The spatially intensive sampling regime employed in the current study using drop cameras recorded the presence of A. senhousia at 212 sites throughout the lower Swan-Canning Estuary. It thus provides clear evidence that not only was A. senhousia present in this estuary but was prominent at many locations and particularly in the lower Canning River and upper half of Melville Water. While this is the first study to explore the distribution of A. senhousia in this estuary since its apparent disappearance in the summer of 2000, large mats of dead shell were trawled up during otter trawling sampling for prawns during the summer of 2017 (Brian Poh, pers. comms). The occurrence of these dead individuals followed an extreme rainfall event resulting in pronounced water column stratification in Melville Water and the formation of a ‘dead zone’ where dissolved oxygen concentrations fell below 2 mg L− 1 in bottom waters (Department of Parks and Wildlife, 2017).
It would be thus reasonable to assume that the presence, distribution and even the persistence of this species in the Swan-Canning Estuary is dynamic, potentially undergoing die offs followed by re-colonisations. This conclusion is consistent with the finding of previous studies in that the beds of A. senhousia usually disappear after a couple of years, possibly due to low oxygen and salinities following substantial freshwater flow (National Introduced Marine Pest Information System, 2002). Furthermore, populations are thought to be highly dynamic, presumably due to highly variable recruitment (NIMPIS, 2002; McDonald and Wells, 2010) and mortality even though this mussel species is known to be tolerant to low oxygen concentrations and relatively low salinities (Government of Western Australia, 2005). Although, it cannot be ruled out that re-colonisation of the benthic habitats could be from re-introductions through the nearby port, given the speed of the recovery it is more likely that re-colonisation is from residual population within the estuary system.
Unlike the native mytildae, Mytilus galloprovincialis and Xenostrobus securis, which colonises hard structure in the Swan-Canning Estuary, A. senhousia has a preference for soft sediments, which comprises much of the estuary bottom. In contrast, mussel beds comprise a structurally complex matrix of living mussels, shells, sediment, and debris, and are considered among the most diverse temperate ecosystems containing several hundred species (Smith et al., 2006) and thus invasions of A. senhousia, such as those in the Adriatic Sea and Tyrrhenian Sea, have been found to have gross positive effects on the benthic community Munari (2008). In Mission Bay, California, an invasion of A. senhousia was also found to contribute substantially to the diets of the fish, the Yellowfin Croaker Umbrina roncador, Spotfin Croaker Roncador stearnsii and Sargo (Anisotremus davidsonii) and the birds the Willet Catoptrophorus semipalmatus and Marbled Godwit Limosa fedoa (Crooks 2002). Although beds of A. senhousia beds have been postulated to compete with seagrasses in other systems (NIMPIS, 2002), the data collected during the present study found that seagrass throughout the study area was almost exclusively found in depths of < 2 m and thus any competition by A. senhousia with seagrass would be limited to a narrow depth range.
The use of simple drop camera, comprising a readily available off-the-shelf camera and simple stainless-steel frame (total cost = AUD500) was cost-effective and facilitated the collection of data at a fine spatial scale (100–200 m) across an area of 28 km2 over the 15 days of sampling. This is a larger extent than could be achieved in the same timeframe using traditional methods involving the collection of sediment from grabs or cores. Moreover, visual analysis of the photos would also be less destructive and time intensive than extraction, identification and quantification of specimens from sediment samples as has been done in many previous studies. It should be noted, that our survey was conducted at a time of year when environmental conditions are stable and that microtidal estuaries, such as the Swan-Canning Estuary are less turbid than macrotidal systems thus facilitative the use of visual methods (Tweedley et al., 2016). The drop camera method employed here is considered is suitable for species such as A. senhousia that form distinct benthic matts and whose spatial distribution and abundance may change markedly and so be missed by less exhaustive sampling.