Biological characteristics of Acanthopagrus butcheri. Changes following
environmental degradation and comparisons of restocked and wild fish.
Alan Cottingham BSc (Hons)
This thesis is presented for the degree of Doctor of Philosophy
2015
Murdoch University, Western Australia
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.
Alan Cottingham
ii
It is not the strongest of the species that survive, nor the most intelligent, but rather
the one most adaptable to change. Clarence Darrow 1925.
iii
The subject of this thesis, the Black Bream Acanthopagrus butcheri.
iv
Abstract
The Black Bream Acanthopagrus butcheri, which is the subject of this thesis, completes its life cycle within estuaries and has particularly plastic biological characteristics. The overarching aims of the thesis were 1) to determine the ways in which A. butcheri has responded to marked environmental changes in the Swan River Estuary and 2) to track the biological performance of restocked A. butcheri in the Blackwood River Estuary in which the stock of this species had become depleted.
Between the early 1990s and mid-2000s, the deeper waters of the upper Swan River
Estuary became increasingly hypoxic, due to a reduction in the flushing of nutrients and organic material from the estuary as a result of declining rainfall and thus freshwater discharge. Over that same time period, the condition, growth and length at maturity of female and male A. butcheri declined and the age at maturity increased. Furthermore, catch data implied that, during that period, the larger A. butcheri showed a marked tendency to move from the deeper hypoxic waters into the shallows, occupied by smaller A. butcheri.
This habitat compression led to greatly increased densities in the shallows. It is therefore proposed that the detrimental effects of hypoxia and high densities led to the above retrograde changes in the biological characteristics of A. butcheri.
The development of a year-effect growth model demonstrated that the growth of
A. butcheri in the Swan River Estuary also differed substantially among years between
2007/08 and 2013/14. Annual growth of one year old A. butcheri in those seven years were positively correlated with temperature during the main growing period, but not with freshwater discharges in the preceding wet ‘winter’ months. That positive correlation with temperature, which is consistent with the metabolic theory of ecology, contrasts with the decline in growth between the early 1990s and mid-2000s, when temperatures were increasing. This implies that the influence of temperature on growth between those two
1 periods was overridden by other factors, i.e. hypoxia and increased densities, but then, in
2007-14, became the most important factor, when fish became concentrated in relatively similar densities in nearshore waters. Although there was no evidence that the overall abundance of A. butcheri in the Swan River Estuary had changed markedly between the early 1990s and early 2010s, there is circumstantial evidence that the stock of this species declined during the 1980s and 1990s in the Blackwood River Estuary, due to fishing pressure and/or environmental changes.
Concern for the status of the stock of A. butcheri in the Blackwood River Estuary led to a study aimed at determining the efficacy of using restocking to replenish the population of this species in this system. The growth and maturity schedules of the 2001 and 2002 year classes of A. butcheri, which had been cultured and introduced into the
Blackwood River in south-western Australia at seven and four months old, respectively, were determined from samples collected regularly between 2002 and 2014. Restocked A. butcheri could always be distinguished from its wild stock because their otoliths retained the pink coloration of the alizarin complexone with which they had been stained prior to release. The growth and maturity schedules of restocked fish were only slightly inferior to those of its wild stock and the mean gonad weights of these two groups did not differ significantly in any month. As increasing numbers of restocked A. butcheri attained the minimum legal length (MLL) of 250 mm for retention, their contribution to the commercial fishery increased from 6% in 2005 to 74% in 2010. That contribution subsequently declined to 39% in 2012 and 10% in 2014, due predominantly to the introduction of the very strong 2008 year class in the commercial catches, the first substantial recruitment into the population since 1999. As restocked fish were estimated as contributing ~55% to the eggs produced in 2008, substantial numbers of the 2008 year class were derived from spawning by restocked fish. The results of this and a previous
2 genetic study imply that restocking is an effective and appropriate way for replenishing stocks of an estuarine species such as A. butcheri, especially as its recruitment is highly episodic.
Protection of riverine discharge into the estuaries of south-western Australia is likely to be the most effective management strategy to maintain the ‘health’ of estuaries and their fisheries. It follows that a thorough understanding of the hydrology of estuaries will become increasingly important to managing these systems as they become subjected further to the effects of climate change and increases in the demand for water resources by human populations. As A. butcheri completes its life cycle within its natal estuary and has plastic biological characteristics, it is an ideal candidate for use as an indicator of the health of an estuary and hypothesising on the effects of climate change.
3
Acknowledgements
I would first and foremost like to thank Emeritus Ian Potter for supervising my research.
His endless dedication, advice and encouragement not only facilitated the completion of this thesis but will continue to inspire me throughout my scientific career. My gratitude is expressed to Professor Norm Hall for his supervision, in particular, his essential advice for the statistical and modelling component of this thesis. I would also like to acknowledge Dr
Alex Hesp for his supervision and assistance. My gratitude is also expressed to Chris
Hallett and Amanda Buckland for the many days of assistance in the field and to Gavin
Sarre, Steeg Hoeksema, Kim Smith, Michelle Gardner, Joel Williams, Eloise Ashworth and Daniel Yeoh for providing data and to Greg Jenkins and Trevor Price who made the studies on restocking possible. I also thank other members of the Centre for Fish and
Fisheries for their input, James Tweedley, Fiona Valesini, Thea Linke, Calais Tink, Lauren
Veale, Ben French and Elaine Lek.
Financial and other support was provided by Ian Stagles of the West Australian Fish
Foundation, Recfishwest, Murdoch University, Western Australia Department of Fisheries,
Swan River Trust and Department of Water.
Above all else, I would like to thank my mum Marion Cottingham and my gorgeous fiancé
Tek Green for their support throughout the duration of my PhD.
4
Publications
Papers published from work undertaken in this thesis.
The following paper was published from Chapter 3 of this thesis.
Cottingham, A., Hesp, A, Hall, N.G., Hipsey, M. and Potter, I.C., 2014. Marked deleterious changes in the condition, growth and maturity schedules of Acanthopagrus butcheri (Sparidae) in an estuary reflect environmental degradation. Estuarine, Coastal and Shelf Science 149, 109-119.
The following paper has been based on Chapter 4 of this thesis.
Cottingham, A., Hall, N.G., Potter, I.C., 2016. Factors influencing the growth of Acanthopagrus butcheri (Sparidae) in a eutrophic estuary have changed over time. Estuarine, Coastal and Shelf Science. 168, 1-11.
The following papers were published from Chapter 5 of this thesis, with the first of these papers having biological and genetic components, of which I was responsible for the former.
Gardner, M.J., Cottingham, A., Hesp, S.A., Chaplin, J.A., Jenkins, G.I., Phillips, N.M., Potter, I.C., 2013. Biological and genetic characteristics of restocked and wild Acanthopagrus butcheri (Sparidae) in a south-western Australian estuary. Reviews in Fisheries Science 21, 441-452.
Cottingham, A., Hall, N.G., Potter, I.C., 2015. Performance and contribution to commercial catches and egg production by restocked Acanthopagrus butcheri (Sparidae) in an estuary. Estuarine, Coastal and Shelf Science 164, 194-203.
5
Reports published from work undertaken in this thesis.
Williams, J., Cottingham, A., Potter, I.C., 2014. Spatial and temporal distributions of spawning by Black Bream (Acanthopagrus butcheri) in the Blackwood River Estuary. Report to West Australian Fish Foundation.
Cottingham, A., Potter, I.C., 2014. Continuation of the monitoring of restocked and wild Black Bream Acanthopagrus butcheri in commercial catches in the Blackwood River Estuary. Report to West Australian Fish Foundation.
Cottingham, A., Buckland, A., Potter, I.C., 2013. Results and implications of monitoring the biological characteristics of restocked and wild Black Bream, Acanthopagrus butcheri, in the Blackwood River Estuary. Report to West Australian Fish Foundation.
Gardner, M.J., Cottingham, A., Phillips, N.M., Hesp, S.A., Chaplin, J.A., Jenkins, G.I., 2010. Biological performance and genetics of restocked and wild bream in the Blackwood River Estuary. Report to the West Australian Fish Foundation.
Valesini, F.J., Hallett, C.S., Cottingham, A., Hesp, S.A., Hoeksema, S.D., Hall, N.G., Linke, T.E., Buckland, A.J., 2010. Development of biotic indices for establishing and monitoring ecosystem health of the Swan-Canning Estuary. Report to the Swan River Trust.
6
Table of contents Abstract ...... 1 Acknowledgements ...... 4 Publications ...... 5 Chapter 1 ...... 10 General introduction ...... 10 1.1 Definition of an estuary ...... 10 1.2 Productivity and food webs in estuaries ...... 11 1.3 Classification of the ways fish use estuaries; the guild approach ...... 12 1.4 Anthropogenic impacts on estuaries ...... 16 1.5 General characteristics of macro- and microtidal estuaries ...... 18 1.6 Estuaries of south-western Australia ...... 19 1.7 Acanthopagrus butcheri ...... 24 1.8 Objectives of thesis ...... 27 Chapter 2 ...... 28 General material and methods ...... 28 2.1 Sampling methods ...... 28 2.2 Sampling regime ...... 28 2.3 Laboratory procedures and measurements ...... 31 2.3.1 Ageing ...... 31 2.3.2 Growth ...... 34 2.3.3 Length and age at maturity ...... 35 Chapter 3 ...... 37 Marked deleterious changes in the condition, growth and maturity schedules of Acanthopagrus butcheri in the Swan River Estuary reflect environmental degradation ...... 37 3.1 Introduction ...... 37 3.2 Materials and Methods ...... 41 3.2.1 Description of the Swan River Estuary ...... 41 3.2.2. Physico-chemical data and sampling regimes ...... 43 3.2.3 Assessment of body condition ...... 45 3.2.4 Growth and reproductive biology ...... 47 3.2.5 Measures of relative abundance ...... 47 3.3 Results ...... 48 7
3.3.1 Decline in freshwater discharge and relationship with hypoxia ...... 48 3.3.2 Body condition and growth ...... 50 3.3.3. Lengths and ages at maturity ...... 54 3.3.4 Catch rates in deeper waters and densities in shallow water ...... 58 3.4 Discussion ...... 59 3.4.1 Condition and growth...... 60 3.4.2. Length and age at maturity...... 62 Chapter 4 ...... 65 Factors influencing the growth of Acanthopagrus butcheri in a eutrophic estuary have changed over time...... 65 4.1 Introduction ...... 65 4.2 Materials and methods ...... 67 4.2.1 Temperature and freshwater flow ...... 67 4.2.2. Sampling regime and processing of fish ...... 68 4.2.3 Year-effect growth model ...... 70 4.2.4 Relationship between growth and temperature and freshwater discharge ...... 74 4.3 RESULTS ...... 74 4.3.1 Temperature and freshwater discharge ...... 74 4.3.2 Model parameter selection and simulation ...... 75 4.3.3 Model results ...... 78 4.3.4 Relationship between growth and temperature and freshwater discharge ...... 83 4.4 DISCUSSION ...... 84 Chapter 5 ...... 89 Performance and contribution to commercial catches and egg production by restocked Acanthopagrus butcheri in the Blackwood River Estuary...... 89 5.1 Introduction ...... 89 5.2 Materials and Methods ...... 92 5.2.1 Collection of restocked and wild Acanthopagrus butcheri ...... 92 5.2.2 Ageing and growth ...... 95 5.2.3 Reproduction and fecundity ...... 96 5.3 RESULTS ...... 99 5.3.1 Growth ...... 100 5.3.2 Maturation ...... 102 5.3.3 Contribution of restocked fish to commercial catches ...... 107 8
5.3.4 Egg production ...... 110 5.4 DISCUSSION ...... 111 5.4.1 Survival and growth of restocked fish ...... 111 5.4.2. Comparisons of the performance of restocked and wild fish ...... 111 5.4.3 Contributions of restocked and wild fish to commercial gill net catches ...... 113 Chapter 6 ...... 116 General Discussion ...... 116 6.1 Swan River Estuary ...... 116 6.1.1 Influence of hypoxia ...... 116 6.1.2 Influence of temperature ...... 118 6.1.3 Recruitment ...... 120 6.1.4 Management ...... 123 6.2 Blackwood River Estuary ...... 123 6.3 Conclusions ...... 127 REFERENCES ...... 130
9
Chapter 1
General introduction
1.1 Definition of an estuary
Estuaries are among the most productive of all aquatic ecosystems (Schelske & Odum,
1962: Kennish, 2003; Hoellein et al., 2013) and thus provide valuable nursery areas for many commercially and recreationally important fish species (Lenanton & Potter, 1987;
Able & Fahay, 2010; Potter et al., 2013). Estuaries are also complex, with each being influenced by a suite of biophysical processes, whose combinations differ and, as a consequence, the environmental characteristics of these systems vary markedly (Monbet,
1992; Tagliapietra et al., 2009; Potter et al., 2010). This explains why early attempts to produce a description of estuaries that applies to all of these systems throughout the world were not entirely successful.
The following definition of Pritchard (1967) is widely quoted. “An estuary is a semi-enclosed coastal body of water, which has a free connection with the open sea and within which seawater is measurably diluted with fresh water derived from land drainage”.
This definition accommodates the macrotidal estuaries of the northern hemisphere, where large tidal forcing and strong freshwater flow maintains “a free connection with the open sea”. However, as Day (1980) pointed out, many estuaries in temperate regions of the southern hemisphere, such as those in southern Africa and southern Australia, can become periodically isolated from the ocean through the formation of a bar at their mouths, and some estuaries in this region can become hypersaline during the warm, dry months when evaporation rates are high. The definition of Day (1980), which took into account those characteristics, has recently been refined by Potter et al. (2010) as follows. “An estuary is a partially enclosed coastal body of water that is either permanently or periodically open to the sea and which receives at least periodic discharge from a river(s), and thus, while its
10 salinity is typically less than that of seawater and varies temporally and along its length, it can become hypersaline in regions when evaporative water loss is high and freshwater and tidal inputs are negligible”.
1.2 Productivity and food webs in estuaries
Estuaries contain large amounts of organic carbon, derived from runoff of biogenic material, and nutrients from the surrounding land (Hopkinson et al., 2005), which fuel the food chain in these systems (Canuel et al., 1995; McLusky & Elliott, 2004). The organic carbon is derived from three main sources. Allochthonous sources comprise mainly plant material that enters the estuary via runoff or from tributaries (Raymond & Bauer, 2001).
This plant material is broken down by microbial decomposers, e.g. heterotrophic bacteria, and the organic carbon in the resultant detritus then becomes available to primary consumers, e.g. detritivores (Cebrian, 1999; Sobczak et al., 2005). Autochthonous sources of carbon, represented by vascular plants and macroalgae, are also routed through decomposers, with only ~20% of the carbon ingested being consumed directly by herbivores (Cebrian, 1999). In contrast, microalgae, e.g. phytoplankton and microphytobenthos, are rich in nitrogen and essential fatty acids and are directly metabolized by consumers (Furnas, 1990). While the amount of microphytobenthos can be comparable with that of phytoplankton in some estuaries (Underwood & Kromkamp,
1999), phytoplankton is typically the dominant resource base for supporting the production of higher trophic levels (Caddy, 2000; Breitburg et al., 2009; Gillson, 2011).
The primary consumers in estuaries, i.e. animals that feed on detritus and plant material, include zooplankton in the water column and benthic invertebrates, such as certain bivalve molluscs and polychaetes, and fish (Whitfield, 1989; Elliott & Hemingway,
2002). These members of the estuarine fauna are consumed by secondary consumers,
11 which range from gastropod molluscs to small and large carnivorous fish and birds
(Schelske & Odum, 1962; Baird, 1999). The differences in feeding mode among fishes in estuaries led to Elliott et al. (2007) constructing the following feeding mode functional groups (FMFG); zooplanktivore, detritivore, herbivore, omnivore, piscivore and zoobenthivore. It was recognised, however, that most species assigned to any one of these groups also fed, but to a lesser extent, on the food that characterised another group. Indeed, this was so extreme in some cases that a seventh group was constructed, i.e. miscellaneous/opportunist, for species that fed on such a diverse range of food that they could not readily be assigned to one of the above specialised FMFGs. The food consumed by the Black Bream Acanthopagrus butcheri, the subject of the current study, varies markedly among estuaries and can include large volumes of macrophytes in some systems
(Sarre & Potter, 1999; Chuwen et al., 2007) and therefore belongs to the last FMFG.
1.3 Classification of the ways fish use estuaries; the guild approach
It has long been recognised that estuaries are ‘used’ in different ways by the fish species that are found within these systems (e.g. Cronin & Mansueti, 1971; Haedrich, 1983;
Whitfield, 1999; Elliott et al., 2007; Able & Fahay, 2010). The various schemes that have been developed for defining that usage have recently been refined by Potter et al. (2013), who subdivided the fishes into four categories, which, each comprised a number of guilds.
The four main overarching categories were marine for species that spawn at sea; estuarine for populations where individuals complete their life cycle within the estuary; diadromous for species that migrate between the sea and freshwater, and freshwater for species that spawn in freshwater.
At the guild level, distinctions were made between those marine species that typically enter estuaries ‘accidentally’ and usually in low numbers (marine straggler) and
12 those that enter estuaries in substantial numbers, usually as juveniles (marine-estuarine opportunist) and those that require sheltered estuarine habitats as juveniles (marine- estuarine dependent). In the case of the estuarine category, the most important distinction is between species that are found only in estuaries (solely estuarine) and those that are represented by populations in both estuarine and marine waters (estuarine & marine). The diadromous category comprises anadromous, catadromous and amphidromous species.
The freshwater category is far less represented in estuaries than the marine category (Potter et al., 2013).
Examples of the main guilds derived from studies in different regions of the world are given below, drawing on inter alia studies in Europe (e.g. Claridge et al., 1986; Franco et al., 2008), North America (e.g. Able & Fahay, 2010), South America (e.g. Plavan et al.,
2010) and Africa and Australia (e.g. Whitfield, 1998; Blaber, 2000; Potter et al., 1990,
2014).
The species of marine straggler are frequently stenohaline, which are thus not adapted to living in the low and or variable salinities often present in the main body and distal regions of estuaries and are consequently most frequently found in the high salinity areas near the estuary mouth (Potter et al., 2013). Many elasmobranchs, such as the Bull
Shark Carcharinas leucas are typical marine stragglers, but this guild also includes numerous species of teleosts such as the Spanish Mackerel (Scomberomorus maculatus),
Red Porgy (Pagrus pagrus) and American Harvestfish (Peprilus paru).
The marine-estuarine opportunists, such as the European Bass (Dicentrarchus labrax), Bluefish (Pomatomus saltatrix), Southern Kingfish (Menticirrhus americanus),
Sea Mullet (Mugil cephalus) and Atlantic Herring (Clupea harengus) are typically euryhaline. They are thus able to penetrate the middle and upper regions of the estuary where salinities are reduced. Several marine-estuarine opportunist species make an
13 important contribution to fisheries in coastal marine waters and sometimes also in estuaries
(Lenanton & Potter, 1987; Blaber, 1997; Costa et al., 2002; Dolbeth et al., 2008; Able &
Fahay, 2010).
Because estuaries are of recent origin and thus transitory in a geological sense,
Hedgpeth (1982) has argued that marine species are unlikely to depend on estuaries for completing their life cycle. However, a few species, such as the Oval Moony
(Monodactylus falciformis) are regarded as dependent on estuaries in southern Africa, as their juveniles are unable to tolerate the ‘turbulent’ conditions that occur along the coast and thus require the more sheltered waters of estuaries for survival during early life
(Whitfield, 1999; Potter et al., 2013).
The solely estuarine category includes the Black Bream (Acanthopagrus butcheri),
Common Goby (Pomatoschistus microps) and Estuarine Round Herring (Gilchristella aestuaria) with the last of these species also abundant in certain freshwater lakes
(Lenanton & Potter, 1987; Potter et al., 2013). The populations of some solely estuarine species are very substantial in those estuaries when the conditions are consistently particularly favourable for breeding and larval survival during and after the spawning period. Species such as A. butcheri grow to a sufficient size and provide good angling and quality of flesh to make a crucial contribution to recreational fisheries and also in some estuaries to commercial fisheries (Lenanton & Potter, 1987; Kailola et al., 1993; Hutchins
& Thompson, 2001; Jenkins et al., 2010).
Good examples of species that can complete their life cycles in both estuaries and marine waters are provided by the Estuary Cobbler (Cnidoglanis macrocephalus),
Longsnout Pipefish (Syngnathus temminckii), Western Gobbleguts (Ostorhinchus rueppellii) and the Australian Anchovy Engraulis australis (Potter et al., 2013; Veale et al., 2014). Several species in this category exhibit parental care of their eggs, larvae and/or
14 young (Blumer, 1982; Wilson et al., 2003). While the populations of C. macrocephalus in estuaries and coastal marine waters have been shown to be genetically distinct (Ayvazian et al., 1994), it is likely that the marine and estuarine assemblages of some species form a continuum (Able & Fahay, 2010).
Estuaries provide a crucial route between the areas where spawning and feeding occur. Catadromous species, such as the European Eel (Anguilla anguilla) spawn at sea and grow in rivers (Potter et al., 2013). In contrast, anadromous species, such as many salmonids, e.g. Pink Salmon (Oncorhynchus gorbuscha) and Atlantic Salmon (Salmo salar) and several species of lampreys, e.g. Sea Lamprey (Petromyzon marinus) and
Pouched Lamprey (Geotria australis) and the Hilsa Shad (Tenualosa ilisha) and Allis Shad
(Alosa alosa), spawn in rivers and grow at sea (McDowall, 1988). Several anadromous species make an important contribution to commercial fisheries (Hard et al., 2008; Al-
Dubakel, 2011). Amphidromous species, such as members of the goby subfamily
Sicydiinae, are particularly prevalent in rivers of small islands in the tropical Indo Pacific, from which they are swept out to sea as larvae, but eventually return to rivers where they become mature and spawn (McDowall, 1988; Tweedley et al., 2013). Such movements enable species to be dispersed across the marine divide and into the rivers of the recently- emerged islands of this region.
While some freshwater species (stragglers) can occur sporadically in the upper and low salinity reaches of estuaries, few freshwater species penetrate the middle part of the estuary in numbers and would therefore constitute the freshwater estuarine opportunist guild (Potter et al., 2013). Good examples of freshwater estuarine opportunist species are provided by the Mozambique Tilapia (Oreochromis mossambicus) and Three-Spined
Stickleback (Gasterosteus aculeatus).
15
1.4 Anthropogenic impacts on estuaries
Estuaries have been widely exploited by human populations and sometimes to the detriment of these systems (de Jong et al., 2002; Kennish, 2003; Lotze et al., 2006).
Estuaries have been used, for example, as important docking points for ships and for transporting resources from other regions, thereby stimulating the development of towns and cities along their shores. This then inevitably resulted in estuaries being used as a repository for waste, which in cases, such as with certain heavy metals, lead to their accumulation through the food chain, thereby posing a potential risk to the health of the upper level components of the food chain in the estuary and to humans (Oze et al., 2005).
Runoff from the surrounding towns, agriculture and catchments has often resulted in a greatly increased input of nutrients into the estuary and therefore the development of severe cases of eutrophication (de Jong et al., 2002). Typical symptoms of eutrophication include an increased frequency of macroalgal growths (Fig. 1.1) and phytoplankton blooms, some of which are toxic, and prolonged periods of low dissolved oxygen concentrations and fish mortalities (McComb, 1995; Whitall et al., 2007; Thronson &
Quigg, 2008; Paerl et al., 2014). The tributary rivers of many estuaries have also been dammed to provide a source of freshwater for various purposes. This has led to a reduction in freshwater discharge and thus to a decrease in the flushing of the estuary, which is important for the maintenance of a well-oxygenated and ‘healthy’ system.
The degradation of temperate estuaries in recent decades as a result of, for example, eutrophication and development of hypoxia, have led to these systems being the most degraded of all marine environments (Jackson et al., 2001). Various studies have shown that increases in hypoxia can have detrimental effects on the species composition of the benthic macroinvertebrates. This results from the differential ability of, for example, crustaceans and polychaetes, to survive such perturbations, which alter the
16 biogeochemistry of the sediment and modifies nutrient cycling between the benthos and water column (Rakocinski et al. 1997; Gaston et al. 1998; Middleburg & Levin 2009).
Hypoxia can also lead to changes in certain biological characteristics of key species in those estuaries. These include reductions in body condition and growth (Eby et al., 2005) and declines in productivity in areas particularly affected by hypoxia (Diaz & Rosenberg,
2008; Breitburg et al., 2009). Furthermore, exposure to hypoxia has also been shown to have a negative effect on maturation schedules, either through endocrine disruption (Wu et al., 2003) or indirectly, as a result of phenotypic plasticity that depends on growth of the individual (Morita & Morita, 2002).
Figure 1.1 Macroalgae collected in a seine net during sampling for fish in the Peel-Harvey
Estuary.
17
1.5 General characteristics of macro- and microtidal estuaries
Estuaries were formed by rising sea levels during the Flandrian transgression (Perillo,
1996). In tidal-dominated regions, estuaries generally consist of a landward tapering funnel, with the width and depth increasing exponentially at their mouths (Perillo, 1996).
While salinity declines progressively landward along the lengths of these systems, the strong tidal amplitude in estuaries, such as the Humber and Ouse Estuary, results in the salinity at any one point along those lengths varying markedly during each tidal cycle
(Uncles et al., 2006). The combination of large tidal amplitude, wide and deep entrance channels and relatively consistent freshwater input ensure that these macrotidal estuaries are well mixed (Elliott & McLusky, 2002; Savenije, 2012). The strong tidal effects also lead to the creation of extensive and highly productive inter-tidal areas, which is very important for wading birds (Baird et al., 1985; McLusky et al., 1992).
Many estuaries on, for example, the eastern seaboard of the United States, such as
Chesapeake Bay and the Neuse River Estuary, are microtidal. The waters in these systems are only partially mixed, have a relatively long residence time and contain a limited inter- tidal area (Luettich et al., 2002; Reynolds-Fleming & Luettich, 2004; Kemp et al., 2005;
Xiong & Berger, 2010).
In a large continental land mass in the southern hemisphere, such as Australia, the estuaries in the north are located in a macrotidal region, and these possess many of the characteristics of their counterparts in the northern hemisphere, whereas those in the south are situated in a microtidal region (ICSM, 2014). In Australia, the microtidal estuaries include the Swan River and Blackwood River estuaries, in which the studies for this thesis were carried out. A combination of changes in sea level, long-shore drift and infilling by sediments, derived from runoff from catchments, has ensured that estuaries in temperate
18 regions of the southern hemisphere typically remain relatively shallow (Hodgkin & Hesp,
1998; Cooper et al., 2008; Schumann, 2013).
1.6 Estuaries of south-western Australia
South-western Australia has a Mediterranean climate. Since the majority of precipitation in this microtidal region thus occurs between late autumn, i.e. May, and mid spring, i.e.
October (Wright, 1997; Hoeksema & Potter, 2006; Australian Bureau of Meteorology,
2015), the volume of freshwater discharge is also highly seasonal in systems such as the
Swan River Estuary and Blackwood River Estuary, typically peaking between mid winter and early spring (Hoeksema & Potter, 2006; Western Australia Department of Water,
2015). As a consequence of the small tidal influence and the restriction of substantial freshwater discharge to the wet, cool period of the year, the water retention times in, for example, the upper Swan River Estuary, can range from as short as <0.2 days in winter to as long as a year in summer (Chan & Hamilton, 2001).
As a consequence of the long residency time in summer, the phytoplankton in these microtidal systems exhibit high rates of photosynthesis and therefore attain far higher concentrations than in typical macrotidal systems. This is particularly the case when the estuary becomes nutrient enriched (McComb, 1995; Steckis et al., 1995). The estuaries of south-western Australia are inherently susceptible to eutrophication, mainly due to restricted tidal flushing, particularly during the protracted dry period, which results in long residence times and the accumulation of organic material in bottom waters. Furthermore, since the soils in this region are nutrient deficient, agriculture has relied heavily on the addition of fertilizers for stimulating crop and pasture growth (McComb & Davis, 1993).
As the soils in south-western Australia show little ability to retain nutrients, phosphorous in particular is readily leached and enters the estuary. Thus, for example, between 1997
19 and 2006, 251 tonnes of nitrogen and 26 tonnes of phosphorous were estimated to have entered the Swan River Estuary each year (Swan River Trust, 2009).
Many estuaries in the low tidal amplitude regions of south-western Australia and southern Africa become closed intermittently, seasonally or for longer periods, through the formation of a sand bar at their mouths (Day, 1981; Potter et al., 1993; Hodgkin & Hesp,
1998; Cooper, 2001). In Western Australia, the Swan River Estuary has remained permanently open since the removal of a rock bar at its mouth in 1886 to provide a deep entrance for ship access (Brearley, 2005). The mouth of the natural entrance of the Peel-
Harvey Estuary, just to the south of the Swan River Estuary, is kept open by dredging the sand that accumulates there (Brearley, 2005). While the locations of the natural entrance to the Swan River and Peel-Harvey estuaries have remained the same, that of the Blackwood
River Estuary has varied and occasionally becomes closed and thus likewise has to be artificially kept open (Brearley, 2013). More comprehensive descriptions of the Swan
River and Blackwood River estuaries are given in Chapter 2. The prevalence of estuaries that become closed through the formation of a sand bar at their mouths increases in an eastward direction along the south coast of Western Australia (Hodgkin & Hesp, 1998).
This largely reflects progressively declining rainfall that occurs from west to east along the south coast (Hodgkin & Hesp, 1998).
South-western Australian estuaries characteristically comprise three morphologically distinct regions, i.e. a short, narrow entrance channel, shallow and wide central basin(s), and the relatively narrow and long saline lower reaches of the tributary rivers (Day, 1981; Hodgkin & Hesp, 1998). In permanently-open estuaries, such as the
Swan River Estuary, salinities in the entrance channel typically remain euhaline for much of the year, except during winter when surface waters can become fresh (Chan &
Hamilton, 2001). In the basins, salinities are typically only euhaline during
20 summer/autumn, whereas during winter, the formation of a salt wedge results in stratification of the water column with salinities in the deeper waters remaining euhaline whereas the surface waters can become fresh. While the basin of the Swan River Estuary is only completely flushed during very wet winters, the upper estuary is fully flushed in most winters (Kurup et al., 1998; Twomey & John, 2001). When freshwater discharge subsides, the homogenous conditions found in the upper estuary during late winter and early spring are disrupted by the tide-driven progression of the salt wedge. Although stratification is typically evident throughout the upper estuary during the warmer months of the year
(Stephens & Imberger, 1996), as the salt wedge progresses upstream, the overlying freshwater and seawater below gradually mix (Twomey & John, 2001). As the salt wedge approaches its most landward position, ~60 km from the estuary mouth, salinities in the downstream region of the upper estuary become euhaline, whereas those furthest upstream become brackish (Swan River Trust, 2014). At the onset of winter, or following unseasonal freshwater discharge, the salt wedge reforms and gradually retreats downstream towards the estuary basin (Douglas et al., 1995). The consistent cyclical changes in salinity throughout the year in the different regions of south-western Australian estuaries are a major determinant of the compositions of the flora (e.g. Thompson, 2001; Twomey &
John, 2001) and fauna (e.g. Loneragan & Potter, 1990; Kanandjembo et al., 2001a,b;
Wildsmith et al., 2011) of these estuaries.
The different species of fishes found in south-western Australian estuaries often exhibit pronounced spatial partitioning among the three morphologically and physico- chemically different regions of these systems (e.g. Loneragan et al., 1987; Loneragan &
Potter, 1990; Potter & Hyndes, 1999; Potter et al., 2014). For example, among the five most abundant species of atherinid in the Swan River Estuary, two species occupy predominantly the lower estuary, two are mainly found in the middle estuary and one
21 largely in the upper estuary (Prince et al., 1982). Likewise, among the five most numerous species of gobiid, one lives mainly in the lower estuary, while the other four are largely restricted to the upper estuary (Gill & Potter, 1993). The Black Bream Acanthopagrus butcheri, which is much larger than those atherinid and gobiid species and is the focus of the current thesis, spends most of the year in the upper estuary, i.e. riverine reaches, but is often swept downstream into the basin by the heavy freshwater discharge that occurs in winter (Sarre & Potter, 2000). Thus, fishing for this iconic recreational species is largely focused on the upper estuary (for further details see Chapters 3, 4 and 5).
Certain estuaries in south-western Australia have been subjected to extreme eutrophication as a result of the input of nutrients from surrounding agricultural and urban development (McComb et al., 1981; Cloern et al., 2014). In the Peel-Harvey Estuary, this was manifested in massive growths of macroalgae and the toxic cyanobacterium Nodularia spumiginea. The eutrophication was so extreme that a deep channel was constructed between this large estuary and the ocean to increase greatly the flushing of nutrients out of the system. In the Swan River Estuary, ~60 km to the north of the Peel-Harvey Estuary, there have been periodic large blooms of microalgae, including Karlodinium micrum, which is toxic andtherefore contributed to major mortalities of Black Bream
Acanthopagrus butcheri (Smith, 2006; Hallett et al., 2015), the main recreational fish species in southern Australian estuaries (Lenanton & Potter, 1987; Ferguson &Ye, 2008;
Hindell et al., 2008; Williams et al., 2012). However, mortalities of this species at other times were presumably related to the hypoxia that resulted from extreme eutrophic events.
The Swan River Estuary and also the Blackwood River Estuary have also suffered from the effects of the decline in freshwater discharge and flushing that has occurred during the last 20-30 years as a result of a reduction in rainfall (see Chapter 3). The marked reduction in rainfall is a consequence of anticyclonic events that now persist over the
22 southern Indian Ocean and which cause a weakening of the north-westerly winds and a reduction in the transport of land-bound moist air (Feng et al., 2010). These changes are reflected in the annual rainfall at Perth Airport, close to the Swan River Estuary, declining from 845 mm y-1 during the 1960s to 684 mm y-1 in the 2000s (Fig. 1.2). An even greater decline in rainfall occurred at Cape Leeuwin, in the proximity of the mouth of the
Blackwood River Estuary, where rainfall declined from 1066 mm y-1 to 712 mm y-1 over the same period (Fig. 1.2). These changes in rainfall over time have resulted in reduced freshwater discharge, which has led to the accumulation of nutrients and other organic material in the deeper waters and thus in exceptional circumstances led to increased levels of hypoxia in those waters (Tweedley et al., 2014). The effects of these changes are explored in Chapters 3 to 5.
Figure 1.2 Total annual rainfall (mm) recorded at Perth Airport (top) and Cape Leeuwin (bottom) between 1960 and 2014. Horizontal bars represent decadal averages.
23
The compositions of the fish faunas in the different regions of south-western
Australian estuaries often differ markedly and typically undergo marked seasonal changes, which, in the middle and upper reaches, are largely related to changes in salinity (e.g.
Potter & Hyndes, 1999; Kanandjembo et al., 2001; Hoeksema & Potter, 2006). For example, in summer, semi-anadromous, marine-estuarine and true estuarine species are typically most abundant in the upper reaches of the estuaries, whereas in winter, after freshwater discharge has increased and salinities declined in the upper reaches, these species are typically more abundant in the estuary basins, or in the case of semi- anadromous species, migrate to sea (Kanandjembo et al., 2001).
As with other microtidal estuaries in the southern hemisphere, a number of species complete their life cycles in south-western Australian estuaries (Potter et al., 1990; Potter
& Hyndes, 1999; Veale et al., 2014). The ability of these estuarine species to breed successfully in these systems has been attributed to the relatively stable water conditions that persist in the middle and upper reaches of these estuaries during the summer months, when these species typically spawn, a stability attributed to the low tidal amplitude and lack of freshwater flow at that time of year (Potter & Hyndes, 1999). In south-western
Australia, ‘estuarine’ species make a substantial contribution to the overall abundance of fishes in estuaries (Potter & Hyndes, 1999; Kanandjembo et al., 2001).
1.7 Acanthopagrus butcheri
The Black Bream Acanthopagrus butcheri (Sparidae), which attains a maximum length of
533 mm and weight of 3,450 g (Hutchins & Thompson, 2001), is abundant in many estuaries in southern Australia (Gomon et al., 2008). Although A. butcheri can swim several kilometres in a day (Hindell et al., 2008; Sakabe & Lyle, 2010), it typically resides for much of the year in the upper regions of estuaries, where it spawns during the spring
24 and early summer (Sarre & Potter, 2000; Nicholson et al., 2008; Williams et al., 2012).
This sparid can be swept out of the estuary during particularly high freshwater discharge and in south-eastern Australia may occasionally enter other estuaries (Lenanton et al.,
1999; Burridge & Versace, 2006). The results of studies of the reproductive biology and distribution of the various life stages of this species in south-western Australian estuaries demonstrate that individuals of this sparid typically complete their life cycle within their natal estuary in this region (Potter & Hyndes, 1999; Sarre & Potter, 2000). Such a conclusion is consistent with the fact that the genetic compositions of A. butcheri in the various estuaries in south-western Australia are significantly different (Chaplin et al.,
1998), as is also the case in south-eastern Australia (Burridge & Versace, 2006).
Acanthopagrus butcheri is one of the most important recreational finfish species in the estuaries of southern Australia (Smith, 2006; Jenkins et al., 2010). In south-western
Australia, A. butcheri was also historically an important commercial species (Lenanton &
Potter, 1987; Loneragan et al., 1987; Kailola et al., 1993), although this is no longer the case due to governmental buy-backs of existing licences, e.g. in the Swan River Estuary. In estuaries such as Culham Inlet, the abundance of this sparid has become so reduced, mainly through mortality associated, in particular, with the development of highly elevated salinities brought about by reduced freshwater discharge, that it is no longer fished commercially in those systems (Norriss et al., 2002; Hoeksema et al., 2006).
Acanthopagrus butcheri is also an important aquaculture species and has been introduced into a large number of artificial impoundments throughout south-western
Australia (Partridge et al., 2004; Smith et al., 2009). Several illegal introductions into natural lakes have also occurred, with a population being sustained in Lake Clifton for many years (Smith & Norriss, 2011). In some estuaries, such as the Blackwood River
25
Estuary, restocking has been used to increase the stock following a prolonged period of low catches (Potter et al., 2008).
Aspects of the biology of A. butcheri, such as growth, vary markedly among estuaries (e.g. Hobday & Moran, 1983; Coutin et al., 1997; Morison et al., 1998; Sarre &
Potter, 2000; Hoeksema et al., 2006). For example, Sarre & Potter (2000) showed that A. butcheri grew more rapidly early in life and reached a larger maximum length in the Swan
River Estuary than in each of three other estuaries they studied, i.e. Wellstead,
Nornalup/Walpole and Moore River estuaries. For example, by the age of 3 and 6 years, female A. butcheri had attained total lengths of 266 and 368 mm, respectively, in the Swan
River Estuary compared with 146 and 232 mm, in the Moore River Estuary, which is only
~85 km further north. The early growth and maximum length in the Swan River Estuary were, however, slightly less than in a saline coastal lake, i.e. Lake Clifton (Sarre & Potter,
2000).
Although the populations of A. butcheri in different south-western Australian estuaries have been shown to be genetically distinct (Chaplin et al., 1998), an aquaculture study of this species demonstrated that, when cultured under the same conditions, the growth of the juveniles produced from broodstock from the permanently-open Swan River and Moore River estuaries were essentially the same (Partridge et al., 2004). Thus, the growth of this sparid in south-western Australian estuaries is influenced largely by environmental factors rather than its genetic composition (Partridge et al., 2004). Thus, in the Moore River Estuary, this species consumes far larger amounts of plant material, such as Cladophora sp., than it does in the Swan River Estuary, whereas the reverse is the case with bivalve molluscs, such as Xenostrobus securis and Fluviolanatus subtorta, which have a greater nutritional value than plant material. Furthermore, the far greater densities of A. butcheri in the Moore River Estuary than the Swan River Estuary were also
26 considered to contribute to A. butcheri growing less rapidly in the Moore River Estuary
(Sarre et al., 2000).
The lengths and ages at which A. butcheri in the estuaries attained maturity also differed substantially and maturity tended to occur at a smaller size and older age in those estuaries in which growth was less rapid (Sarre & Potter, 2000). For example, the lengths and ages at which 50% of fish attain maturity, L50s and A50s, respectively, of females of
218 mm and 2.2 years in the Swan River Estuary differed markedly from the corresponding values of 157 mm and 3.3 years in the Moore River Estuary.
The above examples of marked variations in several of the biological characteristics of A. butcheri among estuaries with different environmental conditions emphasise that these characteristics are particularly plastic in this sparid. This species is thus likely to provide an excellent ‘model’ for exploring how changes in the environment can influence the expression of the fundamental attributes of a species, such as condition, growth and maturity schedules.
1.8 Objectives of thesis
The two main overarching aims of this thesis are as follows. 1) Test the hypothesis that the detrimental changes to the environment of the Swan River Estuary have been accompanied by a decline in the condition, growth and length at maturity of A. butcheri and an increase in its age at maturity. 2) Compare the growth and maturity schedules of restocked and wild
A. butcheri in the Blackwood River Estuary to determine whether cultured fish perform as well as the wild stock and estimate the contribution of the restocked fish to the commercial fishery. These aims are developed in each of the subsequent relevant chapters.
27
Chapter 2
General material and methods
2.1 Sampling methods
The biological characteristics of Acanthopagrus butcheri have been studied in eleven estuaries in south-western Australia and in Lake Clifton, a coastal salt lake (Sarre & Potter,
1999, 2000; Potter et al., 2008; Chuwen, 2009; Fig. 2.1a). This thesis focuses on two of these estuaries, the Swan River Estuary (Fig. 2.1b), which contains the most extensive time series of data, and the Blackwood River Estuary (Fig. 2.1c), in which substantial numbers of cultured A. butcheri have been introduced.
The information provided below covers those aspects of the sampling gear and methodology, ageing procedures, analyses of growth and determination of reproductive characteristics that are common to the studies reported in Chapters 3 to 5. Other details of the sampling regimes for the different studies of A. butcheri in the Swan River Estuary are given in Chapters 3 and 4, and those for the study undertaken in the Blackwood River
Estuary are provided in Chapter 5.
2.2 Sampling regime
The Swan River and Blackwood River estuaries were sampled using 21.5 and 41.5 m long seine nets and a 160 m long composite gill net (see Figure 2.3). The 21.5 m seine net, which consisted of a 1.5 m wide bunt of 3 mm mesh and two 10 m long wings (each comprising 4 m of 3 mm mesh and 6 m of 9 mm mesh), swept an area of 116 m2, while the
41.5 m seine net, which contained a 1.5 m bunt made of 9 mm mesh and two 20 m long wings comprising 25 mm mesh, swept an area of 274 m2. The 21.5 m net was laid parallel to the bank and then hauled onto the shore, whereas the 41.5 m net was deployed in a semi- circle from the bank using a small boat before being hauled on to the shore.
28
Figure 2.1 a) Map showing the locations of the estuaries in south-western Australia for which biological data have been collected for Acanthopagrus butcheri and the morphology of the b) Swan River Estuary and c) Blackwood River Estuary, the two estuaries in which the studies for this thesis were conducted. The box in the map of Australia in the top right hand corner of a) shows the region where the Swan River and Blackwood River estuaries are located in Australia.
29
Figure 2.2 Photographs of the Swan River Estuary (top left, courtesy of Kristian Maley) and Blackwood River Estuary (bottom left, courtesy of Simon Neville) and examples of the sites sampled by seine netting in the Swan River Estuary (top right) and the Blackwood River Estuary (middle right) and by gill netting in the Blackwood River Estuary (bottom right).
The composite gill net, which comprised eight 20 m long panels, each with a different stretched mesh size, i.e. 38, 51, 63, 76, 89, 102, 115 and 127 mm, was set, just after sunset, parallel to the shore in water depths of 2 to 6 m.
All fish were anaesthetised in an ice slurry immediately after capture.
30
Figure 2.3 Examples of the different fishing methods used to collect Acanthopagrus butcheri, showing a gill net (top left) and 21.5 m seine netting (bottom left) and a 41.5 m seine net being deployed (top right) and the commercial fisher boxing his catch in the Blackwood River Estuary (bottom right).
2.3 Laboratory procedures and measurements
2.3.1 Ageing
Each A. butcheri from the Swan River and Blackwood River estuaries, apart from those caught by the commercial fisher in the Blackwood River Estuary, were measured to the nearest 1 mm (total length), weighed to the nearest 0.1 g and sexed by macroscopic examination of their gonads. The A. butcheri supplied by the above commercial fisher had been filleted (but with their gonads retained). While their lengths and sexes could be
31 recorded, it was not possible to obtain their weights. The smallest fish, which could not be sexed, were designated randomly (and in equal numbers) as females and males.
The sagittal otoliths of each A. butcheri were removed, cleaned and stored. All otoliths were initially examined using a dissecting microscope under reflected light. The otoliths of cultured fish that were released into the Blackwood River Estuary were clearly identifiable through the purple colour produced in the central region of their otoliths by the alizarin complexone stain (Fig. 2.4; Chapter 5). Previous studies had shown that the opaque zones in all A. butcheri otoliths could be clearly identified without sectioning when there were up to six such zones, but this was not always the case with otoliths containing a greater number of such zones (Sarre & Potter, 2000). Thus, when otoliths contained ≤6 opaque zones, the numbers of their zones were counted using whole otoliths, but, when there were > 6 such zones, the numbers were counted in sectioned otoliths (see Figure 2.4).
For sectioning, otoliths were embedded in clear epoxy resin, sectioned transversely
(ca 400 µm) through their primordia and mounted on glass microscope slides using DePX mounting adhesive. As marginal increment analysis had previously shown that the opaque zones in the otoliths of A. butcheri are typically formed annually (Sarre & Potter, 2000), validation of the applicability of using opaque zones in otoliths for ageing this species was not required. To assess the level of precision of the ageing of A. butcheri, the numbers of opaque zones in a subsample of 200 otoliths from fish covering a wide size range (and using whole and sectioned otoliths) were counted independently by two readers, i.e., A.
Hesp and myself, and compared using the coefficient of variation (CV) (Chang, 1982;
Campana, 2001).
The equation is: {[√∑ ( ) ⁄( )]⁄ } , where CV j is the age precision estimate for the jth fish, X ij is the ith age determination of the jth fish, X j is the mean age estimate of the jth fish, and R is the number of time each fish is aged. The overall
32
CV was taken as the average of the CVs for all 200 fish. This analysis produced a value of
0.2% for the CV for A. butcheri, indicating a high degree of precision for the ageing of this species, given that a CV value of up to 5% is considered acceptable for medium to long- lived species (Campana, 2001).
The age of individual A. butcheri, estimated from the number of opaque zones in their otoliths, took into account 1) that newly-formed opaque zones typically become delineated from the otolith periphery by the beginning of November (Sarre & Potter,
2000), 2) a birth date, which corresponded to the approximate peak in the mean standardised gonad weight and 3) the date of capture.
Figure 2.4 Comparisons of whole otoliths from 4 month old cultured Acanthopagrus butcheri (top left) before and after the juveniles were immersed in alizarin complexone, which left a prominent stain on their otoliths, enabling the cultured fish to be distinguished following their release into the Blackwood River Estuary (see Chapter 6). A whole otoliths from a 2 year wild A. butcheri (bottom left), a sectioned otolith from a 15 year old wild A. butcheri (top right), , and a whole otolith from a cultured A. butcheri showing the retention of the purple stain in its the primordium after 12 years (bottom right).
33
2.3.2 Growth
Using non-linear regression, the von Bertalanffy growth model (VBGM) was fitted separately to the lengths at age of females and males of A. butcheri caught in the Swan
River and Blackwood River estuaries. The von Bertalanffy growth equation is
ˆ ˆ Lt L1 exp kt t0 , where Lt is the expected total length at age t (years), L is the
-1 asymptotic length (mm), k is the growth coefficient (year ) and t0 is the hypothetical age
(years) at which fish would have zero length.
The growth curves for females and males of A. butcheri in each estuary were compared using likelihood-ratio tests (Cerrato, 1990). For each comparison, the differences between the negative log-likelihood (NLL) was obtained by fitting a common growth curve to the data for both sexes and by fitting separate growth curves to each sex, and the resultant test
n statistic multiplied by two. The NLL was calculated as NLL ln(2 ) lnˆ 1, 2 where n is the number of fish, ln is the natural logarithm, π is the natural constant pi, and
ˆ 2 n L j L j ˆ is determined as: where L and Lˆ are the observed and expected j1 n j j lengths of the j’th fish. The hypothesis that the growth of the two sexes could appropriately be represented by a single growth curve was rejected at the α = 0.05 level of significance if
the above test statistic exceeded ( ), where q is the difference between the numbers of parameters in the two approaches, i.e. 3 (Cerrato, 1990).
The above approach was then adopted to compare the growth of A. butcheri in different periods in both the Swan River Estuary and between restocked and wild A. butcheri in the Blackwood River Estuary. Precise details of the comparisons made in these two estuaries are given in chapters 3 and 5, respectively.
34
2.3.3 Length and age at maturity
The gonads of each A. butcheri were weighed to the nearest 0.01 g and assigned to one of the following stages on the basis of their macroscopic appearance: I/II = virgin or resting adult, III = developing, IV = maturing, V = prespawning, VI = spawning, VII = spent and
VIII = recovering spent (Laevastu, 1965). The trends exhibited by the prevalence of the different stages in ovarian development in sequential months, allied with those of gonad weights, were used to determine the spawning period and the birth date of A. butcheri., taken as the approximate time of peak spawning.
Logistic regression analysis was used to determine the probability P that a female or male of a given length possessed gonads at stages III-VIII during the spawning season in the Swan River and Blackwood River estuaries. The logistic equation is
{ [ ( ) ( ) ( )]}, where L is the total length of the fish in mm and L50 and L95 are the lengths at which 50% and 95% of fish attain maturity, respectively. The values of L50 and L95 were estimated by minimising the negative log- likelihood (NLL) using R (R Core Team, 2014), determined as
NLL M j lnPj 1 Pj ln1 Pj where Mj is the maturity class of the j’th fish, i.e. 0 j
= immature and 1 = mature, and Pj is the probability of the j’th fish being mature.
Likelihood-ratio tests were used to determine whether, for each sex, the ogives relating maturity to length were different and, if so, whether this could be attributed to a difference in the values of L50. The analysis was repeated using OpenBUGS (Surhone et al., 2010), a successor to WinBUGS (Lunn et al., 2000), with two chains, 1,004,000 iterations, a burn- in of 1000 and a thinning interval of 20, to obtain estimates of the approximate 95% credible intervals for the L50s and L95s and the expected proportion of mature fish at each of a range of total lengths. The prior distribution for each parameter was represented by a non-informative normal distribution. The above approach was also used to relate the
35 probability that a female or male possessed mature gonads to the age of the fish and to test whether, for each sex, the ogives relating maturity to age were different and, if so, whether this could be attributed to a difference in the values of A50. OpenBUGs was again employed, but with 2,004,000 iterations of two chains, a burn-in of 4000 and thinning interval of 40.
36
Chapter 3
Marked deleterious changes in the condition, growth and maturity schedules of
Acanthopagrus butcheri in the Swan River Estuary reflect environmental degradation
3.1 Introduction
The high productivity of estuaries (Schelske & Odum, 1962; Whittaker & Likens, 1975;
Elliott & Whitfield, 2011) helps account for the importance of these systems for many marine fish species throughout the world and especially during juvenile life (Blaber &
Blaber, 1980; Potter et al., 1990; McLusky & Elliott, 2004; Able & Fahay, 2010). This rich food source is also crucial for those species that complete their life cycles within estuaries, which are particularly abundant in temperate microtidal regions of the southern hemisphere, such as those of south-western Australia (Potter & Hyndes, 1999; Whitfield,
1999; Chuwen et al., 2011).
In view of their ecological significance, it is disturbing that temperate estuaries are also among the most degraded of all marine ecosystems (Jackson et al., 2001), with the prevalence of those that have become eutrophic due to anthropogenic impacts increasing markedly during recent decades (Kennish & Townsend, 2007; Vaquer-Sunyer & Duarte,
2008). Furthermore, eutrophication in estuaries that have strong vertical stratification leads to the accumulation of organic matter in their deeper waters and thus ultimately to hypoxia
(Diaz, 2000; Eby et al., 2005; Breitburg et al., 2009). Hypoxic conditions can be lethal to fish (Kramer, 1987; Eby et al., 2005; Breitburg et al., 2009) or lead to reductions in their body condition, growth rates and metabolism (Pichavant et al., 2000, 2001; Wu, 2002; Eby et al., 2005) and have detrimental effects on their early development (Wannamaker &
Rice, 2000; Thronson & Quigg, 2008; Hassell et al., 2008a, b) . While fish sometimes have the opportunity to move from hypoxic to better oxygenated habitats (Pihl et al., 1992;
37
Howell & Simpson, 1994; Thiel et al., 1995), such as from deeper to nearshore, shallow waters in estuaries, this can lead to density-dependent effects on growth (Eby & Crowder,
2002; Campbell & Rice, 2014) and/or to exposure to suboptimal temperatures and salinities (Kramer, 1987; Craig et al., 2001; Eby et al., 2005).
Most studies of the ways in which reductions in growth are reflected in changes in the maturity schedules of fish populations have been undertaken on populations subjected to substantial fishing pressure (e.g. Morita & Fukuwaka, 2007; Stahl & Kruse, 2008). In many cases, such reductions in growth were accompanied by a decline in the length at maturity (Morita & Morita, 2002; Morita & Fukuwaka, 2007) and an increase in the age at maturity, which is consistent with, for example, the model of Stearns (1992) in which such trends represent a balance between reducing fecundity and increasing mortality. Certain studies support this model (Helle & Hoffman, 1998; Morita & Morita, 2002; Grover, 2005;
Morita & Fukuwaka, 2007), but none has demonstrated quantitatively, and in a composite manner, that the body condition and growth decline and maturation schedules of a species in an estuary change when that system becomes degraded.
During the last 20 years, the Swan River Estuary has become increasingly eutrophic (Stephens & Imberger, 1996; Hamilton & Turner, 2001; Thomson et al., 2001;
Robson & Hamilton, 2003; Hallegraeff et al., 2010; Kristiana et al., 2012). Thus, while phytoplankton blooms were rarely observed in the 1980s (Zammit et al., 2005), they occurred intermittently in the 1990s (Viney & Sivapalan, 2001) and in most years in the
2000s (Robson & Hamilton, 2003; Hallegraeff et al., 2010; Kristiana et al., 2012). Indeed, this system was one of the two most hypertrophic of the 131 estuarine coastal ecosystems worldwide for which data were recently collated by Cloern et al. (2014). The upper reaches of the Swan River Estuary often experience strong stratification during the warm summer months, when freshwater discharge is typically limited and the salt wedge penetrates
38 rapidly upstream along the bottom of the estuary. As the deeper waters are poorly flushed, they have a long residence time (Chan & Hamilton, 2001),substantial amounts of organic material are thus retained in those waters, resulting in high biological oxygen demand and therefore reductions in oxygen concentrations (Norlem et al., 2014; Tweedley et al., 2014).
This problem has become exacerbated over the last three decades by a decline in freshwater discharge (Western Australian Department of Water, 2015), due to a reduction in rainfall (Australian Bureau of Meteorology, 2014), and a greater penetration of the salt wedge upstream. This has led to elevated nutrient concentrations (Robson et al., 2008) and a greater prevalence of prolific algal and cyanobacterial blooms (Twomey & John, 2001), some of which are toxic (Orr et al., 2004; Mooney et al. 2009) and have resulted in an increased prevalence of fish kills (Smith, 2006; Kristiana et al., 2012).
Acanthopagrus butcheri is abundant in the upper reaches of estuaries throughout southern Australia (Sarre & Potter, 2000; Nicholson et al., 2008; Williams et al., 2012). As with other sparids, certain biological characteristics of A. butcheri are plastic, varying markedly among estuaries with different environmental characteristics. Thus, for example, among four south-western Australian estuaries, the total lengths reached after three years ranged widely from 146 to 266 mm (Sarre & Potter, 2000) and the L50s and A50s for maturity for females ranged markedly from 157 to 218 mm and 1.9 to 4.3 years, respectively. It is also evident that maturity tends to occur at a smaller size and older age in those estuaries in which growth is less rapid (Sarre & Potter, 1999, 2000), which is consistent with the model of Stearns (1992).
The present study has determined certain biological characteristics of A. butcheri in the Swan River Estuary in 2007-11 and compared these with those derived from data for that estuary in 1993-95 (Sarre & Potter, 1999, 2000). These comparisons were used to test the hypothesis that increased environmental degradation in this estuary during recent years
39 has been accompanied by a decline in the body condition, growth and length at maturity of
A. butcheri and an increase in its age at maturity. The trends exhibited by the catch data derived from fishery-independent sampling over time were compared to determine whether there was evidence that an increase in the extent of hypoxia in deeper waters had been accompanied by a reduction in the relative abundance of A. butcheri in those waters and a resultant increase in nearshore, shallow waters.
40
3.2 Materials and Methods
3.2.1 Description of the Swan River Estuary
The Swan River Estuary, which is located on the lower west coast of Australia ( 31o 57’ S and 116o 04’ E), flows through Perth, the capital city. This permanently-open estuary enters the Indian Ocean at the port city of Fremantle and, with an area of 55 km2, is the second largest estuary in south-western Australia (Brearley, 2005). The Swan River
Estuary has two main tributaries, the Swan and Canning rivers (Fig. 3.1).
Figure 3.1 Map showing the location of the sites sampled in the upper region of the Swan
River Estuary.
As with many other south-western Australian estuaries, the Swan River Estuary comprises three morphologically distinct regions (Fig. 2.1). The lower estuary comprises a long (~7.5 km) and narrow (~200 m) entrance channel, while the middle estuary consists
41 of two basins, i.e. Melville Water and Perth Water. Melville Water is 12 km long and 2 km wide and has a maximum depth of 21 m. Perth Water is 2.5 km long, 1.5 km wide, and typically ~1 m deep (Hodgkin, 1987; Thomson et al., 2001). The upper estuary, which includes the 38 km long saline and tidally-influenced lower reaches of the Swan River, and of the corresponding 6 km region in the Canning River (Stephens & Imberger, 1996).
The Swan River Estuary and its catchments (141,000 km2) have been extensively modified since European settlement in 1829 (Smith, 2006). Prior to settlement, tidal exchange was restricted, due to the shallowness of the entrance channel of this estuary, i.e. maximum depth was 2 m, which acted as a partial barrier between the estuary and the ocean. Since the 1890s, the entrance channel has been deepened (ca 13 m) and widened on several occasions, thereby increasing tidal exchange (Brearley, 2005). As a result, the salinities in the two basins and upper estuary now remain elevated for several months of the year (Thomson et al., 2001). The damming of several tributary rivers, e.g. the Helena
River by the Mundaring Weir and the Canning River by the Canning Dam, has resulted in a marked reduction in freshwater flow and thus of discharge into the estuary. The extent of such reductions was augmented by the marked reduction that has occurred during recent years in the amount of freshwater entering the catchment of this estuary annually, with these declining to about half of the level recorded in the years prior to 1975 (Smith, 2006).
The Swan-Avon River Catchment, which contributes ca 85% of the total freshwater input to the estuary, has also been ~70% cleared (Kurup & Hamilton, 2002; Brearley,
2005). To manage flash flooding in the 1950-60s (largely resulting from the clearing of land), a 187 km stretch of the Swan-Avon River was straightened and deepened, and riparian vegetation and snags removed (Brearley, 2005). It is estimated that the above modifications led to a five-fold increase in stream flow, a forty-fold increase in sedimentation and a sixteen-fold increase in nutrient input (Viney & Sivapalan, 2001).
42
3.2.2. Physico-chemical data and sampling regimes
Measurements of freshwater discharge, entering the Swan River Estuary at gauging station
616011 upstream of the study area, were provided by the Western Australian Department of Water. That department also provided weekly vertical profiles of oxygen concentration, temperature and salinity, recorded since 1995 using a YSI sonde (model 6600V2) or equivalent, at five of its sampling stations (KIN, MAY, RON, STJ, SUC) within the study area. The average oxygen concentration in the bottom 1 m of the water column at each station was calculated for each weekly sample from 1995 up to and including those in
2010, the last year for which data have thus far been collated (data collection and calculations by M. Hipsey, University of Western Australia). Linear interpolation between stations was then used to determine the average weekly oxygen concentrations along the central channel of the upper estuary for each year. Oxygen concentrations <4 mg L-1 are regarded as hypoxic. The percentages of the total distance along the channel that was hypoxic in each week were then determined and used to calculate each annual mean.
During the fieldwork for this thesis, Acanthopagrus butcheri was sampled at nine sites in the upper Swan River Estuary (Fig. 3.1). Nearshore, shallow waters were sampled using the 21.5 and 41.5 m long seine nets in each season between winter 2007 and autumn
2011, and thus seasonally in four consecutive 12 month intervals (Table 3.1). Gill netting was undertaken in waters adjacent to each seine net site at the same time (season) as seine netting between winter 2007 and autumn 2009 (i.e. seasonally over two consecutive 12 month intervals), after which it was discontinued as it was catching few fish. The data on condition and growth of A. butcheri caught by the above seine and gill netting are subsequently regarded as representing the 2007-11 period.
43
Table 3.1 Regime for sampling Acanthopagrus butcheri in the upper Swan River Estuary using seine and gill nets between 1993 and 2011.
Sampling seasons Period Net type Number of sites Spring 1993-Autumn 1995 1993-95 41.5 m seine 6 (lower sites) Spring 1993- Summer 1995 Gill 9 Summer 2000-Spring 2000 2000 Gill (catch rate only) 9 Spring 2003-Winter 2004 2003-04 21.5 m seine 6 ( lower sites) 41.5 m seine 6 (lower sites) Gill 9 Winter 2007-Autumn 2009 2007-11 21.5 m seine 9 41.5 m seine 9 Gill 9 Winter 2009-Autumn 2011 2007-11 21.5 m seine 9 41.5 m seine 9
The data derived from samples collected seasonally from nine sites in 2007-11 (see
above) were collated with those obtained monthly between winter 1993 and autumn 1995
for the six lower sites using a 41.5 m long seine net and for those obtained monthly
between spring 1993 and summer 1995 for all nine sites using the same 160 m gill net
(Sarre & Potter, 2000; Table 3.1). This earlier period is subsequently referred to as the
1993-95 period when making comparisons with the condition and growth of A. butcheri in
2007-11. Comparisons were also made with the data from a subsample of A. butcheri
collected in each season between spring 2003 and winter 2004 using both seine nets
(21.5 m and 41.5 m) at the same six lower sites and the gill net at the same nine sites as in
1993-95 (S. Hoeksema, unpublished data; Table 3.1). Each net had the same dimensions
and mesh sizes as in the other studies. This 12 month interval is subsequently referred to as
the 2003/04 period.
44
Catches of A. butcheri during the summer and spring of 2000, obtained as part of another study that involved gill netting at the same nine sites as in 1993-95 and 2007-11
(S. Hoeksema, unpublished data), were used to provide catch per unit effort data for 2000 to complement those for other years.
3.2.3 Assessment of body condition
Following the approach of Froese (2006), the length-weight relationships of each sex of A. butcheri in 2007-11 and 1993-95 were calculated from data for ten randomly-selected fish from each of three length categories, i.e. <160, 180-220, >240 mm, in each calendar season, except for winter in which ten fish were not always available for each of these length categories due to the pronounced tendency for A. butcheri to be flushed downstream in that season when freshwater discharge is far greater than in other seasons (Sarre &
Potter, 1999). The parameters period and period of the length-weight relationships,
period period were estimated by fitting linear equations of the form:
( ) ( period) period ( ) period , to the masses (g) and total lengths (mm) for the 90 females and 90 males in the 2007-11 and 1993-95 periods. In this equation, ln refers to the natural logarithm, and are the observed mass and length, respectively, of the j’th fish, and the deviations of the logarithms of the observed masses from the values predicted by the regression equations were assumed to be normally distributed with means
of zero and variances that differed between the two periods, i.e. period ( period). In the case of both sexes, the linear equations were fitted simultaneously to the data for the two periods using R (R Core Team, 2013). Likelihood-ratio tests were employed to compare, for each sex, the model that assumed a common length-weight relationship for the two periods against the best fitting (i.e. smaller negative log-likelihood, NLL) of the
45 two three-parameter models, which assumed that either the slope or the intercept (but not both) remained constant, and that latter model against the model that assumed that both the intercept and slope of the regression equation differed between the two periods.
Approximate 95% confidence intervals for period and period were estimated as the 2.5 and
97.5 percentiles of the distributions of 1000 parameter estimates obtained by drawing (with replacement) a sample from the data for each sex and period (with the same sample size as the original sample) and refitting the linear regression equation to that sample. When calculating the expected masses of fish from their total lengths, the correction factor
exp( ̂period ) was employed (Beauchamp & Olson, 1973) to account for the bias associated with back-transformation of the predicted values from the values of their natural logarithms. Thus, for each period and sex, the back-transformed allometric equation
period became [ period exp( ̂ )] , where ̂period is the estimated variance of the residuals from the fitted linear equation.
The allometric condition factor, , for fish j of mass and total length for
each period and sex was calculated as ⁄ using the value of period that was estimated for the length-weight relationship for that period (Tesch, 1968). A t-test was used to compare, for each sex, the mean value of the resulting estimates of CF for 2007-11
period with that for 1993-95. As period exp( ), where ( period), the expected
̅̅̅̅ value of for the period and sex was calculated as period period exp( period⁄ ).
Approximate 95% confidence intervals for ̅̅̅ ̅period were calculated for each period and sex from the 1000 bootstrapped samples that were drawn to estimate the confidence intervals for the parameters period and period of the weight-length relationships.
46
3.2.4 Growth and reproductive biology
Likelihood-ratio tests demonstrated that the growth of females and males in the Swan
River Estuary in both 2007-11 and in 2003/04 were significantly different (P<0.01), and the same had previously been shown to apply in 1993-95 (Sarre & Potter, 2000). Thus, comparisons between the growth curves in these three periods were undertaken separately for each sex. As the tests demonstrated that the growth curves differed between these periods (see Results), likelihood-ratio tests were also used to determine which of the growth parameters for each sex in those three periods likewise differed.
As described in Chapter 2, logistic regression analysis was used to determine the probability P that a female or male of a given length possessed gonads at stages III-VIII in
November, i.e. just after the mid-point of the spawning period in October, when no immature fish would be expected to become mature during the current spawning season and would have included all fish that had spawned or were destined to spawn in a given spawning season. This procedure used data derived from samples collected in the
Novembers of 1993 and 1994 collectively, and the Novembers in each year between 2007 and 2010, collectively.
3.2.5 Measures of relative abundance
The numbers of A. butcheri caught on each sampling occasion in each season except winter (see above for rationale), using the 41.5 m seine net at six sites in nearshore, shallow waters of the upper estuary in different years (Table 3.1), were converted to a density (fish 100 m-2). The numbers of fish caught using the multi-mesh gill net at nine sites in offshore, deeper waters of the upper estuary on each sampling occasion were expressed as number of fish gill net-1. Although A. butcheri typically occupies the lower and saline regions of the tributary rivers, which constitute the upper estuary, many
47 individuals are flushed downstream during winter when freshwater discharge increases sharply (Sarre & Potter, 1999). The limited data for this season were thus not included in analyses of the densities and catch rates. Note also that, except in 2000, the mean densities and catch rates and 95% confidence interval for a ‘year’ were calculated using consecutive seasonal data for the spring in one year and summer and autumn of the following year and thus encompass seasons in two consecutive calendar years.
Prior to subjecting the densities and catch rates for the different ‘years’ to Analysis of Variance (ANOVA), examination of these data demonstrated that both required ln(x+1) transformations to meet the test assumption of homogeneous dispersions among a priori groups (see Clarke & Warwick, 2001). If ANOVA detected a significant difference between the densities and/or catch rates among years (P<0.05), Scheffé’s test was used to identify where the variable differed significantly between each pair of years.
3.3 Results
3.3.1 Decline in freshwater discharge and relationship with hypoxia
The annual discharge of freshwater entering the upper estuary declined markedly during the 20 years between 1992 and 2011 (Fig. 3.1, r2=0.35, P<0.01). The influence of this discharge on oxygen concentrations is reflected in the close relationship between the annual values for the average extent of hypoxia and discharge (Fig. 3.2, r2=0.50, P<0.005), reflecting the tight coupling between the persistence of the salt-wedge and freshwater discharge.
48
Figure 3.1. Freshwater discharge into the Swan River Estuary in the 20 years between 1992 and 2011. Data provided by the Western Australian Department of Water.
Figure 3.2. Relationship between the annual average extent of hypoxia (%) in the upper Swan Estuary and freshwater discharge between 1995 and 2010. The extent of hypoxia in each year was estimated by interpolating available oxygen profile data for each sampling cruise within the study domain for each weekly profile and then averaging the values for each year.
49
3.3.2 Body condition and growth
The three-parameter model, representing the relationship between the natural logarithms of body mass and length for females of A. butcheri in the Swan River Estuary in 1993-95 and
2007-11, when using linear equations with a common slope and different intercepts
(NLL=-234.2), provided a better fit than the alternative three-parameter model, which assumed a common intercept and different slopes (NLL=-234.0). It also provided a significantly better fit (P<0.001) than the two-parameter model that assumed a common linear relationship between body mass and length in the two periods (NLL=-217.0). A further increase in model complexity, by allowing both the intercepts and slopes to differ
(NLL=-234.6), failed to significantly improve the fit (P>0.05). The results for males were similar to those for females, with the model assuming a common slope and different intercepts (NLL=-229.6) producing a slightly better fit to the data than that assuming a common intercept and different slopes (NLL=-229.3) and improving the fit significantly
(P<0.001) over that provided by the model with a common slope and common intercept
(NLL=-216.4). Again, extension of the model to allow both the intercepts and slopes to differ (NLL=-229.9) failed to significantly improve the fit (P>0.05). Thus, the linear relationships between the natural logarithms of body mass and total length for each sex in
1993-95 and 2007-10 were best represented by equations with different intercepts and a common slope (Table 3.2).
The mean values of the allometric condition factors for each sex in 1993-95 differed significantly (both P<0.001) from those in 2007-11 (Table 3.2). As the power terms of the length-weight relationships for neither sex differed significantly between
1993-95 and 2007-11, the allometric condition factors for fish from the two periods were directly comparable as a common power term could be employed to calculate their values.
50
Thus, between 1993-95 and 2007-11, the expected values of the allometric condition factor for females and males declined by 6 and 5%, respectively.
period Table 3.2. Estimated values of parameters of relationships ( period ) between body mass (g) and total length (mm) for Acanthopagrus butcheri collected from the Swan River Estuary in 1993-95 and 2007-11, with estimates of expected values of the allometric condition factors for these periods, ̅̅̅ ̅period, and with bootstrap estimates of 95% confidence intervals. Females Males 1993-95 2007-11 1993-95 2007-11 -6 -6 -5 -6 period Estimate 9.370 10 8.883 10 1.147 10 1.086 10 Lower 7.603 10-6 7.153 10-6 9.389 10-6 8.921 10-6 Upper 1.152 10-5 1.091 10-5 1.423 10-5 1.350 10-5 period Estimate 3.116 3.116 3.080 3.080 Lower 3.076 3.076 3.040 3.040 Upper 3.155 3.155 3.116 3.116 -6 -6 -5 -6 ̅̅̅ ̅period Estimate 9.391 10 8.851 10 1.149 10 1.089 10 Lower 7.622 10-6 7.170 10-6 9.408 10-6 8.942 10-6 Upper 1.155 10-5 1.093 10-5 1.426 10-5 1.353 10-5
The von Bertalanffy growth curves provided a good fit to the lengths at ages of females and males of A. butcheri in 1993-95 and 2007-11 (Table 3.3; Fig. 3.4). The curve for the females in 1993-95 lay above and differed significantly (P<0.001) from that of this sex in 2007-11 and the same was true for those of males in these two periods (Fig. 3.5).
Furthermore, the growth parameters k and L∞ for each sex differed significantly between periods (all P<0.001) and were less in 2007-11 than in 1993-95. The values for the t0 of both sexes were slightly more negative in 2007-11 than in 1993-95 (both P<0.001).
Based on the von Bertalanffy growth parameters, the TLs of female A. butcheri at ages two and six years were 142 and 248 mm, respectively, in 2007-11 and thus far less than the corresponding TLs of 203 and 366 mm in 1993-95. Likewise, the estimated TLs of males of 144 and 235 mm, respectively, at those same ages in 2007-11 were far less than the corresponding values of 199 and 354 mm in 1993-95.
51
Table 3.3. von Bertalanffy growth curve parameters and their 95% confidence limits for females and males of Acanthopagrus butcheri caught in the upper Swan River Estuary in the 1993-95, 2003-04 and 2007-11 periods. L∞, asymptotic total length (mm); k, growth -1 2 coefficient (years ); t0, hypothetical age (years) at which fish would have zero length; r , coefficient of determination. n is the number of fish. 2 Period L∞ k t0 r n 1993-95 Females Estimate 441 0.29 -0.12 0.94 680 Lower 427 0.27 -0.19 Upper 454 0.31 -0.06 Males Estimate 424 0.29 -0.19 0.99 754 Lower 413 0.27 -0.26 Upper 436 0.31 -0.13 2003/04 Females Estimate 322 0.35 -0.13 0.96 111 Lower 284 0.26 -0.21 Upper 371 0.44 -0.06 Males Estimate 351 0.28 -0.13 0.99 88 Lower 279 0.19 -0.42 Upper 300 0.42 -0.13 2007-11 Females Estimate 341 0.19 -0.84 0.89 3438 Lower 321 0.17 -0.92 Upper 362 0.22 -0.76 Males Estimate 288 0.25 -0.76 0.90 3248 Lower 274 0.22 -0.85 Upper 306 0.28 -0.68
The growth curves for the more limited samples for females and males in 2003-04 lay between and differed significantly from those for the corresponding sex in 1993-95 and
2007-11 (Fig. 3.5, all P<0.001). In the case of each sex, the value for k in 2003-04 differed significantly from that in 2007-11 (both P<0.001), and the same was true for females for
2003-04 vs 1993-95 (P<0.05). The L∞ for females and males in 2003-04 each differed significantly from that of the corresponding sex in 1993-95 and 2007-11 (all P<0.001), except with females in 2003-04 vs 2007-11 (P>0.05). The values for t0 in 2003-04 were significantly different from those for females (P<0.001) and males (P<0.05) in 2007-11.
52
Figure 3.4. von Bertalanffy growth curves for female and male Acanthopagrus butcheri caught in the Swan River Estuary in 1993-95, 2003-04 and 2007-11.
53
Figure 3.5. Comparisons of von Bertalanffy growth curves for female and male Acanthopagrus butcheri in the Swan River Estuary in 1993-95, 2003/04 and 2007-11.
3.3.3. Lengths and ages at maturity
The smallest TL of the mature female and male A. butcheri during the spawning period were 170 and 151 mm, respectively, in 1993-94, and 144 and 143 mm, respectively, in
2007-10. As the prevalence of mature females and males increased with length, the percentage of mature individuals in each corresponding length class was less in 1993-94 than in 2007-10 (Fig. 3.6). Although the ogive describing the relationship between the proportion of mature females and total length in 1993-94 was not significantly different from that for 2007-10 (NLL=75.5 vs 78.7 for a common curve, P<0.05), and the same was true for males (NLL=56.3 vs 57.9, P>0.05), the L50s at maturity for females and males in
1993-94 (174 and 172 mm) were substantially greater than the 155 and 156 mm for the corresponding sex in 2007-10 (Table 3.4, Fig. 3.6).
54
Table 3.4 Estimates of the total lengths (mm) and ages (years), and associated 95%
credible intervals, at which 50% (L50 and A50) and 95% (L95 and A95) of female and male Acanthopagrus butcheri were mature in the Swan River Estuary in 1993-95 and in 2007- 10. ML = Maximum likelihood. Lower and upper credible limits were determined using OpenBUGS. n is the number of fish.
Period Sex Statistic nlength nage 1993-95 Females ML estimate 174 210 1.9 2.2 156 103 Lower 159 194 1.6 2.0 Upper 187 234 2.0 2.6 Males ML estimate 172 189 2.0 2.1 196 136 Lower 152 175 1.4 1.8 Upper 181 206 2.0 2.2 2007-10 Females ML estimate 155 189 2.5 4.7 422 422 Lower 151 181 2.4 4.4 Upper 161 198 2.7 4.1 Males ML estimate 156 183 2.5 4.5 329 329 Lower 150 176 2.3 4.2 Upper 160 192 2.7 4.9
Only one fish had reached maturity by the end of its first year of life in 2007-10
(Fig. 3.7). However, while most females and males had become mature at the end of their
second year of life in 1993-94, only 56% of females and 35% of males had attained
maturity at this age in 2007-10 (Fig. 3.7). The ogives describing the relationships between
the proportion of mature females and age in 1993-94 and 2007-10 were significantly
different (NLL=82.9 vs 99.7, P<0.001) and the same was true for males (NLL= 55.2 vs
94.8, P<0.001). The A50 of 1.9 years for females at maturity in 1993-94 was less than and
significantly different to the 2.5 years for females at maturity in 2007-10 (P<0.001) and the
same significant trend (P<0.001) was exhibited by the corresponding values of 2.0 and 2.5
years for males at maturity in those periods (Table 3.4, Fig. 3.7).
55
Figure 3.6. Percentage frequency of occurrence of Acanthopagrus butcheri with immature (white bars) and mature gonads (grey bars) in sequential 20 mm total length classes during the spawning seasons of this species in the Swan River Estuary in different years. Females in (a) 1993 and 1994 collectively and (b) 2007-10 and males in (c) 1993 and 1994 collectively and (d) 2007-10. Logistic curves (solid lines) and their 95% credible intervals (dotted lines) for the probability that a fish at a given length is mature were estimated using OpenBUGS. Sample size for each length class is shown. Comparisons of logistic curves for females and males in 1993 and 1994 (solid line) and 2007-10 (dashed line) are shown in e and f.
56
Figure 3.7. Percentage frequency of occurrence of Acanthopagrus butcheri with immature (white bars) and mature gonads (grey bars) in sequential ages during the spawning seasons of this species in the Swan River Estuary in different years. Females in (a) 1993 and 1994 collectively and (b) 2007-10 and males in (c) 1993 and 1994 collectively and (d) 2007-10. Logistic curves (solid lines) and their 95% credible intervals (dotted lines) for the probability that a fish at a given age is mature were estimated using OpenBUGS. Sample size for each length class is shown. Comparisons of logistic curves for females and males in 1993 and 1994 (solid line) and 2007-10 (dashed line) are shown in e and f.
57
3.3.4 Catch rates in deeper waters and densities in shallow water
ANOVA demonstrated that the mean annual catch rates of A. butcheri, derived from samples obtained using gill nets at nine sites in offshore, deeper waters, differed significantly among years (F (5, 186)=4.4, P<0.01). Mean catch rates ranged downwards from a maximum of 3.7 fish net-1 in 1993-94 to a minimum of 0.3 fish net-1 in 2007-08
(Fig.3.8), with Scheffé’s test showing that the mean catch rates in 1993-94 and in 2007-08 were significantly different (P<0.01).
Figure 3.8. Back-transformed mean and associated 95% confidence intervals for a) catch rates derived from gill net samples in offshore, deeper waters and b) densities derived from seine net samples in nearshore, shallow waters of Acanthopagrus butcheri in the Swan River Estuary in the various years of sampling. Number of samples for each year is shown.
58
The mean densities of A. butcheri, derived from samples obtained using a 41.5 m seine net at six sites in nearshore waters, differed significantly among years (F(6,
17)=14.2, P<0.001). The mean densities in 1993-94 and 1994-95 were ≤1.0 fish 100 m-2 and thus far less than in 2003-04 and 2007-08 to 2010-11, which ranged from 3.6 to 8.4 fish 100 m-2 (Fig. 3.8). The mean density of fish in 1993-94 and 1994-95 differed significantly from those in 2003-04, 2007-08, 2008/09, 2009-10 and 2010-11 (P<0.001).
3.4 Discussion
The present study provides the first demonstration that the condition factor, growth and maturity indices of a fish species, whose entire life cycle is confined to estuaries, have all changed in association with marked detrimental modifications to the environment.
Furthermore, each change was shown to occur in the direction hypothesised from the results of other relevant studies (see below) and the model of Stearns (1992). When considering the implications of the inter-period comparisons of particularly the growth of
A. butcheri, it is important to recognise that the growth curves for 1993-95 and for 2007-11 represent the cumulative effects of the growth of the individuals during those periods and, in the case of older fish, that which occurred prior to the commencement of those two periods. The possibility that changes in the biological characteristics of A. butcheri were induced by an increase in intensity of fishing, as described for a number of other species
(e.g. Sharpe & Hendry, 2009; Devine et al., 2012; Diaz & Heino, 2014), can be rejected for the following reasons. The small seasonal commercial fishery for A. butcheri in the
Swan River Estuary has been negligible since 2007, due to a government buy-back of commercial fishing licences (Smith et al., 2014), and the recreational catch per unit effort data indicate that the abundance of A. butcheri has remained relatively stable since the early 1990s (Smith, 2006; Smith et al., 2014).
59
3.4.1 Condition and growth
As body condition is a reliable indicator of the health of a fish (Peig & Green, 2009), it is highly relevant that the allometric condition factor for female and male A. butcheri in the upper Swan River Estuary declined significantly between 1993-95 and 2007-11.
Furthermore, this decline in condition was accompanied by a marked reduction in the growth of both sexes, with the growth curve for the intervening 2003-04 period occupying an approximately intermediate position between the earlier and later periods. Indeed, on the basis of von Bertalanffy growth parameters, the predicted lengths at two and six years in age declined between 1993-95 and 2007-11 by as much as 32 and 34%, respectively, with females, and by 67 and 73%, respectively, with males.
The significant decline in the condition factor and growth of A. butcheri in the Swan
River Estuary between 1993-95 and 2007-11 parallels the observation by Eby et al. (2005) that the condition and growth of the Atlantic Croaker Micropogonias undulatus in the
Neuse River Estuary was least in years when the extent of hypoxia was greatest. The results of controlled laboratory experiments, in which M. undulatus was exposed to four and ten weeks of hypoxia subsequently supported the conclusion that the condition and growth of this species can be adversely affected by reduced oxygen concentrations (Mohan et al., 2014). A number of laboratory studies on other fish species have produced similar results, particularly with growth (e.g. Kramer, 1987; Pichavant et al., 2001; Roberts et al.,
2011). Declines in the condition and growth of fishes exposed to hypoxia have been attributed, inter alia, to a reduction in metabolism and changes in the quantity and quality of potential food (Pichavant et al., 2001; Wu, 2002; Eby et al., 2005; Powers et al., 2005).
In the context of food, the volumetric contributions made to the diets of A. butcheri in the
Swan River Estuary by low-calorie food, such as algae, macrophytes and detritus, increased from 15 to 30% between 1993-95 and 2007-11, whereas high-calorie prey, such
60 as bivalve molluscs, declined from 64 to 19% (cf Sarre et al., 2000; Linke, 2011). These trends parallel those recorded during studies of M. undulatus in the Neuse River Estuary, which showed that, following exposure to hypoxia, the density of food and particularly of bivalves (Macoma spp.) declined and was thus considered to be one of the factors that led to reductions in the growth (Eby et al., 2005). Unfortunately, there are no quantitative data to facilitate comparisons between the quantity and quality of the benthic macroinvertebrates and macroalgae of A. butcheri in offshore and nearshore waters in the two periods and which are known to constitute the diet of this species (Sarre et al., 2000;
Chuwen et al., 2007).
In the context of factors that could also have attributed to the decline in the condition and growth of A. butcheri between 1993-95 and 2007-11, it is pertinent that data, derived from gill and seine net catches, provided strong circumstantial evidence that the abundance of A. butcheri in offshore, deeper waters has declined, whereas that in nearshore, shallow waters has increased markedly. This therefore constitutes habitat compression of the type induced by hypoxia among demersal fish in the Neuse River
Estuary on the east coast of North America (Eby et al., 2005), pelagic fish in tropical waters in the eastern Pacific (Prince & Goodyear, 2006) and certain species in other environments (e.g. Pihl et al., 1991; Tyler & Targett, 2007; Craig, 2012). A tendency, during the past 10 to 20 years, for A. butcheri to have moved from deeper to shallow waters, and the demonstration that its density in the shallow waters has become far greater, raises the strong possibility that density-dependent effects may also have played an important role in leading to a decline in the growth of this species over time. It is thus noteworthy that density dependent factors were also considered responsible for the decline in growth of M. undalatus in the Neuse River Estuary (Eby et al., 2005). In addition, the use of fishery-independent data demonstrated that the growth of the 0+ and 1+ age classes
61 of the Red Drum Sciaenops ocellatus were negatively related to the density of their own age classes (Bacheler et al., 2012). Furthermore, analysis of long-term records of size at age and biomass data for 16 fish populations revealed that the growth of individuals in nine of those populations was inversely related to density (Lorenzen & Enberg, 2002).
3.4.2. Length and age at maturity
As the ogives relating maturity to length in females and to age in both sexes of A. butcheri differed significantly between 1993-94 and 2007-10, the maturity characteristics of this species have changed between the two periods. It was particularly striking that, between
1993-94 and 2007-10, the L50 of females and males at maturity both declined by 10%, respectively, whereas, in contrast, the A50 of females and males at maturity increased by the equivalent of 32 and 25%, respectively. The direction of the changes between the two periods is consistent with the model of Stearns (1992), which predicted that reductions in growth will lead to maturity occurring at a smaller size and older age (see also Chen &
Mello, 1999; Morgan & Colbourne, 1999). This finding is also consistent with the results of a three year tag-recapture study, which demonstrated that the slower-growing residents of the White Spotted Charr Salvelinus leucomaenis reached maturity at a smaller size but older age than faster-growing individuals, a feature considered to reflect adaptive phenotypic plasticity (Morita & Morita, 2002). For A. butcheri, the changes in both the length and age at maturity were pronounced and developed within a relatively short period, highlighting the plasticity of the biological characteristics of this species and thus its ability to respond to the effects of environmental change.
The dynamic nature of estuaries means that few species are able to complete their life cycle within these systems. Most of the exceptions are small and short-lived species and thus mature at a young age and often at the end of the first year of life. The investment
62 of energy to reproduction at an early age is a particularly useful strategy for species that reside in unpredictable environments and therefore subjected to high natural adult mortality. This strategy thus enables sufficient offspring to be produced and thereby ensure the production of future generations. This type of strategy contrast with that of many fish that reside in stable environments, which are long-lived and mature at an old age. Based on the relationship between maximum age and age at maturity, derived by Froese and
Binohlan (2000) from 432 populations, a fish with a maximum age of 30 years would, on average, attain maturity at ~ 9 years of age.
Species, such as A. butcheri, have adopted an intermediate strategy, in which they mature at an early age but have a relatively long life span (up to ~30 years).
The flexibility of species, such as A. butcheri, to allocate energy into growth and/or reproduction also allows them to delay maturity, in favour of allocation of energy to somatic growth, when resources become limited (Grover, 2005). Although delayed maturation and the resultant increase in generation time can lead to reduced population growth rates, such a phenomenon is far more pronounced in short-lived than long-lived species (Bjørkvoll et al., 2012). Thus, as A. butcheri matures at a young age, has a relatively long life span, and a single large female can produce seven million eggs in a single spawning season, a delay in maturation would not be likely to have a conspicuous deleterious effect on population growth. This has helped account for A. butcheri remaining abundant in the upper Swan River Estuary, even though conditions there have deteriorated over the past 20 years. In summary, the results of this study demonstrate that the deleterious changes that have occurred in the environmental conditions of the permanently- open microtidal Swan River Estuary since the 1990s have been accompanied by significant changes in certain biological characteristics of A. butcheri. The marked decline in the condition and growth of A. butcheri is consistent with the established effects of hypoxia,
63 which are known to result in inter alia reductions in metabolism and changes in the quantity and/or quality of food. The declines in condition and growth between the two periods are, however, also considered to reflect density-dependent effects as fish have increasingly tended to aggregate inshore. The reduction in growth would also account for
A. butcheri attaining maturity at a smaller size and older age. The results of this study thus contribute to our understanding of how, and the extent to which environmental degradation can influence the biological characteristics of a demersal fish species and will therefore be of value to fisheries and environmental managers when considering the changes that could occur with other species exposed to the impacts of a decline in water quality.
64
Chapter 4
Factors influencing the growth of Acanthopagrus butcheri in a eutrophic estuary
have changed over time.
4.1 Introduction
An understanding of the growth of individuals in a fish stock is crucial for determining the status of that stock and thus for assessing its resilience to fishing pressure (Beverton &
Holt, 1957; Hilborn & Walters, 1992; Quinn & Deriso, 1999; Haddon, 2011). The growth of fish in a population is traditionally described by fitting a von Bertalanffy growth model to the lengths at age of individuals in that population, using data collected over several years (Fabens, 1965; Ricker, 1979; Quinn & Deriso, 1999; Jennings et al., 2009). This enables the expected lengths at age of fish to be estimated for the period for which lengths and ages were recorded, irrespective of the extent to which growth varies among years.
The overall von Bertalanffy growth curves of the females and males of the estuarine species Acanthopagrus butcheri in the Swan River Estuary in south-western
Australia in 1993-95 were compared with those in 2007-11 (Chapter 3). This showed that growth was less in the latter period, when environmental conditions had declined due to detrimental effects produced by reductions in freshwater discharge, as a result of declining rainfall. These detrimental effects included a pronounced increase in the extent of hypoxia in deeper water, which, on the basis of other studies, would have inhibited the growth of fish that remained within those waters (Pichavant et al., 2000, 2001; Eby et al., 2005). The increases in hypoxia also led, however, to the larger A. butcheri tending to move from deeper waters into the shallow and better oxygenated waters, where, as a consequence, densities increased markedly, but subsequently remained relatively constant through 2007-
11. It was thus proposed that density-dependent effects made a major contribution to the
65 reduction in the growth of A. butcheri between those two periods. As this reduction occurred when water temperatures were increasing (Western Australia Department of
Water, 2015), and growth would thus have been expected to increase (Brown et al., 2004), it was assumed to have been due to factor(s) other than temperature.
Few studies have explored the extent to which the growth of individuals of an estuarine species, i.e. whose life cycle is completed within the estuary (Potter et al., 2013), vary either between or within periods in a given system. However, a detailed modelling study, based on samples collected in 1993-98 and 2003-08, demonstrated that, in south- eastern Australia, the growth of otoliths of the Estuary Perch Percalates colonorum was related more closely to temperature than to any of the other variables examined
(Morrongiello et al., 2014). Although the growth of P. colonorum was also related to freshwater discharge, this effect was not as pronounced as with certain other estuarine species, such as the Barramundi Lates calcarifer in a tropical estuary, where growth was greatest under high flow (Robins et al., 2006). In contrast, there were indications that the growth of the Spotted Seatrout Cynoscion nebulosus in San Carlos Bay, Florida, was negatively related to freshwater discharge (Bortone et al., 2006). The use of a common curve to describe the growth of individuals in a particular population, based on data collected over a period, assumes that, within each year of life, all year classes experienced the same environmental conditions, density-dependent effects and other factors that influence growth (Szalai et al., 2003). To take into account possible inter-annual variation in such conditions, Szalai et al. (2003) employed a von Bertalanffy growth curve with time-varying parameters to predict the growth increments for each age class in each successive year. They described annual changes in L∞ as a random walk on the log scale and assumed that the relationship between L∞ and k was linear, thereby enabling the parameters of that relationship to be estimated. That model was subsequently modified by
66
He & Bence (2007) to allow each year-specific von Bertalanffy growth parameter to vary over time and then developed a further model to incorporate a cohort effect as well as a year effect. When fitting these models, they employed a hierarchical Bayesian approach, which permitted statistical inferences to be drawn as to the direction and magnitude of any potential correlations among those parameters.
The first aim of this study was to determine the extents to which the growth of female and male A. butcheri in the Swan River Estuary varied among years within 1993-95 and 2007-11, the periods for which overall growth had previously been determined
(Chapter 3), but extending the latter period to incorporate data subsequently acquired for
2012-14. A year-effect growth model was thus developed, based on the ‘Fabens year- effect’ model of He & Bence (2007), but fitted using a simpler maximum likelihood approach. The resultant data, and those for some intermediate years, were employed to elucidate how the growth of this sparid changed during more than two decades. The hypothesis was then tested that, in 2007-14, annual growth in the shallows, where the majority of fish were now located, was positively related to temperature during the warmest and main growth phase (late spring to early autumn). The data were also used to explore whether annual growth in 2007-14 was related to freshwater discharge in the immediately preceding mid-winter to early spring, when the vast majority of freshwater discharge occurs.
4.2 Materials and methods
4.2.1 Temperature and freshwater flow
Mean annual values of the maximum daily air temperatures, recorded between 1990 and
2014 at Perth Airport in the vicinity of the Swan River Estuary, were taken from the
67 website of the Australian Bureau of Meteorology (2015). A linear equation was used to describe the relationship between these temperatures and years. A one-tailed t-test was then employed to determine whether the Pearson’s correlation coefficient relating temperature to year was statistically significant and thus confirm that air temperature increased over that period. The annual freshwater discharges entering the Swan River
Estuary at gauging station 616011, upstream of the study area, in each year between 1990 and 2014 were taken from the website of the Western Australia Department of Water
(2015). A log-linear equation was used to describe the curvilinear relationship between freshwater discharge and year. Back-transformed predicted values for freshwater discharge were corrected to account for the bias associated with the log transformation (Beauchamp
& Olson, 1973). A one-tailed t-test was employed to determine whether Pearson’s correlation coefficient between the log-transformed values of freshwater discharge and year was statistically significant and thereby confirm that, as previously shown for 1992 to
2011 (Chapter 3), annual freshwater discharge declined between 1990 and 2014.
4.2.2. Sampling regime and processing of fish
Acanthopagrus butcheri was sampled at nine sites in the upper Swan River Estuary (Fig.
3.1 in Chapter 3). Nearshore, shallow waters were sampled using 21.5 and 41.5 m long seine nets in each season between spring 2007 and winter 2011. Offshore waters were sampled by gill netting in waters adjacent to each seine net site. Gill netting was undertaken at the same time (season) as seine netting between spring 2007 and winter
2009, after which it was discontinued as it was catching few fish. Acanthopagrus butcheri was also caught at intervals between the spring and following winter in 2011 to 2014 using seine and gill nets in the same region of the upper Swan River Estuary as the above nine sampling sites.
68
Acanthopagrus butcheri had previously been caught in the upper Swan River
Estuary between spring 1993 and winter 1995 by using the 41.5 m long seine net at six of the above nine sites in nearshore, shallow waters, and by employing a gill net (comprising panels of the same dimensions and mesh sizes as described above) at all nine sites in offshore waters (Sarre & Potter, 2000). Full details of the sampling regime and net dimensions are given in Chapter 2.2. Additional samples of A. butcheri were collected irregularly during other studies from the upper Swan River Estuary between spring 2003 and winter 2007.
Throughout the subsequent text, the data for each of the 12 months between the spring of one year and the winter of the next year (which thus commences when
A. butcheri spawns) is subsequently referred to as a ‘year’, e.g. spring 2008 to winter 2009 is referred to as 2008/09. Note that there were continuous data for the spring to winter seasons in the seven ‘years’ from 2007/08 to 2013/14, which formed the focus of many of the comparisons.
Following their capture, A. butcheri were euthanised in an ice slurry and transported to the laboratory where they were sexed and their total length measured to the nearest 1 mm. The otoliths of each fish were removed and the growth zones in whole otoliths counted, when their number was ≤6, and in sectioned otoliths when that number was >6. The number of growth zones in the otoliths of each fish, together with the birth date (peak time of spawning) and knowledge of when the new opaque zone becomes delineated, were used to age that fish. Full details for the rationale for this procedure are given in Chapter 2.3.
69
4.2.3 Year-effect growth model
The growth model developed in this study, subsequently termed the ‘year-effect model’, has essentially the same structure as that of He & Bence (2007), but differed as follows. (1)
Deviations from the growth curve are assumed to be normally-distributed, and thus the predicted lengths at age represent the mean rather than median length at age. (2)
Parameters include the lengths at age of each year-class (except that of the 0+ age class) at the start of the sampling period during which fish were collected, rather than parameters for the lengths of fish at a specific age, i.e. 3 years old. (3) The lengths at age zero for fish that entered the population in the first and subsequent years of sampling were calculated using the von Bertalanffy growth equation and the growth parameters for that year. (4) The model was fitted by maximising the log-likelihood rather than employing a hierarchical
Bayesian approach.
The modified year-effect model was used to calculate, separately for each sex, year-specific growth parameters and expected lengths at different ages in each year. As a single growth model could not be fitted to the entire series, due to gaps in the data, the year-effect model was fitted first to the lengths at age of fish collected during 1993/94 and
1994/95 and then to those for each year between 2007/08 and 2013/14.
For the year-effect growth model, t1 and tn denote the first and last year of sampling, respectively, and A represents the maximum age at the beginning of year t1 for the year classes of fish present in that year. Note that these fish would be A + n – 1 years old at the beginning of the last sampling period, i.e. the beginning of year tn. The model assumes that, within each year t (t1 ≤ t ≤ tn), fish grow in accordance with a von Bertalanffy growth curve with year-specific growth parameters, L∞t and kt, which represent the asymptotic length and a coefficient that determines the rate at which fish of a given length approach that asymptotic length, and t0, the age at which, in theory, the fish would have
70 zero length, and is assumed to be constant for all year classes. The model also assumes that fish do not undergo negative growth, and that the length at age 0 years of fish within each year class y, which was recruited to the population during the sampling period, may be calculated using the von Bertalanffy growth curve and the growth parameters for that year, i.e. for t = y, and the common value of t0. The initial lengths at the beginning of year t1 of the fish of year classes already recruited to the population before the first year of the sampling period, i.e. fish of ages 1 to A years, vary and are represented as the parameters
to .
Let ̂ represent the expected length of a fish from year class y at the start of year t, where and where the age of the fish at that time is years. From the above assumptions, it follows that ̂ ̂ if ̂ , otherwise
̂ ̂ ( ̂ )[ exp( )]. It also follows that, when captured at decimal age years, the length ̂ of the j’th fish, which was of year class , may be estimated as ̂ ̂ if ̂ , otherwise ⌊ ⌋ ⌊ ⌋ ⌊ ⌋
̂ ̂ ( ̂ ) { exp [ ( ⌊ ⌋)]}, where the ⌊ ⌋ ⌊ ⌋ ⌊ ⌋ ⌊ ⌋ year in which the fish is caught is calculated as ⌊ ⌋, and where ⌊ ⌋ represents the
‘floor’ function of x, i.e. the greatest integer less than or equal to x. The above equation represents the expected length following a period of growth of ⌊ ⌋ years from an initial length ̂ at the start of year t, where growth is in accordance with a von ⌊ ⌋
Bertalanffy growth model with the growth parameters for the year of capture, i.e. ⌊ ⌋ and . For year classes to , ̂ { exp[ ]} where is ⌊ ⌋
̂ common. Finally, for year classes ,
When fitting the above model, the log-likelihood (LL) was calculated assuming that the observed length at age of each fish at its date of capture was normally distributed, with 71 common variance, about the estimated length at age at that date. The parameters estimated when fitting the model comprised the year-specific values of and (where
), a common value of , and the initial lengths for .
For each set of lengths at age for each sex in 1993-95 and 2007-14, a traditional von Bertalanffy growth curve, with parameters , k and , was first fitted by maximizing
LL. The year-effect model was then fitted, using common values for and k for each year and a common , to obtain estimates of the parameters for the initial lengths for
. ‘All subsets regression’ (Berk, 1978) was then used to identify the model with year-specific values of and which, as parameters were successively added (while retaining common values for and k for the remaining years), provided the best representation of the lengths at age at each successive level of increasing complexity, i.e. with A + 4, A + 5, …, A + 2n +1 parameters. With the addition of each parameter, the
Akaike Information Criterion (AIC) of the best fitting model at each level of complexity was compared with that of the best fitting model at the previous level of complexity. The model with the smaller AIC value was accepted as the better of the two models. AIC was calculated as AIC = 2np - 2LL, where np is the number of parameters.
The observed lengths at age of each year class within each yearly sample were resampled (with replacement) to produce samples of the same size as in the original data for that year class. The resampled data for the different year classes were then combined and the model fitted (as described above) to the resultant data set. This bootstrapping process was repeated 1000 times. The approximate 95% confidence limits for the parameters and estimated length of fish within each year class at each age were taken as the 2.5 and 97.5 percentiles of the estimates produced by the resampling procedure. The means of the predicted lengths were plotted against the means of the observed lengths at age of the 2+, 3+ and 4+ age classes, which had each passed through two or more summers
72 of growth and were represented by substantial numbers, to demonstrate that the model provided a good representation of the data.
A simulation study was used to confirm that, for each sex in 1993-95 and 2007-14, a model of the form described above was capable of producing reliable parameter estimates when applied to the type of data recorded for A. butcheri in the Swan River
Estuary. Thus, the model was fitted to simulated lengths at age to test whether it could recover the specified ‘true’ values of the model parameters used to generate the simulated data. Simulated lengths at age were generated for each of the observed ages at capture of the fish within each year class in each annual sample, assuming that these lengths were normally distributed about the mean lengths at age calculated using the different values of
L∞ and k for each year and a common standard deviation. The values of these parameters and the standard deviation of ‘observed lengths at age’ about the predicted lengths at age were those that had been determined when first fitting the year-effect model to each sex within each period. The same sample sizes for each year class within each annual sample, as in the original data set, were used when generating each simulated data set. The process was repeated for 1000 simulated data sets and the resultant parameters compared with the parameters employed to generate the data. Overall percentage bias was estimated separately for L∞, k and the parameters representing the initial lengths for each sex and
each period as [∑ ( ( ̂ ) )]⁄ , where is the specified value for parameter i and ̂ is the mean of that parameter estimated for the n = 1000 simulated data sets. Percentage Root Mean Square Error (%RMSE) of that estimate was calculated as
∑ √ ∑ (( )⁄ ) .
73
4.2.4 Relationship between growth and temperature and freshwater discharge
The year specific growth parameters were used to calculate annual length increments for females and males of A. butcheri in each year in 2007/08 to 2013/14. The length (growth) increments were calculated for fish with a TL of 115 mm, the approximate TL attained at the end of the first year of life. Mean annual growth increments were plotted against the mean annual values of the maximum daily air temperatures in the months between
November and March, i.e. warmest part of the year and when growth mainly occurs, and against mean freshwater discharge in the immediately preceding months of June to
September, i.e. when the vast majority of rainfall occurs. A one-tailed t-test of the Pearson correlation coefficient was used to test whether annual growth increments between
2007/08 and 2013/14 were positively related to temperature as hypothesised and a two- tailed t-test employed to test whether those increments were related to freshwater discharge, for which there was no clear hypothesis.
4.3 RESULTS
4.3.1 Temperature and freshwater discharge
Mean annual values of the maximum daily air temperatures increased progressively and in an essentially linear manner between 1990 and 2014, with the least value of 23.5°C recorded in 1990 and the greatest of 26.1°C in 2010 (Fig. 4.1, r = 0.73, P<0.001). The linear equation relating temperature (T) to year (Y) was T = 0.07Y – 109.5 and thus mean annual values of the maximum daily air temperatures increased on average by 0.07°C y-1 over that period. Conversely, the mean annual freshwater discharge decreased in an exponential fashion with year (Fig. 4.1), as described by the log-linear equation relating freshwater discharge (D) and year, i.e. ln(D) = -0.08Y + 159 (r = -0.63, P<0.001). Mean annual freshwater discharge ranged from 5.6 to 21.9 m3s-1 between 1990 and 1999, from
74
2.6 to 9.4 m3s-1 between 2000 and 2009 and never exceeded 4.8 m3s-1 between 2010 and
2014.
Figure 4.1 Mean annual values of the maximum daily air temperatures, recorded at Perth Airport in the vicinity of the Swan River Estuary (Australian Bureau of Meteorology, 2015), and mean annual freshwater discharge, recorded at gauging station 616011 upstream of the study area (Western Australian Department of Water, 2015) in each year between 1990 and 2014. The fitted lines are also displayed.
4.3.2 Model parameter selection and simulation
The inclusion of parameters representing the initial lengths of the females and males of
A. butcheri with ages ≥ 1 year at the start of 2007-14 resulted in better fits of the model to the lengths at age for both sexes than when using traditional von Bertalanffy growth models for that period. Thus, the inclusion of the initial length parameters led to the AIC decreasing from 37384 to 36961 for females and from 35836 to 35489 for males. The AICs for females in 2007-14 continued to decline with the addition of nine of the twelve year- specific growth parameters (L∞s and ks) for those years (AIC = 36203, Tables 4.1, 4.2) and the same trend was true for males with the addition of ten of the twelve year-specific growth parameters (AIC = 34710).
75
Table 4.1 Year-specific von Bertalanffy growth parameter estimates, derived by fitting a year-effect growth model to the lengths at age of females and males of Acanthopagrus butcheri caught in the Swan River Estuary in 2007-14 and 1993-95. Approximate lower and upper 95% confidence limits were derived by bootstrapping. Italics are used to designate parameters that remained common among years.
Females Males von Bertalanffy parameters von Bertalanffy parameters -1 -1 Year k (y ) L∞ (mm) t0 (y) k (y ) L∞ (mm) t0 (y) 2007/08 Average 0.32 318 -0.43 0.34 301 -0.44 Lower 0.29 271 -0.39 0.25 273 -0.52 Upper 0.37 340 -0.47 0.36 332 -0.42 2008/09 Average 0.52 192 -0.43 0.51 193 -0.44 Lower 0.36 187 -0.39 0.38 187 -0.52 Upper 0.55 243 -0.47 0.54 236 -0.42 2009/10 Average 0.14 582 -0.43 0.20 417 -0.44 Lower 0.11 413 -0.39 0.16 338 -0.52 Upper 0.21 717 -0.47 0.26 509 -0.42 2010/11 Average 0.43 282 -0.43 0.34 340 -0.44 Lower 0.29 252 -0.39 0.25 324 -0.52 Upper 0.51 379 -0.47 0.36 423 -0.42 2011/12 Average 0.32 318 -0.43 0.57 236 -0.44 Lower 0.29 271 -0.39 0.45 192 -0.52 Upper 0.37 340 -0.47 0.98 258 -0.42 2012/13 Average 1.03 161 -0.43 0.98 165 -0.44 Lower 0.75 117 -0.39 0.77 122 -0.52 Upper 3.68 187 -0.47 2.47 179 -0.42 2013/14 Average 0.32 280 -0.43 0.34 269 -0.44 Lower 0.29 249 -0.39 0.25 258 -0.52 Upper 0.37 308 -0.47 0.36 329 -0.42 1993/94 Average 0.42 373 -0.14 0.20 451 -0.54 Lower 0.35 352 -0.24 0.14 399 -0.78 Upper 0.49 403 -0.06 0.26 546 -0.36 1994/95 Average 0.42 373 -0.14 0.32 451 -0.54 Lower 0.35 352 -0.24 0.20 399 -0.78 Upper 0.49 403 -0.06 0.42 546 -0.36
76
Table 4.2 Parameters representing the initial lengths ( ) of the females and males of Acanthopagrus butcheri with ages ≥ 1 year at the start of 2007-14 and 1993-95. Gaps denote no fish of that age class.
2007-14 1993-95 Age (years) Females (mm) Males (mm) Females (mm) Males (mm) 1 106 105 110 138 2 135 132 208 212 3 165 163 268 257 4 205 190 329 310 5 246 214 316 6 274 235 364 7 311 288 387 8 375 9 377 10 397
As with 2007-14, the inclusion of parameters representing the initial lengths of the females and males of A. butcheri with ages ≥ 1 year at the start of 1993-95 improved the model fits for both sexes beyond those produced by traditional von Bertalanffy growth models (AIC = 5491 vs 5543 for females and 6059 vs 6074 for males). In contrast to 2007-
14, however, the inclusion of separate year-specific L∞s and ks for females in the model did not further improve the fit in 1993-95. The inclusion of separate year-specific ks, but not
L∞s (Table 4.2), did improve slightly, however, the fit of the model for males in 1993-95
(AIC = 6054 vs 6059).
As demonstrated by the overall absolute percentage biases and %RMSEs, the year- effect model estimated accurately the year-specific parameters and the initial length parameters for both sexes and periods (Table 4.3).
77
Table 4.3 Overall percentage bias and percentage root mean squared error (%RMSE) in model estimates of the year-effect growth parameters L∞, k and initial length parameters
( ) at the start of each sampling period for female and males of Acanthopagrus butcheri ≥ 1 year in the Swan River Estuary in 1993-95 and 2007-14.
Females Males Parameter %Bias %RMSE %Bias %RMSE
L∞2007-14 (mm) -1.58 3.41 1.21 3.03 -1 k2007-14 (y ) 2.91 5.64 -3.39 7.14
L∞1993-95 (mm) -0.30 0.42 -1.15 1.62 -1 k1993-95 (y ) -0.13 0.19 0.01 0.14
Length parameters2007-14 (mm) 0.14 0.23 -0.03 0.04
Length parameters1993-95 (mm) 0.12 0.20 0.01 0.04
4.3.3 Model results
The lengths at age predicted by the year-effect model for females and males of A. butcheri in the two years in 1993-95 and seven years in 2007-14 did not exhibit any conspicuous systematic deviations from the observed lengths at age (Fig. 4.2). There were, however, discontinuities between the curves describing the lengths at age of successive age classes in each year (i.e. 12 month interval). These discontinuities reflect the fact that the growth of fish of a particular age varies among years and thus fish of the same age attain different lengths in the different years. For example, for females at 4.0 years of age, the predicted mean length of 217 mm in 2010/11 was greater than the 192 mm in 2009/10 (Fig. 4.2). The inter-annual variations in lengths at age are reflected in differences in both of the year- specific von Bertalanffy growth parameters (k and L∞) for each sex in 2007-14 (Table 4.1).
Thus, for example with females, the mean values for k ranged from 0.14 to 1.03 y-1 and those for L∞ ranged from 161 to 582 mm.
78
Figure 4.2 Lengths at age of females and males of Acanthopagrus butcheri caught in the Swan River Estuary in the two years in 1993-95 and four years in 2007-11 and the lengths at age predicted by the fitted year-effects growth model.
79
The means of the lengths at age of capture of females and males of A. butcheri of the 2+, 3+ and 4+ age classes in each year, predicted by the year-effect growth models in the 1993-95 period, were greater than in each of the seven years in 2007-14 (Fig. 4.3).
Thus, for example, the means of the predicted lengths for 2 year old females in 1993/94, i.e. 241 mm, and 1994/95, i.e. 214 mm, were greater than in each of the seven years in
2007-14, which ranged from 145 to 202 mm. Moreover, the differences between the means of the predicted lengths for 1993-95 and 2007-14 increased with age. For example, in the case of females, the means of the predicted lengths for 1993/94 and 2007/08 differed by 70 mm for 2 year old fish and by 127 mm for 4 year old fish (Fig. 4.3).
The means of the predicted lengths at age of capture for females and males of
A. butcheri of the 2+, 3+ and 4+ age classes in each year in the 1993-95 and 2007-14 periods were very similar to the corresponding means of the observed lengths at capture
(Fig. 4.3). The means of the observed lengths of each age class of fish in the intervening years between 1993-95 and 2007-14 (for which the data were obtained using a slightly different sampling regime and thus not included in the year-effect model) declined in a manner consistent with the trends exhibited by the means of the predicted and observed lengths of those age classes in the preceding and subsequent years (Fig. 4.3). The mean observed lengths for each age class followed a smooth trend over the 21 years of sampling.
Thus, the means of the observed lengths of both sexes declined progressively from 1993-
95 to ~2007/08 and then rose until ~2011/12, before declining again in 2013/14. For example, with 4 year old females, the means of the observed lengths decreased from >300 mm in the 1990s to 192 mm in 2007/08 and then rose to 233 mm in 2012/13 before falling to 215 mm in 2013/14 (Fig. 4.3).
80
Figure 4.3 Mean lengths at age of female and male Acanthopagrus butcheri (and 95% confidence intervals) for sequential age classes caught in each year in the 1993-95 and 2007-14 periods, as predicted by the year-effect growth models (closed circles) and for the observed lengths at age at capture (open circles). Mean observed lengths at age (and 95% confidence intervals) for the smaller samples of fish collected in 2003/04, 2004/05, 2005/06 and 2006/07 are also shown.
81
On the basis of the growth parameters of the year-specific growth model and an initial length of 115 mm (the average predicted length at the completion of the first year of life), female A. butcheri increased in length by 88 mm in both 1993/94 and 1994/95 (Fig.
4.4). These annual increments were greater than those estimated for females in each of the seven years in 2007-14, which ranged from 29 to 63 mm. The trends exhibited by the annual length increments by males followed a very similar pattern, with the annual increment of 63 mm in 1993/94 and 82 mm in 1994/95 being greater than in each year between 2007-14, apart from 2010/11 in which the maximum value of 64 mm was virtually the same as in 1994/94 (Fig. 4.4).
Figure 4.4 Mean values (and 95% confidence intervals) for predicted annual increases in the lengths of one year old females and males of Acanthopagrus butcheri during each year in the 1993-95 and 2007-14 periods, using a common starting length of 115 mm, the average length of fish when they commence their second year of life.
82
4.3.4 Relationship between growth and temperature and freshwater discharge
The predicted annual growth increments for A. butcheri in the years between 2007/08 and
2013/14, based on a length of 115 mm at the commencement of each year, were positively correlated with mean annual values of the maximum daily air temperatures in each month between November to March with both females (Fig. 4.5, r =0.68, P<0.05) and males (Fig.
4.5, r = 0.77, P<0.05). The same growth increments were not significantly correlated with freshwater discharge in the preceding months of June to September, i.e. when the vast majority of rainfall occurs, in the case of either the females (Fig. 4.5, r = 0.26, P>0.05) or the males (Fig. 4.5, r = - 0.01, P>0.05).
Figure 4.5 Relationship between annual growth increments for females and males of A. butcheri between 1 October and 30 September in each year of 2007/08 to 2013/14 and (a, b) mean annual values of the maximum daily air temperatures in each month between November and March and (c, d) mean of the daily freshwater discharge between June and September. Linear equations relating growth increments to temperature are shown.
83
4.4 DISCUSSION
The year-effect model, developed for this study, provided a better description of the growth of female and male Acanthopagrus butcheri in the Swan River Estuary in the two years in 1993-95 and seven years in 2007-14 than that produced by the traditional von
Bertalanffy growth model. It thus enabled an exploration of whether the growth in different years was significantly related to environmental variables that had been selected because they were known to influence growth in other species. Such an approach allowed factors that influence growth to be identified, without specifying the precise functional relationships required to include these factors as covariates in a growth model (Szalai et al., 2003; He & Bence, 2007).
The year-effect model demonstrated that the growth of both sexes of A. butcheri in this estuary varied markedly among years in 2007-14, the period for which there were data for a number of sequential years. These inter-annual differences are emphasised by the discontinuities in the growth trajectories derived for sequential age classes in each year from the year effect model. The differences in growth reflect the plasticity of certain biological characteristics of A. butcheri and its ability to respond to changes in environmental conditions. This type of plasticity in this species and other sparids is also reflected in the marked differences between the biological characteristics of such species in different water bodies in the southern hemisphere (e.g. Sarre & Potter, 2000; Griffiths et al., 2002; Partridge et al., 2004; Chuwen et al., 2007; Sim-Smith et al., 2012; Wakefield et al., 2015).
As A. butcheri was not sampled in eight of the years between 1994/95 and 2007/08 and sampling for the other four years in that period was opportunistic, the year-effect model was not applied to the lengths at age for fish in this period. It is highly relevant, however, that the mean lengths predicted for an age class in 1993/94 and 1994/95 and in
84 each year between 2007/08 and 2013/14 were very similar to the means of the observed lengths for those age classes in each of those years. Thus, the means of the observed lengths for the years in the intervening period in which sampling was conducted are considered broadly representative of those that would have been derived from the year- effect model, if sufficient data had been available to construct a reliable model.
Consequently, the decline in the means of the observed lengths at age for the 2+, 3+ and 4+ age classes, from their maxima in 1993/94 to their minima in 2007-10, followed by a slight rise and then decline to 2013/14, reflect a genuine smooth pattern of change in the growth of A. butcheri over more than two decades.
The smoothness of the changes may be due to growth being continuous over years rather than changing abruptly at intervals, with the means of the observed lengths of each age class of A. butcheri representing the cumulative effect of the present and past growth of the individuals constituting each age class. In contrast, the mean annual growth increments, predicted by the year-effect model for a fish with a TL of 115 mm at the beginning of each year, represent the amount of growth of a fish during the ensuing year.
The greater annual growth increments recorded for females and almost invariably males in
1993/94 and 1994/95 than in any year between 2007/08 and 2013/14 are consistent with the trends exhibited by the means of the observed lengths over more than two decades.
The above trends emphasise that the growth of A. butcheri in the Swan River
Estuary declined between the early 1990s and more recent years and are thus entirely consistent with the differences recorded previously for overall growth, based on the lengths at age data for 1993-95 and 2007-11 (Chapter 3). As temperature increased markedly between 1993-95 and 2007-14, the decline in growth runs counter to the relationship recorded with temperature for many species and is inconsistent with the metabolic theory
85 of ecology (Brown et al., 2004; Angilletta et al., 2004; Weber et al., 2015). Temperature is thus not considered responsible for the decline in growth between periods.
The years between 1993-95 and 2007-11 were accompanied by a decline, not only in growth, but also in body condition and length at maturity and by an increase in age at maturity (Chapter 3). These changes were consistent with those which, on the basis of studies of other species, would be expected if the environment had undergone deleterious changes (Stearns, 1992; Eby et al., 2005; Morita & Fukuwaka, 2007). It was thus highly relevant that the deeper waters of the upper Swan River Estuary became increasingly hypoxic between 1993-95 and 2007-11 (Chapter 3), which would presumably have led to a decline in the metabolic rate of A. butcheri and thus account, in part, for the decline in growth and changes in maturity schedules. As the abundance of large A. butcheri in deeper waters, the typical habitat of these individuals, declined markedly between these two periods, whereas that in nearshore, shallow waters increased sharply, hypoxia is also assumed to have led to large fish moving from deep waters to the better oxygenated nearshore shallow waters. It was thus proposed that the marked increase in density in nearshore waters resulted in a density-dependent inhibitory effect on growth and changes in the other biological characteristics listed above.
In contrast to the marked increase in the densities of A. butcheri in the shallow waters of the estuary between 1993-95 and 2007-11, the mean densities in those nearshore waters remained similar throughout the four years in the latter period and that trend continued through to the end of the present study in 2014 (A. Cottingham unpublished data). The marked inter-annual variations in growth in 2007-14 were thus presumably attributable to factor(s) other than density. It is therefore relevant that, in 2007-14, the annual growth increments of both the females and males of A. butcheri, during their second year of life, were significantly correlated with temperature, which is consistent with the
86 trends exhibited by many fish species and predictions based on the metabolic theory of ecology regarding growth rates of younger fish (Brown et al., 2004; Angilletta et al., 2010;
Neuheimer et al., 2011). It also parallels the results of a study by Morrongiello et al.
(2014), which demonstrated that the growth of the Estuary Perch Percalates colonorum in south-eastern Australia was related more closely to temperature than to any of the other variables examined.
Freshwater inflow into an estuary has often been regarded as important in influencing the biological characteristics of a fish species in estuaries (Gillanders &
Kingsford, 2002; Robins et al., 2006; Bortone et al., 2006; Glover et al., 2013). For example, it supplies nutrients which promote primary and thus secondary productivity, thereby enhancing the growth of fish in these systems (Darnaude, 2005). It is therefore relevant that the growth, and also condition, of A. butcheri in the Swan River Estuary, were greater in 1993-95 than in 2007-11 and thus when freshwater discharge was significantly greater (Chapter 3). In this case, however, the beneficial effects arise from greater flushing in 1993-95 and thus the removal of excessive amounts of nutrients and a reduction in the levels and extent of hypoxia in this eutrophic estuary. The present study demonstrated that, in the shallows of this estuary in 2007-14, when flow remained relatively low and the densities of A. butcheri were relatively constant, the annual growth increments of one year old fish were not significantly correlated with freshwater discharge in June to September, when the vast majority of discharge occurs. Although Morrongiello et al. (2014) found that the growth of P. colonorum was related to freshwater discharge, this relationship was far less influential than temperature.
In summary, the year-effect model demonstrated that the growth of both female and male A. butcheri varied among successive years within a seven year period (2007-14). The results also demonstrated that, over those seven years, the growth of both sexes of
87
A. butcheri was related to temperature, but not to freshwater discharge. The overall growth and other biological characteristics of A. butcheri have previously been shown to decline between 1993-95 and 2007-11, during which there was a marked deterioration in environmental conditions and particularly in deeper waters, which became increasingly hypoxic. It was proposed that the resultant increased aggregation of A. butcheri in the better oxygenated nearshore shallow waters led to a density dependent effect on the growth and associated characteristics of this species. This decline in growth was unlikely to be related to temperature as this variable increased between those two periods. Thus, the factors proposed as influencing growth between 1993-95 and 2007-11, i.e. hypoxia and density-dependent effects, differed from the major factor, i.e. temperature, that influenced the growth of A. butcheri over the seven years between 2007 and 2014.
As A. butcheri is confined to its natal estuary for the whole of its life cycle, the individuals of this species are exposed, throughout life, to any ongoing deleterious changes in the environmental quality within the estuary. This feature, allied with its plastic biological characteristics, makes A. butcheri an ideal candidate for exploring the ways and extent to which certain key biological characteristics of a fish species respond to detrimental and other changes in its environment. This species thereby serves as a valuable indicator of the health of an estuary and for hypothesising on the effects of climate change.
88
Chapter 5
Performance and contribution to commercial catches and egg production
by restocked Acanthopagrus butcheri in the Blackwood River Estuary.
5.1 Introduction
Restocking is increasingly being used to restore fish stocks that have become depleted through over-exploitation and to meet the global demand for seafood (e.g. Bell et al., 2006;
Gomez & Mingoa-Licuanan, 2006; Strøttup et al., 2008; Loneragan et al., 2013).
Furthermore, certain fish stocks have been enhanced through the introduction of cultured fish. For example, the several million Japanese Flounder Paralichthys olivaceus and Red
Sea Bream Pagrus major, which had been hatchery-reared and released into Kagoshima
Bay, Japan, contributed 26 and 36% to the commercial catches of those two species in that bay (Kitada & Kishino, 2006). In Australia, the ~350,000 Sand Whiting Sillago ciliata and
86,000 Dusky Flathead Platycephalus fuscus, which were hatchery-reared and released as juveniles into the Maroochy Estuary, Queensland, subsequently contributed as much as 52 and 28%, respectively, to the commercial catches of those two species in that system
(Butcher et al., 2000).
The use of restocking overcomes the need to impose politically and socially less acceptable options for the recovery of a depleted stock, such as reducing size limits and total allowable catch and closing areas to fishing (Travis et al., 1998; Bell et al., 2006).
Many culture and release programmes have failed, however, to demonstrate that they increased the yield or abundance of the stock (Blankenship & Leber, 1995; Hilborn, 1998;
Bell et al., 2008). The introduction of the cultured fish of a species can have, however, a negative effect on the wild population of that species through, for example, increasing
89 inbreeding, introducing diseases and leading to a decrease in the abundance of wild fish
(Einum & Fleming, 2001; Lorenzen et al., 2012; Camp et al., 2014).
Restocking is a particularly useful method of replenishing a depleted stock of those species that complete their life cycle within estuaries or inland waters and whose stocks cannot thus be augmented naturally by recruitment from outside that system (Lenanton et al., 1999; Taylor et al., 2005; Ward, 2006). Furthermore, if restocking does have a deleterious impact on the genetic composition of a species in such water bodies, it is restricted to the population within the system into which the individuals were introduced
(Potter et al., 2008).
The Black Bream Acanthopagrus butcheri is a very important recreational fish species in the estuaries of southern Australia and is fished commercially in some of these systems (e.g. Lenanton & Potter, 1987; Jenkins et al., 2010; Chapter 3). This sparid, which can live for ~30 years (Morison et al., 1998; Potter et al., 2008), completes its life cycle within its natal estuary and typically spends most of the year within its upper reaches, where it spawns during spring and early summer (Potter & Hyndes, 1999; Sarre & Potter,
1999; Williams et al., 2013; Chapter 3).
Several stocks of A. butcheri in south-western Australia were reported by Lenanton et al. (1999) to have become depleted. It is not clear, however, whether increased fishing pressure and/or environmental changes were responsible for the decline in the abundance of this species in the Blackwood River Estuary, which supports a recreational fishery and a small commercial gill net fishery (Prior & Beckley, 2007; Potter et al., 2008). As the recruitment of A. butcheri into a population can be highly episodic, the recreational and commercial fisheries in some systems rely almost entirely on a few year classes (Morison et al., 1998). Concern for the status of the stock of A. butcheri in the Blackwood River
Estuary led to a study aimed at determining the efficacy of using restocking to replenish
90 the population of this species in this system (Potter et al., 2008). Brood stock from this estuary were thus used in 2001 and 2002 to culture juveniles, whose otoliths were stained with alizarin complexone to allow these fish to be distinguished from the wild stock in samples collected in years following their release (Potter et al., 2008). This marking method has been successfully employed with other species, such as the Chum Salmon
Oncorhynchus keta and the Pink Salmon Oncorhynchus gorbusha (Sato et al., 2011) and the Black Rockfish Sebastes schlegeli (Nakagawa et al., 2007).
The results of sampling A. butcheri in the Blackwood River Estuary in the period up to November 2005 indicated that, in comparison with wild fish, the cultured fish had not grown quite as rapidly. Furthermore, in contrast to the wild stock, approximately 33% of restocked fish at four years old had not reached maturity (Potter et al., 2008). It was suggested that these small disparities may have been related to the effects of hatchery- rearing. The data did show, however, that, by 2005, a few cultured fish had reached the minimum legal length (MLL) of 250 mm for the retention of this sparid by commercial and recreational fishers.
Examination of individuals in fishery-independent samples and in commercial gill net catches of A. butcheri in the Blackwood River Estuary, during the nine years that followed the completion of the earlier study in 2005, demonstrated that the alizarin complexone stain remained prominent in the otoliths of restocked fish throughout that period and that cultured fish were contributing to the commercial fishery. Biological data from the earlier study and recent periods were thus collated to compare statistically the biological performance of cultured and wild A. butcheri, which had coexisted for more than a decade, and to estimate the relative contribution of cultured and wild stock
A. butcheri to commercial catches. The combined data were used to determine whether, when comparisons were made using data collected over more than a decade, (1) the
91 restocked fish grew less rapidly than wild stock fish, and the extent and significance of any differences; (2) all surviving restocked fish eventually became mature and, if so, determine whether the length and age at maturity of restocked fish differed from those of the wild stock; (3) restocked fish made a substantial contribution to successive annual commercial catches of A. butcheri and the extent of any such contributions, and (4) restocked fish eventually made a major contribution to egg production and if there was strong circumstantial evidence that the progeny of the restocked fish could have contributed to commercial catches.
5.2 Materials and Methods
5.2.1 Collection of restocked and wild Acanthopagrus butcheri
Fifty-six female and 50 male A. butcheri were collected from the Blackwood River Estuary and taken to the Australian Centre for Applied Aquaculture Research, Challenger Institute of Technology, where they were maintained under light/dark and temperature conditions that paralleled those in the estuary. When these fish became mature in the springs of 2001 and 2002, i.e. at the same time as A. butcheri spawns in the estuary, they were used as brood stock for culturing A. butcheri (Potter et al., 2008). Fertilised eggs were produced by simultaneously stripping all females and males and subsequently by the natural spawning of these individuals on the following day.
The larvae and juveniles were reared as reported in Partridge et al. (2004) and the otoliths of all juveniles were stained with alizarin complexone as described in Potter et al.
(2008). The 70,000 juveniles cultured in 2001 were released into the Blackwood River
Estuary at ~ seven months old in winter 2002, while the 150,000 reared in 2002 were released at ~ four months old in autumn 2003 (Potter et al., 2008). During subsequent
92 sampling, these cultured individuals were clearly identifiable through the purple colour produced in the central region of their otoliths by the alizarin complexone stain.
In the earlier study, the results of which were reported in Potter et al. (2008), restocked and wild stock A. butcheri were obtained from the upper Blackwood River
Estuary (Fig. 5.1) by fishery-independent gill netting at regular intervals between 2000 and
2005 (Table 5.1). The composite sunken gill net used by Potter et al. (2008), which comprised six 20 m long and 2 m high panels, each containing a different mesh size, i.e.
38, 51, 63, 76, 89 and 114 mm, was set at dusk and retrieved at first light on the following morning. Fishery-independent samples were also obtained in 2006, using gill nets with the same dimensions as described above, and in 2013 and 2014, but with two additional panels being added with mesh sizes of 102 and 127 mm (see Chapter 2.2 for details of gill net).
Samples were also collected using a 41.5 m seine net in 2000-05, 2012 and 2014 and by rod and line fishing in 2000-05 and 2012-14 (Table 5.1). The dimension of the 41.5 m seine net is given in detail in Chapter 2.2.
Random subsamples of the catches of A. butcheri taken by the commercial gill net fisher in the Blackwood River Estuary were obtained annually between 2005 and 2014, except for 2008 and 2011. The commercial (sunken) gill net was 20 m long and 2 m high and made of 105 mm stretched mesh, which was chosen by the fisher so that it essentially caught only fish with total lengths ≥ 250 mm, the MLL. While fishery-independent sampling was undertaken at different times of the year (Table 5.1), the commercial fisher set his gill net in winter and spring to catch A. butcheri as they are flushed downstream by the heavy freshwater discharge that occurs at that time of the year (Potter et al., 2008). The commercial subsamples were provided as frames (i.e. with fillets removed), together with their intact gonads.
93
Figure 5.1 Map showing the six sites at which Acanthopagrus butcheri was sampled in the Blackwood River Estuary by fishery-independent methods (open circles) and the stretch of the estuary in which the commercial gill net fisher operated. Arrow in insert map location of the Blackwood River Estuary in Western Australia.
The sampling and collection of data on restocked and wild A. butcheri in the
Blackwood River Estuary between 2009 and 2014 were undertaken as part of the work for this thesis, as were all analyses of data collected during the full duration of this project. As mentioned previously, the results of sampling prior to 2005 were reported in Potter et al.
(2008), while the previously unanalysed data for 2005 to 2009 were obtained as an extension of that earlier study.
94
Table 5.1 Numbers of Acanthopagrus butcheri caught by fishery-independent sampling and in subsamples obtained by commercial gill netting in the Blackwood River Estuary between 2000 and 2014. Unless otherwise stated, the fishery-independent gill net comprised eight 20 m long panels, each with a different mesh size, i.e. 38, 51, 63, 76, 89, 102, 114 and 127 mm. Data for 2000-05 were reported in Potter et al. (2008). n is number of fish.
Year Season Method n 2000-05 All Fishery independent gill net (38-89 & 114 mm), 1451 seine net and rod & line 2005 Winter Commercial gill net 271 2006 Winter Commercial gill net 103 Winter Independent gill net (38-89 &114 mm) 102 2007 Winter/spring Commercial gill net 344 2009 Winter/spring Commercial gill net 383 2010 Winter Commercial gill net 290 2012 Autumn-spring Commercial gill net 220 Summer Independent seine net & rod & line 56 2013 Winter/spring Commercial gill net 320 Spring/summer Independent gill net & rod & line 147 2014 Winter Commercial gill net 216 Summer/winter Independent gill & seine net & rod & line 192
5.2.2 Ageing and growth
Each A. butcheri caught by fishery-independent sampling and by the commercial fisher was measured to the nearest 1 mm (total length) and sexed through macroscopic examination of its gonads. The methods used for ageing fish and describing growth are given in detail in Chapter 2.3. The ageing of each fish took into account the date of capture and a birth date of 1 November, the mid-point in the spawning period (see Results). In brief, the otoliths of each fish were removed and those of restocked fish identified through
95 the presence of the purple alizarin complexone stain. von Bertalanffy growth curves were fitted to the lengths at age of the restocked and wild stock fish, which for both groups comprised the 2001 and 2002 year classes collectively. Note, however, that while all fish caught by fishery-independent methods were used, the commercially-caught fish were restricted to those >5 years old, i.e. the minimum age by which the total length of all fish would be expected to have reached 250 mm (see Results), the MLL for retention of A. butcheri.
Although preliminary analyses, using likelihood-ratio tests, demonstrated that the von Bertalanffy growth curves of the females and males of both the restocked and wild stock of A. butcheri differed significantly (P<0.001 and 0.01, respectively), the differences between the lengths at age of the two sexes at all ages in both the restocked and wild stock fish were always less than 5%. As such differences are unlikely to be of major biological significance, the lengths at age for both sexes of the restocked fish and of the wild stock fish have been pooled for comparing the growth of these two groups of fish, an approach adopted in a number of studies on other species (e.g. White et al., 2002; Hesp et al., 2004).
Likelihood-ratio tests were then used to determine whether the von Bertalanffy growth curves of the restocked and wild stock fish differed and, if so, the extent of any differences in lengths at age across the ages sampled. If the latter differences exceeded 5%, likelihood- ratio tests were then used to ascertain whether the growth coefficients, k, and/or the asymptotic lengths, L∞, of the restocked and wild stock fish were significantly different.
5.2.3 Reproduction and fecundity
The gonads of each restocked and wild A. butcheri were examined macroscopically and allocated to one of the following maturity stages described by Sarre & Potter (1999), i.e.
I/II = virgin/resting adult, III = developing, IV = maturing, V = prespawning, VI =
96 spawning, VII = spent and VIII = recovering spent. Previous histological studies of ovaries had demonstrated that assignment of gonads to these stages on the basis of macroscopic appearance was appropriate (Sarre & Potter, 1999; Chuwen, 2009).
For determining the length and age at maturity, A. butcheri with gonads at stages
I/II during the spawning period were considered immature, whereas those with gonads at stages III to VIII were considered mature, i.e. destined to spawn or had spawned during that period. This procedure was adopted after careful recent examination of the pattern of sequential change in gonadal stages in the months immediately prior to and during the spawning period in the Swan River Estuary (see Chapter 3). The designation of fish with gonad stages III and IV as mature leads to lower estimates of the L50 and A50 than when they are classified as immature, as in earlier studies of A. butcheri in the Blackwood River
Estuary (Potter et al., 2008) and other estuaries (Sarre & Potter, 1999). Note also that the reproductive schedules were restricted to data derived from samples collected by fishery independent methods and did not include commercial samples as the exact month of collection was not always recorded.
The trends exhibited in sequential months by the prevalence of the different stages in ovarian development and mean gonad weights were used to determine the spawning period of A. butcheri in the Blackwood River Estuary. The gonad weights and lengths of restocked and wild stock females ≥ the corresponding L50s at maturity (see Results) were standardised for length (231 mm) using ANCOVA. Prior to analyses, the total length and gonad weight for each individual was ln(x + 1) transformed, where ‘ln’ is the natural logarithm. To account for bias introduced with the logarithmic transformation, when ln( ̂) is back-transformed, the correction factor exp( ̂ ) was employed (Beauchamp & Olson,
1973).
97
Logistic regression analysis was used to determine, for both the restocked and wild stock, the probability P that a female or male of a given length and age possessed gonads at stages III-VIII during the spawning season in the Blackwood River Estuary, i.e. October,
November and December. Likelihood-ratio tests were employed to determine whether the ogive relating maturity to length for both the females and males of restocked fish differed from that for the corresponding sex of wild stock fish and, if so, whether this was accompanied by differences in the values of the L50 and/or L95, the lengths at which 50 and
95%, respectively, of fish were expected to be mature. The same approach was used to determine whether the ogives relating maturity to age of the restocked females and males differed from that of the corresponding sex of wild stock fish and, if so, whether this was accompanied by differences in the values of the A50 and/or A95, the ages at which 50 and
95%, respectively, of fish were expected to be mature. The reader is referred to Chapter 2.3 for details of the maximum likelihood method used to determine the relationship between the length and age at maturity and the Bayesian approach employed to determine the 95% credible limits.
Comparisons of the relationships between the lengths and the ages at maturity of the females and males of restocked fish (2001 and 2002 year classes collectively) with those of the wild stock employed data collected during the spawning period, with the data for wild stock fish of the 1999 and 2000 year classes being added to those of the wild stock of the 2001 and 2002 year classes. This increased the number of fish in length and age classes of the 2001 and 2002 year classes of the wild stock that were poorly represented in, or absent from, samples collected during the spawning season.
The relative contributions of each year class of the wild stock and restocked
A. butcheri to egg production in the Blackwood River Estuary in 2007 and 2009 were estimated using the lengths of individual fish in each year class in the commercial catches
98 and the equation of Sarre & Potter (1999), which described the relationship between batch fecundity and the lengths of females in the Swan River Estuary. These two years were chosen for analysis because they straddled 2008, for which no commercial samples were available (see Results), and because the 2008 year class was very well represented in commercial samples from 2012-14, i.e. when they exceeded the MLL for A. butcheri of 250 mm. The total number of eggs produced (F) in 2007 and 2009 by each year class of the
- wild stock and restocked fish was thus calculated as ∑ [ ], where is the total length of the j’th fish, with the summation covering all fish within each year class, yc, separately for both the restocked fish and wild stock in each year, y. This enabled the relative percentage contribution to total egg production by each year class of wild stock and restocked A. butcheri in 2007 and 2009 to be estimated . As the
GSIs of females during the spawning period in the Swan River Estuary were similar to those in the other two estuaries that were sampled during the main part of that period (Sarre
& Potter, 1999), the relationship between batch fecundity and length is considered unlikely to differ markedly among systems.
5.3 RESULTS
During the 12 years of sampling since cultured fish were first introduced into the
Blackwood River Estuary, i.e. 2002 to 2014, a total of 1584 restocked fish were obtained from fishery-independent sampling and subsamples of the commercial gill net catches.
These restocked fish comprised 193 individuals of the 2001 cohort and 1391 of the 2002 cohort.
99
5.3.1 Growth
The von Bertalanffy growth curve provided a good fit to the lengths at ages of the wild stock and of the restocked fish of the same two year classes collectively, i.e. 2001 and
2002 (Fig. 5.2a-c, Table 5.2). The growth curves for the restocked fish lay below and differed significantly from that of wild stock fish (P<0.001). The differences between the lengths at age across all ages from 1 to 12 years were between 7 and 11%. While the growth coefficient (k) of 0.29 y-1 for restocked fish was less than and did not differ
-1 significantly (P>0.05) from the 0.31 y of wild stock fish, the L∞ of 323 mm for restocked fish was less than and differed significantly (P<0.01) from the 350 mm of wild stock fish
(Fig. 5.2c). The growth curve for the limited age range of the 2008 year class lay between the curves for the restocked and wild stock fish.
Table 5.2 von Bertalanffy growth parameters and their 95% confidence intervals for wild stock and restocked Acanthopagrus butcheri in the Blackwood River Estuary. L∞, -1 asymptotic length (mm); k, growth coefficient (year ); t0, hypothetical age at which fish would have zero length (years). r2, coefficient of determination; n, number of fish in sample.
-1 2 L∞ (mm) k (year ) t0 (years) r n Wild stock fish Estimate 350 0.31 -0.70 0.88 163 Lower 336 0.26 -1.01 Upper 364 0.36 -0.40 Restocked fish Estimate 323 0.29 -0.79 0.90 1321 Lower 318 0.26 -0.93 Upper 328 0.30 -0.64
Based on the von Bertalanffy growth equation, the TLs of female restocked A. butcheri at two and six years of age were 179 and 278 mm, respectively, and thus less than the corresponding TLs of 198 and 306 mm for the wild stock.
100
Figure 5.2. von Bertalanffy growth curves fitted to the lengths at age of (a) wild stock and (b) restocked Acanthopagrus butcheri of the same year classes, i.e. 2001 and 2002, collectively, and (c) comparison of the growth curves shown in a and b for wild stock fish (long dashed line), restocked fish (continuous line) and the 2008 year class (short dashed line).
101
5.3.2 Maturation
The mean monthly gonad weights of the females of both the restocked and wild stock fish rose to a sharp peak in October and then declined progressively in the following months
(Fig. 5.3). The mean gonad weights of restocked and wild stock fish did not differ significantly (P>0.05) in any month.
Figure 5.3 Mean gonad weights and 95% confidence intervals in sequential months for wild stock (closed circles) and restocked (open circles) females of Acanthopagrus butcheri with lengths ≥ L50 at maturity and which were caught by fishery-independent sampling in the Blackwood River Estuary between 2000 and 2014. Gonad weights have been standardised for fish of a common length (230 mm) using ANCOVA. Black bars, summer and winter; white bars, autumn and spring. Sample size for each month is shown.
102
All wild stock females <179 mm and all restocked females <155 mm were immature (Fig. 5.4a,b). The wild stock females in the 180-199 mm and subsequent length classes were mature (Fig. 5.4a). The prevalence of mature restocked females increased from 25% in the 140-159 mm length class to 76% in the 180-199 mm length class, and all restocked females ≥240 mm were mature (Fig. 5.4b).
The ogive describing the relationship between the proportion of mature restocked females and total length differed significantly from the corresponding ogive for wild stock females (P<0.01, Fig. 5.4a-c, Table 5.3). The L50s at maturity of 174 mm for restocked females was greater than the 165 mm of wild stock, but the difference was not significant
(P>0.05). The L95 of 224 mm for restocked females was greater than the 168 mm for wild stock females and the difference was significant (P<0.001, Fig. 5.4c).
Table 5.3 Estimates of total lengths (mm) and ages (years) and associated 95% credible limits at which 50% (L50 and A50, respectively) and 95% (L95 and A95, respectively) of the females and males of the wild stock and restocked Acanthopagrus butcheri were mature. ML = Maximum Likelihood. n is the number of fish.
L50 L95 A50 A95 (years) n (mm) (mm) (years) Wild stock fish Females ML Estimate 165 168 1.7 2.0 87 Lower 150 153 1.4 1.8 Upper 179 196 1.9 2.2 Males ML Estimate 149 159 1.2 1.6 133 Lower 141 138 1.1 1.4 Upper 153 203 1.3 2.0 Restocked fish Females ML Estimate 174 224 2.1 3.2 139 Lower 166 210 1.8 2.8 Upper 184 245 2.2 3.9 Males ML Estimate 154 177 1.8 2.3 176 Lower 146 168 1.6 2.2 Upper 160 194 1.9 2.8
103
Figure 5.4. Percentage frequency of occurrence, in sequential 20 mm length classes, of females and males of wild stock and restocked Acanthopagrus butcheri with immature (white bars) and mature gonads (grey bars), derived from samples collected during the spawning season of A. butcheri. Logistic curves (solid lines) and their 95% credible intervals (dotted lines) for the probability that a fish at a given length is mature are superimposed on a, b, d and e. Sample size for each length class is provided. Comparisons
104 of logistic curves for wild stock (dashed line) and restocked (continuous line) females and males are shown in c and f. All wild stock males <147 mm and all restocked males <146 mm were immature
(Fig. 5.4d,e). While all wild stock males in the 160-179 mm and subsequent length classes were mature, about 25% of the restocked males in the 160-179 mm length class were immature (Fig. 5.4d,e).
The ogive describing the relationship between the proportion of mature restocked males and total length did not differ significantly from the corresponding ogive for wild stock males (P>0.05, Fig. 5.4d-f, Table 5.3). Although the L50 of 154 mm and L95 of 177 mm at maturity for restocked males were both greater than the corresponding values of 149 and 159 mm, respectively for wild stock males, the differences were likewise not significant (P>0.05, Fig. 5.4f).
No wild stock or restocked females reached maturity by the end of its first year of life (Fig. 5.5a,b). While virtually all wild stock females had become mature by two years of age, only 54% of restocked females had become mature by that age and some restocked females had not become mature by 3 and 4 years of age.
The ogive describing the relationship between the proportions of mature restocked females and age was significantly different (P<0.001) from the corresponding ogive for wild stock fish (Fig. 5.5a-c, Table 5.3). The A50 for maturity of 2.1 years of restocked females was greater than the corresponding value of 1.7 years for wild stock females, but this difference was not significant (P>0.05). The A95 for restocked females of 3.2 years was appreciably greater (P<0.001) than the 2.0 years for wild stock females (Fig. 5.5c).
Some wild stock males attained maturity at the end of their first year of life and all became mature at the end of their second year of life, whereas no restocked males became mature at the end of their first year of life and nearly 20% were still immature at the end of their second year of life (Fig. 5.5d,e).
105
Figure 5.5. Percentage frequency of occurrence, in sequential ages, of females and males of wild stock and restocked Acanthopagrus butcheri with immature (white bars) and mature gonads (grey bars), derived from samples collected during the spawning season of A. butcheri. Logistic curves (solid lines) and their 95% credible intervals (dotted lines) for the probability that a fish at a given age is mature are superimposed on a, b, d, e. Sample size for each length class is provided. Comparisons of logistic curves for wild stock (dashed line) and restocked (continuous line) females and males are shown in c and f.
106
The ogive describing the relationship between the proportions of mature restocked males and age differed significantly (P<0.001) from the corresponding ogive for wild stock fish (Fig. 5.5d-f , Table 5.3). The A50 of 1.8 years and A95 of 2.3 years for restocked males were both greater than and differed significantly (P<0.001) from the corresponding values of 1.2 and 1.6 years, respectively, for the wild stock males (Fig. 5.5f).
5.3.3 Contribution of restocked fish to commercial catches
In any description of the length compositions of A. butcheri in commercial catches from the Blackwood River Estuary, it must be borne in mind that the stretched mesh size of the commercial gill net (105 mm) was chosen by the commercial fisher so that it essentially caught only A. butcheri with lengths ≥250 mm, the MLL for retention of this species. The commercial catches of A. butcheri in 2005 comprised almost exclusively wild stock fish, with their lengths ranging from 257 to 412 mm and producing a modal length class of 280-
299 mm (Fig. 5.6a). The samples in 2005 did contain, however, two restocked fish of the
2002 year class and 14 restocked fish of the 2001 year class, whose lengths collectively ranged from 255 to 295 mm. In 2006, the overall modal length class decreased to 260-279 mm, which was due to the recruitment into the fishery of appreciable numbers of restocked fish of the 2002 year class, whose lengths ranged from 253 to 325 mm. The modal length class of restocked fish, produced predominantly by the 2002 year class, increased from
260-279 mm in both 2006 and 2007 to 280-299 mm in 2009 and 2010. During this period, the lower end of the overall distribution of lengths was dominated by cultured fish, whose lengths were almost invariably <340 mm, whereas the upper end contained several wild fish with lengths extending up to 426 mm (Fig. 5.6a).
107
Figure 5.6 a) Length and b) age-frequency distributions for samples of commercial gill net catches of Acanthopagrus butcheri in the Blackwood River Estuary between 2005 and 2014. White and black bars represent the 2001 and 2002 year classes of restocked fish, respectively, and the grey bars the wild fish, i.e. those whose otoliths did not contain the purple colouration of the alizarin stain that characterises those of cultured fish. Arrows track the 1999 year class of the wild stock and the 2002 year class of restocked fish, which were the dominant year classes between 2006 and 2010.
108
In contrast to 2010, when virtually all fish with lengths <320 mm were restocked
fish, the majority of A. butcheri <320 mm caught in 2012, and even more particularly in
2013 and 2014, were fish whose otoliths were not stained with alizarin complexone, and
were thus spawned naturally in the wild (Fig. 5.6a).
The ages of A. butcheri in commercial catches in 2005, which comprised almost
exclusively wild stock fish, ranged from 2 to 10 years old, with one fish 15 years old and
another 18 years old (Fig. 5.6b). The samples in that year were dominated by the 5+ and to
a lesser extent the 4+ age class, i.e. fish spawned in 1999 and 2000, respectively. The 1999
year class remained the most abundant year class of wild stock fish in the years up to 2010,
being represented, for example, by substantial numbers of the 7+ age class in 2007 and of
the 10+ age class in 2010. However, restocked fish, mainly of the 2002 year class,
increasingly dominated the catches in this period, with their contributions to commercial
catches increasing progressively from 32% in 2006 to 74% in 2010 (Fig. 5.6b).
Although there had been little or no recruitment of year classes of wild stock fish into the fishery in 2006 to 2010, this situation changed dramatically in 2012, when a strong new cohort, representing the 2008 year class, appeared in samples, at which time they were
3 years old (Fig. 5.6b). This strong recruitment was reflected in the marked increase in the prevalence of smaller fish in 2012 to 2014 (Fig. 5.6a). The 2008 year class became so strong in 2013 and 2014, when its individuals were 4 and 5 years old, respectively, that it dominated the age-frequency distributions for those years (Fig. 5.6b). The dominance of this year class in 2013 and 2014 accounts, in part, for the decline in the relative contribution of restocked fish from 74% in 2010 to 19% in 2013 and 10% in 2014 (Fig. 5.6b).
109
5.3.4 Egg production
On the basis of the relationship between fecundity and total length (Sarre & Potter, 1999) and the lengths and numbers of fish in each year class, the contribution to egg production by the 1999 year class of wild stock females was 25% in 2007 and 18% in 2009, in which years these fish were 8 and 10 years old, respectively (Fig. 5.7). These were the greatest contributions by any year class of wild stock females in both years and particularly in 2007.
The contribution to total egg production by females of restocked fish was 54% in 2007 and
55% in 2009, due predominantly to that made by the 2002 year class. Thus, the restocked fish made a greater contribution than wild stock fish to total egg production in 2007 and
2009 (Fig. 5.7).
Figure 5.7 Percentage contributions made to egg production by the different year classes of wild stock and restocked Acanthopagrus butcheri in 2007 and 2009 and by the wild stock and restocked fish overall in those two years. Data were derived from the lengths of individuals of each age class in commercial gill net samples and employed the fecundity- length relationship for A. butcheri in Sarre & Potter (1999). 110
5.4 DISCUSSION
5.4.1 Survival and growth of restocked fish
Although the number of A. butcheri cultured in 2001 and released into the estuary in 2002 was nearly half that cultured in 2002 and released in 2003, i.e. 70,000 vs 150,000, only 6% of the restocked A. butcheri taken by commercial gill netting between 2005 and 2014 belonged to the 2001 cohort. The far lower survival of the 2001 than 2002 cohort of cultured fish may have been related to differences in environmental conditions at the time of their release. Thus, whereas the 2001 cohort was released in winter, when freshwater discharge was increasing and water temperatures were at their minimum, the 2002 cohort was released during autumn, when freshwater discharge was limited and water temperatures were still relatively high. The more ‘benign’ water conditions in autumn than winter would have afforded the individuals of the 2002 cohort a better opportunity to
‘adjust’ to their new environment before facing the rigours of heavy freshwater discharge in winter. The longer time spent in the hatchery by the 2001 cohort (7 vs 4 months) may also have had a negative impact on their ability to adjust to their environment.
5.4.2. Comparisons of the performance of restocked and wild fish
This study demonstrated that, in terms of growth and maturity schedules, restocked
A. butcheri did not perform as well as its wild stock in the Blackwood River Estuary.
However, any consideration of the efficacy of using restocking as a tool for replenishing the depleted stock of this species in this system should take into account the extent of the differences and how the performance of cultured fish compares with that of this species in other estuaries. Thus, while the L∞s of restocked and wild stock fish differed significantly, the difference was only 8% and the rates, k, at which the total lengths at age of restocked
111 and wild stock fish approached their respective asymptotic lengths, were not significantly different. The fact that the growth curve, fitted to the lengths at age of the 2008 year class lay between that for the restocked and wild stock, suggests that the growth of this year class is intermediate, recognising, however, that the difference between the growth of the restocked and wild fish was relatively small.
On the basis of the von Bertalanffy growth equation, restocked fish took 4.3 years to reach the MLL of 250 mm for A. butcheri and thus longer than the 3.4 years taken by wild fish. However, the MLL was reached at an earlier age by restocked fish in the
Blackwood River Estuary than recorded by Sarre & Potter (2000) for this species in 1993-
95 in two other south-western Australian estuaries (~6.7 and 7 years) and at a similar age in another estuary in that period (4.4 years). While restocked A. butcheri took longer to reach the MLL in the Blackwood River Estuary than required in a fourth system, the Swan
River Estuary, in that earlier period, i.e. ~2.8 years, they attain the MLL far more rapidly than in that same system in recent years, i.e. ~7 years, when it had become severely degraded (Chapter 3).
Some restocked fish, and particularly their females, were still immature in the length and age classes at which all wild stock fish had become mature. This accounts for the L50 and A50 at maturity, and, to an even greater extent, the L95 and A95, all being greater for restocked females than wild stock females, recognising that the differences were statistically significant only in the case of the L95s and A95s. Although restocked males matured at a greater length and age than wild stock males, the differences were statistically significant only for the A50s and A95s at maturity. Furthermore, restocked females attained maturity at an age very similar to the 1.9 years for A. butcheri in 1993-95 in the Swan
River Estuary, ~300 km further north, and at a slightly younger age than the 2.5 years estimated for that system in 2007-11 (Chapter 3).
112
Although the differences in growth and maturity schedules of restocked and wild
A. butcheri in the Blackwood River Estuary were relatively small, they parallel the trend in other studies for the long-term performance of the cultured fish of a species to be inferior to that of its wild stock (Jonsson & Jonsson, 2006; Lorenzen et al., 2012).
5.4.3 Contributions of restocked and wild fish to commercial gill net catches
The progressive rise, in the relative contributions of restocked fish to the annual commercial catches of A. butcheri, from negligible levels in 2005 to about three-quarters in
2010 is attributable to two factors. First, through growth, the number of restocked fish that attained the MLL of 250 mm increased progressively each year. Second, the recruitment into the commercial fishery of all year classes of the wild stock of A. butcheri, except for the 1999 year class, was relatively low between 2005 and 2010, probably representing poor spawning success in earlier years. It is thus evident that restocked fish made a major contribution to the commercial catches for a number of years and thus, implicitly, also to those of recreational fishers.
The decline in the contribution of restocked fish to commercial catches in 2012 was largely due to an increase in the contribution of the 2008 year class. The subsequent increase in the proportion of the 2008 year class in 2013 and 2014 reflected an increase in the numbers of that year class that would have reached the MLL and a decline in the abundances of both the 2002 cohort of restocked fish and the 1999 cohort of wild stock fish as a result of natural and fishing mortality.
The question now arises as to whether the progeny of cultured fish could potentially have contributed to the substantial number of individuals that comprised the
2008 year class. On the basis of the A50 at maturity of ~2 years, the restocked A. butcheri of the 2002 year class would, on average, have matured in 2004 and thus could potentially
113 have spawned in that and subsequent years. However, as mentioned above, the age compositions in commercial catches imply that spawning success was very limited between 2000 and 2007. Thus, as the vast majority of wild stock fish in 2007 and 2009, and thus implicitly in 2008, were ≥8 years in age, the relatively strong spawning success in
2008 was presumably due to spawning by cultured fish (particularly of the 2002 year class) and/or wild stock fish or to interbreeding between these two groups.
Although commercial catch samples were not available in 2008, it is relevant that restocked fish were estimated to contribute 54 to 55% to the total egg production by legal- sized A. butcheri in the Blackwood River Estuary in 2007 and 2009. Restocked fish would thus presumably have made a similar contribution in 2008, the year in which recruitment into the population, and thus implicitly spawning success, was particularly strong.
The spawning success of the restocked individuals of some species has been shown to be inferior in certain respects from that of the wild stock of that species, due, for example, to differences in the timing of breeding and/or selection of suitable nest sites
(Fleming & Petersson, 2001, McGinnity et al., 2003; Weir et al., 2004). However, as the monthly trends in ovarian weights and locations of capture were the same in restocked fish as wild stock fish, the two groups spawn at the same time and in the same region of the estuary. Furthermore, as discussed earlier, the differences in the reproductive schedules of the restocked and wild stock fish were relatively small. It is also relevant that the results of a thorough genetic study strongly indicated that the prevalence of inbreeding was not demonstrably greater among the restocked than wild stock A. butcheri (Gardner et al.,
2013).
In summary, this study has shown that the growth of restocked A. butcheri in the
Blackwood River Estuary was only slightly less than that of wild fish likewise hatched in
2001 and 2002 and therefore exposed to the same environmental conditions for the twelve
114 to thirteen years since those cultured fish were introduced into this system. It also demonstrated that an appreciable number of restocked fish were still present in the estuary, eleven years after their introduction to this system, and that all restocked fish, which survived to five years of age, reached maturity and therefore potentially could spawn and contribute to future generations of A. butcheri in this estuary. In addition, the restocked fish were shown to make a very substantial contribution to the commercial gill net fishery for this sparid in the Blackwood River Estuary in 2007-10. There is also strong circumstantial evidence that the cultured fish had contributed progeny that were caught by the commercial fisher for A. butcheri in the Blackwood River Estuary. The results of this study provide a rare example of a continuing investigation into the consequences of a fish restocking programme in an estuary and demonstrate that it is a worthwhile proposition to consider the use of restocking with appropriately cultured fish to replenish a stock of a species such as A. butcheri, whose individuals are confined to their natal estuary.
Acanthopagrus butcheri is also a very good candidate for restocking in the Blackwood and some other estuaries as natural recruitment is so episodic in these systems (Morison et al.,
1998) and density dependent interactions between wild and restocked fish would be likely to be low.
115
Chapter 6
General Discussion
6.1 Swan River Estuary
6.1.1 Influence of hypoxia
As discussed in the General Introduction, the opportunities provided by estuaries for agriculture, industry and urban development have been widely exploited, resulting in the input into these systems of organic material, other nutrients and toxic materials, thereby leading inter alia to eutrophication, hypoxia and mortality among their faunas (de Jong et al., 2002; Kennish, 2002; Breitburg et al., 2009). This has led to estuaries in temperate regions becoming the most degraded of all marine environments (Jackson et al., 2001).
Such degradation has become so pronounced in the Swan River Estuary that it was rated one of the two most hypertrophic of the 131 estuarine and coastal ecosystems worldwide for which Cloern et al. (2014) collated data.
Because the estuaries in south-western Australia are poorly flushed, they are particularly prone to accumulating nutrients and thus to eutrophication (Tweedley et al.,
2013, 2014). This poor flushing is the result of the small tidal water movements in the estuaries of this microtidal region and to the restriction of most freshwater discharge to the wet ‘winter’ months. As freshwater discharge is very restricted during summer, flushing is limited at the time when water temperatures are greatest and biological oxygen demand
(BOD) is therefore also greatest. Furthermore, the problem of removing organic material and other nutrients from the Swan River Estuary through flushing has been exacerbated during recent years by a marked decline in the volume of freshwater discharge entering the estuary in winter and early spring as a consequence of a reduction in rainfall (Chapter 3).
116
The results reported in Chapter 3 demonstrated that the decline in freshwater discharge in the Swan River Estuary between the early 1990s and late 2000s led to progressive increases in hypoxia in its upper region. These marked deleterious changes in environmental conditions were shown to be accompanied by reductions in the body condition, growth and length at maturity of Acanthopagrus butcheri and by an increase in age at maturity (Fig. 6.1, Chapter 3). The direction of each of these changes is consistent with the stress that would be expected on the individuals of a species exposed to detrimental conditions (Kramer, 1987; Pichavant et al., 2001; Stearns, 1992; Eby et al.,
2005; Roberts et al., 2011). It was concluded that the changes in biological characteristics were attributable to the effects of hypoxia directly and to a density-dependent effect. The latter was produced by the increase in density that occurred through larger A. butcheri moving from the deeper hypoxic waters into the better oxygenated shallow waters where the smaller fish are found. Such habitat compression parallels that observed with certain pelagic fish species in the eastern tropical Pacific and Atlantic oceans, where, as a response to hypoxia, they became concentrated in the better oxygenated surface waters (Prince &
Goodyear, 2006; Stramma et al., 2012), which, in one study, were shown to extend downwards to the thermocline at 25 m, below which lies the cold hypoxic water (Prince &
Goodyear, 2006).
Although the development of severe hypoxia in the deeper waters led to habitat compression and an overall decline in the biological performance of A. butcheri in the
Swan River Estuary between the early 1990s and the early to mid-2000s, the limited CPUE data indicated that this did not result in a marked overall decline in the abundance of this species in the upper region of this estuary between these years (Chapter 3).
117
Figure 6.1 Graphical representation of the changes in that occurred in the environment of the Swan River Estuary between 1994/95 and 2007/08 and the accompanying changes in certain biological characteristics of Acanthopagrus butcheri. Symbols courtesy of the Integration and Application Network (ian.umces.edu/symbols/)
6.1.2 Influence of temperature
On the basis of the metabolic theory of ecology (Angilletta et al., 2004; Brown et al., 2004;
Weber et al., 2015), the growth rate of A. butcheri in the Swan River Estuary would be expected to increase with increasing temperature. However, although the annual mean value of the maximum daily air temperatures for the months between November and
March, when growth mainly occurs, increased by ~1.0 °C between the end of the first sampling period (1994/95) and the beginning of the final sampling period (2007/08), the overall growth of A. butcheri decreased between these periods. Thus, any potential influence of temperature on the changes in growth between the earlier and later periods was over-ridden by other factors, i.e. hypoxia and increased densities. However, between
2007 and 2014, when A. butcheri resided predominantly in shallow waters and densities remained high and freshwater discharge consistently low, growth was positively correlated
118 with temperature. As temperature also influences the frequency and amount of food ingested, as well as the rate of digestion and absorption (Brander, 2009), the increased temperatures were also presumably accompanied by an increased availability to utilize prey. The optimum temperature for growth has also been shown to be higher for younger fish (Árnason et al, 2009; Brander, 2009). Thus, although increasing temperatures may result in faster growth of small fish, larger fish may fall outside their optimal temperature range and growth becomes depressed (Árnason et al, 2009; Brander, 2009).
Although the growth of A. butcheri in the Swan River Estuary during recent years was related to temperature (Chapter 4), comparisons of the growth of this sparid in four estuaries in 1993-95 (Sarre & Potter, 2000) suggested that there is no consistent relationship between growth and temperature when considered across estuaries. Thus, the growth rates of juveniles were lowest in the Moore River Estuary, the most northern and therefore warmest of these systems, and were far less than those in the Wellstead Estuary, which is much further south and therefore cooler.
An example of the extent to which the growth of A. butcheri can be influenced by different environmental conditions is provided by the results of a study in which A. butcheri were cultured separately from brood stock from the Moore River and Swan River estuaries and the growth of the resultant juveniles recorded for ~270 days (Partridge et al.,
2004). Despite the growth of A. butcheri being far slower in the Moore River Estuary than the Swan River Estuary, the cultured juveniles grew at the same rate when maintained under identical conditions in the laboratory. This clearly implies that, in some way, conditions in the Moore River Estuary were far less favourable for growth than those in the
Swan River Estuary. It therefore appears relevant that the densities of the juveniles of A. butcheri in the Moore River Estuary were greater than in the other three estuaries studied by Sarre & Potter (2000) and that the diet of this species in that estuary comprised a far
119 higher component of macrophytes than those in the other three estuaries (Sarre et al.,
2000). Thus, density-dependent effects and/or quality of available food could account for the differences in the growth of A. butcheri in those systems. In addition, the results of the above culture experiment demonstrate that, although the populations in the two estuaries are genetically distinct (Chaplin et al., 1998), the differences in the growth of A. butcheri in the two systems are not genetically based.
6.1.3 Recruitment
The studies reported in this thesis for the Swan River Estuary were aimed at comparing the condition, growth and reproductive schedules of A. butcheri in two periods between which this estuary has undergone very pronounced anthropogenic changes. The sampling regime was thus not designed to explore rigorously the extent of any inter-annual differences in recruitment and the factors that influence any such variations. However, the data compiled for the Swan River Estuary does allow some tentative conclusions to be made regarding recruitment, taking into account the results of extensive studies on the recruitment of
A. butcheri in estuaries in eastern Australia (Nicholson et al., 2008; Jenkins et al., 2010,
2015; Williams et al., 2012, 2013).
The age compositions of A. butcheri in the Swan River Estuary in the years between 2007 and 2014 showed that recruitment varied to some degree among years, with, for example, the 2005 year class well represented and the 2006 year class poorly represented (Fig. 6.2). A more extreme situation was recorded for A. butcheri in the
Gippsland Lakes in south-eastern Australia where, on the basis of the age composition in samples collected between 1993 and 1996, recruitment was highly episodic and thus the commercial and recreational fishers relied very heavily on just two age classes (Morison et
120 al., 1998). Recruitment of A. butcheri in this estuary was shown also to be episodic in subsequent years (Jenkins et al., 2010).
Studies in south-eastern Australian estuaries demonstrated that the recruitment of
A. butcheri was negatively correlated with freshwater discharge during the spawning season and positively correlated with the level of stratification in that period (Jenkins et al.,
2015). However, on average, freshwater discharge peaks earlier and much more sharply in the Swan River Estuary than in the Mitchell River, which flows into the Gippsland Lakes
(Fig. 6.3). Furthermore by November, i.e. just after peak spawning, it has declined to relatively far lower levels in the Swan River Estuary than in the Mitchell River. In at least relative terms, the trend exhibited by freshwater discharge in the Swan River Estuary differed from that in the Mitchell River. An examination of the vertical salinity profiles in the Swan River Estuary in November provided no clear evidence that a year of good recruitment is related to the presence of a well-defined halocline. It would thus appear that the factors influencing recruitment in the Swan River Estuary and south-eastern Australian estuaries differ and that inter-annual variations in recruitment are not as pronounced in the
Swan River Estuary.
It would be worthwhile to establish standardised sampling regimes for A. butcheri in south-western Australia, along the lines of that in Victoria (Jenkins et al., 2010;
Williams et al., 2012, 2013), aimed at collecting larvae in the period immediately following peak spawning and representative samples of 0+ fish just after the spawning period. This would enable the relationship between recruitment and freshwater discharge and thus salinity stratification to be explored in detail.
121
Figure 6.2 Age-frequency distributions of Acanthopagrus butcheri caught in the Swan River Estuary between 2007 and 2014. Solid and dashed lines track the 2005 and 2006 year classes, respectively.
Figure 6.3 Mean monthly freshwater discharges into the Mitchell River (gauging station 224203) and Swan River (gauging station 616011) recorded between 1990 and 2015 as percentages of annual totals. Data were provided by Victorian Water Measurement Information System (http://data.water.vic.gov.au/ monitoring.htm) and the Western Australian Department of Water (2015), respectively.
122
6.1.4 Management
It has to be borne in mind that the CPUEs for A. butcheri during the last 25 years were derived from the catches of a limited number of recreational fishers and that the commercial fishery declined markedly during that period as a result of a government buy- back of licences (Chapter 3; Smith, 2006; Smith et al., 2014). Thus, caution must be exercised when drawing conclusions from those CPUEs. It would be useful if reliable recreational catch and effort data for A. butcheri in the Swan River Estuary (and other estuaries) could be collected on a regular basis as such data would facilitate exploration more fully of the trends in abundance of this species and the factors influencing that abundance.
6.2 Blackwood River Estuary
The results of the study in the Blackwood River Estuary demonstrate that, in terms of growth and maturity schedules, the A. butcheri that had been cultured from brood stock from this estuary performed almost as well as the resident wild stock (Chapter 5). They also showed that, once the cultured fish had reached the MLL of 250 mm, they made a very substantial contribution to the commercial catches for A. butcheri in this estuary (Fig.
6.4). The results also provide strong circumstantial evidence that the restocked fish made a significant contribution to the successful spawning of A. butcheri in the Blackwood River
Estuary in 2008. The importance of the restocked fish to the population and fishery for
A. butcheri in the Blackwood River Estuary cannot thus be overestimated.
Although restocking can overcome the need for imposing stringent size and bag limits and fishing closures, prior to any restocking, a review of alternative management options require evaluation. Furthermore, any restocking would need to be undertaken in a manner that maintains the biological integrity of the wild stock and does not have a significant effect on other species or the environment. 123
Figure 6.4 Graphical representation showing the introduction of cultured fish (pink) in 2002/03 and their growth and contribution to the commercial fishery between 2005 and 2014. Symbols courtesy of the Integration and Application Network (ian.umces.edu/symbols/).
The above results for the Blackwood River Estuary can now be considered in the context of those for the Swan River Estuary to develop concepts as to the why the stock of
A. butcheri in the Blackwood River Estuary declined in the past (Lenanton et al., 1999).
The age-frequency distributions of the wild stock of A. butcheri in commercial catches from 2005 to 2014 emphasise that, in contrast to the Swan River Estuary, natural recruitment of this species in the Blackwood River Estuary was very poor in most years.
The basis for this poor recruitment probably represents a continuation of the unidentified problem that led to a decline in the fishery in earlier years. It is thus relevant to revisit how
A. butcheri in the Swan River Estuary responded, in its behaviour, to the detrimental environmental changes that occurred in that system.
The data presented in Chapter 3 provide strong evidence that, as freshwater discharge declined, the deeper waters of the upper Swan River Estuary became increasingly hypoxic. During the same period, the mean annual freshwater discharge in the
Blackwood River Estuary declined significantly (Fig. 6.5, P<0.05, r = -0.35). It is thus
124 relevant that oxygen concentrations at the bottom of the water column in the deeper waters of the Blackwood River Estuary were frequently <2 mgL-1 between 2001 and 2010, and particularly so since 2006 (Brearley, 2013). In contrast, oxygen concentrations were usually >2mgL-1 in 1999 to 2001 and fragmentary records indicate that oxygen concentrations were higher in earlier years (Brearley, 2013). There is thus very strong circumstantial evidence that, as in the upper Swan River Estuary, the decline in freshwater discharge in the Blackwood River Estuary was accompanied by an increase in the extent of hypoxia in the deeper waters of this system.
Figure 6.5 Mean freshwater discharge in the Blackwood River Estuary in each year between 1990 and 2015.
The question now arises as to why the abundance of A. butcheri apparently declined in the Blackwood River Estuary and yet, from the restricted CPUE data, did not change markedly in the Swan River Estuary. The problem posed by hypoxia to large A. butcheri in the deeper waters of the Swan River Estuary was overcome by their tending to move into the wide and better oxygenated shallow margins of this estuary (Chapter 3). The corresponding margins of the Blackwood River Estuary are, however, far narrower (Fig.
6.6). It is thus proposed that there was insufficient nearshore habitat to accommodate many
125 large, as well as numerous small A. butcheri, and thus the overall abundance of this species in the Blackwood River Estuary declined. At the same time, some large A. butcheri are present in nearshore, shallow waters and, from the results of our early, preliminary studies using gill nets, are in greater abundance in the shallows than in deeper waters. The view that the distribution of fish changes in response to decreasing flow and increasing hypoxia in deeper waters would explain why, in the past, commercial fishers for A. butcheri changed from setting their gill nets in deeper waters to shallow, nearshore waters, a practice still adopted by the sole commercial fisher for A. butcheri in the Blackwood River
Estuary (T. Price, pers. comm.).
Figure 6.6 Graphical illustration of a cross section of the upper region of a) the Swan River Estuary and b) Blackwood River Estuary showing the differences in their morphology. Symbols courtesy of the Integration and Application Network (ian.umces.edu/symbols/).
The decline in freshwater discharge and increasing hypoxia in the Blackwood River
Estuary may also have had a detrimental effect on the larvae of A. butcheri in this system.
Work in south-eastern Australia demonstrated that the larvae of this species aggregate in the vicinity of the halocline and thus in the same region of the water column as its zooplankton prey (Williams et al., 2012, 2013). However, preliminary studies by Joel
Williams (pers. comm.) indicate that, in contrast, the larvae of A. butcheri are distributed below the halocline and in much deeper water in the Blackwood River Estuary. Williams
126 suggests that this represents a mechanism for avoiding predation in the region around the halocline, which, through the presence of greater visibility due to much lower turbidity, would make the larvae more susceptible to being consumed by visual predators. However, migration of the larvae into the deeper waters of the Blackwood River Estuary would take the larvae into areas where the oxygen concentrations and abundance of prey are greatly reduced. The proposal that low oxygen concentrations could lead to substantial mortality of the early life cycle stages of A. butcheri is consistent with the results of Hassell et al.
(2008a,b).
From the above, it follows that the success of the cultured fish for restocking A. butcheri could be due to the bypassing of the sensitive larvae phase. If this proves to be the case, restocking may be the best way of replenishing depleted stocks of this species in other estuaries.
6.3 Conclusions
The conclusion that marked declines in rainfall and thus freshwater discharge in the Swan
River and Blackwood River estuaries has led to hypoxia and other related detrimental effects for A. butcheri raise the question of how widespread such effects are in south- western Australian estuaries. It is thus relevant that the volume of freshwater discharge entering other estuaries in south-western Australia has also declined (Fig 6.7). Further studies are thus clearly required to determine whether this has also led to similar detrimental effects on A. butcheri. If this proves to be the case, there is clearly a need to ensure that discharge levels in those estuaries are not impeded any further through, for example water abstraction and farm dams and, better still, are improved. Indeed, protecting natural flow regimes is likely to be the most effective management strategy to maintain the
127
‘health’ of estuaries and their fisheries (see also Gillson, 2011). It follows that a thorough understanding of the hydrology of estuaries will become increasingly important to managing these systems as they become subjected further to the effects of climate change and increases in the demand for water resources by human populations.
In view of anticipated climatic and other anthropogenic changes, it would be very useful to have a long time series of data for a well-studied species whose biological characteristics can be used to monitor the biological effects of those changes.
Acanthopagrus butcheri possesses many of the required characteristics to act as an indicator species. It is widely distributed and relatively abundant in many estuaries in southern Australia and is relatively long-lived for an estuarine species. It is also confined to its natal estuary and thus subjected to essentially the same environment throughout the whole of their life (Bortone, 2003). As A. butcheri is an important recreational and commercial fish species, there is a solid database on its biological characteristics (e.g.
Morison et al., 1998; Sarre & Potter, 1999, 2000; Jenkins et al., 2010, 2015; Williams et al., 2013). Furthermore, the plasticity of certain biological characteristics and the durability of A. butcheri means that this species will respond to environmental differences (Partridge et al., 2004) and is capable of surviving relatively severe environmental changes.
128
Figure 6.7 Freshwater discharge into five estuaries in south-western Australia in the years between 1990 and 2015. Data provided by the Western Australia Department of Water.
129
REFERENCES
Able, K.W., Fahay, M.P., 2010. Ecology of Estuarine Fishes: Temperate Waters of the Western North Atlantic. Baltimore, Maryland. Johns Hopkins University Press, pp. 566. Al-Dubakel, A.Y., 2011. Commercial fishing and marketing of Hilsa Shad Tenualosa ilisha (Hamilton-Buchanon, 1822) in Basrah-Southern IRAQ. Emirates Journal of Food & Agriculture 23, 178-186. Anderson, C.N.K., Hsieh, C, Sandin, S.A., Hewitt, R., Hollowed, A., Beddington, J., May, R.M., Sugihara, G., 2008. Why fishing magnifies fluctuations in fish abundance. Nature 452, 835-839. Angilletta, M.J., Cooper, B.S., Schuler, M.S., Boyles, J.G., 2010. The evolution of thermal physiology in endotherms. Frontiers in Bioscience (elite Edition) 2, 861-881. Angilletta, M.J., Steury, T.D., Sears, M.W., 2004. Temperature, growth rate, and body size in ectotherms: fitting pieces of a life-history puzzle. Integrative and Comparative Biology 44, 498-509. Árnason, T., Björnsson, B., Steinarsson, A., Oddgeirsson, M., 2009. Effects of temperature and body weight on growth rate and feed conversion ratio in turbot (Scophthalmus maximus). Aquaculture 295, 218-225. Australian Bureau of Meteorology, 2015. Climate Data Online. Location: Perth Airport. http://www.bom.gov.au/climate/data/?ref=ftr. Ayvazian, S.G., Johnson, M.S., McGlashan, D.J., 1994. High levels of genetic subdivision of marine and estuarine populations of the estuarine catfish Cnidoglanis macrocephalus (Plotosidae) in southwestern Australia. Marine Biology 118, 25-31. Bacheler, N.M., Buckel, J.A., Paramore, L.M., 2012. Density-dependent habitat use and growth of an estuarine fish. Canadian Journal of Fisheries and Aquatic Sciences 69, 1734-1747. Baird, D., 1999. Estuaries as ecosystems: a functional and comparative analysis. In: Allanson, B., Baird, D. (eds) Estuaries of South Africa. Cambridge, Cambridge University Press, pp. 269-287. Baird D., Evans, P.R., Milne, H., Pienkowski, M.W., 1985. Utilisation by shorebirds of benthic invertebrate production in intertidal areas. Oceanography and Marine Biology, 23, 575–597.
130
Beauchamp, J.J., Olson, J.S., 1973. Corrections for bias in regression estimates after logarithmic transformation. Ecology 54, 1403-1407. Bell, J. D., Bartley, D. M., Lorenzen, K., Loneragan, N. R., 2006. Restocking and stock enhancement of coastal fisheries: Potential, problems and progress. Fisheries Research 80, 1-8. Bell, J.D., Leber, K.M., Blankenship, H. L.,Loneragan, N.R., Masuda, R., 2008. A new era for restocking, stock enhancement and sea ranching of coastal fisheries resources. Reviews in Fisheries Science 16, 1-9. Berk, K.N., 1978. Comparing results of backward and forward selection and all subsets regression in theory and examples. Technometrics, 20, 1-6. Beverton, R.J.H., Holt, S.J., 1957. On the dynamics of exploited fish populations. Fisheries Investigations Series 2, Volume 19. U.K. Ministry of Agriculture and Fisheries, London, pp. 533. Bjørkvoll, E., Grøtan, V., Aanes, S., Sæther, B.-E., Engen, S., Aanes, R., 2012. Stochastic Population dynamics and life-history variation in marine fish species. The American Naturalist 180, 372-387. Blaber, S.J.M., 1997. Fish and fisheries of tropical estuaries. Fish and Fisheries Series 22. London, Chapman and Hall, pp. 367. Blaber, S.J.M., 2000. Tropical Estuarine Fishes: Ecology, Exploitation and Conservation Fish and Aquatic Resources Series 7. London, Blackwell Science, pp. 372. Blaber, S.J.M., Blaber, T.G., 1980. Factors affecting the distribution of juvenile estuarine and inshore fish. Journal of Fish Biology 17, 143-162. Blumer, L.S., 1982. A bibliography and categorization of bony fishes exhibiting parental care. Zoological Journal of the Linnean Society 76, 1-22. Bortone, S.A., Martignette, A.J., Spinelli, J.P., 2006. Spotted seatrout (Family Sciaenidae) growth as an indicator of estuarine conditions in San Carlos Bay, Florida. Florida Scientist 69, 127-139. Brander, K.M. 2009. Fisheries and Climate. In Steele, J.H., Thorpe, S.A., Turekian, K.K. (eds) Climate and Oceans: A derivative of Encyclopaedia of Ocean Science, Second Edition, pp. 550-557. Brearley, A., 2005. Ernest Hodgkin’s Swanland. Estuaries and Coastal Lagoons of South- Western Australia. University of Western Australia Press, pp. 550.
131
Brearley, A., 2013. Revisiting the Blackwood River and Hardy Inlet, 40 years of change. An environmental review of the Blackwood River Estuary Western Australia 1974- 2010. Ernest Hodgkin Trust for Estuary Education and Research, pp. 88. Breitburg, D.L., Hondorp, D.W., Davias, L.A., Diaz, R.J., 2009. Hypoxia, nitrogen, and fisheries: Integrating effects across local and global landscapes. Annual Reviews in Marine Science 1, 329-349. Brown, J.H., Gillooly, J.F., Allen, A.P., Savage, V.M., West, G.B., 2004. Toward a metabolic theory of ecology. Ecology 85, 1771-1789. Bruce, L.C., Cook, P.L.M., Teakle, I., Hipsey, M.R., 2014. Hydrodynamic controls on oxygen dynamics in a riverine salt-wedge estuary, the Yarra River Estuary, Australia. Hydrology and Earth System Sciences 18, 1-14. Burridge, C.P., Versace, V.L., 2006. Population genetic structuring in Acanthopagrus butcheri (Pisces: Sparidae): Does low gene flow among estuaries apply to both sexes? Marine Biotechnology 9, 33-43. Butcher, A., Mayer, D., Smallwood, D., Johnston, M., Williams, L.E., Clapham, S., 2000. Assess the effectiveness of stocking. In Maroochy Estuary Fish Stocking Program 1995–99. Final Report. Queensland: Department of Primary Industries, pp. 52-75. Caddy, J.F. 2000. Marine catchment basin effects versus impacts of fisheries on semi- enclosed seas. ICES Journal of Marine Science 57, 628-640. Camp, E.V., Lorenzen, K., Ahrens, R.N M., Allen, M.S., 2014. Stock enhancement to address multiple recreational fisheries objectives: an integrated model applied to red drum Sciaenops ocellatus in Florida. Journal of Fish Biology 85, 1868-1889. Campana, S.E., 2001. Accuracy, precision and quality control in age determination, including a review of the use and abuse of age validation methods. Journal of Fish Biology 59, 197-242. Campbell, L.A., Rice, R.A., 2014. Effects of hypoxia-induced habitat compression on growth of juvenile fish in the Neuse River Estuary, North Carolina, USA. Marine Ecology Progress Series 497, 199-213. Canuel, E.A., Cloern, J.E., Ringelberg, D.B., Guckert, J.B., Rau G.H., 1995. Molecular and isotopic tracers used to examine sources of organic matter and its incorporation into the food webs of San Francisco Bay. Limnology and Oceanography 40, 67-81. Cebrian, J., 1999. Patterns in the fate of production in plant communities. The American Naturalists 154, 449-468.
132
Cerrato, R.M., 1990. Interpretable statistical test for growth comparisons using parameters in the von Bertalanffy equation. Canadian Journal of Fisheries and Aquatic Sciences 47, 1416-1426. Chan, T.U., Hamilton, D.P., 2001. Effect of freshwater flow on the succession and biomass of phytoplankton in a seasonal estuary. Marine and Freshwater Research 52, 869-884. Chang, W.Y.B., 1982. A statistical method for evaluating the reproducibility of age determination. Canadian Journal of Fisheries and Aquatic Sciences 39, 1208-1210. Chaplin, J.A., Baudains, G.A., Gill, H.S., McCulloch, R., Potter, I.C., 1998. Are assemblages of black bream (Acanthopagrus butcheri) in different estuaries genetically distinct? International Journal of Salt Lake Research 6, 303-321. Chen, Y., Mello, L.G.S., 1999. Growth and maturation of cod (Gadus morhua) of different year classes in the Northwest Atlantic, NAFO subdivision 3Ps. Fisheries Research 42, 87-101. Chuwen, B.M., 2009. Characteristics of the ichthyofaunas of offshore waters in different types of estuary in Western Australia, including the biology of Black Bream Acanthopagrus butcheri. Ph.D. thesis. Murdoch University, Perth, Australia. http://researchrepository.murdoch.edu.au/1662/ Chuwen, B.M., Platell, M.E., Potter, I.C., 2007. Dietary compositions of the sparid Acanthopagrus butcheri in three normally closed and variably hypersaline estuaries differ markedly. Environmental Biology of Fishes 80, 363-376. Chuwen, B.M., Potter, I.C., Hall, N.G., Hoeksema, S.D., Laurensen, L.J.B., 2011. Changes in catch rates and length and age at maturity, but not growth, of an estuarine plotosid (Cnidoglanis macrocephalus) after heavy fishing. Fisheries Bulletin 109, 247-260. Claridge, P.N., Potter, I.C., Hardisty, M.W., 1986. Seasonal changes in movements, abundance, size composition and diversity of the fish fauna of the Severn Estuary. Journal of the Marine Biological Association of the United Kingdom 66, 229-258. Clarke, K.R., Warwick, R.M., 2001. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation. PRIMER-E, Plymouth, UK, pp. 38. Cloern, J.E., Foster, S.Q., Kleckner, A.E., 2014. Phytoplankton primary production in the world's estuarine-coastal ecosystems. Biogeosciences 11, 2477-2501. Cooper, A., Wright, I., Mason, T., 2008. Geomorphology and sedimentology. In: Allanson, B.R., Baird, D. (eds), Estuaries of South Africa. New York, Cambridge University Press, pp. 5-25.
133
Cooper, J.A.G., 2001. Geomorphological variability among microtidal estuaries from the wave-dominated South African coast. Geomorphology. 40, 99-122. Costa, M.J., Cabral, H.N., Drake, P., Economou, A.N., Fernandez-Delgado, C., Gordo, L., Marchand, J., Thiel, R., 2002. Recruitment and production of commercial species in estuaries. In: Elliott, M., Henderson, P.A. (eds), Fishes in Estuaries. Oxford, UK. Blackwell Science Ltd, pp. 54-123. Coutin, P., Walker, S., Morison, A., 1997. Black bream – 1996. Compiled by the Bay & Inlet Fisheries and Stock Assessment Group. Fisheries Victoria Assessment Report No. 14. Fisheries, Victoria. Což-Rakovac, R., Strunjak-Perović, I., Topić Popović, N., Hacmanjek, M., Šimpraga, B., Teskeredžić, E., 2002. Health status of wild and cultured sea bass in the northern Adriadic Sea. Veterinary Medicine.-Czech 47, 222-226. Craig, J.K., 2012. Aggregation on the edge: effects of hypoxia avoidance on the spatial distribution of brown shrimp and demersal fishes in the Northern Gulf of Mexico. Marine Ecology Progress Series 445, 75-95. Craig, J.K., Crowder, L.B., Gray, C.D., McDaniel, C.J., Henwood, T.A., Hanifen, J.G., 2001. Ecological effects of hypoxia on fish, sea turtles, and marine mammals in the northwestern Gulf of Mexico. In: Rabalais, N.N., Turner, R.E. (eds), Coastal Hypoxia: Consequences for Living Resources and Ecosystems. American Geophysical Union, Washington, DC, pp. 269-291. Cronin, L.E., Mansueti, A.J., 1971. The biology of the estuary. In: Douglas, P.A., Stroud, R.H. (eds), A symposium on the biological significance of estuaries, Washington, D.C., Sport Fishing Institute, pp. 14-39. Darnaude, A.M., 2005. Fish ecology and terrestrial carbon use in coastal areas: Implications for marine fish production. Journal of Animal Ecology 74, 864–876. Day, J.H., 1980. What is an estuary? Southern Africa Journal of Science 76, 198. Day, J.H., 1981. The nature, origin and classification of estuaries. In: Day, J.H. (ed.), Estuarine Ecology: with particular reference to Southern Africa. Cape Town, Balkema, pp. 147-178. de Jong, V.N., Elliott, M., Orive, E., 2002. Causes, historical development, effects and future challenges of a common environmental problem: eutrophication. Hydrobiologia 475, 1-19.
134
Devine, J.A., Wright, P.J., Pardoe, H.E., Heino, M., 2012. Comparing rates of contemporary evolution in life-history traits for exploited fish stocks. Canadian Journal of Fisheries and Aquatic Sciences 69, 1105-1120. Diaz, R.J., 2000. Overview of hypoxia around the world. Journal of Environmental Quality 30, 275-281. Diaz, R.J., 2002. Hypoxia and anoxia as global phenomena. In: Thurston R.V. (ed.), Fish Physiology, Toxicology, and Water Quality. Proceedings of the sixth international symposium, Athens, Georgia: Ecosystems Research Division, pp. 183-202. Diaz, P.B., Heino, M., 2014. What can selection experiments teach us about fisheries- induced evolution? Biological Journal of the Linnean Society 111, 485-503. Diaz, R.J., Rosenberg, R., 2008. Spreading dead zones and consequences for marine ecosystems. Science 321, 926-929. Dolbeth, M., Martinho, F., Viegas, I., Cabral, H., Pardalet, M.A., 2008. Estuarine production of resident and nursery fish species: conditioning by drought events? Estuarine Coastal Shelf Science 78, 51-60. Douglas, G.B., Hamilton, D.P., Gerritse, R., Adeney, J.A., 1995. An investigation of water quality at two sites in the Swan River Estuary, 1993/94. CSIRO Division of Water Resources Report, Floreat, Western Australia. Eby, L.A., Crowder, L.B., 2002. Hypoxia-based habitat compression in the Neuse River Estuary: context-dependent shifts in behavioural avoidance thresholds. Canadian Journal of Fisheries and Aquatic Sciences 59, 952-965. Eby, L.A., Crowder, L.B., McClellan, C.B., Powers, M.J., Peterson, C.H., 2005. Habitat degradation from intermittent hypoxia: impacts on juvenile fishes. Marine Ecological Progress Series 291, 249-262. Einum, S., Fleming, I.A., 2001. Implications of stocking: ecological interactions between wild and released salmonids. Nordic Journal of Freshwater Research 75, 56-70. Elliott, M., Hemingway, K. L. (eds), 2002. Fishes in Estuaries. Blackwell Science, Oxford, pp. 636. Elliott, M., McLusky, D. S., 2002. The need for definitions in understanding estuaries. Estuarine, Coastal and Shelf Science 55, 815-827. Elliott, M., Whitfield, A.K., 2011. Challenging paradigms in estuarine ecology and management. Estuarine, Coastal and Shelf Science 94, 306-314.
135
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, 241-268. Fabens, A.J., 1965. Properties and fitting of the von Bertalanffy growth curve. Growth 29, 265-289. Feng, J., Li, J., Li, Y., 2010. A monsoon-like southwest Australian circulation and its relation with rainfall in southwest Western Australia. Journal of Climate 23, 1334- 1353. Ferguson, G., Q. Ye, 2008. Black Bream (Acanthopagrus butcheri). Stock Assessment Report for PIRSA Fisheries. South Australian Research and Development Institute (Aquatic Sciences), Adelaide, F2008/000810-1. SARDI Research Report Series No. 310. Fleming, I.A., Petersson, E., 2001. The ability of released, hatchery salmonids to breed and contribute to the natural productivity of wild populations. Nordic Journal of Freshwater Research 75, 71-98. Franco, A, Elliott, M., Franzoi, P., Torricelli, P., 2008. Life strategies of fishes in European estuaries: the functional guild approach. Marine Ecology Progress Series 354, 219- 228. Friedrich, J., Janssen, F., Aleynik, D., Bange, H.W., Boltacheva, N., Çagatay, M.N., Dale, A.W., Etiope, G., Erdem, Z., Geraga, M., Gilli, A., Gomoiu, M.T., Hall, P.O.J., Hansson, D., He, Y., Holtappels, M., Kirf, M.K., Kononets, M., Konovalov, S., Lichtschlag, A., Livingstone, D.M., Marinaro, G., Mazlumyan, S., Naeher, S., North, R.P., Papatheodorou, G., Pfannkuche, O., Prien, R., Rehder, G., Schubert, C.J., Soltwedel, T., Sommer, S., Stahl, H., Stanev, E.V., Teaca, A., Tengberg, A., Waldmann, C., Wehrli, B., Wenzhöfer, F., 2014. Investigating hypoxia in aquatic environments: diverse approaches to addressing a complex phenomenon. Biogeosciences 11, 1125-1259. Froese, R., 2006. Cube law, condition factor and weight-length relationships: history, meta-analysis and recommendations. Journal of Applied Ichthyology 22, 241-253. Froese, R., Binohlan, C., 2000. Empirical relationships to estimate asymptotic length, length at first maturity and length at maximum yield per recruit in fishes, with a simple method to evaluate length frequency data. Journal of Fish Biology 56, 758–773.
136
Furnas, M. J., 1990. Insitu growth rates of marine phytoplankton – approaches to measurement, community and species growth rates, Journal of Plankton Research 12, 1117-1151. Gardner, M.J., Cottingham, A., Chaplin, J.A., Hesp, S.A., Jenkins, G.I., Phillips, N.M., Potter, I.C., 2013. Comparisons of the biological and genetic characteristics of restocked and wild individuals of a sparid (Acanthopagrus butcheri) in an estuary. Reviews in Fisheries Science 21, 441-452. Gaston, G.R., Rakocinski, C.F., Brown, S.S., Cleveland, C.M., 1998. Trophic function in estuaries: response of macrobenthos to natural and contaminant gradients. Marine and Freshwater Research 49, 833-846. Gilberto, D.A., Bremac, C.S., Acha, E.M., Mianzan, H., 2004. Large scale spatial patterns of benthic assemblages in the SW Atlantic: the Rio de la Plata estuary and adjacent shelf waters. Estuarine, Coastal and Shelf Science 61, 1-13. Gill, H.S., Potter, I.C., 1993. Spatial segregation amongst goby species within an Australian estuary, with a comparison of the diets and salinity tolerance of the two most abundant species. Marine Biology 117, 515-526. Gillanders, B.M., Kingsford, M.J., 2002. Impact of changes in flow of freshwater on estuarine and open coastal habitats and the associated organisms. Oceanography and Marine Biology 40, 233-309. Gillson, J., 2011. Freshwater flow and fisheries production in estuarine and coastal systems: where a drop of rain is not lost. Reviews in Fisheries Science 19, 168-186. Glover, D.C., Devries, D.R., Wright, R.A., 2013. Growth of largemouth bass in a dynamic estuarine environment: an evaluation of the relative effects of salinity, diet, and temperature. Canadian Journal of Fisheries and Aquatic Sciences 70, 485-501. Gomez, E.D., Mingoa-Licuanan, S.S., 2006. Achievements and lessons learned in restocking giant clams in the Philippines. Fisheries Research 80, 46-52. Gomon, M., Bray, D., Kuiter, R., 2008. Fishes of Australia’s Southern Coast. Sydney, Reed New Holland, pp.928. Griffiths, M.H., Wilke, C., Penney, A.J., Melo, Y., 2002. Life history of white stumpnose Rhabdosargus globiceps (Pisces: Sparidae) off South Africa. South African Journal of Marine Science 24, 281-300. Grover, M.C., 2005. Changes in size and age at maturity in a population of kokanee Oncorhynchus nerka during a period of declining growth conditions. Journal of Fish Biology 66, 122-134. 137
Haddon, M., 2011. Modelling and Quantitative Methods in Fisheries. Second Edition. Boca Raton, CRC Press, pp. 449. Haedrich, R.L., 1983. Estuarine fishes. In: Ketchum, B.H. (ed.), Ecosystems of the world, Vol. 22, Estuaries and enclosed seas. Amsterdam, Elsevier, pp. 183-207. Hallegraeff, G., Mooney, B., Evans, K., 2010. What triggers fish-killing Karlodinium veneficum dinoflagellate blooms in the Swan Canning River system? Swan Canning Research and Innovation Program Final Report Swan River Trust, Perth, Western Australia, pp. 31. Hallett, C.S., Valesini, F.J., Clarke, K.R., Hoeksema, S.D., 2015. Effects of a harmful algal bloom on the community ecology, movements and spatial distributions of fishes in a microtidal estuary. Hydrobiologia. http://doi.org/10.1007/s10750-015-2383-1 Hamilton, D.P., Turner, J.V., 2001. Integrating research and management for an urban estuarine system: the Swan-Canning Estuary, Western Australia. Hydrological Processes 15, 2383-2385. Hard, J.J., Gross, M.R., Heino, M., Hilborn, R., Kope, R.G., Law, R., Reynolds, J.D., 2008. Evolutionary consequences of fishing and their implications for salmon. Evolutionary Applications 1, 388-408. Hassell, K. L., Coutin, P. C., Nugegoda, D., 2008a. Hypoxia, low salinity and lowered temperature reduce embryo survival and hatch rates in black bream Acanthopagrus butcheri (Munro, 1949). Journal of Fish Biology 72, 1623-1636. Hassell, K.L., Coutin, P.C., Nugegoda, D., 2008b. Hypoxia impairs embryo development and survival in black bream (Acanthopagrus butcheri). Marine Pollution Bulletin 57, 302-306. He, J.X., Bence, J.R., 2007. Modeling annual growth variation using a hierarchical Bayesian approach and the von Bertalanffy growth function, with application to Lake Trout in Southern Lake Huron. Transactions of the American Fisheries Society 136, 328-330. Hedgpeth, J.W., 1982. Estuarine dependence and colonisation. Atlantica 5, 57-58. Helle, J.H., Hoffman, M.S., 1998. Changes in size and age at maturity of two North American stocks of chum salmon (Oncorhynchus ketal) before and after a major regime shift in the North Pacific Ocean. North Pacific Anadromous Fish Commission Bulletin 1, 81-89.
138
Hesp, S.A., Hall, N.G., Potter, I.C., 2004. Size-related movements of Rhabdosargus sarba in three different environments and their influence on estimates of von Bertalanffy growth parameters. Marine Biology 144, 449-462. Hilborn, R., 1998. The economic performance of marine stock enhancement projects. Bulletin of Marine Science 62, 661-674. Hilborn, R., Walters, C.J., 1992. Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty. New York, Chapman and Hall, pp. 570. 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. Hobday, D., Moran, M., 1983. Age, growth and fluctuating year-class strength of black bream (Acanthopagrus butcheri) and rainbow trout (Oncorhynchus mykiss). Comparative Biochemistry and Physiology Part A 125, 85-94. 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, 183-201. Hoeksema, S.D., Chuwen, B.M., Hesp, S.A., Hall, N.G., Potter, I.C., 2006. Impact of environmental changes on the fish faunas of Western Australian south-coast estuaries. Project No. 2002/017, Fisheries Research and Development Corporation, pp. 190. 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. Hoellein, T.J., Bruesewitz, D., Richardson, D.C., 2013. Revisiting Odum (1956): A synthesis of aquatic ecosystem metabolism. Limnology and Oceanography 58, 2089- 2100. Hopkinson, C.S., Smith, E.M., 2005. Estuarine respiration: an overview of benthic, pelagic and whole system respiration. In: del Giorgio, P. A., Williams, P.J.L. (eds), Respiration in Aquatic Ecosystems. New York, Oxford University Press, pp. 122-146. Howell, P., Simpson, D., 1994. Abundance of marine resources in relation to dissolved oxygen in Long Island Sound. Estuaries 17, 394-402. Hutchins, B., Thompson, M., 2001. The Marine and Estuarine Fishes of South-western Australia. A Field Guide for Anglers and Divers. Perth, Western Australia, Western Australia Museum, pp.103.
139
Intergovernmental Committee on Surveying and Mapping (ICSM), 2014. Australian Tides Manual. Permanent Committee on Tides and Mean Sea Level, Special Publication 9. pp. 43. Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., Bradbury, R.H., Cooke, R.G., 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, 629-638. Jenkins, G.P., Conron, S.D., Morison, A.K., 2010. Highly variable recruitment in an estuarine fish is determined by salinity stratification and freshwater flow: implications of a changing climate. Marine Ecological Progress Series 417, 246-261. Jenkins, G.P., Spooner, D., Conron, S., Morrongiello, J., 2015. Differing importance of salinity stratification and freshwater flow for the recruitment of apex species of estuarine fish. Marine Ecology Progress Series 523, 125-144. Jennings, S., Kaiser, M., Reynolds, J.D., 2009. Marine Fisheries Ecology. UK, John Wiley & Sons, pp. 432. Jonsson, B., Jonsson, N., 2006. Cultured Atlantic salmon in nature: a review of their ecology and interaction with wild fish. ICES Journal of Marine Science 63, 1162- 1181. Kailola, P. J., Williams, M. J., Stewart, P. C., Reichelt, R. E., McNee, A., Grieve, C., 1993. Australian Fisheries Resources. Canberra: Bureau of Resources Science, Australia and Fisheries Research and Development Corporation, pp. 422. Kanandjembo, A.N., Potter, I.C., Platell, M.E., 2001. Abrupt shifts in fish community on the hydrologically variable upper estuary on the Swan River. Hydrological Processes 15, 2503-2517. Kellison, G.T., Eggleston, D.B., Burke, J.S., 2000. Comparative behaviour and survival of hatchery-reared wild summer flounder (Paralichthys dentatus). Canadian Journal of Aquatic Sciences 57, 1870-1877. Kemp, W.M, Boynton, W.R., Adolf, J.E., Boesch, D.F. Boicourt, W.C., Brush, G., Cornwell, J.C., Fisher, T.R., Glibert, P.M., Harding, L.W., Houde, E.D., Kimmel, D.G., Miller, W.D., Newell, R.I.E., Roman, M.R., Smith E.M., Stevenson, J.C., 2005. Eutrophication of Chesapeake Bay: Historical trends and ecological interactions. Marine Ecology Progress Series 303, 1-29.
140
Kennish, M.J., 2002. Environmental threats and environmental future of estuaries. Environmental Conservation 1, 78-107. Kennish, M.J. (ed.), 2003. Estuarine Research, Monitoring and Resource Protection. New York, CRC Press, pp. 297. Kennish, M.J., Townsend, A.R., 2007. Nutrient enrichment and estuarine eutrophication. Ecological Applications 17, Supplement: S1-S2. Kitada, S., Kishino, H., 2006. Lessons learned from Japanese marine finfish stock enhancement programmes. Fisheries Research 80, 101-112. Kramer, D.L., 1987. Dissolved oxygen and fish behavior. Environmental Biology of Fishes 18, 81-92. Kristiana, R., Antenucci, J.P., Imberger, J., 2012. Using a multi-component indicator toward reducing phytoplankton bloom occurrences in the Swan River Estuary. Environmental Management 50, 237-256. Kurup, G.R., Hamilton, D.P., Patterson, J C., 1998. Modelling the effect of seasonal flow variations on the position of salt wedge in a micro-tidal estuary. Estuarine, Coastal and Shelf Science 47, 191-208. Laevastu, T., 1965. Manual of Methods in Fisheries Biology. Rome: FAO. Lehodey, P., Alheit, J., Barange, M., Baumgartner, T., Beaugrand, G., Drinkwater, K., Fromentin, J.-M., Hare, S.R., Ottersen, G., Perry, R.I., Roy, C., van der Lingen, C.D., Werner, F., 2006. Climate variability, fish, and fisheries. Journal of Climate 19, 5009- 5030. Lenanton, R.C.J., Ayvazian, S.G., Dibden, C.J., Jenkins, G., Sarre, G., 1999. The use of stock enhancement to improve the catch rates of black bream, Acanthopagrus butcheri (Munro) for Western Australian recreational fishers. In: Howell, B.R., Moksness, E., Svåsand, T. (eds), Stock Enhancement and Sea Ranching. Oxford, UK, Fishing New Books, pp. 219-230. Lenanton, R.C.J., Potter, I.C., 1987. Contribution of estuaries to commercial fisheries in temperate Western Australia and the concept of estuarine dependence. Estuaries 10, 28-35. Le Vay, L., Carvalho, G.R., Quinitio, E.T., Lebata, J.H., Ut, V.N., Fushimi, H., 2007. Quality of hatchery-reared juveniles for marine fisheries stock enhancement. Aquaculture 268, 169-180. Linke, T.E., 2011. Trophic interactions among abundant members of the fish fauna in a permanently-open and seasonally-open estuary in south-western Australia. Ph.D. 141
thesis. Murdoch University, Perth, Australia. http://researchrepository.murdoch.edu.au-12662/. Loneragan, N.R., Jenkins, G.I., Taylor, M.D., 2013. Marine Stock Enhancement, Restocking, and Sea Ranching in Australia: Future Directions and a Synthesis of Two Decades of Research and Development. Reviews in Fisheries Science 21, 222-236. Loneragan, N.R., Potter, I.C., 1990. Factors influencing the community structure and distribution of different life-cycle categories of fishes within the shallow waters of a large Australian estuary. Marine Biology 106, 25-37. Loneragan, N.R., Potter, I.C., Lenanton, R.C.J., Caputi, N., 1987. Influence of environmental variables on the fish fauna of the deeper waters of a large Australian estuary. Marine Biology 94, 631-641. Lorenzen, K., Beveridge, M.C.M., Mangel, M., 2012. Cultured fish: integrative biology and management of domestication and interactions with wild fish. Biological Reviews of the Cambridge Philosophical Society 87, 639-60. 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, 49-54. Lotze, H.K., Lenihan, H.S., Bourque, B.J., Bradbury, R.H., Cooke, R.G., Kay, M.C., Kidwell, S.M., Kirby, M.X., Peterson, C.H., Jackson, J.B.C., 2006. Depletion, Degradation, and Recovery Potential of Estuaries and Coastal Seas. Science 312, 1806-1809. Luettich, Jr. R.A., Carr, S.D., Reynolds-fleming, J.V, Fulcher, C.W., & Mcninch, J.E., 2002. Semi-diurnal seiching in a shallow, micro-tidal lagoonal estuary. Continental Shelf Research 22, 1669-1681. Lunn, D.J., Thomas, A., Best, N., Spiegelhalter, D., 2000. WinBUGS - a Bayesian modelling framework: concepts, structure, and extensibility. Statistics and Computing 10, 325-337. McComb, A.J. (ed.), 1995. Eutrophic shallow estuaries and lagoons. Boca Raton: CRC Press, pp. 240. McComb, A.J., Atkins, R.P., Birch, P.B., Gordon, D.M., Lukatelich, R.J., 1981 Eutrophication in the Peel-Harvey Estuarine System, Western Australia. In Estuaries and Nutrients. Neilson, B.J., Cronin, L.E. (eds) The Humana Press. Clifton New Jersey, pp. 323-342.
142
McComb, A.J., Davis, J.A., 1993. Eutrophic waters of southwestern Australia. Fertilizer Research 36, 105-114. McDowall, R.M., 1988. Diadromy in Fishes: Migration between Freshwater and Marine Environments. University Press, Cambridge, pp. 308. McGinnity, P., Prodöhl, P., Ferguson, A., Hynes, R., Ó Maoiléidigh, N., Baker, N., Cotter, D., O’Hea, B., Cooke, D., Rogan, G., Taggart, J., Cross, T., 2003. Fitness reduction and potential extinction of wild populations of Atlantic salmon Salmo salar, as a result of interactions with escaped farm salmon. Proceedings of the Royal Society of London Series B 270, 2443-2450. McLusky, D.S., Bryant, D.M., Elliott, M., 1992. The impact of land-claim on macrobenthos, fish and shorebirds on the Forth estuary, Eastern Scotland. Aquatic Conservation 2, 211-222. McLusky, D.S., Elliott, M., 2004. The Estuarine Ecosystem: Ecology, Threats and Management. Oxford, Oxford University Press, pp. 214. Middleburg, J.J., Levin, L.A., 2009. Coastal hypoxia and sediment biogeochemistry. Biogeosciences 6, 1273-1293. Mohan, J., Rahman, Md. S, Thomas, P., Walther, B., 2014. Influence of constant and periodic experimental hypoxic stress on Atlantic croaker otolith chemistry. Aquatic Biology 20, 1-11. Monbet, Y., 1992. Control of phytoplankton biomass in estuaries: a comparative analysis of microtidal and macrotidal estuaries. Estuaries 15, 536-571. Mooney, B.D., De Salas, M., Hallegraeff, G.M., Place, A.R., 2009. Survey for karlotoxin production in 15 species of gymnodinioid dinoflagellates (Kareniaceae, Dinophyta). Journal of Phycology 45, 164-175. Morgan, M.J., Colbourne, E.B., 1999. Variation in maturity-at-age and size in three populations of American plaice. ICES Journal of Marine Science 56, 673-688. Morison, A.K., Coutin, P.C., Robertson, S.G., 1998. Age determination of black bream, Acanthopagrus butcheri (Sparidae), from the Gippsland lakes of south-eastern Australia indicates slow growth and episodic recruitment. Marine and Freshwater Research 49, 491-498. Morita, K, Fukuwaka, M., 2007. Why age and size at maturity have changed in Pacific salmon. Marine Ecological Progress Series 335, 289-294.
143
Morita, K., Morita, S.H., 2002. Rule of age and size at maturity: individual variation in the maturation history of resident white-spotted charr. Journal of Fish Biology 61, 1230- 1238. Morrongiello, J.R., Walsh, C.T., Gray, C.A, Stocks, J.R., Crook, D.A., 2014. Environmental change drives long-term recruitment and growth variation in an estuarine fish. Global Change Biology 20, 1844-60. Nakagawa, M., Okouchi, H., Hattori, K., 2007. Otolith-marking experiments of juvenile black rockfish Sebastes schlegeli using alizarin complexone. Aquaculture Science 55, 253-257. National Water Commission, 2007. Australian Water Resources 2005; A baseline assessment of water resources for the National Water Initiative, Level 2 Assessment. Water Availability Theme National Perspective. http://www.water.gov.au/Publications/WaterAvailability_National_Perspective_Repor t_May07.pdf Accessed July 2015. Neuheimer, A.B., Thresher, R.E., Lyle, J.M., Semmens, J.M., 2011. Tolerance limit for fish growth exceeded by warming waters. Nature Climate Change 1, 110-113. Nicholson, A.D., Jenkins, G., Sherwood, J., Longmore, A., 2008. Physical environmental conditions, spawning and early-life stages of an estuarine fish: climate change implications for recruitment in intermittently open estuaries. Marine and Freshwater Research 59, 735-749. Norlem, M., Paraska, D., Hipsey, M.R., 2013. Sediment-water oxygen and nutrient fluxes in a hypoxic estuary. In; Piantadosi, J., Anderssen, R.S., Boland, J. (eds), MODSIM2013 - 20th International congress on modelling and simulation. Modelling and Simulation Society of Australia and New Zealand, December 2013, Adelaide, Australia. Norriss, J.V., Tregonning, J.E., Lenanton, R.C.J., Sarre, G.A., 2002. Biological synopsis of the black bream Acanthopagrus butcheri (Munro) (Teleostei: Sparidae). Fisheries Research Report No. 93. Department of Fisheries, Perth. Nostvik, F., Pedersen, T., 1999. Movement patterns and growth of wild cod (Gadus morhua L.) and hatchery-reared cod released as 1-group. In: Howell, R.R., Moksness, E., Svasand, T. (eds). Stock Enhancement and Sea Ranching. Oxford, UK. Fishing News Books, pp. 315-333.
144
Orr, P.T., Jones, G.J., Douglas, G.B., 2004. Response of cultured Microcystis aeruginosa from the Swan River, Australia, to elevated salt concentration and consequences for bloom and toxin management in estuaries. Marine and Freshwater Research 55, 277- 283. Oze, G., Oze, R., Anunuso, C., Ogukwe, C., Nwanjo, H., Okorie, K., 2005. Heavy Metal Pollution Of Fish Of Qua-Iboe River Estuary: Possible Implications For Neurotoxicity. The Internet Journal of Toxicology 3, 1-55. Paerl, H.W., Hall, N.S., Peierls, B.L., Rossignol, K.L., 2014. Evolving Paradigms and Challenges in Estuarine and Coastal Eutrophication Dynamics in a Culturally and Climatically Stressed World. Estuaries and Coasts, 37, 243-258. Partridge, G.J., Sarre, G.A., Hall, N.G., Jenkins, G.I., Chaplin, J., Potter, I.C, 2004. Comparisons between the growth of Acanthopagrus butcheri cultured from broodstock from two estuarine populations that are reproductively isolated and differ markedly in growth rate. Aquaculture 231, 51-81. Peig, J., Green, A.J., 2009. New perspectives for estimating body condition from mass/length data: the scaled mass index as an alternative method. Oikos 118, 1883- 1891. Perillo, G.M.E., 1996. Definitions and geomorphological classifications of estuaries. In: Perillo, G.M.E. (ed.), 1995, Geomorphology and Sedimentology of Estuaries. Developments in Sedimentology 53, 17-47. Pichavant, K., Person-Le-Ruyet, J., Le Bayon, N., Sévère, A., Le Roux, A., Quéméner, L., Maxime, V., Nonnotte, G., Boeuf, G., 2000. Effects of hypoxia on growth and metabolism of juvenile turbot. Aquaculture 188, 103-114. Pichavant, K., Person-Le-Ruyet, J., Le Bayon, N., Sévère, A., Le Roux, A., Quéméner, L., Maxime, V., Nonnotte, G., Boeuf, G., 2001. Comparative effects of long-term hypoxia on growth, feeding, and oxygen consumption in juvenile turbot and European sea bass. Journal Fish Physiology 59, 875-883. Pihl, L., Baden, S.P., Diaz, R.J., 1991. Effects of periodic hypoxia on distribution of demersal fish and crustaceans. Marine Biology 108, 349-360. Pihl, L., Baden, S.P., Diaz, R.J., Schaffher, L.C., 1992. Hypoxia-induced structural changes in the diet of bottom-feeding fish and crustacean. Marine Biology 112, 349- 361.
145
Plavan, A.A., Passadore C., Gimenez, L., 2010. Fish assemblage in a temperate estuary on the Uruguay coast: seasonal variation and environmental influence, Brazilian Journal of Oceanography 58, 299-314. Potter, I.C., Beckley, L.E., Whitfield, A.K., Lenanton, R.C.J., 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., Chuwen, B.M., Hoeksema, S.D., Elliott, M., 2010. The concept of an estuary: A definition that incorporates systems which can become closed to the ocean and hypersaline. Estuarine, Coastal and Shelf Science 87, 497-500. 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, 395- 421. Potter, I.C., Hyndes, G.A., Baronie, F.M., 1993. The fish fauna of a seasonally closed Australian estuary. Is the prevalence of estuarine-spawning species high? Marine Biology 116, 19-30. Potter, I.C., French, D.J.W., Jenkins, G.I., Hesp, S.A., Hall, N.G., DeLestang, S., 2008. Comparisons of the growth and gonadal development of otolith-stained, cultured black bream, Acanthopagrus butcheri, in an estuary. Reviews in Fisheries Science 16, 303- 316. Potter, I.C., Tweedley, J.R., Elliott, M., Whitfield, A.K., 2013. The ways in which fish use estuaries: a refinement and expansion of the guild approach. Fish and Fisheries 16, 230-239. Powers, S.P., Peterson, C.H., Christian, R.R., Sullivan, E., Powers, M.J., Bishop, M., Buzzelli, C.P., 2005. Effects of eutrophication on bottom habitat and prey resources of demersal fishes. Marine Ecological Progress Series 302, 233-243. Pritchard, R.W., 1967. What is an estuary? Physical viewpoint. In: Lauff, G.H. (ed.), Estuaries. Washington D.C., American Association for the Advancement of Science, pp. 3-5. Prince, E.D., Goodyear, C.P., 2009. Hypoxia-based habitat compression of tropical pelagic fishes. Fisheries Oceanography 15, 451-464. Prince, E.D., Goodyear, P., 2006. Hypoxia-based habitat compression of tropical pelagic fishes. Fisheries Oceanography 15, 451-464.
146
Prince, J.D., Potter, I.C., Lenanton, R.C.L., Loneragan, N.R., 1982. Segregation and feeding of atherinids in south-western Australian estuaries. Australian Journal of Marine and Freshwater Research 33, 865-880. Prior, S.P., Beckley, L.E., 2007. Characteristics of recreational anglers in the Blackwood Estuary, a popular tourist destination in southwestern Australia. Tourism in Marine Environments 4, 15-28. Quinn, T.J., Deriso, R.B., 1999. Quantitative Fish Dynamics. Oxford University Press, pp. 542. R Core Team, 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/ Accessed February 2014. Rakocinski, C.F., Brown, S.S., Gaston, G.R., Heard, R.W., Walker, W.W., Summers, J.K., 1997. Macrobenthic responses to natural and contaminant-related gradients in northern Gulf of Mexico estuaries. Ecological Applications 7, 1278-1298. Reynolds-Fleming, J.V., Luettich, R.A. Jr., 2004. Wind-driven lateral variability in a partially mixed estuary. Estuarine, Coastal and Shelf Science 60, 395-407. Ricker, W.E., 1979. Growth rates and models. In Fish physiology Vol. VIII, Academic Press, pp. 677-743. Roberts, J.J., Brandt, S.B., Fanslow, D., Ludsin, S.A., Potnoven, S.A., Scavia, D., Höök, T.O., 2011. Effects of hypoxia on consumption, growth, and RNA:DNA ratios of young yellow perch. Transactions of the American Fisheries Society 140, 37-41. Robins, J., Mayer, D., Staunton-Smith, J., Halliday, I., Sawynok, B., Sellin M., 2006. Variable growth rates of the tropical estuarine fish barramundi Lates calcarifer (Bloch) under different freshwater flow conditions. Journal of Fish Biology 69, 379- 391. Robson, B.J., Bukaveckas, P.A., Hamilton, D.P., 2008. Modelling and mass balance assessments of nutrient retention in a seasonally-flowing estuary (Swan River Estuary, Western Australia). Estuarine, Coastal and Shelf Science 76, 282-292. Robson, B.J., Hamilton, D.P., 2003. Summer flow event induces a cyanobacterial bloom in a seasonal Western Australian estuary. Marine and Freshwater Research 54, 139-151. Sakabe, R., Lyle, J.M., 2010. The influence of tidal cycles and freshwater inflow on the distribution and movement of an estuarine resident fish Acanthopagrus butcheri. Journal of Fish Biology 77, 643-660.
147
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, 103- 122. Sarre, G.A., Potter, I.C., 1999. Comparisons between the reproductive biology of the black bream Acanthopagrus butcheri (Teleosti: Sparidae) in four estuaries with widely differing characteristics. International Journal of Salt Lake Research 8, 179-210. Sarre, G.A., Potter, I.C., 2000. The age composition and growth rates of Acanthopagrus butcheri (Sparidae) vary among estuaries: some possible contributing factors. Fisheries Bulletin 98, 785-799. Sato, S., Watanabe, N., Miyauchi, Y., Chiba, T., Iida, M., Okamoto, Y., Ohnuki, T., 2011. Release of otolith marked salmon from Japan in fall of 2010 and spring of 2011. NPAFC Doc. 1347, pp. 11. Savenije, H.H.G., 2012. Salinity and Tides in Alluvial Estuaries, completely revised 2nd edition. www.salinityandtides.com Accessed 13/7/2013. Schelske, C.L., Odum, E.P., 1962. Mechanisms maintaining high productivity in Georgia estuaries. Proceedings of the Gulf and Caribbean Fisheries Institute 14, 75-80. Schumann, E.H., 2013. Sea level variability in South African estuaries. South African Journal of Science 109, 1-7. Sharpe, D.M.T., Hendry, A.P., 2009. Life history change in commercially exploited fish stocks: an analysis of trends across studies. Evolutionary Applications 2, 260-275. Sim-Smith, C.J., Jeffs, A.G., Radford, C.A., 2012. Variation in the growth of larval and early juvenile snapper, Chrysophrys auratus (Sparidae). Marine and Freshwater Research 63, 1231-1243. Smith, K.A., 2006. Review of fishery resources and status of key fishery stocks in the Swan-Canning Estuary. Fisheries Research Report No. 156, Department of Fisheries, Western Australia, 84 pp. Smith, S.A., Hartman, K.H., Pasnik, D.J., 2003. New and old diseases of cultured perch, Perca flavescens. GAA (Global Aquaculture Alliance) Advocate 6, 10-12. Smith, K., Quinn, A., Holtz, M., Nardi, K., 2014. West coast nearshore and estuarine finfish resources status report. In: Fletcher, W.J., Santoro, K. (eds), Status Reports of the Fisheries and Aquatic Resources of Western Australia 2013/143: The State of the Fisheries. Department of Fisheries, Western Australia, pp. 67-82.
148
Smith, K.A., Norriss, J., 2011. The status of the black bream Acanthopagrus butcheri (Pisces : Sparidae ) population in Lake Clifton , south-western Australia, Journal of the Royal Society of Western Australia 94, 25-28. Smith, K., Norriss, J., Brown, J., 2009. Population growth and mass mortality of an estuarine fish, Acanthopagrus butcheri, unlawfully introduced into an inland lake. Aquatic Conservation Marine and Freshwater Ecosystems, 19, 4-13. Sobczak, W.V., Cloern, J.E., Jassby, A.D., Cole, B.E., Schraga, T.S., Arnsberg, A., 2005. Detritus fuels ecosystem metabolism but not metazoan food webs in San Francisco estuary’s freshwater Delta. Estuaries 28, 124-137. Stahl, J.P., Kruse, G.H., 2008. Spatial and temporal variability in size at maturity of walleye pollock in the eastern Bering Sea. Transactions of the American Fisheries Society 137, 1543-1557. Stearns, S.C., 1992. The Evolution of Life Histories. Oxford, Oxford University Press, pp. 249. Steckis, R.A., Potter, I.C., Lenanton, R.C.J., 1995. The commercial fisheries in three southwestern Australian estuaries exposed to different degrees of eutrophication. In: McComb (ed.), Eutrophic shallow estuaries and lagoons. Boca Raton: CRC Press, pp. 189-203. Stephens, R., Imberger, J., 1996. Dynamics of the Swan River Estuary: the seasonal variability. Marine and Freshwater Research 47, 517-529. Stramma, L., Prince, E.D., Schmidtko, S., Luo, J., Hoolihan, J.P., Visbeck, M., Wallace, D.W.R., Brandt, P., Kӧrtzinger, A., 2012. Expansion of oxygen minimum zones may reduce available habitat for tropical pelagic fishes. Nature Climate Change 2, 33-37. Strøttup, J.G., Overton, J.L., Paulsen, H., Möllmann, C., Tomkiewicz, J., Pedersen, P.B., Lauesen, P., 2008. Rationale for restocking the eastern Baltic cod stock. Reviews in Fisheries Science 16, 58-64. Surhone, L.M., Tennoe, M.T., Henssonow, S.F., 2010. OpenBUGS. Mauritius, Betascript Publishing, pp. 108. Swan River Trust, 2009. Swan Canning water quality improvement plan. Swan River Trust, Government of Western Australia, 128 pp. http://www.swanrivertrust.wa.gov.au/science/river/Documents/ swan_canning_water_quality_improvement_plan_final_dec_2009. pdf.
149
Swan River Trust, 2014. Monitoring and Evaluation. Swan River Vertical Plots. http://www.swanrivertrust.wa.gov.au/swan-river-trust/publications/monitoring-and- evaluation accessed November 2014 Szalai, E.B., Fleischer, G.W., Bence, J.R., 2003. Modeling time-varying growth using a generalized von Bertalanffy model with application to bloater (Coregonus hoyi) growth dynamics in Lake Michigan. Canadian Journal of Fisheries and Aquatic Sciences 60, 55-66. Tagliapietra, D., Sigovini, M., Ghirardini, A.V., 2009. A review of terms and defini- tions to categorise estuaries, lagoons and associated environments. Marine and Freshwater Research 60, 497-509. Taylor, M.D., Palmer, P.J., Fielder, D.S., Suthers, I.M., 2005. Responsible estuarine finfish stock enhancement : an Australian perspective. Journal of Fish Biology 67, 299-331. Tesch, F.W., 1968. Age and growth. In: Ricker W.E. (ed.), Methods for assessment of fish production in fresh waters. Oxford, Blackwell Scientific Publications, pp. 93-123. Thiel, R., Sepulveda, A., Kafemann, R., Nellen, W., 1995. Environmental factors as forces structuring the fish community of the Elbe Estuary. Journal of Fish Biology 46, 47-69. Thomson, C.E., Rose, T., Robb, M., 2001. Seasonal water quality patterns in the Swan River Estuary, 1994–1998, technical report. Swan River Trust, Western Australia. Thronson, A., Quigg, A., 2008. Fifty years of fish kills in coastal Texas. Estuaries and Coasts 31, 802-813. Travis, J., Coleman, F.C., Grimes, C.B., Conover, D., Bert, T.M., Tringali, M., 1998. Critically assessing stock enhancement: an introduction to the Mote Symposium. Bulletin of Marine Science 62, 305-311. Tweedley, J., Bird, D., Potter, I., Gill, H., 2013. Species compositions and ecology of the riverine ichthyofaunas on two Sulawesian islands in the biodiversity hotspot of Wallacea. Journal of Fish Biology 82, 1916-1950. 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. Twomey, L. John, J., 2001. Effects of rainfall and salt-wedge movement on phytoplankton succession in the Swan–Canning Estuary, Western Australia. Hydrological Processes 15, 2655-2669.
150
Tyler, R.M., Targett, T.E., 2007. Juvenile weakfish, Cynoscion regalis, distribution in relation to diel-cycling dissolved oxygen in an estuarine tributary. Marine Ecological Progress Series 333, 257-269. Uncles, R.J., Stephens, J. A., Harris, C., 2006. Runoff and tidal influences on the estuarine turbidity maximum of a highly turbid system: The upper Humber and Ouse Estuary, UK. Marine Geology 235, 213-228. Underwood, G.J.C., Kromkamp, J., 1999. Primary production by phytoplankton and microphytobenthos in estuaries. Advances in Ecological Research 29, 93-153. Vaquer-Sunyer, R., Duarte, C.M., 2008. Thresholds of hypoxia for marine biodiversity. Proceedings of the National Academy of Sciences 105, 15452-15457. Veale, L., 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, 1320-1354. Viney, N.R., Sivapalan, M., 2001. Modelling catchment processes in the Swan-Avon river basin. Hydrological Processes 15, 2671-2685. von Bertalanffy, L., 1938. A quantitative theory of organic growth (inquiries on growth laws II). Human Biology 10, 181-213. Wakefield, C.B., Potter, I.C., Hall, N.G., Lenanton, R.C.J., Hesp, S.A., 2015, Marked variations in reproductive characteristics of snapper (Chrysophrys auratus, Sparidae) and their relationship with temperature over a wide latitudinal range. ICES Journal of Marine Science, doi: 10.1093/icesjms/fsv108. Wannamaker, C.M., Rice, J.A., 2000. Effects of hypoxia on movements and behavior of selected estuarine organisms from the southeastern United States. Journal of Experimental Marine Biology and Ecology 249, 145-163. Ward, R.D., 2006. The importance of identifying spatial population structure in restocking and stock enhancement programmes. Fisheries Research 80, 1-128. Weber, M.J., Brown, M.L., Wahl, D.H., Shoup, E., 2015. Metabolic theory explains latitudinal variation in common carp populations and predicts responses to climate change. Ecosphere 6, 1-16. Weir, L.K., Hutchings, J.A., Fleming, I.A., Einum, S., 2004. Dominance relationships and behavioural correlates of individual spawning success in farmed and wild male Atlantic salmon, Salmo salar. Journal of Animal Ecology 73, 1069-1079.
151
Western Australian Department of Water, 2015. River Monitoring Stations. Government of Western Australia. http://kumina.water.wa.gov.au/waterinformation/wir/reports/publish/616011/g02.htm Whitall, D., Bricker, S., Ferreira, J., Nobre, A.M., Simas, T., Silva, M., 2007. Assessment of eutrophication in estuaries: Pressure – state – response and nitrogen source apportionment. Environmental Assessment 40, 678-690. White, W., Hall, N.G., Potter, I.C., 2002. Size and age compositions and reproductive biology of the nervous shark Carcharhinus cautus in a large subtropical embayment, including an analysis of growth during pre- and postnatal life. Marine Biology 141, 1153-1164. Whitfield, A.K., 1989. The benthic invertebrate community of a southern Cape estuary: structure and possible food sources. Transactions of the Royal Society of South Africa 47, 159-179. Whitfield, A. K., 1998. Biology and Ecology of Fishes in Southern African Estuaries. Grahamstown, J.L.B. Smith Institute of Ichthyology, pp. 223. Whitfield, A.K., 1999. Ichthyofaunal assemblages in estuaries: A South African case study. Reviews in Fish Biology and Fisheries 9, 151-186. Whittaker, R.H., Likens, G.E., 1975. The biosphere and man. In: Leith, H., Whittaker, R.H. (eds), Primary Production of the Biosphere. New York, Springer-Verlag, pp. 305-328 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, 525-538. Williams, J., Hindell, J.S., Swearer, S.E., Jenkins, G.P., 2012. Influence of freshwater flows on the distribution of eggs and larvae of black bream Acanthopagrus butcheri within a drought-affected estuary. Journal of Fish Biology 80, 2281-2301. Williams, J., Jenkins, G.P., Hindell, J.S., Swearer, S.E., 2013. Linking environmental flows with the distribution of black bream Acanthopagrus butcheri eggs, larvae and prey in a drought affected estuary. Journal of Fish Biology 483, 273-287. Wilson, A.B., Ahnesjo, I., Vincent, A.C.J., Meyer, A., 2003. The dynamics of male brooding, mating patterns, and sex roles in pipefishes and seahorses (family Syngnathidae). Evolution 57, 1374-1386.
152
Winkler, G., Dodson, J.J., Bertrand, N., Thivierge, D., Vincent, W.F., 2003. Trophic coupling across the St. Lawrence River estuarine transition zone. Marine Ecological Progress Series 251, 59-73. Wright, W.J., 1997. Tropical-extratropical cloudbands and Australian rainfall: I. climatology. International Journal of Climatology 17, 807-829. Wu, R.S.S., 2002. Hypoxia: From molecular responses to ecosystem responses. Marine Pollution Bulletin 45, 35-45. Wu, R.S.S., Zhou, B.S., Randall, D.J., Woo, N.Y.S., Lam, P.K.S., 2003. Aquatic hypoxia is an endocrine disruptor and impairs fish reproduction. Environmental Science & Technology 37, 1137-1141. Zammit, C., Sivapalan, M., Kelsey, P., Viney, N.R., 2005. Modelling the effects of land- use modifications to control nutrient loads from an agricultural catchment in Western Australia. Ecological Modelling 187, 60-70.
153