20 



normally found in the sample, but having similar characteristics. Surrogate compounds, 

 which are the only means of checking method performance on a sample by sample basis, 

 were conducted on all PAH, pesticide, and PCB analyses. 



PAH Accuracy. Sample GR-1 was selected for MS/MSD analysis. For the matrix 

 spike analyses, 12 of 16 compounds resulted in percent recovery values that were 

 significantly below the QC limits (<35% recovery). The matrix spike duplicates also 

 were significantly below the QC limits with 10 of the 16 compounds at or below the 10% 

 recovery. The laboratory provided a quality control review and suggested that the low 

 recoveries may be a function of low spike concentrations relative to the existing sediment 

 concentration of the single sample selected for matrix spike analyses. The high variability 

 of moismre content in marine sediments may also affect the results where the same 

 moismre content is applied to replicates which may differ. Despite the low recoveries in 

 the MS/MSD analysis for sample GR-1, the accuracy of the PAH data was considered 

 acceptable because of the additional quality control results. All of the surrogate spike 

 recoveries, which were tested for each sample, were within acceptable limits. Recoveries 

 for PAHs were between 42-128%, with more limited ranges for each surrogate tested. 

 Surrogate recovery percentages indicate the ability of the laboratory to detect a known 

 concentration of compounds that behave similar to those tested in a sample. All spiked 

 targets of the spiked blank met the acceptance criteria with percent recoveries ranging from 

 63 to 98%. Finally, an instrument reference material (NIST IRM 1491), using an 

 independent-source standard, was tested and all targets also met the acceptance criteria. 



Pesticides/PCB Accuracy. Sample GR-9 was selected for MS/MSD analysis. 

 Aldrin, gamma-BHC, heptachlor, dieldrin, and 4,4'-DDT were spiked as representative 

 pesticides. All MS/MSD sample recoveries resulted in acceptable accuracy. 

 Pesticide/PCB samples were spiked with two surrogate compounds, tetrachloro-m-xylene 

 (TMX) and decachlorobiphenyi (DCB). Recovery of DCB was acceptable for all samples, 

 and TMX acceptable in all but one sample (26% recovery; acceptance limits are 30- 

 150%). In addition, two spiked laboratory control samples were within acceptance limits, 

 indicating overall acceptable accuracy for pesticide and PCB data. The WHG laboratory 

 investigated the continuing calibration verification (CCV) samples for negative bias 

 because of the lack of any measured pesticide or PCB data in the WLIS reference samples. 

 CCV exceedences were not considered to have biased the data, so pesticide/PCB results 

 were reported with no qualification. 



Metals Accuracy. Sample GR-1 was selected for MS/MSD analysis. The 

 MS/MSD recoveries were considered acceptable for all metals in both the MS and MSD 

 samples except aluminum (Al). Low recoveries of spiked Al were due to the low spiked 

 concentration relative to the natural concentration (19000 mg/kg sample concentration. 



Monitoring Cruise at the WLIS Disposal Site, September 1997 and March 1998 



