Sediment Quality in Puget Sound Year 2 - Center for Coastal ...

analyses will be reported elsewhere by NOAA. After staining with rose bengal, the 1.0 mm
sample fractions were examined under dissection microscopes, and all macroinfaunal
invertebrates and fragments were removed and sorted into the following major taxonomic
groups: Annelids, Arthropoda, Molluscs, Echinodermata, and miscellaneous taxa. Meiofaunal
organisms such as nematodes and foraminiferans were not removed from samples, although their
presence and relative abundance were recorded. Representative samples of colonial organisms
such as hydrozoans, sponges, and bryozoans were collected, and their relative abundance noted.
Sorting QAIQC procedures consisted of resorting 25% of each sample by a second sorter to
determine whether a sample sorting efficiency of 95% removal was met. If the 95% removal
criterion was not met, the entire sample was resorted.
Taxonomic Identification
Upon completion of sorting and sorting QAIQC, the majority of the taxonomic work was
contracted to recognized regional taxonomic specialists. Organisms were enumerated and
identified to the lowest taxonomic level possible, generally to species. In general, anterior ends
of organisms were counted, except for bivalves (hinges), gastropods (opercula), and ophiuroids
(oral disks). When possible, at least two pieces of literature (preferably including original
descriptions) were used for each species identification. A maximum of three representative
organisms of each species or taxon was removed from the samples and placed in a voucher
collection. Taxonomic identification quality control for all taxonomists included re-
identification of 5% of all samples identified by the primary taxonomist and verification of
voucher specimens generated by another qualified taxonomist.
Data Summary, Display, and Statistical Analysis
Toxicity Testing
Amphipod Survival - Solid Phase
Data from each station in which mean percent survival was less than that of the control were
compared to the CLIS control using a one-way, unpaired t-test (alpha < 0.05) assuming unequal
variance. Results were not transformed because examination of data from previous tests has
shown that results of tests performed with A. abdita met the requirements for normality.
"Significant toxicity" for A. abdita is defined here as survival statistically less than that in the
performance control (alpha < 0.05). In addition, samples in which survival was significantly less
than controls and less than 80% of CLIS control values were regarded as "highly toxic". The
80% criterion is based upon statistical power curves created from SAIC's extensive testing
database with A. abdita (Thursby et al., 1997). Their analyses showed that the power to detect a
20% difference from the control is approximately 90%. The minimum significant difference
(i.e., "MSD" of ~ 80% of control response) was used as the critical value in calculations of the
spatial extent of toxicity (Long et al., 1996, 1999a).
Sea Urchin Fertilization - Pore Water
For the sea urchin fertilization tests, statistical comparisons among treatments were made using
ANOVA and Dunnett's one-tailed t-test (which controls the experiment-wise error rate) on the
Page 19