Historical abundance and distribution of the native flat oyster (Ostrea angasi) in estuaries of the Great Southern region of Western Australia help to prioritise potential sites for contemporary oyster reef restoration

Worldwide, shellfish reefs underpin a range of ecosystem services, including water filtration, nutrient cycling and the provision of invertebrate and fish habitat (Breitburg and Miller 1998; Coen et al. 1999; Dealteris et al. 2004; Newell and Koch 2004; Newell et al. 2005; Tang et al. 2011; Grabowski et al. 2012). However, shellfish reefs have declined in many parts of the world (Lotze et al. 2006; Beck et al. 2011; zu Ermgassen et al. 2013; Warnock and Cook 2015; Gillies et al. 2020). In some cases, reef losses have exceeded 90% compared with historical abundance. Several different factors may have contributed to such declines, the most common of which is overfishing, particularly where fishing is undertaken using destructive fishing methods (e.g. dredging) that damage or remove hard substrate (Kirby 2004; Lenihan and Peterson 2004; Lotze et al. 2006; Beck et al. 2011). Other factors, such as disease, excessive nutrient input and sedimentation, compound this initial decline and may reduce or prevent the natural recovery of oyster beds (Lenihan and Peterson 1998; Lenihan et al. 1999; Kirby 2004; Mann and Powell 2007; Ogburn et al. 2007; Beck et al. 2011; Diggles 2013).

Reefs of the flat oyster (Ostrea angasi) were once common in many estuaries along the southern coastline of Australia but were mostly lost between c. 1840 and 1870. Few intact oyster beds remained by the end of the 19th century (Nell 2001; Hamer et al. 2013; Alleway and Connell 2015; Gillies et al. 2018). Similar losses have been noted in South Australia and Victoria (Nell 2001; Hamer et al. 2013; Alleway and Connell 2015). This paper records the ‘ecological history’ of O. angasi in Western Australia, and comments on the timing and potential causes of the declines. The second part of the paper suggests how understanding the historic distribution of these oysters, and the causes of historic declines, may help guide ecosystem restoration efforts that are currently being undertaken in Western Australia.

Primary and secondary literature relating to O. angasi in Western Australia was examined to identify locations where this species once existed. Primary data sources included annual reports from the State Commissioner for Fisheries (e.g. Saville-Kent 1894) and other fisheries reports (e.g. Aldrich 1934), as well as records and actual specimens available through the West Australian Museum.

The museum collections included O. angasi shells for which the sampling location, the name of the collector and the collection date have been recorded. All samples were coded to record whether the animal was alive or dead at the time of collection.

Secondary information sources included newspaper articles (National Library of Australia Trove database; http://trove.nla.gov.au/) and the Albany Library Historical Collection Archives (http://history.albany.wa.gov.au/). No attempt was made to record changes in reef abundance over time because of the paucity of robust quantitative and qualitative information.

Once Oyster Harbour had been selected as a suitable location for a case study on the restoration of O. angasi reefs, it was then necessary to establish a protocol to prioritise the most suitable reef restoration sites within the harbour, taking a wide variety of factors into account. The development of a ‘suitability model’ for potential restoration sites was based on the unpublished protocols recommended by Kate Longley-Woods and Simon Reeves (The Nature Conservancy, Australia). A site selection process was conducted using geospatial information relevant to the requirements of O. angasi. The various competing interests of a range of stakeholders were also taken into account, including recreational users, navigational safety (markers and channels) and commercial interests, such as aquaculture leases and fishing areas. Locations supporting seagrass beds were avoided. Further details of this process are described later in the paper.

The distribution of O. angasi, as indicated from shell collections housed at the Western Australian Museum, suggests that the species is, or has been at some point, broadly distributed throughout the south-west of Western Australia (Fig. 1). However, it is possible that the presence of shells in some areas may have resulted from the movement of live oysters by humans. Samples were recorded from both estuarine environments (e.g. Oyster Harbour and Nornalup Inlet) and protected inshore waters (e.g. Cockburn Sound and Esperance; Appendix 1).

Fig. 1.  Locations of oyster samples collected by the Western Australian Museum between 1919 and 2003. Locations where at least one sample was identified as probably alive at time of collection are marked in bold. Locations where commercial fisheries were known to exist are marked with an asterisk (*).

Fig. 2 shows that, in terms of samples collected by the WA Museum, most were collected during the 1970s and 1980s. These periods coincide with a Museum policy of collection enlargement at that time. The most recent collection came from Mandurah in 2003, and was identified as living at the time of capture.

Fig. 2.  Number of O. angasi individuals collected in any specific year by the Western Australian Museum from 1919 to 2003. The highest number of oysters was collected between 1970 and 1990.

Early explorers found plentiful supplies of O. angasi in Western Australia. There were so many oyster beds in Oyster Harbour that George Vancouver ran his vessel aground on a bank of oysters while attempting to leave the harbour in 1791 (Baudin 1809; Vancouver 1898). To commemorate the event, Vancouver named the estuary Oyster Harbour (Vancouver 1898). Baudin (1809) noted the large size of the flat oysters found in the harbour, but neither he nor Vancouver mapped the precise location of the reefs in the harbour.

A dredge fishery for oysters existed in Oyster Harbour, Princess Royal Harbour and King George Sound from the mid-1800s until c. 1880, forming the basis of a ‘lucrative shellfish trade’ (Saville-Kent 1893a, 1893b, 1894; Thompson 1897). However, Saville-Kent noted that the fishery was significantly depleted at least 15 years before his 1893 visit (Saville-Kent 1893a). Although Saville-Kent did not record the precise extent of the oyster beds, he suggested in 1893 (Saville-Kent 1893b), that ‘many square miles’ of oyster banks still existed in both Oyster Harbour and Princess Royal Harbour. He also suggested that, with a little effort, these beds could be resuscitated (Saville-Kent 1893b). The Chief Inspector of Fisheries (C. F. Gale) visited Albany in 1899 and noted that ‘small beds still exist’ in the deeper waters of Oyster Harbour (Gale 1899).

In 1913, the annual report on the fisheries of the state identified Oyster Harbour as an area where O. angasi could still be obtained (Gale 1900, 1905; Inspector of Fisheries 1912, 1913). The state of oyster stocks in the nearby Irwin and Nornalup inlets was described in a 1934 report on Western Australian Fisheries (Aldrich 1934). Although not noted in that report, it appears that in Nornalup Inlet O. angasi beds were present until at least the 1920s (Linton 1923). Linton (1923) described the beds in Nornalup Inlet as ‘the best oyster country of any of the inlets [in the South West]’. He suggested that depletion of the stock was unlikely because no dredging occurred in the inlet (Linton 1923). However, in what appears to be a contradictory statement, Senddan (1923) noted in a report in the Western Mail that although the oysters in Nornalup were ‘rather too large to eat, some steps are required to preserve the oyster beds from total destruction’.

Overfishing and a lack of appropriate management protocols have contributed to oyster population declines in South Australia, Victoria and New South Wales (Nell 2001; Hamer et al. 2013; Alleway and Connell 2015). An inappropriate fishing method (dredge fishing) was often used to harvest oysters, and beds were often fished to exhaustion. Another problem was that oyster shell was used in the production of lime for mortar, and this was another factor that contributed to the loss of oyster beds. In some cases, shell beds were so completely destroyed that recovery by settlement of spat was limited because hard substrate was no longer available. Similar problems contributed to the loss of oyster beds in Oyster Harbour (Warnock and Cook 2015).

Environmental changes that occurred in both Princess Royal Harbour and Oyster Harbour may have further inhibited the recovery of oyster beds. For example, land clearing in the Oyster Harbour catchment that occurred in the 1950s and 1960s increased sedimentation rates in the King and Kalgan rivers, as well as in Oyster Harbour (Hodgkin and Clark 1990a), that may have smothered benthic habitats, such as oyster beds and seagrass meadows.

A recent study in Nornalup and Wilson inlets near Albany did not find any live individuals, but O. angasi shells were abundant in some locations (Western Australian Department of Environment and Conservation and Western Australian Marine Parks and Reserves Authority 2009). Similarly, no live specimens were found in any south-west estuaries during routine sampling in the 1980s and 1990s (Hodgkin and Clark 1988, 1989a, 1989b, 1990a, 1990b, 1999). However, these studies were not specifically aimed to sample O. angasi. Therefore, it is possible that remnant populations may have been missed.

In 2002, 10 flat oysters were collected from Oyster Harbour and their mitochondrial DNA was typed for 16S and cytochrome oxidase 1 (CO1). Seven of the collected oysters were identified as O. angasi, whereas three were identified as Ostrea edulis (Morton et al. 2003).

In 1893, Saville-Kent (Commissioner of Fisheries for Western Australia) attempted to rebuild the oyster stocks in Princess Royal Harbour in order to ‘resuscitate’ the fishery. He suggested that spat settled regularly on moored vessels and other solid substrates in the harbour, but that there was insufficient hard substrate to develop large populations (Saville-Kent 1893b). He identified a small waterway that was open to the sea via a culvert as a location to establish a population of O. angasi. Live brood stock oysters were collected and placed on wooden frames with wire netting mesh, or on wooden planks attached to bricks. By May 1895, the oysters appeared to be doing well but, by October 1896, the waterway had filled with silt and seaweed (Learoyd 1896), smothering the oysters and facilitating the growth of mud worms. Learoyd (1896) suggested that by October 1896, very few O. angasi were still alive and all were infested with the mud worm. Oyster Harbour was then suggested as a better location to develop oyster beds, and a different species of oyster, Saccostrea glomerata, was recommended by Learoyd (1896).

Further attempts to culture O. angasi in hatcheries did not occur until the 1990s. A local company, Ocean Foods International Pty Ltd (OFI), secured leases in both Oyster Harbour and King George Sound and developed a local hatchery. However, it soon became apparent that the local Ostrea spp. was susceptible to the protozoan parasite Bonamia exitiosa, which is lethal to many oyster species (Dinamani et al. 1987; Hine and Jones 1994; Adlard 2000; Cranfield et al. 2005). Following a high level of mortality, OFI began producing an alternative oyster species, namely Saccostrea glomerata, which was less affected by the Bonamia parasite

Once all the historical data on the previous distribution of O. angasi in the estuaries of south-west Western Australia had been examined, they were used to indicate the most suitable locations to restore O. angasi reefs. The presence of O. angasi shells or live oysters, either currently or in historical records, potentially indicates that, at some point in time, the waterbody where they were found was a suitable location for the species. On this basis, Oyster Harbour was selected as a suitable location for a case study on the restoration of O. angasi reefs. Following this, it was then necessary to establish a protocol to prioritise the most suitable reef restoration sites within the harbour, taking a wide variety of factors, and stakeholder views, into account. This prioritisation process is described below.

To optimise the success of the new reefs, a site selection process was conducted using geospatial information relevant to the requirements of O. angasi, and taking into account the various competing interests of a range of stakeholders. An assessment was undertaken comparing tolerance levels of O. angasi with local water quality data (salinity, temperature and dissolved oxygen concentration) collected by the Department of Water and Environmental Regulation (DWER) in Oyster Harbour (https://estuaries.dwer.wa.gov.au). Other environmental and social factors, such as seagrass habitat and alternative water usage, again accessed from DWER, were then overlayed onto the resulting maps. Scores were assigned to each parameter and an overall ‘suitability score’ calculated for each location. The parameters used in this process are listed in Table 1. Although all parameters outlined in Table 1 were considered, only those that were relevant for the suitability model (i.e. as indicated in rationale and notes) were included. Areas that achieved a zero score were considered unsuitable for oyster reef restoration, whereas areas with the highest scores were given the highest prioritisation for reef deployment. The resulting priority area map is shown in Fig. 3. Buffers were designed to surround areas considered not suitable (e.g. aquaculture leases and navigation channels). The maps produced from this process were used to gauge community acceptability of reef construction within the highest-priority zones.

Table 1.  Parameters considered in preparing suitability scores for oyster reef restoration sites in Oyster Harbour
DO, dissolved oxygen; MPA, marine protected area

Fig. 3.  Map showing suitability scores for oyster reef restoration in Oyster Harbour. The green dots represent the positions of the newly constructed restoration sites.

Following this process, the most suitable areas to reconstruct O. angasi reefs were chosen and restoration efforts commenced. In late 2019, The Nature Conservancy constructed ~1650 m² of reef from 1000 tonnes (Mg) of small limestone boulders. Shortly after, ~1 million O. angasi spat, reared in the Albany Shellfish Hatchery, were seeded onto the reef in an effort to restore a sustainable oyster population in Oyster Harbour. Although it is still too early to accurately assess success, initial results suggest that the newly constructed reefs are on an ecological trajectory consistent with becoming fully functioning shellfish reefs in the future.

In terms of future restoration efforts within Oyster Harbour, the same site suitability maps could be used and a similar process could be followed in other bays and estuaries in Western Australia where future oyster reef restoration may be considered.

The lack of sufficient robust information on O. angasi distribution throughout the south-west of Western Australia makes quantitative evaluation of the current status of the stock very difficult, but available information suggests that the current distribution of O. angasi is considerably reduced from historical levels. Overfishing appears to have caused the initial decline, with destructive fishing methods exacerbating the problem. Factors such as poor water quality, increased sedimentation and the loss of seagrass are likely to have contributed to the fact that stocks have not been able to recover naturally. This lack of natural recovery over the century that followed the effective end of the O. angasi fishery suggested that urgent restoration efforts were required if the ecosystem services provided by oyster reefs were to be restored. A process of site selection and prioritisation has been outlined for Oyster Harbour that used the historical presence of O. angasi as a starting point, and it is suggested that this approach could be used in the process of identifying and prioritising suitable restoration sites in other parts of the state.

All data held by corresponding author and are available as required.

The authors declare that they have no conflicts of interest.

This project was made possible by financial contributions from The Thomas Foundation, the Recreational Fishing Initiatives Fund (Recfishwest) and the Great Southern Development Commission (Great Southern Community Chest Fund).