No abstract

Proliferation of species introduced for aquaculture can threaten the ecological and economic integrity of ecosystems. We assessed whether the non-native Pacific oyster, Crassostrea gigas, has proliferated, spread and overgrown native Sydney rock oysters, Saccostrea glomerata, in Port Stephens, New South Wales (NSW), Australia, following the 1991 decision to permit its aquaculture within this estuary. Sampling of seven rocky-shore and four mangrove sites immediately before (1990), immediately after (1991–1992) and nearly two decades after (2008) the commencement of C. gigas aquaculture did not support the hypotheses of C. gigas proliferation, spread or overgrowth of S. glomerata. The non-native oyster, uncommon immediately before the commencement of aquaculture, remained confined to the inner port and its percentage contribution to oyster assemblages generally declined over the two decades. C. gigas populations were dominated by individuals of <40-mm shell height, with established adults being rare. Only at one site was there an increase in C. gigas abundance that was accompanied by S. glomerata decline. The failure of C. gigas in Port Stephens to cause the catastrophic changes in fouling assemblages seen elsewhere in the world is likely to reflect estuarine circulation patterns that restrict larval transport and susceptibility of the oysters to native predators.

The Australian edible oyster industry has been severely impacted by disease and declining yields since the 1970s. Selective breeding of Saccostrea glomerata is one measure addressing these problems by producing fast-growing, disease-resistant oysters. Farmers report that selected oysters have different growth characteristics than their wild counterparts using conventional grow-out methods. This study investigated how different grow-out methods influence commercially valuable oyster characteristics including shell length, shape, surface growth deformities and meat condition. In June 2015, selectively bred S. glomerata spat were deployed in two estuaries (Hawkesbury River and Georges River) in NSW, Australia, using three grow-out methods (fixed trays, Stanway cylinders and floating baskets). In November 2015, oysters were transferred among grow-out methods to test for the effects of changing grow-out methods on oyster growth patterns. Oysters transferred from baskets to cylinders and from trays to cylinders had, on average, deeper and wider shells, a higher meat condition and fewer shell surface deformities than oysters in other grow-out method combinations. However, these oysters were smaller than oysters not grown in cylinders. While there were some differences in growth patterns between the estuaries, overall it was the grow-out methods that most influenced oyster characteristics. This was attributed to differences in the amount and magnitude of movement oysters experienced in the grow-out methods, as recorded by motion sensors. This study demonstrates how grow-out methods can be managed to achieve desired growth trajectories and therefore improve marketability among selective bred S. glomerata. K E Y W O R D Sedible oyster industry, grow-out methods, oyster shell shape, phenotypic plasticity, selective breeding, Sydney rock oyster

Understanding how the density and spatial arrangement of invaders is critical to developing management strategies of pest species. The Pacific oyster, Crassostrea gigas, has been translocated around the world for aquaculture and in many instances has established wild populations. Relative to other species of bivalve, it displays rapid suspension feeding, which may cause mortality of pelagic invertebrate larvae. We compared the effect on settlement of Sydney rock oyster, Saccostrea glomerata, larvae of manipulating the spatial arrangement and density of native S. glomerata, and non-native C. gigas. We hypothesized that while manipulations of dead oysters would reveal the same positive relationship between attachment surface area and S. glomerata settlement between the two species, manipulations of live oysters would reveal differing density-dependent effects between the native and non-native oyster. In the field, whether oysters were live or dead, more larvae settled on C. gigas than S. glomerata when substrate was arranged in monospecific clumps. When, however, the two species were interspersed, there were no differences in larval settlement between them. By contrast, in aquaria simulating a higher effective oyster density, more larvae settled on live S. glomerata than C. gigas. When C. gigas was prevented from suspension feeding, settlement of larvae on C. gigas was enhanced. By contrast, settlement was similar between the two species when dead. While the presently low densities of the invasive oyster C. gigas may enhance S. glomerata larval settlement in east Australian estuaries, future increases in densities could produce negative impacts on native oyster settlement. Synthesis and applications: Our study has shown that both the spatial arrangement and density of invaders can influence their impact. Hence, management strategies aimed at preventing invasive populations reaching damaging sizes should not only consider the threshold density at which impacts exceed some acceptable limit, but also how patch formation modifies this.

Marine diseases represent a significant threat to wild organisms and the ecosystem services they support, yet studies often consider only disease impacts to aquaculture. In eastern Australia, the Sydney rock oyster (Saccostrea glomerata) aquaculture industry is increasingly affected by outbreaks of QX disease caused by parasitic Marteilia sydneyi. The present study considered impacts of M. sydneyi infection on the structure of wild-oyster populations that are dominated by S. glomerata, but that may also include the non-native Pacific oyster, Crassostrea gigas. In the Hawkesbury River Estuary, where cultured S. glomerata has experienced up to 98% QX-induced mortality, we found that disease prevalence was comparatively low among wild S. glomerata, peaking at 14%, and annual infections did not cause seasonal patterns of mortality. Furthermore, C. gigas, a competitor of S. glomerata that is not susceptible to QX disease, was not consistently more abundant at sites with than without the parasite. Overall, our results indicated that relative to cultured counterparts, wild S. glomerata in the Hawkesbury River Estuary is minimally affected by QX disease. Nevertheless, our study showed that diseases of aquaculture stocks have the capacity to infect wild populations, and that longer-term assessment of wild populations at risk is essential.