ASTM E3242-23

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: E3242 − 23
Standard Guide for
Developing Representative Background Concentrations at
Sediment Sites—Data Evaluation and Development
Methodologies
This standard is issued under the fixed designation E3242; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope site CSM that need to be considered when developing repre-
sentative sediment background concentrations. Guide E3344
1.1 This guide describes data visualization, statistical, fo-
describes how to select an appropriate background reference
rensic chemistry and geochemical methodologies (including
area(s) from which to collect sediment samples for laboratory
case studies) used in the evaluation of candidate background
analysis. Guide E3164 describes the sampling methodologies
data sets; this evaluation leads to the development of represen-
to obtain sediment samples in the field (whether from the
tative background data sets for the sediment site. Statistical
sediment site or background reference area[s]), while Guide
methodologies can then be applied to the representative back-
E3163 discusses appropriate laboratory methodologies for the
ground data sets to develop background threshold values
chemical analysis of potential contaminants of concern
(BTVs) that are measures of the upper limit of representative
(PCOCs) in the sediment samples. Relevant content contained
sediment background concentrations for the sediment site. In
in Guides E3344 and E3382 is summarized herein, but the
addition, representative background data sets and sediment site
individual guides should be consulted for more detailed cov-
data sets can be compared using two-sample statistical tests to
erage of these topics.
determine if there are statistically significant differences (at a
1.4 This guide focuses on the approach for the development
specified confidence level) between the two data sets (such as,
of representative sediment background concentrations used for
the median or mean values of the two data sets are significantly
remedial actions performed under various regulatory programs,
different).
including the Comprehensive Environmental Response,
1.1.1 This guide is intended to inform, complement, and
Compensation, and Liability Act (CERCLA). Although many
support, but not supersede the guidelines established by local,
of the references cited in this guide are CERCLA oriented, the
state, tribal, federal, or international agencies.
guide is applicable to remedial actions performed under local,
1.2 Technically defensible representative sediment back-
2 state, tribal, federal, and international cleanup programs.
ground concentrations are critical for several purposes (1).
However, the guide does not describe requirements for each
These include sediment site delineation, establishing remedial
jurisdiction. The requirements for the regulatory entity under
goals and cleanup levels, remedy selection, assessment of risks
which the cleanup is being performed should be reviewed to
posed by representative background concentrations, and estab-
confirm compliance.
lishing appropriate post-remedial monitoring plans.
1.5 This guide is designed to apply to contaminated sedi-
1.3 The overarching framework for the development of
ment sites where sediment data have been collected and are
representative sediment background concentrations at sedi-
readily available. Additionally, this guide assumes that risk
ment sites is presented in Guide E3382. Guide E3240 provides
assessments have been performed, so that the contaminants of
a general discussion of how conceptual site model (CSM)
concern (COCs) that exceed risk-based thresholds have been
development fits into the risk-based corrective action frame-
identified.
work for contaminated sediment sites, while Guide E3382
1.5.1 Furthermore, this guide presumes that the identified
provides a detailed discussion of the elements of a sediment
risk-based thresholds are low enough to pose corrective action
implementation challenges, or the site is subject to recontami-
nation from uncontrolled ongoing anthropogenic or natural
This guide is under the jurisdiction of ASTM Committee E50 on Environmental
sources, or both. In all cases, representative sediment back-
Assessment, Risk Management and Corrective Action and is the direct responsibil-
ground concentrations will be useful for determining the extent
ity of Subcommittee E50.04 on Corrective Action.
Current edition approved Dec. 1, 2023. Published January 2024. Originally
of corrective remedial actions (when used as remedial goals or
approved in 2020. Last previous edition approved in 2020 as E3242–20. DOI:
cleanup levels), evaluating risks posed by representative back-
10.1520/E3242–23
ground concentrations, and establishing appropriate post-
The boldface numbers in parentheses refer to the list of references at the end of
this standard. remedial monitoring plans.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
E3242 − 23
1.6 Units—The values stated in SI or CGS units are to be specifically related to current or historical site-related releases
regarded as standard. No other units of measurement are or activities. E3344
included in this standard.
3.1.1.1 Discussion—The definition of “anthropogenic back-
ground” varies with jurisdiction. In some jurisdictions, the
1.7 This standard does not purport to address all of the
regulator defines anthropogenic background as having both
safety concerns, if any, associated with its use. It is the
human-made and naturally occurring components. (2)
responsibility of the user of this standard to establish appro-
priate safety, health, and environmental practices and deter-
3.1.2 background (aka “reference”), n—a term applied to
mine the applicability of regulatory limitations prior to use.
substances, conditions, or locations that are similar to those
1.8 This international standard was developed in accor-
found at a sediment site but not influenced by current or
dance with internationally recognized principles on standard-
historical releases or activities from the sediment site; these are
ization established in the Decision on Principles for the
usually a combination of naturally occurring (consistently
Development of International Standards, Guides and Recom-
present in the environment but not influenced by human
mendations issued by the World Trade Organization Technical
activity) and anthropogenic (influenced by human activity but
Barriers to Trade (TBT) Committee.
not related to specific current or historical releases or activities
at the sediment site) components. E3382
2. Referenced Documents
3.1.3 candidate background data set, n—a raw (that is,
2.1 ASTM Standards:
unprocessed) background data set obtained either by the
D6312 Guide for Developing Appropriate Statistical Ap-
collection of data from a background reference area(s), or by
proaches for Groundwater Detection Monitoring Pro-
the extraction of background data from the sediment site data
grams at Waste Disposal Facilities
set, or a combination of both. E3382
D7048 Guide for Applying Statistical Methods for Assess-
3.1.3.1 Discussion—The candidate background data set
ment and Corrective Action Environmental Monitoring
must first be evaluated using the steps described in this guide
Programs
to obtain a representative background data set.
D7659 Guide for Strategies for Surface Sampling of Metals
and Metalloids for Worker Protection 3.1.4 contaminant of concern (COC), n—substances identi-
D7720 Guide for Statistically Evaluating Measurand Alarm fied as posing a risk based on a tiered risk assessment and that
Limits when Using Oil Analysis to Monitor Equipment may warrant corrective action. E3382
and Oil for Fitness and Contamination
3.1.4.1 Discussion—Typically, all potential contaminants of
E178 Practice for Dealing With Outlying Observations
concern (PCOCs) identified for a sediment site are evaluated in
E456 Terminology Relating to Quality and Statistics
the risk assessment process. PCOCs that have sediment con-
E1689 Guide for Developing Conceptual Site Models for
centrations greater than risk-based thresholds identified in the
Contaminated Sites
risk assessment process are defined as COCs. Thus, the COCs
E3163 Guide for Selection and Application of Analytical
identified for a sediment site are a subset of the PCOCs
Methods and Procedures Used during Sediment Correc-
identified for that site.
tive Action
3.1.5 distribution, n—as used in statistics, a set of all the
E3164 Guide for Contaminated Sediment Site Risk-Based
various values that individual observations may have and the
Corrective Action – Baseline, Remedy Implementation
frequency of their occurrence in the sample or population.
and Post-Remedy Monitoring Programs
D7720
E3240 Guide for Risk-Based Corrective Action for Contami-
3.1.6 high nondetect, n—a nondetect concentration with a
nated Sediment Sites
highly elevated detection limit; for example, a concentration
E3248 Guide for NAPL Mobility and Migration in Sediment
that resides in the upper decile of the analyte’s distribution (that
– Conceptual Models for Emplacement and Advection
th
is, a detection limit above the 90 percentile of the data set).
E3344 Guide for Developing Representative Sediment
E3382
Background Concentrations at Sediment Sites—Selection
th
of Background Reference Areas
3.1.7 median, n—the 50 percentile in a population or
E3382 Guide for Developing Representative Background
sample. E456
Concentrations at Sediment Sites — Framework
3.1.8 nonparametric, adj—a term referring to a statistical
Overview, Including Conceptual Site Model Consider-
technique in which the distribution of the constituent in the
ations
population is unknown and is not restricted to be of a specified
form. D7048
3. Terminology
3.1.9 outlier, n—see outlying observation.
3.1 Definitions:
3.1.1 anthropogenic background, n—human-made sub-
3.1.10 outlying observation, n—an extreme observation in
stances present in the environment due to human activities, not
either direction that appears to deviate markedly in value from
other members of the sample in which it appears. E178
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
3.1.11 parametric, adj—a term referring to a statistical
contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
technique in which the distribution of the constituent in the
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website. population is assumed to be known. D7048
E3242 − 23
3.1.12 representative background concentrations, potential future conditions and uses for the site. It provides a
n—chemical concentrations that are inclusive of naturally method to conduct the exposure pathway evaluation, inventory
occurring sources and anthropogenic sources similar to those the exposure pathways evaluated, and determine the status of
present at a sediment site but not related to current or historical the exposure pathways as incomplete, potentially complete, or
site releases or activities. E3382 complete.
3.1.13 representative background data set, n—a background 3.2.6 false negative error, n—in statistical tests, also known
data set obtained by evaluating candidate background data as “Type II” error.
3.2.6.1 Discussion—For the purposes of this guide, in site
using the steps described in Guide E3242. E3382
versus background comparisons, the error that occurs when the
3.1.13.1 Discussion—The evaluation determines if there are
statistical procedure does not indicate concentrations above
any data points in the candidate background data set that are
background, when such concentrations are present.
not representative of sediment site background conditions.
These data points are then removed from the candidate
3.2.7 false outlier, n—measurements that are very large or
background data set (using technically justifiable rationale) to
small relative to the rest of the data but represent true extreme
obtain a representative background data set. Typically, this data
values of a distribution and indicate more variability in the
set can be used to develop a BTV, which is a measure of the
population than was expected. (3)
upper limit of representative background concentrations; it is
3.2.8 false positive error, n—in statistical tests, also known
this BTV that is often used as a representative background
as “Type I” error.
concentration.
3.2.8.1 Discussion—For the purposes of this guide, in site
3.1.14 sediment(s), n—a matrix of pore water and particles
versus background comparisons, the error that occurs when the
including gravel, sand, silt, clay and other natural and anthro-
statistical procedure indicates concentrations above
pogenic substances that have settled at the bottom of a tidal or
background, when such concentrations are not present.
nontidal body of water. E3163
3.2.9 population, n—as used in statistics, a comprehensive
3.1.15 sediment site, n—the area(s) defined by the likely
set of values consisting of all possible observations or mea-
physical distribution of COC(s) from a source area and the surements of a certain phenomenon from which a sample is to
adjacent areas required to implement the corrective action. A
be drawn.
site could be an entire water body or a defined portion of a
3.2.10 potential contaminant of concern (PCOC), n—a
water body. E3240
contaminant whose sediment concentrations at the site may
3.1.16 upper tolerance limit (UTL), n—the upper confidence exceed applicable screening levels; this includes chemicals of
limit (with specified confidence level) for a percentile of a
potential environmental concern (COPECs) and chemicals of
distribution. D7659 potential concern (COPCs).
3.1.16.1 Discussion—The UTL is the value below which a
3.2.11 probability plot, n—a plot of ascending observations
specified fraction of the population will be found, with a
in a sample, versus their corresponding cumulative
specified level of confidence. For example, the 95/95 UTL is a
probabilities, based on a specified distribution function.
value for which one would have 95 % confidence that 95 % of
3.2.12 reference element, n—a major element that repre-
the population is below the UTL.
sents the mineral to which a trace element may be adsorbed.
3.2 Definitions of Terms Specific to This Standard:
3.2.13 sample, n—as used in statistics, a group of observa-
3.2.1 arithmetic mean, n—a measure of central tendency
tions taken from a population that serve to provide information
that is the sum of observed values in the sample divided by the
that may be used as a basis for making a decision concerning
sample size.
the population.
3.2.2 background reference areas, n—sediment areas that
3.2.14 sample size, n—as used in statistics, the number of
have similar physical, chemical, geological, biological, and
observations or measurements in the sample.
land-use characteristics as the site being investigated but are
3.2.15 significance level, n—as used in statistical hypothesis
not affected by current or historical site-related releases or
testing, the probability of rejecting a null hypothesis when it is
activities.
true.
3.2.3 background threshold value (BTV), n—a measure of
3.2.15.1 Discussion—Also known as “alpha” (α), it is se-
the upper limit of representative background concentrations.
lected prior to performing a statistical test. The significance
level is commonly set to 0.05, but should be determined on a
3.2.4 cleanup level, n—the prescribed average or point
site-specific basis; consultation with a statistician to choose the
sediment concentration of a chemical that shall not be ex-
optimal significance level may be warranted.
ceeded at the remediated site.
3.2.16 tolerable error rate, n—the specified maximum ac-
3.2.5 conceptual site model (CSM), n—the integrated rep-
ceptable error rate set by the decision maker.
resentation of the physical and environmental context, the
complete and potentially complete exposure pathways and the 3.2.17 trace element, n—an element defined as generally
potential fate and transport of potential contaminants of con-
being present at less than 0.1 weight percent in the sediment
cern at a site. sample; its natural concentrations are typically one or more
3.2.5.1 Discussion—The CSM should include both the cur- orders of magnitude lower than those of the reference ele-
rent understanding of the site and an understanding of the ments.
E3242 − 23
3.2.18 true outlier, n—measurements that are very large or 4.6 Systematic Project Planning and Scoping Process:
small relative to the rest of the data, but are a result of 4.6.1 When applying this guide, the user should undertake a
transcription errors, data-coding errors, or measurement system
systematic project planning and scoping process to collect
problems; or it is not representative of the investigated data information to assist in making site-specific, user-defined
population as confirmed by other lines of evidence.
decisions for a particular project, including assembling an
experienced team of project professionals. These practitioners
3.2.19 upper confidence limit (UCL), n—an upper limit of
should have the appropriate expertise to scope, plan, and
an estimated value, such as the mean, which has a specified
execute a sediment data acquisition and analysis program. This
probability of including the true value, with a specified
team may include, but is not limited to, project sponsors,
confidence level.
environmental consultants, toxicologists, site remediation
3.2.20 upper percentile, n—the value below which a speci-
professionals, analytical chemists, geochemists, and statisti-
fied percentage of observed values falls.
cians.
3.2.21 upper prediction limit (UPL), n—the value below
4.7 Use of Representative Background to Set a Boundary:
which a specified number of future independent measurements
4.7.1 Representative background concentrations for sedi-
will fall, with a specified confidence level.
ments can be used to delineate a sediment corrective action,
4. Significance and Use establishing the boundary of the sediment corrective action
area by distinguishing site-related impacts from representative
4.1 Intended Use:
background concentrations. This application requires the de-
4.1.1 This guide may be used by various parties involved in
velopment of a BTV for the representative background data
sediment corrective action programs, including regulatory
set.
agencies, project sponsors, environmental consultants,
toxicologists, risk assessors, site remediation professionals,
4.8 Use of Representative Background to Establish Cleanup
environmental contractors, and other stakeholders. Levels:
4.8.1 Representative background concentrations for sedi-
4.2 Importance of the CSM:
ments can be used to establish cleanup levels for use in
4.2.1 The CSM should be continuously updated and refined
sediment corrective actions. In cases where risk-based sedi-
to describe the physical properties, chemical composition and
ment cleanup levels are below representative background
occurrence, biologic features, and environmental conditions of
concentrations, background concentrations are typically used
the sediment corrective action project (Guide E1689).
as the cleanup level (4). This ensures that the cleanup levels are
4.3 Reference Material:
sustainable. Any recontamination from ongoing sources will
4.3.1 This guide should be used in conjunction with other
eventually result in surface sediment concentrations greater
ASTM guides listed in 2.1 (especially Guides E3344 and
than the risk-based cleanup level, but the surface sediment
E3382); this guide should also be used in conjunction with the
should still meet a cleanup level based on representative
material in the References at the end of this guide (including
background concentrations, even following recontamination.
1). Utilizing these reference materials will direct the user in
4.9 Use of Representative Background in Risk Assessments:
developing representative background concentrations for a
4.9.1 Representative background concentrations can be
sediment site.
used in the risk assessment process (including human and
4.4 Flexible Site-Specific Implementation:
ecological risk assessments) to understand risks posed by
4.4.1 This guide provides a systematic, but flexible, frame-
background levels of contaminants to human health and the
work to accommodate variations in approaches by regulatory
environment, and the incremental risks posed by site-related
agencies and by the user based on project objectives, site
releases or activities (or both) that result in sediment concen-
complexity, unique site features, regulatory requirements,
trations that exceed representative background concentrations.
newly developed guidance, newly published scientific
Conversely, they can be used to estimate the risk reduction for
research, changes in regulatory criteria, advances in scientific
various contaminants, if sediment is remediated from existing
knowledge and technical capability, and unforeseen circum-
COC concentrations to lower values (that is, representative
stances.
background concentrations).
4.5 Regulatory Frameworks:
4.10 Use of Representative Background in Post-Remedy
4.5.1 This guide is intended to be applicable to a broad
Monitoring Programs:
range of local, state, tribal, federal, or international
4.10.1 Post-remedy monitoring programs can also use rep-
jurisdictions, each with its own unique regulatory framework.
resentative background sediment concentrations either as a
As such, this guide does not provide a detailed discussion of
corrective action target or to understand how post-remedy
the requirements or guidance associated with any of these
concentrations compare to the sources not attributable to
regulatory frameworks, nor is it intended to supplant applicable
current or historical site releases or activities. Typically, source
regulations and guidance. The user of this guide will need to be
control actions taken to ensure that site-related releases are
aware of the regulatory requirements and guidance in the
controlled and will not re-contaminate the post-corrective
jurisdiction where the work is being performed.
action sediments must be developed based on an understanding
of ongoing contributions from representative background.
Ongoing sources unrelated to current or historical site-related
E3242 − 23
releases or activities (that may or may not be subject to source 5. Overview of Representative Background
control actions) must be considered in this evaluation. Concentration Development Process
4.11 Other Considerations:
5.1 Importance of Representative Background:
4.11.1 This guide does not cover all components of a
5.1.1 Multiple sources may contribute to the nature and
program to develop representative sediment background con-
extent of contamination at sediment sites. The largest contri-
centrations.
bution of contamination at sediment sites is typically attributed
4.11.2 The overarching process to develop representative
to current or historical site releases or activities. However,
background concentrations (including CSM considerations) is
contamination can also result from natural or ongoing anthro-
not covered in detail in this guide but is discussed in more
pogenic sources (or both) not related to current or historical site
depth in Guide E3382.
releases or activities. Discharges from combined sewer over-
4.11.3 The selection of a background reference area(s) for
flows (CSOs), industrial outfalls, and storm sewer systems
the sediment site is not covered in detail in this guide but is
(municipal and private) or surface runoff are examples of
extensively described in Guide E3344.
ongoing anthropogenic sources that may be unrelated to
4.11.4 Sediment sampling and laboratory analyses are not
current or historical site releases or activities.
covered in this guide. Guides E3163 and E3164 contain
5.1.2 The off-site contamination not associated with current
extensive information concerning sediment sampling and labo-
or historical site releases or activities is considered a compo-
ratory analyses.
nent of representative background concentrations and will
4.11.5 Data quality objectives are not covered in this guide.
continue to be a source of contamination to the sediment site
Data quality objectives are described in (5).
unless all transport pathways are eliminated. A primary objec-
4.11.6 Background study design considerations are not cov-
tive of determining representative background concentrations
ered in this guide but are described in other references,
is to account for any background chemical input (both natural
including Guides E3163 and E3164, as well as (6, 7).
and anthropogenic) that is expected to continue migrating onto
4.11.7 Geospatial analysis considerations are not thoroughly
the sediment site after the completion of corrective actions.
discussed in this guidance but are discussed in more depth
One of the important principles for management of contami-
relative to environmental evaluations in (8), which focuses on
nated sediment sites is the control of sources of contamination,
quality assurance concerns relative to geospatial analyses.
to the greatest extent practicable, prior to the initiation of
4.11.8 In this guide, only the concentrations of COCs are
corrective actions at the subject site (4, 9). However, it is rarely
considered to be in scope. Residual background radioactivity is
practicable to control all background sources.
out of scope.
5.1.3 Technically defensible representative background con-
4.12 Structure and Components of This Guide: centrations are those that accurately reflect chemical inputs to
a sediment site from natural and ongoing anthropogenic
4.12.1 The user of this guide should review the overall
structure and components of this guide before proceeding with sources unrelated to current or historical site releases or
activities. In addition to informing or establishing technically
use, including:
defensible cleanup levels, representative background concen-
Section 1 Scope
Section 2 Referenced Documents trations can assist in determining site boundaries, identifying
Section 3 Terminology
COCs, establishing and optimizing realistic post-remedy moni-
Section 4 Significance and Use
toring plans, and assessing the performance of corrective
Section 5 Overview of Representative Background Concentration
Development Process
actions.
Section 6 Development of Candidate Background Data Sets
5.1.4 In the absence of representative background
Section 7 Evaluation of Candidate Background Data Sets to Obtain
Representative Background Data Sets concentrations, risk-based cleanup levels may be inappropri-
Section 8 Data Visualization
ately used at sediment sites where representative background
Section 9 Evaluation of High Nondetect Data Points
concentrations are actually greater than the risk-based cleanup
Section 10 Evaluation of Outlying Data Points
levels. Similarly, if the representative background concentra-
Section 11 Forensic Chemistry Evaluation of Organic Contaminants
Section 12 Geochemical Evaluation of Metals
tions have been erroneously developed (for example, by the
Section 13 Methodology Application to Develop a Representative
inappropriate exclusion of some outlier data points [false
Background Data Set from a Candidate Background
outliers]; refer to Section 10), inappropriately low cleanup
Data Set
Section 14 Development of Representative Background
levels could be used in the corrective action evaluation process.
Concentrations
Under both circumstances, surface sediments at sediment sites
Section 15 Comparison of Sediment Site and Representative
Background Data Sets Using Statistical Two-Sample will eventually return to representative background concentra-
Testing
tions at some time after corrective actions are completed and
Section 16 Keywords
cleanup levels will be exceeded. Due to exceedances of the
Appendix X1 Organic and Inorganic Chemistry Overview
Appendix X2 Illustrative Case Studies from One Example Sediment Site inappropriately low cleanup levels, the corrective actions
Appendix X3 Summaries for Outlier Testing and Two-Sample Statistical
would be perceived as failures.
Testing
5.1.5 Attempting to implement corrective actions to achieve
References
concentrations less than representative background is not
sustainable over the long-term and can require considerable
expenditures that serve no environmental or public health
purpose (Guide E3382). The process described in Guide E3382
E3242 − 23
is intended to help promote a scientifically sound approach for tion of background data sets from the sediment site data to
developing representative background concentrations, leading develop candidate background data sets.
to corrective action decisions that avoid costly perceived 5.2.6 Once a candidate background data set is developed,
corrective action failures at sediment sites. The topics covered Sections 6 – 14 describe the process used to evaluate these data
sets to develop representative background data sets for the site
in this guide are a critical component of the process outlined in
Guide E3382 and include developing candidate background and then develop BTVs for these representative background
data sets (see Section 7 for further details on the evaluation
data sets; data visualization of candidate background data sets;
evaluation of candidate background data sets to develop process).
representative background data sets using statistical, forensic
6. Development of Candidate Background Data Sets
chemistry and geochemical methodologies; the development of
various measures of representative background concentrations
6.1 Background Reference Area Characteristics:
for applications at sediment sites using representative back-
6.1.1 The background reference area(s) should have similar
ground data sets; and the application of two-sample statistical
physical, chemical, geological, biological, and land-use char-
tests to compare representative background and sediment site
acteristics as the sediment site. Additionally, such areas should
data sets.
not be influenced by current or historical site-related releases or
activities but should include ongoing sources similar to those
5.2 Overview of Process to Develop Representative Back-
present at the sediment site, as well as a similar land use (Guide
ground Concentrations in Sediment:
E3344). For example, if the sediment site is located in an
5.2.1 Application of background guidance for soil and
industrial area with CSOs, the background reference area(s)
groundwater at upland sites may not be appropriate at sediment
should be located in an industrial area with CSOs. The more
sites. Sediment sites have many different characteristics that
developed the CSM, the more informed the choice of back-
are not present at upland sites (Guide E3248), including
ground reference areas will be. Additional information on the
physical characteristics, geochemical characteristics, biologi-
selection process is provided in Guide E3344 and (10).
cal characteristics, and different contaminant emplacement and
6.1.2 Once an appropriate off-site background reference
transport mechanisms.
area(s) is identified for the sediment site, existing sediment
5.2.2 This guide and its associated guides (Guides E3163,
sample data (if available) should be acquired for the back-
E3164, E3240, E3344, and E3382) have been developed (in
ground reference area(s) and evaluated for data quality and
part) to fill a gap due to the absence of existing guidance from
usability. Existing data from a previous study may be suitable
various regulatory agencies for the development of represen-
for inclusion in the candidate background data set or to inform
tative background concentrations for contaminated sediment
the study design for a new background reference area sediment
sites.
sample collection program. However, the inclusion of existing
5.2.3 Fig. 1 presents the overall framework to develop the
data in the candidate background data set must be assessed on
BTV, which is a measure of the upper limit of representative
a case-by-case basis (7).
background concentrations at a sediment site; this process is
6.1.2.1 Inclusion of existing data in the candidate back-
presented in detail in Guide E3382. As a first step, a thorough
ground data set is inappropriate if different analytical test
understanding of the sediment site is necessary before devel-
methods were used to generate the existing candidate back-
oping the BTV. This can be accomplished by developing a
ground data set and the sediment site data set. Background data
sediment site CSM (refer to Guides E3240 and E3382). As part
sets generated by different analytical laboratories should also
of this CSM, the sediment site PCOCs must be identified.
be subjected to a careful assessment of laboratory standard
operating procedures to ensure comparability of results. Addi-
5.2.4 Once the preliminary sediment site CSM has been
tional statistical comparison of results generated from the
developed, a suitable background reference area (or areas) can
different laboratories are also recommended. The objective is
be identified for sampling (that is, the second step in Fig. 1);
to have comparable background data, even if the data are
the methodology used to do this is presented in detail in Guide
obtained from different laboratories.
E3344.
5.2.5 Candidate background data sets are typically obtained
6.2 Collection of Candidate Background Data Sets from
in two ways: (1) collecting sediment samples from background
Background Reference Areas:
reference areas that have characteristics as similar as possible
6.2.1 Data collection can commence once the background
to that of the sediment site (see Guide E3344 for a detailed reference area(s) are identified. This guide does not address
discussion on the selection of the background reference area),
methods and means of data collection. However, sample
or (2) extracting candidate background data sets from the collection and laboratory analysis methods at the background
sediment site data from portions of the site that have been
reference area(s) should be as similar as possible to those used
unaffected by current or historical site releases or activities (see during sediment site data collection. When collecting data from
Appendix X2 and Ref. (7) for a detailed discussion of a background reference area(s) for comparison to sediment site
background data extraction from the sediment site data set). data, it is critical for background data to be collected in the
Additionally, under certain circumstances data sets from (1) same manner (such as, from the same depth interval) as the
and (2) can be combined to develop a single candidate sediment site under investigation and to use the same analytical
background data set. Section 6 describes the collection of test methodology that was used to determine compliance with
sediment samples from background reference areas and extrac- cleanup levels (ideally, using the same laboratory that analyzed
E3242 − 23
FIG. 1 Process to Develop Representative Sediment Background Concentrations (Modified from Guide E3382)
the samples from the sediment site). This allows a direct 6.2.2 The optimal number of background reference area
comparison between chemical concentrations in background sediment samples needs to be determined on a project-specific
reference area samples to sediment site samples—a crucial step basis by qualified personnel on the project team.
in the development of representative background concentra- 6.2.3 New sediment samples from the off-site background
tions. reference area(s) should be collected and sent to the laboratory
E3242 − 23
for analysis of PCOCs and other parameters (such as total using candidate background data extracted from the sediment
organic carbon and grain size distribution). Upon validation, site data set; or a combination of both if certain conditions are
the laboratory results and suitable existing data would consti- met (see Appendix X2).
tute the candidate background reference area data set. Detailed
7.2 Evaluation Methodologies Summary:
sediment sampling and analysis guidance is provided in Guides
7.2.1 Once developed, candidate background data sets are
E3163 and E3164; guidance on various sampling designs is
then evaluated using the methodologies outlined in the follow-
provided in (6).
ing sections:
6.2.4 Regulatory agency agreement on the scope and scale
7.2.1.1 Visualization of the candidate background data set
of the background reference area sampling effort to obtain a
(Section 8).
candidate background data set is important and should be
7.2.1.2 Evaluation of high nondetect results (Section 9).
captured in a site’s Data Quality Objectives. In the United
7.2.1.3 Evaluation of statistical outlying data points (Sec-
States, the use of the USEPA’s Data Quality Assessment
tion 10).
approach (3) is recommended if no superseding regulatory
7.2.1.4 Forensic chemistry evaluation of organic contami-
guidance is available.
nants (Section 11).
6.2.5 Concentrations from background reference areas may
7.2.1.5 Geochemical evaluation of metals (Section 12).
be characteristic of one or more statistical populations with
7.2.2 If technically justifiable, some data points may be
distinct features. For example, sediment background concen-
excluded from the candidate background data set in the
trations from a basin surrounded by urban developments or
development of a representative background data set (Section
industrial areas (or both) will be distinctly different from those
13).
collected from another portion of the same basin surrounded by
7.3 Uses of Representative Background Data Sets:
agricultural areas. Combining background data sets that repre-
7.3.1 A representative background data set can be used to
sent different statistical populations can lead to erroneous or
develop a BTV for this data set (that is, a measure of the upper
misleading results. Candidate background data sets are those
limit of representative background concentrations for the COC)
that are collected from sampling locations with physical,
as outlined in Section 14.
chemical, geological, biological, and land-use characteristics
7.3.2 A representative background data set can also be used
most similar to the sediment site.
in two-sample statistical comparisons with the sediment site
6.3 Extraction of Candidate Background Data Sets from
data set as described in Section 15.
Sediment Site Data Sets:
6.3.1 Although collecting samples from off-site background
8. Data Visualization
reference areas is typically preferable, in many instances
8.1 Overview:
(especially in urban areas), identification of such areas are
8.1.1 Evaluation of a candidate background data set should
problematic. Under such conditions, candidate background
begin by visualizing the data. There are a number of techniques
data sets may potentially be extracted from sediment site data,
described in this section that can be used, depending on the
as long as part of the sediment site has not been impacted by
characteristics of the candidate background data set. These
current or historical site releases or activities.
plots are exploratory in nature and not all plots shown are
6.3.1.1 Extracting candidate background data from sedi-
required when visualizing the candidate background data set.
ment site data not only maximizes the utility of existing data,
Different plots depicting the same data set are illustrated in the
but also avoids the often complex task of selecting separate
following subsections. Other tools (for example, geographic
background reference areas that adequately display physical,
information system [GIS] post plots) can also be used for data
chemical, geological, biological, and land-use characteristics
visualization.
similar to the site, as described in Guide E3344.
6.3.2 Extraction of candidate background data from sedi-
8.2 Dot Plot:
ment site data often involves utilizing probability plots to
8.2.1 Dot plots represent each concentration in a candidate
segregate site data into impacted versus unimpacted popula-
background data set as an individual dot (see Fig. 2), with
tions for each COC (see Appendix X2).
concentration values listed along the x-axis (12). Samples with
6.3.2.1 Even when data from separate off-site background
similar concentrations appear as vertical stacks, and large data
reference areas are available, an extracted site-specific back-
sets can be accommodated by using one dot to represent a
ground data set can provide (if certain conditions are met)
predetermined number of data points. This plot allows all data
additional data for inclusion in the candidate background data
points to be viewed, and no distributional assumptions are
set. Therefore, an analysis of existing sediment site data is
imposed on the data. Symmetry, bimodal or multi-modal
always recommended. A more complete review of this topic is
groupings, and skewness can be discerned in dot plots.
presented in (7, 11).
8.3 Histograms:
8.3.1 Histograms depict data sets in bar form, with concen-
7. Evaluation of Candidate Background Data Sets to
trations (grouped in “bins,” or intervals) along the x-axis and
Obtain Representative Background Data Sets
the corresponding frequency (or percentages of frequency) for
7.1 Overview: each concentration interval along the y-axis (see Fig. 3). The
7.1.1 Candidate background data sets for the sediment site area of each bar reflects the proportion of that concentration
are generated by sampling a background reference area(s) or interval within the candidate background data set. Histograms
E3242 − 23
FIG. 2 Dot Plot of Lead Concentrations in a Candidate Background Data Set
FIG. 3 Histogram of Lead Concentrations in a Candidate Background Data Set
th
are commonly used in conjunction with goodness-of-fit tests to representing the 25 percentile; the median is represented by a
assess the shape of a data distribution, because the graphs can
symbol within the box. In this example, the upper whisker
reveal such features as symmetry, skewness, and bimodality. extends to the maximum data point and the lower whisker
However, the observed shape of a data set is affected by the bin
extends to the minimum data point. A side-by side configura-
size (for example, 10 mg/kg interval versus 20 mg/kg interval),
tion of box plots permits visual comparison of multiple data
which should be carefully selected. Multiple data sets can be
sets to quickly discern whether the data distributions are
depicted on the same histogram using unique colors or pat-
similar or distinct.
terns.
8.4.2 Users can define the appearance of box plots. Default
settings in statistical software programs typically identify
8.4 Box Plots:
8.4.1 Box plots, or box-and-whisker plots, are used to predefined “outlier values” (for example, values outside 1.5
times the interquartile range) and “extreme values” (for
compare two or more groups of data (3, 13). A box plot (see
Fig. 4) provides a summary view of an entire data set, example, values outside 3 times the interquartile range). These
settings are arbitrary and can be avoided by simply extending
including the range of concentrations, degree of symmetry, and
skewness of the data. The box encloses the central 50 % of the the upper and lower whiskers to the maximum and minimum
data points (“interquartile range”), with the top of the box data points. See Section 10 for discussions of statistical outliers
th
representing the 75 percentile and the bottom of the box and their appropriate treatment.
E3242 − 23
FIG. 4 Box Plots of Lead Concentrations in a Candidate Background Data Set (Left Side) and Following Evaluation in the Representa-
tive Background Data Set (Right Side)
8.5 Probability Plots: determine whether the data set is derived from a single or
8.5.1 A probability plot is often used to visualize
...