In Reply Refer To:
Mail Stop 412
September 28, 2010
Office of Water Quality Technical Memorandum 2010.07
Subject: Changes to the Reporting Convention and to Data Qualification Approaches for Selected Analyte Results Reported by the National Water Quality Laboratory (NWQL)
This memorandum describes changes to data reporting by the NWQL for analytes currently reported to the National Water Information System (NWIS) using the Laboratory Reporting Level (LRL) convention (Childress and others, 1999).
The NWQL is changing the reporting convention for all inorganic analytes that currently use the LRL convention [Attachment A] to simplify data reporting by both USGS personnel and the NWQL, and to minimize potential complications associated with subsequent statistical evaluation of data reported using the LRL convention as noted by Helsel (2005, 2006).
The NWQL also is revising the approach used to qualify results that are reported to NWIS for both inorganic [Attachment A] and organic [Attachment B] analytes reported using the LRL convention. The new approach eliminates application of the estimated (E) remark code to sample results that fall within specified concentration ranges. More descriptive NWIS result-level value qualifier codes are used to identify results in these ranges [see table 13 of Appendix A of the NWIS Users manual (U.S. Geological Survey, 2010)]. Use of the value qualifier codes as described below meets the data qualification reporting requirements of The NELAC Institute (NELAC, 2009).
The following changes will be implemented for samples collected on or after October 1, 2010:
For all inorganic analytes currently reported using the LRL convention [see Attachment A]:
The reporting level* will be set at the long-term method detection level (LT-MDL) concentration.
Concentrations that are less than the LT-MDL will be reported as “less than” the LT-MDL concentration (<LT-MDL).
The “report level type” displayed with the sample result in NWIS will be LT-MDL [see table 14 of Appendix A of the NWIS Users manual (U.S. Geological Survey, 2010)].
Results in the following concentration ranges will no longer be automatically reported with an “E” remark-code:
LT-MDL ≤ Result < 2 LT-MDL
The “n” NWIS result-level value qualifier code denotes results in this range. Note: Under the LRL convention, the LRL value was set at 2 LT-MDL for most inorganic analytes.
For all organic analytes currently reported using the LRL reporting convention [see Attachment B]:
The NWQL will continue to use the LRL convention and the LRL value as the “less than” reporting level concentration for those organic method analytes currently using that convention. Only the elimination of automatic “E” remark-coding of results that are in the concentration ranges noted below will commence on October 1, 2010. The NWQL currently is re-evaluating the processes used to determine detection and reporting levels for organic methods. Additional reporting convention changes for organic methods are pending.
Results in the following ranges will no longer be automatically reported with an “E” remark-code:
LT-MDL ≤ Result < LRL
Results in this range include the “n” NWIS result-level value qualifier code. The LRL is set at 2 LT-MDL (or sometimes higher) for organic analytes.
Results that are less than the lowest calibration standard.
These results include the “b” result-level value qualifier code.
Results that are less than the LT-MDL.
The “t” NWIS result-level value qualifier code is used to denote these results. Only mass spectrometric methods classified as “information rich” report results in this range (Childress and others, 1999). Although all results need to be considered in relation to laboratory and field quality-assurance information, results that are less than the LT-MDL have an increased risk of false positives and should be carefully evaluated in relation to both laboratory and field blanks. These results also have increased uncertainty because they typically are less than the lowest calibration standard.
No changes will be made to the reporting levels or remark codes in NWIS or in the NWQL’s Laboratory Information Management System database for past samples based on the forthcoming changes.
No reporting convention changes will be made for analytes in inorganic or organic methods that are reported using the Minimum Reporting Level (MRL) convention (Childress and others, 1999). The MRL convention uses a censor-limit based reporting level below which no data are reported. The MRL is set at a concentration greater than the analyte’s detection level.
Elimination of the “E” remark code for results falling in the ranges noted above does not indicate increased certainty in the determined concentration relative to previously reported results in these regions that did include the E-code. Although automatic E-coding is being eliminated for the conditions noted above, the E-remark code might be included on reported results for additional performance-based reasons as described in Childress and others (1999).
Currently, a maximum of 3 result-level value qualifier codes, including others besides “b”, “n” and “t” mentioned above, can be associated with each result in NWIS [see table 13 in Appendix A of NWIS User manual (U.S. Geological Survey, 2010)]. Result-level value qualifier codes have been available for use since NWIS version 4.1 was released in 2001. These value qualifier codes have been used by NWQL.
Some methods that report results using the LRL convention do not have established LT-MDLs for the analytes. These include sediment and tissue, recently developed water, and custom analysis methods. The changes outlined in this memo are identical for all analytes classified as Interim Report Level (IRL) type code in Attachments A and B, and are similarly described by substituting IRL for LRL and method detection limit (MDL) for LT-MDL in all text above.
*The reporting level (a generic term) is the “less than” concentration provided when the analyte is not detected or is detected below a minimum (censor-limit based) concentration, which might be at or below the RL value.
Helsel, D.R., 2005, Insider censoring: distortion of data with nondetects: Human and Ecological Risk Assessment, v. 11, p. 1127-1137.
Helsel, D.R., 2006, Fabricating data: How substituting values for nondetects can ruin results, and what can be done about it: Chemosphere, v. 65, no. 11, p. 2434-2439.
Childress, C.J., Foreman, W.T., Connor, B.F., and Maloney, T.J., 1999, New reporting procedures based on long-term method detection levels and some considerations for interpretations of water-quality data provided by the U.S. Geological Survey National Water Quality Laboratory: U.S. Geological Survey Open-file report 99-193, 19 p. Available at http://water.usgs.gov/owq/OFR_99-193/index.html.
The NELAC Institute, 2009, Environmental laboratory sector, volume 1, Management and technical requirements for laboratories performing environmental analysis. Module 4: Quality systems for chemical testing. Available at http://www.nelac-institute.org/docs/standards/2009/EL-V1M4-2009.pdf; accessed July 22, 2010.
U.S. Geological Survey, 2010, User's manual for the National Water Information System of the U.S. Geological Survey, Water-Quality System, version 4.9. Available on-line only at http://wwwnwis.er.usgs.gov/nwisdocs4_9/qw/QW.user.book.html; accessed July 22, 2010.
For questions on the about these changes, please contact the NWQL via firstname.lastname@example.org.
Donna N. Myers /s/
Chief, Office of Water Quality
This memorandum does not supersede any other OWQ Technical Memorandum.
Distribution: All WRD Employees