United States
              Environmental Protection
              Agency
Environmental Monitoring
Systems Laboratory
Las Vegas, NV 89193-3478
              Research and Development
EPA/600/SR-93/242  April 1994
EPA      Project  Summary
              Assessment  and Remediation  of
              Contaminated  Sediments
              (ARCS)  Program—Quality
              Assurance  Program  Plan
              Brian A. Schumacher
               The quality assurance (QA) policy of
              the  U.S. Environmental Protection
              Agency (USEPA) requires every moni-
              toring and measurement project to have
              a written and approved quality assur-
              ance program and project  plan. The
              purpose of this quality assurance pro-
              gram plan is to  specify the policies,
              organization, objectives, and the qual-
              ity evaluation and quality control (QC)
              activities needed to achieve the data
              quality requirements of the Assessment
              and Remediation of Contaminated Sedi-
              ments (ARCS) Program. These specifi-
              cations are used to assess and control
              measurement errors that may enter the
              system at various phases of the pro-
              gram,  (during sampling,  preparation,
              and analysis).
               This Project Summary was developed
              by EPA's Environmental Monitoring and
              Systems Laboratory, Las Vegas, NV, to
              announce key findings of the research
              project that is fully documented in a
              separate report of the same title (see
              Project Report ordering information at
              back).

              Project Description
               The  1987 amendments to  the Clean
              Water Act,' Section 118(c)(3), authorize
              the USEPA  Great Lakes National Pro-
              gram Office (GLNPO) to coordinate  and
              conduct a 5-year study and demonstra-
              tion project relating to the control and re-
              moval  of toxic pollutants in  the Great
              Lakes, with emphasis on removal of toxic
              pollutants from bottom sediments. Five ar-
eas were specified in the Clean Water Act
as requiring priority consideration in locat-
ing and conducting demonstration projects:
Saginaw Bay, Ml; Sheboygan Harbor, Wl;
Grand Calumet River, IN; Ashtabula River,
OH; and Buffalo River, NY. In response,
GLNPO has initiated the ARCS Program.
ARCS  is an integrated  program for the
development and testing of assessment
and remedial action  alternatives for con-
taminated sediments.
  The overall objectives of the ARCS pro-
gram are to:  (1) assess the  nature and
extent of bottom sediment contamination
at selected Great Lakes Areas of Concern
(AOCs), (2) evaluate and demonstrate re-
medial  options, including removal, immo-
bilization, and advanced treatment tech-
nologies, as well as the "no action" alter-
native,  and (3) provide guidance  on the
assessment of contaminated  sediments
and the selection and implementation of
necessary remedial actions in the AOCs
and other locations in the Great Lakes.
  To accomplish the objectives  of the
ARCS program, two committees, one non-
technical workgroup, and three technical
workgroups were established. The names
of the individual workgroups and their ba-
sic responsibilities are:
  Management Advisory Committee: Ad-
vises the GLNPO Director on their per-
ceptions of  the overall  progress of the
ARCS 'program and  reviews annual work
and funding plans for the ARCS program.
  Activities Integration Committee: Over-
sees the ARCS program,  including the
technical  activities of each of the

-------
workgroups, develops and coordinates the
QA/QC program, and coordinates the data
management activities of the ARCS  pro-
gram.
   Toxlcity/Chemlstry Workgroup:  As-
sesses the current nature  and extent of
contaminated sediment problems by study-
Ing the chemical, physical,  and biological
characteristics of contaminated sediments
and their biotic communities, demonstrates
cost-effective assessment  techniques at
the priority consideration areas that can
be used  at other Great Lakes AOCs, and
produces three-dimensional maps show-
Ing the distribution of contaminated sedi-
ments in the priority areas.
   Risk Assessment/Modeling Workgroup-.
Assesses the current and future hazards
presented by the contaminated sediments
to all  biota (aquatic, terrestrial, and hu-
man) under the "no action" alternative and
other remedial  alternatives at the priority
consideration areas, as well as develops
a ranking scheme for site comparison.
   Engineering/Technology Workgroup:
Evaluates and tests available removal and
remedial technologies for  contaminated
sediments, selects promising technologies
for further testing, and performs field dem-
onstrations on as many of the promising
technologies as possible.
   Communication/Liaison Workgroup:
Facilitates the flow of Information from the
technical workgroups and the  overall
ARCS program to the interested public
and provides feedback from the public to
the ARCS program  on needs, expecta-
tions, and perceived problems.
   Expertise  for the three  technical
workgroups (toxicity/chemistry, risk assess-
ment/modeling, and engineering/technol-
ogy) was sought from numerous federal
and state government agencies (USEPA,
U.S. Army Corps of Engineers, U.S.  Fish
and Wildlife Service,  National Oceano-
graphlc and Atmospheric Administration,
Bureau Of Mines, Illinois Natural History
Survey,  New York State Department of
Environmental Conservation), universities
and colleges (Wright State  University,
Michigan State University,  University of
Michigan, State University College of New
York at  Buffalo,  Memphis  State  Univer-
sity, University of Minnesota, Saginaw Val-
ley State College, University of California
at  Santa Barbara, and  private industry
(Lockheed Engineering & Sciences Com-
pany,  Science Applications International
Corporation, and Battelle-Marine Sciences
Laboratory).
   Further discussion of  the primary re-
sponsibilities,  including  sampling  and
analyses, to be performed by the three
technical workgroups is presented in the
following text.
Toxicity/Chemistry Workgroup
  Four  different types of sampling sta-
tions were established for the sediment
toxicity testing by the Toxicity/Chemistry
(T/C) workgroup: reconnaissance stations,
master stations, priority master stations,
and extended priority master stations. A
brief description of the site selection crite-
ria, sampling process, and analyses per-
formed on each station type follows.
  Sampling locations for  the reconnais-
sance stations were selected to give the
greatest possible coverage of the entire
AOC and to obtain  a zone of intensive
sampling around a known  "hot spot". Site
coordinates were obtained using the Lo-
ran C navigation or the global positioning
system. Samples were obtained  using a
Vibra-core* unit. Observations of sediment
color, texture, smell, and layering were
performed on-site. Subsamples of approxi-
mately 61-cm (2-ft) intervals  were col-
lected, placed in 4 L polyethylene bottles,
kept on ice in the field, and stored at the
laboratory at 4°C.  Indicator parameters,
including ammonia, conductivity, metals,
Microtox™  bioluminescence  assay,
organohalogens, pH, sediment grain size
fractions, solvent extractable residue, to-
tal solids, volatile solids, and total organic
carbon (TOC), were analyzed in various
media  (pore water, elutriate, and/or sol-
ids).
  In principle, the  indicator  parameters
correlate with other measurements of con-
tamination and toxicity. Therefore, use of
the indicator parameters allow the detailed
analyses from the few master stations to
be extrapolated throughout the site, based
on  correlations between  reconnaissance
and master station data. Information from
these analyses  and from profiling  data
obtained during  the reconnaissance sur-
vey will be used to prepare three-dimen-
sional contamination maps  during the post-
survey phase.
  The locations of the master stations (in-
cluding  the  priority  and  extended priority
master stations) were selected based on
the availability of historical sediment con-
taminant concentration data and contami-
nant maps  from each AOC, input  from
local authorities, and a desire to provide
some degree of complete geographic cov-
erage in each AOC. Stations were usually
positioned along the sides of the dredged
shipping channel since these shallow  ar-
eas are usually  the location of sediment
deposition  zones.  Collection of the bulk
sediment sample was performed using ei-
ther a Van Veen or Ponar grab sampler.
* Mention of trade names or commericial products does
  not constitute endorsement or recommendation for
  use.
Approximately 15 L of sediment at master
stations and approximately 120 L of sedi-
ment at priority master stations was col-
lected. The sediment from the grabs was
transferred and composited in 5-gal plas-
tic bag-lined buckets.
  Upon completion of the sampling effort,
the sediments were transported to shore
and  homogenized. Homogenization  con-
sisted  of mixing the sediments in a ce-
ment mixer for 15 min. Once the sediment
was determined to  be visually  homoge-
neous, the sample was transferred to la-
beled,  high-density  polyethylene bottles.
A 5-cm headspace was left  in each bottle
to allow for  later sample  homogenization
at the  analytical laboratories. The bottles
were stored  on ice in the field and in walk-
in  coolers at 4 ± 2° C in the dark at the
analytical laboratories.
  Chemical  analyses for all master sta-
tion (including priority and extended prior-
ity  master  stations) samples included:
metals, pH,  acid volatile sulfides, meth-
ylmercury, tributyltin, pesticides, polychlo-
rinated biphenyls (PCBs), polynuclear aro-
matic hydrocarbons (PAHs), dioxins/furans,
and  total organic carbon. Analyses were
performed using standardized, approved
EPA methods. Where standardized  EPA
methods do not exist, written  standard
operating  procedures or published refer-
ences  for the method were provided by
the analytical laboratory.
  To examine the toxicity (actual and po-
tential) of the sediments, pore waters, and
elutriates to  living organisms in the Great
Lakes,  numerous  bioassays were  per-
formed on the sediments from the various
levels  of master stations. At each master
station, a tiered testing approach was  used
to determine the toxicity of the sediments.
Tier I  testing focuses on  acute toxicity
testing   using   Daphnia    magna,
Ceriodaphnla dubia, Chironomus riparius,
Chironomus  tentans,   Selenastrum
capricomutum, and  Microtox™, benthic
community  structure, and mutagenicity
testing while Tier II focuses on partial life-
cycle toxicity  employing  Hyalella azteca
assays. Tier III testing focuses primarily
on   full    life-cycle   toxicity   and
bioaccumulation using Hyalella azteca and
Pimephales promelas. Appropriate water
quality parameters were monitored.
   Priority  master stations were  selected
from the master stations to represent  sedi-
ments with a wide range in the degree of
contamination in each AOC. These sta-
tions underwent the same testing as the
master stations. Additionally, the following
comparative bioassays were performed:
Microtox™,  Selenastrum capricomutum,
Daphnia magna, Hyalella azteca, Lemna
minor,  Pimephales promelas,  Hydrilla

-------
verticillata,  Diaporeia  sp,  Hexagenia
limbata, Panagrellus redivivus, and indig-
enous bacterial enzyme function. This ad-
ditional suite of bioassays was used to
assist in the selection of optimal sediment
toxicity test assays for a given contami-
nant group (PAHs), provide comparisons
with  the  International Joint Commission
recommended test battery, and aid in the
determination of biologically significant con-
taminant levels in "grey" areas where con-
taminants are likely to produce some acute
and chronic toxicity effects.
  Most, if not all, of the sediments se-
lected as  priority  master stations under-
went a bioaccumulation assay using
Pimephales promelas. If the potential ex-
ists for the bioaccumulation of a contami-
nant or suite of contaminants identified in
the sediment, the priority master station
was designated as  an extended priority
master station and the fish tissue under-
went  analysis  for  the  suspected
bioaccumulated contaminants.  The ex- •
tended priority master station sediments
were areas of high contaminant levels (a
"hot spot") in each AOC.
  Fish tumor and abnormality identifica-
tion  on' the brown  bullhead (Atnelurus
nebulosus) were  also performed as part
of the T/C workgroup testing program. The
brown bullhead has been selected as the
primary fish due  to  its intimate contact
with the bottom sediments. Surveys were
conducted in the  Buffalo, Ashtabula, and
Saginaw Rivers to determine the incidence
of external abnormalities and internal tu-
mors.

Risk Assessment/Modeling
Workgroup
  One of the primary objectives of the
Risk Assessment and Modeling (RA/M)
workgroup is to perform hazard evalua-
tions. The phrase "hazard evaluation" re-
fers to the overall  evaluation of impacts to
all receptors of concern resulting from ex-
posure to sediment contaminants and con-
sists of several discrete assessments. The
ultimate purpose of the hazard evaluation
is to determine the existing and future
health risks  and effects (carcinogenic, re-
productive, systemic effects, community
structure impacts)  presented to human and
environmental receptors (aquatic, avian,
mammalian) from direct or indirect con-
tact with sediment contaminants under dif-
ferent remedial options. The hazard evalu-
ation is comprised of four assessments:
exposure, human  health risk, aquatic haz-
ard, and wildlife hazard assessments.
  Two levels of evaluation will be exam-
ined, baseline and comprehensive hazard
evaluations. Baseline human health haz-
ard  evaluations will  be  performed for  all
five AOCs and will be developed from
available  site-specific  information.  The
baseline hazard evaluations describe the
hazards to receptors under present site
conditions or the  "no action"  alternative.
This baseline assessment will examine all
potential pathways by which humans may
incur risk from exposure to sediments at a
given location.
  Comprehensive hazard evaluations will
be  performed for the Buffalo River and
Saginaw River AOCs. These evaluations
describe the  hazards to receptors under
different remedial  alternatives. The reme-
dial alternatives include examining selec-
tive removal  or capping of  "hot spots",
source control,  or dredging of an entire
river, among others. Additionally, the com-
prehensive risk assessment will examine
risk from losses of selected  remedial al-
ternatives.
  Sampling consisted predominantly of the
collection of samples to support the mini-
mass balance/synoptic surveys on  the
Buffalo and Saginaw Rivers. These ef-
forts included the collection of the water
column samples,  simultaneous measure-
ments of river discharge and associated
water quality parameters, and sampling of
fish populations. For the Buffalo River sys-
tem, sampling of  combined sewer outfall
(CSO) discharges were performed.  Two
additional river characterization studies
(sediment  transport  and  sediment
resuspension potential studies) were con-
ducted on the Buffalo River AOC.
  The basic goal in the sampling design
for the  mini-mass balance/synoptic sur-
veys  is to collect information about the
river system during several periods of low
flow (or quasi-steady state) conditions as
well as during at least one high flow event
(after a major storm system has passed
through the AOC or during the spring snow
melt). These data provide information on
the relative importance and amplitude of
point and non-point pollutant sources to
the AOC on both a temporal and a spatial
scale. These same data also serve  as a
primary information source for the mass-
balance, near-field dispersion, far-field dis-
persion, and food chain models to be used
by  the RA/M workgroup. Samples were
collected  from  fixed stations (six in the
Saginaw River AOC and 7 within the Buf-
falo River AOC) for all sampling events to
measure  pollutant influxes to the AOC,
ambient  concentrations within the AOC,
and effluxes to the lake, harbor, or bay.
   Measurements  of the river flow condi-
tions (flow velocity and direction, sediment
load, thermal stratification, etc.) and water
quality parameters (pH, conductivity, tem-
perature,  dissolved oxygen, chlorophyll-a
content) were made simultaneously with
the collection of the water column samples
using a variety of automated measure-
ment systems including:   the Sea-Bird®
Model SBE-25  Sealogger,  Sea-Bird®
SEACAT SBE-16® recorder fitted  with a
Sea  Tech  Transmissometer,  HydroLab
Surveyor II®,  LI-COR® system,  March-
McBirney Model 301 Flow Velocity Meter
and/or Price  and Weathermeasure  cur-
rent  meters.
  Under  both high  and low flow condi-
tions, numerous measurements were taken
throughout the AOC to determine dissolved
contaminant concentrations in the water
column and on the suspended sediment.
Contaminants measured  in  the Buffalo
River included: total PCBs, DDT, dieldrin,
chlordane, benzo(a)pyrene,  benzo(a)-
anthracene, benzo(b)fluoranthene,  benzo-
(k)fluoranthene, chrysene, Pb, and Cu. The
contaminants analyzed in water and par-
ticulate samples collected in the Saginaw
River AOC included:  total PCBs, Pb, Fe,
Cu,  and  Zn.  Conventional water  quality
parameters of sulfides, alkalinity, hardness,
chlorides, TOC, dissolved oxygen content,
and total suspended solids were also mea-
sured in  both AOCs. Analyses were per-
formed using standardized, approved EPA
methods. Where standardized EPA meth-
ods  do not exist, written standard operat-
ing  procedures or published  references
for the method were provided by the ana-
lytical laboratory.
   Fish samples were collected at both the
Buffalo and Saginaw River AOCs to sup-
port  the food  chain modeling efforts. Fish
were collected throughout the entire AOCf
Carp (Cyprinus carpio) was  collected as
the primary fish in the Buffalo River while
walleye  (Stizostedion  vitreum)  was
sampled  in the Saginaw River. Carp were
chosen to be  sampled in the Buffalo River
due  to their abundance and  representa-
tiveness  of the river's  bottom feeders.
Walleye  were selected in the Saginaw
River AOC because of its abundance, the
importance of the walleye fishery at the
AOC,  and past  use  of  the walleye in
bioaccumulation studies.
   The sediment transport  studies  involve
the determination  of the resuspension po-
tential of the bottom sediments. The sam-
pling strategy involves collecting samples
and  testing resuspension potential through-
out  the AOC. The primary consideration
was to perform tests at sites with muddy
bottom sediments since these sediments
are most easily resuspended during natu-
ral high flow events.
   The second river characterization study
involves the collection of total suspended
solids data and other limnological  param-
eters, such as water temperature,  con-
ductivity, and velocity, during high flow

-------
events In the river. These data will be
used  in the calibration of hydrodynamic
and sediment transport models. Collection
efforts were performed throughout the Buf-
falo River AOC during an  event  large
enough to initiate bottom scour of the river
bed.

Engineering/Technology
Workgroup
  Sampling for the Engineering/Technol-
ogy (E/T) workgroup consisted of gather-
ing enough bulk sediments from one or
two locations within an AOC to supply all
the bench scale  remediation processes
with the "same" initial sediment. The sedi-
ments collected were grossly contaminated
with a given class or classes of  contami-
nants. Site selection was based on his-
torical data, the results of the sediment
characterization from the  T/C workgroup
efforts, and discussions among members
of the three technical workgroups.
  Site locations were  marked  on  U.S.
Army  Corps of Engineers sounding charts
after determination via triangulation.  Bulk
samples  were collected from the toe of
the channel using a  crane barge bucket
operation. Sediments were then  scooped
or shoveled from the bucket, working from
top to bottom, to fill approximately twenty
5-gallon  plastic  buckets.  Sample
compositing and homogenization was per-
formed using a cement mixer. Sediments
were  deemed homogeneous by visual in-
spection of texture, color, and water con-
tent. After homogeneity has been obtained,
samples were stored at 4° C in  the  dark
prior to analysis.
  In general, the remedial  processes in
the ARCS program are aimed at the  deg-
radation of organic compounds, such as
pesticides,  RGBs, and PAHs. However,
several remedial processes, such as siev-
ing and cycloning, froth flotation, gravity
separation, and magnetics, were used to
demonstrate  remediation  possibilities for
sediments contaminated with heavy  met-
als. To select the remedial processes for
the organic contaminants, a literature re-
view was performed of existing remediation
technologies. Upon completion of the lit-
erature review, several remedial processes
were  selected by the E/T workgroup for
two different levels of testing, the bench-
scale  tests and pilot-scale demonstrations.
  To  determine if a remediation process
has been successful, contaminant  con-
centrations must be determined prior to
and after the remediation has been com-
pleted. Therefore, testing was performed
on both untreated and treated sediments,
as  well  as  water  and  oil fractions
(remediation by-products) depending upon
the process  being tested.  Parameters
monitored by the E/T workgroup included:
metals, pH, pesticides, PCBs, PAHs, oil
and grease, total organic  carbon, mois-
ture content, conductivity, and total vola-
tile solids. Analyses were performed using
standardized,  approved EPA  methods.
Where standardized EPA methods do not
exist,  written  standard operating  proce-
dures or  published  references  for the
method were provided by the analytical
laboratory.

Quality Assurance Program
  The data collection criteria provide a
balance between constraints of time and
cost and  the quality of data necessary to
achieve the ARCS program research ob-
jectives. The ARCS quality assurance pro-
gram  plan (QAPP) is designed to accom-
plish the following objectives:

  • Establish the QA/QC criteria used to
    control and assess data collection in
    the ARCS program,
  • Provide comparable sampling, prepa-
    ration, and analytical methods,
  • Utilize assessment samples and pro-
    cedures to verify the  quality of the
    data,
  • Perform field and on-site laboratory
    system audits to ensure that all ac-
    tivities are properly performed  and
    that discrepancies when identified are
    resolved, and
  • Evaluate  the data and document the
    results in a final QA report to GLNPO
    management.
  The raw data for the ARCS  program
was collected during three major opera-
tional  phases  consisting of sediment map-
ping,  sampling, and analysis. A certain
amount of data measurement uncertainty
is expected to enter the system  at each
phase. The sampling population itself is a
source of confounded uncertainty that is
extremely difficult to quantify.
  Generally,  the data quality objectives
(DQOs)  for the ARCS program  encom-
pass the overall allowable uncertainty from
sample measurement and  from the sam-
pling  population that the data users are
willing to accept in the analytical results.
Because of the many confounding sources
of uncertainty, overall DQOs for the ARCS
program  are  not  described herein.  This
QAPP focuses on the definition, imple-
mentation, and assessment of Measure-
ment  Quality  Objectives  (MQOs) that are
specified for  the entire sample  prepara-
tion and  analysis phases of data collec-
tion as well as for the verification of the
field sampling phase. The MQOs are de-
fined  according to the following six  at-
tributes:

  • Detectability - the lowest concentra-
    tion of an  analyte that a  specified
    analytical procedure can reliably de-
    tect,
  • Precision - the  level  of  agreement
    among multiple measurements of the
    same characteristic,
  • Accuracy - the difference between an
    observed value  and the "true" value
    of the parameter being measured,
  • Representativeness - the degree to
    which the data collected accurately
    represents the population of interest,
  • Completeness - the quantity of data
    that is successfully collected with  re-
    spect to the amount intended in the
    experimental design, and
  • Comparability -  the similarity of data
    from different sources included within
    individual or multiple data sets; the
    similarity of analytical methods and
    data  from  related projects across
    AOCs.
  Initial MQOs  were established by the
principal laboratories performing a given
type of measurement (inorganic or organic
analyses, bioassays) after discussion and
approval by the members of the T/C, and/
or E/T  workgroups. In most  cases, the
initial  proposed QA  program and MQOs
are equivalent  to the QA program rou-
tinely  implemented at the analytical labo-
ratory. Upon the initiation of the formal QA
program within  the  ARCS  program, the
existing MQOs  were either accepted or
modified with additional requirements to
ensure data quality in the ARCS  program.
The resultant MQOs were then applied to
all  parameters  in the process  of being
analyzed and to all future analyses.
  To  produce data of known quality, par-
ticipating laboratories are required to ana-
lyze certain types of QC samples that are
known to the laboratory staff and that can
be  used by the analysts to identify  and
control  analytical measurement uncer-
tainty. Each QC sample has certain speci-
fications that must be met before data for
that parameter  is considered acceptable.
These specifications include acceptance
limits  and frequency of sample use  re-
quirements. The various  types of  QC
samples for the chemical  and  physical
parameters, as well as the water quality
parameters run  in conjunction with bioas-
says and fish bioaccumulation studies, in-
cluded:   analytical replicaites, field dupli-
cates, reagent  blanks,  reference materi-

-------
als, matrix spikes, matrix spike duplicates,
surrogate spikes for organic analyses, and
ongoing calibration check samples. Addi-
tionally, acceptance criteria and limits have
been established for initial instrument cali-
bration and method detection limits, where
appropriate. QC samples that are unique
to bioassays, fish  bioaccumulation stud-
ies, and/or mutagenicity testing included:
reference toxicants, reference sediments,
pre-exposure sampling,  spontaneous  re-
version rates, and  strain integrity testing.
Secondary confirmation of organism iden-
tification by an independent scientist, ana-
lytical replicates, and relative abundance
of species comparisons within the AOC
and between independent identifications
are required as QA/QC checks  during
benthic community structures determina-
tions.

Quality Assurance
Implementation
  The quality assurance program is imple-
mented through on-site systems audits of
both laboratory and field operations, inde-
pendent  assessments, and  other  proce-
dures used to control and assure the qual-
ity of the  data being collected. Verification
of these data will be accomplished through
a series of manual checks for success in
meeting the ARCS program established
MQOs. All data will be reviewed for the
following  items:

  • completeness of the submitted dataset
    in terms of missing data,
  • completeness of the submitted data
    in terms of  the completeness quality
    assurance objective,
  • formal submission of the data as indi-
    cated by signatures of the PI and
    laboratory QA officer,
  • logbooks, in particular to determine
    holding time violations,
  • raw data including sample weights,
    extract volumes, dilution or  concen-
    tration  factors,  instrument readings
    (e.g.,  chromatograms, quantification
    reports,  etc.), and dates  of analysis,
    where appropriate,
  • proper frequency of use and success-
    ful  completion of  the  established
    MQOs  for QC samples on  a  dated
    per batch basis,
  • method detection limits and their de-
    terminative  data and  dates of  deter-
    mination,
  •  calibration data on  a per instrument
    per analyte  basis,
   •  in-house performance a.udit and other
     QA reports as specified in the  sub-
     mitted QAPPs, and
   •  a discrepancy report indicating at what
     point during the laboratory operations
     the formal ARCS QA  program  was
     initiated and providing a. discussion of
     the QA program at the laboratory prior
     to the  institution of the  ARCS overall
     QA program.

 Data Quality Assessment and
 Reporting
   The assessment of detectability (detec-
 tion  limits)  is accomplished on a param-
 eter basis at two different levels, compli-
 ance with ARCS specified MDLs and cal-
 culation of actual IDLs.  The final results
 will be grouped in tabular  form to allow
 comparisons among the values  for any
 parameter of interest.
   A statistical evaluation procedure that
 has  been  developed by the ARCS QA
 staff is applied to the data to assess pre-
 cision as a function of confounded data
 collection  uncertainty. An additive step-
 function model is used, where an observed
 value of any sediment, elutriate, or water
 characteristic is considered as the sum of
 the "true" accepted value  and an error
 term. Precision is evaluated for  each vari-
 ance segment of the range  of concentra-
 tion for a given analyte.
   The assessment of accuracy is based
 on the ongoing calibration check samples
 and the use of certified reference materi-
 als, standard reference materials, or stan-
 dards for the inorganic and organic analy-
 ses  while  for  the  bioassays  and  fish
 bioaccumulation studies, the assessment
 of accuracy rely upon the  use of refer-
 ence toxicants and the reference sedi-
 ment. The recoveries of matrix and surro-
 gate spikes for the inorganic and organic
 analyses can also be used in the assess-
 ment of accuracy.
   One aspect of sampling representative-
 ness  is assessed by comparing the indi-
 vidual site  locations and AOC  coverage
 with the locations and expected coverage
 DQOs. Representativeness of the homog-
 enization and subsampling procedures  at
the analytical laboratories  may  be as-
 sessed using precision  estimates for the
 analytical and field replicate  samples.
   Upon completion of the ARCS program,
a  comparison will  be made among the
 laboratories that will focus on method dif-
ferences, QC sample results, laboratory
effects, and other QA features of the pro-
gram to  assess program comparability.
Summary statistics  will be used to collate
individual values into pooled groups  that
 enable the data-users to discern trends
 within the overall ARCS program.
   Field sampling  completeness  is  as-
 sessed  by comparing the actual number
 of stations collected to the number re-
 quested during the design phase of the
 ARCS program.  Completeness  of  the
 sample preparation and analytical phases
 is calculated as the number of analyses
 passing the QA requirements divided by
 the number of analyses  performed at a
 given laboratory.
   Each participating laboratory is required
 to produce at least one written report to
 document their QA/QC activities as well
 as several oral laboratory updates at the
 all-hands meetings to be planned through-
 out the duration of the ARCS  program.
 Communications among the various par-
 ticipants in the ARCS program has been
 maintained through conference calls,  site
 visits, releases of preliminary draft data,
 and all-hands and workgroup meetings.
   A final written summary of the QA ac-
 tivities and final results is required from
 each participating ARCS program labora-
 tory, and  should accompany the submis-
 sion  of the laboratory  QA approved
 dataset. Other periodic QA reports will be
 submitted to the  ARCS QA officer,
 workgroup chairs,  and GLNPO staff as
 specified in the laboratory's QAPP.
   The ARCS QA staff will produce a docu-
 mented sample/data tracking system such
 that hardcopy and electronic forms of the
 database can be easily located, identified,
 and collated for use and  distribution by
 the staff at GLNPO.

 Quality Assurance/Quality
 Control  of Historical Databases
   A QA/QC evaluation scale was  devel-
 oped for the RA/M workgroup to allow for
 the objective assessment of historical data
 used in the risk assessments and model-
 ing efforts. Evaluation  scales were  pro-
 duced for inorganic and organic chemistry
 analyses. The verification process will in-
 clude QA/QC compliance checking for ac-
 curacy,  precision,  spike  analyses,
 blanks, detection limits, calibration (initial
 and ongoing), and holding times, as well
 as other QA/QC concerns that can affect
the integrity of the sample and  resultant
data. The  final evaluation of a dataset is
 presented  as a combination of a number
value and  a flag list. The numerical value
for a given parameter or suite of param-
eters is assigned based on the successful
completion  of  each  required QA/QC
sample or measurement. A list of appro-
priate flags are attached to each numeri-
cal rating to indicate where discrepancies

-------
exist between the laboratory data and the
acceptance limits of the required QA pro-
gram. Two different interpretation can be
made using the final ratings. The first in-
terpretation  is based  upon the  formal
ARCS QA program while the second in-
terpretation is based upon the "full poten-
tial" of the submitted dataset in which dif-
ferences between the ARCS QA program
and the QA program implemented during
the generation  of the data can  be ac-
counted.
Data Management System
  The Ocean  Data Evaluation System
(ODES) has been used as the final data-
base repository for the ARCS program.
ODES was designed to support the deci-
sion  making processes associated with
marine/water monitoring programs. Since
ODES was originally designed for saltwa-
ter systems, some modification of data
fields may be required to adapt the sys-
tem for the fresh water environment ana-
lyzed in the ARCS program.
  ODES is comprised of three separate
components: the ODES database, ODES
reporting and graphical tools, and ODES
menu system. Through the ODES menu
system a  user may access information
stored in the ODES database and use the
ODES tools to produce analytical reports.
The ODES database combines source in-
put information with river, harbor, and bay
environmental information including bio-
logical data, sediment pollutant data, wa-
ter quality  data, and field sampling data.
                                                                   •&V.S. GOVERNMENT PUNTING OFFICE: MM - S5MC7/IM24S

-------

-------
   Brian A. Schumacher is with the Environmental Monitoring Systems Laboratory,
     Las Vegas, NV 89193-3478.
   Brian A. Schumacher is the EPA Project Officer (see below).
   The complete report, entitled "Assessment and Remediation of Contaminated
     Sediments (ARCS) Program—Quality Assurance  Program Plan," (Order No.
     PB94-144581/AS; Cost $27.00, 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 Project Officer can be contacted at:
           Environmental Monitoring and Systems Laboratory
           U.S. Environmental Protection Agency
           Las Vegas, NV 89193-3478
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
     BULK RATEE
POSTAGE & FEES PAID
         EPA
   PERMIT NO. G-35
EPA/600/SR-93/242

-------