project in the Swan-Canning Estuary led by The Nature Conservancy, who have undertaken similar work elsewhere in Australia and the world
(e.g. Gillies et al., 2017).
The results of this study indicated that high rates of natural mortality and limited, sporadic recruitment could be a major threat to the longevity and success of shellfish projects in the Swan-Canning Estuary. While the presence of recruitment is a positive sign, the numbers of mussel did not equate to the numbers lost to mortality during the study. This is analogous to the problems with aquaculture-based enhancement, where more fish in does not necessarily mean more fish out (Loneragen et al., 2019). There are several methods commonly used to improve the outcomes of such release programs and they involve altering the i) release site, ii) size-at-release, ii) timing-of-release and iv) number released (e.g. Hines et al., 2008; Tweedley et al., 2017a). Given the proliferation of non-native ascidians on the seeded mussel stakes future reef deployment strategies could be modified to mitigate this impact. This could involve, deploying reefs in areas away from other hard structures (e.g. jetties, channel markers and seawalls) to limit recruitment. Increasing the size and number of mussels seeded to ‘fill’ all available space on the stake and reduce the extent of bare surfaces to discourage ascidian recruitment. If a better understanding of the biology and ecology of the fouling ascidian species was known (e.g. the timing of their reproduction) then mussel reefs could be deployed at a time of the year when larval supplies are limited, allowing the seeded mussel to take hold and prevent colonisation and competition.
95 The preliminary results of a concurrent shellfish project in the Swan-Canning Estuary indicate that the survival of M. galloprovincialis was greater in shallower waters, possibly attributed to increases dissolved oxygen concentrations. Thus, it may be worth understanding whether of survival and recruitment of mussels in shallower depths (i.e. < 2 m) is enhanced and could increase the success of any future reef deployment. However, it is worth noting that, current governmental regulations do not allow the deployment of artificial substrates in depths < 3 m in the Swan-Canning Estuary, and so the deployment of shallow reefs would not be permitted under current management arrangements. Finally, as strategies to mitigate the colonisation of hard substrates by ascidians may not work effectively, and that these species do filter the overlying water column reducing phytoplankton blooms and provide habitat for associated fauna, there is value in understanding the invertebrate communities that ascidian may support and comparing those to shellfish reefs, bare substrate and seagrass beds.
96 References
Abada-Boudjema, A., & Dauvin, J. (1995). Recruitment and life span of two natural mussel populations Perna Perna (Linnaeus) and Mytilus galloprovincialis (Lamarck) from the Algerian coast. Journal of Molluscan Studies, 61(4), 467–481. Acosta, H., & Forrest, B. M. (2009). The spread of marine non-indigenous species via recreational boating: A conceptual model for risk assessment based on fault tree analysis. Ecological Modelling, 220(13-14), 1586–1598.
Adams, J. B., Taljaard, S., Niekerk, L. Van., & Lemley, D. A. (2020). Nutrient enrichment as a threat to the ecological resilience and health of South African microtidal estuaries. African Journal of Aquatic Science, 45(1-2), 23–40. Anderson, M., Gorley, R. N., & Clarke, K. R. (2008). PERMANOVA + for PRIMER. Guide to software and statistical methods, Plymouth, UK, PRIMER-E. Anestis, A., Lazou, A., Pörtner, H. O., & Michaelidis, B. (2007). Behavioural, metabolic, and molecular stress responses of marine bivalve Mytilus galloprovincialis during long-term acclimation at increasing ambient temperature. American Journal of Physiology - Regulatory Integrative and Comparative Physiology, 293(2), 911–921. Aslam, S., Dekker, H., Siddiqui, G., Mustaquim, J., & Kazmi, S. J. H. (2020). Biodiversity on intertidal oyster reefs in the Hab River mouth: 35 new records from Pakistan. Regional Studies in Marine Science, 39, 101415. Azpeitia, K., Ferrer, L., Revilla, M., Pagaldai, J., & Mendiola, D. (2016). Growth, biochemical profile, and fatty acid composition of mussel (Mytilus galloprovincialis Lamarck.) cultured in the open ocean of the Bay of Biscay (northern Spain). Aquaculture, 454, 95–108. Azpeitia, K., Urrutia, M. B., & Mendiola, D. (2018). Analysis of culture production and growth rate patterns of mussel (Mytilus galloprovincialis Lamarck.) cultured in the open ocean of the SE Bay of Biscay for the commercial product development. Aquaculture International, 26, 1211–1228. Barragán, J., & de Andrés, M. (2015). Analysis and trends of the world’s coastal cities and agglomerations. Ocean and Coastal Management, 114, 11–20. Beck, M. W., Brumbaugh, R. D., Airoldi, L., Carranza, A., Coen, L. D., Crawford, C., Defeo, O., Edgar, G. J., Hancock, B., Kay, M. C., Lenihan, H. S., Luckenbach, M. W., Toropova, C. L., Zhang, G., & Guo, X. (2011). Oyster reefs at risk and recommendations for conservation, restoration, and management. BioScience, 61(2), 107–116.
97 Bergström, P., & Lindegarth, M. (2016). Environmental influence on mussel (Mytilus edulis) growth - A quantile regression approach. Estuarine, Coastal and Shelf Science, 171, 123–132. Bishop, M. J., Mayer-Pinto, M., Airoldi, L., Firth, L. B., Morris, R. L., Loke, L. H. L., Hawkins, S. J., Naylor, L. A., Coleman, R. A., Chee, S. Y., & Dafforn, K. A. (2017). Effects of ocean sprawl on ecological connectivity: Impacts and solutions. Journal of Experimental Marine Biology and Ecology, 492, 7–30. Bloomfield, A. L., & Gillanders, B. M. (2005). Fish and invertebrate assemblages in seagrass, mangrove, saltmarsh, and nonvegetated habitats. Estuaries, 28(1), 63– 77. Borsa, P., Daguin, C., & Bierne, N. (2007). Genomic reticulation indicates mixed ancestry in southern-hemisphere Mytilus spp. mussels. Biological Journal of the Linnean Society, 92(4), 747–754. Borthagaray, A. I., & Carranza, A. (2007). Mussels as ecosystem engineers: Their contribution to species richness in a rocky littoral community. Acta Oecologica, 31(3), 243–250. Bourque, D., Davidson, J., MacNair, N. G., Arsenault, G., LeBlanc, A. R., Landry, T., & Miron, G. (2007). Reproduction and early life history of an invasive ascidian Styela clava Herdman in Prince Edward Island, Canada. Journal of Experimental Marine Biology and Ecology, 342, 78–84. Branch, G. M., & Steffani, N. (2004). Can we predict the effects of alien species? A case- history of the invasion of South Africa by Mytilus galloprovincialis (Lamarck). Journal of Experimental Marine Biology and Ecology, 300(1–2), 189–215. Branigan, S., Fitzsimons, J., & Gillies, C. L. (2020). Modern middens: shell recycling for restoring an endangered marine ecosystem in Victoria, Australia. Ecological Management and Restoration, 21(3), 198–204. Brumbaugh, R. D., & Coen, L. D. (2009). Contemporary approaches for small-scale oyster reef restoration to address substrate versus recruitment limitation: A review and comments relevant for the Olympia Oyster, Ostrea lurida Carpenter 1864. BioOne, 28(1), 147–161. Buschbaum, C., Dittmann, S., Hong, J. S., Hwang, I. S., Strasser, M., Thiel, M., Valdivia, N., Yoon, S. P., & Reise, K. (2009). Mytilid mussels: Global habitat engineers in coastal sediments. Helgoland Marine Research, 63(1), 47–58. Cáceres-Martínez, J., & Figueras, A. (1998). Long-term survey on wild and cultured mussels (Mytilus galloprovincialis Lamarck) reproductive cycles in the Ria de Vigo (NW Spain). Aquaculture, 162(1-2), 141–156.
98 Ceccherelli, V., & Rossi, R. (1984). Settlement, growth and production of the mussel Mytilus galloprovincialis. Marine Ecology Progress Series, 16, 173–184. Çelik, M. Y., Karayücel, S., Karayücel, I., Öztürk, R., & Eyüboǧlu, B. (2012). Meat yield, condition index, and biochemical composition of mussels (Mytilus galloprovincialis Lamarck, 1819) in Sinop, south of the Black Sea. Journal of Aquatic Food Product Technology, 21, 198–205. Celik, M. Y., Karayücel, S., & Karayücel, Ý. (2009). Effects of environmental factors on growth and mortality of raft cultivated mussel (Mytilus galloprovincialis) cultivated in lantern nets in Black Sea. AACL Bioflux, 2, 97–108. Cerrato, R. M. (1990). Interpretable statistical tests for growth comparisons using parameters in the von Bertalanffy equation. Canadian Journal of Fisheries and Aquatic Sciences, 47(7), 1416–1426. Chapman, M. G. (2002). Early colonization of shallow subtidal boulders in two habitats. Journal of Experimental Marine Biology and Ecology, 275(2), 95–116. Chapman, M. G., & Blockley, D. J. (2009). Engineering novel habitats on urban infrastructure to increase intertidal biodiversity. Oecologia, 161(3), 625–635. Choi, Y. D. (2004). Theories for ecological restoration in changing environment: Toward ‘futuristic’ restoration. Ecological Research, 19, 75–81. Christianen, M. J. A., Heide, T. Van Der, Holthuijsen, S. J., Reijden, K. J. Van Der, Borst, A. C. W., & Olff, H. (2017). Biodiversity and food web indicators of community recovery in intertidal shellfish reefs. Biological Conservation, 213, 317–324. Chrystal, P. J., Potter, I. C., Loneragan, N. R., & Holt, C. P. (1985). Age structure, growth rates, movement patterns and feeding in an estuarine population of the cardinal fish Apogon rueppellii. Marine Biology, 85(2), 185–197. Clarke, K. R. (1993). Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology, 18(1), 117–143. Clarke, K. R, Chapman, M. G., Somerfield, P. J., & Needham, H. R. (2006). Dispersion- based weighting of species counts in assemblage analyses. Marine Ecology Progress Series, 320, 11–27. Clarke, K. R, & Gorley, R. N. (2015). Primer: User manual/tutorial. Primer-E Ltd., Plymouth, 93. Clarke, K. R., Gorley, R. N., Somerfield, P. J., & Warwick, R. M. (2014). Change in marine communities: an approach to statistical analysis and interpretation. Primer-E Ltd. Clarke, K. R, & Green, R. H. (1988). Statistical design and analysis for a ‘biological effects’ study. Marine Ecology Progress Series, 46, 213–226.
99 Clarke, K. R., Tweedley, J. R., & Valesini, F. J. (2014). Simple shade plots aid better long- term choices of data pre-treatment in multivariate assemblage studies. Journal of the Marine Biological Association of the United Kingdom, 94(1), 1–16. Cloern, J. E., Foster, S. Q., & Kleckner, A. E. (2014). Phytoplankton primary production in the world’s estuarine-coastal ecosystems. Biogeosciences, 11(9), 2477–2501. Coen, L. D., Brumbaugh, R. D., Bushek, D., Grizzle, R., Luckenbach, M. W., Posey, M. H., Powers, S. P., & Tolley, S. G. (2007). Ecosystem services related to oyster restoration. Marine Ecology Progress Series, 341, 303–307. Coleman, M. A., & Connell, S. D. (2001). Weak effects of epibiota on the abundances of fishes associated with pier pilings in Sydney Harbour. Environmental Biology of Fishes, 61(3), 231–239. Coleman, M. A., Wood, G., Filbee-Dexter, K., Minne, A. J. P., Goold, H. D., Vergés, A., Marzinelli, E. M., Steinberg, P. D., & Wernberg, T. (2020). Restore or redefine: Future trajectories for restoration. Frontiers in Marine Science, 7, 1–12. Conley, D. J., Kaas, H., Møhlenberg, F., Rasmussen, B., & Windolf, J. (2000). Characteristics of Danish estuaries. Estuaries, 23(6), 820–837. Cottingham, A., Huang, P., Hipsey, M. R., Hall, N. G., Ashworth, E., Williams, J., & Potter, I. C. (2018). Growth, condition, and maturity schedules of an estuarine fish species change in estuaries following increased hypoxia due to climate change. Ecology and Evolution, 8(14), 7111–7130. Creighton, C., Boon, P. I., Brookes, J. D., & Sheaves, M. (2015). Repairing Australia’s estuaries for improved fisheries production - What benefits, at what cost? Marine and Freshwater Research, 66(6), 493–507. Crisp, J. A., Loneragan, N. R., Tweedley, J. R., D’Souza, F. M. L., & Poh, B. (2018). Environmental factors influencing the reproduction of an estuarine penaeid population and implications for management. Fisheries Management and Ecology, 25(3), 203–219. Daguin, C., & Borsa, P. (2000). Genetic relationships of Mytilus galloprovincialis Lamarck populations worldwide: Evidence from nuclear-DNA markers. Geological Society Special Publication, 177(1), 389–397. Dame, R., Dankers, N., Prins, T., Jongsma, H., & Smaal, A. (1991). The influence of mussel beds on nutrients in the Western Wadden Sea and Eastern Scheldt estuaries. Estuaries, 14(2), 130–138. Dame, R. F., & Prins, T. C. (1997). Bivalve carrying capacity in coastal ecosystems. Aquatic Ecology, 31(4), 409–421.
100 Davenport, J., & Chen, X. (1987). A comparison of methods for the assessment of condition in the mussel (Mytilus edulis). Journal of Molluscan Studies, 53(3), 293– 297. Department of Water. (2014). Monitoring seagrass extent and distribution in the Swan- Canning Estuary. Water Science Technical Series, Report no. 70. Dias, P. J., Fotedar, S., & Snow, M. (2014). Characterisation of mussel (Mytilus sp.) populations in Western Australia and evaluation of potential genetic impacts of mussel spat translocation from interstate. Marine and Freshwater Research, 65(6), 486–496. Elliott, M., & Whitfield, A. K. (2011). Challenging paradigms in estuarine ecology and management. Estuarine, Coastal and Shelf Science, 94(4), 306–314. Elliott, M., Whitfield, A. K., Potter, I. C., Blaber, S. J. M., Cyrus, D. P., Nordlie, F. G., & Harrison, T. D. (2007). The guild approach to categorizing estuarine fish assemblages: A global review. Fish and Fisheries, 8(3), 241–268. Florisson, J. H., Tweedley, J. R., Walker, T. H. E., & Chaplin, J. A. (2018). Reef vision: A citizen science program for monitoring the fish faunas of artificial reefs. Fisheries Research, 206, 296–308. Folpp, H., Lowry, M., Gregson, M., & Suthers, I. M. (2013). Fish assemblages on estuarine artificial reefs: natural rocky-reef mimics or discrete assemblages? PLoS ONE, 8(6), e63505. Ford, J. R., & Hamer, P. (2016). The forgotten shellfish reefs of coastal Victoria: Documenting the loss of a marine ecosystem over 200 years since European settlement. Proceedings of the Royal Society of Victoria, 128(1), 87–105. Gabriele, M., Putrone, V., & Brunetti, R. (1997). Morphometrics and energetic value of Adriatic ascidians. Cahiers de Biologie Marine, 38(3), 169–174. Galimany, E., Wikfors, G. H., Dixon, M. S., Newell, C. R., Meseck, S. L., Henning, D., Li, Y., & Rose, J. M. (2017). Cultivation of the Ribbed Mussel (Geukensia demissa) for nutrient bioextraction in an urban estuary. Environmental Science and Technology, 51(22), 13311–13318. Gardner, J. P. A. (2004). A historical perspective of the genus Mytilus (Bivalvia: Mollusca) in New Zealand: Multivariate morphometric analyses of fossil, midden and contemporary blue mussels. Biological Journal of the Linnean Society, 82(3), 329– 344. Gardner, J. P. A., & Westfall, K. M. (2012). Geographic distribution and molecular identification of a metapopulation of blue mussels (genus Mytilus) in north- eastern New Zealand. Journal of Molluscan Studies, 78(1), 66–73.
101 Geraldi, N. R., Simpson, M., Fegley, S. R., Holmlund, P., & Peterson, C. H. (2013). Addition of juvenile oysters fails to enhance oyster reef development in Pamlico Sound. Marine Ecology Progress Series, 480, 119–129. Gilby, B. L., Olds, A. D., Connolly, R. M., Maxwell, P. S., Henderson, C. J., & Schlacher, T. A. (2018). Seagrass meadows shape fish assemblages across estuarine seascapes. Marine Ecology Progress Series, 588, 179–189. Gilby, B. L., Olds, A. D., Duncan, C. K., Ortodossi, N. L., Henderson, C. J., & Schlacher, T. A. (2020). Identifying restoration hotspots that deliver multiple ecological benefits. Restoration Ecology, 28, 222–232. Gilby, B. L., Olds, A. D., Henderson, C. J., Ortodossi, N. L., Connolly, R. M., & Schlacher, T. A. (2019). Seascape context modifies how fish respond to restored oyster reef structures. ICES Journal of Marine Science, 76, 1131–1139. Gillies, C. L., Castine, S. A., Alleway, H. K., Crawford, C., Fitzsimons, J. A., Hancock, B., Koch, P., McAfee, D., McLeod, I. M., & zu Ermgassen, P. S. E. (2020). Conservation status of the oyster reef ecosystem of southern and eastern Australia. Global Ecology and Conservation, 22, e00988. Gillies, C. L., Crawford, C., & Hancock, B. (2017). Restoring Angasi oyster reefs: What is the endpoint ecosystem we are aiming for and how do we get there? Ecological Management and Restoration, 18(3), 214–222. Gillies, C. L., McLeod, I. M., Alleway, H. K., Cook, P., Crawford, C., Creighton, C., Diggles, B., Ford, J., Hamer, P., Heller-Wagner, G., Lebrault, E., Le Port, A., Russell, K., Sheaves, M., & Warnock, B. (2018). Australian shellfish ecosystems: Past distribution, current status and future direction. PLoS ONE, 13(2), e0190914. Grabowski, J. H., Brumbaugh, R. D., Conrad, R. F., Keeler, A. G., James, J., Peterson, C. H., Piehler, M. F., Sean, P., & Smyth, A. R. (2012). Economic valuation of ecosystem services provided by oyster reefs. BioScience, 62(10), 900–909. Grabowski, J. H., Brumbaugh, R. D., Conrad, R. F., Keeler, A. G., Opaluch, J. J., Peterson, C. H., Piehler, M. F., Powers, S. P., & Smyth, A. R. (2012). Economic valuation of ecosystem services provided by oyster reefs. BioScience, 62(10), 900–909. Grant, S., & Cherry, M. (1985). Mytilus galloprovincialis Lamarck. in Southern Africa. Journal of Experimental Marine Biology and Ecology, 90(2), 179–191. Hallett, C. S., Valesini, F. J., Clarke, K. R., Hesp, S. A. & Hoeksema, S. D. (2012). Development and validation of fish-based, multimetric indices for assessing the ecological health of Western Australian estuaries. Estuarine, Coastal and Shelf Science, 104–105, 102-113.
102 Hallett, C. S., Hobday, A. J., Tweedley, J. R., Thompson, P. A., McMahon, K., & Valesini, F. J. (2018). Observed and predicted impacts of climate change on the estuaries of south-western Australia, a Mediterranean climate region. Regional Environmental Change, 18(5), 1357–1373. Hallett, C. S., Trayler, K. M. & Valesini, F. J. (2019). The Fish Community Index: A practical management tool for monitoring and reporting estuarine ecological condition. Integrated Environmental Assessment and Management, 15, 726-738. Harrison, T. D., & Whitfield, A. K. (2006). Estuarine typology and the structuring of fish communities in South Africa. Environmental Biology of Fishes, 75(3), 269–293.
Harrison, T. D., & Whitfield, A. K. (2008). Geographical and typological changes in fish guilds of South African estuaries. Journal of Fish Biology, 73(10), 2542–2570. Heino, M., & Kaitala, V. (1999). Evolution of resource allocation between growth and reproduction in animals with intermediate growth. Journal of Evolutionary Biology, 12(3), 121–137. Hellyer, C. B., Harasti, D., & Poore, A. G. B. (2011). Manipulating artificial habitats to benefit seahorses in Sydney Harbour, Australia. Aquatic Conservation: Marine and Freshwater Ecosystems, 21(6), 582–589.
Henderson, C. J., Gilby, B. L., Schlacher, T. A., Connolly, R. M., Sheaves, M., Maxwell, P. S., Flint, N., Borland, H. P., Martin, T. S. H., Gorissen, B., & Olds, A. D. (2020). Landscape transformation alters functional diversity in coastal seascapes. Ecography, 43(1), 138–148. Herlyn, M., & Millat, G. (2000). Decline of the intertidal blue mussel (Mytilus edulis) stock at the coast of Lower Saxony (Wadden Sea) and influence of mussel fishery on the development of young mussel beds. Hydrobiologia, 426, 203–210. Hickman, R. W., & Illingworth, J. (1980). Condition cycle of the green-lipped mussel Perna canaliculus in New Zealand. Marine Biology, 60(1), 27–38. Hindell, J. S., Jenkins, G. P., & Womersley, B. (2008). Habitat utilisation and movement of black bream Acanthopagrus butcheri (Sparidae) in an Australian estuary. Marine Ecology Progress Series, 366, 219–229.
Hines, A. H., Johnson, E. G., Young, A. C., Aguilar, R., Kramer, M. A., Goodison, M., Zmora, O., & Zohar, Y. (2008). Release strategies for estuarine species with complex migratory life cycles: Stock enhancement of Chesapeake blue crabs (Callinectes sapidus). Reviews in Fisheries Science, 16(1–3), 175–185. Hines, A. H., Whitlatch, R. B., Thrush, S. F., Hewitt, J. E., Cummings, V. J., Dayton, P. K. & Legendre, P. (1997). Nonlinear foraging response of a large marine predator to
103 benthic prey: eagle ray pits and bivalves in a New Zealand sandflat. Journal of Experimental Marine Biology and Ecology, 216, 191-210. Hodgkin, E. P., & Hesp, P. (1998). Estuaries to salt lakes: Holocene transformation of the estuarine ecosystems of south-western Australia. Marine and Freshwater Research, 49(3), 183–201. Hoeksema, S. D. & Potter, I. C. (2006). Diel, seasonal, regional and annual variations in the characteristics of the ichthyofauna of the upper reaches of a large Australian microtidal estuary. Estuarine, Coastal and Shelf Science 67, 503-520. Hogan-West, K., Tweedley, J. R., Coulson, P. G., Poh, B., & Loneragan, N. R. (2019). Abundance and distribution of the non-indigenous Acentrogobius pflaumii and native gobiids in a temperate Australian estuary. Estuaries and Coasts, 42(6), 1612–1631. Hope, J. H., Lampert, R. J., Edmondson, E., Smith, M. J., & van Tets, G. F. (1977). Late Pleistocene faunal remains from Seton Rock shelter, Kangaroo Island, South Australia. Journal of Biogeography, 4(4), 363-385. Humphries, A. T., & Peyre, M. K. L. (2015). Oyster reef restoration supports increased nekton biomass and potential commercial fishery value. PeerJ, 3, e1111. Jackson, J. B. C., Kirby, M. X., Berger, W. H., Bjorndal, K. A., Botsford, L. W., Bourque, B. J., Bradbury, R. H., Cooke, R., Erlandson, J., Estes, J. A., Hughes, T. P., Kidwell, S., Lange, C. B., Lenihan, H. S., Pandolfi, J. M., Peterson, C. H., Steneck, R. S., Tegner, M. J., & Warner, R. R. (2001). Historical overfishing and the recent collapse of coastal ecosystems. Science, 293(5530), 629–639. Jenkins, G. P., & Wheatley, M. J. (1998). The influence of habitat structure on nearshore fish assemblages in a southern Australian embayment: Comparison of shallow seagrass, reef-algal and unvegetated sand habitats, with emphasis on their importance to recruitment. Journal of Experimental Marine Biology and Ecology, 221(2), 147–172.
Jokela, J., & Mutikainen, P. (1995). Phenotypic plasticity and priority rules for energy allocation in a freshwater clam: A field experiment. Oecologia, 104(1), 122–132. Justic, D., Rabalais, N. N., Turner, R. E., & Dı, R. J. (2009). Global change and eutrophication of coastal waters. ICES Journal of Marine Science, 66(7), 1528– 1537. Kanandjembo, A. N., Platell, M. E., & Potter, I. C. (2001). The benthic macroinvertebrate community of the upper reaches of an Australian estuary that undergoes marked seasonal changes in hydrology. Hydrological Processes, 15(3), 2481–2501. Karayücel, S., Çelik, M. Y., Karayücel, I., Öztürk, R., & Eyüboğlu, B. (2015). Effects of stocking density on survival, growth and biochemical composition of cultured
104 mussels (Mytilus galloprovincialis, Lamarck 1819) from an offshore submerged longline system. Aquaculture Research, 46(6), 1369–1383. Kennish, M. J. (2002). Environmental threats and environmental future of estuaries. Environmental Conservation, 29(1), 78–107. Kincaid, E. S., & Rivera, C. E. D. (2020). Predators associated with marinas consume indigenous over non-indigenous ascidians. Estuaries and Coasts. 1-10. Kovačić, I., Pavičić-Hamer, D., Kanduč, T., & Hamer, B. (2018). Adaptation of cultured mussel Mytilus galloprovincialis Lamarck, 1819 from the northern Adriatic Sea to nearby aquaculture sites and translocation. Acta Adriatica, 58(2), 285–296. La Peyre, M., Furlong, J., Brown, L. A., Piazza, B. P., & Brown, K. (2014). Oyster reef restoration in the northern Gulf of Mexico: Extent, methods and outcomes. Ocean and Coastal Management, 89, 20–28. Lek, E., Fairclough, D. V., Platell, M. E., Clarke, K. R., Tweedley, J. R., & Potter, I. C. (2011). To what extent are the dietary compositions of three abundant, co-occurring labrid species different and related to latitude, habitat, body size and season? Journal of Fish Biology, 78(7), 1913–1943. Lins, D., & Rocha, R. (2020). Cultivated brown mussel (Perna perna) size is reduced through the impact of three invasive fouling species in southern Brazil. Aquatic Invasions, 15(1), 114–126. Lloret, J., Shulman, G., & Love, R. M. (2014). Condition and health indicators of exploited marine fishes. Oxford: Wiley Blackwell. Lodeiros, C. J. M., & Himmelman, J. H. (1996). Influence of fouling on the growth and survival of the tropical scallop, Euvola (Pecten) ziczac (Linnaeus 1758) in suspended culture. Aquaculture Research, 27(10), 749–756. Lök, A., Acarli, S., Serdar, S., Köse, A., & Yildiz, H. (2007). Growth and mortality of Mediterranean mussel Mytilus galloprovincialis Lamarck. 1819, in relation to size on longline in Mersin Bay, Izmir (Turkey - Aegean Sea). Aquaculture Research, 38(8), 819–826. Loneragen, N., Taylor, M. D., & Tweedley, J.R. (2019). A drop in the ocean: marine fish releases in Australia. Austral Ecology. Lorenzen, K., & Enberg, K. (2002). Density-dependent growth as a key mechanism in the regulation of fish populations: Evidence from among-population comparisons. Proceedings of the Royal Society B: Biological Sciences, 269(1486), 49–54. Lowry, M. B., Glasby, T. M., Boys, C. A., Folpp, H., Suthers, I., & Gregson, M. (2014). Response of fish communities to the deployment of estuarine artificial reefs for fisheries enhancement. Fisheries Management and Ecology, 21(1), 42–56.
105 Lutz-Collins, V., Ramsay, A., Quijón, P. A., & Davidson, J. (2009). Invasive tunicates fouling mussel lines: Evidence of their impact on native tunicates and other epifaunal invertebrates. Aquatic Invasions, 4(1), 213–220. Ly, W., Huang, Y., Liu, Z., Yang, Y., Fan, B., & Zhao, Y. (2016). Application of macrobenthic diversity to estimate ecological health of artificial oyster reef in Yangtze Estuary, China. Marine Pollution Bulletin, 103(1–2), 137–143. Lynch, S. A., Darmody, G., O’Dwyer, K., Gallagher, M. C., Nolan, S., McAllen, R., & Culloty, S. C. (2016). Biology of the invasive ascidian Ascidiella aspersa in its native habitat: Reproductive patterns and parasite load. Estuarine, Coastal and Shelf Science, 181, 249–255. Maschette, D., Fromont, J., Platell, M. E., Coulson, P. G., Tweedley, J. R., & Potter, I. C. (2020). Characteristics and implications of spongivory in the Knifejaw Oplegnathus woodwardi (Waite) in temperate mesophotic waters. Journal of Sea Research, 157, e101847. Mastrototaro, F., D’Onghia, G., & Tursi, A. (2008). Spatial and seasonal distribution of ascidians in a semi-enclosed basin of the Mediterranean Sea. Journal of the Marine Biological Association of the United Kingdom, 88(5), 1053–1061. McDonald, J. H., Seed, R., & Koehn, R. K. (1991). Allozymes and morphometric characters of three species of Mytilus in the northern and southern hemispheres. Marine Biology, 111(3), 323–333. McLeod, I. M., Boström-Einarsson, L., Creighton, C., D’Anastasi, B., Diggles, B., Dwyer, P. G., Firby, L., Le Port, A., Luongo, A., Martínez-Baena, F., McOrrie, S., Heller- Wagner, G., & Gillies, C. L. (2020). Habitat value of Sydney rock oyster (Saccostrea glomerata) reefs on soft sediments. Marine and Freshwater Research, 71(7), 771– 781. McLusky, D. S., & Elliott, M. (2004). The estuarine ecosystem: Ecology, threats and management. OUP Oxford. McLusky, D. S., & Elliott, M. (2007). Transitional waters: A new approach, semantics or just muddying the waters? Estuarine, Coastal and Shelf Science, 71(3-4), 359–363. Meyer, D. L., & Townsend, E. C. (2000). Faunal utilization of created intertidal eastern oyster (Crassostrea virginica) reefs in the south-eastern United States. Estuaries, 23(1), 34–45. Monbet, Y. (1992). Control of phytoplankton biomass in estuaries: A comparative analysis of microtidal and macrotidal estuaries. Estuaries, 15(4), 563–571. Moseman, S. M., Levin, L. A., Currin, C., & Forder, C. (2004). Colonization, succession, and nutrition of macrobenthic assemblages in a restored wetland at Tijuana Estuary, California. Estuarine, Coastal and Shelf Science, 60(4), 755–770.
106 Müller, M. N., Dorantes-Aranda, J. J., Seger, A., Botana, M. T., Brandini, F. P., & Hallegraeff, G. M. (2019). Ichthyotoxicity of the dinoflagellate Karlodinium veneficum in response to changes in seawater pH. Frontiers in Marine Science, 6, 1–6. Myrand, B., Guderley, H., & Himmelman, J. H. (2000). Reproduction and summer mortality of blue mussels Mytilus edulis in the Magdalen Islands, southern Gulf of St. Lawrence. Marine Ecology Progress Series, 197, 193–207. Newell, C. R., Short, F., Hoven, H., Healey, L., Panchang, V., & Cheng, G. (2010). The dispersal dynamics of juvenile plantigrade mussels (Mytilus edulis Lamarck) from eelgrass (Zostera marina) meadows in Maine, USA. Journal of Experimental Marine Biology and Ecology, 394(1–2), 45–52. Newell, R. I. E., Fisher, T. R., Holyoke, R. R., & Cornwell, J. C. (2005). Influence of eastern oysters on nitrogen and phosphorus regeneration in Chesapeake Bay, USA. In The Comparative Roles of Suspension-Feeders in Ecosystems (pp. 93-120). Springer, Dordrecht. Obregón, C., Tweedley, J. R., Loneragan, N. R., & Hughes, M. (2019). Different but not opposed: Perceptions between fishing sectors on the status and management of a crab fishery. ICES Journal of Marine Science. Okaniwa, N., Miyaji, T., Sasaki, T., & Tanabe, K. (2010). Shell growth and reproductive cycle of the mediterranean mussel Mytilus galloprovincialis in Tokyo Bay, Japan: Relationship with environmental conditions. Plankton and Benthos Research, 5, 214–220. Parry, G. (1981). The meaning of r- and K-selection. Oecologia, 48(2), 260–264. Peharda, M., Župan, I., Bavčević, L., Frankić, A., & Klanjšček, T. (2007). Growth and condition index of mussel Mytilus galloprovincialis in experimental integrated aquaculture. Aquaculture Research, 38(16), 1714–1720. Peterson, C. H., Grabowski, J. H., & Powers, S. P. (2003). Estimated enhancement of fish production resulting from restoring oyster reef habitat: Quantitative valuation. Marine Ecology Progress Series, 264, 249–264. Petersen, J. K., & Riisgard, H. U. (1992). Filtration capacity of the ascidian Ciona intestinalis and its grazing impact in a shallow fjord. Marine Ecology Progress Series, 88, 9–17. Pineda, M. C., López-Legentil, S., & Turon, X. (2013). Year-round reproduction in a seasonal sea: Biological cycle of the introduced ascidian Styela plicata in the Western Mediterranean. Marine Biology, 160(1), 221–230. Pinheiro, M., Caetano, M., Neuparth, T., Barros, S., Soares, J., Raimundo, J., Vale, C., Coimbra, J., Castro, L. F. C., & Santos, M. M. (2019). Ecotoxicology of deep-sea
107 environments: Functional and biochemical effects of suspended sediments in the model species Mytilus galloprovincialis under hyperbaric conditions. Science of the Total Environment, 670, 218–225. Pizzolon, M., Cenci, E., & Mazzoldi, C. (2008). The onset of fish colonization in a coastal defence structure (Chioggia, Northern Adriatic Sea). Estuarine, Coastal and Shelf Science, 78(1), 166–178. Place, A. R., Bowers, H. A., Bachvaroff, T. R., Adolf, J. E., Deeds, J. R., & Sheng, J. (2012). Karlodinium veneficum-The little dinoflagellate with a big bite. Harmful Algae, 14, 179–195. Platell, M. E., & Potter, I. C. (1996). Influence of water depth, season, habitat and estuary location on the macrobenthic fauna of a seasonally closed estuary. Journal of the Marine Biological Association of the United Kingdom, 76(1), 1–21. Poh, B., Tweedley, J. R., Chaplin, J. A., Trayler, K. M., & Loneragan, N. R. (2018). Estimating predation rates of restocked individuals: The influence of timing of release on metapenaeid survival. Fisheries Research, 198, 165–179. Poh, B., Tweedley, J.R., Chaplin, J. A., Trayler, K. M., Crisp, J. A., & Loneragan, N. R. (2019). Influence of physico-chemical and biotic factors on the distribution of a penaeid in a temperate estuary. Estuarine, Coastal and Shelf Science, 218, 70–85. Potter, I. C., Beckley, L. E., Whitfield, A. K. & Lenanton, R. C. (1990). Comparisons between the roles played by estuaries in the life cycles of fishes in temperate Western Australia and Southern Africa. Environmental Biology of Fishes 28, 143- 178. Potter, I. C., & Hyndes, G. A. (1999). Characteristics of the ichthyofaunas of southwestern Australian estuaries, including comparisons with Holarctic estuaries and estuaries elsewhere in temperate Australia: A review. Australian Journal of Ecology, 24(4), 395–421. Potter, I. C., Tweedley, J. R., Elliott, M., & Whitfield, A. K. (2015a). The ways in which fish use estuaries: A refinement and expansion of the guild approach. Fish and Fisheries, 16(2), 230–239. Potter, I. C., Warwick, R. M., Hall, N. G., & Tweedley, J. R. (2015b). The physico-chemical characteristics, biota and fisheries of estuaries. Freshwater Fisheries Ecology, 48- 79. Potter, I. C., Veale, L., Tweedley, J. R., & Clarke, K. R. (2016). Decadal changes in the ichthyofauna of a eutrophic estuary following a remedial engineering modification and subsequent environmental shifts. Estuarine, Coastal and Shelf Science, 181, 345–363.
108 Prieto, D., Tamayo, D., Urrutxurtu, I., Navarro, E., Ibarrola, I., & Urrutia, M. B. (2020). Nature more than nurture affects the growth rate of mussels. Scientific Reports, 10(1), 1–13. Prieto, D., Urrutxurtu, I., Navarro, E., Urrutia, M. B., & Ibarrola, I. (2018). Mytilus galloprovincialis fast growing phenotypes under different restrictive feeding conditions: Fast feeders and energy savers. Marine Environmental Research, 140, 114–125. Quan, W. M., Humphries, A. T., Shi, L. Y., & Chen, Y. Q. (2012). Determination of trophic transfer at a created intertidal oyster (Crassostrea ariakensis) reef in the Yangtze River Estuary using stable isotope analyses. Estuaries and Coasts, 35(1), 109–120. Reguero, B. G., Beck, M. W., Bresch, D. N., Calil, J., & Meliane, I. (2018). Comparing the cost effectiveness of nature-based and coastal adaptation: A case study from the Gulf Coast of the United States. PLoS ONE, 13(4), 1–24. Riisgård, H. U., Larsen, P. S., & Pleissner, D. (2014). Allometric equations for maximum filtration rate in blue mussels Mytilus edulis and importance of condition index. Helgoland Marine Research, 68(1), 193–198. Rius, M., Turon, X., Dias, G. M., & Marshall, D. J. (2010). Propagule size effects across multiple life-history stages in a marine invertebrate. Functional Ecology, 24(3), 685–693. Rodney, W. S., & Paynter, K. T. (2006). Comparisons of macrofaunal assemblages on restored and non-restored oyster reefs in mesohaline regions of Chesapeake Bay in Maryland. Journal of Experimental Marine Biology and Ecology, 335(1), 39–51. Rose, T. H., Tweedley, J. R., Warwick, R. M. & Potter, I. C. (2019). Zooplankton dynamics in a highly eutrophic microtidal estuary. Marine Pollution Bulletin, 142, 433-451. Rose, T. H., Tweedley, J. R., Warwick, R. M. & Potter, I. C. (2020). Influences of microtidal regime and eutrophication on estuarine zooplankton. Estuarine, Coastal and Shelf Science, 238, 106689. Rothschild, B. J., Ault, J. S., & Goulletquer, P. (1991). The decline of Chesapeake Bay oyster populations: A centuary of habitat destruction and overfishing. International Council for the Exploration of the Sea, 91–152. Sarre, G. A., Platell, M. E., & Potter, I. C. (2000). Do the dietary compositions of Acanthopagrus butcheri in four estuaries and a coastal lake vary with body size and season and within and amongst these water bodies? Journal of Fish Biology, 56(1), 103–122. Schelske, C. L., & Odum, E. P. (1961). Mechanisms maintaining high productivity in Georgia estuaries. Proceedings of the Gulf and Caribbean Fisheries Institute, 14, 75-80.
109 Schlacher, T. A., & Wooldridge, T. H. (1996). How sieve mesh size affects sample estimates of estuarine benthic macrofauna. Journal of Experimental Marine Biology and Ecology, 201(1–2), 159–171. Schotanus, J., Capelle, J. J., Paree, E., Fivash, G. S., van de Koppel, J., & Bouma, T. J. (2020). Restoring mussel beds in highly dynamic environments by lowering environmental stressors. Restoration Ecology. https://doi.org/10.1111/rec.13168. Schulte, M., Burke, R., & Lipcius, R. (2010). Unprecedented restoration of a native oyster metapopulation. Science, 325(5944), 1124–1128. Scyphers, S. B., Powers, S. P., Heck, K. L., & Byron, D. (2011). Oyster reefs as natural breakwaters mitigate shoreline loss and facilitate fisheries. PLoS ONE, 6(8), 1–12. Seed, R. (1969). The ecology of Mytilus edulis (Lamellibranchiata) on exposed rocky shores. Oecologia, 3(3-4), 317–350. Selig, E. R., Hole, D. G., Allison, E. H., Arkema, K. K., McKinnon, M. C., Chu, J., de Sherbinin, A., Fisher, B., Glew, L., Holland, M. B., Ingram, J. C., Rao, N. S., Russell, R. B., Srebotnjak, T., Teh, C. L., Troëng, S., Turner, W. R., & Zvoleff, A. (2019). Mapping global human dependence on marine ecosystems. Conservation Letters, 12, 1–10. Sharma, S., Goff, J., Moody, R. M., Byron, D., Heck, K. L., Powers, S. P., Ferraro, C., & Cebrian, J. (2016). Do restored oyster reefs benefit seagrasses? An experimental study in the Northern Gulf of Mexico. Restoration Ecology, 24(3), 306–313. Simpson, E. H. (1949). Measurement of diversity. Nature, 163(4148), 688-688. Smaal, A. C., Verhagen, J. H. G., Coosen, J., & Haas, H. A. (1986). Interaction between seston quantity and quality and benthic suspension feeders in the Oosterschelde, the Netherlands. Ophelia, 26(1), 385–399. Smart, D. C., Heenan, S., Lokman, P. M., & Lamare, M. D. (2020). Seasonal reproduction of the blue mussel (Mytilus galloprovincialis) from two locations in southern New Zealand. New Zealand Journal of Marine and Freshwater Research, 1–15. Smith, J. R., Fong, P., & Ambrose, R. F. (2006). Long-term change in mussel (Mytilus californianus Conrad) populations along the wave-exposed coast of southern California. Marine Biology, 149(3), 537–545. Sogard, S. M. (1992). Variability in growth rates of juvenile fishes in different estuarine habitats. Marine Ecology Progress Series, 85(1–2), 35–53. Somerfield, P. J., Dashfield, S. L., & Warwick, R. M. (2018). The structure and organisation of integral marine benthic communities in relation to sieve mesh size. Journal of Experimental Marine Biology and Ecology, 502, 164–173.
110 Su, Z., Yan, Y., & Huang, L. (2007). Effect of fouling on feeding, oxygen consumption and waste excretion of pearl oyster Pinctada martensii in Daya Bay cultivation. Marine Science Bulletin, 9, 34–42. Tarjuelo, I., López-Legentil, S., Codina, M., & Turon, X. (2002). Defence mechanisms of adults and larvae of colonial ascidians: Patterns of palatability and toxicity. Marine Ecology Progress Series, 235, 103–115. Taylor, J. J., Southgate, P. C., & Rose, R. A. (1997). Fouling animals and their effect on the growth of silver-lip pearl oysters, Pinctada maxima (Jameson) in suspended culture. Aquaculture, 153(1-2), 31–40. Taylor, M. D., Baker, J., & Suthers, I. M. (2013). Tidal currents, sampling effort and baited remote underwater video (BRUV) surveys: Are we drawing the right conclusions? Fisheries Research, 140, 96–104. Taylor, P., & Whitfield, A. K. (2010). Mass mortalities of fish in southern African estuaries. South African Journal of Aquatic Sciences, 21(1–2), 29–34. Teo, S. L. M., & Ryland, J. S. (1994). Toxicity and palatability of some British ascidians. Marine Biology, 120(2), 297–303. Then, A. Y., Hoenig, J. M., Hall, N., & Hewitt, D. (2015). Evaluating the predictive performance of empirical estimators of natural mortality rate using information on over 200 fish species. ICES Journal of Marine Science, 72(1), 82–92. Thompson, R. J. (1979). Fecundity and reproductive effort in the blue mussel (Mytilus edulis), the sea urchin (Strongylocentrotus droebachiensis), and the snow crab (Chionoecetes opilio) from populations in Nova Scotia and Newfoundland. Journal of the Fisheries Research Board of Canada, 36(8), 955–964. Tolley, S. G., & Volety, A. K. (2005). The role of oysters in habitat use of oyster reefs by resident fishes and decapod crustaceans. Journal of Shellfish Research, 24(4), 1007–1012. Tremblay, R., Myrand, B., & Guderley, H. (1998). Temporal variation of lysosomal capacities in relation to susceptibility of mussels, Mytilus edulis, to summer mortality. Marine Biology, 132(4), 641–649. Tweedley, J. R., Warwick, R. M., Valesini, F. J., Platell, M. E. & Potter, I. C. (2012). The use of benthic macroinvertebrates to establish a benchmark for evaluating the environmental quality of microtidal, temperate southern hemisphere estuaries. Marine Pollution Bulletin, 64, 1210-1221. Tweedley, J. R., Warwick, R. M., Clarke, K. R. & Potter, I. C. (2014). Family-level AMBI is valid for use in the north-eastern Atlantic but not for assessing the health of microtidal Australian estuaries. Estuarine, Coastal and Shelf Science, 141, 85-96.
111 Tweedley, J. R., Warwick, R. M., & Potter, I. C. (2015). Can biotic indicators distinguish between natural and anthropogenic environmental stress in estuaries? Journal of Sea Research, 102, 10–21. Tweedley, J R., Florisson, J. H., Bateman, T. A., Chaplin, J. A., & Gill, H. (2016a). Recreational Fishing Initiatives Fund Final Report: Can recreational fishers provide a cost-effective means for monitoring artificial reefs? Tweedley, J. R., Hallett, C. S., Warwick, R. M., Clarke, K. R., & Potter, I. C. (2016b). The hypoxia that developed in a microtidal estuary following an extreme storm produced dramatic changes in the benthos. Marine and Freshwater Research, 67(3), 327–341. Tweedley, J. R, Warwick, R. M., & Potter, I. C. (2016c). The contrasting ecology of temperate macrotidal and microtidal estuaries. Oceanography and Marine Biology: An Annual Review, 54, 73-171. Tweedley, J. R., Loneragan, N. R., Crisp, J. A., Poh, B., Broadley, A., Bennet, A., Hodson, K., Trayler, K., Jenkins, G. & Chaplin, J. (2017a). Restocking of the Western School Prawn (Metapenaeus dalli) in the Swan Canning Riverpark. Tweedley, J. R., Warwick, R. M., Hallett, C. S., & Potter, I. C. (2017b). Fish-based indicators of estuarine condition that do not require reference data. Estuarine, Coastal and Shelf Science, 191, 209–220. Valesini, F. J., Coen, N. J., Wildsmith, M. D., Hourston, M., Tweedley, J. R., Hallett, C. S., Linke, T. E., & Potter, I. C. (2009). Relationships between fish faunas and habitat type in south-western Australian estuaries. Fisheries Research and Development Corporation Final Report, July 2009. Valesini, F. J., Tweedley, J. R., Clarke, K. R., & Potter, I. C. (2014). The importance of regional, system-wide and local spatial scales in structuring temperate estuarine fish communities. Estuaries and Coasts, 37(3), 525–547. Valesini, F. J., Wildsmith, M. D., & Tweedley, J. R. (2018). Predicting estuarine faunal assemblages using enduring environmental surrogates, with applications in systematic conservation planning. Ocean and Coastal Management, 165, 80–98. Vasconcelos, R. P., Reis-Santos, P., Maia, A., Fonseca, V., França, S., Wouters, N., Costa, M. J., & Cabral, H. N. (2010). Nursery use patterns of commercially important marine fish species in estuarine systems along the Portuguese coast. Estuarine, Coastal and Shelf Science, 86(4), 613–624. Veale, L. J., Coulson, P. G., Hall, N. G., & Potter, I. C. (2016). Biology of a marine estuarine- opportunist fish species in a microtidal estuary, including comparisons among decades and with coastal waters. Marine and Freshwater Research, 67(8), 1128– 1140.
112 Veale, L.J., Tweedley, J. R., Clarke, K. R., & Hallett, C. S. (2014). Characteristics of the ichthyofauna of a temperate microtidal estuary with a reverse salinity gradient, including inter-decadal comparisons. Journal of Fish Biology, 85(5), 1320–1354. Veiga, P., Ramos-Oliveira, C., Sampaio, L., & Rubal, M. (2020). The role of urbanisation in affecting Mytilus galloprovincialis. PLoS ONE, 15(5), e0232797. Villalba, A. (2005). Gametogenic cycle of cultured mussel, Mytilus galloprovincialis, in the bays of Galicia (N.W. Spain). Aquaculture, 51(49), 269–277. Waltham, N. J., Elliott, M., Lee, S. Y., Lovelock, C., Duarte, C. M., Buelow, C., Simenstad, C., & Nagelkerken, I. (2020). UN decade on ecosystem restoration 2021 - 2030 - What Chance for Success in Restoring Coastal Ecosystems? Frontiers in Marine Science, 7, 71. Warwick, R. M., Tweedley, J. R., & Potter, I. C. (2018). Microtidal estuaries warrant special management measures that recognise their critical vulnerability to pollution and climate change. Marine Pollution Bulletin, 135, 41–46. Whitfield, A. K. (1999). Ichthyofaunal assemblages in estuaries: A South African case study. Reviews in Fish Biology and Fisheries, 9(2), 151–186. Whitfield, A. K., Elliott, M., Basset, A., Blaber, S. J. M., & West, R. J. (2012). Paradigms in estuarine ecology - A review of the Remane diagram with a suggested revised model for estuaries. Estuarine, Coastal and Shelf Science, 97, 78–90. Whitfield, A. K. (2017). The role of seagrass meadows, mangrove forests, salt marshes and reed beds as nursery areas and food sources for fishes in estuaries. Reviews in Fish Biology and Fisheries, 27(1), 75–110. Whitfield, A. K., Taylor, R. H., Fox, C., & Cyrus, D. P. (2006). Fishes and salinities in the St Lucia estuarine system — A review. Reviews in Fish Biology and Fisheries, 16(1), 1– 20. Whitmarsh, S. K., Fairweather, P. G., & Huveneers, C. (2017). What is big BRUVver up to? Methods and uses of baited underwater video. Reviews in Fish Biology and Fisheries, 27(1), 53–73. Wilcox, M., Kelly, S., & Jeffs, A. (2018). Ecological restoration of mussel beds onto soft sediment using transplanted adults. Restoration Ecology, 26(3), 581–590. Wildsmith, M. D., Rose, T. H., Potter, I. C., Warwick, R. M., & Clarke, K. R. (2011). Benthic macroinvertebrates as indicators of environmental deterioration in a large microtidal estuary. Marine Pollution Bulletin, 62(3), 525–538. Wildsmith, M. D., Valesini, F. J., & Robinson, S. F. (2017). The value of enduring environmental surrogates as predictors of estuarine benthic macroinvertebrate assemblages. Estuarine, Coastal and Shelf Science, 197, 159–172.
113 Wilson, J. G. (2002). Productivity, fisheries and aquaculture in temperate estuaries. Estuarine, Coastal and Shelf Science, 55(6), 953–967. Wu, Z., Tweedley, J. R., Loneragan, N. R., & Zhang, X. (2019). Artificial reefs can mimic natural habitats for fish and macroinvertebrates in temperate coastal waters of the Yellow Sea. Ecological Engineering, 139, 105579. Yanick, J. F., Heath, J. W., & Heath, D. D. (2003). Survival and growth of local and transplanted blue mussels (Mytilus trossulus, Lamarck). Aquaculture Research, 34, 869–875. Zardi, G. I., Nicastro, K. R., McQuaid, C. D., & Erlandsson, J. (2008). Sand and wave induced mortality in invasive (Mytilus galloprovincialis) and indigenous (Perna perna) mussels. Marine Biology, 153(5), 853–858. Ziegler, S. L., Grabowski, J. H., Baillie, C. J., & Fodrie, F. J. (2018). Effects of landscape setting on oyster reef structure and function largely persist more than a decade post-restoration. Restoration Ecology, 26(5), 933–942.
114