Table 2.

QC samples and acceptance criteria for mercury analysis of fish tissue

QC operation	Frequency	Acceptance limit	Summary of QC findings
Bubbler blank	3 blanks run during calibration and with each batch of up to 20 field samples	50 picograms (pg) of mercury	None of the bubbler blanks were above 50 pg, and none of the averages of the 3 blanks in each of the batches were above 25 pg
Method blank	3 method blanks per batch of up to 20 field samples, with analyses interspersed among the samples in the analysis batch	0.4 nanograms (ng) (400 pg) of mercury, or Less than one-tenth the concentration of an associated sample	None of the method blanks were above 400 pg. Overall, 60% of the method blanks were non-detects (below 40 pg), and 40% of the method blanks were detects from 40 to 230 pg. The lowest of 600 field sample results was 23 ng/g, or 100 times the highest blank
Laboratory control sample (LCS)	1 LCS prepared per batch of up to 20 field samples, analyzed once prior to the analysis of any field samples, and again at the end of each analytical batch, spiked at 4.0 ng	70–130% recovery (5.6–10.4 ng/g)	None of the laboratory control sample results fell outside the acceptance limits. Recoveries ranged from 80 to 119% overall
Reference material sample (laboratory used TORT-3 from the National Research Council of Canada)	1 per batch of up to 20 field samples	75–125% recovery	None of the reference sample results fell outside the acceptance limits. Recoveries ranged from 88 to 105% overall
Matrix spike sample	1 per batch of up to 20 homogenized fillet tissue samples (not required for plug samples due to limited mass) (laboratory opted to run two matrix spike samples per batch)	70–130% recovery	None of the matrix spike sample results associated with the homogenized fillet tissue batches fell outside the acceptance limits. Recoveries ranged from 71 to 119% overall