In Reply Refer To:
Mail Stop 412
                              February 11, 2004


Subject:    Revised Policy for the Approval of U.S. Geological Survey
            (USGS) Water-Quality Analytical Methods

I.  Purpose of Memorandum
This memorandum restates the Water Resources Discipline (WRD) policy for
the approval of water-quality analytical methods contained in Office of
Water Quality (OWQ) Technical Memorandum 98.05, and supersedes it.  It also
includes a description of the process and requirements for obtaining WRD
approval of a new water-quality analytical method or modification to an
approved method, the process for releasing data to the public prior to
method approval, and procedures for handling these data in publicly
accessible databases, in particular NWISWeb (Attachment 1).

II.  Policy
All water-quality analytical data collected by the WRD for release to the
public in data reports, district annual data reports, and in databases that
are directly accessible by the public must be produced using WRD-approved
methods.  This policy was established to ensure that USGS data are of known
and documented quality and that the analytical methods used to produce the
data are thoroughly tested, documented, and available to the public.

Water-quality data produced using unapproved methods can be published in
any kind of interpretive report as long as the methods used are
specifically referenced, or a detailed explanation of the method is
included in the report.  Data obtained using unapproved methods can, and
should, be entered into the National Water Information System (NWIS)
database with appropriate qualifier codes that clearly label the result as
being produced by an unapproved method.  These data must also be flagged in
NWIS so that they will not be released to the public through NWISWeb.

In some instances, the data quality of results produced by a water-quality
analytical method prior to approval may be comparable to the data quality
produced by the method subsequent to approval.  Under these conditions, and
with the concurrence of the Chief of the Office of Water Quality, these
comparable results may be released to the public through publicly
accessible databases, including NWISWeb.  The process required to obtain
this concurrence is outlined in Attachment 1 of this memo.

In other cases, re-evaluation of the quality of data based on laboratory
experience with a large number of environmental samples during the early
stages of implementation of a newly approved analytical method may show
that data for some constituents are not of acceptable quality.  Likewise,
some sample matrices or sample types may result in unacceptable data
quality for some or all constituents.  In either case, if data being
produced are not of acceptable quality, those results may be removed from
NWISWeb through the process outlined in Attachment 1 of this memo.

III.  Sources of Approved Methods
Methods already accepted and published by the following sources have
automatic WRD approval:
a.  Methods published by the U.S. Environmental Protection Agency (USEPA)
b.  Methods published by the American Society for Testing and Materials
c.  Methods published by the American Public Health Association in Standard
  Methods (1995 and revisions)

New methods or modifications of existing methods that are not published in
the sources above must be published in a USGS series publication or similar
peer-reviewed report.  In some cases, a journal article may be substituted
for a report if the article contains all the information requested in
Attachment 1.


American Public Health Association, American Water Works Association, and
Water Environment Federation, 1995, Standard methods for the examination of
water and wastewater (19th ed.): Washington, D.C., American Public Health
Association, variously paged.

U.S. Environmental Protection Agency, 1994, Guidelines establishing test
procedures for the analysis of pollutants (App. B, Part 136, Definition and
procedures for the determination of the method detection limit): U.S. Code
of Federal Regulations, Title 40, revised as of July 1, 1994, p. 635-637
(and revisions).


Water Resources Division Memorandum 98.05 (Superseded): Policy for the
Approval of U.S. Geological Survey (USGS) Water-Quality Analytical Methods.

Questions should be directed to the Chief of the Office of Water Quality.

                                                 Timothy L. Miller /s/
                                                 Chief, Office of Water Quality

This memorandum supersedes Office of Water Quality Technical Memorandum

Distribution:   A, B, S, FO, PO
                      District Water-Quality Specialists
                      OWQ Staff
                      Regional Water-Quality Specialists

Key words: analytical method, database, water quality

Attachment 1:


1.  "Water-quality analytical methods" are chemical, biological, or
physical methods used to measure various properties of hydrologic systems
for purposes of assessing water quality.
2.  "Hydrologic systems" include precipitation, surface water, and ground
water, along with associated atmospheric, biological, and geologic
3.  "Data" refer to the results of the measurements on a hydrologic system
using the water-quality analytical method.  Results are typically
quantifiable numbers, such as analyte concentrations, biological species
names and numbers, or particle size.
4.  "Approved data" are the results of an approved water-quality analytical
method produced under the quality-assurance/quality-control specifications
of the method.  These data may be published in USGS data reports and/or
distributed to the public via publicly accessible databases, including
5.  "Analyte" refers to a constituent, element, isotope, or compound being
determined in a chemical analysis.
6. "Data quality" refers to the properties of the measurement, such as
precision, bias, detection limit, and other relevant measures.

Process for Obtaining Approval of New Methods:
 Any laboratory providing data to WRD can request approval of a new
water-quality analytical method by submitting the following to the Chief,

1.  A cover memorandum, from the originating laboratory, through the
appropriate WRD supervisory chain-of-command, asking for approval of the
method and describing why the method should be approved by WRD.

2.  A published report or USGS report that has completed the review
process, and that describes the water-quality analytical method.  A USGS
report that has not yet received Director's approval must include the
standard manuscript-routing sheet including all signatures, up to final
Director's approval, along with all technical peer-review comments and the
author's responses to those comments.  At least one of the peer reviewers
must be a technical expert from outside the originating laboratory.

3.  The originating laboratory's standard operating procedure (SOP) for the
method in the form and content specified by the originating laboratory's
quality-assurance manual.

Submittals will be reviewed by the OWQ within 3 weeks.  All supporting data
and calculations related to the report shall be available for review upon
request.  Any required changes will be communicated back to the originator
in writing.  For USGS reports that have not yet received Director's
approval, the manuscript routing sheet will be signed by the Chief, OWQ, or
his designee, and returned to the originating laboratory for final report

Upon acceptance of the water-quality analytical method, and contingent upon
final approval of the report for publication, a memo from the OWQ will be
written to document the method as being approved by WRD.

Approval of a new method requires documentation of method performance,
which should include three different matrices; two different concentrations
(one high and one low) and seven replicates of each combination.  All
analytes shall be spiked into each matrix.  For water methods, the three
matrices should include a surface-water, a ground-water, and a
reagent-water matrix.  In some instances, specific water-quality variables,
such as conductivity or dissolved organic carbon, may be known to have
substantial effects on the performance of analytical methods.  Specific
consideration should be given to selecting water sources that span expected
ranges of those key variables.  For other media, the laboratory should
choose three different representative matrices that are appropriate.  When
spiking is not appropriate, method performance may be documented with
standard reference sample precision and accuracy, ideally at multiple
concentrations, if available.

The laboratory method blank shall be fully characterized.  An initial
method detection limit (MDL) shall be determined according to the USEPA
(1994), and must include necessary iterations as specified therein.  Data
for the MDL determination shall include data collected over several days
using different calibration curves.  When the USEPA MDL is not appropriate,
relevant reporting limits will be specified and justified.  Reporting
conventions for the analytical data will be specified and justified,
including numbers of significant figures for different analyte
concentration ranges.  Values reported for non-detections of each analyte
will also be specified and justified.

Process for Obtaining Approval for Modifications to Existing Approved
Changes to WRD-approved water-quality analytical methods must be tightly
controlled. To maintain data quality, method changes expected to affect
data quality must be documented as a new water-quality analytical method
with full validation in a new publication as described above.  On occasion,
it may be appropriate to add analytes to an existing approved water-quality
analytical method.  These changes require documentation in a new
publication using the same procedures outlined for the original method

Changes to approved water-quality analytical methods that are not expected
to affect data quality must be documented in changes to the laboratory's
SOP for that method. These changes shall be recorded as SOP version number
changes, with all supporting data archived with the method file as
specified in the laboratory's quality-assurance manual.  These files must
be available to data users who may need to investigate unexpected
data-quality changes.  Data changes resulting from recalculation or
reinterpretation of old analytical information must also be fully
documented in the laboratory-method file and be available for data users.

Process for Obtaining Approval to Publicly Release Data Produced Prior to
Water-Quality Analytical Method Approval:
The originating laboratory must submit a memorandum to the Chief, Office of
Water Quality, that fully documents the comparability of the subject data
to data quality produced by the method after approval.  All relevant
aspects of the comparability must be addressed, including precision, bias,
holding times, detection limits, data qualifiers, analytical method used,
and comparability of all relevant standard operating procedures.  The time
period involved and all sample identification criteria must be specified
and justified.  Sufficient data must be included to verify all relevant
data-quality indicators such as described above, along with other
indicators that are relevant to the data quality of the specific method.

The memorandum and supporting data and interpretations must be
peer-reviewed by at least two experts external to the laboratory who are
familiar with the specific method.  The reviewers' comments and responses
must be included with the submission memorandum.

Acceptance of the laboratory's request for release of these data to the
public will be documented by memorandum from the Chief, Office of Water
Quality, to the originating laboratory.  This approval must be received by
the originating laboratory prior to release of these data in databases that
are directly accessible by the public, including NWISWeb.

Process for Obtaining Approval for Removing Data from NWISWeb and Changing
Data Qualifiers in the NWIS Database:
The originating laboratory must submit a memorandum to the Chief, Office of
Water Quality, describing, in detail, which data are to be removed,
including constituents, parameter codes, method codes, dates, and sample
identification codes.  A description of the data-quality problem must be
included with relevant examples.  Typically, individual constituents may be
deleted from a method, particular matrices may be found that perform
unacceptably in the method, or an entire method may be found to produce
unacceptable data.

Acceptance of the laboratory's request for removing data from NWISWeb and
changing data qualifiers and flags in NWIS to show a change to unapproved
status will be documented by memorandum from the Chief, Office of Water
Quality, to the originating laboratory.  This approval must be received by
the originating laboratory prior to removal of these data from NWISWeb.

(See attached file: OWQ_04.01_revised_98.05.pdf)
Timothy L. Miller, Chief, Office of Water Quality
U.S. Geological Survey
412 National Center
Reston, VA  20192
703/648-5722 (fax)