Interpretive Reports
Updated Study Reporting Levels (SRLs) for Trace-Element Data Collected for the California Groundwater Ambient Monitoring and Assessment (GAMA) Priority Basin Project, October 2009–March 2013
Davis, T.A., Olsen, L.D., Fram, M.S., and Belitz, K., 2014, U.S. Geological Survey Scientific Investigations Report 2014-5105, 52 p.
ABSTRACT
Groundwater samples have been collected in California as part of statewide investigations of groundwater quality conducted by the U.S. Geological Survey for the Groundwater Ambient Monitoring and Assessment (GAMA) Priority Basin Project (PBP). The GAMA-PBP is being conducted in cooperation with the California State Water Resources Control Board to assess and monitor the quality of groundwater resources used for drinking-water supply and to improve public knowledge of groundwater quality in California. Quality-control samples (source-solution blanks, equipment blanks, and field blanks) were collected in order to ensure the quality of the groundwater sample results.
Olsen and others (2010) previously determined study reporting levels (SRLs) for trace-element results based primarily on field blanks collected in California from May 2004 through January 2008. SRLs are raised reporting levels used to reduce the likelihood of reporting false detections attributable to contamination bias. The purpose of this report is to identify any changes in the frequency and concentrations of detections in field blanks since the last evaluation and update the SRLs for more recent data accordingly. Constituents analyzed were aluminum (Al), antimony (Sb), arsenic (As), barium (Ba), beryllium (Be), boron (B), cadmium (Cd), chromium (Cr), cobalt (Co), copper (Cu), iron (Fe), lead (Pb), lithium (Li), manganese (Mn), molybdenum (Mo), nickel (Ni), selenium (Se), silver (Ag), strontium (Sr), thallium (Tl), tungsten (W), uranium (U), vanadium (V), and zinc (Zn).
Data from 179 field blanks and equipment blanks collected from March 2006 through March 2013 by the GAMA-PBP indicated that for trace elements that had a change in detection frequency and concentration since the previous review, the shift occurred near October 2009, in conjunction with a change in the capsule filters used by the study. Results for 89 field blanks and equipment blanks collected from October 2009 through March 2013 were evaluated for potential contamination bias by using the same approach developed by Olsen and others (2010). Some data collected by the National Water-Quality Assessment (NAWQA) Program for the Southern California Coastal Drainages study unit were included to supplement the GAMA-PBP data. The detection frequency and upper threshold of potential contamination bias (BD-90/90) were determined from field-blank and equipment-blank data for each trace element. The BD-90/90 is the 90th percentile concentration of potential extrinsic contamination calculated by using the binomial probability distribution for greater than 90 percent confidence. Additionally, data from laboratory blanks and blind blanks analyzed by the National Water Quality Laboratory (NWQL) during water years 2010 through 2013, and compiled by the USGS Branch of Quality Systems (BQS), were considered for each trace element. These results were compared to each constituent’s reporting level to determine whether an SRL was necessary to minimize the potential for detections in the groundwater samples, attributed principally to contamination bias. Results of the evaluation were used to set SRLs for trace-element data for about 1,135 samples of groundwater collected by the GAMA-PBP between October 2009 and March 2013.
Ten trace elements analyzed (Sb, As, Be, B, Cd, Li, Se, Ag, Tl, and U) had blank results that did not necessitate establishing SRLs during this review or the review by Olsen and others (2010). Five trace elements analyzed (Al, Ba, Cr, Sr, and V) had blank results that necessitated establishing an SRL during the previous review but did not need an SRL starting October 2009. One trace element (Fe) had field and laboratory-blank results that necessitated keeping the previous SRL (6 micrograms per liter [µg/L]). Two trace elements (Ni and W) had quality-control results that warranted decreasing the previous SRL, and five trace elements (Cu, Pb, Mn, Mo, and Zn) had field, laboratory, or blind blank results that warranted establishing an SRL for the first time or increasing the previous SRL. SRLs for Cu (2.1 µg/L), Pb (0.82 µg/L), Mn (0.66 µg/L), Mo (0.023 µg/L), Ni (0.21 µg/L), W (0.023 µg/L), and Zn (6.2 µg/L) were changed to these levels starting October 2009, based on the BD-90/90 concentration for field blanks or the 99th percentile concentration for laboratory or blind blanks. The SRL for Fe was maintained at 6 µg/L, based on the minimum laboratory reporting level for iron. SRLs for these eight constituents were at least an order of magnitude below the regulatory benchmarks established for drinking water for health and aesthetic purposes; therefore, the practice of reporting concentrations below the SRLs as less than or equal to (≤) the measured value would not prevent the identification of values greater than the drinking-water benchmarks. Co was detected in 99 percent of field blanks, and with a BD-90/90 concentration of 0.38 µg/L, all groundwater results starting October 2009 were coded as “reviewed and rejected.†Co does not currently have a regulatory benchmark for drinking water. The primary sources of contamination for trace elements inferred from this review are the equipment or processes used in the field to collect the samples or in the laboratory. In particular, contamination in field blanks of Co and Mn was attributed to the high-capacity 0.45-micrometer pore-size capsule filters that were in regular use beginning in October 2009 by several USGS programs, including the GAMA-PBP and NAWQA Program, for filtering samples for analysis of trace elements.
The SRLs determined in this report are intended to be used for GAMA groundwater-quality data for samples collected October 2009 through March 2013, or for as long as quality-control data indicate contamination similar to what was observed in this report; quality-control data should be continuously reviewed and SRLs re-assessed on at least a study-unit basis.