United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Las Vegas NV 89193
Research and Development
EPA/600/S8-87/026 Sept. 1987
vvEPA Project Summary
National Surface Water Survey:
Western Lake Survey (Phase I
Synoptic Chemistry) Quality
Assurance Plan
M. E. Silverstein, S. K. Drouse, J. L. Engels, M. L. Faber, and T. E. Mitchell-Hall
The purpose of the National Surface
Water Survey of the National Acid
Precipitation Assessment Program is to
evaluate the present water chemistry of
lakes and streams, to determine the
status of certain biotic resources, and
to select regionally representative sur-
face waters for a long-term monitoring
program to study changes in aquatic
resources. The Western Lake Survey
Phase I is part of the National Surface
Water Survey.
The quality assurance plan and the
analytical methods used during Phase I
of the Western Lake Survey are based
on those used during Phase I of the
Eastern Lake Survey; analytical labora-
tory methods are identical for the two
surveys, but some of the field laboratory
methods were modified for the West.
Sampling protocols are significantly
different in that ground access as well
as helicopter access was used to collect
samples in the West.
Quality assurance requirements and
procedures that are unique to the
Western Lake Survey Phase I are
treated in detail in the Western Lake
Survey Quality Assurance Plan. Quality
assurance requirements and procedures
that were adopted verbatim from the
Eastern Lake Survey Phase I are
referenced in the Western Lake Survey
Quality Assurance Plan and are dis-
cussed in detail in the Eastern Lake
Survey Phase I Quality Assurance
Plan.
This Project Summary was developed
by EPA's Environmental Monitoring
Systems Laboratory, Las Vegas, NV, to
announce key Information that Is pre-
sented In a separate quality assurance
plan ot the same title (see ordering In-
formation at back).
Introduction
The National Acid Precipitation Assess-
ment Program (NAPAP) was initiated at
the request of the Administrator of the
U.S. Environmental Protection Agency
(EPA) to evaluate the extent of the effects
of acidic deposition on aquatic resources
within the United States. When it became
apparent that existing data could not be
used to quantitatively assess the present
chemical and biological status of surface
waters in the United States, the National
Surface Water Survey (NSWS) program
was incorporated as part of NAPAP to
obtain that information. The National Lake
Survey (NLS) component of NSWS com-
prises Phase I Eastern Lake Survey
(ELS-I), Phase I Western Lake Survey
(WLS-I), and Phase II Temporal
Variability.
ELS-I, a synoptic survey of the chemistry
of 1,612 representative lakes in the
Southeast, Northeast, and Upper Mid-
west, was conducted to obtain a regional
and national data base of water-quality
parameters that are pertinent to evalu-
ating the effects of acidic deposition.
Parallel to ELS-I, the WLS-I portion of
NSWS is a synoptic survey of 757 lakes
-------
in the West. WLS-I sampling is conducted
during fall overturn when chemical vari-
ability within a lake is expected to be
minimal as a result of circulation within
the water column. WLS-I is designed to
meet the following objectives for desig-
nated regions of the West:
Determine the percentage (by num-
ber and area) and location of lakes
that are potentially susceptible to
change as a result of acidic deposi-
tion and that have low acid-
neutralizing capacity (ANC).
Investigate the relationships among
water chemistry, regional acidic
deposition patterns, land use, phy-
siographic features, lake morphology,
and basin geometry within and
among regions.
Identify smaller subsets of repre-
sentative lakes for more intensive
sampling in future surveys.
WLS-I was designed to provide statis-
tically comparable data that could be
extrapolated, with a known degree of
confidence, to a regional or national scale.
The conceptual approach emphasized that
the data would not be used to ascribe
observed effects to acidic deposition
phenomena; rather, through comprehen-
sive monitoring activities, WLS-I would
provide information that could be used to
develop correlative, not cause-and-effect,
relationships.
The quality assurance (QA) program
and the analytical methods for WLS-I are
based on those used during ELS-I. The
analytical laboratory methods are identical
for the two surveys. Some field laboratory
methods were modified for WLS-I on the
basis of ELS-I experience and on the
basis of constraints that resulted from
the special geographic limitations associ-
ated with the high-altitude lakes in the
West. New field laboratory protocols also
were added to accommodate changes
between ELS-I and WLS-I field sampling
methods. Sampling protocols are signifi-
cantly different in that ground access as
well as helicopter access is used to collect
samples for WLS-I (Bonoff and Groeger,
1987). The sampling protocols differ from
those used during ELS-I because 455 of
the WLS-I lakes are within designated
wilderness areas that are closed to heli-
copter access. In order to observe the
guidelines and regulations set forth in
the Wilderness Act, almost all lakes
located within wilderness areas that have
been selected for sampling must be
sampled by ground crews of the U.S.
Department of Agriculture Forest Ser-
vice. The ground crews travel to lakes on
foot or on horseback. The lakes that are
not in wilderness areas are sampled by
helicopter crews under the direction of
EPA. The ground sampling protocol devel-
oped for use in sampling the restricted-
access lakes was first evaluated in a
WLS pilot study conducted by EPA's
Region VIII office in the autumn of 1984.
A specialized calibration study is in-
cluded in WLS-I to compare the effects of
the two different sampling methods on
analytical results. The purpose of the
comparison is to derive calibration factors,
if necessary, that can be applied to data
for samples collected by ground crews so
that these data will be equivalent to data
for samples collected by helicopter crews.
This calibration study is designed to (1)
quantify the differences between the two
sampling methods, (2) quantify the effects
of holding samples for different lengths
of time prior to processing, preservation,
and analysis, and (3) quantify any signifi-
cant interlaboratory bias between the two
analytical laboratories that analyze WLS-
I samples.
Two other studies are being conducted
as part of WLS-I. The purpose of one
study, the nitrate-sulfate stability study,
is to compare sample preservation
methods and to study the effects of hold-
ing samples for different lengths of time
before preserving them. The purpose of
the second study, the Corvallis study, is
to compare results for splits of the same
sample when the splits have been
analyzed by different methods. A further
purpose of the Corvalfis study is to deter-
mine whether or not the ICP data can be
substituted in the data base if problems
arise with the standard analysis. Both
studies can provide checks on sampling,
processing, and analytical performance.
Quality Assurance Objectives
The statistical design, sampling and
analytical methods, and QA activities for
WLS-I are structured to meet specific
data quality objectives (DQOs) for the
measurement of sampling, field labora-
tory, and analytical laboratory perfor-
mance. These DQOs are designed to
facilitate checking for chemical variability
and to provide confidence levels for re-
porting population estimates.
The primary DQOs are measures of
precision (expressed as relative standard
deviation), accuracy (expressed as maxi-
mum absolute bias), and detectability
(expressed as an expected value range
and a required detection limit). These
DQOs are applied to each parameter
measured at the lake sampling site, in
the field laboratory, and in the analytical
laboratory. Table 1 summarizes the pri-
mary OQOs used for WLS-I. Certain other \
DQOs also have been considered in the
survey design. Completeness (the quantity
of acceptable data actually collected in
relation to the total quantity that is ex-
pected to be collected) is set at 90 percent
or better for all variables. Comparability
(a measure of the confidence with which
one data set can be compared to another)
is ensured by requiring that standard
procedures be used for laboratory
analyses and that a uniform set of units
be used for reporting data. The calibration
study was performed to ensure that dif-
ferences in the sampling and on-site
analytical procedures used by helicopter
crews and ground crews did not reduce
data comparability. For WLS-I, repre-
sentativeness (the degree to which data
accurately and precisely represent a
characteristic of a population) is maxi-
mized by drawing a systematic random
sample to ensure good geographical
coverage without bias (Landers et al.,
1987).
WLS-I is not intended to characterize
the chemistry of any given lake spatially
or temporally. Therefore, achieving WLS-
I objectives does not require that the only
sample taken from a lake be completely
representative of the lake. In most cases
(except for the lakes to be sampled during
the calibration study) only one sample
per lake is taken during WLS-I. A deter-
mination of whether one sample per lake
is sufficient to achieve the general ob-
jectives of NSWS Phase I, however, can
be made from estimates of "within-lake"
and "among-lakes" variances. Although
some estimates of these variances will
be made for WLS-I lakes in accordance
with the statistical sampling design, more
intensive studies of individual lakes will
be necessary to provide more complete
data on representativeness.
Although the individual sample is not
necessarily representative of the lake,
the subset of lakes sampled should be
representative of the subregional or
regional population of lakes. The sys-
tematic sampling design that was adoptee
for this survey is intended to ensurf
representativeness at this level.
Monitoring Data Quality
The lakes are selected by means of
systematic, stratified design. There ar
three stratification factors: regions, sub
regions, and alkalinity classes. Eac
stratum is an alkalinity class within
subregion within a region. In the We:
(NLS Region 4), all three alkalinity classe
are found within each of the five sut
-------
Table 1. Data Quality Objectives For Precision, Accuracy, and Detectability, Western Lake Survey Phase I
Site'
2,3
2,3
3
3
3
3
1,3
2,3
3
3
3
3
3
3
3
3
3
1,2
3
3
3
3
2
2
Parameter11
Al, Extr actable
Al. Total
ANC
BNC
Ca
cr
Conductance
DIG
DOC
F-, Total
dissolved
Fe
K
Mg
Mn
Na
AW/
N03
pH, Field
pH. Analytical
laboratory
P. Total
SiO2
SO/
True color
Turbidity
Method
Extraction with 8-hydroxyquino-
line into MIBK followed by
AA' (furnace)
AA" (furnace)
Titration and Gran analysis
Titration and Gran analysis
AA" (flame) or ICP9
Ion chromatography
Conductivity cell and meter
Instrumental (acidification, C02
generation, IR detection)
Instrumental (UV -promoted
oxidation, CO2 generation.
IR detection)
Ion-selective electrode and
meter
AA' (flame) or ICP9
AA' (flame)
AA' (flame) or ICP9
AA' (flame) or ICP8
AA' (flame)
Automated color imetry
(phenate)
Ion chromatography
pH electrode and meter
pH electrode and meter
Automated colorimetry
(phosphomolybdate)
Automated colorimetry
(molybdate blue)
Ion chromatography
Comparison to platinum-cobalt
color standards
Instrument (nephelometer)
Units
mg/L
mg/L
neq/L
neq/L
mg/L
mg/L
liS/cm
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
pH units
pH units
mg/L
mg/L
mg/L
PCI/
NTU*
Expected
Range"
0.005 - 1.0
0.005 - 1.0
-100-1.0OO
-10- 150
0.5 - 20
0.2 - 10
5- 1.OOO
0.05 - 15
0.1-50
0.01 - 0.2
0.01 - 5
0.1-1
0.1-7
0.01 - 5
0.5-7
0.01 - 2
0.01 - 5
3-8
3-8
0.005 - 0.07
0.1-25
1 -20
0-200
2-15
Required
Detection
Limits
0.005
0.005
f
t
0.01
0.01
h
0.05
0.1
0.005
0.01
0.01
0.01
0.01
0.01
0.01
0.005
0.002
0.05
0.05
0
2
Precision
Relative Standard Deviation
(RSD) Upper Limit (%f
10(Alconc.X3.01)
20 (A1 cone. O.01)
20(Alconc.5)
10 (DOC cone. <5)
5
10
5
5
10
5
5
10
±0.1'
±0.05'
10 (P cone. >O.O1)
20 (P cone.
-------
Northern Rocky
Mountains (4CJ
Pacific
Northwest (4B)
California (4A)
Central
Rocky
r/_ tj Mountains (4D)
Southern Rocky
Mountains (4E)
Figure 1.
Subregion Boundary
Subregions of the Western United States that are potentially susceptible to acidic
deposition. Western Lake Survey - Phase I.
Field Operations
Field operations are coordinated at field
bases under the supervision of an EPA
field base coordinator. One or two heli-
copter crews (each consisting of a pilot,
an observer, a sampler, and a support
person) operate from each field base
through an EPA duty officer. Each heli-
copter crew makes one or more excur-
sions to lake sites each day. Ten to fifteen
two-person ground crews operate from
each base site through a Forest Service
field manager. Each helicopter crew
samples as many as 10 lakes per day,
and each ground crew samples 1 or 2
lakes per day (or per excursion). The field
base coordinator and the field manager
coordinate the rate of lake sampling so
that no more than 30 field samples are
processed at a field base on any day. This
procedure prevents an overload of sam-
ples arriving at the analytical laboratories.
4
Each field base contains a fully
equipped mobile laboratory that is staffed
by a field laboratory coordinator, a field
laboratory supervisor, and three analysts.
The field laboratory coordinator is re-
sponsible for the overall operation of the
laboratory (e.g., sample tracking and
logistics, data, forms, safety). The field
laboratory supervisor and the analysts
are responsible for sample measurements
made at the field base and for sample
processing. If necessary, the field labora-
tory coordinator also assists with sample
processing. The aliquots prepared by the
field laboratory are described in Table 2.
All field and field laboratory measure-
ments are recorded on the appropriate
field forms and in bound log books.
Training
The WLS-I procedure for training the
helicopter sampling and field laboratory
personnel is identical to the procedure
used during ELS-I. Ground sampling
techniques call for new safety and tech-
nical considerations, so additional training
is provided for ground sampling partici-
pants. This training is carried out at Las
Vegas and at the field bases and remote
sites.
Analytical Procedures and
Internal Quality Control
WLS-I analytical procedures and in-
ternal QC provisions are identical to those
used for ELS-I. These procedures are
described fully in the ELS-I Analytical
Methods Manual (Hillman et al., 1986)
and are summarized in the WLS-I
Analytical Methods Manual (Kerfoot and
Faber, 1987).
Performance and System Audits
Performance Audit Samples
Field synthetic and field natural audit
samples are used to identify problems
affecting data quality that may occur
during sample processing, shipment, or
analysis. These problems could include
sample contamination, sample degrada-
tion, solvent evaporation, and improper
or inaccurate sample analysis.
Waters collected from Big Moose Lake
in the Adirondack Mountains, from Lake
Superior at Duluth, Minnesota, and from
Bagley Lake in the Cascade Range of
Washington State are used as natural
audit samples for the survey. The waters
of Big Moose Lake are acidic; the Lake
Superior waters represent a buffered
system; and Bagley Lake represents a
partially buffered system. Synthetic audit
samples are prepared at a central
laboratory.
The audit samples are shipped to the
analytical laboratories from the field bases
as though the audit samples were aliquot!
of routine lake samples. Every attempt i:
made to ensure that the analytical labora
tory does not recognize the audit sample.'
as different from the routine lake samples
Data are obtained from the analyses o
the field audit samples (1) to judge th<
performance of the field bases and thi
analytical laboratories in the processing
shipment, and analysis of samples am
(2) to establish a statistically valid estimat
of the overall bias and precision of th
analyses and of the stability of a typice
lake sample.
Acceptance windows are establishe
for the measurement of each audit
sample parameter by using a Student'
t-statistic test. Acceptance criteria fc
audit sample values are the same fc
WLS-I as for ELS-I.
-------
Table 2. Allquots, Containers, Preservatives, and Corresponding Analyses,
Western Lake Survey Phase I
Aliquot
(Container Volume)
Preservative and
Description
Analyses
1
(250 mL)
2
(10mL)
3
(250 mL)
4
(125 mL)
5
(500 mL)
6
(125 mL)
7
(125 mL)
Filtered, preserved with HNO3 to pH <2 Ca, Mg. K. Na. Mn, Fe
Filtered, preserved with MIBK-HQ extract Extractable A1
Filtered, no preservative
Filtered, preserved with H2S04 to pH <2 DOC.
Cl', total dissolved F.
SO/', N03', SiO2
Unfiltered. no preservative
Unfiltered. preserved with //2SO< to pH <2 Total P
pH, BNC, ANC.
conductance, DIG
Unfiltered, preserved with HN03 to pH <2 Total Al
Approximately 75 synthetic audit sam-
ples and 150 natural-water audit samples
are scheduled to be processed during
WLS-I. A statistical evaluation of the audit
data, including the setting of audit win-
dows, should provide a good estimate of
the bias and precision of the analytical
methods for each parameter. Further-
more, any change over time in analytical
results for the natural-water audit
samples without a corresponding change
in results for the other audit samples can
be attributed to lack of analyte stability.
The findings of a comparative study
between audit sample types will provide
an estimate of the true maximum holding
times allowable for each type of analyte.
System Audits
System audits (on-site evaluations) are
qualitative evaluations of field and analy-
tical laboratory facilities, equipment, and
operations such as record keeping, data
reporting, and QC procedures. Each field
base and helicopter sampling crew can
expect at least one on-site evaluation
during the course of the survey. In addi-
tion, as many of the 60 ground sampling
crews as possible will be evaluated in the
field. Each analytical laboratory partici-
pating in WLS-I can expect a minimum of
two comprehensive, on-site evaluations
during the survey.
Data Management
The purpose of the data base manage-
ment system is to assemble, store, and
edit data generated during WLS-I and
during other NSWS surveys. The data
base management system also is used to
generate basic survey results, to perform
certain statistical analyses, and to provide
data security.
The WLS-I data are stored in four major
data sets: the raw data set, the verified
data set, the validated data set, and the
final data set. The raw data set is prepared
at Oak Ridge National Laboratory (ORNL).
The field data and the analytical laboratory
data are entered into two separate data
sets which are then compared to remove
data entry errors. Data evaluation, verifi-
cation, and validation procedures provide
the input from which the verified, vali-
dated, and final data sets are prepared.
Data Evaluation and Verification
The field bases and analytical labora-
tories provide preliminary sample data to
the EMSL-LV QA group verbally, by
computer, or by telefacsimile. When the
field and analytical laboratory data are
transmitted through magnetic tapes and
the raw data are made available to the
EMSL-LV QA group, 100 percent of the
data are evaluated and verified on the
basis of available QA/QC information.
The objective of data verification is to
identify data of unacceptable quality and
to correct, flag, or target them for possible
sample reanalysis or for elimination of
the data from future data sets. This data
review process is also important in verify-
ing that contractual requirements have
been met.
After all data have been reviewed, the
analytical laboratories are requested to
submit completed copies of data reporting
forms that were incomplete when pre-
viously submitted, to submit corrections
of previously reported data, to confirm
previous results, and to reanalyze certain
samples that do not meet QA/QC criteria.
The analytical laboratories are directed to
respond promptly so that the results can
be evaluated in time for them to be useful
to the survey.
The verification procedure identifies
suspect data and corrects erroneous data.
The information is collected by the
EMSL-LV QA staff and is placed on
magnetic tapes which are sent to ORNL.
There, the raw data set is updated to
produce the verified data set.
Data Validation
The system of data validation developed
for ELS-I is duplicated for WLS-I. The
validation process begins concurrently
with the verification process. When ORNL
provides the ERL-C staff with a com-
puterized version of the verified data set
through the National Computing Center
at Research Triangle Park (RTF), North
Carolina, the validation process can be
completed.
Known relationships in aquatic chem-
istry and limnology are used to identify
intrasite sample inconsistencies within
data for a set of variables. Intersite
validation consists of comparing single
site values with values for adjacent sites
within a region. Data for groups of sites
are compared and mapped to check for
consistency. After undergoing this re-
viewing process, the data are transferred
to the validated data set.
Development of a Final Data Set
The calculation of population estimates
is difficult if the data set contains missing
values. To resolve the problems in the
validated data set that result from missing
values, a fin^l data set is prepared. This
final set is modified by averaging the field
routine-duplicate pair values that are
within desired precision limits and by
replacing analytical values determined to
be in error during validation. The values
that have been modified fqr or substituted
in the final data set are flagged with the
appropriate data qualifiers.
Related Documents
The WLS-I QA Plan described here is
the final version of the draft plans that
were written before and were modified
during WLS-I activities. Quality assurance
requirements and procedures that were
adopted verbatim from ELS-I are refer-
enced in the WLS-I QA Plan and are
described in detail in the ELS-I QA Plan
(Drouse et al., 1986). Analytical methods
-------
are described in the ELS-I and WLS-I
Analytical Methods Manuals (Hillman et
al., 1986; Kerfoot and Faber, 1987). ELS-
I recommendations that led to WLS-I
protocol changes are discussed in the
ELS-I QA Report (Best et al., 1987). Only
those QA requirements and procedures
that are specific to WLS-I are described
in the WLS-I QA Plan. A QA report that
describes the findings of the survey and
the effectiveness of this QA plan will be
issued after these factors have been
evaluated.
References
Best, M. D., S. K. Drouse, L. W. Creelman,
D. J. Chaloud, 1987. National Surface
Water Survey, Eastern Lake Survey
(Phase I Synoptic Chemistry) Quality
Assurance Report. EPA 600/4-86-011.
U.S. Environmental Protection Agency,
Las Vegas, Nevada.
Bonoff, M. B., and A. W. Groeger 1987.
National Surface Water Survey,
Western Lake Survey (Phase I
Synoptic Chemistry) Field Operations
Report. U.S. Environmental Protection
Agency, Las Vegas, Nevada.
Drouse, S. K., D.C. Hillman, L W.
Creelman, and S. J. Simon, 1986.
National Surface Water Survey, Eastern
Lake Survey (Phase I Synoptic
Chemistry) Quality Assurance Plan.
EPA 600/4-86-008. U.S. Environ-
mental Protection Agency, Las Vegas,
Nevada.
Hillman, D. C., J. F. Potter, and S. J.
Simon, 1986. National Surface Water
Survey, Eastern Lake Survey (Phase I
Synoptic Chemistry) Analytical
Methods Manual. EPA 600/4-86-OO9.
U.S. Environmental Protection Agency,
Las Vegas, Nevada.
Kerfoot, H. B., and M. L. Faber, 1987.
National Surface Water Survey,
Western Lake Survey (Phase I
Synoptic Chemistry) Analytical Methods
Manual. U.S. Environmental Protection
Agency, Las Vegas, Nevada.
Landers, D. H., J. M. Eilers, D. F. Brakke,
W. S. Overton, P. E. Kellar, M. E.
Silverstein, R. D. Schonbrod, R. E.
Crowe, R. A. Linthurst, J. M. Omernik,
S. A. league, and E. P. Meier, 1987.
Characteristics of Lakes in the Western
United States. Volume I. Population
Descriptions and Physico-Chemical Re-
lationships. EPA/600/3-86/054a. U.S.
Environmental Protection Agency,
Washington, D.C.
The information in this document has
been funded wholly or in part by the U.S.
Environmental Protection Agency under
6
contract number 68-03-3249 to Lockheed
Engineering and Management Services
Company, Inc. It has been subject to the
Agency's peer and administrative review,
and it has been approved for publication
as an Agency document.
M. E. Silverstein, S. K. Drouse. J L. Engels, M. L. Faber, and T. E. Mitchell-
Hall are with Lockheed Engineering and Management Services, Las Vegas,
NV89119.
Robert D. Schonbrod is the EPA Project Officer (see below)
The complete report, entitled "National Surface Water Survey, Western Lake
Survey (Phase ISynoptic Chemistry) Quality Assurance Plan," (Order No.
PB 87-214 862/AS, Cost: $18.95, subject to change) will be available only
from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Officer can be contacted at:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas. NV 89193-3478
-------
United States Center for Environmental Research
Environmental Protection Information
Agency Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S8-87/026
0000329 PS
yiST'iSSB'1 *GENC'
afcU"-- R»?PW
------- |