19 



blank, a blank spike, and a blank spike duplicate sample. Sample data were evaluated using 

 protocols developed by the EPA Contract Laboratory Program (CLP; EPA 1988a). 



Samples analyzed for pesticides were extracted 146 days after sample collection, and 

 analyzed 79 days after sample extraction. Samples analyzed for total PCBs were extracted 

 146 days after sample collection, and analyzed 50 days after sample extraction (Table 2-2). 

 EPA guidelines suggest maximum holding times for pesticide and PCB samples of 14 days 

 from sampling to extraction, and 40 days from extraction to analysis. Due to these excessive 

 holding times, the pesticide and PCB data were qualified as estimated, and given the qualifier 

 code J for detected values and UJ for undetected values. 



Each sample analyzed for pesticides was spiked with two system monitoring 

 compounds, or surrogates (dibutyl chlorendate and TCMX). Surrogate QC samples were 

 analyzed as a check on the laboratory's ability to extract known concentrations of compounds 

 not found normally in the sample, and were a measure of laboratory accuracy. Three 

 pesticide samples (MQR2-E, MQR5-A, and MQR6-A) had low TCMX recoveries, and the 

 method blank had a very low dibutyl chlorendate recovery (9%). Since all of the pesticide 

 samples had already been qualified, no further qualifications were necessary. Every PCB 

 sample was also spiked with one surrogate compound (TCMX). One sample exceeded 

 control limits for TCMX recovery (MQR5-A); since this sample had already been qualified, 

 no further qualification was necessary. 



The pesticide and the PCB method blanks were both below detection for all 

 compounds, indicating no laboratory contamination problem. A blank spike and a blank 

 spike duplicate sample were analyzed for both total PCBs and pesticides as an indication of 

 laboratory accuracy and precision. Accuracy was evaluated by calculating the recovery of 

 the spiked compound in the blank. Precision was evaluated by calculating the relative 

 percent difference (RPD) between blank spike duplicate samples. 



Recoveries of total PCBs for both blank spikes were within control limits. Pesticide 

 blank duplicate samples were spiked with five pesticide compounds: lindane, heptachlor, 

 aldrin, dieldrin, endrin, and 4,4'-DDT. Only one recovery of aldrin (124%) in one blank 

 spike sample was above control limits (120%). Accuracy of both pesticide and PCB data 

 was considered acceptable. Precision of both pesticide and PCB data was good; all RPDs 

 were less than 20%. 



Sample MQR5-E resulted in a very high concentration of total PCBs (31 ppm); this 

 result was investigated and confirmed by the NED laboratory. Except for the exceeded 

 holding times, all pesticide and PCB data were considered acceptable. 



Sediment Core Chemistry Data Summary from the MQR Mound, August and December 1991 



