Study location and species
The four species used in this study were collected from subantarctic Macquarie Island (54.6167 degrees S, 158.8500 degrees E), just north of the Antarctic Convergence in the Southern Ocean. Sea temperatures surrounding Macquarie Island are relatively stable throughout the year, with average temperatures ranging from ~4 to 7 degrees C [25]. Collection sites were free from any obvious signs of contamination and did not have elevated concentrations of metals as confirmed by analysis of seawater samples from the collection sites by inductively coupled plasma optical emission spectrometry (ICP-OES; Varian 720-ES).
Toxicity tests were conducted at Macquarie Island over the 2013/14 austral summer, and at the Australian Antarctic Division (AAD) in Tasmania, Australia, from 2013 to 2015. The aquarium at the AAD used for culturing and for holding biota prior to their use in tests was maintained at a temperature of 5.8 degrees C under flow-through conditions (at 0.49L/sec). Individuals for toxicity tests on the island and individuals for return to Australia for culturing were collected from a range of habitats within the intertidal and subtidal zones. All species were highly abundant in each of their respective habitats. The gastropod Laevilittorina caliginosa was collected from pools high on the intertidal zone; the flatworm Obrimoposthia ohlini, from the undersides of boulders from the intertidal to shallow subtidal areas; the bivalve Gaimardia trapesina, from several macroalgae species in high energy locations in the shallow subtidal; and the isopod Limnoria stephenseni, from the floating fronds of the kelp Macrocystis pyrifera, which were located several hundred meters offshore. Test specimens were acclimated to laboratory conditions 24 h to 48 h prior to commencement of tests.
Juvenile flatworms, isopods and gastropods were all products of reproduction in the laboratory at the AAD, and hence their approximate age at testing is known. The flatworms hatched from small (2 mm in diameter) brown eggs, laid on rocks or on the side of aquaria. The flatworms exhibited age based morphological differences; juvenile flatworms were light grey in colour, while the adults were black. The gastropods hatched from small (1 mm in diameter) translucent eggs laid on weed, often in a cluster. For flatworms and gastropods, juveniles were not all the same age at testing due to differing hatching times, with ages ranging from 2 weeks to 3 months. In contrast, juvenile isopods were all the same age. Although brooding isopods were not observed, juveniles were noticed during routine feeding, thus were likely within 2-3 days of being released, 6 months after adults were brought from the field to the aquarium. The tests with these juvenile isopods were done within 1 week of their being observed.
Care was taken to collect adults from the field, for each species, within a narrow size range to minimise differences in ages between individuals tested (Table 1). However, ages of adults individuals used in tests are unknown. The smaller size class of bivalves tested (juveniles: 2.5 plus or minus 0.5 mm, Table 1) was also collected from the field along with the adults (8.0 plus or minus 1.0 mm, Table 1). Based on knowledge on the growth rate of this species (0.8 mm per year; Everson [26], the smaller size class likely represents a young adult of approximately 2.5 to 4 y old, as opposed to a juvenile stage, and adults collected were approximately 9 to 11 y old.
Toxicity tests
A static non-renewal test regime was used for all tests. Two replicate tests were done for each species at each life stage, with the exception of the juvenile isopods, where due to the limited number of individuals available, only one test was done. Longer tests durations of 14 days were done for acute responses due to the longer life span and response to contaminants compared to temperate and tropical species as determined in previous studies [7, 27].
All experimental vials and glassware were washed in 10% nitric acid and rinsed thoroughly with MilliQ water three times before use. Tests were done in lidded polyethylene vials of varying sizes, depending on the size and number of individuals in the test (Table 1). Water was not aerated as DO stayed relatively high for tests due to high dissolution rates in cold water. Acid washed and Milli-Q rinsed mesh (600 micron nylon) was provided for isopods to rest on, while no structure was added to vials for the other test species. Test solutions were prepared 24 h prior to the addition of invertebrates. Five copper concentrations in seawater were prepared using a 500 mg/L Univar analytical grade CuSO4 in MilliQ stock solution, plus a control for each test. Seawater was filtered to 0.45 microns, and water quality parameters were measured using a TPS 90-FL multimeter at the start (d 0) and end (d 14) of tests. Dissolved oxygen (DO) was greater than 80% saturation, salinity was 33 to 35 ppt, and pH was 8.1 to 8.3 at the start of tests. Tests were kept in controlled temperature cabinets set at 6 degrees C under 16:8h light:dark during the summer, and 12:12 for tests during the rest of the year (light intensity of 2360 lux). Temperatures within cabinets were monitored throughout the test using Thermochron iButton data loggers.
Water samples of each test concentration were taken at the start (day 0) and end of tests (day 14). Samples were filtered through an acid and Milli-Q rinsed, 0.45 microns Minisart syringe filter and acidified with 1% ultra-pure nitric acid before being analysed by ICP-OES to determine dissolved metal concentrations. Measured concentrations at the start of tests were within 96% of nominal target concentrations. Averages between measured concentrations at the start and end of tests were made to estimate exposure concentrations, which were subsequently used in statistical analyses to determine point estimates (Table 2). Both survival and sublethal (behavioural) endpoints were used to determine sensitivity to copper. Vials were checked daily and survival and sublethal responses were observed and recorded on days 1, 2, 4, 7, 10 and 14. Tests were terminated when surviving individuals occurred in less than two concentrations, which was generally at 14 d for all species except for bivalves, in which this occurred sooner (7 to 10 d). Gastropods were scored as dead when their operculum was open and there was no response to stimulus (touch of a probe) on the operculum. Flatworms were scored as dead when there was no movement. Bivalves was scored as dead when there was no movement and when the shells were gaping open due to dysfunctional adductor muscles. Isopods were scored as dead when there was no movement of any appendages.
The behavioural endpoint scored for each species was attachment, which indicated healthy and active individuals. For gastropods, this meant the foot was fully extended and attached to experimental vials; for flatworms, the whole body was able to attach (as those affected by copper appeared slightly contracted and could not lie flat); for bivalves, the byssal threads were used to fix individuals to the bottom of the vial, with the siphon also visible and shell slightly open for water exchange; and for isopods, individuals were either holding onto the provided mesh or were swimming, in which case they often reattached to the mesh during observation.
