United States Office of Solid Waste and EPA 540-R-98-038
Environmental Protection Emergency Response OSWER 9230.0-83P
Agency (5102G) PB98-963307
September 1998
v>EPA Quality Assurance Guidance
for Conducting Brownfields
Site Assessments
-------
Quality Assurance Guidance for Conducting
Brownfields Site Assessments
U.S. Environmental Protection Agency
Office of Solid Waste and Emergency Response
Washington, DC 20460
-------
Notice
This guidance document describes key principles and best practices for Brownfields site assessment
quality assurance and quality control based on program experience. It is intended as a reference for
people involved in the Brownfields site assessment process. This guidance manual does not constitute a
rulemaking by the U.S. Environmental Protection Agency (EPA). The policies set forth in this document
are intended solely as guidance. They are not intended, nor can they be relied upon, to create any
substantive or procedural rights enforceable by any party in litigation with the United States. EPA
officials may decide to follow the guidance provided in this directive, or may take action that is at
variance with the guidance, policies, and procedures in this directive, on the basis of an analysis of
specific circumstances. The Agency also reserves the right to change this directive without public notice.
Mention of trade names or commercial products does not constitute endorsement or recommendation for
use.
-------
TABLE OF CONTENTS
Executive Summary ES-1
Section 1. Introduction 1
Section 2. Brownfields Site Assessments 3
Brownfields Site Assessment Purpose 3
Brownfields Site Assessment Process 3
Section 3. Data Quality Objectives 5
The DQO Process 7
Step 1: Stating the Problem 7
Step 2: Identifying the Decision 9
Step 3: Identifying Inputs to the Decision 10
Step 4: Defining the Boundaries of the Study 11
Step 5: Developing a Decision Rule 11
Step 6: Specifying Limits on Decision Errors 12
Step 7: Optimizing the Design 13
Estimating Costs 14
Section 4. Quality Assurance Programs & Sampling Design Strategies 15
Quality Assurance 16
Quality Control 16
Sampling Design Strategies 17
Multi-phase Investigations 20
Types of Samples 20
Background Samples 21
Analytical Methods 21
Quality Control in the Field 23
Field Instrument/Equipment Inspection and Calibration 23
Field Documentation 24
Field System Audits 24
Quality Control in the Laboratory 24
Quality Control Samples 25
Document Control 26
Deliverables for Data Useability Review 27
Improving Data Useability 28
Summary 28
Appendix A - Model Quality Assurance Project Plan A-l
Appendix B - Glossary of Terms B-l
Appendix C - References C-l
Exhibit 1 - Brownfields Site Assessment Process
Exhibit 2 - Summary of the DQO Process
Exhibit 3 - Comparison of Preliminary Assessment and Due Diligence Site Assessments
Exhibit 4 - The Springfield Site
Exhibit 5 - Number of Samples Required to Achieve Given Rates of False Positive and Negative, and
MDRD
Exhibit 6 - Types of QC Samples
in
-------
EXECUTIVE SUMMARY
||\|b any sites across the nation once used for
TIB industrial and commercial purposes are
now abandoned or under-used. Some of these
sites — often referred to as "Brownfields"— are
contaminated; others are perceived or suspected
to be contaminated. In 1993, the Environmental
Protection Agency (EPA) created the
Brownfields Economic Redevelopment
Initiative to empower States, Tribes,
communities, and other stakeholders to work
together in a timely manner to assess and safely
clean up Brownfields to facilitate their reuse.
This guidance document serves to inform
Brownfields site managers of important quality
assurance concepts and issues, and provides a
road map for identifying the type and quality of
environmental data needed to present a clear
picture of the site's environmental conditions.
However, because of the wide range of site-
specific issues, project goals, and the degree of
difficulty that the Brownfields site assessment
team may encounter, this document cannot
anticipate every question likely to arise during
the project. Therefore, when questions arise, it
is hoped that the reader will turn to the
extensively referenced Internet and document
resources provided in Appendix C for more
detailed information.
Brownfields Site Assessments
The Brownfields site assessment requires a team
approach encompassing a range of multi-
disciplinary knowledge and skills. The
Brownfields site assessment should provide
sufficient data of adequate quality to allow
officials to confidently make decisions about the
potential reuse of a Brownfields site.
Brownfields Site Assessment Process
The Brownfields site assessment process
routinely involves one or more of the following
activities: a review of historical records; a field
investigation including sample collection and
analysis; the assessment of data useability; and
an evaluation of cleanup options and costs.
Through careful planning, the Brownfields site
assessment team develops a conceptual site
model and establishes and communicates the
team's goals and how the team will reach those
goals using a Quality Assurance Project Plan
(QAPP).
Quality Assurance/Quality Control
Brownfields team members should understand
the benefits of strong quality assurance (QA)
and quality control (QC) procedures.
Quality assurance is an integrated system of
management activities involving planning,
implementation, assessment, reporting, and
quality improvement to ensure that a process,
item, or service is of the type and quality needed
and expected. Quality control is the overall
system of technical activities (including checks
on sampling and analysis) that measure the
performance of a process against defined
standards to verify that they meet predefined
requirements. Since errors can occur in the
field, laboratory, or office, QC must be part of
each of these functions.
Document Control
Document control is a crucial, but an often
overlooked, component of quality assurance. It
is critical to completion of the last stage of a
Brownfields site assessment — review of data
useability.
Data useability review depends on thorough
documentation of predefined data specifications
and the related events that take place during
implementation of the project
Data Quality Objectives (DQO) Process
Data credibility is one of the most important
challenges facing municipalities, Tribes, and
States managing a Brownfields site assessment.
An important planning tool used to help ensure
ES-1
-------
data credibility is the DQO process.
The DQO process allows the Brownfields site
assessment team to determine the level of data
quality needed for specific data collection
activities, and to estimate the cost associated
with these activities.
SUMMARY OF THE DQO PROCESS
1. State the Problem: What is the purpose of the
project?
2. Identify the decision(s): What are the available
options under consideration?
3. Identify Inputs in the Decision(s): What
information is needed to make informed, defensible
decisions?
4. Define the Boundaries of the Study: What is the
geographical extent and time and budget constraints
for the project.
5. Develop a Decision Rule: Formulate "if...then"
statements that relate the data to the decision they
support.
6. Specify Limits on Decision Errors: Estimate how
much uncertainty will be tolerated in the site
decision(s).
7. Optimize the Design: Identify the most cost-
effective means to gather the data needed. If
obstacles exist, reassess all the steps of the DQO
process to refine decisions and goals until a workable
roadmap or decision tree is produced.
These seven steps are used during the planning
of the Brownfields site assessment process to
ensure that field activities, data collection
operations, and the resulting data meet the
project objectives. The DQO process is
iterative, and the output of one step may affect
prior steps. This may lead the Brownfields site
assessment team to revisit some previous steps
but will ultimately lead to a more efficient data
collection design.
Application of the DQO process is actually a
"common sense" approach that translates broad
consensus-based goals into specific tasks. In
this way, the Brownfields team uses the DQO
process to prepare a road map, which can guide
the project, inform the public and other
interested parties, and bring newcomers to the
project quickly up to speed.
Quality Assurance Project Plan (QAPP)
The Environmental Protection Agency requires
that all Federally funded environmental
monitoring and measurement efforts participate
in a centrally managed quality assurance
program.
Any Brownfields site assessment team
generating data under this quality assurance
program has the responsibility to implement
minimum procedures to ensure that the
precision, accuracy, and completeness of its data
are known and documented.
To ensure this responsibility is met uniformly,
each Brownfields site assessment team should
prepare a written QAPP. The QAPP is a formal
document describing in comprehensive detail
the necessary QA and QC, and other technical
activities that should be implemented to ensure
that the results of the work preformed will
satisfy the stated performance criteria.
The QAPP documents the project planning
process (i.e., the DQO process), enhances the
credibility of sampling results, produces data of
know quality, and saves resources by reducing
errors and the time and money spent correcting
them.
This guidance document includes a description
of a QAPP and template forms to prepare one.
ES-2
-------
SECTION 1. INTRODUCTION
I any sites across the nation once used for
I industrial and commercial purposes are
now abandoned or under-used. Some of these
sites — often referred to as "Brownfields"— are
contaminated; others are perceived or suspected
to be contaminated. In 1993, the Environmental
Protection Agency (EPA) created the
Brownfields Economic Redevelopment
Initiative to empower States, Tribes,
communities, and other stakeholders to work
together in a timely manner to assess, and safely
clean up Brownfields to facilitate their reuse.
Successful Brownfields projects can help
reverse the spiral of unaddressed contamination
and its related problems and help maintain
deterrents to future contamination.
CONCEPTS
QA - (Quality Assurance) an integrated system of
planning, quality control, assessment,
improvement, and reporting.
QC - (Quality Control) a system of technical
activities that measure and control quality so that
data meet users' needs.
QAPP - (Quality Assurance Project Plan) a
document containing detailed procedures for
achieving data quality; generally prepared for all
EPA environmental data collection activities and
approved prior to data collection.
DQOs - (Data Quality Objectives) quantitative and
qualitative statements that define the type,
quantity, and quality of data needed to support the
site decision and acceptable levels of uncertainty
in the data that form the basis for the decision.
DQO Process - a systematic planning tool that
focuses on investigative goals and resultant
decisions to help decision-makers plan to collect
the type and quality of data that meet the
acceptable level of uncertainty.
Because concerns about future environmental
risks and liability can hinder redevelopment, the
Brownfields Initiative seeks to minimize the
uncertainty surrounding actual or perceived
contamination associated with these sites.
Establishing and following comprehensive
quality assurance (QA) procedures during the
collection of environmental data relating to site
contamination helps to minimize uncertainty.
This document provides municipalities, Tribes,
and States with guidance for an overall approach
to quality assurance for Brownfields site
assessments. It includes a description of a
Quality Assurance Project Plan (QAPP) and
forms necessary to prepare one. (See Appendix
A for the QAPP template.) The guidance
presented here provides a road map for
identifying the type and quality of
environmental data needed to present a clear
picture of the environmental conditions of the
site. Knowing the quality of environmental
measurement data will allow municipalities,
Tribes, and States to make site redevelopment
decisions that are both technically sound and
financially feasible.
Section 2 describes the range of environmental
activities expected during Brownfields site
assessments. Section 3 outlines and provides
examples of the data quality objectives (DQO)
process. Section 4 discusses sampling design
strategies and the importance of QA and quality
control. Section 4 also builds on information in
Section 3 by providing more specific sampling
design examples and other information. Section
4 refers the reader to the QAPP template
provided in Appendix A — a series of forms
that can be used to develop a site-specific
QAPP. Section 4 discusses the assessment of
collected data including whether they meet the
objectives of the Brownfields site assessment as
defined during the DQO process. This
document contains a glossary (Appendix B) and
a list of sources of information (Appendix C).
Each section introduces a set of new concepts.
This document does not present step-by-step
instructions on how to conduct all aspects of a
Brownfields site assessment. Instead it outlines
what the various tasks are, and the expertise
necessary for the tasks. For those unfamiliar
with Brownfields site assessments, it will
provide the background necessary to
communicate effectively with other Brownfields
team members, contractors, and laboratories.
-------
SECTION 2. BROWNFIELDS SITE ASSESSMENTS
I Brownfields site assessment should
I provide sufficient data of adequate quality
to allow officials to confidently make decisions
about the potential reuse of a Brownfields site.
The site assessment should minimize the
uncertainties inherent in environmental
investigation and focus on producing data
relevant to site-specific objectives.
Beneficial reuse of Brownfields sites requires a
team approach encompassing a range of multi-
disciplinary knowledge and skills, including
expertise in analytical chemistry, environmental
engineering, geology, sample collection,
statistics, public policy, and economics. Public
satisfaction with the project will also depend on
the Brownfields project team's (team) efforts to
build a consensus and meet the interests of the
public, local community, and commercial sector.
Brownfields Site Assessment
Purpose
Brownfields site assessments are conducted to
facilitate the reuse of properties by determining
whether contamination exists onsite, and if so,
the characteristics of contamination, including
the threat it poses, potential solutions for
cleanup, and the cost of solutions necessary to
prepare the site for redevelopment.
Brownfields Site Assessment
Process
A Brownfields site assessment routinely
involves one or more of the following activities:
a background investigation; a field investigation
including sample collection and analysis; an
evaluation of cleanup options and costs; and the
assessment of the useability of resulting data.
Frequently, the first step is to conduct a site
background or historical investigation to
identify past uses of the property, including
types and amounts of chemicals that may have
been used onsite and waste generation and
disposal activities that may have contributed to
contamination. The team can obtain this
information through review of historical
records, and through interviews with personnel
who may have knowledge of past waste
generation and disposal practices at the site. A
site visit is also helpful for identifying visible
signs of contamination.
A sampling and analysis investigation typically
follows the background investigation. Sampling
and analysis focuses on those areas of concern
identified during the background investigation.
The field sampling activities identify the
contaminants (e.g., arsenic in soil), the
concentrations of those contaminants (e.g., 50
parts per million (ppm)), and the areas of
contamination that should be addressed before
redevelopment can begin (e.g., all areas of
contamination greater than 20 ppm arsenic in
soil).
CONCEPTS
Brownfields Project Team - term applied to the
group of individuals essential to the success of the
project; the team is collectively skilled in analytical
chemistry, environmental engineering, statistics,
economics, public policy, etc.
Brownfields Site Assessment - a process to
determine the feasibility of site redevelopment
through various activities, including background
investigations, site sampling and analysis, and
evaluation of cleanup options and costs.
Another activity is estimating the cost of
cleanup options based on future uses and
redevelopment plans. Information on cleanup
options can be found on EPA's CLU-IN website
located at http://www.clu-in.com/supplyl.htm.
The next activity is assessing whether the data
are sufficient for their intended purpose. For
example, are they sufficiently reliable to
determine that the site does not require cleanup
prior to redevelopment?
This document focuses on preparing for a
Brownfields site assessment (see Exhibit 1
-------
below). Careful planning is critical to ensure
collection of useful data at minimum cost.
Planning tools that will help the team produce
the data they need include the idea of a
conceptual site model, the data quality
objectives (DQO) process, and some important
statistical concepts. These tools help the team
complete the QAPP, which establishes and
communicates the team's goals and how the
Brownfields site assessment will reach those
goals.
The QAPP developed for Brownfields site
assessments should combine planning for the
entire project — management, sampling,
analysis, data review/evaluation, and reporting
— under one cover. The QAPP should be
shared with all members of the Brownfields
project team and contractors performing
sampling and analytical work. During the
preparation of the QAPP, the team should rely
on its contractors and the laboratory it has
chosen to perform analyses to provide assistance
where needed.
Planning for data review is especially important
to an effective QAPP. Data review involves
comparing the actual data generated during site
assessment against the DQOs established during
project planning.
It is expected that a Brownfields team may use
Federal funding for various Brownfields site
assessment activities. These activities can range
from developing an inventory of potential sites
to extensive sampling events at individual sites
to documenting the technical feasibility of
cleanup options. The following sections present
systematic methods for planning a cost-effective
Brownfields site assessment with appropriate
quality assurance that will produce data of
adequate quality to meet project goals.
EXHIBIT 1
Focus of This Document
Conceptual Site Model
DQO Process
QAPP
Data Useabiiity
Assessment
THE BROWNFIELDS
SITE ASSESSMENT PROCESS
-------
SECTION 3. DATA QUALITY OBJECTIVES
I ata credibility is one of the most important
I challenges facing municipalities, Tribes,
and States managing Brownfields assessments.
In accepting a Brownfields grant, the recipient
has agreed to comply with the quality assurance
(QA) provisions set forth in 40 CFR 31.45 (see
40 CFR 31.45 Quality Assurance
If the grantee's project involves environmentally related
measurements or data generation, the grantee shall
develop and implement quality assurance practices
consisting of policies, procedures, specifications,
standards, and documentation sufficient to produce data
of quality adequate to meet project objectives and to
minimize loss of data due to out-of-control conditions or
malfunctions. [53 FR8076, Mar. 11,1988]
box).
Members of the Brownfields team who are
involved in project planning, sample collection,
laboratory analysis, data review, and data
assessment should understand the benefits of
QA and quality control (QC) procedures. These
procedures will be used during the planning of
the Brownfields site assessment to ensure that
field activities and data collection operations,
and the data they generate, meet the objectives
of the project. The DQO process allows the
team to determine the level of data quality
needed for specific data collection activities,
and to estimate the costs associated with these
activities.
The DQO process is actually a "common sense"
approach to translate broad consensus-based
goals into specific tasks. Only after defining the
overall goals of the project can the team identify
the tasks that will produce the data needed to
support decision-making at the end of the
project. In this way, the team uses the DQO
process to prepare a road map, which can guide
the project, inform the public and other
interested parties, and bring newcomers to the
project up to speed.
It is not possible to provide a common set of
DQOs applicable to Brownfields site
assessments because site characteristics,
decisions, and data quality needs vary from site
to site. However, the following is an overview
of the DQO framework and examples of its
application to a hypothetical Brownfields site
assessment.
An overview of the DQO process is summarized
in Exhibit 2 below, and a more thorough
discussion will follow. The DQO process is
iterative, and the output of one step may affect
prior steps. This may lead the team to revisit
some previous steps but ultimately will lead to a
more efficient data collection design.
EXHIBIT 2
SUMMARY OF THE DQO PROCESS
1. State the Problem: What is the purpose of the
project?
2. Identify the Decision(s): What are the available
options under consideration?
3. Identify Inputs in the Decision(s): What
information is needed to make informed, defensible
decisions?
4. Define the Boundaries of the Study: What is the
geographical extent, time, and budget constraints for
the project?
5. Develop a Decision Rule: Formulate "if...then"
statements that relate the data to the decision they
support.
6. Specify Limits on Decision Errors: Estimate how
much uncertainty will be tolerated in the site
decision(s).
7. Optimize the Design: Identify the most cost-
effective means to gather the data needed. If
obstacles exist, reassess all the steps of the DQO
process to refine decisions and goals until a workable
road map or decision tree is produced.
The first step in the DQO process is to develop a
conceptual site model. A conceptual site model
-------
provides an understanding of the site based on
currently available data prior to the site
assessment. It identifies historical uses of the
site, potential exposure pathways, cleanup
concerns, and future land use. As data from the
Brownfields site assessment become available,
they are used to refine the model and give the
team a clearer picture of the site. A well-
defined, detailed conceptual site model will help
the team identify data necessary to support
decisions about the property.
To determine the kinds of data to be collected,
the DQO process translates the goals of the
Brownfields site assessment into qualitative and
quantitative statements that define the type of
data needed to support decisions and that
specify the amount of uncertainty (i.e., the
chance of drawing an incorrect conclusion) the
decision-maker is willing to accept. For
example, different future uses may require that
different environmental standards be met.
Excessive sampling to detect contamination
below the levels required for the planned future
use can waste resources.
Keeping these goals in mind, the DQO process
guides the team to define items such as the
number and types of samples to be collected,
analytical detection limits, and certainty. After
these parameters (the DQOs) are established,
analytical methods and instrumentation can be
selected to develop the most cost-effective
sampling design that will meet these objectives.
Uncertainty — the chance of drawing an
incorrect conclusion — is addressed in Step 6 of
the DQO process. For critical decisions, such as
whether a Brownfields site can be safely reused
as a public recreation area, the degree of
certainty that the site will not pose a threat to
human health must be quite high in order to gain
public acceptance. Because the term "quite
high" is indefinite, the level of safety at the site
may be in question. Quantifying the amount of
uncertainty present in a decision clarifies how
confident the team can be that it is correct. For
example, some decision-makers may require
certainty of 90% to 95% before they feel
comfortable making the decision to allow reuse
of a site for public recreation. The only way to
make such a definitive statement about certainty
is by using a statistical sampling design. This
type of design is discussed in Section 4 of this
document.
Some stakeholders may demand a statistical
expression of certainty in cases where the
planned reuse will allow the public unrestricted
access to potentially contaminated media at a
site (e.g., residential or recreational reuse). For
less critical decisions, for example, industrial
reuse where much of the property is used for
buildings and parking lots, less certainty may be
acceptable or a qualitative statement may be
sufficient.
The DQO process controls the potential for
making decision errors due to uncertainty in the
data by helping the team set limits on the
CONCEPTS
Conceptual Site Model - the conceptual site model
is dynamic in nature. It is initially based on the
best-available information and is updated as
additional data becomes available during the site
assessment.
Uncertainty - the probability of making an
erroneous decision based on available data.
Null Hypothesis - an assumption that is tested by a
scientific investigation (e.g., environmental
investigation); the baseline condition assumed to
be true in the absence of contrary evidence or an
alternative hypothesis (e.g., the earth is flat unless
proven round, the site is dirty unless proven
clean); generally based on the case with the least
desirable consequences.
Decision Error - an incorrect conclusion about a
site (e.g., deciding site cleanup is not needed
when it really is) caused by using data that are not
representative of site conditions due to sampling
or analytical error.
False Negative Decision Error - accepting the null
hypothesis when it is actually false. Implications of
the false negative decision error depend on the
structure of the null hypothesis.
False Positive Decision Error - rejecting the null
hypothesis when it is actually true. Implications of
the false positive decision error depend on the
structure of the null hypothesis.
Screening Assessments - short site inspections
that may have already been conducted, or that the
team may need to conduct, to gather sufficient
data for an effective sampling and analysis plan.
The data gathered will be useful in the initial
conceptual site model.
-------
probability of making a decision error (i.e.,
decision error rate). A hypothesis about the
condition of the site is the basis for this
determination. The hypothesis, referred to as
the null hypothesis, should be designed to guard
against making the decision error that has the
most undesirable consequences. The null
hypothesis is derived from information in the
Statement of the Problem (Step 1 of the DQO
process), including what is known about the site,
the projected site reuse scenario, and the
resources available for study and cleanup.
Typical null hypotheses are the following: "the
site is clean enough" or "the site is too dirty for
the reuse scenario." The team will identify an
alternative hypothesis contrary to the null
hypothesis and the sampling and analysis plan is
then designed to test the null hypothesis by
providing strong evidence to the contrary.
Generally, the more severe consequences of
making the wrong decision at a Brownfields site
occur when the site is actually contaminated
above established health limits, but the decision-
maker acts on data that erroneously indicate that
the site is clean. In this situation, human health
could be endangered if reuse occurs without
cleanup. Therefore, the null hypothesis at a
Brownfields site is likely to be "the site is too
dirty for the reuse scenario," and the site
assessment is then designed to show that the site
is clean, which is the alternative hypothesis.
Additional explanation is provided under Step 6
of the DQO process.
Because of the limited funding for Brownfields
site assessments, it may not be possible to
collect data sufficient to achieve a desired level
of certainty in site decisions. Because
increasing certainty usually requires the
collection of more samples, it can be costly. If
the team can afford to collect and analyze only a
limited number of samples, decision-makers
must take care to communicate only what they
know about the environmental conditions at a
site and how confident they are in that
knowledge.
Limited funding highlights the need for a well-
planned investigation that capitalizes on time-
and cost-saving technologies. By following a
systematic planning process, such as the DQO
process, decision-makers will be able to strike
the best balance between what they want to
know about a property and what they can afford
to know about a property given the realities of
their budget.
The DQO process offers several benefits. By
using the DQO process, the team can establish
criteria for determining when data are sufficient
for site decisions. This provides a "stopping
rule" — a way for the team to determine when
they have collected enough data of sufficient
quality to achieve the desired objectives. In
addition, the DQO process helps the team
establish an adequate level of data review and
documentation. Data review is a process of
assessing data quality based on written
performance-based acceptance criteria (e.g.,
samples must be analyzed by the laboratory
within a specific period of time referred to as
the holding time). Data review also determines
whether the data satisfy the predefined DQOs.
Another benefit of the DQO process is that it
focuses studies by clarifying vague objectives
and identifying the decisions that should be
made prior to the selection of sampling and
analysis parameters. This gives the team
confidence that the data collected will support
the decisions concerning redevelopment of the
site.
The DQO Process
The outputs of the DQO process include the
information the team will need to complete most
of the Quality Assurance Project Plan (QAPP).
Forms D through N of the QAPP template
provide space for describing the sampling and
analysis plan (SAP). (See Appendix A.) The
pre-defined objectives and decision statements
that are a product of the DQO process form the
basis for the SAP. Section 4 of this document
discusses elements of the QAPP and SAP in
more detail with reference to the corresponding
forms in Appendix A.
Step 1: Stating the Problem
-------
The first step of any decision-making process is
to define the problem that has prompted the
study. The team will develop a concise and
complete description of the problem, which can
be documented on form C of the QAPP
template. This description provides the basis for
DQO development and is built on information
collected during the background investigation.
Project planning is perhaps the most critical
component of the Brownfields assessment
process, as it allows team members to determine
fully the goals and scope of sampling events and
the resources necessary to accurately
characterize the site. Therefore, it is important
that all interested parties (including project
managers, engineers, chemists, field sampling
personnel, statisticians, local government
officials, and the public) be involved in the
project from the conceptual design stage. Roles
and responsibilities can be documented on
forms A and B of the QAPP template.
The conceptual site model is an important part
of Step I, Stating the Problem. The conceptual
site model should be updated as additional
information becomes available, but should
initially illustrate the following:
• Potential chemicals of concern;
• Media in which these chemicals may be
present and to which they may migrate
(surface and subsurface soil, surface water,
groundwater, and onsite structures);
• Whether human or environmental receptors
(i.e., targets) at or near the site may be
exposed to contamination; and
• Current and anticipated land use.
The conceptual site model should include maps
and site diagrams that illustrate structures and
areas of potential contamination including
locations of chemical handling, storage, and
disposal. If facility records are unavailable, the
team may find information through State,
Resource Conservation and Recovery Act
(RCRA), and National Pollutant Discharge
Elimination System (NPDES) programs, and
local records offices.
Some information may be available from current
and past owners, lending institutions, and/or
environmental regulatory and real estate
agencies. Some States have laws that require
property owners to disclose the available reports
to prospective purchasers. These reports may
answer some of the team's questions about a
site.
Two common examples of "screening"
assessments are EPA's Preliminary Assessment
(PA) and ASTM's Phase I Assessment (Phase
I), which are briefly described in Exhibit 3
below. Because the scope of these assessments
varies, they may not answer all of the questions
the team wants to address. For example, to find
a discussion about offsite receptors, the team
will want to look at a PA rather than a Phase I.
EXHIBIT 3
COMPARISON OF PRELIMINARY ASSESSMENTS AND DUE DILIGENCE SITE ASSESSMENTS
Preliminary Assessment (PA)
A site may consist of the legal property boundaries and other
areas where contaminants have come to be located.
Level of effort is approximately 120 hours.
Examines all available site-related documents during file
searches of Federal, State, and local agencies.
Surveys on- and offsite pathway potential targets.
Due Diligence - ASTM Phase I
A site within the context of a Due Diligence study is limited to
the legal boundaries of the subject property.
Level of effort is approximately 40 hours.
Examines all "practically reviewable" and "reasonably
ascertainable" site-related documents.
Surveys onsite targets within property boundaries.
-------
EXHIBIT 3
COMPARISON OF PRELIMINARY ASSESSMENTS AND DUE DILIGENCE SITE ASSESSMENTS
Considers hazardous substance migration to offsite human
and environmental targets.
Identifies sources and estimates extent of contamination on-
and offsite.
Does not evaluate sources within secure buildings.
Petroleum products are not considered.
Considers only onsite targets and impacts to the site from on-
and offsite sources.
Identifies sources and estimates extent of contamination only
within property boundaries.
Evaluates sources within all buildings (e.g., asbestos).
Petroleum products are considered.
Source: Site Assessment: A Comparison of Federal Program Methods and Commercial Due Diligence. Journal of Environmental Law & Practice.
p.15-25 March/April 1997
Previously collected analytical data are
particularly important to the conceptual site
model. This information may identify some of
the chemicals onsite and their locations. It also
can indicate the variability of contamination
onsite. The sampling and analytical methods
that were used previously may also prove
helpful to the Brownfields sampling and
analysis plan.
Especially when using previously collected
analytical data, the team needs to review the
information for accuracy and completeness. If
the data are several years old, reported
analytical data and site features may not
represent current site conditions.
Aerial photos are often helpful in reconstructing
the history of a site with multiple prior owners.
Aerial photos are available through
http://mapping, usgs.gov.
The team should also ensure that the conceptual
site model illustrates site conditions that may
lead to an unacceptable threat or that are based
on current and projected future land uses. For
example, if local groundwater is used in
households and businesses, the physical
characteristics of the soil and local
hydrogeology should be understood to better
assess the threat to groundwater resources.
Finally, the team should document the available
resources and relevant deadlines for the study in
the problem statement. This description should
specify the anticipated budget, available
personnel, and contractual vehicles, where
applicable.
An example problem statement follows:
The Springfield team is considering Brownfields
redevelopment of a site comprised of 10 acres
of waterfront property historically used for truck
repair. The site, depicted in Exhibit 4, is
currently fenced and abandoned. Aerial photos
indicate a repair garage located on the western
portion of the property, an office and employee
parking on the east end near the water, and
truck storage between these two areas. A
central fence separates the garage and truck
storage area from the office section of the
property.
Facility records indicate that an underground
fuel storage tank is located behind the northeast
corner of the garage. A previous site
assessment mentions the removal of four
barrels of toluene by the County health
department in 1983 and the presence of a
second storage tank forspent solvent disposal
south of the garage. Analytical results from the
previous site assessment indicate that total
petroleum hydrocarbons (TPH — constituents of
fuel) and volatile organic compounds (VOCs —
constituents of solvents) are present in soils
near the areas of former storage of toluene and
the spent solvent tank. Interviews with past
employees indicate that trucks were washed on
occasion in the truck parking area of the site.
The site appears to present a threat via
exposure to soils; groundwater and surface
water may also be contaminated. The site is
located in a commercial/residential zone. The
city wants to redevelop the site for commercial
and public use.
EXHIBIT 4
THE SPRINGFIELD SITE
-------
Step 2: Identifying the Decision
While environmental field investigations are
designed to satisfy a broad array of objectives,
the goal of a Brownfields site assessment is to
collect adequate environmental data for
decision-makers to determine if the site is
suitable for a specific reuse. This determination
may require several separate but related
decisions.
The "decision statements" usually take the form
of questions that the study will attempt to
answer. Form C of the QAPP template
(Appendix A) provides space to document these
decisions. The decision statements are
important because they indicate alternative
actions and decision performance criteria in
later steps of the DQO process.
Decision Statements for the Springfield site
might include the following:
Historical information indicates that the eastern
portion of the property has not been used for
chemical management activities; the waterfront
section may be reused as a park or other
publicly accessible facility. To assess the
feasibility of this option, the team will make the
following decisions: Will the site need to be
cleaned up before it can be reused as a park?
If cleanup is too expensive, can the site be
redeveloped for another use?
Because historical information indicates that the
western portion of the site is at least partially
contaminated, this area is being considered for
commercial reuse. The team will make the
following decisions to facilitate commercial
financing: Is the site clean enough to attract a
private sector developer? Have issues of
concern to lenders been addressed? What
level of cleanup or other actions is necessary to
answer the questions of developers and
lenders?
For the Springfield site, the team will make two
distinct decisions based on different projected
reuse options for different portions of the site.
Step 3: Identifying Inputs to the Decision
The team should identify the information
needed
to resolve the decision statement(s) and
determine how this information will be
obtained. For example, if groundwater use is a
consideration in the site reuse scenario, the team
should identify how samples of groundwater
will be used to support the reuse decision. The
team should review the conceptual site model to
learn whether existing groundwater data provide
the information needed for the study and
identify data gaps in the model.
The team will identify the characteristics of the
site that need to be measured based on a
threshold value (i.e., a drinking water action
level) that provides the criterion for choosing
among alternative actions. Regulatory
standards, such as State drinking water
standards, usually form the basis for action
levels. If no regulatory threshold or standard
can be identified during this step, the team
should identify information needed to develop a
realistic concentration goal. This information
TRUCK
PARKING AREA
SOLVENT
i STORAGETANK
FORMER LOCATION OF GARAGE
DRUMS CONTAINING TOLUENE
will be critical to the final sampling design.
Later, in Step 7 of the DQO process, Optimizing
the Design, the input identified during this step
is reviewed and refined. The team should be
aware that the decisions made during this step
are "draft" and may be changed during
optimization. The team will make the final
decision on which analytical methods to use in
Step 7.
10
-------
The products of this step include a list of
informational input needed to make the decision
and a list of environmental characteristics that
will be measured. In essence, the output of this
step are actually the input to the decision.
Some input that might be identified for the
Springfield site include the following:
All soil samples in the waterfront area slated for
public access reuse will be collected to depths
specified in State regulations. The samples will
be analyzed for chemicals likely to be present
using a State, EPA, or other analytical method
that meets the objectives of the Brownfields
assessment. To rule out the possibility of other
contaminants of concern, one representative
sample will be analyzed for a broad spectrum of
compounds, including those whose presence is
considered unlikely.
Because previous analytical data from the
western portion of the site indicate
contamination with TPH and VOCs in the former
toluene storage and spent solvent storage
areas, field techniques will be used to confirm
contamination in these areas. Data gaps for
this portion of the site include the following:
whether the underground storage tank has
leaked and if so, whether contamination has
reached groundwater; whether contamination
exists in the truck washing area; and remaining
areas of the site for which no previously
collected data exist. Soils in these areas will be
analyzed fora combination of TPH and VOCs
using a State, EPA, or other analytical method
that meets the objectives of the Brownfields
assessment. To rule out the possibility of other
contaminants of concern, at least one
representative sample will be analyzed fora
spectrum of compounds including those whose
presence is considered unlikely. Groundwater
at the site will be collected and analyzed for
TPH and VOCs using a State, EPA, or other
analytical method that meets the objectives of
the Brownfields site assessment.
Step 4: Defining the Boundaries of the Study
The boundaries of the study refer to both spatial
and temporal boundaries. To define the
boundary of the decision, the team should
identify the geographic area within which all
decisions apply. A spatial boundary could be
the property boundary, a portion of the property,
potential exposure areas, or an area with a
specific reuse projection. For example, a soil-
sampling boundary may include the top 12
inches of soil where truck washing reportedly
took place.
The scale of decision-making is the smallest
area, volume, or time frame of the media for
which the team wishes to make a decision. For
example, to decide whether to clean up the top 2
feet of surface soil, the scale of decision-making
might be related to the method of cleanup; if
contaminated soil will be hauled in a 5-ton
capacity truck, the boundary of decision-making
might be a 2-foot deep, 65-square foot area.
This example is based on the practicalities of
cleanup rather than exposure scenarios.
The team should also define the temporal
boundaries of the decision. The team may find it
impossible to collect data over the full time
period to which the decision will apply. The
team will have to determine the most
appropriate part of that period for gathering data
that reflect the conditions of interest.
Practical boundaries that could also affect
sampling are identified in this step. For
example, seasonal conditions or the
unavailability of personnel, time, or equipment
may make sampling impossible. Form D of the
QAPP template contains space for the team to
document the boundaries of the investigation,
including a project timeline.
Boundaries for the Springfield studies might
include the following:
The park reuse decision applies to the area of
the site east of the fence that divides the garage
and truck parking area from the office area.
This portion of the property is not further
subdivided.
The commercial reuse decision applies to the
areas west of the fence. This portion of the
property is further subdivided into the
groundwater and soil boundaries. The soil is
further subdivided into the subsurface soil
around the underground storage tank, the
11
-------
surface soils in the truck washing area, and the
soils in the remaining portion of the site.
Temporal boundaries are addressed by
performing studies when personnel are
available, during the dry season, following
notification of the local community and receipt of
authorization from site owners.
Step 5: Developing a Decision Rule
The purpose of developing a decision rule is to
integrate the output from the previous steps of
the DQO process into a statement that estimates
the parameter(s) of interest, delineates the scale
of decision-making, specifies the action level,
and describes the logical basis for choosing
among alternative actions.
A decision rule is usually a comparison of a
statistical parameter of interest (such as the
average level of arsenic in soil, or the maximum
level of toluene in groundwater) to a specific
action level. The action level is the contaminant
threshold that defines the conditions around
which the team should select among the
alternative actions and/or take different
directions to solve the problems. For example,
if the action level is exceeded, the team may
choose to clean up the site.
The output for this step is an "if...then"
statement that defines the conditions that would
cause the team to choose among alternative
courses of action. Form D of the QAPP
template provides space to record these decision
rules.
Decision rules for the Springfield site
include the following:
If the average levels of TPH and VOCs in soil
samples are less than selected action levels,
then the redevelopment project can proceed.
If the average levels of TPH and VOCs soil
samples are higher than selected action levels,
then cleanup to the action level is required prior
to reuse.
If the site assessment indicates that
groundwater has become contaminated from
site activities, the team should contact the State
to discuss the impact on redevelopment
scenarios for the western portion of the site.
Step 6: Specifying Limits on Decision Errors
Because of the limitations of environmental
sampling and analysis, the team runs the risk of
making the wrong decision because of
incomplete information. Sampling may not
capture all of the variations in concentrations,
and analyses can only estimate the "true" value.
The team must therefore develop means to limit
or control the impact of errors in estimations on
the likelihood of making a decision error. These
limits should be incorporated into the sampling
and analysis plan during Step 7 of the DQO
process.
Decision-makers are interested in knowing the
true state of some feature of the environment
(e.g., the average concentration of arsenic in the
top twelve inches of soil). The measurement
data that describe this feature can be in error.
Sampling "error" occurs when the sampling
scheme (which determines the sampling
locations) does not adequately detect the
variability in the amount of contaminant in the
environmental matrix from point to point across
the site. Measurement errors can occur during
sample collection, handling, preparation, and
analysis when standard procedures as described
in the SAP are not followed. Report preparation
is another source of error. The sum of the
sampling and measurement errors is called the
total study error.
As mentioned earlier, sampling and
measurement errors can lead to errors in data
thereby causing the decision-maker to select the
wrong course of action. The DQO process helps
the team control these errors through
development and testing of the null hypothesis
and selection of limits for erroneously accepting
or rejecting the null hypothesis.
Although some Brownfields sites are only
perceived to be contaminated, many may be
contaminated at levels exceeding health-based
action levels. In selecting the null hypothesis
12
-------
for a Brownfields site, the team should choose
the circumstances that can have the most severe
consequences. The null hypothesis in Step 6
will usually be "the site is contaminated" and
needs some level of cleanup (i.e., the site is
contaminated above certain action levels). To
test this hypothesis, the team must provide
ample evidence to the contrary or prove that the
site is clean (i.e., any contamination present is
below certain action levels). Erroneously
accepting the null hypothesis as true (false
negative decision error) may unnecessarily
increase the cost of site cleanup because
decision-makers may believe that action is
warranted when it is not. Erroneously rejecting
the null hypothesis (false positive decision
error) can increase the risk of exposure at a
property because a decision-maker may believe
In the Springfield site example, because the
projected reuse of the eastern portion of the
property involves its unrestricted public use, the
null hypothesis is that the eastern portion of the
property is dirty. The team will need to show it
is clean. This hypothesis reflects what most of
the public will probably assume before any
environmental studies occur at the site. The
team already has data showing that the western
portion of the site is contaminated.
To completely avoid any decision errors (100%
certainty), the team would have to sample all
surface soil related to the decision whether to
clean up the surface soil at the site. Because
this is financially infeasible, the team must
collect a number of representative samples from
the area in a manner that reduces the decision
error rate. The analytical results from these
samples will be translated into an estimation of
contamination on part of or the entire site. The
more samples collected, the greater the certainty
the team will have in its decision; however, the
more samples collected, the more costly the
investigation. The team needs to balance the
level of certainty desired with the cost of that
certainty (the cost of additional sample
collection and analysis). Form D of the QAPP
template can be used to document limits on
decision errors.
that no action is warranted when it is.
In contrast to the situation above, if the null
hypothesis were "the site is clean enough," the
team would need to show the presence of
contamination above certain action levels. In
this case, erroneously accepting the null
hypothesis (false negative decision error) can
increase the risk of exposure at a property
because a decision-maker may believe that no
action is warranted when it is; whereas,
erroneously rejecting the null hypothesis (false
positive decision error) may increase the cost of
site cleanup. The amount and quality of data
collected will depend on how the team plans to
control decision error rates.
Limits on the decision errors for the
Springfield site may include the following:
Because of public access reuse projections,
reuse of the eastern portion of the property may
result in children coming into contact with site
soils; therefore, the null hypothesis is that the
site is contaminated. The probability of false
positive decision errors (erroneously rejecting
the null hypothesis or deciding that the site is
clean when it is not) should therefore be
minimized as much as possible. Errors that
increase the probability of leaving soils in place
when they contain substances at levels greater
than the action level — false positive decision
errors — will be considered acceptable no more
than 10% of the time.
Errors that increase the probability of cleaning
up soils when that action is not required — false
negative decision errors — will be considered
acceptable only 10% of the time.
Limits on the probability of errors in decision-
making for the western portion of the site are
not needed because the sample design is
intended to simply define the boundaries of
known contamination.
Step 7: Optimizing the Design
The team should evaluate the cost of sampling
design options that meet the DQO constraints
and select the most resource-effective option.
13
-------
The chosen alternative that meets the DQOs
may be the lowest cost alternative, or it may be
a relatively low-cost design that still performs
well when design assumptions change. Because
the role of statistics is very important when
developing a sampling design that achieves
specified decision error rates, the team member
with statistical expertise should be consulted at
this stage. (See Section 4 for a more thorough
discussion of sampling strategies.)
The team should review the output of the
previous steps to determine exactly how the
selected limits on decision errors will define the
required number and location of samples and the
types of analyses. This step frequently involves
refinement of initial design parameters.
Many different strategies could be employed to
optimize the investigation at the Springfield site;
only a few are presented here. Section 4
describes more options available to the team to
reduce the project's sampling and analysis costs
while meeting DQOs.
Optimizing the SAP for the Springfield site:
The waterfront portion of the property is not
likely to be significantly contaminated because
waste-related activities were neither
documented nor likely conducted there. It
should therefore be possible to reject the null
hypothesis — reject the assumption that the site
is dirty by showing that the site is clean — by
collecting 40 to 50 samples using a statistical
design that maintains a 10% false positive
decision error rate. This allows a 10% chance
that the decision-maker will consider the site
clean when it is not.
For the western portion of the site, the team will
use field techniques to confirm previously
detected levels ofVOCs in the former toluene
storage and spent solvent storage areas. The
same techniques will be used in areas
suspected to be contaminated with VOCs and
TPH — the areas around the underground
storage tank and truck washing area.
This example assumes the variability in site
contamination is approximately 25%. The
effect of this factor on the sampling design is
explained in Section 4. Errors may also result
from mishandling of samples or improper field
procedures. Section 4 of this document
discusses the effect of variability, cross-
contamination and other problems on decision
errors and how quality control samples can be
used to identify and control the impact of some
of these effects.
Estimating Costs
The cost of conducting a Brownfields site
assessment is driven by the adequacy of
available historical data, the type and level of
contamination, the site assessment technologies
used, and the property's projected site reuse.
Estimating costs for Brownfields site
assessments creates unique challenges.
Although the tendency may be to expedite the
planning period, care must be exercised to
ensure that the interest of site owners, investors,
purchasers, and lenders is maintained. The cost
estimates should therefore be developed quickly
while preserving the highest level of accuracy
possible. As stated earlier in this section,
increasing the quality of environmental
measurements will likely increase the cost of a
Brownfields site assessment.
Developing Brownfields site assessment cost
estimates is hindered by a lack of detailed cost
estimating literature that applies to typical
Brownfields sites. The majority of available
information is based on large Federal
government and private sector sites such as
abandoned rail yards and steel mills, not the
smaller former industrial sites such as
automotive repair shops or metal finishing
facilities. Some guides that may assist in the
development of cost estimates are listed in
Appendix C.
Once the team has selected the final sampling
design based on all considerations, including
cost, it should properly document the design.
This will protect the efficiency and
effectiveness of predefined field sampling
procedures, quality control procedures, and
statistical procedures for data analysis. Forms D
through N of the QAPP template provide space
14
-------
for the final sampling design. All drafts of the
sampling and analysis design generated during
the DQO process should be discarded once the
final sampling design is selected and
documented.
A complete discussion of DQOs and their use in
developing the SAP can be found in the
following documents available through EPA's
Quality Assurance Division (QAD) website
(http://es.epa.gov/ncerqa/qa/qa_docs.html):
Guidance for the Data Quality Objectives
Process, September 1994. EPA QA/G-4: EPA
600-R-96-055.
Guidance for Quality Assurance Project Plans,
February 1998. EPA QA/G-5 EPA 600-R-98-
018.
EPA Requirements for Quality Assurance
Project Plans, Draft Final, October 1997. EPA
QA/R-5 (final publication pending).
15
-------
SECTION 4. QUALITY ASSURANCE PROGRAMS &
SAMPLING DESIGN STRATEGIES
I his section describes basic sampling design
I strategies that can be used to optimize
Brownfields sampling and analysis plans (SAP);
introduces the QA and QC concepts; and
discusses terminology used during
implementation of Brownfields site assessment
and the assessment of the resulting data. The
concepts discussed in this section, related to
preparation for collection, analysis, and review
of site assessment data, will be helpful in the
development and preparation of the QAPP for a
Brownfields site assessment.
The last part of Section 4 discusses how the
QAPP integrates all technical and quality
aspects for the life cycle of the project —
including planning, implementation, and
assessment — to produce a project-specific road
map for obtaining the type and quality of
environmental data needed for a specific
decision. Appendix A of this document
contains forms that the team can use directly or
as a guide for writing the QAPP. Some of the
forms may not be necessary depending on site
conditions and sampling design. The following
elements related to sampling should be included
in the QAPP:
• Sampling design (form E of the QAPP
template);
• Sampling methods (form F-l);
• Sample handling and custody (form K);
• Analytical methods (forms F-l and F-2);
• Quality control (form M);
• Instrument/equipment testing, inspection,
and maintenance (forms G and I);
• Instrument calibration frequency (form J);
and
• Data management (form N).
The design and extent of a Brownfields site
assessment will be dictated largely by the
conceptual site model, the availability of
resources, and the required data quality and
level of quality control exercised. The DQO
development process should define all aspects
of the sampling design and these details should
be documented in the QAPP.
CONCEPTS
Sampling Design - scheme for sample collection
that specifies the number of samples collected in a
biased and/or unbiased pattern, as grabs or
composites, etc.
Biased Sampling - collection of samples at
locations based on the judgment of the designer.
Statistical Sampling - collection of samples in a
systematic or random manner.
Multi-phase Sampling - sample collection in
multiple stages; data are used to plan subsequent
rounds.
Adaptive Sampling -when multi-phase sampling is
performed in a single mobilization using field
analytical methods which provide results in 24
hours or less.
Low Bias Analytical Error - when analytical data
indicate that a substance is not present above a
specified concentration, when in fact it is. Low
bias errors can increase the risk of exposure
because a decision-maker may be led to conclude
that no action is warranted when it is.
High Bias Analytical Error - when analytical data
indicate that a substance is present above a
specified concentration, when in fact it is not. High
bias errors can increase the cost of cleanup
because a decision-maker may be led to conclude
that action is warranted when it is not.
Grab Sample - sample from a single location
useful for identifying and quantifying chemicals in
an area where contamination is suspected.
Composite Sample - composed of more than one
discrete sample taken at different locations, useful
to quantify average contamination across a site.
Analytical Method - procedures used to identify
and/or quantify chemicals in a sample.
Measurement Error - the difference between the
true sample value and the measured analytical
result.
Broad Spectrum Analysis - analytical procedure
capable of identifying and quantifying a wide range
of chemicals.
Field Analysis - measurement taken in the field;
results are quick and quantitative or qualitative.
Data Useability - adequacy of data for decisions;
determined by comparing resulting data quality
with predefined quality needs documented in
QAPP (defined during the DQO process).
16
-------
Quality Assurance
QA is an integrated system of management
activities that are part of a project's planning,
implementation, assessment, reporting, and
quality improvement. These activities ensure
that products are of the type and quality
expected.
QA will be an integral part of a Brownfields site
assessment because it provides the road map to
all activities necessary to collect data of known
and adequate quality. Data of adequate quality
will give the team a sufficient level of
confidence in the data to make informed
decisions about the redevelopment of the site
including the following: the threat posed by the
contamination, potential site remediation
alternatives, and additional projects needed to
prepare a site for redevelopment.
The QAPP provides the framework for the
Brownfields project's QA program by outlining
activities that promote the collection of data
with the accuracy and precision required for the
project. Some elements of the QA program
include the following:
• Staff organization and responsibility (form
B of the QAPP template);
• Standard Operating Procedures (SOPs) for
sampling and analytical methods (form F-l);
• Field and laboratory calibration procedures
(forms H and J);
• Routine and periodic quality control
activities (form M);
• Data assessment procedures (form O); and
• Data reduction, validation, and reporting
procedures (forms P, Q-l, Q-2, and R).
Quality Control
Quality control (QC) is integral to the success of
a QA program. It is the overall system of
technical activities that measure the
performance of a process against defined
standards to verify that they meet predefined
requirements. Since errors can occur in the
field, the laboratory, or the office, it is necessary
for QC to be part of each of these functions.
An example of a QC activity is collection of a
rinsate blank sample. When equipment is
cleaned and reused in the field, the sampling
team will collect a sample of the spent rinse
water. Analysis of this sample will show
whether the equipment was sufficiently cleaned
or if hazardous substances have remained on the
equipment that will contaminate the next
sample. The data from the rinsate blank
measure the performance of decontamination
procedures in the field. If contaminants are
found in the rinsate blank, other samples
collected with the same equipment may also be
contaminated and may not meet the stated
requirements established by the DQOs. QA and
QC procedures are discussed below in the
context of expected environmental measurement
activities for Brownfields site assessments.
QA and QC parameters apply to the two primary
types of data — definitive and nondefinitive
data — and whether the data collection activity
is associated with field measurements or
laboratory measurements. The following boxes
provide definitions of these terms.
Definitive Data
Definitive data are documented to be appropriate
for rigorous uses that require both hazardous
substance identification and concentration and
are generated using methods that produce data
suitable for scrutiny of the data validation and
useability criteria described later in this section.
Definitive data are analyte-specific, with
confirmation of analyte identity and
concentration. Methods produce tangible raw
data (e.g., chromatograms, spectra, numerical
values) as paper printouts or computer-generated
electronic files. Definitive data may be generated
at the site or at an offsite location, as long as the
QA/QC requirements of the method are satisfied.
For data to be definitive, either analytical or total
measurement error should be determined. For
further guidance on definitive data, refer to
Guidance for Performing Site Inspections Under
CERCLA, Interim Final, and Guidance for Data
Useability in Risk Assessment. These
documents are available through NTIS at
http://www. ntis.gov/envirn/envirn. htm.
17
-------
Nondefinitive Data
Nondefinitive data are frequently collected during
the first stage of a multi-phase screening
assessment using rapid, less precise methods of
analysis with less rigorous sample preparation.
Nondefinitive data can provide analyte
identification and quantification, although both
may be relatively imprecise. Typically, 10% of
nondefinitive samples or all critical samples are
confirmed using analytical methods and QA/QC
procedures and criteria associated with definitive
data. Nondefinitive data without associated
confirmation data are of unknown quality.
Qualitative, nondefinitive data identify the
presence of contaminants and classes of
contaminants and can help focus the collection of
definitive data, which is generally the more
expensive of the two.
Each site assessment should be guided by a
detailed description of the work to be performed
in the SAP. The SAP takes the conceptual site
model developed in the DQO process and
translates it into a sampling and analysis design
that identifies where, how, how many, and what
types of samples will be collected; how the
samples will be stored and transported; how,
when, and by what method the samples will be
analyzed; and what procedures and records will
be used to track the samples through the
process. Depending on the complexity of the
Brownfields site, multiple SAPs may be needed.
QA and QC parameters should be described in
detail for each of these steps, and include
specific corrective actions to be taken if
difficulties are encountered in the field.
Guidance documents on sampling methods are
listed in Appendix C.
Sampling Design Strategies
Sampling design strategies should factor in the
conditions unique to the site being considered
for redevelopment, including data gaps in the
conceptual site model, exposure potential,
projected site reuse, and available resources.
Step 7 of the DQO process should identify
several possible sampling design strategies.
Some of the variables that may be used in these
strategies to bring down the cost of the project
are described below. The details of the selected
sampling design can be documented on forms E
through N of the QAPP template. The overall
sampling design is described on form E.
Unique site conditions that may call for a certain
strategy include a site with buildings slated for
reuse. In this situation, non-routine sampling
and analysis may be required for unusual sample
matrices, such as building materials.
The main sampling design decision is whether a
statistical (probability based) or judgmental
(nonrandom or biased) sampling design should
be employed. Judgmental sampling is a useful
design when the team wants to characterize
areas of suspected contamination. Statistical
sampling designs are suited for evaluating
trends and estimating the distribution of
contaminants. Sometimes both judgmental and
statistical sampling is required on a single site.
An important distinction of statistical sampling
designs is that they are usually required when
the level of confidence needs to be quantified.
For surface soil sampling in residential reuse
scenarios, a statistical sampling design is likely
to be chosen because a quantitative statement of
the decision error will be needed to show that
the level of any contamination at the site is safe.
Industrial reuse may not require as rigorous a
result or may be possible with a qualitative
statement at sites where exposure is not
possible.
When historical data are unavailable to indicate
discrete areas of contamination, i.e., hot spots, a
useful strategy is to collect samples along a grid.
Grid sampling is designed to cover the entire
site with samples collected at regularly spaced
intervals. The goal of grid sampling is to reduce
the probability of making a decision error. It
allows the calculation of the probability of
remaining undetected hot spots. This technique
could be particularly useful on the previously
unsampled western portion of the Springfield
site where industrial activities are known to
have been carried out. Different types of grids
18
-------
are available and are generally selected based on
the difficulty of layout in the field, sufficient
detection capability, and cost.
Factors affecting the success of grid sampling
include the size and shape of the grid and the
size and shape of the hot spot. If the hot spot is
small relative to grid spacing, the probability of
missing a hot spot will be relatively high. The
cost of grid sampling is determined by the
number of samples, which is determined by the
grid shape and spacing. Closer spacing yields a
higher probability of detecting a hot spot.
If data are needed to determine if groundwater
onsite is contaminated, a statistical sampling
design would be unnecessary and impractical
because of the cost of installation of
groundwater wells. Judgmental groundwater
samples might be collected from nearby wells,
or if the budget can bear it, one or more wells
may be located where the contamination is most
likely to be present. Appendix C identifies
several groundwater sampling and monitoring
guidance documents.
Errors in judgmental sampling may come from
cross-contamination of samples in the field or
improper calibration, maintenance, and use of
field equipment (these issues are discussed later
in this document). To ensure useable data if
these problems arise, the field team should have
previously defined alternative options, such as
routing samples to a fixed laboratory. QC
samples (discussed later in the section) will also
be helpful.
Because of the need for quantification of the
decision error in the public park reuse scenario
— a sampling design that translates the results
from a limited number of samples to an estimate
of the contamination on part of the site — a
statistically based sampling design will be
required.
While this document does provide some
description of how statistics are used to assist in
the decision process, statistical expertise and
support should be enlisted throughout the
project to ensure a defensible environmental
decision is possible at the end of the project.
For more information, refer to EPA Quality
Assurance Division's (QAD) Guidance for the
Data Quality Objectives Process and other
documents listed in Appendix C. To
quantitatively demonstrate that a specific level
of certainty has been achieved for site decisions,
individual data sets must be of sufficient quality,
and overall statistical analysis (which integrates
information from the individual data sets into a
site decision) must be able to support the site
decision.
Section 3 introduced the concept of testing the
null hypothesis and its relation to decision
errors. A false positive decision error is the
erroneous rejection of the null hypothesis; a
false negative decision error is the erroneous
acceptance of the null hypothesis. At the
Springfield site, the null hypothesis is that the
site is dirty. The false positive decision is that
the site is clean when it is dirty; the false
negative decision is that the site is dirty when it
is clean.
If a goal of the project is to document whether
onsite contaminant concentrations are higher
than background concentrations, previous data
may help estimate the relative difference
between the background concentrations and the
onsite concentrations. By using that estimation
and the predefined degree of statistical certainty,
the team can calculate the number of samples
required. The greater the difference between
background and site contaminant levels, the
easier it is to document and quantify that
difference; therefore, fewer samples are needed.
The amount of contaminant variability onsite
also directly impacts the planning and
implementation of sampling design. The higher
the variability, the greater the number of
19
-------
samples that have to be collected to adequately
document that variability. If previous data can
help estimate the amount of variability,
sampling protocols can be optimized during
project planning.
Exhibit 5 can be used in some cases to optimize
the sampling design by keeping samples to a
minimum while maintaining a known level of
confidence. It illustrates the number of samples
required for selected decision error rates given a
statistical sampling design, the requirement to
differentiate the site levels from background
levels (minimum detectable relative difference
(MDRD)), and a 25% variability in contaminant
levels onsite. It also assumes normally
distributed data; for lognormally distributed
data, the data should be transformed to a normal
distribution before the methods described here
can be applied.
EXHIBIT 5
NUMBER OF SAMPLES REQUIRED TO
ACHIEVE GIVEN RATES OF FALSE
POSITIVE AND NEGATIVE, AND MDRD*
False
Positive
10%
10%
20%
30%
20%
20%
False
Negative
10%
10%
10%
20%
20%
10%
MDRD
10%
20%
20%
10%
20%
40%
Number
Samples
42
12
8
19
5
3
*Number of Samples is based on known variability in
contaminant levels represented by a coefficient of variation
(CV) of 25%, random sampling design, and normal
distribution.
Minimum detectable relative difference (MDRD) is
used when discriminating site levels from background levels.
It is the percent difference between the two detected levels.
Source: EPA. Guidance for Data Useability in Risk
Assessment. April 1992.
These parameters, the MDRD and the variability
in contaminant concentrations (coefficient of
20
-------
variation (CV)), can be estimated based on
previously collected data at the site being
assessed or by using data from sites with similar
uses and substances. If the CV is low — the
concentrations on the site do not vary greatly —
fewer samples are needed to achieve the same
decision error rates. The MDRD represents the
difference between concentrations of substances
on the site and substances in the background.
The greater the difference, the easier it is to
distinguish between background and site levels
and therefore, fewer samples are needed for the
same decision error rate.
An underestimation of the actual variability in
existing contamination is the most likely reason
the results would fall short of the desired
confidence level. The team should prepare to
perform the necessary calculation as soon as the
data results become available to determine if the
desired confidence level is being met. If not,
costly resampling or reanalysis of critical
samples may be necessary.
Exhibit 5 also provides an overview of the
interplay of site characteristics and decision
error rates in the sampling design by illustrating
the number of samples needed to meet specified
decision error rates in given circumstances. In
general, the number of samples increases as the
desired error rate decreases. The exhibit shows,
for example, that to achieve a false positive
decision error rate of 20% requires 4 fewer
samples than to achieve a rate of 10%, other
factors being equal. For a given decision error
rate, the number of samples also tends to
increase with decreasing MDRD. For example,
to achieve the same decision error rate, 42
samples are required if background and onsite
contaminant levels differ by only 10%. If
background and onsite levels differ by 20%,
only 12 samples are required to reach the same
decision error rate.
A change in the coefficient of variation, the
factor held constant in Exhibit 5, also can affect
the number of samples required for a given
decision error rate. In general, as the variability
in the concentration of a chemical of concern
increases, more samples are needed to
adequately represent the onsite concentrations
of that chemical.
For the public park reuse portion of the property
in the Springfield example in Section 3, 40 to 50
samples collected in a statistical design
(assuming no need to compare to background
(MDRD) and a variability factor of 25%) will
produce a statistical confidence of 10% false
negative and positive decision error rates
(assuming normally distributed data).
If the Springfield team had not optimized the
sampling and analysis design through the DQO
planning process, either too many or too few
samples may have been collected.
• If too many samples were collected and
unnecessary analyses were conducted,
money that could have been spent on other
projects would have been wasted.
• If too few samples were collected, the data
may be insufficient to confidently allow safe
reuse of the property.
Multi-phase Investigations
A single sampling event may not provide an
adequate characterization of the contamination
onsite, especially when the conceptual site
model contains significant data gaps. In these
situations multi-phase sampling may be helpful.
The need for this sort of investigation should be
identified during the DQO process.
Based on a review of existing site assessment
reports, the team should determine if the
previous data are of sufficient quality and
quantity to guide the sampling design or if some
preliminary information must be obtained to
properly plan for collection of the predefined
samples. The team should base the assessment
of existing data on a review of the documented
quality of the data, and the decision rule criteria
established as part of the DQO process.
If a multi-phase assessment is selected,
preliminary activities should be developed using
guidelines found in ASTM El527 (Standard
21
-------
Practice for Environmental Site Assessment:
Phase I Environmental Site Assessment
Process) or a similar, accepted protocol. When
historical information is scarce, preliminary
sampling designs may include measurement of
contamination in soils, groundwater, and surface
water. To develop this information relatively
inexpensively, field technologies may be used,
including real-time detection instruments such
as a photoionization detector or organic vapor
analyzer. These tools provide ranges of
concentrations of classes of substances. More
definitive results can be obtained with
immunoassays and x-ray fluorescence
instruments, which are discussed later in the
document. Results of the preliminary study
should enable the team to plan a more focused
and cost-effective sampling design.
When no previous data are available to estimate
the variability of the contamination at the site, a
multi-phase investigation may be useful. A
preliminary investigation may be used to
calculate an estimate of variability in site
contaminant concentrations; this estimate can
then be used to determine the number of
samples required to achieve the specified
confidence level or decision error rate. Not
knowing the actual variability in existing
contamination is the most likely reason for the
results to fall short of the desired confidence
level.
Because of the expense of multiple field
mobilizations, a dynamic work plan that
combines two phases of sampling into a single
mobilization would be helpful in keeping down
costs. A dynamic work plan combines field
technologies that quickly provide the data
needed for planning the next phase of data
acquisition with adaptive sampling designs that
are flexible enough to respond to data generated
in the field. For example, field data could be
collected to demonstrate the variability of
chemical concentrations, which influences the
number of samples required to reach a specific
decision error rate.
Types of Samples
The types of samples to be collected and
analyzed are determined during the DQO
process. Like other variables of the sampling
design, the type of sample collected is
dependent on the team's predefined decision
error rate, required degree of accuracy, the
spatial and temporal variability of the media,
and the cost. The paragraphs below discuss
grab and composite samples and how they can
be used in the sampling design.
A sample can be collected discretely or as a
composite sample. Discrete samples, called
grab samples, are taken at a single location and
are useful for identifying and quantifying
chemicals in areas of a site where contamination
is suspected. For example, grab samples
collected around tanks and drums can answer
the questions of whether and what substances
may have leaked from the tanks and drums.
To identify average contamination across a site,
composite samples may be more appropriate.
Composite samples are composed of more than
one discrete sample taken at different locations.
The discrete samples are mixed to obtain a
homogeneous single composite sample and
analyzed. Composite sampling allows sampling
of a larger area while controlling laboratory
analytical costs because several discrete samples
are physically mixed and one or more samples
are drawn from the mixture for analysis. The
drawbacks of composite samples are related to
the averaging of the contamination levels in
discrete samples. Compositing minimizes the
significance of low levels of contamination and
may mask locations where contamination is
above the action level. The number of
composite samples and the number of individual
samples within a composite sample should be
based on the decision error rate goals
established during the DQO process.
Collection of composite samples was not
recommended at the Springfield site described
in Section 3 because the samples were to be
analyzed for compounds that may volatilize
during mixing. If the contaminant of concern
was a metal, then compositing samples to reduce
the number of samples and analyses would have
22
-------
been cost-effective.
The Brownfields team should request standard
operating procedures (SOPs) for sampling
activities from its contractor. (SOPs are
discussed in more detail later in this section.)
The SOPs should describe all aspects of sample
collection, including how to determine if the
sample is representative, handling and custody
requirements, and the sample volume that must
be collected to provide enough material for all
necessary analyses. The laboratory chosen to
perform analyses should be able to provide
volume requirements for selected analytical
methods.
Sometimes SOPs will call for the collection of
extra volumes of a sample. For example, when
sampling for VOCs, loss of the analytes through
breakage of a sample container or through
volatilization (loss of the container's headspace)
can render the sample or analysis useless.
Background Samples
Some action levels are derived from naturally
occurring or background concentrations. In
these cases, teams should collect background or
upgradient samples from nearby areas that are
not impacted by site contamination.
Background samples are analyzed for the same
parameters as the site samples to establish
background concentrations of target analytes
and compounds. They are collected in areas
unaffected by the site, and therefore, indicate
whether the concentration of a particular analyte
in a sample is related to site activities.
When it is necessary to compare site
contamination levels to background levels, it is
helpful to collect and analyze background
samples prior to the final determination of the
sampling design since the number of samples
necessary for a specific decision error will be
reduced if background concentrations are low.
Analytical Methods
The samples called for in the SAP will be
analyzed either onsite or in a laboratory
according to analytical methods selected during
the DQO process. Analytical methods can be
classified based on the medium (e.g., soil,
water) from which the sample was taken, the
sample preparation method, the chemicals for
which the analysis is requested (analytes), the
expected levels of the analytes (detection limits
or ranges), the level of confidence in the results,
and cost. Due to the variety of analytical
methods available, the team should work closely
with the laboratory to select appropriate
methods that meet the predefined DQOs.
The team should pay particular attention to the
action levels for the site decision when selecting
analytical methods. Different methods have
different detection limits, the lowest
concentration of a contaminant that can be
detected by a particular test method or analytical
instrument. When contaminant concentrations
in a sample decrease, that is, as they approach
the detection limit, they become increasingly
difficult to quantify, and the instrument readings
or test results become less reliable.
Measurements that fall below the detection
limits are not reliable, and are usually reported
as less than or equal to the detection limit. The
team will want to select a method with a
detection limit appropriate for the intended use
of the data. For the decision of whether to clean
up the site for public park use, the team will
want to use a method that can accurately
quantify concentrations below the action level.
Analytical methods can be varied during
optimization of the sampling design to produce
more cost-effective results. For example, if a
sampling design option calls for a multi-phase
investigation, the preliminary study may collect
data using a broad spectrum analytical method
to identify classes of compounds. Subsequent
analyses can focus on compounds for which
analyte- and class-specific methods are
available, thereby producing less expensive and
more accurate data than full spectrum methods.
The combination of a multi-phase strategy and
varied analytical methods allows for the
collection of more samples without a loss of
confidence or increase in cost. Also, an early
23
-------
broad spectrum analysis increases the
probability of identifying all contaminants of
concern. This sampling design strategy would
have been useful for the park reuse portion of
the Springfield site for which no previous
sampling data exist if the site history indicated a
high likelihood of contamination.
Forms F-l and F-2 of the QAPP template
provide space for documentation of analytical
methods to be used in the sampling design.
SOPs should be attached to the QAPP for all
analytical methods: standard, non-modified
publicly available methods and nonstandard or
modified methods that accommodate site-
specific conditions. An example of a nonroutine
method is analysis of samples from building
materials that may have become contaminated.
All field methods or mobile laboratory methods
should also have clearly written SOPs.
An alternative approach to selecting analytical
methods is included in the Performance Based
Measurement System (PBMS). This approach
is a partnership-style relationship with the
laboratory that facilitates cost-effective, method
adaptations that best serve site-specific project
needs. Instead of prescribing how to accomplish
a task, PBMS statements of work describe in
objective terms what performance standards
must be met. If the team chooses this approach,
it should work closely with the laboratory to
document site-specific performance of the
method and how this performance supports the
predefined data quality needs. More data on
PBMS can be found on the following webpage:
http://www.epa.gov/epaoswer/hazwaste/test/
pbms.htm.
If the team decides to analyze samples in the
field, they should be aware of any limitations of
the field methods under consideration and
ensure that the DQOs will be met. Two
common types of field analytical methods are
immunoassays and x-ray fluorescence (XRF).
These methods can produce rapid results at
relatively little cost but they may have limited
ability to identify certain contaminants and
reach very low detection limits. Ongoing
technological advances are rapidly expanding
the capabilities and usefulness of field analytical
technologies. Information on performance of
field analytical technologies can be obtained
from the Cleanup Information website at
http://www.du-m.com/supply 1 .htm. See
Appendix C for additional references.
Matrix interferences are common obstacles to
successful sample analysis. For example, a
clayey matrix may not release analytes of
concern during sample preparation causing the
resulting levels to be biased low. The team
should explore and document contingency plans
to guide field work in the event that site-specific
interferences hinder the reliability of a particular
method. Contingency plans, which can be
documented on form O of the QAPP template,
can save considerable time and money by
averting the downtime of expensive field teams
and limiting the costs of producing non-
informative data and of subsequent resampling.
Quality Control in the Field
Field quality control requirements and
documentation of all field sampling and
observations is critical to provide a historical
record for future reviews and analysis of the
useability of the data produced. The official
field log book will contain documentation of
field activities that involve the collection and
measurement of environmental data. Additional
forms may be used in the field to record related
activities as explained below.
SOPs delineate the step-by-step approach that
field personnel will follow in collecting
samples, taking field measurements, calibrating
instruments, etc. Sampling and field analytical
SOPs that may be used during a Brownfields
site assessment include the following:
• Sampling of surface and/or subsurface soil;
• Wipe sampling;
• Sampling of concrete and debris;
• Sampling of groundwater and surface water;
• Use of field analytical instrumentation, such
as onsite GC, GC/MS, XRF, or other field
measurement methods;
• Monitoring well installation and
24
-------
development; and
• Direct push sampling.
Most qualified sampling contractors and State
and Federally certified laboratories develop
SOPs and analytical methods as part of their
overall QA program. These SOPs and analytical
methods can be used to evaluate a potential
contractor's competence to perform field
activities or as a standard for conducting a field
audit (described later in this section). SOPs are
especially important for field activities where
significant error can be introduced into data
measurements. Data review and acceptance
depends on the documentation that SOP
protocols were followed. Any modifications to
the SOPs during field work should be
thoroughly documented. All SOPs used for a
Brownfields site assessment should be included
as appendices to the QAPP (see form F-l of
QAPP template). For further reference, see
Guidance for the Preparation of Standard
Operating Procedures for Quality-Related
Operations (see Appendix C).
SOPs are necessary for field activities including
calibration, decontamination, and preventive
maintenance and should be a part of the SAP.
The field team should document what SOP they
are using in the field and any deviations from
the SOP.
Decontamination protocols describe methods,
tools, and products used to clean reusable
sampling equipment after sample collection to
prevent contaminating the next collected
sample. These protocols are sometimes dictated
by the specific sampling SOPs. A field
preventive maintenance protocol involves
ensuring that all field equipment has been
properly calibrated, charged, and inspected prior
to and at the end of each working day and that
replacement parts are available.
Field Instrument/Equipment Inspection and
Calibration
Sampling and analysis generally requires the use
of varied equipment and tools in the gathering
of environmental data. All field equipment
needs to be inspected to determine if it is
adequate for the media, parameters to be
sampled, and the tests to be performed.
Data may be generated onsite through the use of
real-time equipment, such as a photoionization
detectors, an organic vapor analyzer, or a pH
meter. A more detailed analysis may call for
mobile lab-generated data.
The field-testing and mobile laboratory
equipment should be examined to ensure that it
is in working condition and properly calibrated.
The calibration of field instruments should be
performed according to the method and schedule
of an SOP — usually based on the
manufacturer's operating manual. Calibration
should be performed more often as field
conditions dictate.
Field Documentation
Generally, the Brownfields team records field
activities in ink, in a bound notebook with
prenumbered pages or on a preprinted form. For
each sampling event, the field team provides the
site name and location, date, sampling start and
finish times, names of field personnel, level of
protection, documentation of any deviation from
protocol, and signatures of field personnel.
For individual samples, field teams should
document the exact location and time the sample
was taken, any measurement made (with real-
time equipment), physical description of the
sample, sample number, depth, volume, type of
sample, and equipment used to collect the
sample. This information can be critical to later
evaluations of the resulting data's useability.
Individual samples should be labeled in the
field. Labels should include sample location,
sample number, date and time of collection,
sample type, sampler's name, and method used
to preserve the sample, if applicable. (Sample
preservation involves the treatment of a sample
usually through the addition of a compound that
adjusts pH to retain the sample properties,
including concentration of substances, until it
can be analyzed.) The field team should follow
25
-------
a sample summary table similar to form F-2 of
the QAPP template for each sampling event.
The table should include a listing of the total
number of samples, types of sample matrices, all
analyses planned for each sample differentiating
critical measurements, and other information
that may be relevant to later assessments of the
data's useability.
The team should track the transfer of samples
from the field to the laboratory with chain-of-
custody forms. Information on the chain-of-
custody forms should include name of
laboratory, persons relinquishing and receiving
samples, quantity of sample material,
preservation solutions, test methods requested,
unit of measurements, and signatures of
laboratory personnel. Custody procedures
should be discussed in QAPP template form K:
Sample Handling and Custody Requirements.
Field System Audits
During the initial stages of field activities, a QA
representative from the Brownfields team
should determine whether the field activities are
following the protocols delineated in the QAPP.
If, during the audit, the QA representative
identifies deviations from the prescribed
procedures, the field team manager should take
on-the-spot actions to ensure that field activities
are conducted in accordance with the QAPP.
The QA representative should document any
deficiencies encountered and the corrections
made. Results of the audit should be maintained
at the site assessment office as part of the
document control program. Document control
is discussed later in this section.
Quality Control in the Laboratory
The team should select laboratories that have
defined QA protocols. All laboratories used to
analyze samples should have an overall Quality
Assurance Plan available for review, including
SOPs and analytical methods, internal QA/QC
procedures and logs, and data review
procedures. The team may decide to conduct a
laboratory system audit to ensure that these
plans and procedures are in place and in use.
The team may also decide to audit its laboratory
by submitting performance evaluation (PE)
samples to the laboratory with the other
environmental samples collected at the
Brownfields site. A PE sample is a sample of
known composition provided for laboratory
analysis to monitor laboratory and method
performance. A PE sample can be used to rate
the laboratory's ability to produce analytical
results within the pre-set limits documented in
the QAPP. PE samples may be the simplest and
most cost-effective way to audit a laboratory.
Laboratories that participate in EPA's Contract
Laboratory Program (CLP) and State programs
typically analyze PE samples on a routine basis.
The team should request a copy of the
laboratory's PE results as part of its audit
program. The team should rely on existing audit
information, if available and relevant, to
determine the reliability of a laboratory.
Quality Control Samples
QC samples are collected and analyzed to
determine whether sample concentrations have
changed between the time of sample collection
and sample analysis, and if so, when and how.
For example, cross-contamination may occur
during sampling, and degradation may occur
during storage. A field QC sample is a sample
that is collected and produced in the field. The
laboratory QC sample is prepared by the person
conducting the first step of the sample analysis.
The number of field and laboratory QC samples
used during the project depends on the
analytical method and requirements of the
QAPP. The general rule is that 10% of samples
should be QC samples. This means that if 20
samples are collected, at least two additional
samples should be submitted as QC samples.
Exhibit 6 lists typical QC samples and the data
they provide.
Three basic types of QC samples that are
prepared in the field and laboratory include
blanks, spikes, and replicates. A blank sample
is a clean sample that has not been exposed to
the sample medium being analyzed but is
26
-------
subjected to the same procedures used in the
preparation and analysis of the sample from the
medium being analyzed. The blanks may be
exposed to the same decontamination, transport,
storage, or analytical process as the regular
samples to obtain a baseline value that may be
used later to evaluate other data.
EXHIBIT 6
TYPES OF QC SAMPLES
QC Sample
Information Provided
Blanks
field blanks
rinsate blanks
reagent blanks
method blank
Bias introduced during sampling and analysis
field handling or transport
contaminated equipment
contaminated reagent
any aspect of laboratory analytical system
Spikes
matrix spike
matrix spike duplicate
analysis matrix spike
surrogate spike
Bias introduced in laboratory
preparation and analysis
preparation and analysis precision
instrumentation
analysis
Duplicates, Splits, etc.
co-located samples
field duplicates
field splits
laboratory duplicates
laboratory splits
analysis duplicates
Precision
sampling and analysis precision
precision of all steps after sample collection
shipping and interlaboratory precision
analytical precision
interlaboratory precision
instrument precision
Source: Guidance for Quality Assurance Project Plans.
A spike sample is a sample to which a known
amount of a chemical has been added for the
purpose of determining the efficiency of
recovery of the analytes. These QC samples are
particularly helpful during analysis of complex
matrices (e.g., sediment or sludge). A duplicate
sample is a second sample taken from the same
source at the same time and analyzed under
identical conditions. A split is a duplicate that
is sent to a different laboratory for analysis.
When more than one duplicate is collected it is
called a replicate.
For the portion of the Springfield site slated for
reuse as a public park, the team planned for 40
to 50 samples. Following the 10% rule of
thumb stated above, the team will need four to
five QC samples. Considering the importance
of the accuracy of the data, the team will be
particularly interested in knowing if any bias is
February 1998, EPA QA/G-5 EPA600-R-98-018.
present in the data. Because historical data did
not identify hazardous waste activities on this
portion of the property the team will not be as
concerned about cross contamination. One field
QC sample will be collected to detect any
contamination introduced by field procedures.
Two laboratory QC samples, a matrix spike and
a matrix spike duplicate will be collected to
check bias and precision. The fourth sample, a
co-located sample, might be collected to test the
precision of field collection procedures. Many
of the QC samples are defined in the Glossary of
Terms in Appendix B of this document.
Document Control
Document control is a crucial component of QA.
Although it is critical to completion of the last
stage of a Brownfields site assessment (review
of data useability), it is sometimes overlooked.
Data useability review depends on thorough
documentation of predefined data specifications
and the events that take place during
implementation of the project which may cause
27
-------
the data to fall short of the predefined
specifications.
The team should identify those documents that
will be necessary to check for compliance with
all field and laboratory procedures. As
explained under the next heading of this
document, expertise in analytical chemistry and
statistics are necessary for the last stages of the
assessment of data useability. Therefore, during
planning for the Brownfields site assessment,
these team members should be consulted to
identify the types of documentation they will
need. Critical types of documentation include
the following:
Field Logbook
• Site sketch or map with location of each
sample collection point
• Full descriptions of all deviations from
analytical SOPs, SAP, and the QAPP
• Description of field sampling conditions and
physical parameter data as appropriate for
the media involved
QAPP
• Site description, including surrounding
structures and terrain features, nearby
populations, flow directions of relevant
media, and a description of active industrial
processes
• Description and rationale for sampling
design and procedures and references to all
SOPs
Field SOPs
• Sampling, decontamination, and calibration
procedures
Analytical SOPs
• Analytical methods used, sample tracking
and log-in procedures
Laboratory Deliverable s
• Narrative explanation of level of analytical
data review used by the laboratory and
resulting data qualifiers, indicating direction
of bias based on the assessment of QC
samples (e.g., blanks, field and laboratory
spikes)
• Results for each analyte and sample
qualified for analytical limitations
• Sample quantitation limits (SQLs) and
detection limits for undetected analytes,
with an explanation of the detection limits
reported and any qualifications
• Instrument printouts and logbooks, spectra,
and raw data
Laboratory Notebook
• Full descriptions of all deviations from
analytical SOPs, SAP, and the QAPP
Custody Records
• Chain-of-custody forms
• Laboratory custody records
Additionally, project specific documentation
may include status reports, teleconference
records, and other correspondence.
The documents and records that should be
tracked and secured should be listed on form P
of the QAPP template. The list should identify
the party responsible for producing these reports
and provide directions for where they should be
stored or sent.
Documentation permits the reviewer to trace a
sample from collection to analysis and reporting
of results. The goal of the document control
program is to account for all necessary project
documents produced during planning,
implementation, or analysis.
Grantees under EPA's Brownfields program
should adhere to program requirements for
record retention found in 40 CFR Subpart O.
Requirements include a numerical document
control system, document inventory procedure,
and a central filing system with a designated
person(s) responsible for its maintenance.
Deliverables for Data Useability
Review
Review of the useability of data culminates in
the determination of whether actual data meet
the data objectives. This determination is
28
-------
impossible without the development,
implementation, and documentation of
procedures used for sample collection,
shipment, analysis, and data reporting. This
information will allow QA reviewers to
determine, with reasonable certainty, whether
data collected during the Brownfields site
assessment meet the DQO criteria documented
in the QAPP. As in all aspects of a Brownfields
site assessment, the activities that occur during
this stage of the project should be planned and
documented in the QAPP. Forms Q-l, Q-2, and
R provide space for the team to describe what
procedures will be performed during data
useability review.
Through data useability review, the team
determines whether the project has performed
within the specifications in the planning
objectives. The decision to accept data, reject
data, or accept only a portion of the data, should
be made after consideration and analysis of all
parameters described in the QAPP. The team
should have an analytical chemist and a
statistician for portions of this assessment but
some items do not require special expertise.
The data useability review begins with an
analysis of the data for their own merit and ends
with a statistical reconciliation of the data with
site-specific data quality objectives.
The stages of data useability review generally
begin with an evaluation of the effectiveness of
the sampling operations, their conformance to
the SAP and SOPs, and whether any unusual
circumstances are documented in the field logs.
The five data quality indicators that should be
assessed during the review of field procedures
include completeness, comparability,
representativeness, precision, and bias
(accuracy). These terms are described in
Appendix B of this document.
Some activities that should be carried out
include identifying all samples (including
locations and analytes) called for in the QAPP
and comparing those to sample locations
documented in the field log and on a field map.
These samples are also compared to the sample
results submitted by the laboratory to see if the
predefined number of samples was actually
analyzed. This review should also check if the
predefined analytical methods and detection
limits were used. These reviews should give an
indication of the completeness of the data. Any
discrepancies that cannot be resolved may affect
the level of certainty in the final decision about
site reuse.
The next step is to verify the analytical data.
This activity is often guided by the analytical
method used, and several guidance documents
are available that provide a framework for data
verification based on the analytical method.
QAPP form Q-2 provides space to describe the
process to be used during data verification.
During data verification (some EPA Guidance
documents use the terms verification and
validation interchangeably), an analytical
chemist reviews the results of QC samples to
identify sources of error in the data overall, on a
sample basis and analyte by analyte. The
reviewer will look at sample holding times that
may affect the data from a single sample, and
calibration and analyte recoveries that may
affect the quantification of individual analytes.
After the data verification is complete, the team
should have a better idea of what chemicals are
present at what levels and what threats the site
poses.
The five data quality indicators listed above
should also be applied when verifying the
analytical data. The QAPP should describe the
level of verification that will be required.
The last stage of data useability review is
validation, which should be carried out by the
statistician to determine whether the data can
support their intended use. The QAPP should
explain on form R how the results will be
reconciled with the predefined data
requirements. Methods for determining possible
deviations from planning assumptions should be
described. The QAPP should also specify how
the limitations on data use will be reported.
EPA's Guidance for Data Quality Assessment
provides some information on this determination
(see Appendix C for full reference).
29
-------
Improving Data Useability
The team should plan for corrective actions to
improve data useability when performance fails
to meet objectives for data that is critical to the
Brownfields site assessment. Corrective actions
can be costly (resampling) or relatively
inexpensive (requesting additional information
from the laboratory). Much of this information
is determined by the team as part of Step 6 of
the DQO process: Specifying Limits on Decision
Errors. Corrective actions are intended to
improve data quality and reduce uncertainty,
and may eliminate the need to qualify or reject
data. Some corrective actions include the
following:
• Retrieving missing information;
• Resolving technical or procedural problems
by requesting additional explanation or
clarification from the technical team;
• Requesting reanalysis of sample(s) from the
extract stored at the laboratory;
• Requesting construction and re-
interpretation of analytical results from the
laboratory or team chemist;
• Requesting additional sample collection and
analysis for site or background
characterization; modeling potential impacts
on uncertainty using sensitivity analysis to
determine range of effect;
• Adjusting or questioning data based on
approved default options and routines; and
• Qualifying or rej ecting data for use in the
site assessment.
Summary
This document has presented the DQO process
as a systematic planning tool for cost-effective
site assessments. It has introduced elements of a
QAPP, which helps the team maintain control of
the project and achieve its objectives through
use of QA and QC measurements. It also
discussed elements of sampling strategies and
how to determine whether resulting data meet
project objectives. These tools will help
municipalities, Tribes, and States reduce the
environmental uncertainty associated with
Brownfields sites by producing environmental
measurement data of known and adequate
quality. This will, in turn, support defensible
decision-making and stakeholder satisfaction.
The QAPP template that follows in Appendix A
can be used to guide the design of a QAPP or
the forms can simply be reproduced and
completed as the QAPP.
30
-------
Appendix A - Model Quality Assurance Project Plan
INTRODUCTION
The EPA requires that all environmental monitoring and measurement efforts participate in a centrally
managed quality assurance (QA) program.
Any Brownfields team generating data under this quality assurance program has the responsibility to
implement minimum procedures to ensure that the precision, accuracy, completeness, and
representativeness of its data are known and documented. To ensure the responsibility is met uniformly,
each party should prepare a written QA Project Plan (QAPP) covering each project it is to perform.
The QAPP documents the project planning process, enhances the credibility of sampling results,
produces data of known quality, and saves resources by reducing errors and the time and money spent
correcting them. The QAPP is a formal document describing in comprehensive detail the necessary QA,
quality control (QC), and other technical activities that should be implemented to ensure that the results
of the work performed will satisfy the stated performance criteria.
All QA/QC procedures should be in accordance with applicable professional technical standards, EPA
requirements, government regulations and guidelines, and specific project goals and requirements.
The tables and figures contained in the Appendices to this document can be used to compile the
Brownfields Site QAPP. These forms can be reproduced or downloaded from EPA's Brownfields web
page located at http://www.epa.gov/swerosps/bf/, or similar forms with the same requirements can be
created. Standard Operating Procedures (SOPs) and standard analytical methods should be referenced in
the text and included as appendices to the Brownfields Site QAPP. SOPs should be referenced in the
QAPP by title, date, revision number and the originator's name.
A-l
-------
Brownfields
Quality Assurance Project Plan
TABLE OF CONTENTS
FORMS
Project Management
Form A Title and Approval Page
Form B Project Organization and Responsibility
Form C Problem Definition
Form D Project Description/Project Timeline
Measurement Data Acquisition
Form E Sampling Design
Form F-l Method and SOP Reference Table
Form F-2 Sampling and Analytical Methods Requirements
Form G Preventive Maintenance - Field Equipment
Form H Calibration and Corrective Action - Field Equipment
Form I Preventive Maintenance - Laboratory Equipment
Form J Calibration and Corrective Action - Laboratory Equipment
Form K Sample Handling and Custody Requirements
Form L Analytical Precision and Accuracy
Form M Field Quality Control Requirements/Laboratory Quality Control
Requirements
Form N Data Management and Documentation
Assessment/Oversight
Form O Assessment and Response Actions
Form P Project Reports
Data Validation and Useability
Form Q-l Verification of Sampling Procedures
Form Q-2 Data Verification and Validation
Form R Data Useability
A-3
-------
Form A
Title and Approval Page
Document Title
Prepared by: (Preparer's Name and Organizational Affiliation)
Address and Telephone Number
Day/Month/Year
Project Manager:,
Project QA Officer:
U.S. EPA Project Manager Approval:
U.S. EPA QA Officer Approval:
Signature
Printed Name/Date
Signature
Printed Name/Date
Signature
Printed Name/Date
Signature
Printed Name/Date
A-5
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
FormB
Project Organization and Responsibility
(Fill-in the blanks, if applicable, otherwise insert another project-specific chart.)
Develop an organizational chart that identifies the chain of command of key personnel, including the QA
representative. Include titles, responsibilities, and organizational affiliation of all project participants.
A-6
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Form C
Problem Definition (use multiple pages if needed)
Briefly state the specific problem that the data collection project is designed to solve or the decisions to
be made (i.e., the project objectives). Include relevant characteristics of the site, such as site use history,
suspected locations and identification of contaminants, range of contaminant concentrations, media that
may be affected, and likely migration routes. Cite previous studies that indicate why the project is
needed.
A-7
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
FormD
Project Description (use multiple pages if necessary):
Provide a detailed description of the work to be performed, e.g., identify media to be sampled, whether
field or fixed laboratories will be used, if field analytical methods will be used, likely action levels, work
schedules, required reports, etc.
A-8
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Form D (Cont.)
Project Timeline
Prepare an overall project timetable that outlines beginning and ending dates for the entire project as well as specific activities and products within the project.
Activities (list products)
Activity Start Dates (MM/DD/YY) Activity End
FormE
A-9
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Sampling Design (use multiple pages if needed)
Discuss project sampling design and provide a rationale for the choice of sampling locations for each
parameter/matrix to be sampled during this project, e.g., a judgmental sampling strategy with broad
spectrum analysis using methods from SW-846. Identify action levels. Attach a detailed site map with
anticipated sampling locations. State whether and how field analytical techniques will be used and
identify the number of field analyzed samples that will be sent for confirmation by a permanent
laboratory.
A-10
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form F - 1
Method and SOP Reference Table (use multiple pages if needed)
Use this form to create an SOP Reference Table. The appropriate number/letter reference from this table will be used to complete Forms F-2
through J, and Form L. Attach all referenced Project Analytical and Sampling SOPs to the QAPP.
Include
Analytical Method Reference:
document title, method name/number, revision
number, date
la.
2a.
3a.
4a.
Project Analytical SOPs:
Include document title, date, revision number, and
originator's name
Ib.
2b.
3b.
4b.
Ic.
2c.
3c.
4c.
Project Sampling SOPs:*
Include document title, date, revision number, and originator's name
* Project Sampling SOPs include sample collection, sample preservation, equipment decontamination, preventive maintenance, etc.
A-ll
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form F-2
Sampling and Analytical Methods Requirements (use multiple pages if needed)
Describe the details of the data collection and analysis design for the project. Insert the appropriate SOP number/letter reference in the table.
Form F-l contains the Method and SOP Reference Table. Attach analytical SOPs for sample collection and analysis for each parameter/matrix.
The following is example data.
Parameter
benzo(a)
pyrene
Matrix
soil
Number of
Samples
(include
field QC)
23
Analytical
Method*
CLP Organic
SOWOLM03.2
Sampling
SOP*
Ic
Containers
per Sample
(number, size
and type)
2x 4oz, flint
glass jar,
polyprop cap,
teflon liner
Preservation
Requirements
(temperature,
light, chemical)
stored in dark at
4C
Maximum Holding
Time at Lab
(preparation/
analysis)
extract in 7 days
Insert the appropriate SOP number/letter reference in the above table. Form F-l contains the Method and SOP Reference Table.
A-12
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
FormG
Preventive Maintenance - Field Equipment
Identify the equipment and/or systems requiring periodic preventive maintenance. Cite references on
how periodic preventive and corrective maintenance of measurement or test equipment shall be
performed to ensure availability and satisfactory performance of the systems. Cite descriptions of how to
resolve deficiencies and when re-inspection will be performed. Describe the availability of spare parts
identified in the manufacturer's operating instructions and how SOPs will be maintained.
Instrument
Activity
Frequency
SOP
Ref. *
* Insert the appropriate reference number/letter from Form F-l, Method and SOP Reference Table.
A-13
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
FormH
Calibration and Corrective Action - Field Equipment
Identify all tools, gauges, instruments, and other equipment used for data collection activities that must be calibrated to maintain performance
within specified limits. Reference calibration procedures to be conducted using certified equipment and standards with known relationships to
recognized performance standards. Reference procedures on the maintenance of records of calibration.
Instrument
Activity
Frequency
Acceptance
Criteria
Corrective
Action
SOP
Ref. *
* Insert the appropriate reference number/letter from Form F-l, Method and SOP Reference Table.
A-14
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form I
Preventive Maintenance - Laboratory Equipment
Identify the equipment and/or systems requiring periodic preventive maintenance. Cite references on
how periodic preventive and corrective maintenance of equipment shall be performed to ensure
availability and satisfactory performance. Cite discussions of how the availability of critical spare parts,
identified in the manufacturer's operation instructions and/or SOPs, will be assured and maintained. Cite
corrective actions for calibration check samples that exceed the control limits, drift in the calibration
curve, or if a reagent blank indicates contamination.
Instrument
Activity
Frequency
SOP
Ref. *
* Insert the appropriate reference number/letter from Form F-l, Method and SOP Reference Table.
A-15
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form J
Calibration and Corrective Action - Laboratory Equipment
Identify all tools, gauges, instruments and other equipment used for data collection activities that must be calibrated to maintain performance
within specified limits. Reference calibration procedures to be conducted using certified equipment and/or standards with known relationships to
recognized performance standards. Reference procedures on the maintenance of records of calibration.
Instrument
Activity
Frequency
Acceptance
Criteria
Corrective
Action
SOP
Ref. *
* Insert the appropriate reference number/letter from Form F-l, Method and SOP Reference Table.
A-16
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
FormK
Sample Handling and Custody Requirements (use multiple pages if needed)
Describe the procedures for sample handling and custody. Include chain-of-custody forms; identify the
sampling tags and custody seals the field teams should use. Refer to SOPs for collecting, transferring,
storing, analyzing, and disposing of samples.
A-17
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form L
Analytical Precision and Accuracy (use multiple pages if needed)
Identify the analytical methods and equipment required, including sub-sampling or extraction methods,
laboratory decontamination procedures and materials, waste disposal requirements (if any), and specific
performance requirements (i.e., quantitation limits, precision, and accuracy) for each method.
Analyte
Analytical
Method*
Detection
Limit
(water/soil)
(units)
Quantitation
Limit
(water/soil)
(units)
Precision
(water/soil)
Accuracy
(water/soil)
Insert the appropriate reference number/letter from Form F-l, Method and SOP Reference Table.
A-18
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
FormM
Field Quality Control Requirements
The procedures and requirements contained in EPA Requirements for Quality Assurance Project Plans, October, 1997. EPA QA/R-5 (Draft
Final), or latest revision, should be followed and referenced below.
QC Sample
Duplicate
Equipment Blank
VGA Trip Blank
Cooler
Temperature Blank
Bottle Blank
Other (specify)
Frequency *
5% per parameter per matrix
or
5% per parameter per matrix
or
1 per Cooler
or
1 per Cooler
or
1 per Lot #
or
Acceptance
Criteria
Corrective
Action
* Circle criteria listed or indicate alternative criteria
A-19
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Form M (Cont.)
Laboratory Quality Control Requirements
QC Sample
VGA
Reagent/Method
Blank
Reagent/Method
Blank
Duplicate
Matrix Spike
Performance
Evaluation (PE)
Sample
Other
Other
Frequency *
Daily
or
5% per parameter per matrix
or
5% per parameter per matrix
or
5% per parameter per matrix
or
5% per parameter per matrix
per concentration level
or
Acceptance
Criteria
Corrective
Action
* Circle criteria listed or indicate alternative criteria.
A-20
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Form N
Data Management and Documentation (use multiple pages if needed)
Briefly discuss data documentation and management from field collection and laboratory analysis to data
storage and use. Analytical data packages should include all relevant documents (for example, a
laboratory narrative, tabulated summary forms for laboratory standards, quality control, and field sample
results in order of analysis, raw data for laboratory standards, quality control, and laboratory log book
sheets). Describe procedures for detecting and correcting errors during data reporting and data entry.
Provide examples of any forms or checklists, such as chain-of-custody or field calibration forms.
Types of information to request from the laboratory:
a) Data Results Sheets (include any performance evaluation sample results)
b) Method Blank Results
c) Surrogate Recoveries and Acceptance Limits
d) Matrix Spike/Matrix Spike Duplicate Results and Acceptance Limits
e) Spike/Duplicate Results and Acceptance Limits
f) Laboratory Control Sample Results and Acceptance Limits
g) ICP Serial Dilution Results
h) ICP Interference Check Sample Results
I) Project Narrative which contains all observations and deviations
A-21
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
FormO
Assessment and Response Actions (use multiple pages if needed)
Describe procedures for identifying and correcting any problems encountered during specific project
operations.
A-22
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
FormP
Project Reports (use multiple pages if needed)
Identify the frequency, content, and distribution of project reports that detail project status, results of
internal assessments, corrective actions implemented, and project results. For example, the field team
may be required to submit daily status reports comprised of field log sheets describing any field
measurements taken, number of samples collected and their status (shipped, at lab, or awaiting shipment),
deviations from SOPs, etc.
A-23
-------
Title: Revision Number: Revision
Site Name: Revision Date: Revision
Site Location: Page: of Page:
A-24
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Form Q - 1
Verification of Sampling Procedures (use multiple pages if needed)
Describe the process to be used to review the sampling procedures to verify that they conform to
requirements in the sampling and analysis plan.
A-25
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
Form Q - 2
Data Verification and Validation (use multiple pages if needed)
Describe the process to be used to verify conformance of the analytical data with predefined
requirements. Describe the process to be used to validate conformance of the analytical data to the
predefined needs of the Brownfields site assessment.
A-26
-------
Title: Revision Number:
Site Name: Revision Date:
Site Location: Page: of
FormR
Data Useability (use multiple pages if needed)
A-27
-------
Title:
Site Name:
Site Location:
Revision Number:
Revision Date:
Page: of
Revision
Revision
Page:
Describe the process for determining whether the data successfully meet the requirements for their
intended use. Outline methods to be used to identify anomalies and departures from assumptions in the
sampling and analysis design. Discuss how limitations of the data will be reported.
A-28
-------
Accuracy
Appendix B - Glossary of Terms
A measure of the closeness of an individual measurement or the average
of a number of measurements to the true value. Accuracy is influenced
by a combination of random error (precision) and systematic error (bias)
components which are due to sampling and analytical operations. EPA
recommends that this term not be used and that precision and bias be
used to convey the information usually associated with accuracy.
AnalyteThe chemical for which a sample is analyzed.
ASTM
Background
Bias
Bioaccumulation
Blank
Brownfields Site Manager
Calibration
Calibration standard
American Society for Testing and Materials — An organization which
develops and publishes standard methods of analysis and standards for
materials and procedures.
A level of hazardous substances that approximates the level that would
be present in the medium of concern if the source of contamination
under analysis did not exist.
The systematic or persistent distortion of a measurement process which
causes errors in one direction (i.e., the expected sample measurement is
different from the sample's true value). Bias can result from improper
data collection, poorly calibrated analytical or sampling equipment, or
limitations or errors in analytical methods and techniques.
The tendency of a hazardous substance to be taken up and accumulated
in the tissue of organisms, either directly or through consumption of
food containing the hazardous substance. Bioaccumulation typically
results in increasing concentrations of hazardous substances in tissues of
organisms higher up the food chain.
A sample that has not been exposed to the analyzed sample stream in
order to monitor contamination during sampling, transport, storage, or
analysis. The blank is subjected to the same analytical or measurement
process as other samples to establish a zero baseline value and is
sometimes used to adjust or correct routine analytical results.
Person appointed by the cooperative agreement recipient or lead agency
to oversee cleanups at specific sites.
Comparison of a measurement standard, instrument, or item with a
standard or instrument of higher accuracy to detect and quantify
inaccuracies and to report or eliminate those inaccuracies by
adjustments.
Standards prepared by successive dilution of a standard solution
covering the full concentration range required and expected to be seen in
the samples, for the organic or inorganic analytical method. The
calibration standard must be prepared using the same type of acid or
solvent used to prepare samples for analysis.
B-l
-------
CERCLA
Chain-of-Custody
CLP
Comparability
Completeness
Composite sample
Control Sample
Cooperative Agreement
Data Validation
process
user needs.
Data Verification
Definitive Data
DL
Duplicate Sample
Comprehensive Environmental Response, Compensation, and Liability
Act of 1980, as amended.
An unbroken trail of accountability that ensures the physical security of
samples, data, and records.
U.S. EPA's Contract Laboratory Program. Refers to laboratory
specifications, analytical methods, and QA/QC protocols required for
Superfund and related activities.
The confidence with which one data set can be compared to another.
A measure of the amount of valid data obtained from a measurement
system compared to the amount that was expected to be obtained under
correct, normal conditions.
Non-discrete samples composed of one or more individual samples taken
at different locations at a site. Composite samples are representative of
the average concentrations of contaminants across a large area.
A QC sample introduced into a data collection process to monitor the
performance of the system.
A form of assistance provided by a Federal agency in which substantial
interaction is anticipated between the Federal agency and the assistance
recipient (e.g., State, Tribal, or local government or other) during the
performance of the contemplated activity.
Confirmation through examination and provision of objective evidence
that requirements for a specific intended use have been met. The
of examining the analytical data to determine conformance to
Confirmation through examination and provision of objective evidence
that predefined requirements for a specific intended use have been met.
The process of examining the result of a given activity to verify
conformance to stated requirements for that activity.
Data that are documented as appropriate for rigorous uses that require
both hazardous substance identification and concentration. Definitive
data are often used to quantify the types and extent of releases of
hazardous substances. Guidance for Performing Site Inspections Under
CERCLA, Interim Final, p. 99; Guidance for Data Useability in Site
Assessment, Draft, pp. 13 and 14.
Detection Limit — the lowest concentration or amount of the target
analyte that can be determined to be different from zero by a single
measurement at a stated level of probability.
A second sample taken from and representative of the same population
and carried through all steps of the sampling and/or analytical
B-2
-------
DQOs
procedures in an identical manner. See Field Duplicate, Matrix
Duplicate, and Matrix Spike Duplicate.
Data Quality Objectives — Qualitative and quantitative statements
(derived from the DQO Process) that clarify the objectives of studies,
technical processes and quality assurance programs, define the
appropriate type of data, and specify tolerable levels of potential
decision errors that will be used as the basis for establishing the quality
and quantity of data needed to support decisions.
Also called the Equipment Rinsate. A sample of analyte-free reagent
taken after completion of decontamination and prior to sampling at the
next sample location. It is used to check field decontamination
procedures to ensure that analytes from one sample location have not
contaminated a sample from the next location.
False Positive Decision Error The erroneous decision that the null hypothesis is correct.
False Negative Decision Error The erroneous decision that the null hypothesis is incorrect.
Equipment Blank
Field Blank
Field Duplicate
GC
A blank used to provide information about contaminants that may be
introduced during sample collection, storage, and transport. A clean
sample, carried to the sampling site, exposed to sampling conditions, and
returned to the laboratory and treated as an environmental sample.
An independent sample collected from the same location or source, as
close as possible to the same point in space and time. Duplicates are
stored in separate containers and analyzed separately for the purpose of
documenting the precision of the sampling process. (Laboratory
variability will also be introduced into the samples' results.)
Gas Chromatography — An analytical technique used to analyze
environmental matrices for contaminants.
GC/MS Gas Chromatography/Mass Spectrometry — This is a gas chromatography analyzer combined
with a mass spectrometer detector. The mass spectrometer uses the difference in mass-to-charge
ratio (m/e) of ionized atoms or molecules to separate them from each other and to quantify their
concentrations.
Grab Samples
Hazardous Substances
defined
Holding Time
Discrete samples that are representative of a specific area and a specific
time. Useful in identifying "hot spots" of contamination at a site.
CERCLA hazardous substances, pollutants, and contaminants, as
in CERCLA Sections 101(14) and 101(33).
The period a sample may be stored prior to its required analysis.
Although exceeding the holding time does not necessarily negate the
veracity of analytical results, it causes the qualifying or "flagging" of the
data for not meeting all of the specified acceptance criteria.
B-3
-------
Human Exposure Any exposure of humans to a release of one or more hazardous
substances via inhalation, ingestion, or dermal contact. Amdur, Mary O.,
John Doull, and Curtis D. Klaassen, Toxicology, The Basic Science of
Poisons, Fourth Edition, 1991, p. 14; Hazard Ranking System Guidance
Manual, Interim Final, pp. 153, 259, 293, 317, 363, and 411.
Interference An element, compound, or other matrix effect present in a sample which
interferes with detection of a target analyte leading to inaccurate
concentration results for the target analyte.
Matrix The substrate containing the analyte of interest — examples are soil,
water, sediments, and air. Also called medium or media.
Matrix Duplicate A duplicate field sample used to document the precision of sampling and
homogeneity of a given sample matrix. (Same as field duplicate.)
Matrix Spike (MS) A sample prepared by adding a known mass of target analyte to a
specified amount of matrix sample for which an independent estimate of
target analyte concentration is available. Spiked samples are used, for
example, to determine the effect of the matrix on a method's recovery
efficiency.
Matrix Spike Duplicate (MSD) A split sample, both portions of which are spiked with identical
concentrations of target analytes, for the purpose of determining the bias
and precision of a method in a particular sample matrix.
Maximum Contaminant Maximum concentration of a contaminant allowed in drinking water
Level (MCL) systems by the National Primary Drinking Water regulations: 40 CFR
141.11 (inorganic chemicals) and 141.12 (organic chemicals).
Method Blank A clean sample processed simultaneously with and under the same
conditions as samples containing an analyte of interest through all steps
of the analytical procedure.
Method Detection Limit (MDL) The minimum concentration of an analyte that can be measured and
reported with 99% confidence. It is determined by analysis of samples
with known concentrations at various dilutions. This limit is matrix-
specific (e.g., soils vs. waters).
Null Hypothesis Presumed or baseline condition. In the case of environmental
investigations, generally either that the site is contaminated or that the
site is clean.
ppb Parts per billion; Mg/kg (micrograms per kilogram); ,ug/l (micrograms
per liter).
ppm Parts per million; mg/kg (milligrams per kilogram); mg/1 (milligrams per
liter).
B-4
-------
Precision A measure of mutual agreement among individual measurements of the
same property, usually under prescribed similar conditions, expressed
generally in terms of the standard deviation.
Priority Pollutants List of inorganic and organic analytes commonly tested for in the
National Pollution Discharge Elimination System (NPDES) program.
QA Quality Assurance — An integrated system of management activities
involving planning, implementation, assessment, reporting, and quality
improvement to ensure that a process, item, or service is of the type and
quality needed and expected.
QAPP Quality Assurance Project Plan — A formal document describing in
comprehensive detail the necessary QA, QC, and other technical
activities that must be implemented to ensure that the results of the work
performed will satisfy the stated performance criteria.
QC Quality Control — The overall system of technical activities that
measures the attributes and performance of a process, item, or service
against defined standards to verify that they meet the stated requirements
established by the customer; operational techniques and activities that
are used to fulfill requirements for quality.
QL Quantitation Limit — The level above which quantitative results may be
obtained with a specified degree of confidence.
RCRA The Resource Conservation and Recovery Act of 1976, as amended.
Release Any spilling, leaking, pumping, pouring, emitting, emptying, discharging, injecting, escaping,
leaching, dumping or disposing into the environment (including the abandonment or discharging
of barrels, containers, and other closed receptacles containing any hazardous substance or
pollutant or contaminant). CERCLA § 101(22)
Representativeness A measure of the degree to which the measured results accurately reflect
the medium being sampled. It is a qualitative parameter that is
addressed through the design of the sampling program in terms of
sample location, number of samples, and actual material collected as a
"sample" of the whole.
SAP Sampling and Analysis Plan — Site- and event- specific plan detailing
sampling rationale, protocols, and analyses planned per sample type. A
part of the QAPP.
Screening Data Data that are appropriate for applications that only require determination
of gross contamination areas and/or for site characterization decisions
that do not require quantitative data. Screening data are often used to
specify which areas to sample to collect definitive data. Guidance for
Performing Site Inspections Under CERCLA, Interim Final, pp. 99 and
100; Guidance for Data Useability in Site Assessment, Draft, p. 15.
B-5
-------
SOP
Source Area
Spike
Split Samples
Standard Addition
Standard Curve
Surrogate
SVGA
SVOC
SW-846
Trip Blank
VGA
VOC
Standard Operating Procedure — A written document that details the
method for an operation, analysis, or action with thoroughly prescribed
techniques and steps, and that is officially approved as the method for
performing certain routine or repetitive tasks.
An area of contamination from which substances may have migrated to
other media. Several source areas can be located within a site.
A known quantity of a chemical that is added to a sample for the purpose
of determining (1) the concentration of an analyte by the method of
standard additions, or (2) analytical recovery efficiency, based on
sample matrix effects and analytical methodology. Also called
analytical spike.
Two or more representative portions taken from one sample in the field
or in the laboratory and analyzed by different analysts or laboratories.
Split samples are used to replicate the measurement of the variable(s) of
interest.
The practice of adding a known amount of an analyte to a sample
immediately prior to analysis used to evaluate interferences.
A plot of concentrations of known analyte standards versus the
instrument response to the analyte.
A pure substance with properties that mimic the analyte of interest. It is
unlikely to be found in environmental samples and is added to them to
establish that the analytical method has been performed properly.
Semi-Volatile Organic Analysis or Analyte.
Semi-Volatile Organic Compound. BNA; extractable organic
compound.
U.S. EPA "Test Methods for Evaluating Solid Waste," 1986 (Third
Edition), plus Updates, a publication describing standard methods of
analysis, sampling techniques, and QA/QC procedures.
A clean sample of matrix that is carried to the sampling site and
transported to the laboratory for analysis without having been exposed to
sampling procedures.
Volatile Organic Analysis or Analyte.
Volatile Organic Compound.
B-6
-------
Appendix C - References
Brownfields General
Tool Kit of Information Resources for Brownfields Investigation and Cleanup. EPA 542-B-97-001.
Washington, DC: U.S. Environmental Protection Agency (5102G). 1997.
Road Map to Understanding Innovative Technology Options for Brownfields Investigation and Cleanup.
EPA 542-B-97-002. Washington, DC: U.S. Environmental Protection Agency (5102G). 1997.
SW-846/RCRA General
Test Methods for Evaluating Solid Waste, Physical/Chemical Methods (SW-846), vol.1, ch. 9 "Sampling
Plan." 3rd ed. Washington, DC: U.S. Environmental Protection Agency, 1986.
Test Methods for Evaluating Solid Waste, Physical/Chemical Methods (SW-846), vol.2, ch. 10 "Sampling
Methods." 3rd ed. Washington, DC: U.S. Environmental Protection Agency, 1986.
CLP/Superfund General
A Compendium of Superfund Field Operations Methods, NTIS PB88-181557; EPA 540-P-87-001.
Washington, DC: U.S. Environmental Protection Agency, December 1987.
Data Quality Objectives for Remedial Response Activities - Development Process, EPA 540-G-87-003.
Washington, DC: U.S. Environmental Protection Agency, March 1987.
Guidance for Conducting Remedial Investigations and Feasibility Studies Under CERCLA, NTIS PB89-
184626; EPA 540-G-89-004. Washington, DC: U.S. Environmental Protection Agency, October,
1988.
Sampler's Guide to the Contract Laboratory Program, EPA 540-P-90-006. Washington, DC: U.S.
Environmental Protection Agency, December 1990.
Superfund Analytical Review and Oversight, NTIS PB90-249541; EPA 9240.0-03. Washington, DC:
U.S. Environmental Protection Agency, October 1988.
User's Guide to the Contract Laboratory Program, EPA/9240.0-1. Washington, DC: U.S.
Environmental Protection Agency, December 1988.
General Sampling and Data Guidance - U.S. EPA
Decision Error Feasibility Trials (DEFT) Software for the Data Quality Objectives Process. EPA
QA/G-4D. EPA 600-R-96-056, September 1994. .
DQO Software Tools. U. S. Department of Energy, Office of Environmental Management (EM-76),
Pacific Northwest National Laboratory, .
EMAP QA Terms, Office of Research and Development, Office of Modeling, Monitoring Systems and
Quality Assurance. October 15, 1997. .
C-l
-------
EPA Requirements for Quality Assurance Project Plans, (Draft Final) EPA QA/R-5. Washington, DC:
U.S. Environmental Protection Agency, October 1997.
Field Analytical and Site Characterization Technologies: Summary of Applications. EPA 542-R-97-011.
Washington, DC: U.S. Environmental Protection Agency (5102G). November 1997.
Field Screening Methods Catalog: User's Guide, NTIS PB89-134159; EPA 540-2-88-005. Washington,
DC: U.S. Environmental Protection Agency, September 1988.
Guidance for the Data Quality Objectives Process, EPA QA/G-4: EPA 600-R-96-055. Washington, DC:
U.S. Environmental Protection Agency, September 1994.
Guidance for Data Quality Assessment, EPA Office of Research and Development, EPA 600-R-96-084.
Washington, DC: U.S. Environmental Protection Agency, January 1998.
Guidance for Data Useability in Risk Assessment (Part A), Final. NTIS PB92 - 963356; Publication
9285.7-09A. Washington, DC: U.S. Environmental Protection Agency, April 1992.
Guidance for Data Useability in Risk Assessment (Part B), Final. NTIS PB92 - 963362; Publication
9285.7-09B. Washington, DC: U.S. Environmental Protection Agency, May 1992.
Guidance for the Preparation of Standard Operating Procedures for Quality-Related Operations, EPA
QA/G-6: Final - EPA 600-R-96-027. Washington, DC: U.S. Environmental Protection Agency,
November 1995.
Guidance on Quality Assurance Project Plans, EPA QA/G-5; EPA 600-R-98-018. Washington, DC:
U.S. Environmental Protection Agency, February 1998.
Keith, L.H., Environmental Sampling and Analysis: A Practical Guide. In Print. Washington, DC:
American Chemical Society, 1990.
Quality Assurance/Quality Control Guidance for Removal Activities, Sampling QA/QC Plan and Data
Validation Procedures, Interim Final. NTIS PB90-274481/CCE; EPA 540-G-90-004. Washington,
DC: U.S. Environmental Protection Agency, April 1990.
Sample Custody: NEIC Policies and Procedures, EPA 330-9-78-DDI-R. Washington, DC: U.S.
Environmental Protection Agency, Revised June 1985.
Standard Operating Procedures for Field Samplers, EPA Region VIII, Environmental Services Division,
Denver, CO, March 1986.
Vendor Field Analytical and Characterization Technologies System (Vendor FACTS) v. 3.0. Office
of Solid Waste and Emergency Response (5102G). CD-ROM from NCEPI or download from
Internet at CLU-IN. 1998. .
Gilbert, R.O., Statistical Methods for Environmental Pollution Monitoring, VanNostrand
ReinholdNY, 1987
C-2
-------
Groundwater Sampling and Monitoring
Compendium of ERT Groundwater Sampling Procedures, NTIS/PB91-921274; EPA 540-P-91-007.
Washington, DC: U.S. Environmental Protection Agency, January 1991.
Handbook - Ground Water. EPA 625-6-87-016. Washington, DC: U.S. Environmental Protection
Agency, March 1987.
Practical Guide for Ground-Water Sampling, NTIS PB86-137304; EPA 600-2-85-104. Washington, DC:
U.S. Environmental Protection Agency, September 1985.
Resource Conservation and Recovery Act (RCRA) Ground-Water Monitoring Technical Enforcement
Guidance Document, NTIS PB87-107751. Washington, DC: U.S. Environmental Protection Agency,
September, 1986.
Test Methods for Evaluating Solid Waste, Physical/Chemical Methods (SW-846), vol. 2, ch. 11 "Ground
Water Monitoring." 3rd ed. Washington, DC: U.S. Environmental Protection Agency.
Surface Waters
Compendium of ERT Surface Water and Sediment Sampling Procedures, EPA 540-P-91-005.
Washington, DC: U.S. Environmental Protection Agency, January 1991.
Kitrell, F.W. A Practical Guide to Water Quality Studies of Streams, NTIS PB-196367. Washington,
DC: U.S. Environmental Protection Agency, 1969.
Geophysical Methods
Compendium of ERT Soil Sampling and Surface Geophysics Procedures, EPA 540-P-91-006.
Washington, DC: U.S. Environmental Protection Agency, January 1991.
Geophysical Methods for Locating Abandoned Wells, NTIS PB84-212711. Washington, DC: U.S.
Environmental Protection Agency, July 1984.
Geophysical Techniques for Sensing Buried Wastes and Waste Migration, NTIS PB84-198449; EPA
600-7-84-064. Washington, DC: U.S. Environmental Protection Agency, June 1984.
Subsurface Characterization and Monitoring Techniques, vols.l and 2. EPA 625-R-93-003a.
Washington, DC: U.S. Environmental Protection Agency, May 1993.
Use of Airborne, Surface, and Borehole Geophysical Techniques at Contaminated Sites: A Reference
Guide, EPA 625-R-92-007. Washington, DC: U.S. Environmental Protection Agency, September
1993.
Soils and Sediments
Compendium of ERT Surface Water and Sediment Sampling Procedures, EPA 540-P-91-005.
Washington, DC: U.S. Environmental Protection Agency, January 1991.
C-3
-------
Determination of Background Concentrations of Inorganics in Soils and Sediments at Hazardous Waste
Sites (Engineering Forum Issue), Washington, DC: U.S. Environmental Protection Agency,
December 1995. (http://www.epa.gov/crdlvweb/tsc/issue.htm)
Guidance Document for Cleanup of Surface Impoundment Sites, NTIS PB87-110664; EPA 9380.0-06.
Washington, DC: U.S. Environmental Protection Agency, June 1986.
Methods for Evaluating the Attainment of Cleanup Standards, vol. 1, Soils and Solid Media, NTIS PB89-
234959. Washington, DC: U.S. Environmental Protection Agency, February 1989.
Sediment Sampling Quality Assurance User's Guide, NTIS PB85-233542; EPA 600-4-85-048.
Washington, DC: U.S. Environmental Protection Agency, July 1985.
Hazardous Waste
Compendium ofERT Waste Sampling Procedures, EPA 540-P-91-008. Washington, DC: U.S.
Environmental Protection Agency, January 1991.
Drum Handling Practices at Hazardous Waste Sites, NTIS PB86-165362; EPA 600-2-86-013.
Washington, DC: U.S. Environmental Protection Agency, January 1986.
Guidance Document for Cleanup of Surface Tank and Drum Sites, NTIS PB87-110672; EPA 9380.0-03.
Washington, DC: U.S. Environmental Protection Agency, May 1985.
Handbook for Stabilization/Solidification of Hazardous Wastes, EPA 540-2-86-001. Washington, DC:
U.S. Environmental Protection Agency, June 1986.
Remediation Technologies
Innovative Site Remediation Technology: Phase I (Process Descriptions and Limitations)
vol.1 "Bioremediation" EPA 542-B-94-006 June 1995
vol.2 "Chemical Treatment" EPA 542-B-94-004 September 1994
vol.3 "Soil Washing/Soil Flushing" EPA 542-B-93-012 November 1993
vol.4 "Solidification/Stabilization" EPA 542-B-94-001 June 1994
vol.5 "Solvent/Chemical Extraction" EPA 542-B-94-005 June 1995
vol.6 "Thermal Desorption" EPA 542-B-93-011 November 1993
vol.7 "Thermal Destruction" EPA 542-B-94-003 October 1994
vol.8 "Vacuum Vapor Extraction" EPA 542-B-94-002 April 1995
(Washington, DC: U.S. Environmental Protection Agency, November 1993 to June 1995)
Innovative Site Remediation Technology: Phase II (Design and Application)
vol.1 "Bioremediation" EPA 542-B-97-004 May 1998
vol.2 "Chemical Treatment" EPA 542-B-97-005 September 1997
vol.3 "Liquid Extraction Technologies" EPA 542-B-97-006 May 1998
vol.4 "Solidification/Stabilization" EPA 542-B-97-007 September 1997
vol.5 "Thermal Desorption" EPA 542-B-97-008 September 1997
vol.6 "Thermal Destruction" EPA 542-B-97-009 August 1998
vol.7 "Vacuum Extraction and Air Sparging" EPA 542-B-97-010 May 1998
(Washington, DC: U.S. Environmental Protection Agency, September 1997 to August 1998)
C-4
-------
SOURCES OF DOCUMENTS
U.S. Environmental Protection Agency (no charge)
Center for Environmental Research Information (CERI)
ORD Publications
26 West Martin Luther King Drive
Cincinnati, OH 45268
(513)569-7562
http://es. epa. gov/program/epaorgs/ord/ceri. html
U.S. EPA Technology Innovation Office
Clean-Up Information Internet Site
http://clu-in. com
Public Information Center (PIC) (no charge)
U. S. Environmental Protection Agency
Public Information Center (PIC)
PM-211B
401M Street, S.W.
Washington, DC 20460
(202) 382-2080
e-mail: public-access@epamail. epa. gov
Superfund Docket and Information Center (SDIC) (no charge)
U.S. Environmental Protection Agency
Superfund Docket and Information Center (SDIC)
OS-245
401M Street, S.W.
Washington, DC 20460
(202) 382-6940
http://www.epa.gov/earthl 00/records/aOOl 08. html
National Technical Information Services (NTIS) (cost varies)
National Technical Information Service
U. S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
http://www.ntis.gov/
National Center for Environmental Publications and Information (no charge)
U.S. EPA/NCEPI
P.O. Box 42419
Cincinnati, Ohio 45242-2419
(800)490-9198 Fax (513) 489-8695
http://www.epa.gov/epahome/publications. htm
C-5
------- |