EPA-540/G-90-008
                                        9285.7-09A
                                        April 1992
Guidance for Data Useability in
           Risk Assessment
                  (Part A)

                   Finai
          Notice: Guidance for Radioanalytical
          Data Useability in Risk Assessment is
          Given in Part B
             *
         Office of Emergency and Remedial Response
            U.S. Environmental Protection Agency
                Washington, DC 20460

-------
         -f  -")-
                                           NOTICE

The policies and procedures set forth here are intended as guidance to U.S. Environmental Protection Agency and other
government employees. They do not constitute ruluuiiking by the Agency, and may not be relied on to create a
substantive or procedural right cnforceuhle by :iny other person. The U.S. Environmental Protection Agency may lake
action that is at variance with the policies and procedr'es in this guid:uice and may change them at any time without
public notice.

Copies of the guidance c;ui be obtained from:

        National Technical Information Service
        5285 Port Royal Road
        Springfield. VA 22161
        Phone: 703-4X7-4650

-------
                                      Contents
CHAPTER 1   INTRODUCTION AND BACKGROUND	1
    1.1 CRITICAL DATA QUALITY ISSUES IN RISK ASSESSMENT	1
       I.I.I   DaUi Sources	2
       1.1.2   Detection Limits	2
       1.1.3   Qualified Data	2
       1.1.4   Background Samples	2
       1.1.5   Consistency in Data Collection	2
    1.2 FRAMEWORK AND ORGANI/.ATION OFTIIKCilHI>ANCE	3
CHAPTER 2   THE RISK ASSESSMENT PROCESS	7
    2.1 OVERVIEW OF BASELINE HUMAN HEALTH RISK ASSESSMENT AND Till: EVALUATION OF
       UNCF.RTAINTY	7
       2.1.1   Data Collection and Evaluation	11
       2.1.2   Exposure Assessment	13
       2.1.3   Toxicity Assessment	15
       2.1.4   Risk Cliaraclerization	17
    2.2 ROLES AND RF.SPONSIBILITIFS OF KEY RISK ASSESSMENT PERSONNEI	18
       2.2.1   Project CooMination	18
       2.2.2   Gathering Existing Site Data and IX-veloping the Conceptual Model	18
       2.2.3   Project Scoping	18
       2.2.4   Qu.:iliiy Assurance Document Preparation and Review	20
       2.2.5   Budgeting and Scheduling	21
       2.2.6   Iterative Communication	21
       2.2.7   Data Assessment	22
       2.2.8   Assessment and Presentation of Environmental Analytical Data	23
CHAPTER 3   USEABILITY CRITERIA FOR BASELINE RISK ASSESSMENTS	25
    3.1 DATA USEABILITY CRITERIA	26
       3. .1   Data Sources	26
       3. .2   Documentation	29
       3. .3   Analytical Methods and Detection Limits	30
       3. .4   Data Quality Indicators	31
       3. .5   Data Review	34
       3. .6   Reports from Sampling and Analysis to the Risk Assessor	36
    3.2 PRELIMINARY SAMPLING AND ANALYTICAL ISSUES	37
       3.2.1   Chemicals of Potential Concern	40
       3.2.2   Tentatively Identified Compounds	41
       3.2.3   Identification and Quandtation	45
       3.2.4   Detection and Quantitntion Limits and Range of Linearity	47
       3.2.5   Sampling and Analytical Variability Versus Measurement Error	50
       3.2.6   Sample Preparation and Sample Preservation	.'.	54
       3.2.7   Identification of Exposure Pathways	55
       3.2.8   Use of Judgmental or Purposive Sampling Design	55
                                             in

-------
                                   Contents
                                     (cont'd)
      3.2.9   Field Analyses Versus Fixed laboratory Analyses	57
      3.2.10  Laboratory Performance Problems	58

CHAPTER 4  STEPS FOR PLANNING FOR THE ACQUISITION OF USEABLE
   ENVIRONMENTAL DATA IN BASELINE RISK ASSESSMENTS	63
   4.1 STRATEGIES FOR DESIGNING SAMPLING PLANS	63
      4.1.1   Completing the Sampling Design Selection Worksheet	65
      4.1.2   Guidance for Completing the Sampling Design Selection Worksheet	72
      4.1.3   Specific Sampling Issues	76
      4.1.4   Soil Depth Issues	78
      4.1.5   Balancing Issues for Decision-Making	80
      4.1.6   Documenting Sampling Design Decisions	81
   4.2 STRATEGY FOR SELECTING ANALYTICAL METHODS	81
      4.2.1   Completing the Method Selection Worksheet	83
      4.2.2   Evaluating the Appropriateness of Routine Methods	84
      4.2.3   Developing Alternatives When Routine Methods are not Available	87
      4.2.4   Selecting Analytical Laboratories	87
      4 2.5   Writing the Analysis Request	88
   4.3 BALANCING LSSlfES FOR DECISION-MAKING	88
CHAPTER S   ASSESSMENT OF ENVIRONMENTAL DATA FOR USEABILITY IN
   BASELINE RISK ASSESSMENTS	95
   5.1 ASSESSMENT OF CRITERION I: REWRTS TO RISK ASSESSOR	100
      5.1.1   Preliminary  Reports	100
      5.1.2   Final Report	100
   5.2 ASSESSMENT OF CRITERION II: DOCUMENTATION	101
   5.3 ASSESSMENT OF CRITERION 111:  DATA SOURCES	101
   5.4 ASSESSMENT OF CRITERION IV: ANALYTICAL METHOD AND DETECTION LIMIT	102
   5.5 ASSESSMENT OF CRITERION V: DATA REVIEW	102
   5.6 ASSESSMENT OF CRITERION VI: DATA QUALITY INDICATORS	103
       5.6.1   Assessment of Sampling and Analytical Data Quality Indicators	105
       5.6.2  Combining the Assessment of Sampling and Analysis	114
 CHA   ER 6  APPLICATION OF DATA TO RISK ASSESSMENTS	117
   6.» . ASSESSMENT OF THE LEVEL OF CERTAINTY ASSOCIATED WITH THE
       ANALYTICAL DATA	117
       6.1.1  What Contamination is Present and at What Levels?	117
       6.1.2  Are Site Concentrations Sufficiently Different from Background?	119
       6.1.3  Are AH Exposure Pathways and Areas Identified and Examined?	120
       6.1.4  Are All Exposure Areas Fully Characterized?	120
    6.2 ASSESSMENT OF UNCERTAINTY ASSWIATED WITH THE BASELINE RISK ASSESSMENT
       FOR HUM AN HEALTH	121
                                          IV

-------
                              Contents
                                (cont'd)
APPENDICES
   I.    DESCRIPTION OF ORGANICS AND INORGANICS DATA REVIEW PACKAGES	125
   II.   LISTING OF COMMON POLLUTANTS GENERATED BY SEVEN INDUSTRIES	153
   III.   LISTING OF ANALYTES, METHODS, AND DETECTION OR QUANTITATION LIMITS FOR
        POLLUTANTS OF CONCERN TO RISK ASSESSMENT	167
   IV.   CALCULATION FORMULAS FOR STATISTICAL EVALUATION	235
   V.   "J" DATA QUALIFIER SOURCE AND MEANING	239
   VI.   "R" DATA QUALIFIER SOURCE AND MEANING	245
   VII.  SUMMARY OF COMMON LABORATORY CONTAMINANTS, CONCENTRATION REQUIRE-
        MENTS, AND RISK ASSESSMENT IMPLICATIONS	249
   VIII.  CLP ANALYTICAL METHODS SHORT SHEETS AND TCL COMPOUNDS	253
   IX.   EXAMPLE DIAGRAM FOR A CONCEPTUAL MODEL FOR RISK ASSESSMENT	263

-------
                                           Exhibits
1   Data Use-ability Criteria to Plan Sampling, Analysis and Assessment Efforts in Baseline Risk Assessment	3
2   Organization of the Guidance	5
3   Data Relevant to Components of (lie Risk Assessment Process	8
4   Baseline Risk Assessment Process and Typical Sources of Uncertainty	9
5   Range of Uncertainty of Risk Assessment	10
6   Development of Conceptual Site Model	12
7   Generic Equation for Calculating Chemical Intakes	16
8   Roles and Responsibilities of Risk Assessment Team Members	19
9   Example Risk Assessment Checklist for Use in Scoping	20
10  Checklist for Reviewing the Workphin	21
11  Checklist for Reviewing the Sampling and Analysis Plan	22
12  Importance of Data Useability Criteria in Planning for Baseline Risk Assessment	26
13  Data Sources and Their Use in Risk Assessment	28
14  Relative Importance of Documentation in Planning and Assessment	30
15  Relevance of Sampling Data Quality Indicators	31
16  Relevance of Analytical Data Quality Indicators	32
17  Alternative Levels of Review of Analytical Data	34
18  Automated Systems to Support Data Review	35
 19  Data and Documentation  Needed for Risk Assessment	36
20  Importance of Sampling Issues in Risk Assessment	38
21  Sampling Variability and Measurement Error	39
22  Importance of Analytical Issues in Risk Assessment	40
23  Median Coefficient of Variation for Chemicals of Potential Concern	42
 24  Munitions Compounds and Their Detection Limits	43
 25  Summary of Most Frequently Occurring Chemicals of Potential Concern by Industry	44
 26 Steps in the Assessment of Tentatively Identified Compounds	45
 27 Requirements for Confident Identification and Qu;uilitation	45
 28 Relative Impacts of Detection Limit and Concentration of Concern: Data Planning	46
 29 The Relatioaship of Instrument Calibration Curve and Analyte Detection	48
 30 Example of Detection Limit Calculation	49
 31  Measurement of Variation and Bias Using Reid Quality Control Samples	51
 32 Sampling Issues Affecting Confidence in Analytical Results	52
 33  Sources of Uncertainty that Frequently Affect Confidence in Analytical Results	53
 34  Sample Preparation Issues	!	54
 35  Information Available from Different Sampling Techniques	54
 36  Comparison of Sample Preparation Options	56
 37  Identification of Exposure Pathways Prior to Sampling Design is Critical to Risk Assessment	57
 38  Strengths and Weaknesses of Biased and Unbiased Sampling Designs	58
 39  Characteristics of Field and Fixed Laboratory Analyses	59
 40  Strengths and Weaknesses of Field and Fixed Laboratory Analyses	60
 41  Examples of Spatially and Temporally Dependent Variables	64
 42  Examples of Sampling Designs	65
                                                  vii

-------
                                          Exhibits
                                            (cont'd)

43  Applicability of Sampling Designs	66
44  Common Sampling Designs	67
45  Hierarchical Structure of Sampling Design Selection Worksheet	68
46  Factors in Determining Total Number of Samples Collected	72
47  Relationships Between Measures of Statistical Performance and Number of Samples Required	73
48  Number of Samples Required to Achieve Given Levels of Confidence, Power and MDRD	76
49  Confidence Levels for the Assessment of Measurement Variability	77
50  Soil Depth Sampling Worksheet	79
51  Automated Systems to Support Environmental Sampling	81
52  Method Selection Worksheet	82
53  Automated Systems to Support Method Selection	84
54  Common Laboratory Contaminants and Interferences by Organic Analyte	85
55  Common Laboratory Contaminants and Interferences by Inorganic Analyte	86
56  Comparison of Analytical Options for Organic Aiuilytes in Water	90
57  Comparison of Analytical Options for Organic Analytes in Soil	91
58  Comparison of Analytical Options for Inorganic Aiuilytes in Water and Soil	92
59  Comparison of Analytical Options for Organic and Inorganic Analytes in Air	93
60  Data Useability Assessment of Criteria	95
61  Minimum Requirements, Impact if Not MCI. and Corrective Actions for Data Useability Criteria	96
62  Corrective Action Options When Data Do Not Meet Performance Objectives	97
63  Data Useability Worksheet	98
64  Relative Importance of Detection Limit and Concentration of Concern: Data Assessment	103
65  Consequences of Alternative Sampling Strategies on Total Error Estimate	104
66  Use of Quality Control Data for Risk Assessment	105
67  Steps to Assess Sampling Performance	110
68  Recommended Minimum Statistical Performance Parameters for Risk Assessment	Ill
69  Basic Model for Estimating Total Variability Across Sampling and Analysis Components	114
70  Combining Data Quality Indicators From Sampling and Analysis into a Single Assessment of Uncertainty... 115
71  Data Useability Criteria  Affecting Contamination Presence	118
72  Data Useability Criteria  Affecting Background Level Comparison	119
73  Data Useability Criteria  Affecting Exposure Pathway and Exposure Area Examination	120
74  Data Useabilily Criteria  Affecting Exposure Area Characterization	121
75  Uncertainty in Data Collection and Evaluation Decisions Affects the Certainty of the Risk Assessment	122
                                                 VIII

-------
                                       Tips*
    The analytical data objective for baseline risk assessments is that uncertainty is known and
    acceptable, not that uncertainty be reduced to a particular level,  (p. 3)
    To maximize data useability for the risk assessment, the risk assessor must be involved from
    the start of the Rl.  (p. 7)
    All data can be used in the baseline risk assessment as long as their uncertainties are clearly
    described, (p. 11)
    Uncertainty in the analytical data, compounded by uncertainty caused by the selection of the
    transport models, can yield results that are meaningless or that cannot be interpreted, (p. 14)
    Uncertainties in toxicological measures and exposure assessment are often assumed to be
    greater than uncertainties in environmental analytical data; thus, they are assumed to have a
    more significant effect on the uncertainty of the risk assessment, (p. 17)
    Analytical data collected solely for other purposes may not be of optimal use to the risk
    assessment, (p. 20)
    Effective planning improves the useability of environmental analytical data in the final risk
    assessment.
    (p. 25)
    Use historical analytical data and a broad spectrum  analysis to initially identify the chemicals
    of potential concern or exposure areas, (p. 26)
    Jo expedite the risk assessment, preliminary data should be provided to the risk assessor as
    soon as they are available,  (p. 35)
    To protect human health, place a higher priority on preventing false negatives in sampling
    and analysis than on preventing false positives, (p.  41)
    Use preliminary data to identify chemicals of potential concern and to determine any need to
    modify the sampling or analytical design, (p. 41)
    Specific analysis for compounds identified during library search can be requested, (p. 41)
    The closer the concentration of concern is to the detection limit, the greater the possibility of
    false negatives and false positives, (p. 47)
    The wide range of chemical concentrations in  the environment may require multiple analyses
    or dilutions to obtain useable data. Request results from all analyses, {p. 47)
    Define the type of detection or quantitation limit for reporting purposes; request the sample
    quantitation limit for risk assessment,  (p. 47)
    When contaminant levels in a  medium vary widely, increase the number of samples or
    stratify the medium to reduce  variability, (p. 50)
    Sampling variability typically contributes much more to total error than analytical variability.
    (p. 50)
    Field methods can produce legally defensible  data if appropriate method QC is available and
    if documentation is adequate,  (p. 57)
    To minimize the potential for false negatives, obtain data from a broad spectrum analysis
    from each medium and exposure pathway, (p. 58)
    The CLP or other fixed laboratory sources are most appropriate for broad spectrum analysis
    or for confirmatory analysis, (p. 58)
    Solicit the advice of the chemist to ensure proper laboratory selection and to minimize
    laboratory and/or methods performance problems that occur in sample analysis, (p. 58)
    Use of the Sampling Design Selection Worksheet will help the RPM or statistician determine
    an appropriate sampling design,  (p. 65)
* For further information, refer to the text. Page numbers are provided.

                                            ix

-------
                                    Tips
                                   (cont'd)
While other designs may be appropriate in many cases, stratified random or systematic
sampling designs are always acceptable, (p. 65)
If the natural variability of the chemicals of potential concern is large (e.g., greater than 30%),
the major planning effort should be to collect more environmental samples, (p. 72)
At least one broad spectrum analytical sample is required for risk assessment, and a
minimum of two or three are recommended for each medium in an exposure pathway, (p.
73)
Collect and analyze background samples prior to the final determination of the sampling
design since the number of samples is significantly reduced if little background
contamination is present, (p. 75}
Systematic sampling supp'smented by judgmental sampling is the best strategy for
identifying hot spots, (p. 75)
Focus planning efforts on maximizing the collection of useable data from critical samples, (p.
78)
The ability to combine data from different sampling episodes or different sampling
procedures is a very important consideration in selecting a sampling design but should be
done with caution, (p. 78)
Ensure that critical requirements and priorities  are specified on the Method Selection
Worksheet so that the most appropriate methods can be considered,  (p. 83)
Use routine methods wherever possible since method development is time-consuming and
may result in problems with laboratory implementation,  (p. 83)
Analyte-specific methods that provide better quantitation can be considered for use once
chemicals of potential concern have been identified by broad spectrum analysis, (p. 84)
All results should be reported for samples analyzed at more than one dilution, (p. 85)
Field analysis can be used to decrease cost and turnaround time providing data from a broad
spectrum analysis  are available, (p. 89)
Focus corrective action on maximizing the useability of data from critical samples,  (p.  97)
Use preliminary data as a basis for identifying sampling or analysis deficiencies and taking
corrective action, (p. 100)
Problems in data useability due to sampling can affect all chemicals involved in the risk
assessment; problems due to analysis may only affect specific chemicals, (p. 100)
Qualified data can usually be used for quantitative risk assessments,  (p.  105)
Anticipate the need to combine data from different sampling events and/or different
analytical methods,  (p.  107)
Determine the distribution of the data before applying statistical measures, (p. 109)
Determine the statistical measures of performance most applicable to site conditions before
assessing data useability.  (p. 110)
 Use data qualified asUorJ for risk assessment purposes,  (p. 113)
 The major concern with false negatives is that  the decision based on  the risk assessment may
not be protective of human health, (p. 117)
False negatives can occur if sampling is not representative, if detection limits are above
concentrations of concern, or if spike recoveries are very low. (p. 117)
False positives can occur when blanks are contaminated or spike recoveries are very high. (p.
 118)
 Statistical analysis may determine if site concentrations are significantly above background
concentrations when the differences are not obvious,  (p. 120)
 The primary planning objective is that uncertainty levels are acceptable, known and
 quantitatable, not  that uncertainty be eliminated,  (p. 121)
                                                                                          -V.V

-------
                                           PREFACE
The U.S. Environmental Protection Agency (EPA) has
established a Data Usability Workgroup to develop
national guidance Tor determining data  useability
requirements needed for environmental data col led ion
on  hazardous waste sites under the Comprehensive
Environmental Response, Compensation, and Liability
Act of 1980 (CERCLA) as amended by the Supcrfund
Amendments and Reauthorization Act of 1986(S ARA).
Data useability is the process of assuring or determining
that the quality of data generated meets the intended use.
This guidance has been designed by the Risk Assessment
Subgroup of the Data Useability Workgroup to provide
data users with a nationally consistent basis for making
decisions about the minimum quality and quantity of
environmental analytical  data thai  are sufficient  to
support Superlund risk assessment decisions, regardless
of which  parties conduct the investigation.   This
document  is the first part (Part A) of the lwo-p:ul
Guidance for Data Useability in RixkAsxexxment. Part
B of this guidance addresses radioanalytical issues.

Risk Assessment  Guidance for Super/and (RAGS),
Volume I:  Human Health Evaluation Manual. Part A
(EPA 1989a) serves as a general guidance document for
the risk assessment process. Building upon RAGS,  an
"interim final" version of Guidance for Data Useahility
in Risk Assessment was issued by the Risk Assessment
Subgroup of the Data Useability Workgroup in October
1990.  The guidance was issued as "interim final" in
order to obtain and incorporate comments and criticisms
from data users who tested it in real-world situations.

The authors acknowledge the significant helpof all who
nave provided comments and criticisms.  The results
indicate that many people react favorably to the guid;uice
and find it useful in planning a risk assessment or in
evaluating assessments already underway.  Issues were
identified where guidance in the interim final needed to
be supplemented or discussed in more detail.  These
issues include providing a more detailed discussion of
sampling strategies, incorporating ground water factors,
addressing .soil  depth for exposure, and obtaining
background data.  Issues concerning data reporting
formats, validation and use of non-CLP data, and
tentatively identified compounds were also identified.
The final version of the guidance provides greater detail
in the discussion of these and other issues.

This  guidance provides  direction for planning and
assessing analytical data collection activities for the
baseline human health risk assessment, conducted as
part of the remedial investigation (RI) process.
Although the guidance addresses the baseline risk
assessment within the RI, it i* appropriate fur use in
the new Superfund Accelerated  Cleanup Model
(SACM) where data need* for risk a.s.se.sxrm:nt are
considered at the  onset of site evaluation.   Site-
specific conditions may often require sampling  or
analysis beyond (he basic recommendations given in
this guidance. The guidance docs t it directly address
the use of ecological data  for purposes other than
baseline risk assessments for human health, although
some considerations have been included when data may
be used for both ecological and human health evaluation.

Tliis guidance complements guidance provideu in RAGS
(EPA 1989a), Guidance for Conducting Remedial
Investigations and Feasibility Studies Under CERCLA
(EPA 1988a), and Data Quality ObjectivesforRemedial
Response Activities: Development Process (EPA 1987a).
RAGS provides the framework for making data quality
assessments in  baseline risk assessments, and this
guidance  supplements and strengthens  important
technical deuiils of the framework by providing direction
on minimum requirements for environmental analytical
data used  in baseline risk assessments.  As such, it
complements and builds upon Agency guidance for the
development and use  of data quality objectives in all
data collection activities.

Tin's guidance is addressed primarily to the remedial
project managers (RPMs) who have the principal
responsibility for leading the data collection and
assessment activities that support the human health risk
assessment and, secondarily, to risk assessors who must
effectively communicate their data needs to the RPMs
i'nd use the data provided to them.  Chemists, quality
assurance specialists, statisticians, hydrogeologists and
other technical experts involved in the RI process can
use tliis guidance to optimize the  useability of data
collected in the RI for use in baseline risk assessments.

Comments on the guidance should be sent to:

    Toxics Integration Branch
    Office of Emergency and Remedial Response
    401 M Street. SW (OS-230)
    Washington, DC  20460
    Phone: 202-260-9486
                                                   XI

-------
                                         ACKNOWLEDGEMENTS

               This guidance was developed by nn EPA workgroup with membership from EPA Headquarters, EPA Regional offices
               and representatives of the contractor community.  The EPA Risk Assessment Subgroup of the Data Useability
               Workgroup provided valuable input regarding the content, approach and organization of the guidance. Members of the
               Risk Assessment Subgroup, responsible for generating this guidance, have experience in human health risk assessment,
               remedial project management, chemistry, toxicology, hydrogeology, and quality assurance.  Technical review was
               provided by  lexicologists, chemists, quality assurance specialists, engineers, project managers, and statisticians from
               both EPA and contractor staff.

               Leadership for development of the "interim final" version of this guidance was provided by Data Useability Workgroup
               Region III Co-chairpersons Chuck Sands [currently at the Analytical Operations Branch (AOB)J and Claudia WJters,
               and Ruth Bleyler of the Toxics Integration Branch (TIB).

               Leadership for development of the "final" version of this guidance was provided by Ruth Bleyler and Lisa Matthews of
               TIB and Chuck Sands of AOB.  We wish to acknowledge Region V and Region VI for their assistance with the
               implementation effort for the final version of the guidance.

               Members of the Risk Assessment Subgroup include:
                              Ruth Bleyler
                              Richard Drunker
                              Rex Bryan
                              Matt Charsky
                              Skip Ellis
                              Gwen Hootcn
                              Dawn loven
                              Peter Isaacson
                              Cindy Kaleri
                              Jim LaVelle
                              Jim Luey
                              Jon Rauscher
                              Chuck Sands
                              Robin Smith
                              Pat Van Leeuwen
                              Chris Weis
                              Leigh Woodruff
Toxics Integration Branch
USEPA Region HI
Viar & Comp;uiy
Office of Waste Programs Enforcement
CH2M HILL
USEPA Region VIII
USEPA Region III
Viar & Company
USEPA Region VI
USEPA Region VIII
USEPA Region VIII
USEPA Region VI
Analytical Operations Branch
CH2M HILL
USEPA Region V
USEPA Region VIII
USEPA Region X
               Additional Workgroup participation includes:
                              Wayne Bennan
                              Ann Marie Burke
                              Dorothy Campbell
                              Judy Hsieh
                              Mark Moese
                              Sheila Sullivan
                              Haas Waetjen
ICF
USEPA Region 1
USEPA Region VIII
USEPA Region I
Ebasco
USEPA Region V
Office of Waste Programs Enforcement
\
\
 I:
                                                             xlii

-------
                                          Chapter 1
                          Introduction and Background
lliis guidniKX* was developed by the U.S. nnvinmmciibtl
I'rotection Agency (1:PA) for remedial project managers
(RPMs). risk assessors. :uid contractors. It is published
in two pans; this document is Pan A.  Part B solely
addresses useability issues in radioanalylical sampling
and analysis for risk  assessment.  Doth pans ot tin's
guidance are designed to assist RPMs in maximi/ing
the u stability of environmental analytical data collected
in the remedial investigation (RI) process for baseline
human health risk assessments.  Since RPMs, with
assistance from technical experts, oversee the preparation
of workplans and sampling and analysis plans for RI
data collection,  it is important Tor them to understand
the types, quality and quantity of data needed by risk
assessors, and the impact that Uu. r data collection
decisions have on the  Ic.-cl of certainty of baseline risk
assessments for human health.  This guidance provides
detailed approaches and basic recommendations for
both obtaining and interpreting data for risk assessment
that specifically address:

   •  1 low todesign RI sampling and analytical activities
     that meet (he data quantity and data quality needs
     of risk assessors.

   •  Procedures for assessing the quality of  the data
     obuiined in the RI,

   •  Options for combining environmental analytical
     data of varying levels of quality from different
     sources and  incorporating them into  the risk
     assessment.

   •  Procedures for determining the level of certainty
     in the risk  assessment based on the uncertainty in
     the environmental analytical data, and

   •  Guidelines on (he timing and execution of the
     various activities in  order to most  efficiently
     produce dclivcrablcs.

Although the guidance addresses the baseline risk
assessment within the RI. it is appropriate for use in the
new Supcrfund Accelerated Cleanup Model (SACM)
where data needs for risk assessment arc considered at
the onset of site evaluation.

Risk assessors  should be an integral pan of the RI
planning process to ensure that adequate environmental
analytical data of acceptable quality and quantity for the
risk assessment are  collected during the  RI.  This
guidance assists risk assessors in communicating their
environment;!! analytical data needs to the RPMs. Risk
assessors should work closely with the RPMs to identify
and recommend sampling designs and analytical
nn.-i.HHls that will maximi/e the quality ol the baseline
risk assessment for human health within the site-related
and budgetary constraints of (lie RI, and will produce
consistent risk assessments useful to risk managers.

This guidance provides a number of worksheets and
exhibits that can be used as bases for the organi/ation of
sampling or analytical phuming or assessment processes.
However, implementation of guidance will be siie-
spcciftc, and site personnel should develop and modify
these guid:uice materials to best suit the conditions at
their site.

Although ecological data useability is not addressed
.specifically in this guidance, the chemical data obtained
from site cluiracteri/ation are useable for certain elements
of the ecological  assessment.  In an  ecological
assessment, the chemicalsof potential concern and their
priorities may be different than those of (lie human
health risk assessment.  For example, iron is rarely of
concern in human health risk assessments,  but high
levels of iron may pose a threat to aquatic species, lico-
guidance documents relevant toriskassessment include
Risk Assessment Guidance for Su/ierfund. Volume II:
Environmental Evaluation Manual (HPA l'JX%). ECU
Update (HPA I'J'Jla) and Ecological Assessment of
Hazardous Waste Sites:  A Field and Laboratory
Reference (I-PA l«J89c).


1.1  CRITICAL DATA QUALITY ISSUES
     IN  RISK ASSESSMENT

Five basic environmental data  quality issues  arc
frequently encountered in risk assessments.  This
guidance provides procedures, minimum requirements,
.ind other information to resolve or minimize  the effect
of these issues on the assessment of uncertainty in the
risk assessment. The issues affect both (he planning for
and the assessment of analytic;!! data for use  in RI risk
assessments.  The following sections describe these
issues and their impact on data useability. and highlight
the resolutions of these issues.
                  Acronyms
 CLP     Contract Laboratory Program
 El'A     U.S. F.nviruntneniul Protection Agency
 QAPjl'   quality assurance project plan
 RAGS    Risk Assessment Guidance for Supcrfund
 RI       remedial investigation
 RI'M     remedial project manager
 SACM   Superfutxl Accelerated Cleanup Model

-------
1.1.1  Data Sources

Data  users  must  select sampling and  analytical
pnvedures and providers appropriate to the data needs
of each risk assessment.  Practical tradeoffs among
detection limits,  response time, documentation,
analytical costs, and level of uncertainty should be
considered prior to selecting sampling designs, analytical
methods, and service providers.

The Contract Laboratory Program (CLP) has been the
principal source of analytical data for investigations at
hazardous waste sites. The CLP requires adherence to
specific data acceptance criteria which results in data of
known analytical quality produced in a standardized
package. Another principal source of analytical data is
the EPA Regional laboratory, which often produces
data similar  in quality to that  of the CLP.  Other
analytical sources, such as Held analysis  or  fixed
labor-atone i (EPA, stale, or private), can also produce
dataof acceptable quality. Accordingly, RPMs and risk
assessors should seek the source of data that best meets
the data quality needs of the risk assessment. Section
4.2 provides guidance for selecting analytical sources.

Field analytical data have been used primarily to aid in
making decisions during sampling.  However, recent
ad vances in technology, when accompanied by sufficient
and appropriate quality control measures, allow field
analytical data to be used in risk assessments with more
frequency and more confidence than in the  past. Dy
using field analyses, RPMs can increase the number of
samples to better characterize the site and significantly
decrease sample turnaround time (to provide real-time
 decision-making in the field) as long as acceptable data
 quality is maintained.   Guidance for assessing the
 useabil ity and applicability of field analytical data in the
 risk assessment process is also provided in Section 4.2.

 For any source of monitoring data. RPMs must ensure
 that data quality objectives, analytical methods, quality
 control requirementsandcriteria,levclof documentation,
 and degree and assignment of responsibilities for quality
 assurance oversight are clearly documented in the quality
 assurance project plan (QAPjP). In addition, the RPM
 is responsible for the enforcement of these parameters.
 For non-Supcrfund-lead  analyses, the potentially
 responsible party, state, or federal agency determines
 and documents these parameters. The QAPjP  is then
 submitted to the RPM for review. In all cases involving
 risk assessment, the RPM should always seek the source
 of data that best meets the data quality needs of the risk
 assessor. The data source chosen must generate data of
 known quality.
1.1.2  Detection Limits

Selecting the analytical method to meet the required
detection limits is fundamental to the useability of
analytical data in risk assessments. In addition, the type
of detection limit, such as method detection limit or
sample quanlitation limit, used in making data quality
decisions affects the certainty of the risk assessment.
Guidance for making these decisions is provided in
Section 4.2. Preliminary remediation goals, as defined
in Risk Assessment Guidance for Superfund (RAGS)
Volume I: Human Health Evaluation Manual. Pan B
(EPA 1991b). provide criteria to be considered in
evaluating the adequacy of detection limits.

1.1.3  Qualified Data

Laboratories, and individuals conducting independent
data review, affix coded qualifiers to data when quality
control requirements or other evaluation criteria are not
met.  Data reviewers assess these and many  other
criteria to determine the useability of data.  Qualified
data must be used appropriately in risk assessments.
Data arc almost always useablc in the risk assessment
process, as long as the uncertainty in the data and its
impact on the risk assessment are thoroughly explained.
Section 5.6 describes procedures  for incorporating
qualified data and data of varying analytical quality into
the risk assessment.

 1.1.4  Background Samples

 In conducting a risk assessment, it is critical todislinguish
 site contamination  from  background levels due to
 anthropogenic or naturally occurring contamination in
 order to determine the presence or absence of
 contamination and to compare with background risk.
 Analytical data reported near method detection limits
 and sample results qualified during data review
 complicate the use of background sample  data to
 determine site contamination. Planning for the collection
 of a sufficient number of background samples from
 representative locations increases the certainty in
 decisions about the significance of site contamination.
 Section 4.1 discusses how statistical analysis and
 professional judgment can be combined to design a
 sampling program for collecting adequate background
 data.

  1.1.5  Consistency in  Data Collection

  Data  collection  activities may  vary among  parties
  conducting RIs. Consistency in all Superfund activities
  is increasingly crucial.  All  parties  collecting

-------
environmental analytical data Tor  baseline risk
assessments for human health should use guidance
provided in Risk Assessment Guidance for Suptrfand
(RAGS) Volume I: Human Health Evaluation Manual.
Part A (EPA 1989a) and this guidance to ensure thai
baseline risk assessments for human health are conducted
consistently and are protective of the public health.


1.2 FRAMEWORK AND ORGANIZA-
    TION OF THE GUIDANCE

This guidance is organized following the usual sequence
used to determine the useability  of environmental
analytical data for baseline  human  health risk
assessments.  Exhibit 1 illustrates the conceptual
framework for the guidance.  Six criteria are used to
evaluate data useability for baseline risk  assessments
for human health:

   • Data sources,

   • Documentation.

   • Available analytical services in lermsof analytical
    methods and detection limits,
  • Data quality indicators,

  • Daui review, and

  • Reports to risk assessor.

These criteria address the five major data quality issues
described in Section  1.1 and other issues thai impact
data useability in the risk assessment. The data useability
criteria are applied in Rl planning to guide the design of
sampling plans juid select analytical methods for the
data collection effort. The criteria arc employed again
to assess the useability of the analytical data collected
during the RI, and of data from other studies and
sources,  such as site inspections. This guidance also
describes how to determine the uncertainties in the risk
assessment based on the level of uncertainty of the
environmental analytical data, determined using the
data use-ability criteria.

    «•• The analytical data objective for baseline
    risk assessments is that the uncertainty is
    known and  acceptable,  not  that  the
    uncertainty be reduced to a particular level.
            EXHIBIT 1. DATA USEABILITY CRITERIA TO PLAN SAMPLING,
                         ANALYSIS AND  ASSESSMENT EFFORTS
                              IN BASELINE RISK ASSESSMENT

DEFINING
DATA USEABIUTY
CRITERIA (3.1)
• Data Sources
• Documentation
• Analytical Method*
and Detection Limits
• Data Quality
Indicator*
• Data Review
• Report* to Risk
Assessor



PLANNING
SAMPLING
CONSIDERATIONS
• Preliminary Sampling
Issues (3.2)
• Strategies for
Designing
Sam pi ing Plans (4.1)

ANALYTICAL
CONSIDERATIONS
• Preliminary Analytical
Issues (3.2)
• Strategy (or Selecting
Analytical Methods
(4.2)

-
ASSESSING
DATA USEABILITY
CRITERIA (5.0)
• Reports to Risk
Assessor
• Documentation
• Data Sources
• Analytical Methods
and Detection Limits
• Data Review
• Data Quality
Indicators

*-
DETERMINING
LEVELS
OF
CERTAINTY
FOR
BASELINE
RISK
ASSESSMENT
(6.1)



-------
Exhibit 2 summarizes the purpose of each chapter of
this guidance and highlights how the chapters can best
assist RPMs and risk assessors. Worksheets, assessment
tables, and other aids are used extensively throughout
the guidance. These are tools that can be used "as is,"
or they can be modified for use or used as the basis for
site-specific worksheets or summaries. Chapter contents
are summarized below.

   •  Chapter 2—The Risk Assessment Process: This
     chapter explains the purpose and objectives of a
     baseline human health  risk assessment  and
     describes the four basic elements of a risk
     assessment:  data collection and  evaluation,
     exposure assessment, toxicity assessment,  and
     risk characterization.  The chapter discusses the
     uncertainties associated with the risk assessment
     process and emphasizes the impact of analytical
     data quality on each element.  The roles  and
     responsibilities of the RPM, the risk assessor, and
     others involved in planning and conducting data
     collection activities to support the risk assessment
     are described.

   • Chapter 3—Useability Criteria for Baseline Risk
     Assessments:   Six criteria are defined in  this
     chapter for interpreting the importance of sample
     collection, analytical techniques, and data review
     procedures to the useability of analytical data in
     risk assessments.  The sampling and analytical
     issues that need to be addressed in using these
     criteria are discussed.  The chapter stresses the
     need to consider and plan for risk assessment data
     requirements in the early design stages of the RI.

   • Chapter 4—Steps for Planning for the Acquisition
     of Useable Environmental Data in Baseline Risk
     Assessments:  This chapter provides explicit
     guidance for designing  sampling plans  and
     selecting analytical methods based on the  data
     quality requirements of baseline risk assessments.
     Worksheets for sampling design  selection, soil
     depth sampling, and method selection are provided
     at pan of the step-by-step guidance for making
     data collection decisions for individual sites.

   • Chapter 5—Assessment of Environmental Data
     for Useability in Baseline Risk Assessments: This
chapter explains how to assess the useability of
site-specific data for risk assessments after data
collection according to the six criteria defined in
Chapter 3.  For each assessment criterion, the
chapter defines minimum data requirements and
explains  how  to determine actual performance
comp:\red to performance objectives and execute
appropriate corrective actions for data critical to
the risk assessment. The chapter also describes
options available to risk assessors for incorporating
analytical data from different sources and varying
levels of quality into the baseline risk assessment.

Chapter 6—Application of Data  to  Risk
Assessments: This chapter details procedures for
determining  the  overall level of uncertainty
associated with the risk assessment. The discussion
addresses  characterization of contaminant
concentrations within exposure areas, detemining
the presence or absence of chemicals of potential
concern,  and distinguishing site contamination
from background levels.

Appendices—The appendices provide analytical
and  sampling technical reference materials,
including descriptions  of  generic organic and
inorganic  data review packages;  listings of
common industrial pollutants; analytical methods
and detection or quanlitation limits (see Section
3.2.4  for  definitions);  common  laboratory
contaminants; calculation formulas for statistical
evaluation;  information on  analytical  data
qualifiers;  a summary  of Contract  Laboratory
Program methods with corresponding Target
Compound List compounds and Target Analyte
List anaytes; and an example of a conceptual site
model.

Index—The  index provides  cross-references
throughout the guidance. This is important because
Chapters 3,  4, and 5 present planning and
assessment issues as complementary discussions
that can be viewed independently.

Tips—Tips, marked with a «•, are incorporated
into the  text of the chapters.  These tips  draw
attention to key  issues in  the text  but are not
intended to summarize the discussion in the chapter.

-------
Chapter 1
Introduction and Background

• Presents critical data useability issues.
• Specif*» audience to be primarily RPMs and risk assessors.
• Defines scope and specifies organization of the guidance.
    Chapter 2
    The Risk Assessment Process
    •  Explains the elements o; a risk assessment and the impact of analytical data quality on each
      element.
    •  Defines the uncertainty J in the risk assessment process.
    •  Describes tho roles :>( the risk assessor. RPM and others involved with the risk assessment
      planning and assessment process.
       Chapter 3
       Ueeability Criteria for Baseline Risk Assessments
       •  Defines six criteria for assessing data useability: data sources, documentation, analytical
          methods/detection Kmits, data quality indicators, data review, and reports to the risk assessor.
       •  Applies criteria to sampling and analytical issues.
            Chapter 4
            Step* for Planning for the Acquisition of Useable Environmental Data In Baseline Risk
            As sees mente

            * Provides guidelines for designing sampling plans and selecting analytical methods.
            • Provides worksheets to support sampling design selection, soil depth sampling.
             and analytical method selection.
              Chapter 5
              Assessment of Environmental Data for Useability in Baseline Risk Assortments

              •  Describes minimum requirements for useable data.
              •  Explains how to determine actual performance compared to objectives.
              •  Recommends corrective actions for critical data not meeting objectives.
              •  Describes options for combining data from different sources and of varying quality into the risk
                 assessment.
                   Chapter 6
                   Application of Date to Risk Assessment*

                   •  Provides procedures to determine the uncertainty of the analytical data.
                   •  Explains how to distinguish site from background levels of contamination and determine the
                      presence (absence) of chemicals of potential concern.
                   •  Discusses how to characterize contaminant concentrations within exposure areas.
                       Appendices

                       •  Provide technical reference materials for sampling  and analysis.
                       •  Describe data review packages and meanings of selected date qualifere.
                                                                                                      21-002-002

-------
                                         Chapter  2
                         The Risk Assessment  Process
This chapter is an overview of the data collection and
evaluation issues that affect the quality and useability of
baseline human health risk assessments. Ecological
risk assessment is not discussed in this guidance. The
discussion focuses on how the quality of environmental
analytical data influences the  level of certainty of the
risk assessment and stresses the importance of
understanding data limitations in characterizing risks to
human health.

The chapter has two sections. Section 2.1 is an overview
of baseline human health  risk assessment  and the
significance of uncertainty in each stage of the risk
assessment process. Section 2.2 summarizes the roles
and responsibilities of key participants in  the risk
assessment process.


2.1 OVERVIEW OF BASELINE
     HUMAN HEALTH RISK
     ASSESSMENT AND THE
     EVALUATION OF UNCERTAINTY

The approach to the baseline human health risk
assessment process used for exposure to chemicals of
potential concern  is well established.  The National
Research Council (NRC) prepared a comprehensive
overview of this process (NRC 1983). which has become
the foundation for subsequent  EPA guidance (EPA
1986a. EPA 1989a, EPA 1989b). RAGS, Pan A (EPA
1989a), discusses in detail the human health baseline
risk assessment process which is used in the Superftnd
program.

The risk assessment process has four components:

   • Data collection and evaluation,

   • Exposure assessment,

   • Toxicity assessment, and

   • Risk characterization.

Exhibit 3 lists information sought in each component of
the baseline risk assessment.

Uncertainty analysis is often viewed as the last step in
the risk characterization process. However, as discussed
in detail in RAGS. Pan A, uncertainty analysis is a
fundamental element of  each component of risk
assessment, and the results for each component require
an explicit statement of the degree of uncertain ty. These
 results are the bases for estimating the degree of
uncertainty in the risk assessment as a whole.  This
chapter reviews the issues that determine the level of
uncertainty in each component of risk assessment.

    ••  To maximize data useability for the risk
    assessment,  the risk assessor must be
    involved from the start of the Rl.

The import1.rice of obtaining analytical data that fulfill
the needs of risk assessment cannot be overstated. The
risk assessor must be involved from the start of the risk
assessment process to help establish the scope of the
investigation and the design of the sampling and analysis
program.

All analytical data collected for baseline risk assessment
must be evaluated for their useability. The procedures
for evaluating the adequacy of the data are documented,
along with  the resulting estimates of  the  levels of
certainty. Limitations in the analytical data are not the
only source of uncertainty in risk assessment. Exhibit
4 identifies some typical sources of uncertainty, inherent
in each component of the risk assessment, which restrict
the depth and breadth of the evaluation. This guidance
deals only with the uncertainty inherent in data collection
and evaluation.  Consult RAGS. Part A, for a more
complete discussion of these and other uncertainties.
                 Acronyms

  ATSDR  Agency for Toxic Substances and Disease
           Registry
  DQO    data quality objective
  EPA     U.S. Environmental Protection Agency
  CIS      Geographical Information System
  I1EAST  Health Effects Assessment Summary Tables
  IRIS     Integrated Risk Information System
  LOAEL  lowest-observable-adverse-effect level
  NOAEL  no-observable-adverse-effcct level
  NRC    National Research Council
  PAH    polycyclic aromatic hydrocarbon
  PCD     polycblorinated biphenyl
  QA      quality assurance
  QAPjP   quality assurance project plan
  QC      quality control
  RAGS   Risk Assessment Guidance for Supeifund
  RfC     reference concentration
  RfD     reference dose
  RI      remedial investigation
  RME    reasonable maximum exposure
  RFM    remedial project manager
  SAP    sampling and analysis plan
  SOP    standard operating procedure
  UCL    upper confidence limit

-------
    EXHIBIT 3.  DATA RELEVANT TO COMPONENTS OF
              THE RISK ASSESSMENT PROCESS
 Risk Assessment
    Component
                     Data
 Data Collection and
     Evaluation
•  Background monitoring data for all affected media.

•  Environmental data for all relevant media.

•  List of chemicals of potential concern.

•  Distribution of sampling data.

•  Confidence limits surrounding estimates of
   representative values.
Exposure Assessment
•  Release rates.

•  Physical, chemical and biological parameters, for
   evaluating transport and transformation of she-
   related chemicals.

•  Parameters to characterize receptors according to their]
   activity, behavior and sensitivity.

•  Estimates of exposure concentrations for all
   chemicals, environmental media and receptors
   at risk.

•  Estimates of chemical intake or dose for all
   exposure pathways and exposure areas.
 Toxicity Assessment
   Toxicity values for all chemicals, exposure
   pathways, and exposure areas of concern.

   Uncertainty factors and confidence measures for
   RfDs; weight-of-evidence classifications for cancer
   slope factors.
 Risk Characterization
 • Hazard quotients and indices.

 • Estimates of excess lifetime cancer risk.

 • Uncertainty analysis.
                                                                        21-002-003

-------
                              EXHIBIT 4.  BASELINE RISK ASSESSMENT PROCESS AND
                                             TYPICAL SOURCES OF UNCERTAINTY
                                    f
                            Exposure Assessment

                     •  Assumptions regarding intake
                        factors, population characteristics,
                        and exposure patterns may not
                        adequately characterize exposure
                        and may result in underestimates or
                        overestimates of risk.

                     •  The degree to which release or
                        transport models are represen-
                        tative of physical reality may
                        overestimate or underestimate risk.

                     •  Inappropriate selection of detection
                        limit can result in overestimate or
                        underestimate of risk.

                     •  Assumption of 100% bioavail-
                        ability of chemicals in environ-
                        mental media (sol in particular) may
                        result in overestimates of risk.

                     •  Assumption that chemicals of
                        potential concern do not degrade or
                        transform in the environment may
                        result in underestimates or
                        overestimates of risk.

                     •  Incremental risks associated with
                        exposure to site-related chemicals
                        of potential concern cannot be fully
                        characterized and may result in
                        underestimates of risk.

                     •  Methods used to estimate inhalation
                        exposure to volatfles.  suspended
                        particulates or dust may
                        overestimate intake and risk.

                     •  Very few percutaneous absorption
                        factors are available for chemicals
                        of potential concern. Exposure
                        from dermal contact may be over-
                        estimated using conservative
                        default values.
i
I
                    Source:  Adapted from EPA 1989a.
   Data Collection and
       Evaluation

Use of inappropriate method
detection limits may result in
underestimates of risk.

Results may overestimate or
underestimate risk when an
insufficient number of
samples are taken.

Contaminant loss during
sampling may result in
underestimates of risk.

Extraneous contamination
introduced during sampling
or analysis may result in
overestimation of risk.
   Risk Characterization

Risk/dose estimates are
assumed to be additive in the
absence of information on
synergism and antagonism.
This may result in over-
estimates or underestimates
of risk.

Toxicity values are not
available for all chemicals of
potential concern.  Risks
cannot be quantitatively
characterized (or these
compounds and may result in
underestimates of risk.

For some chemicals or
classes (e.g., PCBs. PAHs).
in the absence of toxicity
values, the cancer slope
factor or RID of a highly toxic
class member is commonly
adopted.  This approach may
overestimate risks.
         1
   Toxicity Assessment

 Critical toxidty values are
 derived from animal studies
 using high dose levels.
 Exposures in humans occur
 at low dose levels.
 Assumption of linearity at
 low dose may result in
 overestimates or under-
 estimates of risk.

 Inappropriate selection of
 detection limit can result in
 overestimates or under-
 estimates of risk.

 Extrapolation of results of
 toxicity studies from
 animals to humans may
 introduce error and
 uncertainty, inadequate
 consideration of
 differences in absorption,
 pharmacokinetics, and
 target organ systems, and
 variability in population
 sensitivity.

> There is considerable
 uncertainty in estimates of
 toxicity values. Critical
 toxicity values are subject
 to change as new evidence
 becomes available. This
 may result in overestimates
 or underestimates of risk.

> Use of conservative high to
 low dose extrapolation
 models may result in
 overestimation of risk.
                                                                                                                      21-002-OM

-------
Risk assessment can be a simple operation, using only
screening-level data, or can be comprehensive, requiring
a robust data set designed to support statistical analyses.
Exhibit 5 discusses the range of uncertainty of baseline
risk assessment The first column in Exhibit 5 defines
the range of the analysis from a low to a high degree of
uncertainty. The second column describes the associated
data useability and limitations in the risk analysis.

   •  The first level of analysis  in Exhibit  5 is a
     quantitative risk assessment based on a sampling
     program that can be statistically analyzed. The
     assessment explicitly bounds and quantitates the
     uncertainty  in all estimates.  This analysis may
     strive to attain an ideal based upon the complexity
     of the site. The assessment is "quantitative" in that
     numeric estimates are derived for potentially
     adverse non-carcinogenic and carcinogenic effects,
     and in that the level of certainty is quantified.

   •  The second level of analysis in Exhibit 5 is a
     quantitative assessment based on a limited number
     of samples or on  data that cannot be fully
                 quantitated. The risk characterization may include
                 numeric estimates of excess lifetime cancer risks
                 and the calculation of hazard indices. However,
                 the level  of analytical  uncertainty for these
                 measures may be significant but is either not
                 quantitated or is estimated. Given the limitations
                 of the analytical data, only a qualitative evaluation
                 of the analytical uncertainty  is feasible.  Most
                 baseline risk assessments fall within this category.
                 Bias may need to be determined for its effect on
                 predicted exposures and conscijuent risk.

                 The third level of the continuum is a qualitative
                 assessment of risk. The assessment is qualitative
                 because no numeric measures can be derived to
                 indicate the potential for adverse effects, and the
                 level of certainty cannot be assessed. The risk to
                 human health is considered only in general terms.
                 Qualitative assessments are based upon limited
                 sources of historical information, such as disposal
                 records, circumstantial evidence of contamination,
                 or preliminary site assessment data.
         EXHIBIT 5.  RANGE OF UNCERTAINTY OF RISK ASSESSMENT
            Range of Analyses
             Description/Limitations
      Quantitative Assessment of Risk:

      Uncertainty minimized, quantified.
      and explicitly stated.  Resulting or
      final uncertainty may be highly
      variable (either high or low).
Risk assessment conducted using well-designed,
robust data sets and models directly applicable to site
conditions. Sampling program, based on geostatistical
or random design, will support statistical analysis of
results. Statistical analysis used to characterize
monitoring data. Confidence limits or probability
distributions may be developed for all key Input
variables.
       Quantitative Assessment of Risk:

       Magnitude of uncertainty
       unknown. No explicit quantitative
       estimates provided. Qualitative,
       tabular summary of factors
       influencing risk estimates may be
       provided for determination of
       possible bias in error.
Risk assessment conducted using data set of limited
quality and size.  No meaningful statistical analysis can
be conducted.  Results of risk assessment may be
quantified but uncertainty surrounding these measures
cannot be quantified. Only a qualitative statement is
possible. The majority of baseline risk assessments
typically fall within this category.
       Qualitative Assessment of Risk:

       Only qualitative statement of
       uncertainty is possible.
       Uncertainty is high.
 Risks cannot be quantified due to insufficient monitoring
 or modeling data. Qualitative statement of risks based
 on historical information or circumstantial evidence of
 contaminantion is provided.  This evaluation must be
 considered a preliminary, screening level assessment.
                                                                                               21-0(8-006
                                                   10

-------
    *• All data can be used in the baseline risk
    assessment as long as their uncertainties
    are clearly described.

Risk assessments must sometimes be conducted using
data of limited quantity and of differing quality. When
RPMs and other technical experts involved in the RI
understand the quantity and quality of data required in
risk assessments, they are better able to design data
collection programs to meet these requirements.

2.1.1  Data Collection and Evaluation

Overview of methods for data collection and
evaluation. Data collection begins with a statement of
the risk assessment purpose and a conceptual model of
thecurrent understanding of the problems tobe addressed
tor (lie site under investigation. The model draws from
all available historical data (EPA  1989a).  It is first
created with a  best  estimate of the  types and
concentrations of chemicals, or of key chemicals that
are likely to be present, given the history of the site. Site
records, site maps, the layout of existing structures,
topography, and readily observable soil, water and air
characteristics on and off the site help  to estimate
chemicals of potential concern, likely imporuint exposure
pathways, potentially exposed populations, and likely
temporal  and spatial variation. All of these elements
comprise the conceptual model (Exhibit 6 and Appendix
IX).  Once the conceptual model has been developed
and information has been disseminated to project staff,
(he site Ls scoped to identify data gaps and requirements
for the baseline risk assessment.
Several key issues that are part of the development  of
data quality objectives (DQOs) should be addressed at
scoping (Neptune, el. al. 1990):

    •  The types of data needed  (e.g..  environmental,
      toxia>logical),

    •  How the data will be used (e.g.,  site character-
      ization, extent of plume, etc.. what chemicals of
      concern will drive the risk-based decision), and

    •  The desired level of certainty for the conclusions
      derived from the analytical data (e.g.. what are the
      probabilities of false positive and false negative
      results as a function of risk and concentration).

 Carefully designed sampling  and analysis programs
 minimize  the subsequent  need to qualify the
 environmental data during the data assessment phase.
 The objective of the data collection effort Ls to produce
 data that can be used to assess risks to human heal th with
 a known  degree of certainty.
A complete list of chemicals of potential concern is
produced when the analytical data have been collected
and evaluated. This list of analytcs is the focus of the
risk assessment. EPA no longer advocates the selection
of "indicator compounds." because this practice may
not accurately relied the total risk from exposure to
multiple site chemicals of potential concern, nor does it
improve the quality or accuracy of the risk assessment
(El'A 1989:0.

Uncertainty in djita collection and evaluation.  Four
principal decisions must be made during data col lection
and cv:tluation in the risk assessment:
   •  The presence and levels of contaminants at the site
     at a predefined level of detail.

   •  It the levels of site-related chemicals differ
     significantly  from their background levels,

   •  Whether the analytical data are adequate to identify
     :ind  ex:unine exposure pathways and exposure
     :irc:is. and

   •  Whether the :uialylical data are adequate to fully
     characterize exposure areas.

These decisions are examined in detail in subsequent
chapters. The discussion in this section introduces basic
concepts.

Determining  what contamination Is present and at
what level. Once a site is suspected to be contaminated
and chemicals of potential concern have been identified,
the levels of chemical contamination in the affected
environmental media must be quantitated to derive
exposure and intake estimates. Estimates of the site
contamination  must  be  produced,  with  explicit
descriptions of the degree of certainty associated with
 the concentration values.
 Variability in observed concentration levels arises from
 a combination of variance in sampling characteristics of
 the  site,  in sampling  techniques, and in laboratory
 analysis. The key issue in optimizing the useability of
 data for risk assessment is to understand, quantify, and
 minimize these  Variabilities.

 HPA's objective  is to protect  human  health and the
 environment. Therefore, the design of RI programs is
 intended to minimize two potential errors:

   •  Not detecting site contamination that Ls actually
      present (i.e., false negative values), and

    •  Deriving site concentrations that do not accurately
      characterize the magnitude of contamination.
                                                     II

-------
                        EXHIBIT 6.  DEVELOPMENT OF CONCEPTUAL SITE MODEL
    Identify Chemicals of Potential Concern

   . Historical data on former useage of site.

   . Results from earlier analyses.

   • Potential background chemicals.

   . Mobility, toxicity and degradation
     characteristics.
                        Identify Site Characteristics
                                              Identify Population Characteristics
                          Detailed site map, locating areas of
                          storage, use and disposal of chemicals
                          of potential concern.
                                                On-site and nearby off-site
                                                population.
                                                                     • Land use (current and future)
                                                                       (e.g., residential, industrial,
                                                                       recreational).
   Geological, hydrogeological and soil
   characteristics information.
                          Surface and subsurface topography.

                          Meteorological data.
                                                Receptors at risk.
   . Sources of release.
              Identify Exposure
             Pathways (e.g.. Soil
                 Ingestion)
                                    Identify Exposure
                                   Pathways (e.g.. Air
                                       Inhalation)
                                                    Identity Exposure
                                                 Pathways (e.g.. Dermal
                                                       Contact)
Identify Exposure
     Areas
Identify Exposure
     Areas
Identify Exposure
     Areas
Identify Exposure
     Areas
Identify Exposure
     Areas
Identify Exposure
     Areas
                                                    Develop Conceptual Site Model
                                                                                                                            21-002-006

-------
DeternilninglfsiteconcentratioasdinVrsignificantly
from background concentratioas.  A fundamental
decision in baseline risk assessments is whether the site
poses an increased risk to human health  and the
environment. The decision depends on the degree of
certainty that the background concentrations are
significantly different  from the concentrations of the
chemicals of potential concern at the site.  Generally,
this question can be confidently answered only if the
design of the  sampling  program accommodates the
collection of both site  and background samples and if
the selection of analytical methods is appropriate.
The differences  between site  and  background
concentrations is evaluated by  comparing observed
levels of chemicals of potential concern at the site with
measured background concentrations  of the same
chemicals  in the same  environmental media.
Statistically, this is a test of the null hypothesis, that the
mean concentration of achemical at the study area is not
significantly different  from the mean concentration of
the chemical at the background location. (Historical on-
site  levels or nearby  off-site levels may  be u :d u
supplement background data. An example of an off-site
area is  the 4-mile radius  used  for the air exposure
pathway in the Hazard Ranking System.) If data from
background samples are clearly different from the results
of site monitoring (e.g.. mean chemical concentrations
differ coasistendybytwoordcrsofrnagnitude),sla(istical
analysis of the data may not be necessary.  Under such
circumstances, RAGS indicates that the primary issue is
establishing a reliable representation of the extent of the
contaminated area. Determining extent of conuunination
is not discussed in this guidance and involves tiiflcrcnt
decisions, DQOs. and sampling designs. If tin; results
of site monitoring arc less than two orders of magnitude
above background, the procedures used for  sampling
and analysis for risk assessment should follow the
recommendations of Chapter 4.

The null hypothesis is always evaluated and accepted or
rejected with a specified level of certainty. This level of
certainty is defined by (he significance, or confidence,
level. A type I error is the probability that the null
hypothesis is rejected when in fact it is  true (which
contributes  to false positive conclusions). A type  II
error Is the probability that the null hypothesis is accepted
 when it is false (a false negative conclusion).  How
 sampling and analysis design affects the likclituxx) of
 these two types of errors is described in Chanter 4.

 Evaluating whether analytical data are adequate to
 identify and examine exposure pathways and their
 exposure areai. Identifying and delineating exposure
 pathways and their exposure areas are important  in
 identifying  potentially  exposed populations and for
developing  intake estimates. In the baseline risk
assessment, the  risk assessor  combines data on
conuunination with information on  human activity
patterns to identify exposure pathways and to determine
the exposure area.  The ability to accomplish this
depends on the adequacy of analytical data.
Sampling should be designed to provide representative
data for exposure areas at a site, to address hot spots, to
evaluate  the transport of site-related chemicals  of
potcnlhtl concern, and to facilitate the identification of
all exposure pathways. A well-designed sampling and
analysis program results in data of known quality and
quantification of  spatial and temporal variability; it
specifies how to interpret the magnitude of observed
values (such as by comparison with background levels
or some  other benchmark).   Analytical data should
characterize the extent of contamination  at the site in
three dimensions.

Evaluating whether analytical data are adequate to
fully  characterize exposure areas.  Heterogeneity
should be considered in  the environmental medium
under evaluation. Hot spots  need to be identified and
char, icterized. Neptune, et. al. 1990, have proposed the
concept of an "exposure unit" as the area over which
receptors integrate exposure. This concept establishes
a basis for summarizing the results of monitoring and
transport modeling. The sampling and analysis program
must be designed to enable the risk assessor to refine the
initial ch:irac(criza(ion of exposure pathways and to
spatially and temporally identify the critical areas of
exposure.
 2.1.2  Exposure Assessment

 Overview of methods for exposure assessment. The
 objectives of the exposure assessment are:
   •  To identify or define the source of exposure,

   •  To define exposure pathways along with each of
      their components  (e.g.,  source, mechanism of
      release, mechanism  of  transport, medium of
      transport, etc.),

   •  To  identify potentially exposed  populations
      (receptors), and

   •  To measure or estimate the magnitude, duration.
      :ind frequency of exposure to site contaminants for
      each receptor (or receptor group).

 Actions at h:i/ardous waste sites arc based on an estimate
 of the reasonable maximum exposure (RM1I) expected
 to occur under both current and future conditions of land
 use (EPA 1989a). EPA defines the RME as (he highest
 exposure that is reasonably expected to occur at a site
                                                    13

-------
over time. RMRs are estimated for individual pathways
and combined across exposure pathways if appropriate.
Once potentially exposed populations are identified,
environmental concentrations at points  or exposure
must be determined or projected. Intake estimates (in
mg/kg-day) arc then developed Tor each chemical of
potential concern using a conservative estimate of the
average concentration to which receptors arc exposed
over the exposure period. (RAGS recommends a 95%
upper confidence limit (UCL) on the arithmetic mean.)
The concentration estimate is then combined with other
exposure parameters (e.g., frequency, duration, and
body weight) to calculate intake.

In the risk assessment report, estimates of intake are
accompanied by a full description (including sources)
of the assumptions made in their development.  This
information may be used subsequently in sensitivity
and uncertainty analyses in the risk characterization.

Uncertainty analysis in exposure assessment.
Exposure assessments can introduce  a great  deal of
uncertainty into the  baseline risk assessment process.
Small measures of  uncertainty in each of the  input
purame.ers which comprise an exposure scenario may
result in substantial uncertainty in the final assessment.
The largest measure of uncertainty is associated with
characterizing transport and transformation of chemicals
 in the environment, establishing exposure settings, and
deriving estimates of chronic intake.  The ultimate
 effect of uncertainty in the exposure assessment is an
 uncertain estimate of intake.
 'Hie following sections discuss the significance of the
 unccruiinty in the analytical data set on selected aspects
 of exposure assessment. Foramorccompletediscussion
 of Uie exposure assessment process, the reader is re ferred
 to RAGS, Part A.

 Characterizing  environmental fate, identifying
 exposure pathways, and  identifying  receptors at
 risk. An evaluation of the transport and transformation
 of chemicals in the environment is conducted for several
 reasoas:
    •   To understand the behavior of  site-related
       chemicals of potential concern,

    •  To project the ultimate  disposition  of  these
       chemicals,

    •  To identify exposure pathways and receptors
       potentially at risk, and

    •  To characterize environmental concentrations at
       the point of exposure.

  These evaluations  cannot be accomplished with any
  degree of certainty if the analytical data are inadequate.
Monitoring data ;ux* most appropriately used to estimate
current or existing exposure when direct contact with
contaminated environmental  media is the primary
concern.  Modeling may be required, however, in order
toevaluate the potential for future exposure, or exposure
at a distance from the source of release, or to predict
present concentrations where measurement is«xicos(ly.
In each case, success in estimating potential exposures
depends heavily on the adequacy of the analytical data.

Environmental fate and transport assessment often uses
models  to estimate  concentrations in  environmental
media at points distant from the source of release.
Models,  of necessity, are simplifications  of a real,
physical system.  Consequently, it  is critical that the
limitations of the model (the way that the model differs
from reality)  be understood and  considered when
applying the model  to a particular site. The degree to
which the model differs from reality (in critical areas of
thcanalysis)contribu(eslotheunceruuiityoftJ)eaiuily.sis.
Transport models are commonly selected for their utility
in describing or interpreting a set of monitoring data.
Chemical transport  models must be carefully selected
for their ability toincaningfullycharactcri/e the behavior
of chemicals in the environmental medium for the
specific site under investigation.   Models  that are
inappropriate for the geophysical conditions at the site
will result in errors in  the exposure assessment.  For
example, the model may be  designed  to predict
contaminant movement through sand, while soils at the
site are primarily made up of clay. Additionally, if the
analytical data set is severely limited in size or does not
accurately characterize the nature of contamination at
 the situ, a transport model cannot be properly selected or
 accurately calibrated.   This introduces  additional
 uncertainly.

     «•  Uncertainty  in the analytical data,
     compounded  by uncertainty caused by the
     selection of the transport models, can yield
     results that are meaningless or that cannot
     be interpreted.

  Estimating chemical  intake.  Uncertainties in all
  elements of the exposure assessment come together.
  and are compounded, in the estimate of intake. It is here
  that the professional judgment of the risk assessor is
  particularly important. The risk assessor must examine
  and interpret a diversity of information:
    • The nature, extent and magnitude ofcontamination,

    • Results of environmental transport modeling,

    • Identification of exposure pathways and areas.
                                                      14
                                                      c

-------
  •  Identification of receptor groups currently exposed
     and potentially exposed in the future, and

  •  Activity patterns and sensitivities of receptors and
     receptor groups.

D ased on this information, the risk assessor characterizes
the exposure setting and quantifies all pai ameters needed
in  the  equations to estimate  intake (EPA 1989a).
Chemical intake is a function of the concentration of the
chemical at the point  of contact, the  amount of
contaminated medium contacted per unit lime or event,
the exposure frequency and duration, body weight, the
ability of the chemical to penetrate the exchange
boundary, and the average time period during which
exposure occurs. Exhibit 7 is the generic form of the
intake equation used in exposure assessment.

The specific form of the intake equation varies depending
upon the exposure pathway under consideration (e.g.,
ingestion, inhalation,  dermal contact) (EPA 1989a).
Each of the variables in  these equations,  including
chemical concentration,  is  commonly characterized as
a point estimate.  However, each intake variable in the
equation has a range of  possible values.  Site-specific
characteristics determine  the  selection  of  the nost
appropriate values. In an effort to increase consistency
among Supcrfund risk assessments, EPA has established
standardized exposure parameters to be used when site-
specific data are unavailable (EPA 1991b).  Note that
the combination of all factors selected should result in
an estimate of reasonable maximum exposure for each
chemical in each pathway (EPA 1989a).

For most risk assessments, it may not be possible, nor
necessarily advantageous, to  develop a quantitative
uncertainty analysis: In  these cases, a summary of
major assumptions and their anticipated effects on final
exposure estimates should be included (o  provide a
qualitative characterization of the level of certainty in
the intake estimates.

2.1.3  Toxicity Assessment

 Overview of methods  for toxicity assessment.  The
objectives of toxicity assessment are to evaluate the
 inherent toxicity of the compounds at the site, and to
 identify and select toxicity values to  evaluate the
 significance of receptor exposure to these compounds.
Toxicity assessments rely on scientific data available in
 the  literature  on adverse effects on  humans and
 nonhuman species.
 Several values of loxicity are important in human health
 risk assessments. Reference doses (RfDs) and reference
 concentrations (RfCs) are used for oral and inhalation
 exposure, respectively, to evaluate non-carcinogenic
and developmental effects; cancer slope factors and unit
risk estimates are  used for the oral and inhalation
pathways for carcinogens.

RfDs and Rl'Cs are values developed by EPA to evaluate
the potential for non-carcinogenic effects in humans.
The RfD is defined as  an estimate (with uncertainty
spanning an order  of magnitude or more) of a daily
exposure  level for human populations, including
sensitive sub-populations, that is likely lobe without an
appreciable risk of adverse health effects over  the
period of exposure (EPA 1989a). Subchronicor chronic
RfDs may be derived for a chemical for intermediate or
long-term exposure scenarios. These values are typically
derived from the no-obscrvable-adverse-effect level
(NOAEL) or the lowest-observable-advcrse-ef feet level
(LOAEL) and the application of uncertainty and
modifying factors (EPA 1989a).  Uncertainty factors
arc used to account for the variation in sensitivity of
human sub-populations and the uncertainty inherent in
extrapolating the results of animal studies to humans.
Modifying factors account for additional uncertainties
in the studies used  lo derive the NOAEL or LOAEL.

Cancer slope factors and unit risk values are defined as
plausible, upper-bound estimates of the probability of
cancer response in an exposed  individual, per unit
intake over a lifetime exposure period (EPA 1989a).
EPA commonly develops slope factors for carcinogens
with wcight-of-evidence classifications that reflect the
likelihood mat the toxicant is a human carcinogen (EPA
 1989a).

To reduce variability in toxicological values used for
risk assessment, a standardized hierarchy of available
toxicological data  is  specified for Supcrfund.  The
primary source of information for these data  is the
Integrated Risk Information System (IRIS) database
(EPA 1989d).  IRIS consists  of verified RfDs, RfCs,
cancer slope factors, unit risks, and other health risk ami
EPA regulatory information. Data in IRIS are regularly
 reviewed and updated by an EPA workgroup. If toxicity
 values arc not available in IRIS, the EPA Health Effects
Assessment Summary Tables (HEAST) (EPA 1990a)
 are used as a secondary current source of information.
 Additional sources of toxicity information are provided
 in RAGS.

 The loudly assessment is conducted parallel with the
 exposure assessment, but may begin as early as the data
 collection and evaluation phase.   As chemicals of
 potential concern ore identified at the site, the toxfcologbit
 begins to identify the  appropriate  toxicity values.  A
 well-designed sampling and analysis program facilitates
 timely identification of the chemicals that will be the
 focus of the risk assessment.
                                                     15

-------
             EXHIBIT?. GENERIC EQUATION FOR
              CALCULATING CHEMICAL INTAKES
      Where:
                         tCx fCRxEFDN   JL
                         tCX V   BW  )* AT
                       intake; the amount of chemical at the exchange
                       boundary (mg/kg body weight-day)
         Chemical-related variable

                 C »   chemical concentration; the average
                       concentration contacted over the exposure
                       period (e.g., trig/liter water)

         Variables that describe the exposed population

                 OR «  contact rate; the amount of contaminated
                       medium contacted per unit time or event (e.g.,
                       liters/day)

                 EFO « exposure frequency and duration; describes how
                       long and how often exposure occurs. Often
                       calculated using two terms (EF and ED):

                          EF «• exposure frequency (days/year)

                          ED B exposure duration (years)

                 BW • body weight; the average body weight over the
                       exposure period (kg)

          Assessment-determined variable

                 AT »  averaging time; period over which exposure is
                       averaged (days)
Source: RAGS (EPA 1989«).
                                                                    11-002-007
                                   16

-------
Uncertainty analysis and toxicity assessment.  The
toxicity assessment is another contributor to uncertainty
in risk assessment.  Limitations in the analytic;d data
from environmental samples affect the results of the
toxicity assessment, but not to the extent that they affect
other components of the risk assessment process. Data
on physical and chemical parameters that may influence
bioavailability can influence route-to-route and vehicle-
related adjustments to toxicity values. The selection of
appropriate toxicity values is influenced by monitoring
data from environmental samples to the extent that this
information assists in identifying chemicals of potential
concern, exposure pathways, and the time periods over
which exposure may occur. Based on this information,
the lexicologist identifies sub-chronic or chronic RfDs,
RfCs, and cancer slope factors for oral, dermal, and
inhalation exposure pathways.
A list of toxicity values for risk assessment should
include an indication of the degree of certainty associated
with these values. Weight-of-evidence classifications
provide a qualitative estimate of certainty and should be
included in the discussion  of cancer slope factors.
Uncertainty and  modifying factors used in deriving
RIDs and RfCs should also be included in the discussion
of non-carcinogenic effects.

2.1.4  Risk Characterization
 Overview of methods for risk characterization. The
 last step  in  the baseline  risk assessment  is  risk
characterization.  This is the process of integrating the
 results of the exposure and toxicity assessments, by
 comparing estimates  of  intake with appropriate
 toxicological values to determine  (he likelihood of
 adverse effects in potentially exposed populations. Risk
 characterization  is considered  separately for
 carcinogenic  and non-carcinogenic effects, because
 organisms typically respond  differently  following
 exposure to carcinogenic and non-carcinogenic agents.
 For non-carcinogenic effects, lexicologists recognize
 the existence of a threshold of exposure below which
 there is likely to be no appreciable risk of adverse health
 impacts in an exposed individual. It is the current EPA
 position that exposure to  any level of carcinogenic
 compounds is considered  to carry a risk of adverse
 effect, and that exposure is not characterized by the
 existence of a threshold.
 EPA's procedure for calculating risk from exposure to
 carcinogenic compounds (EPA 1986a,   EPA 198.9a,
 EPA 1989b)uscsanon-threshold,dose-responsemodel.
 The model is used to calculate a cancer slope factor
 (mathematically, the slope of the dose-response curve)
 for each chemical.  Generally, (be cancer slope factor is
 used in conjunction with the chronic dally intake to
 derive a probabilistic upperbound  estimate of excess
 lifetime cancer risk to the individual.
                                                     17
The dose-response model most commonly used by EPA
in deriving ihe cancer slope estimates is linearized and
multistage. The mathematical relationship of the model
assumes that the dose-response relationship is linear in
the low-dose portion of the curve (EPA 1989a). Given
this assumption, the slope factor is a constant, and risk
is directly proportional to intake.

The recommended practice for evaluating the potential
for non-carcinogenic effects is to compare the RfD of a
given chemical to the estimated intake of Ihe potentially
exposed population from a given exposure  pathway
(EPA 1989a). This ratio (intakc/RfD) is  termed the
"hazard quotient." It is not a probabilistic estimate of
risk, but simply a measure of concern, or an indicator of
the potential for adverse effects.  A  more detailed
discussion of risk characterization ispresented in RAGS.
Further discussion of methods for risk characterization,
and of specific factors such as metabolic rate factors,
gender differences, and variable effects due to multiple
chemicals of potential concern, is available from many
sources (EPA 1988a, EPA 1989b, EPA 1989c).

Uncertainty analysis in risk characterization.  No
risk assessment is certain. Risk assessment Is a process
that provides an estimate of potential (present  and
future) individual risk, along with the limitations or
uncertainties associated with the estimates. The most
obvious effect of limitations in the analytical data on
risk characterization is the ability to accurately estimate
the potential for adverse effects in potentially exposed
 individuals. Clearly, if the available monitoring data do
notfacilitatcamcaningfuldctcrminaiion of RME values,
 the risk estimates will directly reflect this uncertainty.
     •• Uncertainties in toxicological measures
     and exposure assessment are often
     assumed to be greater than uncertainties in
     environmental analytical data;  thus, they.
     are  assumed to have a more significant
     effect  on the  uncertainty of the risk
     assessment.
 Resource and time constraints often limit the opportunity
 to develop a well-designed and comprehensive data set.
 Risk assessments must be conducted using the available
 information, even when there is no opportunity  to
 improve the data set. However, the results should be
 presented with an explicit statement regarding limitations
 and uncertainty.

 If possible, a sensitivity analysis should be conducted to
 bound the results of risk assessments. A simple approach
 might consist of establishing  the range of potential
  values (e.g., minimum, most likely, and maximum) for
  key input variables and discussing the influence on the
  resulting risk estimates.  The key variables can then be
  ranked with respect to the magnitude of potential effect
  on the  risk estimates.   In certain instances,  more

-------
quantitative approaches to uncertainly analysis may he
useful  if they can  be supported by the available
information. Combining probability distributions using
Monte Carlo techniques is one commonly cited example
(EPA 1988b. EPA 1989a. Fmkel 1990). An overview
of recommended methods for assessment of uncertainty
in risk characterization  is presented in  RAGS.
Risk*Assistant, a software tool developed for EPA,
provides an uncertainty analysis that determines the
effect on the final risk estimate of using alternative
parameter values, indicates the relative contribution of
each pathway to risks from the contaminated media, and
(for carcinogenic risks) determines the percentage of
total risk from a contaminant in each medium (Thistle
Publishing 1991).  A  more detailed consideration of
uncertainty analysis in risk assessment may be found in
Methodology for Characterization of Uncertainty in
Exposure Assessment (EPA 1985) and Confronting
Uncertainty inRiskManagenieni:AGuiilefor Decision-
Makers (Finkel 1990).


2.2 ROLES AND RESPONSIBILITIES
     OF KEY RISK ASSESSMENT
     PERSONNEL

The risk assessor generally enlists the participation of
 individuals with specific skills and technical expertise.
The quality and utility of the baseline risk assessment
 will ultimately depend on the planning and interaction
 of these technical professionals. Key participants include
 (he RPM and the  risk assessor, who are  primarily
 responsible for ensuring that data collected during the
 RI are useable for  risk assessment activities.  Other
 participants  include"hydrogeologists, chemists.
 statisticians, quality assurance staff, and other technical
 support personnel involved in planning and conducting
 the RI.   Exhibit 8 summarizes the roles and
 responsibilities of the risk assessment participants.

 2.2.1  Project Coordination

 All data collection activities that support the risk
 assessment are coordinated by the RPM. The RPM's
 responsibilities begin upon site listing and continue
 through deletion of the site from the National Priorities
 List.   A network of technical experts,  including
 representatives of other agencies involved in human
 health or environmental/ecological assessments or
 related issues, is established at the start of the RI. This
 ensures that the potential for adverse effects to human
 health and the environment is adequately assessed during
 the RI. To successfully plan and direct the sampling and
 analysis  effort, the RPM  must facilitate interaction
 among key participants.
2.2.2  Gathering Existing Site Data
        and Developing the Conceptual
        Model

The RPM is responsible for gathering and evaluating all
historical and existing site data. This is an important
clement in phuming the scope of the risk assessment and
data collection, and in determining additional data needs.
Sources of information especially  pertinent for risk
assessment include data from potentially responsible
parties, industrial records identifying chemicals used in
processes, preliminary natural resource studies, Agency
for Toxic Substances and Disease Registry (ATSDR)
health studies, environmental impact statements,
transport manifests, site  records, site inspection
documents, and site visits.  Aerial photographs and site
maps showing past and present locations of structures
and transportation corridors should also be collected.
The RPM should also consider the application of  a
computer-based Geographical Information  System
(GIS) as a major tool.

The RPM should ensure that a broad spectrum analysis
was conducted at the site for all media and should
review  industry-specific records to minimize the
potential for false negatives.  From the inspection of
historical data and broad  spectrum  analyses,  a
preliminary list of the chemicals of potential concern is
prepared to assist in scoping and  in developing the
conceptual model of the site.  Once all the existing
historical site data have been collected, the RPM works
with  the risk assessor to develop a conceptual model.
The conceptual model is a depiction and discussion of
the current understanding of the contamination, the
sources of release to the environment, transport
pathways, exposure pathways, exposure areas and
receptors at risk. Preliminary identification of potential
exposure pathways at the site  under investigation  is
 particularly important for the design of a thorough data
 collection effort. The conceptual site model should be
 provided to all key participants in the RI during the
 project scoping and should be included in the workplan.
 As work progresses and the site is better characterized,
 the RPM and the risk assessor  should update the
 conceptual model.

 2.2.3  Project Scoping

 The  adequacy of the sampling and analysis effort
 determines the qualityoflheri.sk assessment. Therefore,
 it is imperative that  the risk  assessor be an active
 member of RI planning and continue to be involved
 during the entire course of the project.

-------
           EXHIBIT 8.  ROLES AND RESPONSIBILITIES OF
                   RISK ASSESSMENT TEAM MEMBERS
Remedial project manager
• Directs, coordinates and monitors all activities.
• Establishes network with other data users including federal, state and local agencies.
• Creates conceptual model.
• Gathers existing site data.
• Organizes scoping meetings.
• Controls budget and schedule.
• Guides preparation of QA documents.
• Ensures that the risk assessor receives preliminary analytical data.
• Contributes to data assessment.
• Develops preliminary list of chemicals of potential concern.
• Resolves problems affecting P.I objectives, including risk assessment issues (e.g., resampling,
  reanalysis).
Risk assessor
• Reviews all relevant existing site data.
• Assists the RPM in developing the conceptual model and the preliminary list of chemicals of potential
  concern.
• Contributes to recommendations on sampling design, analytical requirements, including chemicals of
  potential concern, detection limits and quality control needs during project scoping.
• Helps to refine (he conceptual model.
• Communicates frequently with the RPM, hydrogeologist and chemist to ensure that data collection
  meets needs.
• Reviews and contributes to SAP and QA documents.
• Assesses preliminary data as scon as available to verify conceptual site model.
• Specifies additional needs.
• Assesses reviewed data for useability in risk assessment.
• Communicates all site activities with specific groups, such as chemists.
• Prepares risk assessment.
 Hydrogeologlst, chemist and other technical support
 • Provides technical input to scoping.
 • Prepares/provides input to SAP and QA documents in support of risk assessment data needs.
 • Communicates frequently with the RPM and/or risk assessor on status of data collection and issues
  affecting data.
 • Provides preliminary data to the RPM and/or risk assessor for review.
 • Supports fate and transport modeling for the exposure assessment.
 • Implements corrective actions to improve data useability.
 Quality assurance specialist
 • Responsible for data quality review and technical assistance in preparing QA documents.
 • Provides historical performance QA data or recommendations for appropriate QC.
 • Ensures adequate QA procedures are In place, including field and analytical audits.
                                                                                        21-002-OOS
                                             19

-------
    «• Analytical data collected solely for other
    purposes may not be of optimal use to the
    risk assessment.

Data obtained solely with the aim of characterizing the
nature and extent of contamination at a site may not
fully support the needsof the risk asscssorinquantitating
exposure, and therefore the potential for adverse effects
in human and nonhuman receptors. Data on the nature
and extent of contamination may therefore be rejected
by the risk assessor, requiring an additional round of
sampling. For example, data identifying the boundaries
of the site may not be representative of the level of
contamination within an exposure area. Therefore, it is
important  to maintain the risk  assessment data
requirements as  a high priority throughout remedial
investigations.

Sampling and analysis methods discussed during scoping
should ultimately be based on site-specific data needs.
The RPM, risk assessor, hydrogeologist, statistician.
and project chemist must mainiain open communication
during scoping and throughout the RI to ensure that this
occurs. Data review and deliverable requirements should
be determined during the scoping meetings so that these
specifications can be included  in the sampling and
analysis plan (SAP) for the  RI.  The RPM  should
prepare a checklist of considerations for the scoping
meetings and provide it  to all  individuals involved.
Exhibit 9 presents an example checklist of items useful
for risk assessment to be considered by the RPM during
scoping.  Chapters 3 and 4 give specific guidance for
planning the data collection  efforts to support risk
assessments.

2.2.4  Quality Assurance  Document
        Preparation and Review

After scoping, the RPM guides  the preparation of the
workplan and quality assurance documents.  The
workplan, the SAP. and the quality assurance project
plan (QAPjP) should document the combined decisions
of the RPM, risk assessor, and other project staff.
                             EXHIBIT 9. EXAMPLE RISK ASSESSMENT
                                 CHECKLIST FOR USE IN SCOPING
                        • Has all historical inlormation been gathered and characterized
                          and is it appropriate and available for use?

                        • What sample matrices should be investigated?

                        • What analytical methods should be used?

                        • Are the methods appropriate tor risk assessment, given
                          specific contaminants present and their toxic'rty?

                        • Wilt any special quality control requirements be necessary?

                        • Who will conduct the anarysis (e.g., which type of laboratory)?

                        • What analytical data sources should be used (fixed laboratory
                          and/or field analysis)?

                         • What sampling designs are appropriate?

                         • How many samples will be needed?

                         • How will the data review be accomplished?

                         • What types of deliverables will be required? Specify the types of
                          delh/erables required from both laboratory and data validation.

                         • What budget or other limitations constrain data collection (e.g.,
                          due date, contractor availability)?
                                                   20


-------
I^irticular emphasis isplaccd on establishing confidence
limits, acceptable error, and level of quality control
(discussed in Chapter 3). This facilitates cost-effective
design of the sampling and analytical program and
minimizes the collection of data of limited use for risk
assessment.

The risk assessor reviews the workplan  and SAP to
ensure that (he relevant data quality issues, sampling
design, analytical needs, and data assessment procedures
are adequately addressed for risk assessment. Exhibits
10 and 11 provide checklists to aid the review of the
workplan and SAP.

2.2.5  Budgeting and Scheduling

As the overall site manager, the RPM must address and
balance risk assessment data needs with other data use
needs, such as health and safety, treatability studies,
transport, and the nature and extent of contamination.
The risk assessor is responsible for identifying specific
data  requirements  for  risk  assessment and
communicating these needs to the RPM.  The RPM is
responsible  for developing and implementing the
schedule for acquiring the data.  Balancing costs and
services while adhering to  the schedule is a major
responsibility of the RPM.

'Hie RPM must coordinate tlte use of analytical services.
Data from  different analytical sources  provide the
flexibility needed to balance cost with sampling needs
and lime constraints. The advantages anddisad vantages
of field analyses and fixed laboratory analyses should
be considered, as described in Chapters 3 and 4. The
risk assessment participants can assist in the development
of field sampling plans and the selection of appropriate
analytical methods that will provide the risk assessor
with a set of useable data, within the budgeting and
scheduling constraints of the RPM.

2.2.6  Iterative Communication

Continuing, open, and frequent communication among
the participants is critical to the success of the Rl :utd
baseline risk assessment. A single meeting or discussion
is rarely adequate to ensure that all relevant issues have
been addressed. Development of the risk assessment
within the RI report is  an iterative process of action.
feedback, and correction or adjustment.

After re view of the workplan, the SAP, and the QAPjP,
the  RPM monitors the flow of information.  The risk
asscssorassists the RPM toensure that the data produced
are incompliance with therequirementsof the workplan
and SAP.  Key questions they consider once the data
become available are:

  •  Have correct sampling protocols been followed?

  •  1 lave all critical samples been collected?
        EXHIBIT 10.  CHECKLIST FOR REVIEWING THE WORKPLAN
   •  Does the workplan address the objectives of baseline risk assessment?

   •  Does the workplan document the current understanding of she history and the physical setting?

   •  Have historical data been gathered and assessed?

   •  Has information on probable background concentrations been obtained?

   •  Does the workplan provide a conceptual site model for the baseline risk assessment, including a
     summary of the nature and extent of contamination, exposure pathways of potential
     concern, and a preliminary assessment of potential risks to human health and the environment?

   •  Does the workplan document the decisions and evaluations made during project scoping,
     including specific sampling and analysis requirements for risk assessment?

   •  Does the workplan address all data requirements for the baseline risk assessment and explicitly
     describe the sampling, analysis and data review tasks?
                                                                                         21-002-010
                                                 21

-------
          EXHIBIT 11. CHECKLIST FOR REVIEWING THE SAMPLING
                                   AND ANALYSIS PLAN
     •   Do the objectives of the QAPjP and the field sampling plan meet risk assessment needs
         established in the scoping meeting?

     •   Are QA/QC procedures provided for in the SAP adequate for the purposes of the baseline
         risk assessment?

     •   Have the data gaps for risk assessment that were identified in the Rl workplan been
         adequately addressed in the SAP?

     •   Are there sufficient QC samples to measure the likelihood of false negatives and false
         positives, and to determine the precision and accuracy of resulting data?

     •   Have analytical methods been selected that have detection limits adequate to quantitate
         contaminants at the concentration of concern?

     •   Have SOPs been prepared for sampling, analysis and data review?

      •   Will the sampling and analysis program result in the data needed for the baseline risk
         assessment:

           -  to address each medium, exposure pathway and chemical of potential concern,
           -  to evaluate background concentrations,
           ••  to provide detail on sample locations, sampling frequency, statistical design and analysis,
           --  to evaluate temporal as well as spatial variation, and
           -  to support evaluation of current as well as future resource uses?
  •  Have the samples been analyzed as requested?

  •  Are data arriving in a timely fashion?

  •  Have appropriate sample quantitation limits/detec-
    tion limits been achieved?

  •  Has quality assurance been addressed as stated in
    UieSAPandQAPjP?

  •  Have the data been reviewed as stated in the SAP?

  • Is the quality of the analytical data acceptable for
    their intended use?

Based upon theseconsideraiions, the RPM.riskassessor
and other technical team members must jointly determine
if any corrective actions are needed, such as requesting
additional sampling, using alternative  analytical
methods, or reanalyzing samples.

2.2.7   Data Assessment

The RPM and risk assessor work with other JP* ticipants
to identify a list of chemicals of potential concern and
                                     21-002411
decide on data review procedures. This information is
developed during project scoping and incorporated into
the workplan and SAP. The RPM, risk assessor, and
project chemist should agree on  the type and level of
data review required for both positive and "non-detect"
results. Typically, the RPM assesses the overall data
reviewed by the chemist, and the risk assessor reviews
data relevant to risk assessment, unless other
arrangements have been established and explicitly stated
in the SAP.

The risk assessor may request preliminary data, or
results that have received only a partial review, in order
toexpedite the reassessment to save time and resources.
Preliminary data can be used to validate the conceptual
model or to begin the toxicity assessment. The data may
also indicate aneed formodifying sampling or analytical
procedures. However, preliminary data should not be
used in calculating risk. Once the full analytical data set
is obtained, the RPM and risk assessor should consult
with the project chemist and statistician to assess the
utility of all available information.
                                                22

-------
2.2.8  Assessment and Presentation
        of Environmental Analytical
        Data

Once environmental data are evaluated in the data
review process, the risk assessor develops a final data
set for use in the baseline risk assessment All chemicals
of potential concern should now be identified. The risk
assessor prepares summary tables containing the
following information:

   •  Site name and sample locations,

   •  Number of samples per defined, representative
     areaof each medium (e.g., do not count background
     samples together with other samples),

   •  Sample-specific results,

   •  Analyte-specific sample quantitation limits,

   •  Number of values above the quantitation limit.
  •  Measures of central tendency (e.g., 95% UCL on
     the  arithmetic  mean of the environmental
     concentration),

  •  Specifications for the treatment of detection or
     quantitation limits and treatment of qualified data,
     and

  •  Ranges of concentrations.

All assumptions, qualifications, and limitations should
be explicitly stated in the tables.  The risk assessor
provides the final data summary tables to the RPM,
project hydrogeologist, project chemist, and other
appropriate project staff for re view. These are the data
that will be  used  in the baseline risk assessment to
determine the potential risk to human health.  It is
essential, therefore, that this information consists of the
best data available and reflects the collective review of
the key participants in the risk assessment. An example
of such a set of data is given in Appendix I.
                                                   23

-------
                                          Chapter 3
        Useability Criteria for  Baseline  Risk  Assessments
This chapter applies data useability criteria to data
collection planning efforts to maximize the useability of
environmental analytical data in baseline risk
assessments.  It also addresses preliminary issues in
planning sampling and analysis programs.

The chapter has two sect ions. Section 3.1 discusses data
useability criteria involved in risk  assessment and
suggests ways they can be applied to ensure data are
useable. Section 3.2 presents preliminary sampling and
analysis issues including identification of chemicals of
potential concern, available sampling and analytical
strategies or  methods, and probable sources  of
uncertainly.

Before scoping the RI, it is critical for successful planning
that the RPM develop a conceptual site model (Exhibit
6)  in consultation  with the risk assessor and all
appropriate  personnel.  This chapter provides the
background information necessary  to plan for the
acquisition of environmental data for baseline risk
assessments.  The  quality  of a risk assessment is
intimately tied to the adequacy  of the sampling and
analysis plan (SAP)  developed during the RI.

    *• Effective planning improves the useability
    of environmental analytical data in the final
    risk assessment.

Data needs for  baseline risk assessments are not
necessarily met by data the RPM acquires to identify the
nature and extent of contamination at a Supcrfund site.
For example, asampling strategy designed to determine
the boundaries of a conuuninated area may not provide
data to quamitate concentrations within an exposure
area.   The risk  assessment may also require more
precision and accuracy, and lower detection  limits.
Accordingly,  the risk assessor  should be an  active
member of the team planning  the RI and must be
consulted from the start of the planning process.

Four fundamental decisions for risk assessment are to
be made with the  data acquired during the  RI, as
discussed in Chapter 2.

   • If the sampling design is representative, the
     question of what contamination is present and at
     what concentration is an analytical problem. Key
     concerns are the probability of false negatives and
     false positives. The selection of analytical methods,
     laboratory performance, and type and amount of
     data review affects these issues for both site and
     background samples.

   • Assuming that chemicals  of potential concern
     have been identified, the second question involves
     background levels of conuuninution.  Are site
     concentrations  sufficiently elevated  from true
     background levels to indicate an increased risk for
     human health due to site contamination?

  •  All exposure pathways and exposure areas must
     be identified and examined. The two decisions
     concerning exposure pathways and areas primarily
     involve identifying and sampling the media of
     concern.

  •  The final decision involves characterizing exposure
     areas.  Sampling  and  analysis  must  be
     representative and satisfy performance objectives
     determined during the planning process.

RI planning and implementation of RI plans affect the
certainty of chemical identification and quantitation.
Therefore, the RI needs to collect useable environmental
analytical data to enable the risk assessor to make these
decisions.
                  Acronyms

 AA        atomic absorption
 CLP       Contract Laboratory Program
 CRDL      contract required detection limit
 CRQL      contract required (juanlilation limit
 DQI       data quality indicator
 DQO       data quality objective
 GC        gas chromatography
 MRS       Hazard Ranking System
 ICP        inductively coupled plaxma
 IOL       instrument detection limit
 LOL       limit of linearity
 LOQ       limit of quantitation
 MDL       method detection limit
 MS        mass spectrometry
 OVA       organic vapor analyzer
 PA/SI      primary assessment/site inspection
 PAH       polycyclic aromatic hydrocarbon
 PCB       polychlorinated biphenyl
 PQL       practical quantitation limit
 QA        quality assurance
 QC        quality control
 QAPjP     quality assurance project plan
 QTM       Quick Turnaround Method
 RI         remedial investigation
 RI/FS      remedial investigation/feasibility study
 RPM       remedial project manager
 RRF       relative response factor
 RRT       relative retention time
 SAP       sampling and analysis plan
 SOP       standard operating procedure
 SQL       sample quantitation limit
 TIC       tentatively identified compound
 TRIS       Toxic Release Inventory System
 XRF       X-ray fluorescence
                                                   25

-------
3.1  DATA USEABILITY CRITERIA

Exhibit 12 lists the six data useability criteria involved
in planning for the risk assessment, summarizes the
importance of each criterion to risk assessment, and
suggests actions to take during the planning process to
improve the useability of data. The following sections
define each criterion and describe its effect on risk
assessment

3.1.1  Data Sources

The data sources selected during the RI planning process
depend on the type of data required and their intended
use.  Data collected prior to the RI are considered
historical; data collected during the RI are considered
current and are usually specified in the RI planning
process. Data may be analytical or non-analytical. The
same analytical data requirements apply, whether the
data are current or historical. Field screening methods
can be used, and sufficient documentation produced, to
act as an initial source of data. The minimum criteria lor
 analytical data are discussed in Chapter 5.
                                                        Data sources prior  to  remedial investigation.
                                                        Historical data sources are useful for determining
                                                        sampling locations and analytical approaches in the RI.
                                                        Early site inspections may locate  industrial process
                                                        information that suggests chemicals of potential concern.
                                                        Historical data indicate industry-specific analytes and
                                                        general levels of contamination and trends (hat are
                                                        useful for identifying exposure pathways, for developing
                                                        the sampling design, and for selecting analytical methods.
                                                        Historical analytical data are often available from the
                                                        preliminary assessment/site  inspection  (PA/SI),
                                                        including reports on the physical testing, screening, and
                                                        analysis of samples. Other sources of analytical data for
                                                        baseline risk assessment include the Hazard Ranking
                                                        System (HRS) documentation, site records on removal
                                                        and disposal, and industry-specific systems for chemical
                                                        discharge permits.  Results from analyses by state  or
                                                         local governments may also indicate chemicals  of
                                                         potential concern. Exact locationai data for historical
                                                         samples should be obtained whenever possible.

                                                             «• Use historical analytical data and a broad
                                                             spectrum analysis to initially identify the
                                                             chemicals of potential concern or exposure
                                                             areas.
                                                          The quality of historical data must be determined prior
                                                          to their use in the RI. For historical analytical data to be
Exhibit 13 identifies available data sources and their
primary uses in the risk assessment process. Historical
and current analytical data sources are briefly discussed
below.
              EXHIBIT 12. IMPORTANCE OF DATA USEABILITY CRITERIA
                    IN PLANNING FOR BASELINE RISK ASSESSMENT
       Data
     Usability
     Criterion
                                   Importance
          Suggested Action
   Data Sources
   (3.1.1)
                 Data sources must be comparable if data are combined for
                 quantitative use in risk assessment. Plans can be made in
                 the RI for us* of appropriate data sources so that data
                 compatibility does not become an issue.
Use data from different data sources together to
balance turnaround Ime. quality of date, and
cost  Consult with a chemist or statistician to
assess compatibility of data sets.
   Documentation
   (3.12)
                 Deviations from the SAP and SOPs must be documented
                 so that the risk assessor w» be aware of potential
                 imitations In the data. The risk assessor may need
                 additional documentafon. such as field records on weather
                 conditions, physical parameters and site-spedAc geology.
                 Date useabto for risk assessment must be linked to a
                 specific location.
Review the workplan and SAP and. if
appropriate. SOPs. As the date arrive, check
for adherence to the SAP so that corrective
action such as resampling may be taken and still
adhere to tie project timetable.

Stress importance of chabvof-custody for
sample point Identification In RI planning
meetings.
   Analytical
   Methods and
   Detection
   Limits
   (3.1.3)
                 The method chosen must test for Ihe chemical of potential
                 concern at • detection fcntt that wH meet the concentration
                 levels of concern tnapplcablemaWces. Samples may
                 have to be reanalyzed at • tower detection Bmlt If he
                 detection In* to not tow enough to confirm the presence
                 and amount of mntairinafen.
 Participate with chemist In selecting methods
 with appropriate detection Imits during RI
 planning. Consultation with a chemist Is
 required when a method's detection Bmtt is at or
 above the concentration level of concern.
                                                                                                    21-002-012
                                                       26

-------
           EXHIBIT 12. IMPORTANCE OF DATA USEABILITY CRITERIA
                  IN PLANNING FOR BASELINE RISK ASSESSMENT
                                                  (Cont'd)
    Data
Usability
 Criterion
Data Quality
Imfcators
(3.1.4)

Completeness
Comparability
Representa-
tiveness
Precision
Accuracy
 Data Review
 (3.1.5)
 Reports
 to Risk
 Assessor
 (3.1.6)
                                  Importance
Completeness for critical samples must be 100%.
Unforeseen problems during sample collection (as defined
In Chapter 4) and analysis can affect data completeness.
If • sample data set for risk assessment is not complete.
more samples may have to be analyzed, affecting Rl time
and resource constraints.

The risk levels generated in quantitative risk assessment
may be questionable if Incompatible data sets are used
together.
Sample data must accurately reflect the site
characteristics to effectively represent the site's risk to
human health and the environment. Hot spots and
exposure area media must have representative data.
                 If the reported result b near the concentration of concern.
                 H Is necessary to be as precise as possible in order to
                 quantify the likelihood of false negatives and false
                 positives.
 Quantitative accuracy information Is critical when results
 are reported near the level of concern.  Contamination in
 the field, during shipping, or in the laboratory may bias the
 analytical results. Instruments that are not caltorated or
 tuned according to Statement of Work requirements may
 also bias results. The use of data that Is biased may affect
 the interpretation of risk levels.
 Use of preliminary data or partially reviewed data can
 conserve time and resources by allowing modification of
 the sampling plan whSe tie Rl is in process. Critical
 anafytes and samples used tor quantitative risk
 assessment require a fun data review.
 Data reviewers should report data in • format hat provides
 readability as wel as clarifying information. SQLs. a
 narrative, and quaifiers that are July explained reduce the
 time and effort required In Interpreting and using the
 analytical results. Limitations can be readily identified and
 documented in the risk assessment report
                                                                Suggested Action
Define completeness in the SAP for both the
number of samples and quantity of useabte data
needed to meet performance objectives.
Identify critical samples during scoping. The
SAP should be reviewed by the RPM before
Initiation of sampling.

Plan to use comparable methods, sufficient
quality control, and common units of measure for
different data sets that will be used  together, to
faciilate data compatability. Consult with a
chemist to ensure comparibility of data sets.

Discuss  plans for collection of sufficient number
of samples, a sample design that accounts for
exposure area media, and an adequate number
of samples for risk assessment during scoping
and document plans in the SAP. This guidance
may be modified by Region-specific guidelines.

Plan for  the use of QC samples (duplicates,
replicates and/or collocated samples) applicable
to risk assessment before sampling activities
begin.  Assess confidence limits from the QC
data on the basis of thu sampling design or
analytical method used.

Plan and assess QC data (blanks, spikes,
performance evaluation samples) to measure
bias in sampling and analysis. Consult a
chemist  to Interpret data qualified as
•estimated" that are near a concentration of
concern.
 Decisions regarding level and depth of review wll
 conserve time and project resources and should
 be made In conjunction with the RPM and
 analytical chemist "Non-delecr results require
 a full review.
 Prescribe a report formal during scoping, and
 include it in the SAP. Communicate with the
 potential data reviewer to aid the definition of a
 specific report formal. Region-specific
 guidelines may apply.
                                                                                                           ai-002-012-01
                                                         27

-------
                     EXHIBIT 13. DATA SOURCES AND THEIR
                              USE IN RISK ASSESSMENT
          Available Data
             Source*
 Data Type
             Primary U«e(t)
       PA/SI data
Analytical
 Scoping and planning
 Identifying data trends
 Determining historical background levels
       HRS
       documentation
Site records,
manifests.
PA/SI,
analytical
 Quantitating the risk assessment
 Identifying trends
 Planning (by identifying the chemicals present)
       Site records on
       removal and disposal
Administrative
 Planning (by identifying the chemicals present)
       Toxic Release
       nventory System
       TRIS) (Industry-
       Specific)
Chemical
discharge
 Planning (by identifying the chemicals present)
       Site, source and
       media characteristics
       as found in PA/SI data
       and reference
       materials
Physical
parameters
(e.g., meteor-
ological,
geological)
 Determining fate and transport
 Defining exposure pathways
       -ield screening
 Analytical
 Performing a preliminary assessment
 1 Characterizing the site
       Field analytical
 Analytical
1 Quantitating the risk assessment
' Characterizing the site
       Fixed laboratory,* both
       CLP and-Ron-CLP
       (EPA. state. PRP.
       commercial)
 Analytical
• Quantitating the risk assessment
• Providing a reference
• Broad screen
• Confirming screening data
• Characterizing a she      	
          Mobile laboratories often have the same instrumentation available as fixed laboratories,
          with the exception of ICP or MS.
                                                                                      21-002-OU
useful in the quantitative risk assessment, sampling
design, sampling and analytical techniques, and detection
limits must be documented, and the data must have been
reviewed.

Historical analytical data of unknown quality may be
used in developing the conceptual model or as a basis
for scoping, but  not in determining representative
exposure concentrations. Analytical data from the PA/
SI that meet minimum data useability requirements (see
Section 5.1.1) can be combined with data from the RI to
                        estimate exposure concentrations. Similarly, historical
                        data of lower quality may be used if the concentrations
                        are confirmed by subsequent RI analyses.

                        Data sources for the remedial investigation. It may
                        be efficient to use a variety of data sources during an RI.
                        For example, analytical services providing a  rapid
                        turnaround of estimated data can be used to estimate tbr.
                        three-dimensional extent of contamination or to "chase"
                        a groundwater  pollutant plume.  Rapid turnaround
                        analytical services include  field analysis or Quick
                                                  28

-------
Turnaround Method (QTM) analyses under the Contract
Laboratory Program (CLP).  On the other hand, if an
unexpected situation arises, such as the discovery of
buried drums on the site, it may be appropriate to
procure the analytical services of a local commercial
laboratory.   Data requiring a rapid  turnaround are
typically produced from streamlined analytical methods.
and a certain percentage should be analyzed  using a
confirmatory method, such as CLP analytical services.
The planning process for the RI identifies gaps in the
available analytical data and determines additional data
collection requirements. Three types of analytical data
sources can be used during the RI to acquire analytical
data for a risk assessment. These include Held screening.
Held analyses, and fixed laboratory analyses.

   •  Field screens are performed using chemical field
     lest kits, ion-specific probes, and other monitoring
     equipment,  but should  be confirmed by other
     techniques.  Field screening is usually performed
     to provide a preliminary assessment of the type
     and  level of concentration of the chemicals of
     potential concern.

   •  Field analyses are performed using instruments
     and  procedures  equivalent to fixed laboratory
     analyses; they produce legally defensible data if
     QC procedures are implemented. Field analyses
     are usually  performed as part of  an integrated
     sampling  and analysis  plan to quantitate  risk
     assessment and site characterization.

   •  Fixed laboratory analyses are particularly useful
     for broad spectrum and confirmation analyses.
     They often  provide  more detailed information
     over a wider range of analytes than field analyses.
     Fixed laboratory analyses are critical toquantitative
     risk  assessment and site characterization.

A discussion of issues related to fieldand fixed laboratory
analyses is presented in Section 3.2.9.

Analytical services constitute a significant portion of
the Superfund budget and should be conserved when
possible.  CLP costs do not appear on the remedial
investigation/feasibility study (RI/FS) project budget.
Analyte-spccific methods may be used  for chemicals
identified after a broad  spectrum analysis by CLP or
other fixed laboratory analysis, and may provide more
accurate results. Site samples analyzed by CLP routine
analytical services take an average of 35 days to produce
results and data review will add to the overall turnaround
time. Other data sources, such as a mobile laboratory or
CLP QTM or special analytical services, can quickly
produce good "first look" results which can be followed
up immediately whileon site. Mobile laboratory services
can replace some CLP services if analytical capabilities
are adequately demonstrated by method validation data
and if m inimum QC requirements are met (seep. 59). At
least 10% of sample analyses should be confirmed by
fixed laboratory analysis in all situations.

3.1.2  Documentation

Data collection and analysis procedures must be
accurately documented to substantiate the analysis of
the sample, conclusions derived from the data, and the
reliability of the reported analytical data. Plans should
be prepared during the RI scoping to document data
collection activities.  This RI documentation can be
used later to evaluate completeness, comparability,
representativeness,  precision,  and accuracy of the
analytical data sets. Four major types of documentation
are produced during an RI:

   •  Thesamplingandanalysisplan,includingaquality
     assurance project plan (QAPjP),

   •  Standard operating procedures (SOPs),

   •  Field and analytical records, and

   •  Chain-of-custody records.

Sampling and analysis plan. The scoping meetings
and  the SAP must clearly establish  the end use
requirements for data. The data quality indicators for
assessing results against stated performance objectives
should also be documented in  the SAP (see Section
3.1.4). The SAP includes the QAPjP and information
required in the SOPs, field and analytical records, and
chain-of-custody records (EPA 1989a).
Standard  operating procedures and field and
analytical  records.  SOPs for field and analytical
methods must be written for all field and laboratory
processes. Adherence to SOPs provides consistency in
sampling and analysis and reduces the level of systematic
errorassociated withdatacollectionandanalysis. Exhibit
14 lists the types of SOPs, field records, and analytical
records that are usually associated wUhRI data collection
and analyses, and relates the importance of each to the
risk assessment.
All deviations from the referenced SOPs should be pre-
approved by the RPM and documented. Samples that
are  not collected or analyzed in accordance with
established SOPs may be of limited use because their
quality cannot be  determined.

Chaln-of-custody.  The technical team must decide
during scoping what data may be used for cost recovery
actions, and plan  accordingly for the use of full-scale
chain-of-custody or  less formal chain-of-custody
procedures. Full-scale chain-of-custody is required for
                                                    29

-------
                    EXHIBIT 14.  RELATIVE IMPORTANCE OF
            DOCUMENTATION IN PLANNING AND ASSESSMENT
Documentation
Sampling and Analysis Plan
Selection and identification of sampling points
Sample collection SOP
Analytical procedures or protocols
SOP for data reporting and review
QA project plan
Method-specific QC procedures
QA/QC procedures
Documented procedures for corrective action
SOP for corrective action and maintenance
Sample preservation and shipping SOP
SOPs for sample receipt, custody, tracking and storage
SOP for installation and monitoring of equipment
Chain-of-Custody
Documentation records linking data to sample location
Sampling date
Sample tags
Custody seals
Laboratory receipt and tracking
Field and Analytical Records
• Field log records
• Field information describing weather conditions, physical parameters
or site-specific geology
• Documentation for deviations from SAP and SOPs
• Data from analysis - raw data such at instrument output, spectra,
chromatograms and laboratory narrative
• Internal laboratory records
Importance 1
Critical
High
High
High
High
Medium
Medium
Medium
Medium
Medium
Low
Low
Critical
Critical
High
Low
Low
High
High
High
High
Low
KEY Critical - Essential to the useability of data for risk assessment
High • Should be addressed in planning for risk assessment
Medium • Primarily impacts how data are qualified in risk assessment
Low • Usually has little effect on useability of data for risk assessment
                                                                               J1-002-014
cost recovery and enforcement actions, but does not
affect a quantitative determination of risk. Full-scale
chain-of-cusuxly includes sample labels and formal
documentation that prove the sample was not tampered
with or lost in the data collection and analysis process.
Sample identity must be verifiable from the collector's
notebook and laboratory data sheets, as well as from a
formal chain-of-custody.

3.1.3   Analytical Methods and
        Detection Limits

The choice of analytical methods is important in  RI
planning. Appropriate analytical methods have detection
limits that meet risk assessment requirements for
chemicals of potential concern and have sufficient QC
measures to quantiuuc target compound identification
and measurement. The detection limit of the method
directly affects the useability of data because chemicals
reported near the detection limit have agreatcrpossibility
of false negatives and false positives. The risk assessor
or RPM must consult a chemist for assistance In choosing
an analytical method when those available have detection
limits near the required action level. Wheneverpossible,
methods should not be used if the detection limits are
above the relevant concentrations of concern.
                                               30

-------
I
                 3.1.4  Data Quality Indicators

                 Data quality indicators (DQIs) arc identified during the
                 development of data  quality objectives (DQOs),  to
                 provide quantitative measures of the achievement of
                 quality objectives. This section discusses each of five
                 DQIs as they relate to the assessment of sampling and
                 analysis.

                    •  Completeness

                    •  Comparability

                    •  Representativeness

                    •  Precision

                    •  Accuracy

                 These indicators :uv evaluated through the review of
                 sampling and  analytical data and accompanying
                            documentation.  The risk assessor may  need to
                            communicate with a chemist or statistician after the data
                            collection process has been completed to evaluate DQIs.
                            Therefore, the SAP, Held and analytical records, and
                            SOPs should  be accessible.   Exhibits  IS and  16
                            summarize the importance  of DQIs to sampling and
                            analysis in risk assessment and suggest planning actions.

                            Each DQI is defined  in this section.  Note that the
                            specific use of the indicators to measure data uscability
                            is different for sampling and analysis.  For example,
                            completeness a^applicd to sampling refers to the number
                            of samples to be collected. Completeness as applied to
                            analytical performance primarily refers to the number
                            of data points that indicate an analytical result for each
                            chemical of interest (e.g., 10 samples analyzed for 25
                            chemicals will produce a total of 250  data points, 10
                            data points for each chemical).
                          EXHIBIT 15.  RELEVANCE OF SAMPLING DATA QUALITY INDICATORS
                              Data Quality
                               Indicator*
           Importance
                                                                                 Suggested Planning Action
                           Completeness
Complete materials enable assessment
of sample representativeness for
Identification of false negatives and
estimation of average concentration.
Stipulate SOPs for sample
collection and handling In
the SAP to specify requirements for
completeness.
                           Comparability
Comparable data give the ability to
combine analytical results across
sampling episodes and time periods.
Use the same sample design across |
sampling episodes and similar time
periods.
                           Representativeness
Representative data avoid false negatives
and false positives (field sampling
contamination).

Non-representative data may result In
bias of concentration estimates.
Use an unbiased sample design.

Collect additional samples a's
required.

Prepare detailed SOPs for handling
field equipment
                           Precision
Variability in concentration estimates may
Increase uncertainty.
                                                                                 Increase number of samples.

                                                                                 Use appropriate sample designs.

                                                                                 Use QC results for monitoring.
                           Accuracy
Contamination during sampling process,
loss of sample from Improper collection or
handling (loss of volatile!) may result In
bias, false negatives, or false positives
and Inaccurate estimates of
concentration.
Use SOPs for sample collection,
handling, and decontamination.

Use QC results for monitoring.
                                                                     31

-------
EXHIBIT 16. RELEVANCE OF ANALYTICAL DATA
            QUALITY INDICATORS
Data Quality
Indicators
Completeness
Comparability
Representativeness
Precision
Accuracy
Importance
Poor data quality or lost samples
reduces the size of the data set
and decreases confidence in
supporting information.
Comparable data allow the ability
to combine analytical results
acquired from various sources
using different methods for
samples taken over the period of
investigation.
Non-representative data or
non-homogeneity of sample
increases the potential for false
negatives or false positives.
Potential for change in sample
before analysis may decrease
representativeness.
Monitoring can indicate the level
of precision.
Precision provides the level of
confidence to distinguish
between site and background
levels of contamination. It is of
primary importance when the
concentration of concern
approaches the detection limit.
Accuracy also provides the level
of confidence to distinguish
between site and background
levels of contamination. As
concentration of concern
approaches the detection limit,
the differentiation includes
confidence In determining
presence or absence of chemical
of potential concern.
l^^__^^BHHl_^HaH^^^^HIB^^^«HHH
Suggested Planning Action
Prepare SOPs to support sample
tracking and analytical procedures,
review, and reporting aspects
of laboratory operations.
Reference anatyte-specific method
performance characteristics.
Reference applicable fate and transport
documentation.
Anticipate field and laboratory
variability.
Include requirement for broad spectrum
analyses across site area.
Ensure sample is mixed and adequately
represents the environment (not
applicable to volatiles).
Include provision for blank (transport,
storage and analytical) QC monitoring.
Use field methods when applicable,
since they have an advantage in
minimizing variability from transport and
storage.
Method OC component and site-specific QC
samples that use external reference are the
best monitoring techniques.
Consider in method selection whether
anticipated site levels are near the MDL and
above action limits.
Broad spectrum screening methods may
have significant negative bias for chemicals
of potential concern. Consider method
accuracy and detection limits if site levels
approach concentrations of concern.
_^^_^^^B_^_l^^^__BB^^^^__i
                                              21-002-018
                      32

-------
Completeness.  Completeness  is a measure  of the
amount of uscnble data resulting from a data collection
activity. The required level of completeness should be
defined in the QAPjPforthe numberof samples required
in the sampling design and for the quantity of uscable
data for chemical-specific data points needed to meet
performance objectives. All required data items must
be obtained for critical samples and chemicals, which
are identified in the QAPjP. Incompleteness in any data
item may bias results as well as reduce the amount of
useablc data.
Problems that occur during data collection and analysis
affect the completeness of a data set. Fewer samples
may be collected and analyzed than originally planned
because of site access problems. Laboratory performance
may be affected if capacity is exceeded, causing data to
be rejected. Some samples may not be analyzed due to
matrix problems.  Samples that are invalid  due to
holding time violations may have to be re-collected or
the data set may be  determined as useable  only to a
limited extent.  Therefore, both advance planning in
identifying critical samples and the use of alternative
sampling  procedures are  necessary 'o e: sure
completeness  of a  data set for the  baseline risk
assessment.

Comparability.   Comparability expresses the
confidence with which data are considered to be
equivalent. Combined data sets arc used regularly to
develop qualitative estimates of risk.  The ability to
compare data sets is particularly critical when a set of
data for a specific parameter is  applied to a particular
concentration of concern.
Comparability for sampling primarily involves sampling
designs and time periods. Typical questions to consider
 in determining sampling comparability  include:

   •  Was  the same approach to sampling taken in two
      sampling designs?

   •  Was  the sampling performed at the same time of
      year  and under similar physical conditions  in the
      individual events?

   •  Wore samples filtered or unfillered?

   •  Were sjimples preserved?

 Typical questions to consider in determining analytical
 comparability include:

   •  Were different analytical methodologies used?

   •  Were detection limits (he same or at least similar?
  •  Were different laboratories used?

  •  Were the units of measure the same?

  •  Were sample preparation procedures the same?

Use routine available methods and consistent units of
measure when data collection will span several different
sampling events and laboratories,  to  increase the
likelihood that analytical results will be comparable.
For field analyses confirmed by laboratory analyses,
careful attention must be taken to ensure that the data
from field and  fixed laboratories are comparable or
equivalent (see Section 3.2.9).  When precision and
accuracy are known, the data sets can be compared with
confidence. Planning ahead for comparable sampling
designs, methods, quality control, and documentation
will aid the risk assessor in combining data sets for each
exposure pathway.

Representativeness.    For  risk  assessment,
representativeness is the extent to which data define the
true risk to human health and the environment. Samples
must be collected to reflect the site's characteristics and
sample analyses must represent the properties of the
field sample.  The  homogeneity of the sample, use of
appropriate handling, storage, preservation procedures,
and the detection of any artifacts of laboratory analyses,
such as blank contamination, arc particularly important.
For  risk assessment, sampling  and analyses must
adequately represent each exposure area or the definition
of an exposure boundary.
Representativeness can be maximized by ensuring that
sampling locations are selected properly, potential hot
spots are addressed, and a sufficient number of samples
are collected over a specified lime span.  The SAP
should describe sampling techniques and the rationale
used to select sampling locations.

Precision.   Precision is a quantitative measure  of
variability, comparing results  for site samples to the
mean, and is usually reported as acocfficicnt of variation
or a standard deviation of the arithmetic mean. Results
of QC samples are  used to calculate the precision of the
analytical or sampling process. Measurement error is a
combination of sample collection and analytical factors.
Field duplicate samples help to clarify  the distinction
 between uncertainty from sampling techniques and
 uncertainty from  analytical  variability. Analytical
 variability can be measured  through the analysis of
 laboratory duplicates or through multiple analyses of
 performance evaluation samples.  If analytical results
 are reported near a concentration of concern, the standard
 deviation or coefficient of variation can be incorporated
 in  standard  statistical evaluations  to  determine  the
 confidence level of the reported data. A statistician or
                                                     33

-------
achcmist should be consulted tomake this determination.
Total variability must be evaluated to assess the precision
of data used to define parameters in risk assessment.
Accuracy. Accuracy is a measure of the closeness of a
reported concentration to the true value. This measure
is usually expressed as bias (high or low) and determined
by calculating percent recovery from spiked samples.
The risk assessor should know the required  level of
certainty for the end use of the data, expressed as DQOs,
when reviewing accuracy information. When results
are reported  at or near a  concentration of concern,
accuracy information is critical.
Accuracy  of identification may be affected by sample
contamination introduced in the field, during shipping,
or at the laboratory. Field and trip blanks should be used
during the  RI to identify contamination and the associated
bias related to sample collection or shipment. Method
blanks, audit samples, and calibration check standards
should be used to monitor laboratory contamination.
Accuracy information may be-of less importance if the
precision  (bias) is known.

3.1.5  Data Review

This  section discusses the importance of alternative
levels of data review to the risk assessment.  The two
major effects of data review on data useability are:
  •  The timeliness of the data review and

  •  The level and depth of review (e.g., entire site,
     specific  sample focus, specific analyte focus,
     amount of QC data assessed).

A tiered approach involving combinations of data review
alternatives is recommended so that the risk assessor
can use preliminary data before extensive review. The
RPM, in conjunction with the risk assessor and the
project chemist, must reach a consensus on the level and
depth of data review to be performed for each data
source, to balance useability of data and resource
constraints. Exhibit 17 summarizes the characteristics
and uses of different levels of data review.

Timing of review. Plans for the timing of the data
review should be made prior to data collection and
analysis. The risk assessor uses  preliminary data in a
qualitative manner to identify compounds for loxicity
studies and, initially, to ascertain trends in concentrations
and distributions of the analy tes of concern, to plan for
additional sampling, and to request additional analyses.
Using data as they become available will usually reduce
the  time needed to  complete the risk  assessment.
However, all data must receive a minimum level of
review before use in the quantitative aspects of risk
assessment.  Iterations on data review is resource
intensive; if they are used, they should be planned
carefully as part of a structured process.
      EXHIBIT 17.  ALTERNATIVE LEVELS OF REVIEW OF ANALYTICAL DATA
Level of
Review
None
Full
Partial
Automated
Sample*
Initial
Initial samples
analyzed for broad
spectrum components
Analytes
All
All
Critical samples for all analytes
or
All samples for critical analytes
All
All
Parameters
Analytical results
All analytical results,
QC, and raw data
Selected analytical
results, QC, or raw
data
Parameters available
to the automated
system. No raw data
are evaluated.
Potential Uses
Qualitatively identify risk
assessment analytes.
Modify SAP.
Quantitatively perform risk
assessment. Modify SAP.
Modify review process.
Improve timeliness,
overall efficiency,
save resources.
Focus on chemicals
of potential concern.
Improve timeliness,
consistency, cost
effectiveness. If data are
electronically transferred to
a database, eliminates
transcription errors.
                                                                                                   Zt-OOJ-017

-------
                     *•  To expedite  the risk assessment.
                     preliminary data should be provided to the
                     risk assessor as soon as they are available.

                 Level and depth of review.  The RPM may select
                 different levels of data review, in consultation with the
                 risk assessor or other data users and the project chemist.
                 All data must have a minimum level of review.  Data
                 review levels can range from all site samples with all
                 reported data to specific key analytes and samples and
                 may be specified in EPA Regional policies. Careful
                 consideration is required in selecting a level of review
                 that is consistent with data quality requirements.
                 A full data review minimizes false positives, false
                 negatives, calculation errors, and transcription errors.
                 "Non-detect" results must be reviewed to avoid "false
                 negative" conclusions. Partial review should be utilized
                 only  after  broad spectrum analysis  results  have
                 undergone full review; it may be useful after chemicals
                 of potential concern have been  identified.  A flexible
                 approach to data review alternatives allows the RPM to
                 balance time and resource constraints.

                 Depth of data review refers to which evaluation criteria
                 are selected, ranging from generalized criteria that may
                 affect an entire data set (e.g., holding time) to analyte-
spccific criteria that may affect only a portion of results
from one sample (e.g., recovery of a surrogate spike for
organic* or analyle spike recovery for inorganics). The
RPM decides the depth of review for each data source,
to provide a balance between use-ability of data  and
resource constraints. Chemicals of potential concern in
the quantitative risk assessment should not be eliminated
from concern without a full data review.

Automated data review systems. Automated data
review systems can be used to assess all samples and
analytes for which (here are computer-readable data in
the format required by the automated system. The depth
of review depends on both the data and the assessment
system. The  primary advantages  of automated data
review systems for the risk assessor are timeliness, the
elimination of transcription errors that can be introduced
during manual review processes, and computer-readable
output which usually includes results and qualifiers.
This information can be transferred to computer-assisted
risk assessment and exposure modeling systems. Exhibit
18 provides a list of software that aid data review and
evaluation.
                                           EXHIBIT 18. AUTOMATED SYSTEMS*
                                                 TO SUPPORT DATA REVIEW
System
CADRE
Computer Assisted Data
Review and Evaluation
•DATA
Electronic Data Transfer
and Validation System
EPA Contact
Gary Robertson
Quality Assurance Div.
USEPA, EMSL-LV
(702)798-2215
William Coakley
USEPA, Emergency
Response Team
(908) 906-0921
Description
An automated evaluation system
that accepts files from CLP format
disk delivery or mainframe transfer
and assesses data based on
National Functional Guidelines tor
Organic (or Inorganic) Data Review
(EPA 1991 e, EPA 1988e) (default
criteria). System accepts manual
entry of other data sets, and rules for
evaluation can be user-defined to
reflect specific information needs.
(Inorganic system is in development.)
An automated review system
developed to assist In rapid
evaluation of data in emergency
response. May be applicable for both
CLP and non-CLP data. System
combines DQOs, pre-established
site ipeciticalions, QC criteria, and
sample collection data with laboratory!
results to determine useability. 1
' Both systems operate on an IBM-compatible PC AT with a minimum ol 640K RAM. 1
A fixed disk li recommended. 1
                                                                                                      11-001411
                                                                    35
1

-------
3.1.6  Reports from Sampling and
        Analysis to the Risk Assessor

Preliminary data reports assist the risk assessor in
identifying sampling or analytical problems early enough
so that corrective actions can be taken during data
collection, before sampling or analysis resources are
exhausted. The risk assessor should request preliminary
data during RI planning and formalize the request in the
SAP. The use of such information may reduce the
overall time required for the risk assessment and increase
the quality of a quantitative risk assessment.
Exhibit 19 lists the final data and documentation needed
to support risk assessment, and rates the importance of
each item. Data are most useable when reported in a
readable  format and accompanied by  additional,
clarifying information. Regional policy usually defines
report structures which specify the format for manual
summaries, for machine-readable data(where required),
and for summary tables from data re view. TheRPMcan
request the data reviewers to provide a data summary
table listing sample results, sample quantitation limits,
and qualifiers on diskette for downloading into Risk*
Assistant (an automated tool to support risk assessment),
                                                     spreadsheets, or other software programs that the risk
                   EXHIBIT 19.  DATA AND DOCUMENTATION NEEDED
                                   FOR RISK ASSESSMENT
Data and Documentation
• Site description with a detailed map indicating site location, showing
the site relative to surrounding structures, terrain features, population or
receptors, indicating air and water ('ow, -nd describing the operative Industrial
process if appropriate.
• Site map with sample locations (including soil depths) Identified.
• Description of sampling design and procedures Including rationale.
• Description of analytical method used and detection limits Including
SQLs and detection limits for non-detect data.
• Results given on a per-sample basis, qualified for analytical limitations
and error, and accompanied by SQLs. Estimated quantities of
compounds/tentatlvely identified compounds.
• Field conditions and physical parameter data as appropriate for the media
Involved in the exposure assessment.
• Narrative explanation of qualified data on an analyte and sample basis,
Indicating direction of bias.
• QC data results for audits, blanks, replicates and spikes from the field and
laboratory.
• Definitions and descriptions of flagged data.
• Hardcopy or diskette results.
• Raw data (Instrument output, chromatograms. spectra).
• Definitions of technical jargon used In narratives.
Importance
Critical
Critical
Critical
Critical
Critical
Critical
High
High
High
Medium
High
Low
KEY Critical • Essential to the useabfflty of data for risk assessment
High » Should be addressed In planning for risk assessment
Medium • Primarily Impacts how data are qualified In risk assessment
Low - HMlttJeeftect on useabtlity of data tor risk assessment

-------
assessor may use.  An example of a recommended
report format for tabular results appears in Appendix I.

The data reviewer should provide a narrative summary,
which is comprehensible to a nonchemist, describing
specific  sampling or analytical problems,  data
qualification flags, detection limit definitions,  and
interpretation of QC data.  This summary must always
be followed and supported by a detailed commentary
that explicitly addresses each item from the narrative on
-. technical basis. Theexplanationfordataqualification
i, he commentary facilitates data use. If a nontechnical
narrative is unavailable, the risk  assessor must (at a
minimum) be provided widiexplanationsof qualification
Hags, detection limits, and interpretation of QC  data
(see Appendices I, V and VI for examples). A chemist
f niliar with the site can be requested to interpret the
   alyticalreview with site-specific information, such as
physical site conditions that affect sample results.


3.2 PRELIMINARY SAMPLING AND
     ANALYTICAL ISSUES

This guidance cannot encompass sampling design in the
assessment of environmental sampling and analysis
 procedures; however, this section does sketch a
 framework for these activities.  It discusses key issues
 for determining the potential impact of sampling and
 analysis procedures on data useability for risk assessment
 and for identifying situations that require statistical or
 methodological support.  The sampling discussion
 primarily focuses on soil issues, bu t some generalizations
 can be made to other media such as sediment or
 ground water. Rulesof thumb, reference tables, statistical
 formats  and  checklists support the  statistical
 understanding  and sophistication of RPMs and risk
 assessors. A Sampling Design Selection Worksheet, a
 Soil Depth Sampling Worksheet, andaMeihodSelection
 Worksheet are tools, presented with step-by-step
 instructions in Chapter 4, to focus planning efforts.

 Sampling issues.  Resolving statistical and  non-
 statistical sampling issues provides the risk assessor,
 project chemist,  and QA personnel with a basis for
 identifying sampling design and data collection
 problems, interpreting the significance of analytical
 error, and selecting methods based on the expected
 contribution of sampling and analytical components to
 total measurement error.  Comprehensive discussions
 of environmental sampling procedures are given in
 Principles of Environmental Sampling (Keith 1987),
 Environmental Sampling and Analysis (Keith 19900).
 Methods for Evaluating  the Attainment of  Cleanup
 Standards (EPA 1989e), and the Soil Sampling Quality
 Assurance User's Guide (EPA 19890.
Several assumptions concerning sampling and associated
statistical procedures have been made to simplify the
discussion in this section:

  •  The RPM and risk assessor are familiar with basic
     environmental sampling and statistical terms and
     logic and have access to a statistician.

  •  Sampling designs are mainly based on stratified
     random or systematic random sampling (grid), or
     variations thereof. Systematic sampling requires
     special  variance  calculations  for  estimating
     statistical performance parameters such as power
     and confidence level; these calculations are not
     provided in this guidance.

   •  Statisticians are  consulted for  any significant
     problems or issues not covered in this guidance.

   •  Superfund contaminant concentrations for a site
     generally  fit  a  log-normal  distribution.
     Measurements of variability are generally given
     in log-transformed units. Overviews of statistical
     methodology include Gilbert (1987) and Koch
     and Link (1971).  Parametric tests in transformed
     units (Aitchison and Brown 1957) have logarithmic
     forms (Seichel  1956).   Graphical methods of
     determining re-transformed means and their 95%
     confidence levels are available (Krige 1978).

   • Quality assurance procedures for sampling and
     analysis are not separate,  even though the
     discussion addresses them separately.

 Exhibit 20 summarizes the importance of each of the
 preliminary  sampling planning issues  to the risk
 assessment, proposes  planning actions to reduce or
 eliminate their effect on data useability, and refers the
 reader to further discussion in the text  Information
 relevant to preliminary  sampling planning can be
 obtained by collecting site maps, photographs and other
 historical and  current document;,  which depict
 production, buildings, sewage and storm drains, transport
 corridors, dump sites, loading zones, and storage areas.
 A reliable and current base map is particularly important.

 Data adequacy. All data users should clearly state the
 level of data adequacy they desire. These statements,
 and the resources that will be committed, should be
 incorporated into the sampling plan objectives. If an
 appropriate levelof uncertainty cannot be determined at
 this stage, an initial goal should be agreed on  for the
 final level of reliability, which may be revised during
 the iterative sampling process. Since each site is unique,
 it may be extremely difficult to attain a given level of
 data adequacy.  An iterative sampling program may
                                                    37

-------
         EXHIBIT 20.  IMPORTANCE OF SAMPLING ISSUES IN RISK ASSESSMENT
                 Issue
            ImporUnct
      Suggested Action
            Chemical* ot Potential
            Concern
            3.2.1)
Chemicals have different falos ol
occurrence and coefficients ol variation.
This Impacts Ihe probability ol false
negatives and reduces conlldonce limits lor
estimates ol concentration.
Increase Ihe number ol samples lor
chemicals wlh low occurrence and/or
high coefficients ol variation.
            Samplng and
            Analytical VariablMy
            versus Measurement
            Error (3.2.5)
Sampling variably can exceed
measurement error by a (actor ol three to
lour (EPA 1969c).
                               Sampling variability increases uncertainty
                               or variability, measurement error
                               Increases bias.
Reduce sampling varlabUy by taking
more samples (using less expensive
methods). This allows more samples
to be analyzed.

Use QC samples to estimate and
control bias. Prepare SOPs lor
handling all field equipment.
            Media Variability
            (3.2.5)
 Samplng problems vary widely by media as
 do variability and bias.
 Design media-specHIc tamping
 approaches.
            Sample Preparation
            and Sample
            Preservation
            (3.2.6)
 Contamination can be Introduced during
 sample preparation, producing labe
 posthres. Filtering may remove
 contaminants sorbed on particles.
 Use blanks at sources ol potential
 contamination.  Cotect lilered and
 unlilered samples.
             Identification d
             Exposure Pathways
             (3.2.7)
 No) all samples taken In a site
 characterization are uselul lor risk
 assessment. Often only a lew samples have
 been taken In the area ol Interest.
 Specifically address exposure
 pathways h samplng designs. Risk
 assessors should participate in
 scoping meeting.
             Use ol Judgmental or
             Purposive Samplng
             Design
             (3.2.8)
 Statistical sampling designs may be costly
 and do not take advantage ol known areas
 d contamination.
 Use Judgmental sampling to examine
 known contaminated areas, then use
 an unbiased method to charade rue
 exposure.
allow a realistic appraisal of the variability present at the
site; a  phased investigation may be warranted, with an
increase in data adequacy at each phase.
Natural variation. It is important to realize that natural
variation (environmental heterogeneity) in both soil
and water systems may be so great that variation due to
field sampling is significantly greater than  that due to
laboratory analysis. For example, laboratory sample-
sample precision is commonly of the order of less than
 1 %. whereas soil sample-sample precision is commonly
between 30% to 40%. Sampling variation is Influenced
by the homogeneity  of material being sampled, the
number of samples, collection procedures, and Ute size
of individual samples.
 Uncertainty  in sampling measurements is additive.
 Exhibit 21 lists the components of sampling variability
 and measurement error. The final error associated with
 mi estimate is  the sum of the errors associated with
 natural variation (Intrinsic randomness, mlcrosimeiure,
 macrostructure), plus sampling error, plus laboratory
                             measurement error.   Poor sampling techniques  can
                             swmnp the natural phenomenon that is being evaluated.
                             Therefore, sampling options must be fully reviewed and
                             the  probable  uncertainty from sampling  must be
                             acceptable.
                             Initial survey sampling plan. A preliminary sampling
                             plan should be chosen that provides a basis for evaluation
                             of overall  sampling goals,  sampling techniques,
                              feasibility, and statistical analysis techniques. General
                              categories of sampling plans include simple random,
                              stratified  random, systematic, judgmental/purposive.
                              and spatial systematic. The features of these different
                              plans are discussed in more detail in Chapter 4.
                              Statistical analysis of the survey data allows evaluation
                              of how well the sampling program is doing. Depending
                              on the contaminant, current technology may allow on-
                              slte "laboratory" analysis of the samples using portable
                              microcomputers and telecommunications.   On-sllc
                              statistical analysis Is also possible. On-slte analysis
                              reduces project completion time and costs.  In a truly

-------
        EXHIBIT 21. SAMPLING
            VARIABILITY AND
        MEASUREMENT ERROR
        Sampling variability: The variation
        between true sample values that is a
        function of the spatial variation in the
        pollutant concentrations.

        Measurement error The variation
        resulting from differences between
        true sample values and reported
        values. Measurement error is a
        function of uncertainty due to the
        following:
         • Sample collection variation
         • Sample preparation/handling/
          preservation/storage variation
         • Analytical variation
         • Data processing variation
iterative sampling campaign, on-site statistical analysis
can guide the sampling teams, maximizing information
capture and minimizing time-related costs.

Analytical  issues.   The  following assumptions
concerning analytical procedures have been made in
this section:
   •  The RPM and the risk assessor are familiar with
     standard analytical chemical procedures.
     Reference books on environmental issues in
     analytical chemistry  are available  and can be
     consulted (ASTM 1979. Manahan 1975. Dragun
     1988. Baudo. el. al., eds. 1990. Taylor 1987).

   •  Chemists are available and will be consulted for
     any significant problems or situations not covered
     in this guidance.

   •  AnalyticalQAproceduresareusedinconjunction
     with and affect sampling QA procedures, even
     though  the discussion treats these procedures
     separately.

Exhibit22 summarizes the importance of each analytical
issue to risk assessment, lists suggested actions during
the planning  process,  and refers the reader to further
discussion in the text.  Each issue is discussed in terms
of its effect on data quality for risk assessment, and how
to anticipate and plan for potential problems. The RPM
should also consult the project chemist to determine the
appropriate sample  volumes or weights  required for
different types of analysis.

Biota  sampling and analytical Issues.  The type of
assessment (e.g., human health or ecological) determines
the type of samples to  be collected.  An ecological
assessment may require analysis of the whole body or of
a specific organ  system of a target species (because
organic, and some inorganic, chemicals of concern are
often concentrated in tissues with high lipid contents).
Human health risk assessment usually concentrates on
edible portions.
Typical sampling considerations  for biota include
specifying the species to be sampled, sampling locations,
tissue to be analyzed, number  of individuals to be
sampled, and the method of analysis of the chemical of
concern.  Biota analyses should  include a method
validation that incorporates  tissues or plant analyte
spikes,  and any available performance  evaluation
materials.  The  purpose of  spiking is to determine
whether the analytes are recoverable from the matrix or
clean-up steps hinder detection of the analyte.

Spiking and duplicate information can be used to assess
method precision and accuracy. The primary source of
performance e valuation materials is the National B ureau
of Standards  repository.  Samples and performance
evaluation materials should be matched by matrix
(species and whole/edible portions).

Volatile analytes are very difficult to measure in biota.
Samples should be stored on dry  ice immediately after
collection. Fat and cholesterol can also block columns
and  impede chromatography for  base/neutral/acid
extractable  tissue analysis.   Gel  permeation
chromatography procedures may only be marginally
effective in clean up, and the lipids present may retain
analytes of concern, thereby reducing recoveries. Plant
matrices are often difficult to digest, and a variety of
digestion procedures  using hydrogen peroxide or
phosphoric acid may be warranted. Tissues for organic
analysis should be wrapped in aluminum foil for
shipment to the laboratory, and tissues for metals analysis
should be wrapped in plastic film. All tissues should be
sent frozen on dry ice.

Air  sampling and analysis issues.  Air sampling
procedures should account for wind speed and direction
as well as seasonal and daily fluctuations; they should
also account for the influence of these factors on the
exposed population (e.g., the largest population may be
potentially exposed in the evening when the wind speed
may be least). The definition of detection limits is very
important for air analyses.  For example, the same
concentration will appear very different if expressed on
a weight/volume basis than on a volume/volume basis.
Sampling strategics may need to distinguish between
paniculate and gaseous forms of chemicals of concern.
It is important to collect media blanks to determine the
type and amount of contamination that may be found.
 Blanks should also be provided to the laboratory for
 spiking to determine analytical precision and accuracy.
                                                    39

-------
            EXHIBIT 22.  IMPORTANCE OF ANALYTICAL ISSUES
                               IN RISK ASSESSMENT
Analytical Issue
Chemicals of
Potential Concern
(3.2.1)
Tentatively Identified
Compounds
(3.2.2)
Identification and
Quantitation
(3.2.3)
Detection Limits
(3.2.4)
Media Variability
(3.2.5)
Sample Preparation
(3.2.6)
Field Analyses versus
Fixed Laboratory Analyse*
(3.2.9)
Laboratory Performance
Problems
(3.2.10)
Importance
Chemicals of potential
lexicological significance may be
omitted.
Identification and quantitation do
not have high confidence.
False negatives may occur when
analytes are present near the
MDL
Significant risk may result at
concentrations lower than
measurable.
Variability and bias may be
introduced to analytical
measurements.
Variability and bias may be
Introduced to analytical
measurements.
Tradeoffs required with regard to
speed, precision, accuracy.
personnel requirements,
identification, quantitation and
detection limits.
Quality of data may be
compromised.
Suggested Action
Examine existing data and site history
for industry-specific wastes to
determine analytes for measurement
Perform broad spectrum analysis.
Be prepared to request further
analyses if potentially toxic
compounds are discovered during
screening. Compare results from
multiple samplings or historical data.
Use technique with definitive
identification (e.g., GC-MS).
Alternatively, use technique with
definitive identification first, followed
by another technique (e.g., GC) to
achieve lower quantitation limits.
Review available methods for
appropriate detection limit.
Use environmental samples as QC
samples to determine recovery and
reproducibility in the sample media.
Select analytical methods based on
sample medium and strengths of the
sample preparation technique.
Consider options and set priorities.
Select experienced laboratory and
maintain communication.
The sample medium should be checked to ensure that
recovery rates are documented.

3.2.1   Chemicals of Potential Concern

Chemicals of potential concern are chemicals that may
be hazardous to human health or the environment and
are identified at thcsite,initiallyfrom historical sources.
Chemicals identified at Superfund sites have varying
rates of occurrence, average concentrations, and
coefficients of variation. Thesedifferencesareafunction
of fate and transport properties, occurrence in different
media, and interactions with other chemicals, in addition
to use and disposal practices. Information on frequency
of occurrence and coefficient of variation determines
the number  of  samples required to adequately
characterize exposure pathways and is essential in
designing  sampling plans.  Low frequencies of
occurrence and high coefficients of variation mean that
more samples will be required to characterize the
exposure pathwaysofinteresL Potential false negatives
                                              40

-------
occur as variability increases and occurrence rates
decrease. From an ecological standpoint, chemicals of
potential concern may be different from those for human
health concerns. For example, copper is an analyte of
high concern from an ecological perspective, but of low
concern from a human health perspective. In addition,
if water quality criteria  are used as  lexicological
thresholds, it should be determined whether the criteria
are based on ecological or human health effects.

    «• To protect human health, place a higher
    priority on preventing false negatives in
    sampling and analysis than on preventing
    false positives.

Data are available for volatiles, extractable organics,
pesticides/PCBs, tentatively identified organic
compounds, and metals (see Appendix 10, for aqueous
and soil/sediment matrices, and releases from industries
known to produce waste commonly found at Supcrfund
sites. Data from CLP Superfund sites are also available
for calculating site-specific coefficients of variation.
Exhibit 23 indicates the occurrence rates and coefficients
of variation for selected chemicals of potential concern
to risk assessors. Many other chemicals (which are not
of concern) may be present without affecting the level
of risk to the exposed population.

    f Use preliminary data to identify chemicals
     of potential concern and to determine any
     need to modify the  sampling or analytical
     design.

The need for  risk assessment indicates that there is
already some knowledge  of contamination at the site.
 Based on available lexicological and site data, the risk
assessor can recommend target chemicals (or chemical
classes) for analysis and desired detection limits. For
example, explosive chemicals are likely to be present at
a former munitions site.  Exhibit 24 presents data on
 munitions compounds, such as feasible detection limits
and health advisory limits.

 Information on industry-specificanalytes is summarized
 in Exhibit 25 and detailed in Appendix II. If historical
 data are incomplete, a broad spectrum analysis should
 be performed on selected  samples from each sampling
 location to provide necessary scoping information.

 The RPM or risk assessor should inform the planning
 team about chemicals of  potential concern at the site,
 exposure pathways, if known, concentrationsof concern,
 and other pertinent information, particularly any
 requirement to distinguish specific states of the chemicals
 of potential concern. Some oxidation states of metals
 (e.g., chromium) are more easily absorbed or are more
 toxic than others, and organically substituted metals
such as mercury arc more toxic than their elemental
slates.  If these concerns are important, analyses that
determine  metal specification rather than elemental
analyses should be performed, if available.  Similarly,
for organic compounds, such as tetrachloroethane,
degradation products or metabolites may be more toxic
than the parent compounds.  In this case, sampling
procedures and analytical methods should include the
parent compound, degradation products, and metabolites
of chemicals of potential concern.

3.2.2  Tentatively Identified
        Compounds

Gas chromatography-mass speclrometry (GC-MS)
analyses categorize organic compounds in two ways.
Target compounds are those compounds for which the
GC-MS instrument  has been specifically calibrated
using authentic chemical standards. A target compound
in an environmental sample is identified by matching its
mass spectrum and  relative retention time (RRT) to
those  obtained  for the authentic  standard during
calibration.   Quantitntion of a  target  compound is
achieved by comparison of its chromntographie peak
area to that of an internal standard compound, normalized
to the relative response factor (RRF) which is the ratio
of the peak areas of the authentic chemical standard and
the internal standard measured during calibration.

    •• Specific analysis for compounds ident-
     ified during library search can be requested.

Tentatively Identified Compounds (TlCs) are any other
compounds which are reported in the sample analysis,
but for which the GC-MS instrument was not specifically
calibrated.  A TIC is  identified by taking its mass
spectrum from the environmental sample, and comparing
it to  a computerized library of mass spectra.
Computerized comparison routines score the various
library spectra for their similarity to the TIC and rank
the spectra most similar to the TIC's spectrum. If the
TIC is reported as a specific compound, it is usually
reported to be one of the compounds whose spectra
were retrieved in the library search. Quantitalion of a
TIC is less accurate than for target compounds, because
the true RRF is not known (since no calibration for this
specific compound was performed). The RRF is assumed
to be  1.0; whereas, measured RRFs below 0.05 and
above 10.0 arc known.

Confidence in the idenlificatlonof aTICcan be increased
 in several ways. The main steps in identifying and
 quantitating TIC data are summarized in  Exhibit 26.
 An analytical chemist trained in the interpretation  of
 mass spectra and chromalograms can review TIC data
                                                   41

-------
                 EXHIBIT 23.  MEDIAN COEFFICIENT OF VARIATION FOR
                          CHEMICALS OF POTENTIAL CONCERN 1
Chemical ol
 'otential Concern
                 Number of Sites                    Number of Sites
Soil/Sediment     at Which Chemical      Water       at Which Chemical
Median %CV2       was detected3     Median %CV2     was detected3
  Chtoromethane                            16.7            61
  Trichloromethana/Chloroform                 53.9            392
  Telrachtoromelhana'Carbontetrachloride        15.4            38
  1,2-Dfcfibroelhane                         17.6            64
  Tetrachtoroettiana                          17.0            56
  Vinyl chloride                             11.0            55
  Tetrachloroelhene                          24.5            392
  Dichloropropane                           19.0            29

  Isophorone                                0.7            74
  Bis (2-chtoroelhyl) efrer                      0.5             10
  1.4-Dichlorobenzene                        0.9            120
  Bis (2-ethylhexyl) phthalate                   0.7            1197
  Benzo{a) pyrene                            0.5            1058
  Styrene                                  16.9            117
  N-nitrosodiphenylamine                      0.5            U2

  DDE                                     4.5            329
  DDT                                     2.9            521
  Dieldrin                                   4.4            274
  Heptachlor                                4.8            249
  Gamma-BHC (lindane)                      6.3            142
  PCB1260                                 0.21          251

  Arsenic                                   40.3           1098
  Beryllium                                271.3           1091
   Cadmium                                134.6           1096
   Chromium                                11.9           1098
   Mercury                                1032.3          1098
   Lead(Pb)                                10.9           1098
   1 List ol chemicals of potential concern Is derived (rom health-based levels and frequency of occurenoe at Super fund
     sites isted hi the CLP Statistical Database. (Number ol sites for which data exist totals 8.900.)
   2 Median percent coefficient ol variation of anaryte concentrations.
   3 November 1988 lo present
                                                    42

-------
              EXHIBIT 24. MUNITIONS COMPOUNDS AND THEIR
                                  DETECTION LIMITS
 Health
Advisory     Acronym
Compound Name
                1
Detection Limit'
     (PPb)
            HMX           Octahydro-1,3,5.7-tetranitro-1,3,5,7-tetrazocine
            RDX           Hexahydro-1,3,5-trinitro-1,3.5-triazine
                            Nitrobenzene
            TNB            1,3,5-Trinitrobenzene
            DNB           1,3-Dinitrobenzene
            Tetryl           Methyl-2,4.6-trinitrophenylnitramine
            TNT            2.4,6-Trinitrotoluene
            2.4 DNT        2.4-Dinitrotoluene
            TAX            Hexahydro-1 -(N)-acetyl-3,5-dinitro-1,3,5-triazine
            SEX            Octahydro-1 -(N)-acetyl-3,5,7-trinitro-1,3.5,7-tetrazocine
            2.6 DNT        2.6-Dinitrotoluene
            2,4.5 TNT       2.4.5-Trinitrotoluene
            2 Am DNT       2-Amino-4.6-dinitrotoluene
            4 Am DNT       4-Amino-2.6-dinitrotoluene
            2.4 DAmNT     2,4-Diamino-6-nitrotoluene
            2,6 DAmNT     2,6-Diamino-4-nitrotoluene
            DIMP           Disopropyl-methylphosphonate
            TNG           Gylcerol trinitrate (Nitroglycerin)
                            Nitrocellulose
            DMMP         Dimethyl methylphosphonate
            NG            Nitroguanadine
                                              5.1
                                              4.2
                                              6.4
                                              5.9
                                              9.1
                                              4.4
                                              6.3
                                              2.3
                                              5.1
       Health advisory complete.
       Health advisory in preparation (1990).
  1
    Depending upon matrix and instrument conditions, these compounds may be chromatographable
    and may be tentatively identified as indicators of the presence of munitions during GO-MS library
    search procedures.
    Detection limits are provided where available. Specific compounds with complete health advisories
    are designated as target analytes with defined detection limits specified in a high performance liquid
    chromatographic method developed and provided by the U.S. Army Toxic and Hazardous
    Materials Agency.
                                                                                          2t •002-024
                                               43

-------
EXHIBIT 25. SUMMARY OF MOST FREQUENTLY OCCURRING
  CHEMICALS OF POTENTIAL CONCERN BY INDUSTRY*
Compound
Acetone
Aluminum
Ammonia
Ammonium Nitrate
Ammonium SuDate
Anthracene
Arsenic
Benzene
Biphenyl
Chlorine
Chlorobenwne
Chromium
Copper
Cydohexane
Dibenzoluran
Oichloromethane
Formaldehyde
Freon
Glycol Etheri
Hydrochloric Add
Lead
Manganeie
Methanot
Methyl Ethyl Ketone
Naphthalene
Nickel
Nitric Acid
Pentachkxophenol
Propytont
Sodium Suffate
Sodium Hydroxide
SuHuricAcW
Tfichloroethene
Toluene
Titanium TetrechkxkJe
Xylen*
t.t.l-trichkxoethan*
Induttry
1




X












X


X
X
X






X
X
X
X



X
a
X

X
X





X



X









X


X
X

X

X





s


X







X




X



X


X






X
X


X
X
X

4











X



X



X





X
X


X
X
X
X



X
s


X


X
X

X


X
X

X

X







X


X









•


X

O(






X






X




X





X
X
X

X

X

7

X
X




X















X




X
X
X
X

X

X

KEY 4 .Electroplating I
1 - Baflery Recycling S • Wood PrtswvattvM •
{•Munfions/Exploiiyn C • leather Twrtng •
3. PeUldde Manufacturing 7 • Petroleum (Wining |
'SummarlMd horn Appendix 11. |
                     44

-------
I
   EXHIBIT 26.  STEPS IN THE
ASSESSMENT OF TENTATIVELY
    IDENTIFIED COMPOUNDS
                   Idtntiflcatlon   •   GO-MS analysis Indicates the
                                    presence of a tentatively
                                    Identified compound.

                                •   Incorporate retention
                                    time/retention Index matching
                                    and use physical
                                    characteristics (boiling point
                                    or vapor pressure) to
                                    determine It Identification Is
                                    reasonable.

                                •   Examine historical data and
                                    Industry-specific compound
                                    lists.

                                •   Reanalyze sample with an
                                    authentic standard.

                   Quantitatlon   •   Assess  known analytical
                                    response characteristics for
                                    similar compounds or similar
                                    compound classes.

                                •   Determine response
                                    characteristics by analysis of
                                    an authentic standard.
                mass spectra and chromatograms can review TIC data
                and eliminate many false positive identifications. The
                use of retention indices or relative retention times can
                confumTICs identified by the GC-MScomputer(Eckel,
                et.al. 1989).  Examination of historical data, industry-
                specific compound lists, compound identifications from
                iterative sampling episodes, and analyses performed by
                different laboratories may also increase confidence in
                the identification of a TIC. The final identification step
                is to reanalyze the sample after calibrating the GC-MS
                instrument with an authentic standard of the compound
                that the TIC is believed to be.

                If toxic compounds are identified as TICs by this type of
                broad spectrum analysis the RPM or risk  assessor
                should request further analyses to positively identify the
                compound and to accurately quamitatc it.  The risk
                assessor or RPM should discuss data requirements with
                an  analytical chemist to determine the  appropriate
                analytical method.

                Many compounds that appear as TICs during broad
                spectrum analyses belong  to compound  classes.
                Example! of compound classes are saturated aliphatic
                hydrocarbons and polycyclic aromatic hydrocarbons
(PAHs).  The risk assessor may be able to make a
preliminary judgment of toxicity at the compound class
level  without a definitive identification of each
compound present.  For  example, in a sample
contaminated by gasoline, organics analysis  would
indicate a series of TICs as aliphatic hydrocarbons of
increasing size.  These may not be carcinogenic, and
more precise identification may not be required.  If a
similar sample were contaminated with coal tar, larger
hydrocarbons and a series of PAHs would be found
during the analysis. The aliphatic hydrocarbons are not
especially toxic, but the PAH compound class contains
carcinogens and are of greater concern.

3.2.3  Identification  and Quantitation

A risk assessor first confirms chemical identification,
and then determines the level of contamination. This
section summarizes the effects of detection limits and
sample contamination considerations on the confidence
in analyte identification and quantitation. Requirements
for confidence  are  specified in Exhibit  27.   When
anaiytes have concentrations of concern approaching
method  detection  limits,  the confidence in both
identification and quantitation  is low. This case is
illustrated in Exhibit 28.  In addition, confidence in
identifying and quantitating as representative of site


    EXHIBIT 27.  REQUIREMENTS FOR
    CONFIDENT IDENTIFICATION AND
                QUANTITATION
                                                    Identification   •   Analyte present above the IDL

                                                                 •   Organic - Retention lime and/or
                                                                     mass spectra matches authentic
                                                                     standards.

                                                                 •   Inorganic - Spectral absorptions
                                                                     compared to authentic
                                                                     standards.

                                                                 •   Knowledge of blank
                                                                     contamination (II any).

                                                    Quantitation   •   Instrument response known
                                                                     from analysis of an authentic
                                                                     standard.

                                                                 •   Detected concentration above
                                                                     the limit ol quantitation and
                                                                     within the limit of linearity
                                                                     (Instrument response not
                                                                     saturated).
                                                                   45

-------
              EXHIBIT 28. RELATIVE IMPACTS OF DETECTION LIMIT
             AND CONCENTRATION OF CONCERN: DATA PLANNING
                    Relative Position of Method
                    Detection Limit (MDL) and
                 Concentration of Concern (COC)
       Consequence
            Confidence MDL
             knits
                                                           Non-Detects and
                                                           Detects Useable
                                                             Possibility of
                                                          False Positives and
                                                           False Negatives
                                                           Non-Detects Not
                                                               Useable
                                                           Detects Useable
                                                           Possibility of False
                                                               Negatives
conditions is potentially diminished if the chemicals of
potential concern are present as contaminants from
laboratory or field procedures. This section identifies
analytes and cites situations in which this is most likely
to occur.

The first requirement of analysis is confidence in the
identification of chemicals of potential concern.
Identification means that the chemical was present in
the environmental sample above the detection limit.
Chemicals can be correctly  identified at  lower
concentrations than are suitable for accurate quantitntion.
If lower quantitation  limits are  required for risk
assessment purposes, a larger initial sample size may be
processed, or the sample extract may be concentrated to
a smaller final volume. I lowever, concentration of an
cx'.r.ict to a smaller volume, or increasing the sample
size, may saturate the instrument in the presence of
matrix interferences. The RPM should discuss these
issues with an analytical chemist to determine the best
approach. A further discussion of limits of quantitation
is presented in Section 3.2.4. and Appendix III.

To ensure maximum confidence in the identification of
an organic chemical contaminant, an  instrumental
technique, such as imus spectromctry, that provides
definitive results is necessary. Although alternative
techniques arc availably GC-MS determination is the
best available procedure for confident identification or
confirmation of volatile and extractable organic
chemicals of potential concern. The application of this
technique minimizes the risk of error in qualitative
identification and measures chemicals of potential
concern at environmental levels above the detection or
quantllation limits  listed in Appendix III.   In cases
where  the target detection limit  is too low to allow
                                                 46

-------
but more definitive, instrumental techniques can be
used.

The identification of inorganic chemicals is more certain.
A reported concentration determined by  atomic
absorption (AA) spcctroscopy or inductively coupled
plasma (ICP) atomic emission spcctroscopy is generally
considered evidence of presence at the designated level
reported, provided  there  is no interference.  If
interferences exist, the laboratory should try to
characterize the type of interferences (background,
spectral or chemical) and take the necessary steps to
correct them.

3.2.4   Detection and Quantitation
         Limits and  Range of Linearity

The following discussion is intended to provide the
RPM and risk assessor with an understanding of the
various ways that detection or quanlitation limits can be
reported. The term "detection limit" is frequently used
without  qualification.   However, there are several
methods for calculating detection limits. The RPM
should consult with the project chemist and the risk
assessor whenever analytical methods are to be selected,
     Common Detection and Quantitation Limits
     Instrument detection limit. The IDL includes
     only the instrument portion of detection, not
     sample preparation, concentration/dilution
     factors, or method-specific parameters.
     Method detection limit.  The MDL is the
     minimum amount of an  analyte that can be
     routinely identified using a specific method.
     The MDL can be calculated from the IDL by
     using sample size and concentration factors
     and assuming 100% analyte recovery.
     Sample quantitatlon limit. The SQL is the
     MDL adjusted to reflect sample-specific action
     such as dilution or use of a smaller sample
     aliquot for analysis due to matrix effects or the
     high concentration of some analytes.
     Contract required quantitatlon (detection)
     limit. The CRQL for organics and CRDL for
     inorganics are related to the SQL that has been
     shown through laboratory validation to be the
     lower limit for confident quantitation and to be
     routinely within the defined linear ranges of
     the required calibration procedures.
     Practical quantitation limit.  The PQL.
     defined in SW846 methods, is the lowest level
     that can be reliably achieved within specified
     limitsof precision and accuracy during routine
     laboratory operating conditions.
and specify the nature of the detection limits that must
be reported; it is the laboratory's responsibility to adhere
to tliis requirement. If no requirement has been specified,
then the laboratory should be requested to explicitly
describe the typos of  the detection limits it reports.
Detection limits can be calculated for the  instrument
used for measurement, for the analytical method, or as
a sample-specific quantitation limit.  The risk assessor
should request that the sample quantitation limit (SQL)
be reported whenever possible.  The term "detection
limit" should be considered generic unless the specific
type is defined. Exhibit 29 illustrates the relationship
between instrument response and the quantity of analyte
presented to the analytical system (i.e., a  calibration
curve).
    •• The closer the concentration of concern
    is to  the detection  limit,  the greater the
    possibility of false negatives  and false
    positives.

    *•  The wide range of chemical concen-
     trations in the environment may require
     multiple analyses or dilutions to  obtain
     useable data.   Request  results from all
     analyses.

The definitions that follow are intended to  provide the
RPM and risk assessor with an understanding of the
various methods for  calculating detection limits, the
terms used to describe specific detection limits, and the
limitations associated  with identification  and
quantitation of chemicals of potential  concern  at
concentrations near specified detection limits.
Understanding the different terms used  to describe
detection limits helps avoid reporting problems. Exhibit
 30 provides examples of calculations of the three most
 commonly reported types of detection limits.

     *•  Define the type of detection or  quanti-
     tation limit for reporting purposes; request
     the sample quantitation limit for risk
     assessment.

 lastrument detection limit. The instrument detection
 limit (IDL) includes only the instrument portion of
 detection, not sample preparation, concentration/dilution
 factors, or method-specific parameters.  The IDL is
 operationally defined as three times the standard
' deviation  of seven  replicate analyses at the lowest
 concentration that is statistically different from a blank.
 This represents 99% confidence that the signal identified
  is the result of the presence of the analyte. not random
  noise. The IDL is not the same as the method detection
  limit.  Use of the IDL should be avoided for risk
  assessment.
  Method detection limit. The method detection limit
                                                     47

-------
                      EXHIBIT 29. THE RELATIONSHIP OF INSTRUMENT
                      CALIBRATION CURVE AND ANALYTE DETECTION
                            Region of
                            Unknown Identification and
                            Quantitation
                                              Region of Known
                                                Quantitation
                                                                     Region of
                                                                    Less Certain
                                                                    Quantitation
                           2? Region
                           ^of Less
                           ^Certain
                           Zldentifi-
                                                  IDL  • Instrument Detection Limit
                                                  MOL - Method Detection Limit
                                                  LOQ - Limit of Quantitation
                                                  LOL - Limit of Linearity
                                                Concentration
                         IDL     MDL  LOQ
          LOL
Method detection limit The method detection limit
(MDL) is the minimum amount of an analyte that can be
routinely identified using a specific method. The MDL
can be calculated from the IDL by using sample size and
concentration factors and assuming 100% analyte
recovery. Thisestimateofdetectionlimitmaybebiased
low because recovery is frequently less than 100%.
MDLs are operationally determined as three times the
standard deviation of seven replicate spiked samples
run according to the complete method.  Since this
estimate includes sample  preparation effects, the
procedure  is more  accurate  than  reported  IDLs.
However, the evaluation is routinely completed on
reagentwater. AsarcsulLpotcntiallysignificantmalrix
interferences that decrease analyte recoveries are not
addressed.
The impact of an MDL on risk assessment is illustrated
in Exhibit 28. When planning to obtain analytical data,
the risk assessor knows the concentration of concern or
preliminary remediation goal. When the concentration
of concern of an analyte is greater than the MDL, to the
extent that the confidence limits of both the MDL and
concentration of concern do not  overlap, then both
"non-dftect" and "detect" results can be used with
confidence. There willbeapossibilityoffalsepositives
and false negatives if the confidence limits of the MDL
and  concentration of concern overlap.  When the
concentration of concern is sufficiently less than the
MDL that the confidence limits do not overlap, then
there is a strong possibility of false negatives and only
"detect" results are useable.
                                                 48

-------
          EXHIBIT 30.  EXAMPLE OF DETECTION LIMIT CALCULATION
     IDL « 3 x SO* of replicate injections

               Example:    100 ppb pentachlorophenol standard

                     II:    SO B 5 ppb

                 Then:    IDL « 3x5 ppb = 15 ppb

     MDL » 3 x SD of replicate analyses (extraction and injection)

               Example:    100 ppb pentachlorophenol spiked in sample producing average measured
                           concentration of 50 ppb (not all anatyte is recovered or measured)

                     If:    SD«18ppb

                 Then:    MDL = 3x 18 ppb = 54 ppb


     Incorporate calculation of MDL from IDL

      SQL « MDL corrected for sample parameters

               Example:    100 ppb pentachlorophenol with MDL of 57 ppb

                     If:     Dilution factor = 10  (sample is diluted due to matrix interference or high
                           concentrations of other analytes)

                 Then:    SQL » 10 x 57 ppb = 570 ppb

       SD B Standard Deviation
Sample quantltatlon limit. The SQL is the MDL
adjusted to reflect sample-specific action such as dilution
or use of smaller aliquot sizes than prescribed in the
method.  These adjustments may be due to matrix
effects or the high concentration of some analytes. The
SQL is the most useful limit for the risk assessor and
should always be requested.
For the same chemical, the SQL in one sample may be
higher than, lower than, or equal to SQL values for other
samples.  In addition, preparation or analytical
adjustments,  such as dilution of the sample for
quantitation of an extremely high level of one chemical,
could result in non-detects for other chemicals included
in the analysis, even though these chemicals may have
been present at trace quantities in the undiluted sample.
The risk assessor should request results of both original
and dilution analyses in this case.  Since the reported
SQLs lake into account sample characteristics, sample
preparation, and analytical adjustments, they arc the
most relevant quanlilation limits  for evaluating  non-
detected chemicals.

Contract required quantitation (detection) limit.
The CLP specifies a contract required quanlitation limit
                                                                                          21*402-090
(CRQL) for organics and a contract required detection
limit (CRDL) for inorganics. Each of these quantities is
related to the SQL that has been shown through laboratory
validation to be the lower limit for confident quanlitation
and to be routinely within the defined linear ranges of
the required calibration procedures.
The use of CRQLs and CRDLs attempts to maintain the
analytical  requirements within performance limits
(which are based upon laboratory variability using a
variety of instruments). CRQLs are typically two to five
times the reported MDLs and they generally correspond
to the  limit of quantitation.

Practical quantitation limit. The practical quantitation
limit (PQL), defined in SW846 methods, is the lowest
level  that can be reliably achieved within specified
limitsof precision and accuracy during routine laboratory
operating conditions.  It is important to note that the
SQL and PQL are not equivalent. Use of PQL values as
measures of quantitation  limits should be  avoided
wherever possible in risk assessment.
Other quantitation measurement!.  The  limit of
quantitalion (LOQ) is the level above which quantitative
                                                 49

-------
results may be obtained with a specified degree of
confidence.  At analyte concentrations close to, but
above the MDL. the uncertainty in quantitation  is
relatively high. Although the presence of the analyte is
accepted at 99% confidence, the reported quantity may
be in the range of ±30%.  Ten times the standard
deviation measured for instrument detection  is
recommended todemonstlatealevel at which confidence
is maximized (Borgman 1988).

The limit of linearity (LOL) is the point at or above the
upper end of the  calibration curve at which the
relationship between the quantity present and the
instrument response ceases to be linear (Taylor. 1987).
Instrument response usually decreases at the LOL, and
the concentration reported is  less than  the amount
actually present in  the sample  because of instrument
saturation. Dilution is necessary to analyze samples in
which analyte concentrations are above the  LOQ.
However, dilutions correspondingly increase SQLs.
Data should be requested from both diluted and undiluted
analyses.

3.2.5   Sampling and Analytical
         Variability Versus
         Measurement Error

 Sampling and analytical variability and measurement
 error are two key concepts in data collection. Each is
 discussed in the context of evaluating strategies for the
 collection and analysis of both site and background
 samples.

 Exhibit 21 defines sampling variability and measurement
 error. Most SAPs are a necessary compromise between
 cost and confidence level. Basically, two types of
 decisions must be made in planning:

   •  What statistical performance b necessary to
      produce the quality of data appropriate to meet the
      risk assessor's sampling variability performance
      objectives and

    • What types  and numbers of QC samples are
      required to detect and estimate measurement error.

      *•  When contaminant levels In a medium
      vary widely, increase the number of samples
      or stratify the medium to reduce variability.

  Sampling plans attempt to estimate and minimize both
  sampling variability and measurement error. Sampling
  variability affects the degree of confidence and power
  the risk assessorcanexpectfrom the results. Confidence
  is the ability to detect a false positive hypothesis, and
  power is the ability to detect a false negative. Power is
  more important for risk assessment. An estimate of the
sampling variability that is a function of the spatial
variation in the concentrations of chemicals of potential
concern is obtained by calculating the coefficient of
variation for each chemical.  When the coefficient of
variation is less than 20% and a substantial quantity of
data are available, the effect of spatial and temporal
variation on concentrations of chemicals of potential
concern is minimal, and the power and certainty of
statistical tests is high (EPA 1989c).

Spatial variability can  be analyzed after an initial
sampling effort through simple statistical summation or
through the use of variogram analysis, a part of the
geostatistics.  EPA has developed software to assist a
risk assessor  in  this  analysis:   Geostatistical
Environmental Assessment Software (GEOEAS) (EPA
1988c) and Geostatistics for Waste Management
(GEOPACK) (EPA 1990b).

Measurement error is estimated using the results of QC
samples and represents the difference between the true
sample value and the reported value. This difference
has five basic sources: the contaminant being measured,
sample collection procedures, sample handling
procedures, analytical procedures, and data production
procedures.  Measurement error  due to analytical
procedures is discussed in Section 3.2 under analytical
issues. Measurement error due to sampling is estimated
by examining  the  precision of results from  field
duplicates. The minimum recommended number of
field duplicates is 1 for every 20 environmental samples
(5%). A minimum of one set of duplicates should be
 taken per medium sampled unless many strata are
 involved; five  sets are recommended.  Exhibit  31
 summarizes the types and uses of QC samples in defining
 variation and bias in measurement.

     *• Samplingvariabilitytypicallycontributes
     much more to total error than analytical
     variability.

 In summarizing the discussion of sampling variability
 and measurement error, one finding puts the concepts in
 perspective: "An analysis of the components of total
 error from soils data from an NPL site sampled for PCB s
 indicated that 92% of the total variation came from the
 location of the sample  and 8% from the measurement
 process" (EPA 19890. Of the 8%, less than 1% could
 be attributed to the analytical process. The rest of the
 8% is attributable to sample collection, sample handling,
 data processing and pollutant characteristics. Sampling
 variability Is often three to four times that introduced by
 measurement error. Exceptions to this observation on
 the components of variation or sources of error occur in
  instances of poor method performance for specific
  analytes.
                                                   50
                                                                                                            r

-------
             EXHIBIT 31.  MEASUREMENT OF VARIATION AND BIAS
                     USING FIELD QUALITY CONTROL SAMPLES
                 Quality Control
                 Sample Types
          Variation or Bla» Measured
                Field duplicate


                Field blank
                Field rtnsate
                Trip blank
Provides data required to estimate the sum of
subsampllng and analytical variances.

Provides data required to estimate the bias due to
contamination Introduced during field sampling or
cleaning procedures. Also measures contamination at
laboratory. Compare with laboratory method blank
to determine source of contamination.

Provides data required to estimate the sum of the bias
caused by contamination at the time of sampling from
sampling equipment and by analysis and data handling.
Indicates cross-contamlnaiion and potential contamination
due to sampling devices.

Provides data required to estimate the bias due to
contamination from migration of volatile organlcs into the
sample during sample shipping from the field and sample
storage at the laboratory.
                Source: EPA 1960c.
Media or matrix variability.  Appropriate samples
must be collected from each medium of concern and, for
heterogeneous media,  from designated strata.
Stratification reduces variability  in  results  from
individual strata, which can be different layers or surface
areas.  Media to  be sampled should  include  those
currently uncontaminated but of concern, as well as
those  currently contaminated.   For media of a
heterogeneous nature (e.g., soil, surface water,  or
hazardous waste),  strata  should be established and
samples specified by stratum to reduce variability, the
coefficient of variation and the required number of
samples.

Sampling considerations vary according to media. The
sampling concern may involve contaminant occurrence,
temporal variation, spatial variation, sample collection.
or sample preservation. Exhibit 32 indicates potential
sampling problem  areas for each medium.  Problem
areas are classified relative to other media.  RPMs can
use this exhibit to plan for possible sampling problems
in the data collection design. Sampling designs must be
structured to identify and characterize hot  spots.
Information needed for fate and transport  modeling
should be obtained during a site sampling investigation.
                   This information also differs by the medium of concern
                   (EPA 1989a).

                   The type of medium in which a chemical is present
                   affects the potential sensitivity, precision, and accuracy
                   of (he measurement Sharp distinctions occur in applying
                   a single method to media such as water, oil. sludge, soil,
                   or tissue. Medium or matrix problems are indicated by
                   the presence of analytical interferences, poor recovery
                   of analytes from the matrix, physical problems such as
                   viscosity (flow parameters), and paniculate content that
                   affect sample processing. Exhibit 33 shows the sources
                   of uncertainty across media.  Spiked environmental
                   samplcsmonitortheeffect of these sourcesof uncertainty
                   on the accuracy of recovery of target compounds from
                   the matrix.  Duplicates quantify the  effect of these
                   parameters on precision. The method must be chosen
                   carefully if a difficult medium such as oily waste or soil
                   is to be analyzed. Routine methods usually specify the
                   medium or media for which they are applicable.

                   Method detection and general confidence in analytical
                   determinations are also often affected by specific media
                    types and by analytical interference.  The impact of
                    matrix interference on detection limits, identification.
                                                   51

-------
             EXHIBIT 32.  SAMPUNG ISSUES AFFECTING CONFIDENCE
                                  IN ANALYTICAL RESULTS
Major
Sampling
Issues
Contaminant
Migration
Temporal
Variation
Spatial
Variation
Topographic/
Geological
Properties
Hot Spots
Sample
Collection
Sample
Prepa radon/
Handing
Sample
Storage
Sample
Preservation
Key: VV -
V •
Problem Likelihood by Medium
Ground Surface Hazardous
Soil Water Water Air BloU Watte
VV

vv vv
vv
vv vv
V
VV V
vv
vv
V V VV
VV V
VV V V VV
V
vv
VV VV V
VV VV V V
vv vv vv
vv vv
Ukely source ol significant sampling problem.
Potential source ol sampling problem.
Source: Modified horn Keith I990b.
and  quantitation is  illustrated by the following
discussions (which are not meant to be comprehensive).

  •  Oil and hydrocarbons affecting GC-MS analyses,

  •  Pbthalates and non-pesticide chlorinated
     compounds that can  interfere with pesticide
     analyses, and

  •  Iron spectral interference affecting ICP sample
     results.

Oil and hydrocarbons. The presence of appreciable
concentrations of oil and other hydrocarbons may
interfere with the extraction or concentration process.
Also, even at low concentrations, oil in a sample usually
produces a large series of chromatographic peaks that
interfere with thedetection ofother chemicalsof potential
concern during gas chromatography. Any chemicals of
potential concern that may elute concurrently from the
GC column are obscured by the hydrocarbon response
and may not present a  distinct spectrum.  Also,
hydrocarbons that are present in significant quantity are
often identified as TICs, potentially adding a large
number of compounds for consideration by the risk
assessor.
During RI planning, the risk assessor should determine
if there is a potential for hydrocarbon contamination,
through knowledgeofhistoricalsite use and examination
of historical data. The laboratory can be instructed to
add cleanup protocols to the analysis, or  to use a
supplemental analysis for which the hydrocarbons are
not interferences (e.g., electron capture detection for
halogenated compounds).

Phthalatcs  and  non-pesticide   chlorinated
compounds. Phthalates interfere with pesticide analyses
by providing a detector response similar to that for
chlorinated compounds. Phthalates and non-pesticide
                                                 52

-------
         EXHIBIT 33. SOURCES OF UNCERTAINTY THAT FREQUENTLY
                 AFFECT CONFIDENCE IN ANALYTICAL RESULTS
Degree of Significance by Medium
Source of
Uncertainty Soil Water
SAMPLING
Design
Contamination
Collection
Preparation
Storage
Preservation
LABORATORY
Storage
Preparation
Analysis
Reporting
ANALYTE-SPECIFIC
Volatility
Photodegradation
Chemical Degradation
Microbial Degradation
Contamination
VV
VV
V
VV
VV
VV/V
VV
VV
V
VV
VV
V
V
VV
VV
VV
VV
vv/v
V
V
VV
VV
Air Biota
VV
V
VV
VV
VV
VV
V V
V VV
V
V
VV
Hazardous
Waste
V
V
VV
VV
VV

KEY:
VV » Likely source of significant error or uncertainty.
V • Potentially source of significant error or uncertainty.
VV/V • Magnitude of effect determined by examination of data.
chlorinated compounds are often present in greater
concentrations than the pesticides of concent. Pesticide
data are often required at low detection limits and,
thercfore.GC-MS analyses are not usedforquantitation.
In these cases, a gas chromatographic analysis using
electron capture detection is more sensitive, providing
a wider useful range of detection. The phthalates and
chlorinated compounds can coelute with chemicals of
potential concern, thereby obscuring the detection of
target  analytes and  raising  the analyte-speciflc
quantitation limit.   Phthalates and chlorinated
compounds also produce additional  peaks on  the
chromatogram that can be interpreted as false positive
responses to pesticides.  A  second analysis using a
differentcolumn provides anextrameasureof confidence
in identification. Alternatively, sample extracts from
positive analyses can be further concentrated  for
confirmation by GC-MS if concentrations of analytes
are sufficient.
Iron.  Large quantities of iron in a sample affect the
detection and quantitation of other metallic elements
analyzed by ICP atomic emission spectrosoopy at
wavelengths near the iron signals.  The strong iron
response overlaps nearby signals, thereby obscuring the
results of potentially toxic elements present at much
lower concentrations. An interference check sample for
ICP analyses monitors the cffectof such elements. High
concentrations of iron are analyzed  with low
concentrations of other metals in these samples to
indicate whether iron interfered with metal detection at
lower concentrations.  If spectral interferences are
observed, data may be qualified as overestimated. The
risk assessor or RPM should consult the project chemist
to determine if a particular method requires a
performance check.
                                                53

-------
*=•-
                3.2.6  Sample Preparation and
                        Sample Preservation

                Some samples require preparation in the field to ensure
                that the results of analyses reflect the true characteristics
                of the sample.  Sample  filtration and compositing
                procedures are discussed  in this section. Exhibit 34
                summarizes  the issues which the various  sample
                preparation methods address. Exhibit 35. outlines the
                primary information gained with the various sampling
                techniques.

                          EXHIBIT 34.  SAMPLE
                         PREPARATION ISSUES
     EXHIBIT 35.  INFORMATION
   AVAILABLE FROM DIFFERENT
      SAMPLING TECHNIQUES
Issue
Sample
Integrity
Source of
Analyte
Media
Analyte
Speciation
Large
Number of
Samples to
be Analyzed
Action
Preservation — acids, biocides
(may be applicable to volatile*
or metals).
Unfiltered samples - measure
total analytes
Filtered samples - discriminate
sorbed and unsorbed analytes
Choice of sample preparation
protocols affects analyte
special! on
Composite samples
(However, this raises the
effective detection limit in
proportion to the number of
samples composited.)
                 Filtration. If the risk assessor needs to discriminate
                 between the amount of analyte present in true solution
                 in a sample and that amount sorbed to solid particles,
                 then the sample must be filtered and analyses should be
                 performed for both filtered and Unfiltered compounds.
                 Some samples,  such as tap water, are never filtered
                 because there is no paniculate content Filtration should
                 be performed in the field as soon as possible after the
                 sample has been taken and before any preservative has
                 been added to the sample, nitration often does not
                 proceed smoothly. It is common practice only to filter
                 a small proportion of all samples taken, and to perform
                 analyses for  the total content of the analyte in the
                 majority of samples. Filtered samplesgenerallyprovide
                 a good indication of the fraction of contaminant likely
                 to be transported over large distances horizontally in a
                 plume. However, in the immediate vicinity of a source
                 or point of exposure, unfiltercd samples may be valuable
                 in providing an indication of suspended material that
Sample
Type
Filtered
Unfiltered
Grab
Composite
Information
Can differentiate sorbed
and unsorbed analytes.
Total amount of analyte
in sample is measured.
Can be used to locate
hot spots.
Can provide average
concentrations over an
area at reduced cost.


                                    21-002-03$

may act as a source or sink of dissolved contaminants
and may therefore modify overall transport.
Compositing.  Reducing the number of samples by
compositing  is also a form of sample preparation.
Compositing may be performed to reduce analytical
costs, or in  situations where the risk assessor has
determined that an average value will best characterize
an exposure pathway. Compositing cannot be used to
idendfy hot spots, but can be effective when averaging
across the exposure area. Caution should be exercised
when compositing since low level detects can be
averaged out and become non-detects.
Preservation. Sample characteristics can be disturbed
by post-sampling biological activity or by irreversible
sorption of analytes of concern onto the walls of the
sample container. A variety of acids and biocides used
for preservation are discussed in standard works such as
Standard Methods for the Examination of Water and
Waste-water (Clesceri, el. al., eds. 1989). Samples are
also usually shipped with ice to reduce biological activity.
Preparation.  Several factors in sample preparation
affect analytical data.  These factors include sample
matrix, desired detection limit, extraction solvent,
extraction efficiency, sample preparation technique.
and whether the analysis is performed in the field or in
a fixed laboratory.  In  addition, parameters such as
turnaround time may preclude the use of some sample
preparation alternatives.

 An extraction method  must be  able to release the
chemicals of concern from the sample matrix.  For
example, organic solvents will extract non-polar organic
 compounds from water. Polar and ionic compounds
                                                                 54

-------
(such  as unsymmetrically  halogen-substituted
compounds, phenols, and carboxylic acids) may require
additional techniques for extraction from water.  The
choice of solvent is also  critical to the extraction
efficiency. Melhanol would be expected to extract a
larger quantity of volatile organic material from soils or
sediments than from water. For inorganic analyses, the
matrix may require additional acidification to dissolve
metal salts that have precipitated from the solution.

Sample preparation procedures for organic analytes are
applied based on volatility.  Volatile organics are
analyzed using head-space or purge and trap techniques.
Extraction alternatives for the analysis of less volatile
(extractable) organic chemicals  include separatory
funnels, Soxhlet extraction apparatus, continuous liquid-
liquid extractors, and solid phase cartridges. Details of
these extraction options can be obtained from the project
chemist.  Strengths and weaknesses of each of these
preparation procedures are described in Exhibit 36.

For inorganic analyses, the sample matrix Is  usually
digested in concentrated acid. The released metals are
introduced into the instrument, then analyzed by flame
AA or ICP atomic emission spectrophotometry. The
selection of the acid for digestion influences the detection
limit because different acids have different digestion
abilities.

   •  If digestion is not used, the sample measurement
      corresponds to a determination of soluble metals
      rather than total metals. If soluble metals have a
      greater toxicological significance, this difference
      may be important to the risk assessment.

   •  If the sample is filtered in the field or the laboratory
      before  digestion, any metals  associated with
      particulates are removed before  analysis.   If
      particulars are an exposure pathway in the risk
      assessment,   sample  filteration  would
      underestimate risk.

 The analytical request must specify if the sample is to be
 filtered and whether or not it is to be digested (to
 measure soluble metals). Unless otherwise specified.
 samples are usually digested but not filtered.

 3.2.7  Identification of Exposure
         Pathways

 Exposure pathways and their components,  such as
 source, mechanism of release, etc.. should be designated
 prior to the design of the sampling procedures. For the
 risk assessment, at least one broad spectrum analytical
 sample is required and two or three are recommended
for each medium and potential source in an exposure
pathway. If the site sampling design fails 10 consider all
exposure pathways and media, additional samples will
be required.

Current and future exposure pathways may be limited to
particular areas of a site.  If sampling activity can be
concentrated in these areas, the precision and accuracy
of the data supporting risk assessments can be improved.

Risk assessment requires characterization of each
exposure area for the site. Samples not falling within
the areas of potential concern are not used in  the
identification of chemicals of potential concern nor in
the calculation  of  reasonable maximum exposure
concentration.  Depending on exposure pathways, the
riskassessor may utilize only asmallnumberof samples
that were collected at a site. Exhibit 37 shows why the
identification of exposure pathways is critical to the
sampling design in order to maximize the  number of
samples that are useable in the risk assessment.

3.2.8   Use of Judgmental or
         Purposive Sampling Design

Judgmental or purposive designs that specify sampling
points based on existing site knowledge may be
appropriate for the initial phase of site sampling or when
the risk assessment is performed using few samples. In
such instances, non-statistical approaches may be more
effective in accomplishing the purpose of the  risk
assessment for human health, than statistical designs
with unacceptably large sampling variability.

J udgmental samples can be incorporated into a statistical
design if the samples designate the area of suspected
contamination as an exposure  area or stratum.   The
judgmental samples are then selected randomly or within
 a grid in the area of known contamination. Under the
 procedures described, the initial judgmental samples
 are not considered biased for the exposure area. Exhibit
 38 summarizes some strengths and weaknesses of biased
 and unbiased sampling designs.

 Resource constraints sometimes restrict the number of
 samples for the risk assessment and therefore potentially
 increase the variability associated with the results. When
 the number of samples that can be taken is restricted,
 judgmental sampling may identify  the chemicals of
 potential concern, but cannot estimate the uncertainty
 of chemical quantities. The reasonable maximum
 exposure or upper confidence limit cannot be calculated
 from results of a  judgmental design.  Bias can be
 avoided with the procedures described in the previous
 paragraph.
                                                    55

-------
            EXHIBIT 36.  COMPARISON OF SAMPLE PREPARATION OPTIONS
  Fraction
  4 Matrix
  Preparation
                Strength*
                                                                                  Wtaknesses
Volatile
SoWVaier
Head-space
                  Purge and Trap
Extractabto
Organic*
in Water
 Separately
 Funnel
                  Continuous
                  Extraction
                  Sotd Phase
                  Extraction
 Extractable
 Organic* In
 Soil
 Inorganic*
                  Sonicalkm
                  Soxhlet
                  Extraction
 Add Digestion
                   0.45 um
                   Membrane
                   Filtration

                   Direct Aspiration
Rapid, tirrple. potentially automated and
minimal interferences H standards an
prepared using sample media to minimize
the efleds ol Ionic strength variability
between samples and standards.
                    General/ recommended lor this analysis
                    (comparabSilles); can be automated;
                    broadty appicable and alow* concentration
                    factor; good recoveries across analyte 1st

                    High precision and recoveries lor waters.
Relatively rapid processing and low set-up
costs; relatively high PAH recovery.
                     Minimal matrix problems; generally higher
                     analytical precision and high phenol
                     recoveries; overal high extraction
                     efficiency (accuracy).
                     Very rapid, simple technique: samples can
                     be extracted In the field lor laboratory
                     analysis; potentially low MDL in a clean
                     matrix.
                     Rapid sample preparation; relatively low
                     solvent requirement; good efficiency ol
                     anatyte recovery/matrix exposure to
                     solvent.
 Relatively routine requirement (or direct
 analytical support; relatively good
 exposure of sample to solvent K sample
 texture appropriate; relatively low initial
 CO*.

 Dissolves paniculate*; provides results lor
 total metals.

 Isolates dissolved metal* specie*.
                     No preparation required; provide* results
                     for dissolved metal*.
Qualitative identification; comparison of
concentration possble but quantitative
standardization is difficult, especially true
lor complex matrix (e.g.. particulates and
day In sol): no mechanism for
concentration; application and sensHMty
are very anatyte-specJIic.

Sacrifice of either highly volatile anatytes or
Inadequate purge of low volatility analytes:
dependent on purge and trap parameters.
Sols have variable response dependent on
soil characteristics. Efficiency of sol purge
is not monitored.

Generally low recovery of target analytes;
high potential for matrix problems: poor
method precision.

Lower recovery of PAH and phlhalates
(especially higher molecular weight):
time-consuming procedure and high initial
set-up costs: more potential for
contamination.

Procedure has limited available performance
data. Presence of interference and matrix
problems can alted extraction efficiency
and data quality. Each batch ol extraction
medium must be tested lor efficiency by
recovery ol standards, preferably in the
same matrix.  Breakthrough (loss) occurs at
high sample concentrations.

Labor intensive; constant attention to
procedure: relatively high Initial cost.
Methytene chloride/acetone solvent mixture
resuls in many condensation product* and
often in method blank contamination.

Relatively high operating cost-replacement
 apparatus: solvent: lor some matrices may
 not provide eflidenl sample/solvent  contact
 (e.g.. channeling, very slow sample  output).
 Some compounds are acid Insoluble:
 dkjestkxi may promote interference effects.

 Filtration problem* In field; doe* not provide
 a total melals assay; I* an extra step ki
 •ample coUclton.

 Paniculate* •Meet sample Introduction.
                                                                                                                              tl-OM-MI
                                                                56

-------
                  EXHIBIT 37. IDENTIFICATION OF EXPOSURE PATHWAYS PRIOR TO
                       SAMPLING DESIGN IS CRITICAL TO RISK ASSESSMENT
Examples ot sampling design missing exposure areas of concern:
Systematic Grid:

x x / y'x x
x x.-' Xx x x
• *
* *
X/* /X X X
X..'* ..'X .X X X
* • 1
(A)
Random:
* V *
/ / x
X / / X X
/x/
* * V
« *
* *
« *
• *
/' X/' X
(A)
No samples
lor exposure
pathway A
and
five lor B
(B)
No samples
lor exposure
palhwayB
and
three lor A
(B)
3.2.9  Field Analyses Versus Fixed
        Laboratory Analyses

Field analyses are typically used to gather preliminary
information to reduce errors associated with spatial
heterogeneity, or to prepare preliminary maps to guide
further sampling. Fiekl analyses are often conducted
during the RI  to provide data  to determine worker
protection levels, the extent of contamination, well
screen casing depths, and the presence of underground
contamination, and to locate hot spots. For many sites,
field analyses can often provide useful data for risk
assessment  The analyses provide semi-quantitative
results, often free of significant matrix interference, that
can be used quantitatively if confirmed by a quantitative
analysis from fixed laboratories.

Field instruments are usually divided into three classes:.
field portable instruments that can be carried by a single
person, field transportable instruments that can be moved
and used in  the field or in a mobile laboratory, and
mobile laboratory instruments that are installed in a
trailer for transport to a site. Instrumentation used may
be GC. X-ray  fluorescence (XRF), or organic vapor
analyzer (OVA). Examples and applications of these
instruments might include on-site GC analysis of soil
gas  to  indicate the presence of  underground
contamination, XRF for soil lead analyses, and the
OVA to detect volatile organics. reported hi benzene
equivalents rather than in standard units of concentration.

Analytical methods that have traditionallybeen restricted
to off-site laboratories can now be employed in the field.
In addition, the quality of field instrumentation has
unproved steadily, allowing for be tier measurements at
the site.   Rugged versions of fixed  laboratory
instrumentation, such as XRF and GCs, can often be
performed in trailers if adequate ventilation and power
supplies are available.  With field  analyses,  greater
numbers of samples can be analyzed with immediate, or
very short, holding times with no shipping and storage
requirements.  At least 10% of field analyses should be
confirmed  by fixed  laboratory analyses to ensure
comparability.

    *• Field  methods  can produce  legally
     defensible data if appropriate method QC is
     available and if documentation is adequate.
                                                57

-------
       EXHIBIT 38. STRENGTHS AND
WEAKNESSES OF BIASED AND UNBIASED
             SAMPLING DESIGNS
Sampling
Design
Biased
(judgmental.
purposive)


Unbiased
(random,
systematic
grid.
geostallstical)

Strength*
• Uses knowledge of
local ion
• Fewer resources
• Timeliness
• Focuses samping
effort

• Ability to calculate
uncertainty
• AbJily to determine
upper confidence
NmK
• Representativeness
• Reduces probability
of false negative
Weaknesses •
• Inabiliytoralculale I
uncertainty I
• InabiSty to determine
upper confidence
liml
• Decreases
representativeness
• Increases
probabiity of false
negatives
• Resource intensive
• May require
statistician
• Timeliness
• More samples
required
 Significant QA oversight of  field analyses  is
 recommended to enable the data to be widely used.
 Field analysts performance data are often not available—
 in part because of the variety of equipment and operating
 environments, variety of sample matrices, and relative
 "newness" of certain technologies. Therefore, an in-
 field method  validation program is recommended.
 Spikes and performance evaluation materials should be
 incorporated, if available in addition to other standard
 QC measures such as blanks, calibration standards, and
 duplicates.

 The precision and accuracy of individual measurements
 may be lower in the field than at fixed laboratories, but
 the quicker turnaround and the possibility of analyzing
 a larger number of samples may compensate for this
 factor.  A final consideration is the qualifications of
 operators in the Held. The RPM, in consultation with
 chemists and quality assurance personnel, should  set
 proficiency levels required for each instrument class
 and  decide whether proposed  instrument operators
 comply with these specifications.

 Fixed laboratory analyses are particularly useful  for
 conducting broad  spectrum  analyses  for target
 compounds, to avoid the possibility of false negatives.
 They generally provide more information for a wider
range of analyics th;ui field analyses, and are generally
more reliable than field screening or field analytical
techniques.

    ••  To minimize the potential for false neg-
    atives, obtain data from a broad spectrum
    analysis from each medium and exposure
    pathway.

Fixed laboratory analysis commonly  uses mass
spcctromctry for  organic analyses, which provides
greatly enhanced abilities for compound identification.
For inorganics, A A spcctroscopy or ICP atomic emission
spcctroscopy should be used for reliable identification
'of target analytes. Once the broad spectrum analysis
and contaminant  identification  has occurred,  other
methods may be employed that offer lower detection
 limits, better quantitate specific analytes of concent,
 and that may be less expensive.

     «• The CLP or other fixed laboratory sources
     are most appropriate for broad spectrum
     analysis or for confirmatory analysis.

 Characteristics such as turruiround time, detection and
 identification ability of the instruments, precision and
 accuracy requirements of the measurements, and
 operator  qualifications should  be considered when
 selecting field or fixed laboratory instrumentation.
 Exhibit 39 compares the characteristics of field and
 fixed laboratory analyses. The risk assessor and RPM
 should consult the project chemist to consider the
 available options and make a choice of analysis based
 on method parameters, turnaround time, and cost, as
 well  as  other data requirements pertinent  to risk
 assessment needs (e.g., legal defcnsibility). Exhibit 40
 compares the strengths and weaknesses of field and
 fixed laboratory analyses.

 3.2.10 Laboratory Performance
          Problems

 The RPM should  be aware  of problems that occur
  during laboratory analyses, even though the resolution
  of such  problems are usually handled by the project
  chemist. This section discusses common performance
  problems and explains how to differentiate laboratory
  performance  problems from  method  performance
  problems.

      «•  Solicit the advice of the chemist to en-
      sure proper  laboratory  selection and to
      minimize  laboratory  and/or  methods
      performance problems that occur in sample
      analysis.
                                                   58

-------
                EXHIBIT 39.  CHARACTERISTICS OF FIELD AND
                          FIXED LABORATORY ANALYSES
Characteristic
Prevention of
false negatives
Prevention of
false positives
Analytical
Turnaround Time
Sample
Preparation
Field Analysis
Immediate analysis
means volatiles not lost
due to shipment and
storage.
No sample to sample
contamination during
shipment and storage.
Data available
Immediately or in up to
24 to 48 hours
(additional time
necessary for data
review).
Limited ability to prepare
samples prior to
analysis.
Fixed Laboratory
Analysis
More extensive sample
preparation available to
increase recovery of
analytes.
Contamination by
laboratory solvents
minimized by storage
away from analytical
system.
Data available in 7 to 35
days unless quick
turnaround time
requested (at increased
cost).
Samples can be
extracted or digested,
thereby increasing the
range of analyses
available. |
Laboratory performance problems mayoccurforroutine
or non-routine analytical services and can happen with
the most technically experienced and responsive
laboratories. Laboratory problems include instrument
problems and down-time, personnel inexperience or
insufficient training, and overload of samples.  Issues
that may appear to be laboratory problems, although
they are actually planning problems, include inadequate
access to standards, unclear requirements in the analytical
specifications, difficulty in implementing non-routine
methods, and some sample-related problems.  Another
problem for the RPM may be a lack of laboratories with
appropriate experience or available capacity to meet
analytical needs. These problems can usually be averted
by "up-front" planning and by a detailed description of
required analytical specifications.

   •  Instrument problems can be revealed with a unique
     identifier for each instrument in the laboratory that
     is reported with (he analyses. Calibration and
                   21-002-03*

performance standards, such as calibration check
standards, internal standards, or system monitoring
compounds, should be specified in the analytical
method to monitor performance of each instrument.
In addition, the use of instrument blanks should be
specified (to avoid the possibility of carry-over
during the analysis).

Some degradation in data quality may appear
when new personnel are operating or when the
sample load for a laboratory is high. The contrib-
uting personnel for each analysis should be
identifiedclearlyinlaboratory records andreports,
and qualifications of personnel required in contracts
should be documented.

Sample and  method problems can often  be
distinguished from laboratory problems if they are
not associated with aspccific instrument or analyst.
A review of method QC data should distinguish
between laboratory and sample problems.
                                                  59

-------
                    EXHIBIT 40. STRENGTHS AND WEAKNESSES OF FIELD
                               AND FIXED LABORATORY ANALYSES
     Analysts*
                  Strengths
             WMkn»s»*»
Reid -Portable XRF
(Metals)
Extremely high volume sampling and analysis;
compatible with sophisticated sampling and
data handling software. Detection limit may be
above laboratory Instrument values but
applicable to specific site levels of interest
Confirmation technique recommended.
Comparability may require external
standardization of calibration because
quantitation Is based on soil surface area
versus a soil volume. Results often lower
(nan from AA analyses.
Field GC
Rapid analysis supporting high volume sampling
[or variety of volatile and extractable organic
target compounds (includes pesticides/PCBs).
Minimization of sample handling variability and
data quality indicators comparable to fixed
laboratory methods.
                                                                       Requires prior site knowledge to ensure
                                                                       applicability to specific conditions (e.g.,
                                                                       soil-gas may not be appropriate for
                                                                       Investigation In sandy area). Confidence
                                                                       In Identification is matrix- and site-specific
                                                                       and highly variable depending on sample
                                                                       complexity. Confirmation technique
                                                                       recommended.
Mobile Laboratory
XRF. AA (Metals)
Combines the high volume sample capacity of
field analyses with the detection limits, data
quality and confidence associated with
laboratory analyses.
Requires significant resources, time,
and personnel to transport maintain
and operate; generally most appropriate at
high volume sites, especially remote.
Mobile laboratory
Luminescence
 Rapid survey of analytes that routinely
 require sample preparation (e.g., PAHs and PCBs).
 Detection limits can be adjusted within limits to
 site-specific concentrations ol concern.
Technique has had minima] use In EPA
site investigation. Comparability may
be an Issue and require extensive
confirmatory analyses.
 Mobile Laboratory
 GC, GC-MS
 Combines high volume capacity of field
 analyses with Increased confidence In
 identification (GC-MS) or Improved data
 quality (GC).  GC methods may be Identical
 to laboratory procedures but quality is
 Intermediate due to site conditions (e.g..
 temperature, humidity and power requirements).
Same weaknesses as for mobile
laboratory Inorganics. An additional
weakness Is the increased training
requirements and decreased availability
of experienced GC-MS operators for
totally Independent system operation.
Possibiity of site contamination and
cross-contamination.
 Fixed Laboratory
 XRF.AA.ICP
 (Metals - Available
 Routine Methods)
 Highest comparability and representativeness.
 Data quality. Including detection limits,
 generally predictable. Efficient match of analyses
 required to Instrument (e.g.. multiple analyses
 run simultaneously by ICP).
 Slow delivery of data; increased
 documentation requirement due to
 the number of particlpants-felattvely
 high sample cost
 Fixed Laboratory
 GC & GC-MS
 (Organic* • Available
 Routine Methods)
 Highest comparability and representativeness.
 Necessary confirmation of qualitative
 Identification. Data quality and detection
 limits generally predictable. In depth
 analysis and sample archives for follow-up
 testing.                       	
 Same weaknesses as for Axed
 laboratory metals; anaryte-spedfic
 performance.
 ICP • Inductively Coupled Plasma Spectroscopy. Graphite AA » Graphite Furnace (electrothermal) Atomic Absorption
 SpecUoscopy. Flame AA » Flame Atomic Absorption Spectroscopy. ICP-MS »Inductively Coupled Plasma-Mass
 Spectroscopy. XRF * X-Ray Fluorescence.  GC « Gas Chromatography. GC-MS • Gas Chromatography-Mass
 Spectrometry. AA « Atomic Absorption Spectroscopy.
                                                                                                           11402-040
                                                       60

-------
                   EXHIBIT 40. STRENGTHS AND WEAKNESSES OF FIELD
                              AND FIXED LABORATORY ANALYSES
                                                (Cont'd)
  Analysis*
             Strengths
            Weaknesses
ICP
Simple, automated, extremely rapid; can assay
metals simultaneously; can detect ppb levels.
Subject to salt or iron interferences; lacks
detection capability at low levels; not
suitable for less than 20 ppb Arsenic, Lead,
Selenium, Thallium, Cadmium, Antimony;
required-background and interelement
correction.
Graphite AA
Simple, automated; can assay most metals; can
assay low level metals; can detect ppb levels.
Lower precision and accuracy result unless
methods of standard additions used.
Method is time-consuming; requires
background correction; requires matrix
modifiers; subject to spectral interferences.
Graphite tube requires replacement
frequently.
Rama AA
Simple, rapid, very suitable for high concentration
sodium and potassium assays; commonly used and
rugged.
Not as sensitive as graphite AA; salts can
interfere; limited by lamp capabilities;
detects ppm levels.
ICP-MS
                Rapid; can detect low levels; accurate.
                                               Method is subject to isobaric molecular and
                                               ion interferences. Nebulization, transport
                                               process, and memory physical
                                               interferences occur. Method is relatively
                                               new and is expensive.  Specialized training
                                               is required.
 ICP-Hydride
                Rapid; can detect low levels of Antimony, Arsenic,
                Selenium; Hydride formation eliminates spectral
                interferences.
                                                Dependent on analyte oxidation state;
                                                especially sensitive to copper interference.
                                                Method is relatively new. Specialized
                                                training is required.
 ICP «Inductively Coupled Plasma Spectroscopy.  Graphite AA » Graphite Furnace (electrothermal) Atomic Absorption
 Spectroscopy. Flame AA « Flame Atomic Absorption Spectroscopy. ICP-MS »Inductively Coupled Plasma-Mass
 Spectroscopy. XRF « X-Ray Fluorescence. GC » Gas Chroma tog raphy.  GC-MS = Gas Chromatography-Mass
 Spectrometry. AA » Atomic Absorption Spectroscopy.
                                                                                                21-002-04041
                                                   61

-------
                                        Chapter 4
       Steps for Planning for the Acquisition  of Useable
      Environmental Data  in Baseline Risk Assessments
This chapter provides planning guidance to the RPM
and risk assessor for designing an effective sampling
plan and selecting suitable analytical methods to collect
environmental analytical data for use in baseline risk
assessments.  It is important to understand  that the
variances inherent in  both sampling and analytical
designs combine to contribute to the overall level of
uncertainty.  The  chapter also provides a number of
charts and worksheets that should be useful in planning.
It is important to remember that these are provided for
guidance only. Each Repion, or the staff at an individual
site, may modify these for their use or develop their own
materials.

The chapter has two sections. The first section of the
chapter describes the process of selecting a sampling
design strategy and developing a sampling plan to
resolve the four fundamental risk assessment decisions
presented in Chapter 2:

   • What contamination is present and at what levels?

   • Arc site concentrations sufficiently different from
     background?

   • Are all exposure pathways and exposure areas
     identified and examined?

   • Are all exposure areas fully characterized?

 A Sampling Design Selection Worksheet and a Soil
 Depth Sampling Worksheet are used as data collection
 and decision-making tools in this process. Guidance for
 evaluating alternative sampling strategies and designing
 statistical sampling plans is included.

 The second section of the chapter provides guidance on
 selecting the methods for analyzing samples collected
 during the RI. A Method Selection Worksheet is used
 to compile the list of chemicals of potential concern and
 to determine analytical priorities so that the most suitable
 combination of methods is selected.

 The risk assessor or RPM, in consultation with other
 technical experts, will probably complete several
 worksheets, representing different media,  exposure
 pathways,  potential sampling strategies, chemicals of
 potential concern, and analytical priorities. This is done
 to compile sufficient information to communicate basic
 risk assessment requirements to the RPM, and to ensure
 that these requirements are addressed in the sampling
 and analysis plan (SAP).

 The selection of sampling plans and analytical methods
 should be based on the performance measures discussed
in this chapter. These measures are assessed by data
quality indicators that quantify attainment of the data
quality objectives (DQOs) developed by the RPM for
the total data collection and evaluation effort.
4.1  STRATEGIES FOR DESIGNING
     SAMPLING PLANS

This section provides guidance forevaluating alternative
sampling strategies.  Risk assessment may involve
sampling many media at a site: groundwater, surface
water, soil, sediment, industrial sludge, mine tailings, or
air. The strategics for sampling different media often
vary. For example, random stratified sampling may be
the appropriate method for examination of soils at a site.
but the positioning of groundwater monitoring wells is
seldom done on a random basis. Sampling designs for
soils and sediments  are usually created to examine
spatial distribution and heterogeneity of chemicals of
concern.  Groundwater  sampling plans examine the
                  Acronyms

  AA        atomic absorption
  BNA      base/neuiral/acid
  CAS       Chemical Abstracts Service
  CLP       Contract Laboratory Program
  CV        coefficient of variation
  CVAA     cold vapor atomic absorption
  DQO      data quality objective
  EMMI     Environmental Monitoring Methods Index
  EMSL-LV  Environmental Monitoring Systems
             Laboratory - Las Vegas
  EPA      U.S. Environmental Protection Agency
  GC        gas chromatography
  GFAA     graphite furnace atomic absorption
  CIS       Geographic Information System
  GPC      gel permeation chromatography
  ICP       inductively coupled plasma
  MDL      method detection limit         «
  MDRD    minimum delectable relative difference
  MS       mass spectrometry
  PA/SI     primary assessment/site inspection
  PCB      polychlorinated biphenyl
  QA       quality assurance
  QC       quality control
  RAS      routine analytical services
  RI        remedial investigation
  RME      reasonable maximum exposure
  RPM      remedial project manager
  SAP      sampling and analysis plan
  VOA      volatile organic*
  XRF      X-ray fluorescence
                                                  63

-------
extent of a plume containing the chemical of concern,
and also often examine seasonal or temporal variability
in chemical concentrations. Exhibit 41 summarizes the
relative variation in spatial and temporal properties for
different types of measurement.

The terms stratum and strata are used frequently in this
section. A stratum is usually a physically defined layer
or area; it can also be a conceptual grouping of data or
site characteristics that is used in statistical analysis.

Sampling guidance in  this section is focused on
determining the spatial  extent and variability of the
concentration  of chemicals of potential concern.
Therefore, it applies most directly to soils and sediments.
Some EPARegions have developed samplingguidances
for groundwater, and the RPM and risk assessor should
consult these whenever available.

Examples of common sampling designs are given in
Exhibit 42, and their overall  applicability is shown in
       Exhibit 43. Schematic examples of some of the designs
       are illustrated in Exhibit 44.

       The objective of the sampling plan is to determine a
       strategy  that collects data representative of site
       conditions.  The data must have acceptable levels of
       precision and accuracy, obtain minimum required levels
       of detection for chemicals of potential concern, and
       have acceptable probabilities of false positives and false
       negatives. Meeting these objectives involves optimizing
       the confidence in concentration estimates and the ability
       to detect differences between site and background levels.
       To accomplish these objectives, the RPM can optimize
       the number of samples, the sampling design, or the
       efficiency of statistical estimators (e.g., mean, standard
       deviation, and standard error).

       Increasing the number of samples may increase initial
       costs, depending on whether fixed or field analytical
       methods are used for analysis, but it is necessary in
                     EXHIBIT 41.  EXAMPLES OF SPATIALLY AND
                        TEMPORALLY DEPENDENT VARIABLES
                                             Relative Variation In Measurements
                                                        Attributable to:
                   Measurement
         Geophysical Measurements
         Soil-Gas Measurements
         Weather/Air Quality
                                                                        Usually Large
Usually Small
Surface Water Quality
          Physical Soil Properties
          Soil Moisture
          Aquifer Properties
                                                                         Usually Small
Usually Large
Groundwater Flow
          Concentration of Groundwater
          Contaminants
                                                                                         21-002-041
                                                  64

-------
     EXHIBIT 42. EXAMPLES OF
          SAMPLING DESIGNS
De*lgn
Judgmental/
Purposive
Classical Random
Classical Stratified:
Random
Systematic
Cluster
Composite
Systematic:
Random
Grid
Search
Surrogate
Phased
Geostatistical
Example* of Application
Monitoring Wells
Hot Spots
Background Soil
Drums at Surface
Waste Piles
Soil from Boreholes
Soil from Test Pits
Determine Concentrations of
Chemicals of Potential
Concern in Soil
Concentrations of Chemicals
of Potential Concern. Surface
Soil Characteristics
Contaminant Hot Spots
Gas Detector Measurements
Extent of Contamination
Distribution of Contamination
certain situations (see Section 4.1.2).  The sampling
design can often be improved by stratifying within a
medium to reduce variability, or by selecting a different
sampling approach, such as a geostatistical procedure
termed "kriging."  Improving the efficiency  of the
statistical  estimators involves specifying the type of
data distribution if parametric procedures are being
used, or switching from nonparametric to parametric
procedures if distributional assumptions can be made.

Exhibit 45 is a Sampling Design Selection Worksheet,
structured to assist design selection for the most complex
environmental situation, which is usually soil sampling.
The worksheet contains the elements needed to support
the decisions for RI sampling design to meet data
requirements for risk assessment.  The RPM and risk
assessor may use this worksheet or use it as a model to
create one specifically suited to their needs. The final
site sampling plan must meet the  data useability
requirements of risk assessment The final procedure
for sampling design should be selected based on the
specific reason for sampling (e.g.. defining a boundary
or obtaining an average over some surface or volume).
The worksheet should be completed for each medium
and exposure pathway at the site. Once completed, this
initial set of worksheets can be  modified to assess
alternative sampling strategies.  Completion of a set of
worksheets  (i.e.. a worksheet for each medium and
exposure pathway at a site, based on a single sampling
strategy) specifies the total number of samples to be
taken for an exposure pathway, and sample breakdown
according to type (i.e., field  samples, quality control
samples, and background samples).

The remainder of this section  is a step-by-step guide to
completing tne Sampling Design Selection Worksheet.
Chemicals of potential concern listed on the Sampling
Design Selection Worksheet should be the same as
those used for the Method Selection Worksheet (Exhibit
52).

4.1.1  Completing the Sampling
        Design Selection Worksheet

    f Use of the Sampling  Design Selection
    Worksheet will help the RPM or statistician
    determine an appropriate sampling design.

Pathway.mediumand design alternatives. Sampling
procedures  used in environmental sampling are either
unbiased or biased. Classical  and geostatistical models
are unbiased in terms of  sample evaluation and
hypothesis  testing.  The classical model is based on
random, or stratified random procedures, and the
geostatistical model on optimizing co-variance.
Systematic  grid sampling can be utilized by either the
classical or geostatistical model. Biased, or j udgmental/
purposive, design requires the use of different approaches
to planning and evaluation.

    «•  While other designs may be appropriate
     in  many cases,  stratified random or
     systematic sampling designs are always
     acceptable.

   • Classical model: The classical model uses either
     a random or stratified random sampling design. It
     is appropriate for use in sampling any medium to
     define the representative concentration value over
     the exposure area. It is not subject to judgmental
     biases, and produces known estimates and
     recognized statistical measures and guidelines. A
     stratified random design provides the RPM and
     risk assessor with great flexibility.  If the nature
      and extent of the exposure areas are not yet well
      defined, a pilot random study can be conducted
      and the results included in the final design. The
      data can be averaged for any exposure area. The
      classical model is the basis for calculating
                                                  65

-------
        EXHIBIT 43. APPLICABILITY OF SAMPLING DESIGNS
                                           Objective of Sampling
                                Estimate
                                Chemical
                             Concentration
                               Distribution
Evaluate
 Trends
 Identify
Hot Spots
      Judgmental/
      Purposive
      Classical Random
      Classical Stratified

         Random

         Systematic
                   Maybe

                   Maybe
      Systematic:

         Random

         Grid
       Geostatistical
confidence levels, power, and minimum detectable
relative differences (MDRDs).

Geostatistical model:  Geostatistical techniques
are good for identifying hot spots and can be used
for calculating reasonable maximum exposure
(RME).  These techniques require complex
judgmental or purposive calculation procedures.
Even with the use of available computerprograms,
a statistician .should be consulted because different
                            21-002-043

approaches to estimating key parameters can
produce different estimates.

Systematic grid sampling:  Systematic grid
sampling procedures are good for identifying
unknown hot spots and also provide unbiased
eslimatesof chemical occurrence and concentration
(Gilbert 1987) useful in calculating the RME.
Systematic sampling can be used in geostatistical
or classical  estimation models.  Variance
                                          66

-------
     EXHIBIT 44. COMMON SAMPLING DESIGNS
 Simple Random
   Sampling
     Cluster
    Sampling
                                                   Clusters
Stratified Random
    Sampling
Stratified Systematic
     Sampling
                   Strata
                    Strata
 Systematic Grid
    Sampling
 Systematic Random
     Sampling
                                                    21-002-044
                           67

-------
        EXHIBIT 45. HIERARCHICAL STRUCTURE OF SAMPLING DESIGN
                            SELECTION WORKSHEET
    Parti
Medium Sampling
   Summary
                                 Exposure Pathway II
                              Exposure Pathway I
    Part II
Exposure Pathway
   Summary
                                                                  Exposure Area 0
                                                               Exposure Area C
                                    Part III
                                Number of Samples
                                 in Exposure Area
Exposure Area B
                                                               Exposure Area A
                                                                 Part III
                                                             Number of Samples
                                                              in Exposure Area
                                                                            21-002-045

-------
            EXHIBIT 45. PARTI: MEDIUM SAMPLING SUMMARY
               SAMPLING DESIGN SELECTION WORKSHEET
                                  (Cont'd)
A. Site Name	
C. Medium: Groundwater. Soil. Sediment, Surface Water. Air
            Other (Specify)	;	
D. Comments:	
B. Base Map Code.
E. Medium/
Pathway
Code


Exposure Pathway/
Exposure Area Name
_..
Column Totals
	 — 	
F. Number of Samples from Part II
Judgmental/
Purposive


Back-
ground


Statistical
Design


Geo-
metrical
orGeo-
statistical
Design


QC


Q: Grand Total:
Row
Total



                                                                       H-MB-04S41
                                     69

-------
         EXHIBIT 45. PART II: EXPOSURE PATHWAY SUMMARY
               SAMPLING DESIGN SELECTION WORKSHEET
                                     (Cont'd)
H.
Chemical of Potential Concern
and CAS Number

1.
Frequency
of
Occurrence

J. Estimation
Arithmetic
Mean
—
Maximum

K.
CV

L.
Background

M. Code (CAS Number) of Chemical of Potential Concern Selected as Proxy
N. Reason for Defining New Stratum or Domain (Circle one)
   1. Heterogeneous Chemical Distribution
   2. Geological Stratum Controls
   3. Historical Information Indicates Difference
   4. Field Screening Indicates Difference
   5. Exposure Variations
   6. Other (specify)	
O. Stratum or Exposure Area
Name and Code
<

P.
Reason

Q. Number of Samples from Part III
Judgmental/
Purposive

R. Total (Part I. Step F)
Back-
ground


Statistical
Design


Geo-
metrical
orGeo-
statlstlcal
Design


OC


Row
Total


                                                                                    lt-OOJ-045-OJ
                                            70

-------
f9*&*
                               EXHIBIT 45. PART III: EXPOSURE AREA SUMMARY
                                   SAMPLING DESIGN SELECTION WORKSHEET
                                                         (Cont'd)
               O. Stratum or Exposure Area
               E. Medium/Pathway Code

               S. Judgmental or Purposive Sampling
                  Comments: 	
                                                             Domain Code _
                                                             Pathway Code.
                   Use prior site information to place samples, or determine location and extent of contamination.  Judgmental or
                   purposive samples generally cannot be used to replace statistically located samples.

                   An exposure area and stratum MUST be sampled by at least TWO samples.
                   Number of Samples
                   Number of Background Samples
               T.  Background Samples
                   Background samples must be taken for each medium relevant to each stratum/area. Zero background samples
                   are not acceptable. See the discussion on page pp. 74-75.
                                                                    (<40% if no other information exists)
U.  Statistical Samples
    CV of proxy or chemical of potential concern
    Minimum Detectable Relative Difference (MDRD)
    Confidence Level	(>80%)  Power of Test

    Number of Samples
    (See formula in Appendix IV)
               V.  Geometrical Samples
                   Hot spot radius
                          . (Enter distance units).
                   Probability of hot spot prior to investigation
                   Probability that NO hot spot exists after investigation
                   (see formula in Appendix IV)

               W.  Geostatistfcal Samples

                   Required number of samples to complete grid +
                  ' Number of short range samples
                                                .(0 to 100%)
                                                	(enter only if >75%)
               X.  Quality Control Samples
                   Number of Duplicates
                   Number of Blanks
               Y.  Sample Total for Stratum
                    (Part II. Step U)
                                (Minimum 1:20 environmental samples)	
                                (Minimum 1 per medium per day or 1 per sampling
ratum
Judgmental/
Purposive

Back-
ground

Statis-
tical
Design

Geo-
metrical
orGeo-
statistical

QC



Row
Total

9ijytf.ru m»
                                                             71

-------
    calculations required loeslimate confidence limits
    on theaverageconcentrationareavailable (Caulcutt
     1983).   Systematic sampling is powerful for
    complete site or exposure area characterization
    when the exposure area is  known to be
    heterogeneous.

Determining number of samples. Four factors need to
be considered in determining the total number of samples
required (see Exhibit 46):
  •  Exposure areas,

  •  Statistical performance objectives (based on site
     environmental samples),

  •  Quality assurance objectives  (based on QC
     samples), and

  •  Background samples (based on MDRD).
variation for a chemical of potential  concern  and
measures of performance is the basis for determining
the number of samples necessary to provide useable
data for risk assessment.

    •r If the natural variability of the chemicals
    of potential concern is large (e.g., greater
    than 30%), the major planning effort should
    be to collect more environmental samples.

The number of samples can be calculated given a
coefficient of variation, a required confidence level or
certainty, a required statistical power, and an MDRD.
Exhibit 47 illustrates the relationships between  the
number of samples required given typical values for the
coefficient of variation and statistical performance
objectives.  Calculation formulas in Appendix IV
facilitate the examination of effects beyond the examples
cited.
  EXHIBIT 46.  FACTORS IN DETERMINING
TOTAL NUMBER OF SAMPLES COI LECTED
    Number of Exposure Are** That wOl be Sampled
    (p. 74)

     • Media within exposure area
     • Strata wkhin exposure area medhxn

    Number of Sample* for Each Exposure Area
    Grouping Given Required Stattstfcal Performance
     • Confidence (1- a), where a is the probabiity ol a
       type I error
     • Powef(1-p).whor«|Jbth« probability d a type II error
     • Minimum detectable relative difference

    Number of Quality Control Samplea (p. 7«)

     • Field duplicate (cotocaled)
     • Field duplicate (spit)
     • Blar* (trip, field, and equipment (rinsate))
     • Field evaluation

    Number of Background Sample* (p. 74)

     • Number o< ste samples collected
     • Mioknurn detectable relative diflerence
 The number of environmental site samples is ultimately
 controlled by performance requirements, given the
 statistical sampling design. The relationship between
 numberof samples and measures of performance depends
 upon the variability of the chemicals of potential concern,
 which is measured by the coefficient of variation. In
 other words, the relationship between the coefficient of
4.1.2  Guidance for Completing the
        Sampling Design Selection
        Worksheet

This section provides  step-by-step instructions for
completing the Sampling Design Selection Worksheet
shown in Exhibit 45.

Part I: Medium Sampling Summary

A. Enter the Superfund site name.

B . Enter a code that uniquely identifies a base map of
    the site or the exposure unit.

    All sampling events should be identified on a map
    or in a database such as a Geographical Information
    System (CIS).

 C. Identify the medium to be sampled (e.g., soil,
    groundwater, industrial sludge, mine  tailings.
    smelter slag, etc.).

 D. Enter any comments required to  describe the
    exposure area, and other information such as the
     RPM's name.

 E.  Enter a medium/pathway code that  has been
     assigned for the risk investigation.

 F.  Specify the exposure pathway (e.g., ingestion of
     soil).

     Leave this entry blank for now,  then enter the
     number of samples for each  category that have
     been selected from Part II (Step R) of the worksheet
     when completed.
                                                   72

-------
•v  r
EXHIBIT 47. RELATIONSHIPS BETWEEN MEASURES OF STATISTICAL
                       PERFORMANCE AND NUMBER
                          OF SAMPLES REQUIRED
Samples Required to Meet
Minimum Detectable
Coefficient
of Variation (%) Power (%)
10
15
20
25
30
35
Note:
Source:
95
95
95
95
95
95
Confidence Relative Difference
Level (%)
90
90
90
90
90
90
5%
36
78
138
216
310
421
10%
10
21
36
55
78
106
Number of samples required in a one-sided one-sample t-test to achieve a
minimum detectable relative difference at confidence level and power. CV
on geometric mean for transformed data.
EPA1989C.



20%
3
6
10
15
21
28
based
_
                Sample types are broken out by sample type:

                  •  Judgmental/Purposive,

                  •  Background,

                  •  Statistical design (e.g.. stratified random
                     sampling).

                  •  GeomctricalorgcostatisUcaldesignOncluding
                     hot spot sampling), and

                  •  Quality control samples.

                *• At least one broad spectrum analytical
                sample is required for risk assessment, and
                a  minimum  of two or three  are
                recommended for each medium in an
                exposure pathway.

             G. Enter the grand total of all samples within a specific
                medium.
                                                                              21-002-047

                                           Part II:  Exposure Pathway Summary

                                           H.  List the chemicals or potential concern and their
                                               CAS numbers.

                                               List the known or suspected chemicals of potential
                                               concern based on historical data. This will generally
                                               be from the PA/SI.

                                           I.   List the frequency of occurrence (%).

                                               The frequency of occurrence is the  percent of
                                               samples in which the chemical of potential concern
                                               has been identified. This may be obtained from
                                               site-specific data or calculated from historial (PA/
                                               SI) data (r fate and transport modeling.

                                           J.  Enter an estimate of the average (arithmetic mean)
                                               and maximum concentration of the chemical of
                                               potential concern.

                                               Historical data or data from similar sites can be
                                               used to derive these values. More sampling will
                                               usually be necessary to determine  statistically
                                                            73

-------
    significant differences if these values arc close to
    background levels or to the levels of detection.

K.  Estimate the coefficient of variation.

    The coefficient of variation (CV) can be estimated
    from site-specific data or from data from similar
    sites. The number of samples necessary to produce
    useable data will generally increase as the CV
    increases.   The definition of separate strata or
    domains should be  investigated if a CV is above
    50%.  Exhibit 23 contains a listing of historical
    values for CVs that may be used as an estimate in
    the absence of site-specific data.

L.  Estimate background concentration.

    Background concentration estimates should be for
    each medium relevant to each strata/area.  Site-
    specific data are preferred, but data from similar
    sites can be utilized.

M. Select a proxy chemical of potential concern.

    Choose a proxy from  the  list of chemicals of
    potential concern to develop sampling plans. Note
    that a proxy that  has the highest CV, lowest
    frequency of occurrence, or whose concentration at
    the site is closest to background levels will require
    the most samples.

 N. Develop the reason for defining new strata or areas.

       •  Heterogeneous Chemical Distribution:  If a
          chemical can be shown to have dissimilar
          distributions  of concentration in different
          areas, then the areas should be subdivided.
          For  example^hot spots may be considered
          separately.

       •  Geological Stratum Controls: Knowledge of
          local geologic conditions can  be used to
          produce separate areas where similar statistical
          distributions are likely to exist In particular,
          different "stratigraphic" layers may produce
          distinct strata.

        •  Historical Information: Historical information
          on  production, discharge or storage of
          chemicals of potential concern can be used to
          identify separate areas.

        • Field Screening: Field analytical results can
           be  used to locate sub-populations that are
           mapped into exposure areas.

        •  Exposure Variations:  Information or
           variations in behavior patterns, land use or
           receptor groups can beusedtoidentify separate
           areas.
      •  Other reasons can be used to produce separate
         sampling areas, such as observed stress on
         vegetation, oily appearance of soils, or the
         existence of refuse, etc.

O.  List the stratum or area name and code.

    The stratum or area identifies sub-areas on the site
    base-map.

P.  Annotate reason from Step N.

Q.  List the number  of samples  estimated after
    completing Part III of this worksheet.

R.  List the number  of samples  estimated after
    completing Part II and Part in of this worksheet.

Part III: Exposure Area Summary

S.  Enter j udgmental/purposive sampling comments.

    A minimum of three to fivejudgmental or purposive
    samples  must be  used to  sample a stratum  or
    exposure area. Historical or prior site information
    can be used to locate sampling positions to determine
    the  extent and magnitude of contamination.
    Chemical field screening, geophysics, vegetation
    stress, remote sensing, geology, etc.  can also be
    used to guide judgmental sampling. Judgmental or
    purposive samples  are  not  recommended for
    estimating average and maximum  values within a
    stratum or domain area, but they  can be used in
    geostatistical kriging  estimations  and can  be
    included in calculating risk.

 T. Identify background samples.

    For statistical purposes, a sufficient number of
     background samples must be taken to determine
     the validity of the null hypothesis that there is no
     difference between mean values of concentration
     in the site  and the background  samples at  the
     desired level of confidence.  Early sampling and
     analysis of background samples will indicate the
     ease  with  which background  levels can be
     discriminated, and allow modifications to be made
     to the SAP if necessary.

     Background samples must be  taken  for each
     exposure pathway.  As with QC  samples, results
     from  the background  sample should be assessed
     early to sec if background  levels will severely
     impact the  sampling design.  The number of
     necessary background samples increases as the
     variability of the background values  increases.
     Background samples should not be used in the
     estimation of average or maximum values within a
      stratum or exposure area,  but they can be used in
                                                     74

-------
   kriging estimations.  In those instances where
   background levels are close to on-site contamination
   levels, it may be  necessary to collect as many
   background samples as site samples. Small numbers
   of background samples increase the probability of
   atypell.false negative error(i.e., that nodifference
   exists between site and background when a
   difference does, in fact, exist). However, rigorous
   statistical analyses involving background samples
   may be unnecessary if site and non-site related
   contamination clearly differ.

   w Collect and analyze background samples
   prior to the  final determination  of the
   sampling  design since  the number of
   samples is significantly  reduced if little
   background contamination is present.

   Background levels ofcontaminants vary bymedium
   and the type of contamination. • If a detectable
   background level  of a  contaminant occurs
   infrequently, the number of background samples
   analyzed might be kept small. Metals often have
   high rates of detection  in background samples.
   Some pesticides, such as DDT. are anthropogenic
   and also have high rates of detection in particular
   matrices.  Anthropogenic background levels are
   also found in sites near industries and urban areas.
   It is important to distinguish detection, or lack of
   detection, in a single sample from a false positive
   or false negative result. Results from single samples
   are different estimators than those from statistical
   parameters from  pooled samples.  Background
   sampling must be increased  in the following
   situations:

      •  Contamination exists in more than one
         medium,

      •  Expected coefficients of variation in chemicals
         of concern are high and confirmed by actual
         data,

      •  Relative differences between site and
         background levels are small, and

      •  Site concentrations and concentrations of
         concern are low.

U.  Identify statistical samples.

    Samples should be systematically or randomly
    located. The number of samples can be calculated
    using the CV of the proxy variable, the required
    MDRD, the required confidence level and power of
    the lest, and the appropriate statistical formula and
    appropriate charts.
   For example, using the equation in Appendix IV:
    Where Zo and Z, arc obtained from the normal
    distribution tables for significance levels a
    and D respectively; a is the probability of the
    false positive error rate, and B is the probability
    of the false negative error rate.

    Then,  if a is 0.2 (20%) and the confidence
    level is 80% then Z0 is 0.842. If B is 0.05 (5%)
    then the power is 95% and Z, is 1.648.

    If the MDRD is 20% and the CV is 30%, then
    D = MflKQ which equals 0.666
         CV
    and n>15 samples are required.
V.  Identify samples from geometrical design.

    *• Systematic sampling supplemented by
    judgmental sampling is the best strategy
    for identifying hot spots.

    For example, using the equation in Appendix IV:
    Where R = 20m

    and A = 37.160m1

    and X = 0.3  Probability that a hot spot is in the
                exposure area from "historical
                records" or from field screening or
                geophysical tests.

    and C = 0.2  The acceptable "walk away"
                probability that a hot spot exists
                after a sampling grid has been
                done.
    then:

        D « 2.7, R = 54.8 m, and
        n a 27,160/54.82 =12.37

    Therefore 12 samples are required.
     Note that the requirements for IS samples from a
     statistical sampling approach can be met in this
     example if the hot spot search is augmented by
     randomly locating two additional samples.  The
     results for number of samples from U and V are not
     additive.

 W.  Identify samples from gcostatistical design.

     Ageostatistical sampling pattern should be designed
     at the early stage of planning. A statistician should
     be consulted to develop the design.
                                                   75

-------
X.  Quality Control Samples

    Generally, duplicates should be taken at a minimum
    of 1 duplicate for every 20 environmental samples
    (EPA 19890.  However, this frequency may be
    modified based on site conditions.  For example.
    the number of duplicates and other QC samples
    may be set high for the beginning of site sampling,
    evaluated after several duplicates to  determine
    routine measurement error, and  subsequently
    adjusted according to observed performance. The
    information in Exhibit 48 shows that confidence in
    measurement error increases sharply when four or
    more pairs of duplicate samples are taken per
    medium.  Critical samples are recommended for
    designation as duplicates in the Q A sampling design.
      EXHIBIT 48. NUMBER OF SAMPLES REQUIRED
      TO ACHIEVE GIVEN LEVELS OF CONFIDENCE,
                  POWER, AND MDRD 1
       Confldtnc* (1-a)    Powtf(l-O)   MDRD   No. of Samptot
          90%
          90%,
          60%

          60%
          80%,
          80%
                 90%
                 90%,
                 90%
                 80%
                 80%,
                 90%
10%
20%
20%
10%
20%
40%
42
12
 8
IB
 6
 3
       1 Value* (or numbtf ot uirplM art battd on a CV ol 25 V

       2 Th« minimum recommtntftd ptrlormanct m>«ui« lor iM auMimtnt
        aw confidcno* (80%) and pow*f (90%).

        Source: EPA 1989C.
  Y.
Blanks provide  an estimate of bias due  to
contamination  introduced  by  sampling,
transportation, carryover during field filtration,
preservation, or storage. At least one field blank
per medium should be collected each day, and at
least one blank must be collected for each sampling
process (EPA 1989f).
Examine results from duplicate and blank samples
.as early as possible in the sampling  operation to
ascertain if presumed sampling characteristics are
accurate and discover areas where the sampling.
strategy requires modification. For a more detailed
discussion of the types and use of QC samples see
A Rationale for the Assessment of Errors in the
Sampling of Soils (EPA 1990c).

Calculate the sample total for stratum or exposure
area (enter in Part II, Step U).
4.1.3  Specific Sampling Issues

Selection of performance measures.  Quantitative
data qua! ity indicators based on performance objectives
should be proposed for completeness, comparability,
representativeness, precision, and accuracy during
planning.   Performance measures are specified as
minimum  limits for each stratum.  Based on the
coefficients of variation of the analyte concentrations,
these limits will determine the numbers of samples
required. The actual values or objectives are determined
by the level of acceptable uncertainty, which includes
that  associated  with  hot  spot  identification.
Recommended minimum criteria are specified in Exhibit
48 for statistical performance measures associated with
the uncertainty  in risk assessment: confidence level,
power, and MDRD. Recommended minimum criteria
for measurement error and completeness for critical
samples are discussed in the following sections.
Setting minimum acceptable limits for confidence
level, power,  and minimum  detectable relative
difference.  Confidence level, power, and MDRD are
three measures of sampling design precision.  These
measures are ultimately determined by the coefficient
of variation of chemical concentration and the number
of samples. Each measure is briefly defined as follows:
   •  Confidence level:  The confidence level  is 100
      minus a,  where a is the percent probability of
      taking action when no action is required (false
      positive).

    •  Power. Power is  100  minus B, where 6 is the
      percent probability of not taking action when
      action is required (false negative).

    •  Minimum detectable relative difference: MDRD
      is the percent difference required between site and
      background concentration levels  before the
      difference can be detected statistically.

  The power and ability to detect differences between site
  concentration levels compared to background levels are
  critical for risk assessment.  Given a CV, the required
  levels of confidence, power, and MDRD significantly
  affect the number of samples. Exhibit 48 illustrates the
  effect when the CV is equal to 25%.

  It is important to note that the number of samples
  required to meet  confidence and power requirements
  will be low if the acceptable MDRD is large; that is, if
  site  contamination is easily discriminated  from
  background levels.

  Determining required precision  of measurement
  error. Field duplicates and blanks  are the major field
  QC  samples  of importance to  the precision of
  measurement error.  Duplicates provide an estimate of
                                                     76

-------
total measurement error variance, including variance
due to sample collection, preparation, analysis, and data
processing. They do not discriminate between-batch
error variance. If the duplicate is collocated, contaminant
sample variation caused by a heterogeneous medium is
also included in the measure.  The precision of the
measurement error estimate is subject to the number of
duplicates on which the estimate is based. Exhibit 49
gives the estimated precision of the measurement error
based on the number of duplicate pairs.  With three
duplicates, the true measurement error variance could
be as much as 13.89 times the observed variance, if a
95% level of confidence is required.  The resources
needed for the collection and analysis of duplicates
depend on the magnitude and variability  of the
concentration of concern for the chemicals of potential
concern.
   • Little room for measurement error exists if the  <
     level of concentration of concern is near the method
    detection limit, and the precision of the estimate of
    measurement error is critical.

  • If the  natural variability of the chemicals of
    potential concern is relatively large, the major
    planning effort will be to collect more samples
    from the exposure areas, rather than collecting
    more QC samples. More detailed discussions of
    the use of QC measures and selection of the
    appropriate number of QC samples may be found
    in A Rationale for the Assessment of Errors in the
    Sampling of Soils (EPA 1990c).

Planning for 100% completeness for criticalsamples.
Certain samples in a sampling plan may be designated
by the RPM or risk assessor as critical in determining
the potential risk for an exposure area. For example, if
only one background sample is taken fora given medium
and exposure area, then that sample would be considered
                      EXHIBIT 49. CONFIDENCE LEVELS FOR THE
                    ASSESSMENT OF MEASUREMENT VARIABILITY
Number of Interval for 95% Confidence that Measurement Error Is Within Limits
Duplicate _
Pair Samples Observed True
Variance (s2) Variance
2
3 '
4
s
6
7
8
9
10
15
20
25
60
100
.27
.32
.36
.39
.42
.44
.46
.47
.49
.54
.58
.62
.70
.77
4
4
4
4
4
4
4
4
4
4
4
4
4
4
o2
o2
o2
o2
o2
o2
o2
o2
o2
o2
f
o2
o2
o2
4
4
4
4
4
4
4
4
4
S
^
£
4

Observed
Variance (s2)
3951
13.89
8.26
6.02
4.84
4.14
3.67
3.33
3.08
2.40
2.08
1.91
1.61
1.35
2
s • Observed variance (precision of an estimate).
2 •
e •True variance (population variance). |
Note:
Source:
Assumes data are or have been transformed to normal distribution. |
EPA 1990C.




                                                  77

-------
"critical."  All data associated with such a sample must
he complete. The only acceptable level of completeness
for critical samples is HK)%.

    *• Focus planning efforts on maximizing
    the collection of useable data from critical
    samples.

Hot spots and the probability of missing a hot spot.
Hut spots are primarily an issue in soil sampling. The
RPM and risk assessor must determine whether hot
spots exist in the exposure area and the probable size of
the hot spot.  This information can often be deduced
from historical data and assisted by judgmental sampling,
although judgmental sampling alone cannot produce
estimates  of the probability that a hot spot has been
missed. Procedures for determining the probability of
missing a hot spot are not as effective in random designs
as in systematic and geostitistical designs. However, a
search strategy which stratifies the area based on grids
and then randomly samples within each grid can be used
within the classical technique.  Systematic and
geostatistical design  approaches  provide  the best
approach to unknown hot spot identification.
Appendix IV describes  numerical procedures and
assumptions to determine the probability that a given
systematic design will detect a hot spot and provides a
calculation formula based on a geometrical approach.
To  employ this formula, the  distance  between grid
 points and the estimated size of the hot spot as a radius
 must be specified.

 Historical data comparability. The RPM may wish to
 assess historical data along with current results or may
 anticipate that the current data will need to be compared
 with results from future sampling activities.  Consult a
 statistician in either of these cases to determine if the
 current sampling design will allow the production of
 dataof known comparability. Factorsolher than statistics
 may need to be considered when attempting to combine
 data  from different sampling  episodes.  Physical
 properties of the site such as weather patterns, rainfall
 juid geologic characteristics of different exposure areas
 may need to be considered. Temporal effects, such as
 the seasonality or time period of sampling, or seasonal
 heightof a water table, may also be important. Analytical
 methods  have been  modified over lime and many
 required detection limits have been revised.

      ••The ability to combine data from different
      sampling episodes or different sampling
      procedures is a very important consideration
      In selecting a sampling design but should
      be done with caution.
4.1.4  Soil Depth Issues

The appropriate depth or depths to take soil samples can
be a major issue in determining a sampling design.
Exhibit SO is a worksheet designed to help the RPM and
risk assessor to determine an appropriate soil sampling
depth. The conceptual site model (Exhibit 6) provides
the basis for completing this worksheet. The nature and
depth of soil horizons at the site should be established
wherever possible. Features such as porosity, humic
content, clay content, pH, and aerobic status often affect
the movement or fate of chemicals of potential concern
through a soil. As with other worksheets provided in
this guidance, this worksheet is intended as a guide or
basis for development. RPMs. in consultation with the
risk assessor and other staff, can revise or modify this
worksheet as appropriate to the  site.  Consider both
current and future land use scenarios in soil exposure
areas because of the sorptive and retentive properties of
soils.

Completing the Soil Depth Sampling Worksheet

1.   Land Use Alternatives

     A.  Identify current or future land use.

     B.  Identify exposure scenario.

         The exposure scenario should be identified for
         current or future land use. Identify the scenario
         according to Role of Baseline Risk Assessment .
         in SuperfundRemedy Selection Decision (EPA
         \99lc) soul Human Health Evaluation Manual
         Supplemental Guidance: Standard Default
         Exposure Factors (EPA 1991d). A residential
         exposure scenario should be used whenever
         there are, or may be, occupied residences on or
         adjacent to the site. Unoccupied sites should
         be  assumed to be residential  in  the future
         unless residential land use is unreasonable.
          Sites that are surrounded by operating industrial
          facilities can be assumed to remain as industrial
          areas unless there is an indication that this
          assumption is not appropriate. Other potential
          land uses, such as recreation and agricultural,
          may be used if appropriate.

  2.   Chemicals of Potential Concern

      A.  Specify class of chemical.

          Circle the classes of chemicals of potential
          concern (e.g., volatile organlcs (VOAs),
          semivolatile organics (semi-VOAs), inorganics
          or metals, or special class) that apply.
                                                     78

-------
EXHIBIT 50. SOIL DEPTH SAMPLING WORKSHEET
Step 1 • Land Use Specifications*
1 A (check one)
Current .
_ Future ;
_ Current & Future, Same
Sampling Depth Considerations
Step 2: Chemicals of Concern
A Class: VOAs. Metals,
semi-VOAs. Special
(e.g., PCBs. dioxin)
B Physical Properties: Mobile.
Soluble, or teachable
Step 3: Soi Characteristics

C Particle Size
0 Concern for Migration to Other
Media. (Air. SW. sediments.
ftlAA

Step 4: Vegetative Cover
Heavy/Sparse/Intermittent
Step 5: Other Factors

»
1B (check one)
_ Residential _ Recreational
Commercial/Industrial _ Agricultural
Other (Specify)
StepS. Expected
Depth of Contamination
by Chemicals of
Potential Concern
Surface Units Subsurface


Step 7. Exposure Pathways
Ingestlon Dermal Inhalation



StepB. Representative
Sample Depths
(units 	 J

* The complexity of a site determines if multiple worksheets are necessary to distinguish between current and future land use scenarios
(e.g.. mix of residential and commercial use for different areas of a site, possible future residential use. etc.).

-------
    B.  Record physical properties.

        Circle the physical properties of the chemicals
        of potential concern that apply.  These
        properties can be estimated from factors such
        as the octanol/watcr  partition coefficient,
        Henry's law constant, and water  solubility
        appropriate to each chemical.

3.  Soil Characteristics

    A.  Record the taxonomic designation of the soil,
        if known.

    B.  Record the organic matter content of the soil.

    C.  Record the most common particle size of the
        soil.

    D.  Identify  any concern for. migration of  the
        chemicals of potential concern to other media
        (e.g.,  air, sediment,  surface water, and
        groundwater).

4.  Vegetative Cover

    Circle whether the vegetative cover of the site is
    heavy, sparse or intermittent.

5.  Other Factors

    List other factors or considerations that influence
    the desired depth of soil sampling.  For example,
    geological factors (e.g., depth to groundwater or
    bedrock) could influence soil sampling.

 6 . Expected Depth of Contamination by Chemicals
    of Potential Concern

     Enter expected depth (and units) of contamination
     by chemicals of potential concern, given the
     chemicals, soil characteristics and vegetative cover.
     Depth can be influenced by disposal practices or
     deposition patterns, soil characteristics, vegetative
     cover, and physical and chemical properties of the
     chemicals of potential concern.

 7.  Exposure Pathways

     Enter exposure pathways by chemicals of potential
     concern, soil characteristics and vegetative cover.
     Physical and chemical properties of the chemicals
     of potential concern will influence their activity in
     theexposure path way(e.g.,VOAs and theinhalation
     pathway). Soil characteristics and vegetative cover
     will also influence the exposure  pathway (e.g.,
     groundwater and water ingestion pathway).
8.  Representative Sample Depths

    Record representative sample depths (including
    units) indicated by the data completed in Steps 2
    through 7.
           Basic Soil Depth Definitions

   Surface dust is the top 0 to 2 inches of soil that can
   be carried by the wind and tracked into houses.

   Surface soil is the top 0 to 6 inches of soil. If the
   surface is-arass covered, surface soil is considered
   the 2 inches below the grass layer.

   Subsurface soil can typically range from 6 inches
   to 6 or more feet in soil depth. Forexample.atsites
   with potential soil moving activity, soil depths
   greater than 6 feel could be of concern in risk
   assessment.
 Other Performance Measures. Other performance
 measures may be designated to facilitate the monitoring
 and assessment of sampling. For example, field spikes
 and Held evaluation or audit samples can be used to
 assess the accuracy and comparability of results. Field
 matrix  spikes are routine samples spiked with the
 contaminant of interest in the field and do not increase
 the number of field samples. Field evaluation samples
 are of known concentration, which are introduced in the
 field at the earliest stage possible and subject to the same
 manipulation as routine samples.   Field  evaluation
 samples will increase the total number of samples
 collected.  Performance measures for field  spikes and
 evaluation samples are expressed in terms of percent
 recovery.  Difficulties associated with field spiking,
 especially  in soil, have resulted in  limited  use of this
 practice (EPA 19890.

 4.1.5  Balancing Issues for Decision-
         Making
 Completing a number of Sampling Design Selection
 Worksheets (Exhibit 45) for different exposure areas,
 media, and sampling design alternatives will enable the
 RPM and risk assessor to compare ande valuate sampling
 design  options and consequences and  select  the
 appropriate sampling design  for  each  medium and
 exjK)sure path way. Practical tradeoffs bet ween response
 time, analytical costs, number of samples, sampling
 costs, and level of uncertainty can then be weighed. For
 example, perhaps more samples can be collected if less
 expensive analyses are used. Or, if the risk assessment
 is based on a point source, collection of additional
  samples  to estimate chemical concentrations and
  distribution can be avoided.
                                                     80

-------
Computer programs are useful tools in developing and
evaluating sampling strategies, especially in trading off
costs against uncertainty, and identifying situations
when additional samples will notsigniflcantly affect the
useability of the data  (i.e., the point of diminishing
returns).  Each automated system has specific data
requirements and is based on specific site assumptions.
The major systems that support environmental sampling
decisions are listed, contacts for information given, and
brief descriptions provided in Exhibit 51.

4.1.6  Documenting Sampling Design
        Decisions

It is important to document the primary issues considered
in balancing tradeoff to accommodate resource concerns
and their impact on data useability. Fully document all
final sampling design decisions, including the rationale
for each decision. During the course of the RI, continue
to document pertinent issues that arise and any sampling
plan modifications which are implemented.


4.2 STRATEGY FOR SELECTING
    ANALYTICAL METHODS

This section describes how to use the Method Selection
Worksheet shown in Exhibit 52 as a data collection and
decision-making tool to guide the selection of analytical
methods that meet the needs of the risk assessment and
to select the most appropriate method for each analyte.
The RPM and risk assessor should consult the project
chemist and use this worksheet in method  selection.
Alternatively, it can be a model to create a worksheet
specifically suited to their needs. Methods selected in
this process may be routine or non-routine.
                    EXHIBIT 51. AUTOMATED SYSTEMS* TO SUPPORT
                                ENVIRONMENTAL SAMPLING
System
Data Qually Objective
(Training) • Expert
System
ESES
Environmental Samplng
(Plan Design) • Expert
System
GEOEAS
Geoslalislicai
Environmental
Assessment Software
SCOUT
MuRlvarlate Statistical
Analysis Package
ASSESS
EPA Contact
Dean Neptune
USEPA
Quafty Assurance
Management Stall
(202) 260-9464
Jell Van Ee
Exposure Assessment Dtv.
USEPA, EMSL-LV
(702) 798-2367
Evan EngKind
Exposure Assessment Dtv.
USEPA. EMSL-W
(702) 798-2248
Jell Van Ee
Exposure Assessment Dtv.
USEPA. EMSL-tV
(702) 798-2367
Jed Van Ee
Exposure Assessment Dtv.
USEPA. EMSU.V
(702) 798-2367
Description I
Training system designed to assist In I
ilannlng ol environmental •
nvesligations based on DQO process. I
Expert system designed to assist In
planning sample colsction. Includes
models that address statistical design.
QC. sampSng procedures, sample
handling, budget, and documentation.
Current system addresses metal
contaminants In a sol matrh. (Expanded
appication under development, contact
EMSL-LV.)
Collection ol software tools lor
two-dimensional geoslalbtical analysis
ol spatially dMrbuted data points.
Programs Include lie management.
contour mapping, krlging, and variogmm
analysis.
A collection ol statistical programs that
accept GEOEAS files lor muHlvarhite
wialysls.
System designed to assist In
assessment ol error In samplng ol soils.
Esllmales measurement error variance
components. Presents scatter pica ol
OC data and error plots to assist In
determining the appropriate amount ol
QC samples.
' AliY^enwwMwnoniriylBMeortvalblePCATwrthimWrnumolMOKRAM. AtteeddUkte


                                                 81

-------
EXHIBIT 52. METHOD SELECTION WORKSHEET
L Analyte*
A.
Chemical or Class of
Chemicals of
Potential Concern

B.
Reporting
Requirement
(YorN)

Y« Total reported for compound class.
.N * Each anatyte reported separately.
II. Medium


III Critical Parameters
A.
Turnaround
Time
(enter hours
or days)

8.
ID Only or
ID Plus
Quant
(lOorlD+Q)

C.
Concen-
tration of
Concern 0
(or PRO)
•
0.
Required
Method
Detection
Limit3

IV. Routine Available Methods4

.

4 Method detection limit should be no greater than 20* of concentration of concern.
Refer lo Appendix III for specific methods. Recommend consultation with chemist and/or automated methods search to determine all methods available.
(Exhibit 53 lists computer systems that support method selection.)

-------
   *•  Ensure that critical requirements and
   priorities are specified on  the Method
   Selection  Worksheet so that the most
   appropriate methods can be considered.

   •   Routine methods arc issued by an organization
       with appropriate responsibility (e.g., state or
       federal agency with regulatory responsibility,
       professional  organization), are validated,
       documented, and  published, and contain
       information on  minimum performance
       characteristics such as detection limit, precision
       and accuracy, and useful range.

   •   Non-routine methods address situations with
       unusual or problematic matrices, low detection
       limits  or new  parameters, procedures or
       techniques; they often contain adjustments to
       routine methods.

   *•  Use routine methods wherever possible
   since  method  development is time-
   consuming and may result in problems with
   laboratory implementation.

4.2.1  Completing the Method
       Selection Worksheet

1. Identify uniilyte*.

   List the chemicals  of potcntiiil concern to risk
   assessment for the site on the Method Selection
    Worksheet.  Use the same list of chemicals that
   appears on  the Sampling  Design Selection
   Worksheets.  Under Column ID, indicate whether
   the concentration for each analyte should be reported
    separately, or the total  for the compound class
   reported.

2.  Identify medium for analysis.

    Specify the analysis medium (e.g., soil, sediment,
    groundwater, surface water, air, biota).

3.  Decide on critical parameter*.

    Specify the required data turnaround lime (I II A) as
    the number of hours or days from the time of
    sample collection.   Indicate  whether  chemical
    identification alone is desired or identification plus
    quantUation (HID).  Specify the concentration of
    concern (I HC) and requ ired detection or quantitat ion
    limit (HID).
                                t
4.  Identify routine available method*.

    Use the final worksheet column, in consultation
    with the project chemist, to list the methods available
    that satisfy the requirements in the preceding steps.
    Reference sources and software ore available to
   assist in identifying routine analytical methods
   applicable for environmental samples (Exhibit 53).
   The most common routine methods for organics
   and inorganics analyses for risk assessment arc
   listed in Appendix III. The methods in the appendix
   are from the following sources:

   •    Contract  Laboratory Program  (CLP)
        Statements of Work for Routine Analytical
        Services (EPA 1990d. EPA 1990e).

   •    Test Methods for  Evaluating Solid Waste
        (SW846): Physical/Chemical Methods (EPA
        1986b),

   •    Standard Methods for the Examination of
        Water and Wastewater (Clesceri. el. a/., eds.
        1989). and

   •    EPA Series  200.  300. 500, 600 and 1600
        Methods (EPA 1983. EPA 1984. EPA 1988d.
        and EPA 1989g).

   Other sources of methods are:

   •    Field Analytical Support Project (FASP) (EPA
        1989h).

   •    Field Screening  Methods Catalog (EPA
        1987b).

   •    Field Analytical Methods Catalog,

   •    ERT Standard Operating Guidelines,

   •    Close Support Analytical Methods,

   •    A CompentliumofSuperfundFieldOperations
        Methods (EPA 1987c).

   •    Association  of Official Analytical Chemists
        (AOAC). and

    •    American Society for Testing and Materials
        (ASTM).

Several computer-assisted search and artificial
intelligence-based  tools are available, including  the
Environmental Monitoring  Methods Index (EMMI),
the Smart Methods Index, and a computerized reference
book on analytical methods. Some of these systems arc
designed as  (caching tools, as well as informational
compendia.  All offer the ability to rapidly search and
compare lists of chemicals and method characteristics
from accepted reference sources.   Exhibit 53 lists
software products that aid method selection, identifies
contacts for information, and gives a short description
of the product.

-------
       EXHIBIT S3. AUTOMATED SYSTEMS*
        TO SUPPORT METHOD SELECTION
System
Environmental
Monlowg
Method* Index
(EMMI)
Smart Methods
Inde*
Geophysical
Technque*
Eipert System
EPA Sampang
and Analyst*
Data Baw
Con feet
W. A. TeMard
USEPA
Oltic* ol Water
(202)260-7120
John Nooenno
Dually Aasuranee Dtv.
USEPA. EMSl-LV
(702I7M-21IO
Atto Meggeta
Advanced Monlorng
Dtv.
USEPA. EMSL-W
(702) 7M-22S4
l*»t*Pv*«*her»
1-400-272-7737
Descriptor)
An automated sorting and
Mltctlon software package thai
curientry conlalnt ovt( 800
methods and over 2600
analytes Itont molt than 80
regulating and non-regulating
Ms. These are ctost-
relerenced to tacitttata Mlectbn
baaed on required needs (e.g.,
anayia detection tmi.
Instrument).
Natural language eipert system
prototype that pr ovldea
Interactive queries ol database*
croM-reterenced by method.
anayte. and performance
feature*.
An tipert system that tugged*
and rank* geophysical
techniques including soil-gas, lor
apcrikabil(y ot use bated on
Me-ipecilc chaiactenXie*.
A three-volume tet * diskette*
and a printed manual provide*
a search ol samclng and
anaryteal method summaries
from a menu-driven program ol
ISO EPA-approved method*.
The database can be Marched
by melhod. analyte. malm, and
various QA consideration*.
' Al system* trtl run on any IBM-comp*)*** PC AT with • mrwnum ol 640* RAM.


4.2.2  Evaluating the Appropriate-
        ness of Routine Methods

    •v  Analvte-specific methods that provide
    better quantitation can be considered for
    use once chemicals of potential concern
    have been identified by a broad spectrum
    analysis.

Choice of the proper method is critical to the acquisition
of useable data.  See Section 3.2 for a more detailed
discussion.  Routine methods provide data of known
quality for the analysis of chemicals and sample types
described in the method. Data quality issues (precision,
accuracy, and interferences) arc usually described in the
melhod.  Consult the project chemist and  examine
available methods with respect to the criteria defined on
the Method Selection Worksheet. It may be helpful to
divide the analyte list into categories based on the types
of analysis. For example, a requirement for chromium,
cadmium, and arsenic data could not be generated by the
same analysis as data for chlorinated hydrocarbons
because of sample extraction and treatment procedures.
 It may be possible to use several methods independently
and combine the data sets for risk assessment purposes.
This is done routinely by the CLP, where inorganics
(elemental analysis),  volatiles,  extractable organics,
and pesticides are analyzed by different methods. In
some cases, no routine method or series of methods will
be able to satisfy all criteria and compromises must be
considered.  The RPM, with the advice of the risk
assessor, must then determine  which  criteria are of
highest priority and which can be modified. For example,
if a low detection limit is of high priority, turnaround
time  and cost of analysis will likely  increase.
Alternatively,  low detection  limit and  precision
requirements may need to be modified if an initial broad
spectrum analysis is of high priority to quickly determine
the largest number of chemicals present at the site.

Turnaround time. Turnaround time is determined by
the available instrumentation, sample  capacity, and
methods requirements.  Turnaround times for  field
analyses can be as short as a few hours, while those for
fixed laboratory analyses include transport lime and
range from several days to several  weeks.  Field
instruments can provide the quickest results, especially
if the data do not go through a formal review process.
However, the  confidence in chemical  identification.
and particularly quantitation, may not be as high.  In
general, methods with quick turnaround limes may be
less precise and have higher detection limits. If data are
needed quickly, a field method can be used for initial
results and a fixed laboratory method used to produce
more detailed results (or confirm the earlier results).
thereby increasing the confidence in field analyses.
Sample quantitation limits.  Risk assessment  often
requires a sample  quantitation limit at or below the
detection limit for routine methods for many chemicals
 of (oxicological concern (see Section 3.2.4).  The sample
 quantitation limits vary according to the size, treatment,
 and analysis of each individual sample. The quantitation
 limits for chemicals in water samples are often far lower
 than for the same chemicals  in soils because of co-
 ex tractable components in the soil. Interferences known
 for the method may hinder acquisition  of data of
 acceptable quality and are more pronounced near the
 method detection limit. Compare documented melhod
 interferences  with site conditions to identify potential
 method problems. Some common sourcesof interference
  in organic and inorganic analyses are summarized In
  Exhibits 54 and 55.   If needed sample  quantitation
  limits cannot be met by available methods, consult the
  project chemist for  the feasibility of detection at the
  desired level in the required sample type.  The chemist
  can help determine if method adaptation can resolve the
  problem, or if a non-routine method of analysis can be
  used.
  Uxcf ul rang*. The useful range of a melhod is the range
  of concentration of chemicals for which precise and
  accurate results can be generated. This range Is analytc-
  speclfic.  The lower end of the useful  range Is the
  method detection limit, often genericnlly referred to as
                                                    84
                                                                                    	-—**--•»	i-frfctMeJirian

-------
      EXHIBIT 54.  COMMON LABORATORY CONTAMINANTS AND
                   INTERFERENCES BY ORGANIC ANALYTE
Contamination
or
Interference
Fat/Oil

Sulfur

Phthalate
Esters

Laboratory
Solvents


Fraction
Extractable
organics,
pesticides, and
PCBs

Extractable organics.
chlorinated and
phosphorus-
containing pesticides

Chlorinated
pesticides. PCBs,
and extractable
organics

Volatile organics
(methylene chloride.
acetone, and
2-butanone)


Matrix
Tissue,
waste.
soils

Sediment.
waste,
soils

All

All

Effects on
Analysis
Increased
detection limit,
decreased
precision/
accuracy
Presence/
absence,
detection limits.
precision/
accuracy
False positive
identification
(pesticides and
extractable
organics) or
positive bias
(pesticides and
extractable
organics)
False positive
identification or
positive bias

Removal /
Action
GPC (all groups), florisil
(pesticides), acid
digestion (PCBs only)



Rorisil, GC-MS
confirmation of identity
(pesticides. PCBs).
evaluation of reagents
and method blanks for
contamination

Confidence in data use I
based on interpretation 1
of blank data 1
* ' H
Source: EPA1986a. I
                                                                                         21-002-0(4
the  "detection limit."  If a lower detection limit is
required, use of a larger sample or smaller fin;il extract
volume can sometimes compensate.  However, any
interfering chemicals lire also concentrated, thereby
producing greater interference effects. Above the useful
range, the response may not be linear and may affect
quiuiiitaUon. This causes inaccurate and/or imprecise
measurements.  Reducing the sample size for analysis
or diluting the extracted material may  bring the
concentration within (he useful range. With Individual
environmental samples, some chemicals are sometimes
present at the low end of the useful range of the method.
while others arc above the useful range. In this situation.
two analyses, at different effective dilutions, ore
necessary to produce accurate and precise data on all
chemicals.   If  detailed  criteria Tor performing and
reporting such actions are not already pan of the
analytical Statement of Work, then the laboratory should
be instructed to notify (he RPM if this situation occurs.
to allow for sufficient time for rcanalysis wilhin ihe
specified holding lime. All relevant analyses should be
reported to maximize the useability of both detected and
non-detected analytcs.

    +• All results should be reported tor samples
    analyzed at more (nan one dilution.

Precision and accuracy. Routine methods often specify
precision and accuracy with respect to specific anulytcs
(chemicals) and matrices (sample media). I lowcver. be
aware that environmental samples arc often difficult to
analyze because of the complexity of the matrix IK the
                                                85

-------
EXHIBIT 55. COMMON LABORATORY CONTAMINANTS AND
      INTERFERENCES BY INORGANIC ANALYTE
Analyte
Arsenic
Beryllium
Cadmium
Chromium
Lead
Mercury
Selenium
Cyanide
Technique
GFAA
ICP
ICP
GFAA
ICP
GFAA
ICP
GFAA
ICP
CVAA
GFAA
ICP
Cotorimetrio/
•pectrophotometric
Interference
Iron, Aluminum
Aluminum
Titanium, Vanadium
None except possible
sample matrix effects
Iron
Calcium
Iron, Manganese
Sulfate
Aluminum
Sullide. High Chloride
Iron, Aluminum
Aluminum
Acids, Sulfide.
Chlorine oxidizing
agents
Removal/
Action
Background correction
(not deuterium) (Zeeman).
If above 100 ppm,
correction factor utilized.
If above 100 ppm,
correction factor utilized.
Background correction
for matrix effects.
If above 100 ppm,
correction factor utilized. I




Add calcium, standardize •
suppression, background 1
correction. •
If above 100 ppm, •
correction factor utilized. •
Lanthanum nitrate •
addition as matrix I
modifier, background •
correction. 1
If above 100 ppm, 1
correction factor utilized. •
Remove interferences with 1
cadmium carbonate •
(removes sulfide). I
potassium permanganate I
(removes chloride), excess •
hydroxylamine sullate •
(removes free chlorine). 1
Alternate wavelength lor •
analysis, background •
correction (not deuterium) •
(Zeeman). •
Above 100 ppm, •
correction factor utilized. •
Increase pH to > 12 In field to I
remove acids, cadmium •
carbonate (removes sulfide), I
ascorbic acid (removes tree 1
chlorine). •
Key: ICP • Inductively coupled plasma. I
GFAA • Graphite furnace atomic absorption. I
CVAA • CoW vapor atomic absorption. |
                                                  I1-OM-M44I
                         86

-------
presence of a large numberof contaminants; this usually
results in lower levels of precision and accuracy than
those cited in the method.

4.2.3  Developing Alternatives When
        Routine Methods are not
        Available

If routine methods are not available to suit the parameters
of interest, it is often due to one or more of the following
factors:

  •  The  detection limit of  commonly  available
     instrumentation has been reached, and a lower
     detection limit is required for the risk assessment,

  •  An unusual combination of chemicals are  of
     potential  concern,

  •  The sample matrix is complex, and

  •  The chemicals  of potential concern or other
     analytical parameters are unique to a particular
     site.

 Consult an analytical chemist for specific guidance on
 the potential limitations of alternative approaches. These
 may include adaptation of a routine method or use of a
 non-routine method. Be aware that certain conditions,
 such is extremely low detection  limits for some
 chemicals, may be beyond  the capability  of current
 analytic al technology. Turnaround times and costs may
 also be increased.

 Adaptation of routine methods. Adapting routine
 methods may be a solution when routine methods will
 not  provide the desired data even after compromises
 have been made  with respect to parameters such as
 turnaround time and cost. Using the completed Method
 Selection Worksheet as the starting point, work closely
 with an  analytical chemist  to  formulate  suitable
 modifications to the routine method.  Evaluate and
 document any effects on data quality (hat will result
 from the modifications.
 Within the CLP. such analyses can  be obtained by
 special analytical requests.  Before analysis of site
 samples, it is advisable to confirm a laboratory's ability
 to perform the adapted method with preliminary  data.
 Use of non-routine methods.  Existing non-routine
 methods  that  meet criteria can  be  used if a routine
 method cannot be adapted to provide the necessary data.
 Such analyses can be found in the research literature.
 usually catalojvd by analytc or instrument. On-line
 computerized search services can be of considerable
 help in identifying such methods. Work interactively
 with an analytical chemist in reviewing selected methods.
Recognize that non-routine analyses require u greater
level of capability and experience from the analytical
laboratory,  and that turnaround lime can be longer
because the method may need alteration during analysis
if problems develop.
Development  of new methods.  Developing new
methods should be the option of last resort. The KPM,
risk assessor, and project chemist should consider
recommending the development of new methods only
for chemicals of substantial potential concern that cannot
currently be analyzed at appropriate limits of detection.
Although designing a method based on data available
for a given instrument and  analytes  may seem
straightforward,  the process  is time-consuming  and
expensive.  Unforeseen problems can often arise when
the method is implemented in the laboratory.  Problems
can occureven when laboratory personnel have superior
training and experience. Consider the following points
when requesting the development of a new method:
   •  If possible, select a laboratory with a recognized
     reputation for performance and flexibility  in a
     related area. Treat laboratory personnel as partners
     in the development process. This is true whether
     a commercial or a government laboratory is used.

  . •  Identify sources  for authentic standards of the
     chemicals in question  to support method
     development.  Computerized databases such as
     the EPA EMMI (see Exhibit 53) may be useful for
     such a determination.

   •  Be aware that turnaround time for useable data
     may be long (potentially several months) because
     of the likelihood of trying different approaches
     before discovering an acceptable procedure.

 4.2.4 Selecting Analytical Labora-
        tories

 In selecting a laboratory to produce analytical data for
 risk assessment purposes, identify and  evaluate the
 following laboratory qualifications:

   •  Possession of appropriate instrumentation and
      trained personnel to perform the required analyses.
      as defined in the analytical specificatioas,

   •  Experience in performing the  same or similar
      analyses.

   •  Performance evaluation results from formal
      monitoring or accreditation programs,

   •  Adequate laboratory capacity  to perform  all
      analyses in Uie desired timcfnune,
                                                   87

-------
  •  Intni-laboratory QC review of all generated data,
     i:'-V vx'itdent of the data generators, and

  •  Adequate laboratory protocols for method
     performance documentation and sample security.

For non-routine analyses, the laboratory should have
highly trained personnel and instrumentation not
dedicated to production work, especially if new methods
or untested modifications are requested.

Accreditation programs monitor the level of quality of
laboratory performance within the scope of their charters.
Many of these  programs periodically provide
performance evaluation samples that the laboratories
must analyze within certain limits in order to maintain
their status.  Prior to laboratory selection, request that
laboratories provide information about their performance
in accreditation programs.  This information can be
used for evaluation of laboratory quality, in the case of
similar matrices and analytes. Laboratory adherence to
standards of performance such as the Good Laboratory
Practices Standards (Annual Book ofASTM Standards)
also provides a measure of laboratory quality.

4.2.5  Writing the Analysis Request

Include the following items in the analysis request:

   •  A  clear, complete description of  the  sample
      preparation, extraction, and analysts procedures
      including detailed performance specifications. For
      adaptation of routine methods, specify the routine
      method and explicitly state  alterations with
      applicable references.

   •  Documented reporting requirements.

   •  Laboratory access to required authentic chemical
      standards.

   • A mechanism for the laboratory  to obtain EPA
      technical assistance in implementing method
      modifications or performing non-routine methods.

 If the analysis request is for a non-routine method,
 reference the published material with a  detailed
 specification of procedures and requirements prepared
 by  the analytical chemist who has been working with
 the RPM and risk assessor.  The  specification must
 include the frequency, acceptance criteria, and corrective
 action requirements for each of the following:

    •  Instrument standardization, including tuning and
       initial and continuing calibration.
  • QC check samples such as surrogate compound
    and internal standard recoveries,

  • Method blank perform;ince (permissible level of
    contamination),

  • Spike sample recovery requirements,

  • Duplicate analysis requirements, and

  • Performance evaluation or QC sample results.

Allow time for the laboratory to review the analysis
request and question any part of the description that
seems unclear or unworkable according to its experience
with the analytes or sample matrix. Preliminary data,
such as precision and accuracy data on a subset of the
analytes, can be requested to determine if the laboratory
can implement the proposed method. Should the criteria
not be met in the preliminary analyses, the analytical
chemist should advise  the laboratory  on additional
method modifications to produce the required data. In
some cases, even qualitative data can be used to note the
presence of chemicals of potential concern.

In all cases,  require  the  laboratory performing the
analyses to contact the project chemist at the first sign
of a problem that may affect data quality. The RPM and
the site technical team can then judge the magnitude of
the problem and determine appropriate corrective action.


4.3 BALANCING  ISSUES FOR
     DECISION-MAKING

Resource issues. Resource limitations are a major
reason for sampling design modification. The number
of samples required to achieve desired performance
measures may exceed resource availability. Modifying
the sampling design  and  the efficiency of statistical
estimators can reduce sample size and costs, and improve
overall timeliness for the risk assessment. Analytical
 methods such as field analyses may also reduce cost.
 Systematic and geostatistical sampling designs can
 often achieve the required performance measures with
 fewer samples than classical random sampling (Gilbert
 1987).   Pilot sampling can be used to verify  initial
 assumptions of the  SAP,  increase  knowledge of
 contaminant distribution, and support SAP modifications
 to reduce  the number of samples.  Explain resource
 issues and record potential design modifications in
 documentation developed during planning.
 Completing a number of Sampling Design Selection
 Worksheets (Exhibit 45) for different exposure areas,
                                                    88

-------
media, and sampling design alternatives will enable the
RPM and risk assessor to compare and evaluate sampling
design options  and consequences and  select the
appropriate  sampling  design  for each medium and
exposure pathway.

Computer programs are useful tools in developing and
evaluating sampling strategies, especially in trading off
costs against uncertainty, and identifying situations
when additional samples will not significantly affect the
useability of the data (i.e., the point of diminishing
returns).  Each  automated system has  specific data
requirements and is based on specific site assumptions.
The major systems that support environmental sampling
decisions are listed, contacts for information given, and
brief descriptions provided in Exhibit SI.

Documenting design decisions.  It is important  Co
document the primary issues considered in balancing
tradeoffs to accommodate resource concerns and their
impact on data useability. Several compromises among
options are discussed in this section.   Features  of
analytical options available for organic and inorganic
analytes are summarized in Exhibits 56 through 59.
Fully document all final sampling and analytical design
decisions, including the rationale for each decision.
During the  course of the RI,  continue to document
pertinent issues  that arise and any plan modifications
which are implemented.
The goal of balancing issues in the selection of analytical
methods is  to obtain the best analytical performance
without sacrificing risk assessment requirements. The
selection of analytical methods often involves tradeoffs
among the required detection limit, number of analytes
involved, precision and-accuracy, turnaround time, and
cost Some choices may conflict with others.

Cost should be considered only after the most appropriate
methods have been determined.  Methods requiring
specialized instrumentation, such as high resolution
mass spcctrometry, will be more expensive. Methods
for use on matrices such as soil, can be more expensive
than similar methods for a simpler matrix such as water.
Less expensive methods often have  higher detection
limits and less specific confirmation of identification.
However, the turnaround times are often quicker and a
larger number of samples can be analyzed. This often
significantly increases sampling precision and reduces
the probability of missing hot spots. Less expensive
methods are often chosen if the site has already been
characterized by broad spectrum analyses. In evaluating
routine  methods, consider whether analysis of more
samples through use of less expensive  methods can
provide a similar level of data quality to that achieved
through the use of more expensive methods on fewer
samples. By remaining aware of the effect of individual
issues on the data quality, the RPM can determine the
optimum choices.

    ••  Field analysis can be used to decrease
    cost and turnaround time,  providing data
    from a broad spectrum analysis are
    available.

In addition to turnaround time tor analysis, time must
also be scheduled for data review. This will not hinder
the availability of  laboratory  and field data for
preliminary  use if a tiered data review sequence  is
incorporated.

When using the tiered approach, consider the use of split
samples (i.e., sending sample splits for analysis by field
and fixed laboratories).  Quantitative comparison can
then be made between the precision and accuracy of the
field analyses and  those of the fixed  laboratory.
Confirmation of identification by both field and fixed
laboratories also increases data  confidence and
useability. It is recommended that field methods should
be used with at least a 10% rate of confirmation  or
comparison by fixed laboratory analyses.
                                                    89

-------
           EXHIBIT 56. COMPARISON OF ANALYTICAL OPTIONS
                   FOR ORGANIC ANALYTES IN WATER
   Method
MDL
Quantitative
Confidence
Timeliness
Precision &
 Accuracy
FIELD SCREEN/FIELD ANALYSIS (Assumes preparation step)
   GC(PCB)
   GC (Pesticides)
   GC (VOA)
   G C (Soil Gas)
   GC (BNA)
   PHOTO VAC
   Detector
FIXED LABORATORY

   CLP RAS
    VOA
    BNA
    Pesticides
    Dioxin

   CLP LOW CONC
    GC
    VOA
    BNA
   500 SERIES
    GC
    VOA
    BNA

   600 SERIES
    GC
    VOA
    BNA

   SW846
    GC
    VOA
    BNA
                                                                      21-002-045
                                     90

-------
EXHIBIT 57. COMPARISON OF ANALYTICAL OPTIONS
        FOR ORGANIC ANALYTES IN SOIL
Quantitative
Method MDL Confidence Timeliness
FIXED LABORATORY
CLP RAS
VOA V
BNA V
Pesticides
Dioxin (2,3.7.8 TCDD) V
SW846
GC V
VOA V
BNA V
1600 SERIES
GC V
VOA V
BNA V
Dioxin V
FIELD SCREEN
GC(PCB) V V V
GC(Pesticides) V
GC(VOA) V V
GC(Soil Gas) V V
GC(BNA) V V
PHOTO VAC
Detector V
Precision &
Accuracy Comparability

V
i
V V
V
V
V
V V
V V
V V
V V
v
V
V
V
Key: V * Method strength
                                                   21402-05S41
                      91

-------
               EXHIBIT 58. COMPARISON OF ANALYTICAL OPTIONS
                 FOR INORGANIC ANALYTES IN WATER AND SOIL
                      Quantitative
                      Confidence   Timeliness
Precision &
 Accuracy    Comparability 2
FIXED LABORATORY
 CLP RAS
   ICP
   GFAA          V
   Flame AA
  200 Series
   GFAA
   AA
   ICP-MS3
   ICP-Hydride

FIELD SCREEN
   XRF
 Key: \ = Method strength
  CLP Inorganic water assays are more accurate and precise than soil assays.
  ICP and GFAA are comparable at medium to high ppb levels. For As, Pb, Se, Tl and So at less than
  20 ppb, GFAA is the method of choice.
  ICP-MS and ICP-Hydride methods are relatively new; therefore, precision, accuracy, and comparability
  estimates based on large statistical sampling are not available.
                                           92
                                                                                 21-OOZ-OM-OZ

-------
          EXHIBIT 59. COMPARISON OF ANALYTICAL OPTIONS* FOR
                 ORGANIC AND INORGANIC ANALYTES IN AIR
                                             Precision &
                                              Accuracy    Comparability
        Quantitative
        Confidence
FIXED LABORATORY
 CLP VOA
  Cannister
  Tenax
2-5 ppb
2-30 ppb
(for most)
             0.00001-
             0.001 ug/m3
  Key:  V » Method strength
  The methods described are new Statements of Work.
                                        93
                                                                          ti-oot-wtoa

-------
                                       Chapter  5
    Assessment of Environmental  Data  for Useability  in
                          Baseline Risk Assessments
This chapter provides guidance for the assessment and
interpretation of environmental data for use in baseline
human health risk  assessments.   Ecological risk
assessments follow a similar logic but may differ in
some details of sampling and analytical methodologies
and minimum data requirements. The discussion of
data assessment is presented as six steps that define the
assessment process for each data useability criterion.
Exhibit 60 lists the six criteria in the order that a risk
assessor would evaluate them. It also gives references
to the sections in this chapter where they are further
discussed.

       EXHIBIT 60. DATA USEABILITY
         ASSESSMENT OF CRITERIA
                  CRITERION I

                 Reports to Risk
                   Assessor
                     (5.1)
                     I
                  CRITERION II
                  Documentation
                     (5.2)
                  CRITERION III

                  Data Sources
                      (5.3)
                      1
                  CRITERION IV

                Analytical Method and
                  Detection Unit
                      (5.4)
                   CRITERION V

                   Data Review
                      (5.5)
                      ±
                   CRITERION VI

                   Data Quality
                    Indicators
                      (5.6)
The four basic decisions to be made from data collected
in the RI arc:

  •  What contamination Is present and at wh'at levels?

  •  Are site concentrations sufficiently different from
     background?

  •  Are all exposure pathways and exposure areas
     identified and examined?

  •  Are a^-exposure areas fully characterized?

The  uncertainty associated with each data useability
criterion affects the level of confidence associated with
each of these decisions.

How to conduct the data assessment Theriskassessor
or RPM examines the data, documentation, and reports
for each assessment criterion (I - VI) to determine if
performance is within the limits specified in the planning
objectives.   The data assessment  process for each
criterion should be conducted according to the step-by-
step procedures discussed in  this chapter.  Minimum
requirements  are listed for each criterion. Potential
effects of not meeting the minimum requirements are
also discussed and corrective action options  are
presented. Exhibit 61 summarizes the major impact on
assessment if the minimum requirements associated
with each data useability criterion have not been met.
                 Acronyms

  CLP     Contract Laboratory Program
  CV      coefficient of variation
  CRDL   contract required detection limit
  CRQL   contract required quanlitation limit
  DQO    data quality objective
  GC      gas chromatography
  ICP     inductively coupled plasma
  MDL    method detection limit
  MS      mass spcctromelry
  QA      quality assurance
  QC      quality control
  RAGS   Risk Assessment Guidance for Superfund
  RI      remedial investigation
  RME    reasonable maximum exposure
  RPD    relative percent difference
  RPM    remedial project manager
  SAP    sampling and analysis plan
  SOP    standard operating procedure
  SQL    sample quantitation limit
                                                 95

-------
    EXHIBIT 61. MINIMUM REQUIREMENTS, IMPACT IF NOT MET, AND
         CORRECTIVE ACTIONS FOR DATA USEABILITY CRITERIA
 Data UseabUlty
    Criterion
        Minimum
     Requirement
    Impact on Rlf k
Assessment If Criterion
        Not Met
     Correct) v«
       Action
5.1  Reports to Risk
    Assessor
 Site description
 Sampling design with
 sample locations
 Analytical method and
 detection liml
 Results on per-sampte basis.
 quaifled tor analytical
 limitations
 Sample quantHation limits and
 detection limits for non-
 detects
 Field conditions lor media
 and environment
 Preliminary reports
 Meteorological data
 Field reports
•  Unable to perform
   quanttaltve risk
   assessment
•  Request missing
   information
•  Perform qualitative
   risk assessment
 5.2 Documentation
 Sample resuls related to
 geographic location
 (chain-of-custody records.
 SOP*, field and analytical
 records)
 •  Unable to assess
   exposure pathways
 •  Unable to Identify
   appropriate
   concentration for
   exposure areas
                                                                          •  Request locations
                                                                             identified
                                                                          •  Resampling
 5.3  Data Sources
 Analytical data results for
 one sample per medium
 per exposure pathway
 Broad spectrum analysis for
 one sample per medium
 per exposure pathway
 Field measurements data
 for metja and environment
 • Potential for false
   negatives or false
   positives
 • Increased variabifty hi
   exposure modeling
   Resampling or
   re analysis for
   critical samples
 5.4 Analytical
     Method and
     Detection Umt
  Routine (federally
  documented) methods used
  to analyze chemicals of
  potential concern in crlical
  samples
 • Unqualified precision
   and accuracy
 • False negatives
    Reanalysis
    Resampling or
    reanarysis for critical
    samples
    Documented
    statements of
    limitation for non-
    critical samples
 5.5 Data Review
• DeCned level of data review
  (oral data
  5.6 Data Quality
     tndkjftors
• SampBng vanab»y
  quantified for each anaryte
• QC samples to Identify and
  quantify precision and
  accuracy
• Sampfingand
  analytical precision and
  accuracy quantified
    Potential for fats*
    negatives or falsa
    positives
    Increased variability and
    bias due to analytical
    process, calculation
    errors or transcription
    Perform data
    review
    Unable to quantify
    contdence levels for
    uncertainty
    Potential tor false
    negatives or false
    positives
    Rexamptngfor
    critical samples
    Perform qualitative
    risk assessment
    Perform
    quantitative
    risk assessment
    for non-critical
    samples with
    documented
    dbcuseionof
    potential Imitations
                                              96

-------
The following activities should be performed for each
assessment criterion:

   •  Identify or determine performance objectives and
     minimum data requirements.

     Quantitative or qualitative performance objectives
     should be specified in the sampling and analysis
     plan for all components of the acquisition of
     environmental data (as discussed in Chapter 4).
     The first step in assessing each criterion is to
     assemble these performance objectives and note
     any changes. Performance objectives should also
     be compared with  the  minimum acceptable
     requirements for data useability presented in this
     chapter.  These minimum requirements can be
     adopted as performance objectives if objectives
     were not specified. For example, the requirement
     that there must be a broad spectrum analysis for at
     least one sample in each medium for each exposure
     area would be  a performance objective, if
     performance were not specified during planning.

   •  Determine actual performance compared to
     performance objectives.

     The next step in the assessment of each criterion is
     to examine results to determine the performance
     that was achieved for each data useability criterion.
     This performance should then be compared with
     the objectives established during planning. Take
     particular note of performance for samples or
     analyses that are critical to the baseline risk
     assessment  All deviations from the objectives
     .should be noted. In those cases where performance
     was better than that  required in the objective, it
     may be useful for assessment of future activities to
     determine if this  is  due to unanticipated
     characteristics of the si te or to superior performance
     in some stage of the data acquisition. Corrective
     action Is the next step where performance does not
     meet performance objectives for data critical to
     the risk assessment.

   • Determine and execute any corrective action
     required.

     *• Foots corrective action on maximizing
     the useabilityof data from critical samples.

 Corrective action should be taken to improve data
 useability when performance fails to meet objectives
 for data critical u. 'herisk assessment Corrective action
• options are  described in Exhibit  62.  These options
 require communication among the risk  assessor, the
 RPM, and the technical team. Sensitivity analysis may
 be performed by the risk assessor to estimate the effects
of not meeting performance requirements given the
certainty of the risk assessment. Corrective actions may
improve data quality and reduce uncertainty, and may
eliminate the need to qualify or reject data.

     EXHIBIT 62.  CORRECTIVE
   ACTION OPTIONS WHEN DATA
   DO NOT MEET PERFORMANCE
              OBJECTIVES
     •   Retrieve missing information.

     •   Resolve technical or procedural
         problems by requesting additional
         explanation or clarification from the
         technical team.

     •   Request reanalysis of sample(s)
         from extract.

     •   Request construction and
         re-interpretation of analytical results
         from the laboratory or the project
         chemist.

     •   Request additional sample
         collection and analysis for site or
         background characterization.

     •   Model potential impact on risk
         assessment uncertainty using
         sensitivity analysis to determine
         range of effect.

     •   Adjust or impute data based on
         approved default options and
         imputation routines.

     •   Qualify or reject  data for use in risk
         assessment.
                                      21-002-062
 Using a worksheet to organize the data assessment.
 The level of certaintyassociatcd with the data component
 of risk assessment depends on the amount of data that
 meet performance  objectives.   The risk assessor
 determines whether  the data for each  performance
 measure are satisfactory (data accepted), questionable
 (data qualified) or unsatisfactory (data rejected). The
 worksheet provided in this chapter may be used as a
 guide or organizational tool.

 Use the Data Useability Worksheet, Exhibit 63, to
 document data assessment  decisions.   Record the
 decision as accepted, accepted with qualification, or
 rejected for use in the risk assessment for each data
                                                  97

-------
               EXHIBIT 63. DATA USEABILITY WORKSHEET
Data Useability Criterion
   Reports to Risk Assessor
   Documentation
   A.WorkPlan/SAP/QAPP
    C. Field and
      Analytical Records
    Data Sources

    A. Analytical
    B. Non-analytical
IV  Analytical Methods
 V   Data Review
 Decision: Accept. Qualified Accept, Reject
                                             98

-------
                   EXHIBIT 63.  DATA USEABILITY WORKSHEET
                                           (Cont'd)
Data Useability Criterion
VI
Data Quality Indicators
A. Completeness
B. Comparability
C. Representativeness
D. Precision
E. Accuracy ~*
Decision
Sampling
Analytical
Combined
Sampling
Analytical
Combined
Sampling
Analytical
Combined
Sampling
Analytical
Combined
Sampling
Analytical
Combined















Comments





• fl
Decision: Accept, Qualified Accept, Reject •
                                                                                         21-002-M341
useability criterion.  Outline the justification for each
decision in the comments section.

The remainder of this chapter explains how to assess
data using the data useability criteria. Assessment of
Criterion I involves identifying the data and
documentation required for risk assessment (Section
5.1). Assessment of Criteria II through V examines
available data and results in terms of the assessment of
data useability criteria for documentation (Section 5.2),
data sources (Section 5.3), analytical method and
detection limit (Section 5.4), and data review (Section
5.5). Criterion VI includes the assessment of sampling
and analytical performance (Section 5.6) according to
five data  quality indicators:   completeness,
comparability, representativeness, precision, and
accuracy.
                                                99

-------
5.1  ASSESSMENT OF CRITERION I:
     REPORTS TO RISK ASSESSOR

     Minimum Requirements

   •  Site description.

   •  Sampling  design with sample  locations.
     related to site-specific data needs and data
     quality objectives.

   •  Analytical method and detection limit.

   •  Results on per-sample basis qualified for
     analytical limitations.

   •  Sample quantitation limits and detection
     limits for non-detects.

   •  Field conditions for media and environment.

   •  Preliminary reports.

   • Meteorological data.

   • Field reports.	

 Data and documentation supplied to  the risk assessor
 must beevaluatedfor completeness and appropriateness.
 and to determine if any changes were made to the work
 plan or the sampling and analysis plan (SAP) during the
 course of the work. The SAP discusses the sampling
 and analytical design and contains the quality assurance
 project plan and data quality objectives (DQOs), if they
 have been developed.  The risk assessor should receive
 preliminary and final  data reports, as described in the
 following sections.

 5.1.1   Preliminary Reports

     *•  Use preliminary data as  a basis for
     identifying sampling or analysis deficiencies
     and taking corrective action.
  Preliminary analytical data reports allow the risk assessor
  to begin assessment as soon as the sampling and analysis
  effort has begun.  These initial reports have three
  functions:

    •  The risk assessor can begin to characterize the
       baseline risk  assessment on the basis of actual
       data. Chemicals of interest will be identified and
       the variability in concentration can be estimated.

    •  Potential problems in sampling or analysis can be
       identified and the need for corrective action can be
       assessed. Forcxamplc, additional samples maybe
       required, or the method may need to be modified
       because of matrix interferences.
  •  RI schedules are more likely to be met if the risk
     assessment process can begin before the final data
     reports are produced.

The major advantage of preliminary review of data by
the  risk assessor is the  potential  for feedback and
corrective action while the RI is still in process. This
can improve the quality of data for risk assessment.

5.1.2  Final Report

    «*•  Problems in data useability due to sam-
    pling usually can affect all  chemicals
    Involved rn the risk assessment; problems
    due to analysis may only affect specific
    chemicals.

The minimum data reports and documentation needed
to prepare the risk assessment are:

   • A description of the site, including a detailed map
     showing the location of each sample, surrounding
     structures, terrain features, receptor populations,
     indications of air and water flow, and a description
     of the operative industrial process (if any),

   • A description and rationale for the sampling design
     and sampling procedures,

   • A description of the analytical methods used,

   • Results for each analyte and each sample, qualified
     for analytical limitations, and a full description of
     all deviations from SOPs, S APs, and QA plans,

    •  Sample quantitation limits (SQLs) and detection
      limits for undetected analytes. with an explanation
      of  the detection limits reported and  any
      qualifications,

    •  A narrative explanation of the level of data review
      used and the resulting dataqualifiers. Thenarrative
      should indicate the direction of bias, based on the
      assessment of the results from QC samples (e.g.,
      blanks and field and laboratory spikes), and

    • A description  of field conditions and physical
      parameter data as  appropriate for  the media
      involved in the exposure assessment.

  It may not be possible to perform a quantitative baseline
  risk assessment if any of these materials are not available
  and cannot be obtained.  The RPM or risk assessor
  should attempt to retrieve missing delivcrables from the
  source.

  Additional reports and data that are useful to the risk
  assessor, such as data results on Contract Laboratory
  Program (CLP) diskettes, ore listed InExhibit 19. Access
                                                    100

-------
to this information can improve the efficiency and
quality of the risk assessment. However, not having
access does not necessarily require the data to be qualified
or rejected. Minimum requirements for reports to the
risk assessor are listed in Exhibit 61.
5.2 ASSESSMENT OF CRITERION II:
     DOCUMENTATION

     Minimum Requirements

   •  Sample results related to geographic location
     (chain-of-cusuxly records. SOPs, field and
     analytical records).                     	

Three types of documentation must be assessed: chain-
of-custody records. SOPs. and  field  and analytical
records. Chain-of-custody records for risk assessment
must document the sample locations and the date of
sampling  so  that sample results can be  related to
geographic location and specific sample containers. If a
sample result cannot be related to a sampling date and
the point of sample collection, the results are unuseable
for quantitative risk assessment. Full scale chainof-
custody procedures (from sample collection  through
analysis) are required for enforcement or cost recovery.

SOPs describe and specify the procedures to be followed
during sampling and analysis. They are QA procedures
that increase the probability that a data collection design
will be properly  implemented.  SOPs also  increase
consistency  in performing tasks and,  as a result,
determine the level of systematic error md reduce the
random error associated with sampling and analysis.
Knowledge that SOPs were developed and followed
increases  confidence that the quality of data can be
determined, and the level of certainty in risk assessment
can be established. The existence of SOPs  for each
process or activity involved in data collection is  not a
minimum requirement, but SOPs can be useful if data
problems occur,  particularly in assessing (he
comparability of data sets.

Field and analytical records document the procedures
 followed  and the  conditions of the procedure.  Field
and analytical  records,  such as field logs and raw
 instrument output, may be useful to the risk assessor as
back-up documentation, but they are not minimum
requirements. QC data from blanks, spikes, duplicates,
replicates, and standards should also be accessible, in
either raw or summary formats, to support qualitative or
quantitative assessments of the analytical results. Like
SOPs, such records are critical to resolving problems in
interpretation, but they may not directly affect the level
of certainty  of  the  risk assessment.   Minimum
requirements for  documentation are listed in Exhibit
61.
5.3 ASSESSMENT OF CRITERION III:
     DATA SOURCES

     Minimum Requirements

  •  Analytical sample  data  results  for each
     medium within an exposure area.

  •  Broad spectrum analysis for one sample per
     medium per exposure area.

  •  Field measurements data for media and
     environment.

Data source assessment involves the evaluation and use
of historical and current analytical data.  Historical
analytical data should be evaluated according to data
quality  indicators  and  not source (e.g.,  analytical
protocols may have changed significantly over time).

The minimum  analytical data requirement for risk
assessment is that results are produced for each medium
within an exposure area using a broad spectrum analytical
technique, such as GC-MS methods for organic analytes
or ICP for inorganic analytes.  The useability of data
will almost always increase as more broad spectrum
analyses are performed for each exposure area.  The
absence of a broad spectrum  analysis from a fixed
laboratory results in an increased  probability of false
negatives; all chemicals of potential concern at the site
may not be identified.  In the absence of a broad
spectrum analysis, the best corrective action is to take
additional samples. If additional samples cannot be
obtained, the probability of false  negatives and faLe
positives should be considered high, and the level of
certainty of the risk assessment is decreased.

The broad spectrum analysis, and  any other analytical
data, are subject to the basic documentation and data
review  requirements discussed in this chapter. The
location of the sample data point must be known, as well
as the method and SQL achieved for analytical results.
Guidance for the  assessment of analytical data to
determine false positives and  false negatives and the
precision and  accuracy of concentration  results  is
provided in Section 5.6.1.

Field measurements of physical characteristics of the
site, medium, or contamination source are a critical data
source, whose omission can  significantly  affect the
ability of the risk  assessor to perform a quantitative
assessment Physical site information is also required to
 performcxposure fate and transport modeling. Examples
                                                  101

-------
of such data arc p;uliclc size, pi I, clay content and
poros ity of soils, wind direction and speed, topography.
and percent vegetation. RAGS, Part A. Exhibit 4-2.
"Examples  of Modeling Parameters  for  Which
Information May Need to be Obtained During a Site
Sampling Investigation," (EPA 1989a) provides a list of
data elements according to medium modeling category.
These measurements must be collected during sampling.
The use of default options and routines to estimate
missing values allows the use of the model but increases
the uncertainty associated with the exposure assessments.


5.4 ASSESSMENT OF CRITERION IV:
     ANALYTICAL METHOD AND
     DETECTION LIMIT

     Minimum Requirements

   • Routine  (federally documented) methods
     used to analyze chemicals of potential
     concern in critical samples.	

 The risk assessor compares SQLs or method detr-tioi:
 limits (MDLs) with analyte-specific results to determine
 their consequence given the concentration of concern.
 Assessment of preliminary data  reports provides an
 opportunity to review the detection limits early  and
 resolve any problems. When a chemical of potential
 concern is reported as not detected, the result can only
 be used with confidence if the quantitalion limits reported
 are lower than the corresponding concentration of
 concern. The minimum recommended requirement is
 that the MDL be no more than 20% of the concentration
 of concern,  so that the SQL will also be below the
 concentration of concern. Chemicals identified above
 this ratio of detection limit to concentration of concern
 can be used with good confidence. For example, if the
 concentration of concern for arsenic in groundwater is
  70 ug/L for an average daily consumption of 2  L of
  water by a 70 kg adult, the detection limit of a suitable
  method for examination of groundwater samples  from
  such a site should be no greater than 14 ug/L. Minimum
  requirements for analytical methods and detection limits
  are listed in Exhibit 61.

  If the concentration of concern is less than or equal to the
  detection limit, and the  chemical of concern is nol
  detected, do not use zero  in the calculation of the
  concentration term.  When the MDL reported for an
  analytc is near to the concentration of concern,  the
  confidence in both identification and quantisation may
  be low. This is illustrated in Exhibit 64.  Information
  concerning non-dciects or detections at or near detection
limits should be qualified according to the degree of
acceptable uncertainty, as described in Section 5.6.1.

The concentration of concern for ecological risk may be
different than the concentration of concern for human
health risk. In addition, aquatic life criteria should be
examined to determine if they are based on ecological
or human health risk.


5.5 ASSESSMENT OF CRITERION V:
     DATA REVIEW

     Minimum Requirements

   »  Defined level of data review for all data.	

Data review assesses the quality of analytical results
and is performed by a professional with a knowledge of
the analytical procedures.  The requirement for risk
assessment is that only data that have been reviewed
according to a specified level or plan will be used in the
quantitative risk assessment. Any analytical errors, or
limitations in data that are identified by the review, must
be noted in the risk assessment if the data are used. An
explanation for qualifiers used must be included with
the review report.

 All data should receive some level of review.  The risk
 assessor may  receive data  prior to  the quantitative
 baseline risk assessment that were not reviewed.  Data
 that have not been reviewed must be identified because
 the lack of review increases the uncertainty for the risk
 assessment These data may lead  to false positive or
 false negative assessments and quartitation errors.
 Unrevicwed data may also contain  transcription errors
 and calculation errors.  Data may be used in the
 preliminary assessment before  review, but must be
 reviewed at a predetermined level before use in the final
 risk assessment.

 Depending upon data user requirements, the level and
 depth of the data review are variable. The  level and
 depth of the data review may be determined during the
 planning process and must include an examination of
  laboratory and method performance for the samples and
  analytcs involved. This examination includes:

     • Evaluation of data completeness.

     • Verification of instrument calibration,

     • Measurement of laboratory precision  using
      duplicates; measurement of laboratory accuracy
       using spikes,

     • Examination of blanks for contamination.
                                                    102

-------
                                  EXHIBIT 64. RELATIVE  IMPORTANCE OF DETECTION LIMIT
                                  AND CONCENTRATION OF CONCERN: DATA ASSESSMENT
                                         Relative Position of Method
                                          Detection Limit (MDL) and
                                       Concentration of Concern (COC)
               Consequence
                                         Confidtnc*   Ml
                                         limits
COC
                                    Concentration
Conlid«nc»
   limits
                                                                          Non-Detects and
                                                                          Detects Useable
                                                         COC
                                    Concentration
                                                                            Possibility of
                                                                         False Positives and
                                                                          False Negatives
                                        COC
                                                   M
                                    Concentration
               Non-Detects Not
                  Useable

               Detects Useable

              Possibility ol False
                 Negatives
                  •  Assessment of adherence to method specifications
                    and QC limits, and

                  •  Evaluation of method performance in the sample
                    matrix.

                Specific data review procedures arc dependent upon the
                method and data user requirements.  Section 5.6.1
                details procedures for evaluating QC samples for
                laboratory and method performance. CLP data review
                procedures are performed according to criteria outlined
                in National Functional Guidelines for Organic Data
                Review (EPA 1991e) and Laboratory Data Validation:
                Functional Guidelines for Evaluating Inorganics
                Analyses (EPA  1988e). Minimum requirements for
                data review are listed in Exhibit 61.
T
        5.6 ASSESSMENT OF CRITERION VI:
            DATA QUALITY INDICATORS

            Minimum Requirements

          • Sampling variability quantitated for each
            annlytc.

          • QC samples required to identify and
            qunntitale precision and accuracy.

          • Sampling and analytical precision and
            accuracy quantified.	

        The assessment of data quality indicators presented in
        this chnpicr is significant to determine data useabilily.
    103

-------
EXHIBIT 65. CONSEQUENCES OF ALTERNATIVE SAMPLING
        STRATEGIES ON TOTAL ERROR ESTIMATE
                                                Consult*
                                               Statistician
Group Data by
MectunVStratum

Judgmental
Model
+• By Analyte *
/ Multiple >v No

Yes
*.

Non-Statistical
Treatment


                                   Accept
                                   Probability
                                   Missing Hot
                                    Spot?
                           10*

-------
                    *• Qualified data can usually be used for
                    quantitative risk assessments.

               The assessment of data quality indicators for either
               sampling or analysis involves the evaluation of five
               indicators:  completeness,  comparability, represen-
               tativeness, precision, and accuracy. Uncertainties in
               completeness, comparability, and representativeness
               increase the probability of false negatives and fake
               positives when the data are used to test particular
               hypotheses as part of the site evaluation. This increase
               in uncertainty can affect the confidence of chemical
               identification. Variation in completeness, comparability.
               representativeness, precision, and accuracy affects the
               uncertainty of estimates of average concentration and
               reasonable maximum exposure (RME).  Once  the
                indicator is examined or a numerical value is determined,
                the re.su Its can be compared to the performance objectives
                established during RI planning.  This comparison
                determines the useability of the data and any required
                corrective actions.

                A summary of the minimum requirements for data
                quality indicators Is presented in Exhibit 61, and the
                evaluation process is illustrated in Exhibit 65. Specific
                requirements for each indicator are presented in the
                following sections.
5.6.1   Assessment of Sampling and
        Analytical Data Quality
        Indicators

The major activity in determining the useability of data
based on sampling is assessing the effectiveness of the
sampling operations performed. Samples provided for
analysis must answer the four  basic decisions to be
made with  Rl data in risk assessment (cited  at the
beginning of this chapter) that are translated into .site-
specific objectives based on scoping and planning
decisions.

Independent djtta review evaluates laboratory results,
not sampling. Determining the useability of analytical
results begins with the review of QC samples and
qualifiers to  assess analytical performance of the
laboratory and the method.   It is  more  important to
evaluate the effect on the data than to determine the
source of the error. The data package is reviewed as a
whole for some criteria; data arc reviewed at the sample
level for other criteria, such as  holding lime. Factors
affecting the accuracy of identification and the precision
and accuracy of quantiiation of individual chemicals.
such as calibration and recoveries, must be examined
analyte-by-analyte. The qualifiers used in the review of
CLP data are presented and their effect on data quality
is discussed in this section.   Exhibit 66 presents a
                              EXHIBIT 66. USE OF QUALITY CONTROL DATA FOR RISK ASSESSMENT
I
Quality Control Criterion
Spikes (High Recovery)
Spikes (Low Recovery)
Duplicates
Blanks
Calibration
Tun*
Internal Standards ,
(Reproducbttv) 3
Internal Standards
(High Recovery)
Internal Standards
(Low Recovery)
Effect on Identification When
Criterion Is not Met
-
False Negative1
None, unless anatyte found
in one duplicate and not the
other. Then either false
positive or false negative.
False Positive
-
False Negative
-
-
Fabe Negative1
Ouanttattve Blaa
HtQn
Low
•S3
High
Higher
Low2
-
-
Low
High
u«e
Use data as upper limit.
Use data as lower limit.
Use data as estimate-poor precision.
Set confidence level 5x blank.
Use data above confidence level.
Use data below confidence level
as estimate.
Use data as estimate
unless problem is extreme.
Reject data or examine raw data and
use professional judgment.
Use data as estimate-poor precision.
Use data as lower limit.
Use data as upper limit.
1 False negative only fk«y I recovery!* near zero.
* Effect on bias determined by examination of data lor each individual analyte.
3 Include* surrogates and system monitoring compounds.
                                                                   105

-------
summary of the QC samples and the data use implications
of qualified data. Corrective action options are shown
in Exhibit 62.

Sample media can be more complex than expected in
environmental analysis.  For example, sludge or oily
wastes may  contain  interfering  chemicals whose
presence cannot be predicted in precision and accuracy
measurements.  The risk assessor must examine the
reported precision [relative percent difference (RPD)]
and accuracy [percent recovery (% R)] data to determine
useability. Ranges used for rejection and qualification
of CLP data have been determined based on the analysis
of target compounds in environmental media.'  These
ranges, documented in the Functional Guidelines (EPA
1991e. EPA  1988e) can be used in the absence of
specifications in the planning documents.
                Completeness.  Completeness for sampling is
                calculated by the following formula:
                Percent      _ (Number of Acceptable Data Points ^ x 100
                Completeness     Total Number of Samples Collected

                This measure of completeness is useful for data collection
                and analysis management but misses the key risk
                assessment issue, which is the total number of data
                points available and acceptable for each chemical of
                potential concern. Incompleteness should be assessed
                to determine if an acceptable level of data useability can
                still be obtained or whether the level of completeness
                must be increased, either by further sampling or by other
                corrective action.  Any decrease in the  number of
                samples from that specified in the sampling design will
                affect the final results.   In this case, the option of
                obtaining more  samples should be reviewed.
        Minimum Requirements
           for Completeness
 Impact When Minimum
Requirements Are Not Met
    Corrective Action
       •  Percentage of sample
         completeness determined
         during planning to meet
         specified performance
         measures.

       •  100% of all data for ar alytes
         in critical samples (at least
         one sample per medium per
         exposure area).

       •  All data from critical samples
         considered crucial.
         Background samples and
         broad spectrum analyses are
         usually critical.
•  Higher probability of false
   negatives.

•  Reduction in confidence
   level and power.

•  A reduction in the number of
   samples reduces she
   coverage and may affect
   representativeness. Data for
   critical samples have
   significantly more impact
   than incomplete data for
   non-critical samples.

 •  Useability of data is
   decreased for critical
   samples.

 •  Useability of data is
   potentially decreased for •
   non-critical samples.

 • Reduced ability to
   differentiate she levels from
   background.

 • Impact of incompleteness
   generally decreases as the
   number of samples
   increases.
•  Resampling or reanalysis to
   fill data gaps.

•  Additional analysis of
   samples already at
   laboratory.

•  Determine whether the
   missing data are crucial to
   the risk assessment (i.e..
   data from critical samples).
                                                                                                21-002-0*1
                                                     106

-------
Typical causes for sample attrition include siie conditions
preventing sample collection (e.g., a well runs dry),
sample breakage, and invalid or unuseable analytical
results. Incompleteness can increase the uncertainty
involved in risk assessments by reducing the available
number of samples on which identification and estimates
of concentration of chemicals at the site are based. The
reduction in the number of samples from the original
design further affects representativeness by reducing
site coverage and increases  the variability  in
concentration estimates. Only the collection of additional
samples will resolve the problem, unless the samples
involved were duplicates or splits.  In this case, or if the
cause was laboratory performance, the extracts may be
considered for reanalysis.

Completeness for analytical data is calculated by the
following formula:

Percent          (Number of Acceptable Sample;^ x 100
Completeness      Total Number of Samples Analyzed

The completeness for analytical data required 'or ri k
assessment is defined as the number of chemical-specific
data results for an exposure area in an operable unit that
are determined acceptable after data review.
                An analysis is considered complete if all data generated
                are determined to be acceptable measurements as defined
                in the SAP. Results for each analyte should be present
                for each sample.  In addition, data from QC samples
                necessary to determine precision and accuracy should
                be present. QC samples and the effects of problems
                associated with these samples are discussed later in this
                section.

                Comparability. Comparability is not compromised
                provided that the sampling design is unbiased, and the
                sampling design or analytical methods have not changed
                over time. If any of these factors change, the risk
                assessor may experience difficulties in combining data
                sets to estimate the RME. The determination of the
                RME is based on the principal of estimating risk over
                lime for the exposure area. The ideal situation occurs
                when samples can be added within the basic design,
                decreasing the level of uncertainty.

                    <*• Anticipate the need to combine data from
                    different sampling events and/or different
                    analytical methods.

                Comparability  is a very important qualitative data
                indicator  for analytical assessment  and is a critical
        Minimum Requirements
           for Comparability
 Impact When Minimum
Requirement* Are Not Met
  Corrective Action
       •  Unbiased sampling design or
         documented reasons for
         selecting another sampling
         design.

       •  The analytical methods used
         must have common analytical
         parameters.

       •  Same units of measure used
         in reporting.

       •  Similar detection limits.

       •  Equivalent sample
         preparation techniques.
• Non-additivity of sample
  results.

• Reduced confidence, power,
  and ability to detect
  differences, given the
  number of samples
  available.

• Increased overall error.
   For Sampling:

Statistical analysis of effects
of bias.

   For Analytical Data:

Preferentially use those data
that provide the most
definitive identification and
quantitation of the chemicals
of potential concern.  For
organic chemical
identification, GC-MS data
are preferred over QC data
generated with other
detectors. For quantitation,
examine the precision and
accuracy data along with the
reported detection limits.

Reanalysis using comparable
methods.
                                                                                                21-OCB-Ott
                                                    107

-------
parameter when considering the combination of data
sets from different analyses for the same chemicals of
potential concern.  The  assessment of data quality
indicators determines if analytical results being reported
are equivalent to data obtained from similar analyses.
Only comparable data sets can readily be combined for
the purpose of generating a single  risk assessment
calculation.

The use of routine methods simplifies the determination
of comparability because all laboratories use the same
standardized procedures and reporting parameters. In
other cases, the risk assessor may have to consult with
an analytical  chemist  to evaluate whether different
methods are sufficiently comparable to combine data
sets. The RPM should request complete descriptions of
non-routine methods. A preliminary assessment can be
made by comparing the analytes, useful range, and
detection limit of the methods. If different units of
measure have been reported, all measurements must be
converted to a common set of units before comparison.

Representativeness.   Representativeness of data is
critical to  risk assessments.  The results of the risk
assessment will be biased to the degree that the data do
not reflect the chemicals and concentrations present in
the exposure area or unit of interest. Non-representative
chemical identification may result in false  negatives.
Non-representative estimates of concentration kvels
may be higher or lower than the true concentration.
Non-representative sampling can  usually only be
                resolved by additional sampling, unless the potential
                limitations of the risk assessment are acceptable.
                It is important to determine whether any changes have
                occurred in the actual sample collection that convert an
                originally unbiased sampling plan into a biased sampling
                episode. Bias in unbiased designs is difficult to assess
                because no measure of the true value is known. Bias is
                assumed in non-statistical designs.

                Representativeness  is primarily a  planning concern.
                The solution is in the design of a sampling plan that is
                representative.  Once the design is  implemented, only
                the sampling variability is evaluated during  the
                assessment process, unless contamination occurs in the
                QC samples or blanks, or problems exist during sample
                preparation that affect sample results.  Incompleteness
                of data potentially  decreases representativeness and
                increases the potential for false negatives and the bias in
                estimations of concentration.

                Representativeness  is determined  by examining  the
                sampling  plan, as  discussed in  Section 3.2.   In
                determining  the representativeness of the data,  the
                evaluator examines the degree to which the data meet
                the performance standards of the method and to which
                the analysis represents the  sample submitted to the
                laboratory.   Analytical  data  quality affects
                representativeness since data of low quality may be
                rejected for use in risk  assessments.  Holding time.
                sample preservation, extraction procedures, and results
          Minimum Requirements
           for Representativeness
            Sample data representative
            of exposure area and
            operable units.

            Documented sample
            preparation procedures.
            Filtering, compositing, and
            sample preservation may
            affect representativeness.

            Documented analytical data
            as specified In the SAP.
 Impact When Minimum
Requirements Are Not Met
  Bias high or low in estimate
  ofRME.

  Increased likelihood of false
  negatives.

  Inaccurate identification or
  estimate of concentration
  that leads to inaccurate
  calculation of risk.

  Remaining data may no
  longer sufficiently represent
  the site if a large portion of
  the data are rejected, or If all
  data from analyses of
  samples at a specific location
  are rejected.
  Corrective Action
Additional sampling.

Examination of effects of
sample preparation
procedures.

For critical samples.
reanalyses of samples or
resampling of the affected
site areas. For non-critical
samples, reanalyses or
resampling should be
decided by the RPM In
consultation with the
technical team.

If the resampling or
reanalyses cannot be
performed, document in the
site assessment report what
areas of the site are not
 represented due to poor
 quality of analytical data.
                    M-002-Ctt
                                                      108

-------
from analyses of blanks affect the representativeness of
analytical data (see Appendix V).

Precision.  The two basic activities performed in the
assessment of precision  are  estimating sampling
variability  from the  observed  spatial variation and
estimating the measurement error attributable to the
data collection process.  Assumptions concerning the
sampling design and data distributions must be examined
prior to interpreting the results.  This examination will
provide the basis for selecting calculation formulas and
knowing when statistical consultation is required.
The type of sampling design selected is critical to the
estimation of sampling variability  as discussed in
Sections 3.2  and 4.1.  If the sampling design is
judgmental, the nature of the sampling error cannot be
determined and estimates of the average concentrations
of analytes may not be representative of the site.

     <*• Determine the distribution of the data
     before applying statistical measures.
               The nature of the observed chemical data distribution
               affects  estimation procedures.  The estimation of
               variability and confidence intervals will become complex
               if the distribution cannot be assumed normal or to
               approximate normal when transformed to log normal.
               Estimates of the  95% upper confidence  limit of the
               average concentration for the RME should be based on
               an analysis of the frequency distribution of the data
               whenever the database is sufficient to support such
               analysis. Statistical tests may be used to compare the
               distribution of the observed data with the normal or log
               normal  distribution (Gilbert 1987).  Graphs of data
               without statistical test results may also be acceptable for
               some data sets. Statistical computer software can assist
               in the analyses of data distribution.

               Sampling  variability. Exhibit 67  summarizes the
               assessment procedures for the evaluation of variability
               from different sampling procedures.  The estimation of
               confidence levels, power, and minimum detectable
               relative differences requires assumptions about  the
               coefficients of variation from sampling variability for
        Minimum Requirements
             (or Precision
 Impact When Minimum
Requirements Are Not Met
    Corrective Action
       • Confidence level of 80% (or
         as specified in DQOs).

       • Power of 90% (or as specified
         in DQOs).

       • Minimum detectable relative
         differences specified in SAP
         and modified after analysis of
         background samples if
         necessary.

       • One set of field duplicates or
         more as specified in the SAP.

       • Analytical duplicates and
         splits as specified in the SAP.

       • Measurement error specified.
• Errors in decisions to act or
  not act based on analytical
  data.

• Unacceptable level of
  uncertainty.

• Increased variability of
  quantitative results.

• False negatives for
  measurements near the
  detection limits.
      For Sampling:

• Add samples based on
  information from available
  data that are known to be
  representative.

• Adjust performance
  objectives.

      For Analysis:

• Analysis of new duplicate
  samples.

• Review laboratory protocols
  to ensure comparability.

• Use precision measure-
  ments to determine
  confidence limits for the
  effects on the data.

• The risk assessor can use
  the maximum sample results
  to set an upper bound on the
  uncertainty in the risk
  assessment if there is too
  much variability in the
  analyses.

                       21-002-064

-------
           EXHIBIT 67. STEPS TO ASSESS SAMPLING PERFORMANCE
            1.  Confirm statistical assumptions.

            2.  Summarize anaryte detection data by strata: media within site or site subgroups
               and strata within media.

            3.  Transform analyte concentration data so distribution is approximately normal.

            4.  Calculate the coefficient of variation for each analyte detected.

            5.  Using Exhibit 47 'Relationships Between Measures of Statistical Performance
               and Number of Samples Required,' look up the range of power, confidence
               level  and minimal detectable relative differences for the calculated
               coefficient of variation.

            6.  Compare the statistical performance measures required to those achievable
               given the coefficient of variation and sample size.

            7.   If the performance objectives are achieved, go to  Step 9.

                If the required statistical performance levels are not met, then additional samples
                must be taken or one or more of the performance parameters must be changed.

                If samples are to be added. Exhibit 47 and the calculation formulas in Appendix
                IV  can be used to determine the number needed.

            8.   If the performance parameters are to be changed, the parameter  to be changed
                should be the one which will increase the probability of taking unnecessary
                action as opposed to unnecessary risk.

            9.   Examine the results of the QC samples. Sample results must be considered to
                be qualitative if no results are available for QC samples.

            10.   If the QC sample results indicate possible bias through contamination, take
                appropriate corrective action.
                                                                                       21-002-0*7
each chemical of potential concern.  The RPM or risk
assessor should discuss the implications of these
assumptions with a statistician  to determine their
potential impacts on data useabitity.
    f Determine the statistical measures of
    performance most applicable to site
    conditions before assessing data useability.

Once the statistical assumptions and observed analyte
variability are known, selected statistical performance
measures can be assessed to determine (he data quality
achieved.   Additional samples may be needed, or
modified DQOs required, as a result of evaluating
sampling variability.  Three issues are involved in the
assessment of required statistical performance:

  •  Level of certainty or confidence,

  •  Power, and

  •  Minimum delectable relative difference.

The  required level for each of these performance
measures should be included in the SAP as DQOs. The
user's data quality  requirements defined  by  these
statistical measures determine the number of samples
that are taken during data collection.  Recommended
minimum statistical performance parameters for
                                                 110

-------
discriminating contaminant  concentrations  from
background levels in risk assessment are provided in
Exhibit 68.

     EXHIBIT 68. RECOMMENDED
         MINIMUM STATISTICAL
 PERFORMANCE PARAMETERS FOR
           RISK ASSESSMENT
   Null Hypothesis: On-srle Contaminant
     Concentrations are not Higher than
               the Background
     Confidence level:
     80% minimum, reject null when true (take
     unnecessary action).
           o
     Power.
     90% minimum, accept null when false (fail to
     take action when action is required).

     Minimum detectable relative difference:
     10% • 20%.  usually depends on concentration
     of concern.
   1  (1 -false positive estimate) or (1 -a).
   2  (1-false negative estimate) or (1 -P).
  Source: EPA 1989f.
                                         21-OOt-OH
 First, summarize the sample results at the analyte level
 by stratum and strata within media to determine whether
 the performance objectives have been met. Sampling
 error Is not relevant  if a particular combination of
 stratum and analyte yields only a single data point In
 that case, assessment proceeds to that of analytical error
 for that stratum and analyte combination.

 The distribution for stratum and analyte combinations
 with multiple data points should usually be examined
 for normality and transformed  to log  normal. The
 coefficient of variation is calculated for each stratum
 and analyte combination. If the distribution resulting
 from  the transformation  is not normal, a new
 distributional model will need  to be  identified and
 validated in consultation with  a statistician.  Non-
 parametric procedures which require no distributional
 assumptions may also be used.

 Conversely, the statistical performance achieved «.:Ji be
 determined, given the coefficient of variation. This
 performance should be compared to the requirements
 stated in planning. If the performance objectives are
 achieved, the risk assessor can proceed to the assessment
 of measurement error.
If the required statistical performance objectives are not
met, additional samples must be taken, or one (or more)
of the performance parameters must be changed. If
samples are added, the tables and formulas provided in
Chapter 4 and Appendix IV can be used to calculate the
number of samples required. Ifaperformance parameter
is changed, it should be the one that will increase the
probability of taking unnecessary action as opposed to
an increased probability of unnecessary risk.   The
uncertainty level will then be reduced first, the minimum
detectable relative difference will be increased second.
and the level of power will be reduced last. Minimum
recommended levels for performance parameters in
risk assessment in the absence of site-specific DQOs are
80% confidence levels, 90% power,  and 10-20%
minimum detectable relative differences (EPA 19890.
Exhibit 68 summarizes the recommended DQOs for
statistical performance parameters.

Measurement error. Measurement error is estimated
using the results of field duplicate samples.   Field
duplicates determine total within-batch measurement
error, including analytical error if the samples are also
analyzed as laboratory duplicates. The estimate is of the
difference between analytical values reported for
duplicates. This type of variation has four basic sources:
sample collection procedures, sample handling and
storage procedures, analytical procedures, and data
processing procedures.

The formula for computing the relative percent difference
between duplicates is:

        V     xlOO
 where R, and R2 are the results from the first and second
 duplicate samples, respectively. Precision is a measure
 of the repeatability of a single measurement and is
 evaluated from the results of duplicate samples and
 splits.

 Low precision can be caused by poor instrument
 performance, inconsistent application of method
 protocols, or byadifficult, heterogeneous sample matrix.
 The last effect can be distinguished from the others by
 evaluation of laboratory QC data.

 If split samples have been analyzed by different methods
 or different laboratories, then data users have a measure
 of the  quality of  individual techniques.  Splits are
 particularly effective when one laboratory is a reference
 laboratory. Ifbothsetsofdataexhibitthe same problems,
 then laboratory performance can usually be ruled out as
 a source of error. Splits are also useful when using non-
 routine methods or comparing results from different
 analytical methods.
                                                   Ill

-------
Accuracy. Accuracy is a measure of overestimation or
underestimation of reported concentrations and is
evaluated from the results of spiked samples.   The
procedure for determining accuracy will vary according
to differences in the number of measurements and the
precision of the estimates. Data that are not reported
with confidence limits cannot be assigned weights
based on precision and should not be combined for use
(Taylor 1987).
Spiked samples are particularly useful in the analysis of
complex sample types because they help the reviewer
determine the extent of bias on the sample measurement.
A set of standards at known concentrations is mixed into
a portion of the sample or into distilled water prior to
sample preparation and analysis. The analytical results
are compared to the amount spiked to determine the
level of recovery. It is important to note that unless
every sample is spiked, spike recoveries indicate only a
trend rather than a specific quantitative measure.

Accuracy is con trolled primarily by the analytical process
and is reported as bias. The absolute bias of a sampling
design cannot be determined unambiguously berause
the true value of the chemicals of concern in the exposure
area can never be known. However, statistically based
sampling designs described in Chapter 4 are structured
to produce unbiased results.

 Bias can be estimated using field  spikes on field
evaluation or audit samples to assess the accuracy and
                 comparability of results. These estimates will reflect
                 the effects of sample collection, handling, holding time,
                 and the analytical process on the result for the sample
                 collected.

                 Bias  is estimated for  the measurement process by
                 computing the percent recovery (%R) for the spiked or
                 reference compound as follows:


                 — D   (Meagurqd Anyminl *  Amount in I^mpilce^j Sampled » IOQ
                                  Amount Spiked
                 Because of the inherent problems associated with the
                 spiking procedure and the interpretation of recovery,
                 spikes are considered minimum requirements only if
                 specified in the SAP. Field matrix spikes are currently
                 not recommended for use in soils (EPA 19890.

                 Field blanks are evaluated to estimate the potential bias
                 caused by contamination from sample collection,
                 preparation, shipping and/or storage. Results for the
                 analysis of field blanks indicate whether contamination
                 resulted in bias, but they are not estimates of accuracy.
                 Bias pertaining to analytical recoveries is computed as
                 follows:
                 Percent
                 Btu
(Me
aired Amount-Amount in Umpiked Simple^ » 100
          Amount Spiked
       Minimum Requirement*
             for Accuracy
  Impact When Minimum
Requirement* Are Not Met
             Corrective Action
         Field spikes to assess
         accuracy of non-detects and
         positive sample results if
         specified in the SAP.

         Analytical spikes as
         specified in the SAP.

         Use analytical methods
         (routine methods whenever
         possble) that specify
         expected or required
         recovery ranges using
         spikes or other QC
         measures.

         No chemicals of potential
         concern detected In the
         blanks.
•  Increased potential for false
   negatives. If spike recovery
   is low, it is probable that the
   method or analysis is biased
   low for that analyte and
   values of all related samples
   may underestimate the
   actual concentration.

 •  Increased potential for false
   positives. If spike recovery
   exceeds 100%, interferences
   may be present, and ft Is
   probable that the method or
   analysis is biased high.
   Analytical results
   overestimate the true
   concentration of the spiked
   analyte.
        •  Consider resampling at
           affected locations.

        •  No correction factor is
           applied to CLP data on the
           basis of the percent recovery
           In calculating the analyte
           concentration.

        •  H recoveries are extremely
           low or extremely high, the
           risk assessor should consult
           with an analytical chemist to
           identify • more appropriate
           method for reanalysis of the
           samples.
                                                                                                21-ooz-oes
                                                     112

-------
Blanks are of primary concern for the analysis of bias
involved in sampling because  of  the difficulty in
performing field spikes and the availability of appropriate
reference standards and matrix for evaluation samples.

Results from blanks can be used to estimate the extent
of high bias in theevent of contamination. The following
procedures  should be  implemented to prevent the
assignment  of false positive values due to blank
contamination:

   • If the  field blanks are contaminated  and the
     laboratory blanks are not, the RPM or risk assessor
     can conclude that contamination occurred prior to
     receipt of the samples by the laboratory. If the
     contamination is significant (i.e., it will interfere
     with the determination of risk), consider resampling
     at affected locations.

   • If it is  not possible to resample, the RPM or risk
     assessormust assess the effect ofthe contamination
     on the potential for false positives. Often, this
     determination can be made by examining data
     from samples located nearby.  If all samples and
     blanks show the same level of apanicutar chemical,
     the presence ofthe chemical in the samples is most
     likely due to contamination.

   • If the  laboratory  blanks are contaminated, the
     laboratory  should  be required to rerun the
     associated analyses. This is especially important
     in the case of critical analy tes or samples. Before
     reanalyses, the laboratory must  demonstrate
     freedom from contamination by providing results
     of a clean laboratory blank. Note: If laboratory
     blanks are contaminated, field blanks will generally
     also be contaminated.

    • If reanalysis is not possible, then the sample data
     must be qualified.  The Functional Guidelines
     provide examples  of blank qualification.
     Chemicals detected  in  the associated samples
     below the action level defined in the Functional
     Guidelines are considered undetected.

 Data  qualifiers.  All data generated by the routine
 analytical services of the CLP are reviewed and qualified
 by Regional representatives according to the guidelines
 found in the Functional Guidelines as modified to fit the
 requirements  of the individual Regions.

     *• Use data qualified as U or J for  risk
     assessment purposes.

 Analytes qualified  with a U  are considered "not
 detected." If precision and accuracy are acceptable (as
 determined by the QC samples), data are entered in the
 data summary tables in the data validation report as the
SQL or corrected quantitation limit (MDL corrected for
dilution and percent moisture), and qualified with a U.
Note that the same chemical can be reported undetected
in a series of samples at different concentrations because
of sample differences.

Data  qualified with an  R  are rejected because
performance requirements in the sample or in associated
QC analyses  were not met.  For example, if a mass
spectrometer "tune" is not within specifications, neither
the identification nor quantitation of chemicals can be
accepted with confidence. Extremely low recoveries of
a chemical in a spiked sample might also warrant an R
designation for that  chemical in associated samples
because ofthe risk of false negatives (see Appendix VI).

Data qualified with al present a more complex issue. J-
qualified data  are considered "estimated" because
quantitation in the sample or in associated QC samples
did not meet specifications.  The justification for
qualifying the data should be explained in the validation
report  Draft revisions of the Functional Guidelines
propose that the justification be included on a qualifier
summary table submitted with the validation report.

Data can  be  biased  high  or  low when qualified as
estimated.   The  bias can often be determined by
examining the results of the QC samples. For example,
if interfering levels of aluminum are found in inorganic
analysis of the interference check sample, the sample
results are probably biased high because the signal
overlap is added to the signal being reported. When
volatile organic compounds are qualified J for holding
time violations, the results are usually biased low because
some of the volatile  compounds may have volatilized
during storage.

Data associated with contaminated blanks are not
consideredestimatedandarenot flagged J. Thepresence
of the blank contaminant chemical in the analytical
samples is questionable at levels up to 5 to 10 times
those found in the blank, depending on the nature of the
analyte. An action level is determined for each chemical
based on the quantity found in the blank.  Data above the
action level are accepted without qualification and data
between the contract required quantitalion limit (CRQL)
and the action level are qualified U (undetected).

Estimated organics and inorganics data that are below
 the CRQL or contract required detection limit (CRDL)
 are qualified as UJ. This qualifier signifies that the
 quantitation  limit is  estimated because the QC rvaiilts
 did not meet criteria specified in the SAP.

 Other qualifiers may be added to the analytical data by
 the laboratory. A set of qualifiers (or flags) has been
 defined by the CLP for use by the laboratories to dqpote
                                                   113

-------
problems with the analytical data. These qualifiers and
their potential use in risk assessment are discussed in
RAGS (EPA 1989a).

5.6.2  Combining the Assessment of
        Sampling and Analysis

Once the quality of the sampling and analysis effort has
been assessed using the Five data quality indicators,
combine the results to determine the overall assessment
of a particular indicator across sampling and analysis.
Combining the assessment  for completeness,
comparability, and representativeness is discussed in
this sectionasaqualitativeprocedure. Statistical models
are available for combining data sets with different
variability and bias. The risk assessor should consult a
chemist or statistician if the magnitude of the sampling
and analysis effort warrants the use of a formal statistical
treatment of comparability.

The basic model for estimating total variability across
sampling and analysis components is presented in Exhibit
69.  An example  of a non-statistical approach  to
combining the assessment results is given in Exhibit 70.
 Using this approach, each  data quality  indicator is
                   assessed to determine whether a problem exists in either
                   sampling or analysis. This assessment leads to different
                   combinations of problem determination. For example,
                   completeness may have been a problem in sampling
                   [YES]  but not a problem in  analysis [NO]; the
                   combination is [YES/NO].

                   Basic guidance is given on the combinations of sampling
                   and analysis once assessment patterns based  on the
                   determination of aproblem have been established. This
                   guidance is qualitative in nature and  is presented to
                   assist in organizing the data assessment problem for the
                   application of professional judgment If the assessment
                   pattern is [NO/NO], the issue of combining results is not
                   a problem. Conversely, if the pattern is [YES/YES], the
                   issue of combining results is an issue of the effects of the
                   combined magnitudes. Instances of combined sampling
                   and analysis problems for a single indicator will have
                   significant effects on the risk assessment uncertainty.
                   The most complicated assessment pattern to interpret is
                   encountered when a problem occurs in one area but not
                   in another (e.g.. in sampling but not in analysis). This
                   situation is briefly discussed for each indicator in the
                   following sections.
                 EXHIBIT 69.  BASIC MODEL FOR ESTIMATING
                 TOTAL VARIABILITY ACROSS SAMPLING AND
                               ANALYSIS COMPONENTS
                    o,    =
                   where
  °p

i total variability
: measurement variability
: population variability
                   where     o§  « sampling variability (standard deviation)
                              n   • handling, transportation and storage variability
                              
-------
EXHIBIT 70. COMBINING DATA QUALITY INDICATORS FROM
 SAMPLING AND ANALYSIS INTO A SINGLE ASSESSMENT
                      OF UNCERTAINTY
                   Assessment of Problems
                                           Combined Sampling
                                              and Analytical .
                                              Determination
 Data Quality
  Indicators
Sampling    Analytical
                                                 YES/NO

                                                 NO/YES
 Completeness
                                                 YES/YES

                                                  YES/NO

                                                  NO/YES
 Comparability
                                                 YES/YES

                                                  YES/NO

                                                  NO/YES
Representativeness
                                                 YES/YES

                                                  YES/NO

                                                  NO/YES
                                                 YES/YES

                                                  YES/NO

                                                  NO/YES
  The combination [NO/NO] indicates that the data quality indicator will not affect the
  level of uncertainty In data useability.
                                                            21402-070
                               115

-------
Completeness. A sample is considered incomplete for
all analytes. Analytical incompleteness is usually related
to particular analytes. In this instance [YES/YES], the
effect on the risk assessment will vary  according to
chemical.  For some chemicals, the data points will be
lost in both sampling and analysis.
The effects of a loss in the number of sample points for
a particular chemical can be substantial.  For example,
if collection of 10 samples was planned and one sample
could not be collected because of site access problems,
one  was  broken in transport, and  the laboratory
experienced analysis problems with three samples for
the chemical of potential concern causing the data to be
rejected, then only five data points remain.

If the assessment pattern is [YES/NO], the effects are
distributed across  all chemicals involved in the risk
assessment If the pattern is [NO/YES],  the effects are
localized  to the particular chemical affected.

Comparability. Comparability problems in sampling
are primarily due to different sampling designs and time
 periods.  Seasonal variations are treated like spatial
 variations because the risk assessment is calculated as
 risk over time. Data can be averaged and considered as
 a single data set.  For analytical data,  comparability
 problems are related primarily to the use of different
 methods and laboratories.  A pattern of [YES/YES] will
 indicate that the risk assessor will have considerable
 difficulty in  :ombining the  various data  sets into a
 single assessment of risk.  In situations of [YES/NO] or
 [NO/YES], the problem of sampling comparability is
 more difficult to resolve. Models exist for determining
 comparability between methods and integrating results
 across laboratories. These models involve the general
 statistical approach to confirming data sets with different
 but known variability and bias (Taylor  1987).
Representativeness. Representativeness in sampling
is critical to the risk assessment. Non-representativeness
affects both false negatives (chemicals not identified)
and estimates of concentration and, therefore, affects
estimates  of RME.   Analytical representativeness
involves the question of whether the analytical results
represent the sample collected.  For example, holding
times and sample preservation can cause the analytical
results not to be representative of the sample collected.
These  questions should  be  treated separately in the
discussion of effects.
Precision.  The contribution  to imprecision  from
sampling variability often exceeds that from analytical
variability in the measurement process.  If precision is
a problem in both sampling and analysis, the risk
assessor should focus on  the  impact  of sampling
variabilityontheestimateorRME. Analytical variability
will be minimal in comparison to the effects of sampling
variability unless the sampling variability is untypically
low and the analytical variability is untypically high.

Accuracy. The assessment of accuracy in  sampling is
focused primarily on  recoveries  from spiked  or
performance  evaluation  samples.   Analytical
performance and  potential blank contamination are
reflected in analytical spike recoveries. If the pattern is
 [YES/YES] tor accuracy, this may require assessment
 of calibration, or of potential blank contaminants, and
 integration of their possible effects by comparison of
 results from laboratory and field QC samples.

 If the accuracy pattern is [NO/YES], then the issue is
 analytical performance. Low variability in sampling as
 measured by low coefficients of variation for chemicals
 of potential concern should increase the risk assessor's
 concern over an analytical accuracy problem.

 Highsamplingvariability(CV>25%)willgreatlyreduce
 the effects of analytical bias on the level of certainty of
 the risk assessment.
                                                     116

-------
                                         Chapter  6
               Application of  Data to Risk Assessments
This chapter provides guidance for integrating the
assessment of data useability to determine the overall
level of uncertainty of risk assessment. This guidance
builds on each of the previous chapters.

   •  Chapter 2 explained the risk assessment process
     and  the roles  and responsibilities  of key
     participants.  Exhibit 5 defined a continuum of
     level of certainty in the baseline risk assessment
     result based on the ability of the risk assessor to
     quantitate  or qualify  the  level  of uncertainty
     associated  with the analytical data.

   •  Chapter 3 denned six data useability criteria and
     examined preliminary issues  that must be
     considered while planning sampling and analysis
     activities to increase the certainty of the analytical
     data collected for the risk assessment.

   •  Chapter 4 presented strategies for planning
     sampling and analysis activities based ou the six.
     data useability criteria.

   •  ChapterSdescribedhowtouseeachdatauseability
     criterion to determine the effect of sampling and
     analysis issuesondataqualityand on the useability
     of data in baseline risk assessment.

 The Data Useability Worksheet (Exhibit 63) assists the
 risk assessor in summarizing data quality across the
 various assessment phases.  This worksheet is the basis
 for this chapter's discussion of the impact of analytical
 data quality on the level of certainty of  the risk
 assessment.
 6.1  ASSESSMENT OF THE LEVEL OF
      CERTAINTY ASSOCIATED WITH
      THE ANALYTICAL DATA
 This section  explains bow to  assess the level of
 confidence in sampling and analytical procedures in the
 context of the four major decisions to be made by the
 risk assessorwith environmental analytical data. Exhibits
 in this section apply the data useability criteria, defined
 in Chapter 3  and appearing on the Data Useability
 Worksheet, to these four decisions. Data useability
 criteria affect the level of confidence involved in each
 decision. The level of certainty in the data collection
 andevaluationcomponentof risk assessment affects the
 overall certainty of the risk estimate.
6.1.1   What Contamination is Present
        and at What Levels?

The risk assessor's first task is to use analytical data to
determine what contamination is present at the site and
at what levels (i.e., what potential exists for increased
risk from the contamination).  Exhibit 71  lists the
criteria from the Data Useability Worksheet that affect
this decision. The most critical analytical data question
to be answered before calculating  the risk is the
probability of false negatives or false positives. False
negatives are of greater concern in risk assessment than
false positives, since false negatives  may result in a
decision that would not be protective of human health.
False positives cause the calculated risk to be biased
high, and are of concern because taking unnecessary
action at a site is costly.

    *•  7776 major concern with false negatives
    is that the decision  based on the risk
    assessment may not be protective ofhuman
    health.
Probability of false negatives. Falsenegatives occur
when chemicals of potential concern are present but are
not detected by the sampling design or the analytical
method.  The probability of false negatives can be
determined by using the following parameters from the
Data Useability Worksheet: analytical methods, data
review,   sampling  completeness,   sampling
representativeness, analytical completeness, analytical
precision and accuracy, and combined error.

     •r False negatives can occur if sampling is
     not representative, if detection limits are
     above concentrations of concern, or if spike
     recoveries are very low.

 Sampling strategies can increase the probability of false
 negatives if too few samples were taken or if sections of
 the  site were not sampled. The probability of false
 negatives increases if sampling of any exposure pathway
 was not representative.

 Knowledge of analyte-specific detection limits is critical
 to determining  the probability of false negatives.
 Recovery  values from spikes, internal standards,
                  Acronyms

  RAGS    Risk AiMssment Guidance for Superfund
  SAP     sampling and analyiis plan
  SOP     standard operating procedure
                                                  117

-------
             EXHIBIT 71.  DATA USEABILITY CRITERIA AFFECTING
                              CONTAMINATION PRESENCE
   Worksheet
   Reference
               Data Useablllty
                   Criterion
Data Collection and
Evaluation Decision
      1
      2B
      2C
      3A
      4
      5
      6A
      6C
      6D
      6E
Reports to risk assessor
Documentation (SOPs)
Documentation  (analytical records)
Data sources (analytical)
Analytical methods
Data review
Completeness (analytical)
Representativeness (sampling)
Precision (analytical)
Accuracy (sampling and analytical)
  What contamination is
   present and at what
         levels?
surrogates, and system monitoring compounds are used
to assess the level of accuracy and precision in laboratory
data and determine whether the detection limits stated in
the analytical methods have been met.

   •  The probability of false negatives for an analyte is
     high if the concentration of concern is at or below
     the detection limit. This probability should have
     been documented during planning if no analytical
     methods were found with detection limits below
     the concentration of concern. If the concentration
     of concern is very near the detection limit, a false
     negative can occur because of "drift" in instrument
     response. This behavior may not be reflected in
     data from spike recoveries or blanks.

   • The probability of false negatives is low if spike
     recoveries are acceptable, or biased high as
     documented during data review, and the detection
     limits are below the concentration of concern for
     each analyte.

   • Theprobabilityoffalsenegativesisdiiectlyrelated
     to the amount of bias if spike recoveries are biased
     low and detection limits are below the concentration
     of concern for each analyte.  The effect  is more
     pronounced the closer the concentration of concern
     is to the detection limits.

   • The possibility of false negatives should be
     carefully evaluated whenever sample extracts have
     been highly diluted (i.e.. diluted beyond normal
     method specifications).

 Probability of false positives. False positives occur
 when a chemical of concern is detected by an analytical
                                                                                           21-002-071
                                   method but is truly not present at the site. Assessment
                                   of the following parameters from the Data Useability
                                   Worksheet can be used to determine the probability of
                                   false positives: analytical methods, data re view, sampling
                                   accuracy, analytical completeness, analytical precision
                                   and accuracy, and combined error.

                                       <*•  False positives can occur when blanks
                                       are contaminated or spike recoveries are
                                       very high.

                                   Sampling and analysis uncertainties connected with
                                   false positives can be assessed by examining the results
                                   of quality control samples.  Blank contamination is the
                                   most important indicator of probability of false positives,
                                   particularly when accompaniedby high spike recoveries.
                                   As described in Chapter 5, samples can be contaminated
                                   during sampling, storage, or  analysis.   Field and
                                   laboratory blanks identify this problem by determining
                                   the level and point of contamination.  Sample matrix
                                   interferences can also cause false positives. High spike
                                   recoveries indicate that matrix interference has occurred.

                                      • The probability of false positives is high if the
                                        chemical of potential concern has been detected in
                                        any blanks.  False positives should be suspected
                                        for any sample value less than 5 times the blank
                                        concentration  (10 times for common laboratory
                                        contaminants). High spike recoveries combined
                                        with blank contamination increase the likelihood
                                        of false positives.

                                      • The probability of a false positive for an analyte is
                                        directly related to the amount of bias if chemicals
                                        of potential concern are detected in blanks and
                                        spike recoveries for the analyte are biased high.
                                                  118

-------
  • The probability of false positives is highest when
    the reported concentration is near the detection
    limit for an analyte.

  • The probability of false positives Is low if chemicals
    of potential concern have not been detected in any
    blanks and spike recoveries are not biased high.

6.1.2   Are Site Concentrations
        Sufficiently Different from
        Background?

Background samples provide baseline measurements to
determine the degree of contamination.  Background
samples are collected and analyzed for each medium of
concern in the same manner as other site samples. They
require the  same degree of quality control and data
review. Background samples differ from other samples
in that the sampling points, as defined in the sampling
and analysis plan (SAP), are intended to be in an area
that has not been exposed to the source of contamination.
Historical data, when available, are particularly useful
in selecting sampling and analysis techniques used to
determine the representative concentrations of chemicals
of potential  concern in background samples. Historical
data can help to delineate physical areas that  are
background and provide a basis for temporal trends in
the concentration of chemicals of potential concern.
Exhibit  72  lists the criteria  from the  Data Useability
Worksheet  that affect this decision.

As pan of the risk assessment process, the risk assessor
must determine  if background  samples  are
uncontaminated. The entire data collection process will
be simplified if chemicals of potential concern are not
found in background samples. If chemicals of potential
concern are found in the background samples, the risk
assessor must determine whether they are at naturally
                                occurring  levels, of anthropogenic origin,  due to
                                contamination during the sampling process, or are site
                                contaminants.

                                Both naturally occurring chemicals and anthropogenic
                                chemicals have significance for  risk assessment.
                                Naturally occurring chemicals are those expected at a
                                site in the absence of human influence.  Metals are
                                naturally occurring chemicals that are often included in
                                risk analysis; they are often present in environmental
                                media in varying concentrations. For example, soils of
                                high organic content, such as humus, would have a low
                                concentration of metals by weight,  while soils with a
                                high clay content would contain higher metal levels.
                                Anthropogenic chemicals are defined in RAGS (EPA
                                1989a) as chemicals that are present in the environment
                                due to man-made,  non-site  sources (e.g., industry,
                                automobiles). Chemicals of anthropogenic origin may
                                include organic compounds  such as phthalates
                                (plasticizers), DDT.orpolycyclicaromatichydrocarbons
                                and inorganic chemicals such as lead (from automobile
                                exhaust).   Guidance highlights for background
                                concentration issues for risk assessment are:

                                   •  Organic chemicals of potential concern found in
                                     background  samples  should not be considered
                                     naturally occurring. They may be present because
                                     they  are either site contaminants or are of
                                     anthropogenic origin. They also could be a result
                                     of contamination during sampling.

                                   •  The risk assessor may eliminate chemicals from
                                     risk assessment calculations if their concentrations
                                     fall within naturally coining levels and are below
                                     the concentration of concern.

                                   •  Contamination ofbackground samples is indicated
                                     if chemical concentrations are higher than naturally
                                     occurring levels. Such contamination may come
             EXHIBIT 72.  DATA USEABILITY CRITERIA AFFECTING
                         BACKGROUND LEVEL COMPARISON
      Worksheet
      Reference
         1
         2A
         3A
         6A
         6B
         60
         6E
          Data Useability
              Criterion
  Data Collection and
  Evaluation Decision
Reports to risk assessor
Documentation (SAP) and historical data
Data sources (analytical}
Completeness (sampling)
Comparability (analytical)
Precision (analytical)
Accuracy (sampling and analytical)
                                                 119
 Are site concentrations
sufficiently different from
     background?
                                                                                         zt-ooz-on

-------
   from anthropogenic sources or from problems in
   sampling or analysis activities. The risk assessor
   may include analytical data with other site data or
   perform a separate risk assessment based on best
   professional judgment.

 • Anthropogenic chemicals should not be eliminated
   from the risk assessment.

 • Statistical analysis may be necessary to determine
   if site levels are distinctly different from those
   found in background samples when background
   results approach site concentration levels.

 • Statistical analysis may be necessary where
   chemicals of potential concern are detected in site
   samples at very low concentrations. It is difficult
   to distinguish a difference between background
    and site sample concentrations at levels close to
    the detection limit.

   *• Statistical analysis may determine if site
   concentrations  are  significantly  above
   background concentrations when the
   differences are not obvious.

6.1.3  Are All Exposure Pathways and
       Areas Identified and Examined?

The identification and examination of exposure pathways
is discussed in detail in RAGS. Exhibit 73 summarizes
the criteria that the risk assessormust assess to determine
the probable level of certainty that all exposure pathways
and areas have been identified and examined.

The nature of the exposure pathways and areas to be
examined is critical to the selection of a sampling design
and analytical methods.  If the pathways and areas are
not identified properly, the resulting characterization
maybe inappropriate. Tberiskassessorshoulddetennine
which pathways and areas are not adequately assessed
                           and determine the effect on the risk assessment if they
                           are excluded from study.  Guidance highlights  for
                           exposure pathway identification for risk assessment
                           are:

                             • Recommend acquisition of additional samples
                               from  the  inadequately represented exposure
                               pathway  or  area  if  feasible.   (Sampling
                               considerations presented in Chapter 3 should be
                               re-examined).

                             • Investigate whether computer simulation modeling
                               is feasible if additional samples cannot be collected
                               from an inadequately represented pathway or area.
                               For example, air flow models could be used to
                               estimate transport of volatile contaminants if the
                               contamination of soil and water at a site is fully
                               characterized but no air samples were obtained.

                              •  Note in the report that the risk could not be
                                determined for a pathway or area, or use simple
                                chemical/physical relationships  to estimate
                                exposure if additional samples cannot be collected
                                from an inadequately represented pathway and no
                                simulation models are appropriate. For example,
                                equilibrium partition coefficients can be used to
                                estimate movement in the vadose zone of soil if
                                insufficient data exist to calibrate a groundwater
                                transport model.

                            6.1.4  Are All Exposure Areas Fully
                                   Characterized?

                            Assessing how well exposure  areas have  been
                            characterized involves  evaluation  of completeness,
                            comparability, and representativeness across analytical
                            and sampling data quality indicators.  Exhibit 74 lists
                            the criteria from the worksheet that affect this decision.
                            To be fully characterized, the exposure area mast have
          EXHIBIT 73. DATA USEABILITY CRITERIA AFFECTING EXPOSURE
                      PATHWAY AND EXPOSURE AREA EXAMINATION
         Worksheet
         Reference
             1
             2A
             3B
             6A
             6B
    Data Useability
        Criterion
Reports to risk assessor
Documentation (SAP)
Data sources (non-analytical)
Completeness (sampling)
Comparability (sampling)
Data Collection and
Evaluation Decision
    Are all exposure
  pathways and areas
     identified and
      examined?
                                                                                    21-002-073
                                                120

-------
been appropriately sampled. Broad spectrum analyses
must also have been conducted for the media of concern
and anrJyte-specific methods used where appropriate.
The uncertainty in data collection and analysis depends
on the evaluation of completeness, comparability and
representativeness as discussed in Section 5.6. Based
on these indicators, the risk assessor should determine
the magnitude of the effect of data confidence on the
reassessment Guidance highlights forcharacterization
of exposure areas for risk assessment are:

   • Use the data but note the level of confidence
     associated with assessment of the affected exposure
     area if it is not significant.

   • Statistical interpretation  procedures  (e.g..
     sensitivity analysis) may be used if the confidence
     level associated with data for an exposure area is
     significant but does not warrant resampling and
     reanalysis.

    • If the uncertainty associated with the data is high,
     the risk assessor may determine that an exposure
     pathway or area is not fully characterized.


  6.2 ASSESSMENT OF UNCERTAINTY
      ASSOCIATED WITH THE BASE-
      LINE RISK ASSESSMENT FOR
      HUMAN HEALTH

  The level of certainty in making each of the four
  decisions discussed in Section 6.1 contributes to the
                                   overall uncertainty  in data collection  and analysis
                                   components of risk assessment. The critical factor in
                                   assessing the effect of uncertainty on the environmental
                                   analytical data component of risk assessment is not that
                                   uncertainly exists, but rather that the risk assessor is able
                                   to qualify and/or quantitate the uncertainty so that the
                                   decision-maker can make informed decisions.  The
                                   certainty levels  for risk assessment, represented in
                                   Exhibit 75, are based on the ability to quantitate the
                                   uncertainty in analytical data collection and evaluation.
                                   However, data collection and evaluation is only one
                                   source of uncertainty in the  risk assessment.  Other
                                   components of the  risk assessment process, such as
                                   toxicity of  chemicals and  exposure  assumptions,
                                   influence the four decisions to be made and contribute
                                   significantly to  the uncertainty of the baseline risk
                                   assessment.

                                   The most quantitative level of risk assessment occurs
                                    when the  uncertainty in data can be determined
                                    quantitatively.   The next  level occurs  when the
                                    uncertainty can be determined qualitatively, or the
                                    impact of the uncertainty is assessed using sensitivity
                                    analysis. The least desirable situation occurs when the
                                    uncertainty in data is unknown. This situation can occur
                                    if the minimum requirements given in Chapter 5 for the
                                    data useability criteria have not been achieved.

                                        «•  The primary planning objective is that
                                         uncertainty levels are acceptable, known
                                         and quantitatable, not that uncertainty be
                                         eliminated.
                 EXHIBIT 74. DATA USEABILITY CRITERIA AFFECTING
                           EXPOSURE AREA CHARACTERIZATION
     Worksheet
     Reference
              Data Useability
                 Criterion
Data Collection and
Evaluation Decision
        1
        2A
        2B
        2C
        3A
        3B
        6A
        6B
        6C
        6D
Reports to risk assessor
Documentation (SAP)
Documentation (SOPs)
Documentation (field records)
Data sources (analytical)
Data sources  (non-analytical)
Completeness (sampling and analytical)
Comparability (sampling and analytical)
Representativeness (sampling and analytical)
Precision (sampling)
Are all exposure areas
  fully characterized?
                                                                                            21-002-074
                                                  121

-------
                   OF THE RISK ASSESSMENT
Decisions To
  Be Made
     What
contamination is
 present and at
  what levels?
    Are site
 concentrations
   sufficiently
 different from
 background?
Are all exposure
 pathways and
areas identified
and examined?
    Are all
   exposure
   areas fully
characterized?
Risk Assessment
     Process
  Data Collection
  and Evaluation
    Exposure
   Assessment
     Toxicity
   Assessment
       Risk
  Characterization
Nature of Risk
 Assessment
                               Quantitative
                               (uncertainty
                              explicitly stated)
                                                                Quantitative
                                                               (uncertainty not
                                                                   known)
                                                                Qualitative (no
                                                                  uncertainty
                                                                  estimate)
                                                                        21-002-07S
                                   122

-------
                         Appendices
1.   DESCRIFHON OF ORGAN1CS AND INORGANICS DATA REVIEW PACKAGES	125

II.   LISTING OF COMMON POLLUTANTS GENERATED BY SEVEN INDUSTRIES	153

III.  LISTING OF ANALYTES. METHODS, AND DETECTION OR QUANTITAT1ON LIMITS FOR
    POLLUTANTS OF CONCERN TO RISK ASSESSMENT	167

IV.  CALCULATION FORMULAS FOR STATISTICAL EVALUATION	235

V.   T DATA QUALIFIER SOURCE AND MEANING	239

VI.  "R" DATA QUALIFIER SOURCE AND MEANING	245

VII. SUMMARY OF COMMON LABORATORY CONTAM IN ANTS. CONCENTR ATION REQUIRE-
    MENTS. AND RISK ASSESSMENT IMPLICATIONS	249

VIII. CLP ANALYTICAL METHODS SHORT SHEETS AND TCL COMPOUNDS	253

IX.  EXAMPLE DIAGRAM FOR A CONCEPTUAL MODEL FOR RISK ASSESSMENT	263
                                 123

-------
\
i
                                                    APPENDIX I
                   DESCRIPTION OF ORGANICS AND INORGANICS DATA REVIEW PACKAGES


                   The purpose of Appendix I is to familiarize the reader with a model for data review
              deliverables.  This appendix consists of the following items:

                  o  A description of the data reporting format,

                  o  An example  of a data review summary, and

                  o  Example data review forms.

                   Please note  that the example  forms are designed for the validation of Contract
              Laboratory Program (CLP)  data packages.  An example  form is included for each analytical
              fraction (volatiles, semivolatiles, pesticide/Aroclors and  metals) and for samples from
              soil/sediment and aqueous matrices. These forms nevertheless include the necessary
              information for the review  of most types of data (analytical results, sample
              quantitation/detection limits, data qualifiers, etc.) not associated with the CLP.

                                          1.  DATA REPORTING FORMATS

                   Whenever an analytical laboratory is requested to  analyze field samples for a specific
              site, the RPM (in consultation with the technical project team) must ensure that the laboratory
              will provide adequate documentation to support all current and future uses of the data.
              Potential uses of the data can include data validation, monitoring, modeling, risk assessment,
              site characterization, Record  of Decision defense, enforcement, and litigation.

                   Data packages produced by analytical laboratories should contain all the documents that
              were produced or used by the laboratory for that particular analysis.  The required 'documents
              should include a narrative (detailing the exact method performed, deviations from the method,
              problems encountered,  and problem resolution), chain-of-custody records, laboratory logbook
              pages,  and raw data and tabulated  summary forms  for all standards, quality control and field
              samples.

                   The documents should  be organized in a logical manner and the entire data package
              should be paginated. Generally, the laboratory should be required to produce a data package
              with documents ordered in the following manner

                     1)     Narrative
                     2)     Tabulated summary forms for  laboratory standards and quality control samples
                            (in chronological order by type of quality control  sample/standard by date of
                            analysis by instrument)
                     3)     Tabulated summary forms for  field  sample results (in increasing RAS, SAS, or
                            project  sample number order)
                     4)     Raw data for  field samples (in increasing RAS, SAS, or project sample number
                            order)
                     5)     Raw data for  laboratory standards and quality control samples (in chronological
                            order by type of quality control sample/standard by date of analysis by
                            instrument)
                     6)     Laboratory logbook pages
                     7)     Chain-of-custody records
                                                          125

-------
                                   APPENDIX I (continued)

     It is often convenient to require that the laboratory data package resemble as closely as
possible the data packages required by the current CLP RAS SOWs for organics and
inorganics, that the tabulated summary forms provided in those SOWs be utilized and modified
appropriately, and that the data qualifiers in those SOWs be applied to the data as appropriate.
The following sections describe specific requirements for the content of each document
contained in the laboratory data package.

NARRATIVE:

      A narrative must be provided describing the analytical methods and exact procedures
performed by the laboratory, as well as any deviations from the method.  Problems
encountered during analysis,  problem resolution and any factors which may affect the validity
of the data must be addressed. The  narrative must include the laboratory name and RAS,
SAS, or project sample numbers cross-referenced to the laboratory sample identification
numbers, and must be signed and dated by the laboratory manager.

      Any telephone communications between the laboratory and sampling personnel (or other
parties outside of the laboratory) to  resolve sampling discrepancies or analytical problems must
be documented in detail on telephone communication logs.  Those telephone logs must
explicitly detail the problems requiring resolution, the agreed to resolution,  and the names and
affiliations of the communicating parties.  All telephone logs must be appended to the
narrative.

       An example calculation of a positive hit and a detection/quantitation  limit for each type
of sample analysis must  be provided. All equations, dilution factors and information required
 to reproduce the laboratory results must be provided.

 TABULATED SUMMARY FORMS:

 Laboratory Standards and Quality Control Samples

       Tabulated summary forms  must be provided for all  laboratory standards, tunes, blanks,
 duplicates, spikes, and any other types of laboratory quality control samples/standards.  The
 tabulated summary forms must contain information pertinent to the type of laboratory quality
 control sample/standard which was  analyzed.  Typical entries include:  concentrations spiked,
 concentrations detected, spike compound names, results of statistical calculations (%R, %D,
 RPD, RSD,  CV, RRF, SD, etc.), sample identification numbers, dates/times of analysis,
 instrument IDs, lab file IDs, and QC limits.

       The exact format of each tabulated summary form will depend on the particular analysis
 method requested and the quality control procedures specified in that method.  However,
 comprehensive tabulated summary forms must be prepared for all quality control
 samples/standards analyzed  by the laboratory. For example, typical tabulated summary  forms
 for volatile  organics analyses include but are not limited to:

 Surrogate results:  Tabulate the sample identification numbers, surrogate compounds added,
 concentration added, percent recoveries, and QC limits for all standards, blanks, quality
 control samples and field samples.  Flag outliers.

 Matrix spike and matrix spike duplicate results:  Tabulate the matrix spike compounds added,
 concentration added, percent recoveries and relative percent differences for the spiked
 compounds, and  QC limits.  Flag outliers.  List the sample identification numbers.  Results for


                                             126

-------
                                 APPENDIX 1 (continued)

all non-spike compounds must be tabulated on the form used to summarize field sample
results.

Method/laboratory blanks:  Tabulate the sample  identification numbers, lab file IDs, and time
analyzed for field samples and matrix spike samples  which pertain to each blank on a separate
form.  The form must also contain the GC column, instrument ID, laboratory sample
identification number, lab file ID, and date/time of  analysis for the  blank itself.  Results for
each blank must also be tabulated on the form used  to summarize field sample results.

Tuning results:  Tabulate the m/e, ion abundance criteria, and percent  relative abundances and
list the  tune compound name, instrument ID, lab file ID, and date/time of injection which
pertain  to each tune analysis on  a separate  form. The form must also contain tabulated sample
identification numbers, lab file IDs, and date/time of analysis for all field samples, matrix
spike samples, blanks, and standards which pertain to that tune. Fag outliers.

Initial calibration results:  Tabulate the target compound names, relative response factors for
each target and  surrogate compound at each standard concentration,  mean relative response
factors  and  percent relative standard deviations  for all target  and surrogate compounds, and
QC limits for each  initial calibration on a separate form.  The form  must also contain the
concentration of the calibration  standards,  instrument ID,  lab file IDs,  and dates/times of
standard analyses for that initial calibration.  Flag outliers.

Continuing  calibration results:  Tabulate the target compound  names, mean relative response
factors  from initial calibration, relative response factors from  continuing calibration, percent
differences, and QC limits for all  target and surrogate compounds for each continuing
calibration on a separate form.  The form  must  also  contain the concentration of the
continuing calibration standard, instrument ID,  lab file ID, and dates/times of initial and
continuing calibration standard analyses which pertain to that continuing calibration.  Flag
outliers.

Internal standard results:  Tabulate the sample identification numbers,  internal standard
compound names, QC limits, retention times and area counts of the  quantitation ion for each
internal standard compound in the continuing calibration  standard and all field samples,
matrix  spike samples, and blanks which pertain to that continuing calibration on a separate
form.  The form must also contain the instrument ID, lab file ID, and date/time of continuing
calibration standard analysis. Flag outliers.

MDL study results: Tabulate the  target compound names, concentrations spiked and detected
for each MDL spike analysis, and the standard  deviation and  calculated MDL for each target
compound. (Note:  The narrative must explain  the  MDL  procedure utilized to generate the
 values.  The formula and associated constant values  utilized in the calculation of the MDL  for
each analyte must be provided.  The column, instrument  ID, trap composition, and operating
conditions must be clearly displayed on the raw data.)

 Field Samples

The exact format of the tabulated summary form for each field sample will depend on the
particular analysis  method requested.  However, comprehensive tabulated summary forms  must
 be prepared for each field sample analyzed by  the laboratory.  At a minimum, the target
compound names, concentration units, positive  hits  and numerical detection/quantitation limits
and any laboratory qualifier flags for each target compound must be tabulated on a separate
 form.  Definitions must be provided for all qualifier flags used by the laboratory. For each


                                              127

-------
                                 APPENDIX I (continued)

sample, the tabulated form must also contain the RAS, SAS, or project sample identification
number, laboratory name, laboratory sample ID, lab file ID, sample matrix type, and level of
analysis (low, medium, high).  The percent moisture/solids, weights and volumes of sample
prepared/purged/extracted/digested/analyzed, initial and final extract/digest and extract
clean-up volumes, injection volume, clean-ups performed, dilution factor, measured pH, and
dates that sample was received/extracted/digested/analyzed should  be  included as appropriate
to the analysis method.

RAW DATA:

Raw data must be provided by the laboratory for all laboratory quality control samples,
blanks, spikes, duplicates, standards, and field samples.  The exact format and content of the
raw data will depend on the particular analysis  method requested.  However, any and all
instrument printouts, strip chart recordings, chromatograms, quantitation reports, mass spectra
and other types of raw data generated by the laboratory for a particular project must be
provided in the data package. Typical raw data for organic GC/MS analyses includes but  is
not limited to:

       o      Reconstructed total ion chromatograms,

       o      Instrument quantitation  reports containing the  following information:
              laboratory sample identification number, RAS, SAS or  project sample number,
              date and time of analysis, RT and/or scan number of quantitation ion with
              measured area, analyte concentration, copy of area table from data system,
              GC/MS instrument ID, lab file ID, column, trap composition, and operating
              conditions,

        o      Raw and  enhanced mass spectra for all positive field sample results  and daily
              continuing calibration standard reference spectra for all positive field sample
              results,

        o      Mass spectra and three library searched best-match mass spectra for all
              tentatively identified compounds reported, and

        o      Instrument  normalized mass listing and the mass spectrum for each  tune.

Typical raw data  for inorganic analyses includes but is not limited to:

        o     Instrument printouts and strip chart  recordings containing the  following
              information:  laboratory sample  identification  number, RAS, SAS or project
              sample number, date and time of analysis, absorbance/emissions values, analyte
              concentration, instrument ID, lab file ID, and operating conditions, and

        o     Standard curve raw data, plotted standard curves, linear regression  equations,
              and correlation coefficients.

 LABORATORY LOGBOOK PAGES:

       Copies of standards preparation  logs, sample  preparation/extraction/digestion logs,
 sample analysis run logs, personal logs, and any hand written project-specific notes must be
 included.  The initial and final volumes of sample  prepared/purged/extracted/digested, initial
                                             128

-------
                                   APPENDIX I (continued)

and final extract/digest and extract clean-up volumes, injection volumes, and dilution factors
must be clearly labelled.

CHAIN-OF-CUSTODY RECORDS:

      All chain-of-custody records provided to the laboratory during sample shipment or
generated by the laboratory during sample receipt, storage, preparation, and analysis must be
included.  Chain-of-custody records include but are not limited to: signed and dated field
chain-of-custody forms, signed and dated shipping airbills, sample tags, SAS packing lists,
RAS Traffic Reports, internal laboratory receiving  records, and internal laboratory
sample/extract/digest transfer records.
                                             129

-------
                     APPENDIX I (Continued)

                   2. DATA REVIEW SUMMARY

            ORGANIC DATA SUMMARY FORMS UTILIZED
                    BY REGION III IN THE CLP

   DATE:

SUBJECT:


   FROM:


     TO:                                      ~


   THRU:


OVERVIEW

Case       consisted of four (4) low level water and two  (2)  low
level soil samples, submitted for -full organic analyses.  Included
in this data  set was one (1) equipment blank and one (1)  trip
blank.  The trip blank was analyzed for volatiles only.   The
samples were  analyzed as a Contract Laboratory Program  (CLP)
Routine Analytical Service  (RAS).

SUMMARY

All samples were successfully analyzed  for all target compounds
with the  exception of 2-Butanone and 2-Hexanone in  the  volatile.
fraction.  All  remaining instrument and method sensitivities  were
according to  the Contract Laboratory Program  (CLP)  Routine
Analytical Service  (RAS) protocol.

MAJOR PROBLEM

The  response  factors (RF) for 2-Butanone  and  2-Hexanone were less
than 0.05 in  one of the continuing volatile calibration.   The
quantitation  limits for this compound  in  the  affected samples
were qualified unreliable,  "R".   (See  Table I in Appendix F  for
 the  affected  samples.)

 MIKOR PROBLEMS

 Several conpounds  failed  precision criteria for initial  and/or
 continuing,.calibrations.  Quantitation limits and the reported
 results for these  compounds may be biased and, therefore, have
 been qualified estimated,  "UJ"  and "J'«, respectively.  (See  Table
 I in Appendix F for the affected samples) .
                               130

-------
                     APPENDIX I (Continued)

                   2. DATA REVIEW SUMMARY


                                               Page 2 of 3
ftOTES

  o  The soil semivolatile MS/USD analyses were originally
     extracted within the technical and contractual holding
     times.  Re-extractions were required because of surrogate
     recoveries, and these re-extractions were performed outside
     of  holding times.  Surrogate recoveries were again outside
     of the QC limits, therefore,  original sample results are
     being reported.

  o  The maximum concentration of compounds found in the trip
     blanks, field blanks, or method blanks are listed below.
     All samples with concentrations of common laboratory
     contaminants less than ten times (<10X) the blank
     concentration, and uncommon laboratory contaminants less
     than five times (

          Methylene chloride *               7 J
          Acetone *                          9 J

          Bis(2-ethylhexyl)phthalate *      10 J

          *    Common Laboratory Contaminant


  o  The semivolatile MS/MSD analyses had compounds other than
     the spiking compounds present.  The following is a table of
     results and precision estimates for the  non-spiked
     compounds:
                 ffS/MSD Non-Spiked Compounds
                                     Concentration  fucr/Iil
      Compound
Phenanthrene                       150 J    190 o      140 J   16.5
Fluoranthene                       340 J    470 J      440 J   16.3
Benzo(a)anthracene                 290 J    310 J      320 'J    5.0
Chrysene                           290 J    330 J      300 J    6.6
Bis  (2-ethylhexyl) phthalate       160 J    200 J      240 J   20.0
Benzo  (b)pyrene                    190 J    240 J      240 J   12.9
Benzo  (k) pyrene                   230 J    200 J      220 J    7.1
Benzo  (a) pyrene                   240 J    190 J      240 J   12.s

      RSD-  Relative Standard  Deviation
                            131

-------
                    APPENDIX I (Continued)

                  2. DATA REVIEW SUMMARY
                                                      Page 3 of 3

 o The pesticide/pea analyses of all soil sanples and associated
   QC samples had surrogate recoveries in excess of the QC limit.
   Since no positive results were reported for any pesticide or
   PCB compounds for any of the samples in this case no data was
   affected. (See Appendix F).

 o The reported Tentatively Identified Compounds (Tie's) in
   Appendix D have been reviewed and accepted or corrected.

 o All data for Case       were reviewed in accordance with the
   Functional Guidelines for Evaluating Organic Analyses with
   modifications for use within Region III.  The text of this
   report addresses only those problems affecting usability.
ATTACHMEKTS

APPENDIX A - Glossary of Data Qualifiers
APPENDIX B - Data Summary.  These 'include:
      (a)  All positive resales for target compounds with
          qualifier codes where applicable.
      (b)  All unusable detection limits (qualified "R").
APPENDIX C - Results as Reported by the Laboratory for All
                 Target Compounds
APPENDIX D - Reviewed and Corrected Tentatively Identified
             Compounds
APPENDIX E - Organic Regional Data Assessment Summary
APPENDIX F - Support Documentation
                             132

-------
                              APPENDIX I (Continued)

                              3.  DATA REVIEW FORM
o
   9
   I
l



















s
* locati
vt




















1
to
vt

















.
{(
"
1
u
4.




















•ft
Jf
!




















•
5
2
f




















*
•>*
«
O
«A
y
|















^
!
_i
o

•1
0
>
2
w
I
1








,











ir.5lssS i|t_!iMj, ,
3 ** <• •• t V «^ ft* 4 5 * ' •* % ^ W •* •* ^ ^ V C C *V
^ ^E 4 • • u LJ O O U *• MI Y v v x «v wt wt i/f M 5 *•« « «




I
v
c
1
2
VI
a
I
0
*«
it
•o
K <^
O w
1 -3 -D
— «^ O
»• V t. M
« c — ••
£ o a o
•- W > 3C

• •
Ul
?s
£40
** < *. ~
«iH
^sl'-J
•l!2^2i
X 3 00 0
-5 — 0
< fc & U O

















O «l
•< O VI
r ex
o * o —
0 - 7< « C
a » x< «
x — K O O
b C C u T>
?i!!I
O = — •— 
-------
                               APPENDIX I (Continued)




                              3. DATA REVIEW FORM
o
z



















g
1



















1
1
V*



















I
•



















M
M



















I Sampling date



















lAnalysii Date



































_j
D

*
1 Inorganic Analyte
























1
Cl
c
•o
I
M
§
a
I
o
*•»
•o
o
S O
ll
O ~
u o
" a
^ •«
•• O t M
0 U X
£ O ^J O
- U > Z

li/
Analytical Method
F Furnace AA N01
P ICP/ flame AA
CV Cold Vapor
C Colorlmetrle







'









I 5


-------
APPENDIX I (Continued)




3.  DATA REVIEW FORM
o
UJ
0
<
o.






"i*
«
d
j
<
5
§




£ !
*J 2
| |

o 5
o C
« 8
z C
3










•
f>*
•J
bi
*-










e
n
r







c

*
W




















































































s
•*
•
8
^
I
•A



































1
f
































•^

»
1
«
w
'S
V



































M
1



































b
•«
2
?
i































o
u

w
«>
£
1
w
s
«
1































o o o o>n«AOOoiAO«^om^oofy%oooooo
o^«»o O«*V«MO o *" • •* o *- o •- y^ r^ •-
•A IA Irt •/»



r.» luK. IS !». ss s
iiilliililiillllllllilil


a ">
— k.
S "o
i.
0 C
i. O
I X
u *>
» 2
I 5
I I
*_ «-
' i
1 2
** c
c o
1 T: .
•1 *- M
§" a
w. J3

a ~ j:
— o>
_ 0 —
E "$
2 ."° ^
*• »
i si ;
T3 X
0 C 0 -0
w O C. 0
?— Q. *•
*- S «-
a « p
x ••» Q.
O •- M O
&u c
. a •* o
0 "S D w u
— JJ w— •»

*- t/i ** >. O -
O M « — L.
o > ec o 3c «/>

^ o
•o * t."C
° 28."
z w >» •-
C *v T> O
— <_ a. — —
« 3 U O O
W tfc •- U O
K.B»
 135

-------
                              APPENDIX I (Continued)


                             3.  DATA REVIEW FORM
 o
I
R
o
II
I 3
             f
norgtrtl
                             gS°
                             •o«-
                                          —oo
                               136
                                                               1   £

                                                               1   £

                                                               *%l
                                                               w  O i-.  4*
                                                               m  ~- O.  **

                                                               I  5*  a
                                                                i > « o at vi
                                                                  S ^
                                                            <• 3 U O 0
                                                            U •. —0 U

                                                            •«
                                                            £



                                                            < «. a. u u

-------
                          APPENDIX I (Continued)


                          3.  DATA REVIEW FORM
p.
   i
1
               1
                                                              $5
                                                              i!
                                                              2S

                                                              il
                                                              :5$
                                                              !2"2
                                                             III!

                                                             ** * O M
                                                             U *'•>'•<-
                                                             • — —
                                                             w »* *• t*

                                                             trgi.2

                                                             S555
                                    !i
                      i|i-*
   i     ;   i
    f !  !  it
HiU 1  Ml
                           137

-------
APPENDIX I (Continued)




3.  DATA REVIEW FORM
.
fci
o
a.






'•a
!
.J
a.
j
I
\
£
5
1
3
CLFVOLA1









»-













?

1






















SDONa___



i
s































































location
«
I
M

























I
1
Vt
























i
1
b

























M
I

























u
•«
«
a
?
I
V*

























•
t*
•«
O
M
5.
1





















a
0

O
w
s
1
«
IvolatI























-:
*" V u O O £
j .HI s i! Ill !l ! i (i
* S2! *
•**-*« V ••••"•«•«*•»£ *~**JC»*OO£>~ A • 'a' 9 a w fM >S
V ** £ *• C O^fJZOfQbt-UujcYi.ui. • t- J* A .« V £ ? ^
S Ji* ^* 3f * jf W O *J *• O C *• ^•M*JMOO«wl|^O>>C£€NI^24#*U
n ^ SS*^*************S U *"" *"*€•€• ^JS**O * C O 2 C •


3
• if
> *
•V
*- —
o
o *-
tg
0 >.
** ~*
«l
— «»
•a c
i Identlfle
ration* Ide
Jl
So
6 **
— "*•!
- 2 e
sli
ill

« J S •>
w a « M

i??S
S5?^
u o o >
g -,
0 -« 3 «




   138

-------
                                  APPENDIX I (Continued)

                                                   •


                                 3.  DATA REVIEW FORM
c
     8 3
                                                                                  2


                                                                                  1
                                                                                  u

                                                                                  >.
                                                                                 •zt
>• a




^J


go
                                                                                 5~
                                                                                 o a M





                                                                                 III

-------
APPENDIX 1 (Continued)




3.  DATA REVIEW FORM
*%
1
w
Vt
X
S
•-
<
U
AQUEOUS SAHPL






•
o
3t
U
<

M U
>• ac
I £
u ** •
-
£ o u
gar IA
u ^
u u
_l
*-
<
a.
u



















































S
M
|
•
}
vt
























u
•
1
I*























1
h*
1
U
•^
1^
<*•
•
k
























•
^
I
























•
**
•
o
?
"»•
1
V*
























S
S
tte
S
5
o






















J
J
u
1
1
*
«^
•
^
e
>





















-
^
• S g
£ . | „ I 1
* ..r , 1* 2?| i| ^ | §
~ -Sj=*:je jc S2 «o°k««S — * Sf
h*.l =828 8 Sf.I^IIs € £ 5S s £
J2fi slllgl,€ = H5|«|»= o J Jg i{ I
if«sic!i««i«h^i«shsP^MA.ii.e
llxll'Sl335!3?-!^^-^!^? J,^5SaM-|2£S8
|gi|«-€iAA|A2^«&I^iJiJ5--SSlsSt^32^fc5
5i>fi^53-V--.•5^A^a>i^•CtS^«•:i!i^^^25£S^

>
«/
°>
t_
2
g
u
0
J=
w
si
2:5
"i
« —
go
rl
L. «J
og
t.-S
A X

** cx
*i «
«• •
•"• •*- T)
** 6 *
-s8
J-« k
*<
*—
• ^"^
II?
5«



    140

-------
                                    APPENDIX I (Continued)


                                    3. DATA REVIEW FORM
  o
  <
  CL
M
        .
 rv
           1

            I
   i
              •  — •*  —
I  H
3  £  i

1  I


                                        -
                                        141

-------
                                 APPENDIX I ^Continued)

                                 3.  DATA REVIEW FORM
  5
  i
S t3
.M

                                                                            >
                                                                            «- "o
*s
fc-o

                                                                           EOT)
                                                                           — «< •>


                                                                           :.|

                                                                           |g«
                                                                           grr-

-------
                               APPENDIX I (Continued)

                               3. DATA REVIEW FORM
 5
i
  s
  <
  i
1
2
u
!
                                                        4><-«MfMa«f
                               'Jl'i
                                  i  S  Z«-^
                              CM  •~Q'*^UMf£
                              TSeglgSfeV?
                                                                    «•
                                                                    2S
                                                                    i!
                                                         *i
                                                         ?i
                                                         ^2
                                                         ll
                                                         ll
t 1 tat I on Lim
tnate due to
approximated
                                                        Is si

                                                                   S--3'
                                   143

-------
                                 APPENDIX I (Continued)
                                                 •
                                 3. DATA REVIEW FORM
i
           I
                                                                            s
                                                                            u
                                                                            x
                                                                          « C
                                                                          i*
                                                                          32

                                                                          £-
                                                                          12

                                                                          II
                                                                          i.2
                                                                          52 .
                                                                          ~«S

-------
11
  i  S
  i  I
  3
                                  APPENDIX I (Continued)



                                  3.  DATA REVIEW FOR*M
Saaple Location
              i
I
                                                                                §


                                                                                X
                                                                              u
                                                                              >• a
                                                                              la
                                           ? i
                                  JS
                                  *«







                               ^3  «7
                                     us

-------
                                     APPENDIX I (Continued)


                                    3. DATA REVIEW FORM
u
2
       *
       8
 _1
 6
 a
                 *#
                 i
            §
            5
            «k
            s
                                •
           1
I
I
                             a
                             g
                                                                                  o
                                                                                  fc.
                                                                                  **
                                                                                  I



                                                                                  1
                                                                   • **
                                                                  .= £

                                                                  I2
                                                                                s

                                                                                til
                                                                                •D-
                                                                                  SO
                                                                                  »*
                                                                                  &
                                                                                  SM

                                                                                   •
                                                                                    »
                                                                                — Z C)

                                                                                III
                                                                                M O >

-------
                                   APPENDIX I (Continued)



                                   3.  DATA REVIEW FORM
 o
•a
•c
J

e
C
                              o
                              oc
                              u
epo
 
-------
APPENDIX 1 (Continued)




3-  DATA REVIEW FORM
1
°l
•u
s









•a
1
g
|
§
i.

i 5
5 ^
»i 3
s I
5 '
i 1
g J
i c
i §
1
5
V











«v
9
•A
*
t*
1















s
S



g
2
2



























































































ocation
•«
^
f
a
VI



































|
X
a
u*


































i.
w
1
w
•
I
*w




































•A
M
\
H
•E



































•
*«
4
O
•*•
1
Irt



































o
1
M

;




































•
•
o
•;
1
<































3
3

|
U
*
«*
•
1
•
5"

—






























oogoooooooooooooooooooooooooooog
*• '^ *S *^ *^ '^ '^ *^ '^ ^ ^* '^ '^ '^ *^ ''^ *^ '^ '^ '^ '^ *^ *^ *^ *^ *^ '^ '^ '^ '^ ^^ '^
4* ^ i. £
5 £ ^ ^ *
1 |.2S .1 S. §2
g""* vt^A 4f *>>£ C C U £ «•
^- • — v««**«s jc ^ J5 • tiCa««i7CS
|!!|HI! Ifllfhli If
W Q.^ ^*f***>kOwC*w>*^B£4|CeVA4%AA*^fi« O^^^OM^
•*3O**CO^"*V**O^w« • • fc. • 4.C^«OWff^* OOOC CO
m«M40fMO^«.tf^«4Ka.£< o«.«. 2 m.S^oa««-o<*


•
• 9
2.2
xS
<• i*
k
•*
-* o
e *-
ii
:t
*« •»
!!
t*
11
"81
rl
««
1?
II
••
12
.;i
M *•
— 41
«!
g.g
ll!
«r fc —
°l- •
?*«^e
k «l — **
•= — — «
^ss?
*z- *•
» « • •
U ** */ «M
2 ~'S »
e S S 5

u o o >
rs-
u






  148

-------
                                APPENDIX I (Continued)



                                3. DATA REVIEW FORM
•a
     g
     §
|


£
                                                                       i
                       go
                       w <-
                       tg
                       — w

                       u

                       II
                       •8?
                                                                     II
                                                                      gg
                                                                     J £

                                                                      o e
                                                                    •-2J

                                                                      SS
                                                                     a—8


                                        H


                                        *~
^  M *• C
   _oo
S  S SS
                                                                     8 33c

                                   149

-------
                                   APPENDIX I (Continued)


                                   3.  DATA REVIEW FORM
1
y
o
g
§
                                                                                    2.2
                                                                                    25
                                                                                    a*
                                                                                    8--S
                                                                                     o
                                                                                    COO
                                                                                    030
                                                                                    VI O >
                                         ISO

-------
                                     APPENDIX I (Conti-ued)



                                     3. DATA REVIEW FORM
  I
  I
1
      e!
  O
  a

  3
 •o

 ra
I
                                         ISI

-------
                                   APPENDIX I (Continued)


                                  3.  DATA REVIEW FORM
o

a
 >

 E

 1
 I
 «;
 _>
 1


 e
 z
 I
<
<\<
       i
       8
2
2
f
c
K
«;
c

c
     «
     *
     I
                I
1
o

I  2
                                                                            2
                                                                            ti
                                                                          •o —
                                                                            go
                                                                           w
                                                                     « K


                                                                     Is
                                                                     sr

                                                                     ** «j .
                                                                     is
                                         1S2

-------
                                      APPENDIX II
      LISTING OF COMMON POLLUTANTS GENERATED BY SEVEN INDUSTRIES
      Appendix II identifies seven industries that generate waste which contains pollutants that
are known to pose human and environmental hazards.  This appendix is intended  to aid the
reader in three ways:

    o   To assist in the identification of target compounds and potential exposure pathways.

    o   To predict associated contaminants that potentially yield interferences.

    o   To assist in early identification of sites that contain high levels of compounds that
        may not be included as target analytes for routinely available methods.

The data for these tables were obtained by searching the USEPA Toxic Release Inventory
System using the Standard Industrial Classification (SIC) codes listed below:

       Industry                                         SIC Code

       1      Battery Recycling                         3691, 3692
       2      Munitions/Explosives                      2892
       3      Pesticides Manufacturing                  2842, 2879
       4      Electroplating                            3471
       5      Wood Preservatives                       2491
       6      Leather Tanning                          3111
       7      Petroleum Refining                       2911

      The appendix consists of seven tables and depicts the pollutants associated with each of
the seven industries, the CAS number of each pollutant, and the matrices where each pollutant
has been  found.   The list is  not  inclusive of all pollutants or industrial  sources.   The seven
industries were selected based on the  recommendation of the Risk Assessment Subgroup of the
Data  Useability  Workgroup because of the  frequency of occurrence of the pollutants produced
by those industries in Superfund sites.
                                              153

-------
                                                           Appendix II
                                                LISTING OF COMMON POLLUTANTS
                                                GENERATED BY SEVEN INDUSTRIES
                                                INDUSTRY 1: BATTERY RECYCLING
Rank
1
2
3
4
5
6
7
I
9
10
It
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
«*«*
34
35
Compound
LEAD
SODIUM SULFATE (SOLUTION) j
SODIUM HYDROXIDE (SOLUTION)
SULFURICACID
AMMONIUM SULFATE (SOLUTION)
MANGANESE
l.l.l-TRICHLOROETHANE
METHANOL
FREONH3
TRICHLOROETHYLENB
TOLUENE
ZINC
AMMONIA
CADMIUM
ANTIMONY
BARIUM
NICKEL
FORMALDEHYDE
ACETONE
XYLENE (MIXED ISOMERS)
TETRACHLOROETHYLENE
DICHLOROMETHANE
PHENOL
MERCURY
N-BUTYL ALCOHOL
METHYL ETHYL KETONE
METHYL ISOBUTYL KETONE
HYDROCHLORIC ACID
?™TWaiLOROETHANE (METHYL CHLOROFORM)
COBALT
ARSENIC
COPPER
SILVER
ACETONITRILE
CAS Number
7439921
7757826
1310732
7664939
7783202
7439965
71556
67561
76131
79016
108883
7440666
7664417
7440439
7440360
7440393
7440020
50000
67641
1330207
127184
75092
108952
7439976
71363
78933
108101
7647010
7697372
71556
7440484
7440382
7440508
7440224
75058
Air
Y

Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Water Soil
Y Y
Y
Y
Y
Y
Y Y
Y
Y

Y

Y Y
Y
Y Y
Y
Y
Y Y






Y
Y
Y

Y
Y

Other
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

 Rank - Order of Frequency of Occurrence

* Other - Other Matrice. (Biota. Hazardous Warte, Sludge, etc.)

-------
                                                 LISTING OF COMMON POLLUTANTS
                                                 GENERATED BV NT, VEN INDUSTRIES
                                                INDUSTRY 2: MUNITIONS/EXPLOSIVES
Rank
Compound
                                                                    CASNiunber
                                                                        Air
Water
                                                                                                  Soil
Other
1
2
3
4
5
6
7
8
9
10
II
12
13
|4
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
ACETONE
NITRIC ACID
AMMONIUM NITRATE (SOLUTION) ,
PENTACHLOROPHENOL ',
SODIUM SULFATE (SOLUTION)
AMMONIA
SULFURJCACID
METHYL ETHYL KETONE
CYCLOHEXANE
CHLORINE
NITROGLYCERIN
DICHLOROMETHANE
CALCIUM CYANAMIDE
LEAD
ETHYLENEOLYCOL
N-BUTYL ALCOHOL
TERT-BUTYL ALCOHOL
M-XYLENE
METHANOL
ASBESTOS (FRIABLE)
I.I.I.TRICHLOROETHANE
POLYCHLORINATED BIPHENYLS
COPPER
ALUMINUM
2.4-DINITROTOLUENE
GLYCOL ETHERS
BENZENE
BISa-ETHYLHEXYL) ADIPATE
ZINC
DIBUTYL PHTHALATE
SODIUM HYDROXIDE (SOLUTION)
DIETHYL PHTHALATE
67641
7697372
64S4522
87865
7757826
7664417
7664939
78933
110827
7782505
55630
75092
156627
7439921
107211
71363
75650
108383
67561
1332214
71556
1336363
7440508
7429905
121142
79141
71432
103231
•fj *t\£ ££
7440666
84742
1310732
84
-------
                                           LISTING OF COMMON POLLUTANTS
                                           GENERATED BY SEVEN INDUSTRIES
                                        INDUSTRY 3: PESTICIDES MANUFACTURING
Rank
Compound
                                                             CAS Number
                                                                Air
Water
Soil
Other
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
JI
32
33
34
35
36
37
38
39
40
41
42
' Rank
k
Other
SODIUM SULFATE (SOLUTION)
AMMONIA
TOLUENE
SODIUM HYDROXIDE (SOLUTION)
TITANIUM TETRACHLORIDE
METHANOL
DICHLOROMETHANE
XYLENE (MIXED ISOMERS)
CHLOROBENZENE
HYDROCHLORIC ACID
CHLOROPHENOLS
STYRENE
ACRYLONmULE
FORMALDEHYDE
CARBON TETRACHLORIDE
CHLOROTHALONIL
1 .2-DICHLOROETHANE
ACETONE
HEXACHLOROBENZENH
1.1.1-TRICHLOROETHANE
ETHYLENEOLYCOL
GLYCOL ETHERS
1.3-BUTADIENE
CHLOROMETHANE
CAFTAN
TETRACHLOROETHYLENE
CHLORINE
CARBARYL
COPPER
PARATH1ON
ZINEB
PYRIDINE
AMMONIUM NITRATE (SOLUTION)
PHOSPHORIC ACID
CARBON DISULFIDE
I ,2.4-TRICHLOROBENZENE
SULFURICACID
MALEIC ANHYDRIDE
ETHYLBENZENE
2.4-D
BROMOMETHANE
SEC-BUTYL ALCOHOL
- Order of Frequency ofOeeuneace
• Other Matrices (Biota, Hazardous Watte. Sludge, etc.)
7757826
7664417
108883
1310732
7550450
67561
75092
1330207
108907
7647010
106489
100425
107131
50000
56235
1897456
107062
67641
1 18741
71556
107211
79141
106990
74873
133062
127184
7782505
63252
7440508
56382
12122677
110861
6484522
7664382
75150
120821
7664939
108316
100414
94757
74839
78922


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y '
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y


Y
Y
Y

Y

Y
Y

Y


Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y

Y
Y


Y
Y


Y
Y
Y

Y



Y
Y
Y
Y
Y

Y




Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y

Y
Y
Y
Y
Y





-------
                                                      Appendix n
                                           LISTING OF COMMON POLLUTANTS
                                           GENERATED BY SEVEN INDUSTRIES
                                        INDUSTRY 3: PESTICIDES MANUFACTURING
                                                                                                        A
                                                                                                        f
Rank
Compound
                                                             CAS Number
                                                                Air
Water
Soil
                                                                                              Other
43
44
45
46
47
4B
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
' Rank
Other
IJ-AD
CUMENE
M-XYLENE
ASBESTOS (FRIABLE)
FRRONII3
UiaiLOKOUI-NZI-NI- (MIXliO ISOMLKS)
CYCLOHEXANE
2.4-DICHLOROPHENOL
1.4-DICHLOROBENZENE
DICHLOROBROMOMETHANE
TR1FLURAUN
I.2.4-TRIMETHYLBENZENE
METHYL ISOBUTYL KETONE
1.4-DIOXANE
NITRIC ACID
N-BUTYL ALCOHOL
FLUOMETURON
2-METHOXYETHANOL
BISa-ETHYLHEXYL) ADIPATE
PHENOL
ACRYLIC ACID
QUINTOZENE
ALUMINUM
BENZOYL PEROXIDE
0-XYLENE
CHROMIUM
2-PHENYLPHENOL
HYDROGEN CYANIDE
HEXACHLOROCYCLOPENTADIENE
DIOOFOL
BIPHENYL
4-NITROPHENOL
METHYL ETHYL KETONE
TRICHLOROETHYLENE
M-CRESOL
TETRACHLORVINPHOS
DI(2-ETHYLHEXYL) PHTHALATE (DEHP)
TEREPHTHAUC ACID
DICHLORVOS
MANEB
P-XYLENE
- Order of Frequency of Occurrence
_ Other Matrice* (Biota. Hazardou* Waste. Sludge, etc.)
7439921
98828
108383
1332214
7AI1I
/DIJII
25J2I226
110827
120832
106467
75274
1582098
95636
108101
123911
7697372
71363
2164172
109864
103231
108952
79107
82688
1344281
94360
t\fA*tf
95476
7440473
«Uh 4<*<*
99437
74908
7440666
77474
115322
92524
100027
78933
79016
108394
961115
117817
100210
mm
V&iji
12427382
106423


Y
Y
Y
Y
Y
Y
Y
Y
Y
Yr
Y
Y

Y

Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y

Y

Y

Y
Y
Y
Y
Y




Y
Y
Y
Y
Y





Y


Y
Y
Y




Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y



-------
Rank
                                                             Appendix II
                                                 LISTING OF COMMON POLLUTANTS
                                                 GENERATED BY SEVEN INDUSTRIES
                                             INDUSTRY 3: PESTICIDES MANUFACTURING
Compound
                                                                    CAS Number
Air
Water
                                                                                                  SoU
                                                                                                Other
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
10!
102
103
104
105
106
107
108
109
110
111
112
113
114
US
116
117
118
119
120
METHYLENE BROMIDE
CHLORAMBEN
BENZENE
HYDROGEN FLUORIDE
ETHYLENE
C.I. ACID BLUE 9. DISODIUM SALT
DIMETKrXSULFATE
ISOPROPYL ALCOHOL
HYDRAZINE
VINYL CHLORIDE
METHYLENEBIS(PHENYUSOCYANATE)
EP1CHLOROHYDR1N
PROPYLENE
NITRILOTRIACETIC ACI D
ARSENIC
NAPHTHALENE
VINYUDENE CHLORIDE
TRICHLORFON
DIBUTYL PHTHALATE
ANILINE
METHOXYCHLOR
DIETHANOLAMINE
NITROBENZENE
CYANIDE COMPOUNDS
AMMONIUM SULFATE (SOLUTION)
UNDANE
POLYCHLORINATED BIPHENYLS
PROPYLENI OXIDE
2.4-DINmU>PHENOL
PHOSGENE
HEXACHLOROETHANE
CADMIUM
ETHYLENE OXIDE
BENZYL CHLORIDE
4.6-DINITRO-O-CRESOL
CHLOROBENZILATE
74953
133904
71432
7664393
74851
3844459
77781
67630
302012
75014
101688
106898
115071
139139
7440382
91203
75354
52686
84742
62533
72435
111422
98953
57125
7783202
58899
1336363
75569
51285
75445
67721
7440439
75218
100447
534521
510156
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y

Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y

Y
Y
Y
Y



Y

Y


Y




Y

Y


Y
Y
Y
Y
Y
Y
Y



Y




Y



Y
Y


Y




y



Y


Y


Y
Y Y



y
Y

Y


y




Rank - Order of Frequency of Occurrence

Other - Giber Matricec (Biou. Hazardous Wane. Sludge, etc.)

-------
                                                             Appendix II
                                                 LISTING OF COMMON POLLUTANTS
                                                 GENERATED BY SEVEN INDUSTRIES
                                                  INDUSTRY 4: ELECTROPLATING
Rude
         Compound
                                                                    CAS Number
                                                                                  Air
Water
Soil
                                                                                                          Other
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
SULFUKICACID
HYDROCHLORIC ACID
SODIUM HYDROXIDE (SOLUTION)
1 .1 .1-TRICHLOROETHANE
SODIUM SULFATE (SOLUTION)
NITRIC ACID
DICHLOROMETHANE
^itf+mfd
NICKEL
TRICHLOROETHYLENE
CHROMIUM
TETRACHLOROETHYLENE
METHYL ETHYL KETONE
ZINC
FREON1I3
ALUMINUM
COPPER
I1IOSI1IORICACID
TOLUENE
V CAV\
LcAI)
XYLENE (MIXED ISOMERS)
ACETONE
CADMIUM
ETHYLBENZENE
ETHYLENEGLYCOL
CYANIDE COMPOUNDS
AMMONIA
FORMALDEHYDE
GLYCOL ETHERS
CHLORINE
METHANOL
ETHYLENE OXIDE
METHYL ISOBUTYL KETONE
2-METHOXYETHANOL
HYDROGEN FLUORIDE
PHENOL
1 .2-DICHLOROBENZENE
N-BUTYL ALCOHOL
TERT-BUTYL ALCOHOL
BARIUM
VINYUDENE CHLORIDE
2-ETHOXYETHANOL
ISOPROPYL ALCOHOL
7664939
7647010
1310732
71556
7757826
7697372
75092
7440020
79016
7440473
127184
78933
7440666
76131
7429905
7440508
7664382
108883
7439921
1330207
67641
7440439
100414
107211
57125
7664417
50000
79141
7782505
67561
108101
109864
7664393
108952
95501
71363
75650
7440393
75354
110805
A7MO
Wf OJV
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y .
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y



Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

 Rank - Order of Frequency of Occurrence

* Other • Other Matrice* (Biou. Hazardous Waoe. Sludge, etc.)

-------
       Rude
                                                                    Appendix II
                                                        LISTING OF COMMON POLLUTANTS
                                                        GENERATED BY SEVEN INDUSTRIES
                                                         INDUSTRY 4: ELECTROPLATING
Compound
CAS Number
Air
Wtter
                                                                                                        SoU
                                      Other
43
44
45
46
47
41
49
50
51
52
53
54
MANGANESE
HYDROGEN CYANIDE
STYRENE
TETRACHLORV1NPHOS
MELAMINE
N-DIOCTYL PHTHALATE
1.4-DIOXANE
COBALT
NAPHTHALENE
AMMONIUM SULFATE (SOLUTION)
SILVER
PROPYLENE
7439965
74908
100425
961 I 15
IOS781
117840
123911
7440484
91203
7783202
7440224
115071
Y

Y
Y


Y
Y


Y
Y
Y
Y


Y
Y


Y
Y
Y

8
        Rank «• Order of Frequency of Occurrence

        Other - Other Mairice* (BioU. Huudori WtKc, Sludge, etc.)

-------
                                                          Appendix n
                                              LISTING OF COMMON POLLUTANTS
                                              GENERATED BY SEVEN INDUSTRIES
                                              INDUSTRY 5: WOOD PRESERVATION
Rank*
1
2
3
4
5
6
7
S
9
10
II
12
13
14
15
16
17
18
19
20
21
Conpomd
CHROMIUM
NAPHTHALENE
AMMONIA
PENTACHLOROPHENOL
DIBENZOFURAN
ANTHRACENE
COPPER
ARSENIC
FORMALDEHYDE
BIPHENYL
BENZENE
DICHLOROMETHANE
I.l.l-TRICHLOROETHANE
AMMONIUM SULFATE (SOLUTION)
QUINOUNE
PHENOL
ZINC
PHOSPHORIC ACID
O-CRESOL
HYDROCHLORIC ACID
M-CRESOL
CASNmnber
7440473
91203
7664417
87865
132649
120127
7440508
7440382
50000
92524
71432
75092
71556
7783202
91225
108952
7440666
7664382
95487
7647010
108394
Air
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Water
Y
Y

Y
Y
Y
Y
Y

Y
Y


Y
Y
Y
. Y

Y

Y
Soil
Y
Y
Y
Y
Y
Y
Y
Y

Y



Y
Y






k
Other
Y
Y
Y
Y
Y
Y
Y
Y

Y


Y

Y

Y




Rank - Order of Frequency of Occurrence

Other - Other Matrices (Biota. Hazardous Wane. Sludge, etc.)

-------
Rank
                                                            Appendix II
                                                LISTING OF COMMON POLLUTANTS
                                                GENERATED BY SEVEN INDUSTRIES
                                                 INDUSTRY 6: LEATHER TANNING
Compound
CAS Number
Air
                                                                                        Water
                             SoU
                                                                                                         Other
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
AMMONIUM SULFATE (SOLUTION)
SULFUR1CACID
SODIUM HYDROXIDE (SOLUTION)
AMMONIA
TOLUENE
SODIUM SULFATE (SOLUTION)
METHYL ETHYL KETONE
XYLENE (MIXED ISOMERS)
CHROMIUM
CLYCOL ETHERS
METHYL ISOBUTYL KETONE
2-METHOXYETHANOL
ACETONE
2-ETHOXYETHANOL
N-BUTYL ALCOHOL
TETRACHLOROETHYLENE
CYCLOHEXANE
AMMONIUM NITRATE (SOLUTION)
MANGANESE
l.U-TRICHLOROETHANE
DICHLOROMETHANE
DIETHANOLAMINE
METHANOL
ISOPROPYL ALCOHOL
PHOSPHORIC ACID
ETHYLENEGLYCOL
FREON 113
PHENOL
ETHYL ACRYLATE
7783202
7664939
1310732
7664417
108883
7757826
78933
1330207
7440473
79141
108 101
109864
67641
110805
71363
127184
110827
6484522
7439965
71556
75092
111422
67561
67630
7664382
107211
76131
108952
140885
Y
Y

Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y

Y
Y

Y
Y
Y
Y
Y

Y

Y
Y
Y
Y
Y
Y
Y
Y
Y


Y



Y

Y


Y

Y Y
Y
Y
Y Y
Y

Y
Y
Y Y
Y
Y
Y
Y
Y
Y

Y
Y
Y Y


Y

Y





 Rank - Order of Frequency of Occurrence

 Other - Other Matrice* (Biou. Hazardous Wane. Sludje. etc.)

-------
                                                      Appendix n
                                           LISTING OF COMMON POLLUTANTS
                                           GENERATED BY SEVEN INDUSTRIES
                                           INDUSTRY 7: PETROLEUM REFINING
Rude
Compound
                                                             CAS Number
                                                                Air
Water
SoU
Other
1
3
4
5
6
7
S
9
10
11
12
13
14
15
16
17
IS
19
20
21
22
23
24
25
26
27
21
29
30
31
32
33
«JJ
34
35
36
IT
J/
38
39
40
41
^ t
42
Rank
Other
SODIUM SULFATE (SOLUTION)
ALUMINUM
AMMONIA
SODIUM HYDROXIDE (SOLUTION)
SULFUR1CACID
TOLUENE
XYLENE (MIXED ISOMERS)
BENZENE
METHYL ETHYL KETONE
PROPYLENE
PHENOL
DIETHANOLAMINE
ETHYLENE ,
METHANOL
CYCLOHEXANE
U.4-TR1METHYLBENZENE
ETHYLBENZENE
PHOSPHORIC ACID
CHROMIUM
METHYL TERT-BUTYL ETHER
ASBESTOS (FRIABLE)
P-XYLENE
AMMONIUM SULFATE (SOLUTION)
M-XYLENE
CUMENE
ACETONE
CRESOL (MIXED ISOMERS)
HYDROGEN FLUORIDE
0-XYLENB
NAPHTHALENE
NICKEL
CHLORINE
LEAD
METHYL 1SOBUTYL KETONE
ETHYLENE CLYCOL
MOLYBDENUM TRIOWDE
ZINC
HYDROCHLORIC ACID
GLYCOL ETHERS
BARIUM
COPPER
i.i.i-TRicHLOROETHANE
m Order of Frequency of Occurrence
m Other Matricc* (Biota. Kazardon* Wane, Sludge, etc.)
7757126
7429905
7664417
1310732
7664939
108883
1330207
71432
78933
115071
108952
111422
74851
67561
110827
95636
100414
7664382
7440473
1634044
1332214
106423
7783202
108383
98828
67641
1319773
7664393
95476
91203
7440020
7782505
7439921
108101
107211
1313275
7440666
7647010
79141
7440393
7440508
71556


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y



-------
                                                            Appendix II
                                                LISTING OF COMMON POLLUTANTS
                                                GENERATED BY SEVEN INDUSTRIES
                                               INDUSTRY 7: PETROLEUM REFINING
Rank
 MHHIMII
  43
  44
  43
  46
  47
  48
  49
  50
  51
  52
  S3
  54
  55
  56
  57
  58
  59
  60
  61
  62
  63
  64
  65
  66
  67
  68
  69
  70
  71
  72
  73
  74
  75
  76
  77
  78
  79
  80
  81
  82
  83
  84
Compound
                                                                   CAS Number
                                                                       Air
                                                                                       Water
                                                                                                Soil
                                                                                               Other
ANTIMONY
1.3-BUTADIENE
N-BUTYL ALCOHOL
FORMALDEHYDE
EPICHLOROHYDRIN
COBALT
VANADIUM (FUME OR DUST)
CUMENE HYDROPEROXIDE
TERT-BUTYL ALCOHOL
4.4MSOPROPYUDENEDIPHENOL
BUTYRALDEHYDE

CARBON TETRACKLOR1DE
STYRENE
TRICHLOROETHYLENE
MANGANESE

AMMONIUM NITRATE (SOLUTION)
CARBON DISULFIDE

POLYCHLORINATED BIPHENYLS
PHOSPHORUS(YELLOW OR WHITE)
QUINOLJNE
2-METHOXYETHANOL
1 ^DIBROMOETHANE
TETRACHLOROETHYLENE
ANTHRACENE
2.4-DMETHYLPHENOL
HYDROGEN CYANIDE
CHLOROMETHANE
NITROBENZENE
li-DlCHLOROPROPANE
CARBONYLSULFIDE
ACETONITRILE
SILVER
2-ETHOXYETHANOL
THALLIUM
FREONI13
SELENIUM
DICHLOROMETHANE
MERCURY
CADMIUM
7440360
106990
71363
50000
106898
7440484
7440622
80159
75650
80057
123728
92524
56235
100425
79016
7439965
75218
6484522
75150
107062
1336363
7723140
91225
109864
106934
127184
120127
105679
74908
74873
98953
78875
463581
75058
7440224
110805
7440280
76131
7782492
75092
7439976
7440439
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y

Y
Y


Y
Y
Y
Y
Y

Y
Y

Y
Y
Y
Y
Y

Y
Y
Y


Y
Y

Y

Y





Y
Y
Y
Y
Y

Y
Y
Y




Y
Y
Y
Y
Y

Y
Y
Y

Y

Y

Y

Y
Y
Y


Y
Y
Y
Y

Y
Y

Y
Y
Y





Y



Y
Y
Y
Y
Y



Y






Y

Y
Y
Y
Y

Y

Y
Y




Y
Y
Y

Y



Y
Y
Y

Y
Y









Y

Y

Y

Y
Y
  Rude -Order of Frequency of Occurrence
  Other - Other Mtfrke* (Biou, ffcardout Wane. Sludge, etc.)

-------
                                                            Appendix II
                                                LISTING OF COMMON POLLUTANTS
                                                GENERATED BY SEVEN INDUSTRIES
                                                INDUSTRY 7: PETROLEUM REFINING
Rank
Compound
CAS Number
                                                                                 Air
Water
SoU
Other
15
86
17
U
n
90
91
92
93
94
95
96
I.I.2-TRICHLOROETHANE
ARSENIC }
CYANIDE COMPOUNDS
CHLORINE DIOXIDE
ACRYLIC ACID
1 J-DICI ILOROPROPYLENE
1.2-BUTYLENE OXIDE
CHLOROBENZENE
1,4-DIOXANE
DI(2-ETHYLHEXYL) PHTHALATE (DEHP)
BERYLLIUM
CHLOROFORM
79005
7440382
S7I25
10049044
79107
5427S6
I06M7
108907
123911
117117
7440417
67663
Y Y
Y Y

Y
Y
Y
Y

Y

Y


Y Y
Y




Y

Y

Y
 Rude • Order of Frequency of Occurrence

 Other - Other Metric*. (Biou. Huardow Wute, Sludge, etc.)

-------
                                     APPENDIX III
  LISTING OF ANALYTES, METHODS, AND DETECTION OR QUANTITATION LIMITS
               FOR POLLUTANTS OF CONCERN TO RISK ASSESSMENT
     The purpose  of  this appendix  is to  familiarize the reader with  the  variety of EPA
methods  that  are  available for analysis of pollutants of concern  in  risk assessment.   The
appendix facilitates appropriate method selection for pollutants in the matrix of interest.

     Appendix  III consists first of a summary of  definitions of  commonly used detection
limits and quantitation limits.  Tables I. II, and  III depict detection limit estimates achievable
for 33  organic and  inorganic pollutants of potential concern to risk assessment in air, soil, and
water matrices respectively. The detection limits listed herein are provided  for guidance and
may not always be achievable.  Specific quantitation limits are highly matrix-dependent.

     Table  IV provides a  summary of each method of analysis for  these pollutants.  The 33
pollutants listed  were chosen because they are highly toxic and/or have reported cancer risks,
and occur at a frequency of greater than 2% in 141 National  Priorities List (NPL) sites.*

     Tables V-A  and V-B provide an  additional comparison of  analytical mi thodologies for
selected organic compound classes and  inorganic analytes including method  detection ranges
and the applicable analytical system and preparation procedures.
 *Sourc9:  CLP Statistical Database (STAT).
                                              167

-------
Instrumentation
CVAA«
ECO-
ELCD-
FID.
FLAME «
Fluor».
FPD«
GC»
GC-MS-
GFAA.
HPLC«
HYDAA.
ICP«
LC«
MSm
NPD-
PIO>
UV»
                                      APPENDIX Bl
                                       GLOSSARY
Cold Vapor Atomic Absorption
Electron Capture Detector
Electrolytic Conductivity Detector
Flame lonization Detector
Flame Atomic Absorption
Fluorescence
Flame Photometric Detector
Gas Chromatography
Gas Chromatography-Mass Spectrometry
Graphite Furnace Atomic Absorption
High Pressure Liquid Chromatography
Hydride Atomic Absorption
Inductivery Coupled Plasma
Liquid Chromatography
Mass Spectrometry
Nitrogen/Phosphorus Detector
Photoionization  Detector
Ultraviolet
Ouantitatlon/Detectlon Limit*
CRDL *                Contract Retiree Detection Limit
CRQL •                Contract Required Quantitation Limit
EDL *                 Estimated Detection Limit
MDL»                 Method Detection Limit
NA»                  Not Available
PQL«                 Practical Quantitation Umit

Methods/Sample Preparation
 CLP SOW
 Dl
 EPA

 EPA AIR

 EPADW
 EP Extracts
 MCAWW
 QTM
 SDDC
 SMEWW
 SW846
 TO
 XTN
 3510
 3540
 3550
 5030
Contract Laboratory Program Statement of Work
Direct injection of Squid samples; solid samples mixed, then injected
Guidelines Establishing Test Procedures for the Analysis of Pollutants
under the dean Water Act
Compendium of Methods for the Determination of Toxic Organic
Compounds in Ambient Air
Methods for the Determination of Organic Compounds in Drinking Water
Extraction procedure toxteity test extracts
Methods for Chemical Analysis of Water and Wastes
Quick Turnaround Method
Silver dtothyldithiocarbamate
Standard Methods for the Examination of Water and Wastewater
Test Methods for Evaluating Solid Waste
Toxic organic
Extraction methods that could be used Include 3510,3520,3540 and 3550
Separatory Funnel Extraction of Liquid Samples
Soxhlet Extraction of Solid Samples
Sonication Extraction of Solid Samples
Purge and Trap
                                             168

-------
                                                               APPENDIX ID
                                                                 TABLE I
               METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                              AIR MATRICES
    ANALYTE/
    COMMON NAME
    CAS NUMBER
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-   QUANTITATION/
ATION          DETECTION LIMIT
ORGANOCHLORINE
Chlordane
57749
p,p'-DDE
72559
p.p'-DDT
50293
PEST1C1DES/AROCLORS
EPA AIR METHOD TO-4 'Method for the Determination of Organochlorine
Pesticides and Polychlorinated Biphenyls in Ambient Air"
EPA AIR METHOD TO-4 'Method for the Determination of Organochlorine
Pesticides and Polychlorinated Biphenyls in Ambient Air"
EPA AIR METHOD TO-4 'Method for the Determination of Organochlorine
Pesticides and Polychlorinated Biphenyls in Ambient Air"
GC-ECD
GC-ECD
GC-ECD
EDL= >1.0ng/m3
EDL= > l.Ong/m3
EDL= >I.Ong/m3
    VOLATILE COMPOUNDS

-------
                                                             APPENDIX HI
                                                                TABLE 1
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                            AIR MATRICES
 ANALYTE/
 COMMON NAME                                                                                INSTRUMENT-    QUANTITATJON/
 CAS NUMBER             METHOD REFERENCE/TITLE OF METHOD                                 ATION           DETECTION LIMIT
 1.4-dich)orobenzene
 106467
 Benzene
 71432
Chloroethene
(Vinyl Chloride)
75014

Dichlororoethaoe
(Methylene Chloride)
75092
 EPA AIR METHOD TO-1  "Method for the Determination of Volatile Organic           GC-MS
 Compounds in Ambient Air Usifig Tenax Adsorption and Gas Chromatography-
 Mass Spectrometry (GC-MS)' '

 EPA AIR METHOD TO-14  "The Determination of Volatile Organic Compounds         GC-MS
 (VOCs) in Ambient Air Using SUM MA Passivated Canister Sampling and Gas
 Chromatographic Analysis"

 EPA AIR METHOD TO-3  'Method for the Determination of Volatile Organic           GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection*

 EPA AIR METHOD TO-1 "Method for the Determination of Volatile Organic            GC-MS
 Compounds in Ambient Air Using Tenax Adsorption and Gas Chromatography-
 Mass Spectrometry (GC-MS)*

 EPA AIR METHOD TO-14 "The Determination of Volatile Organic Compounds        GC-MS
 (VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
 Chromatographic Analysis*

 EPA AIR METHOD TO-2  'Method for the Determination of Volatile Organic          GC-MS
 Compounds in Ambient Air by Carbon Molecular Sieve Adsorption and Gas
 Chromalognphy-Mass Spectrometry (GC-MS)"

 EPA AIR METHOD TO-3 "Method for the Determination of Volatile Organic           GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection"

 EPA AIR METHOD TO-14 "The Determination of Volatile Organic Compounds         GC-MS
(VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
Chromatographic Analysis"

EPA AIR METHOD TO-14 "The Determination of Volatile Organic Compounds         GC-MS
(VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
Chromatographic Analysis"
 NA
                                                                                                              NA
                                                                                                             NA
 NA
                                                                                                             EDL = 6.0 mg/m3
                                                                                                             NA
                                                                                                             NA
NA
NA

-------
                                                              APPENDIX HI
                                                                TABLE I
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                             AIR MATRICES
 ANALYTE/
 COMMON NAME                                                                                 INSTRUMENT-   QUANTITATION/
 CAS NUMBER            METHOD REFERENCE/TITLE OF METHOD                                 ATION          DETECTION LIMIT
 Dichloromethane
 (Methyleae Chloride)
 75092
 Etbenyl Benzene
 (Styrene)
 100425
Tetrachloroethene
(TetncbJoroethylene)
127184
Tetnchlorometbaoe
(Carbon Tetnchloride)
56235
 EPA AIR METHOD TO-2  'Method for the Determination of Volatile Organic          GC-MS
 Compounds in Ambient Air by Carbon Molecular Sieve Adsorption and Gas
 Chromatography-Mass Spectrometry (GC-MS)"

 EPA AIR METHOD TO-3  'Method for the Determination of Volatile Organic           GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection"

 EPA AIR METHOD TO-14 "The Determination of Volatile Organic Compounds         GC-MS
 (VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
 Chromatographic Analysis"

 EPA AIR METHOD TO-3  "Method for the Determination of Volatile Organic           GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection"

 EPA AIR METHOD TO-1  "Method for the Determination of Volatile Organic           GC-MS
 Compounds in Ambient Air Using Tenax Adsorption and Gas Chromatography-
 Mass Spectrometry (GC-MS)"

 EPA AIR METHOD TO-14  "The Determination of Volatile Organic  Compounds         GC-MS
 (VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
 Chromatographic Analysis"

 EPA AIR METHOD TO-3 "Method for the Determination of Volatile Organic           GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection*

 EPA AIR METHOD TO-14 "The Determination of Volatile Organic Compounds         GC-MS
(VOCs) in Ambient Air Using SUMMA Passivated Canister Sampling and Gas
Chromatographic Analysis"
NA
                                                                                                              NA
EDL - 10 mg/mj
                                                                                                              NA
NA
                                                                                                              EDL = 50 mg/m3
                                                                                                              NA
EDL - 2000 mg/m3

-------
                                                            APPENDIX III
                                                              TABLE I
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                           AIR MATRICES
ANALYTE/
COMMON NAME                                                                               INSTRUMENT-   QUANnTATlON/
CAS NUMBER           METHOD REFERENCE/TITLE OF METHOD                                ATION          DETECTION LIMIT
Tetnchlotomethane
(Carbon Tetnchloride)
56235
Trichloromethane
(Chloroform)
67663
 EPA AIR METHOD TO-2 'Method for the Determination of Volatile Organic          GC-MS
 Compounds in Ambient Air by Carbon Molecular Sieve Adsorption and Gas
 Chromatography-Mass Spectrometry (GC-MS)*

 EPA AIR METHOD TO-3 'Method for the Determination of Volatile Organic          GC-FID/
 Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
 Chromatography with Flame lonization and Electron Capture Detection"

 EPA AIR METHOD TO-14 'The Determination of Volatile Organic Compounds         GC-MS
 (VOCs) in Ambient Air Using SUMMA  Passivated Canister Sampling and Gas
 Chromatognphk Analysis"

 EPA AIR METHOD TO-2 "Method for the Determination of Volatile Organic           GC-MS
 Compounds in Ambient Air by Carbon Molecular Sieve Adsorption and Gas
 Chromatography-Mass Spectrometry (GC-MS)*

 EPA AIR METHOD TO-3 "Method for the Determination of Volatile Organic           GC-FID/
Compounds in Ambient Air Using Cryogenic Preconcentration Techniques and Gas        GC-ECD
Chromatography with Flame lonization and Electron Capture Detection"                . .
NA
                                                                                                           NA
EDL - 2000 mg/m3
                                                                                                           NA
                                                                                                           NA

-------
                                                          APPENDIX III
                                                            TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                    SOIL/SEDIMENT MATRICES
     METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-     QUANTITATION/
ATION           DETECTION LIMIT
 INORGANICS

 Arsenk
 7440382
Beryllium
7440417
7440439
 CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
 Multi-Media. Multi-Concentration'

 MCAWW METHOD 206.2/SW846 Method 7060 'Arsenic (Atomic Absorption,
 Furnace Technique)"

 SW846 METHOD 6010 'Inductively Coupled Plasma Atomic Emission
 Spectroscopy*

 SW846 METHOD 7061 •Arsenic (Atomic Absorption, Gaseous Hydride)'

 CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
 Multi-Media. Multi-Concentration*

 MCAWW METHOD 210.1/SW846 Method 7090 'Beryllium (Atomic
 Absorption, Direct Aspiration)"

 MCAWW METHOD 210.2/SW846 Method 7091  'Beryllium (Atomic
 Absorption, Furnace Technique)"

 SW846 METHOD 6010 "Inductively Coupled Plasma Atomic Emission
 Spectroscopy"

 CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
 Multi-Media, Multi-Concentration*

 MCAWW METHOD 213.1/SW846 Method 7130  •Cadmium (Atomic
Absorption,  Direct Aspiration)*

MCAWW METHOD 213.2/SW846 Method 7131  •Cadmium (Atomic
Absorption,  Furnace Technique)"
  GFAA-ICP


  GFAA


  ICP


  HYDAA

  GFAA-FLAME-
  1CP

  FLAME
  i

  GFAA


  ICP
 GFAA-ICP-
 FLAME

 FLAME
                                                                                         GFAA
CRDL = 2.0 rag/kg


MDL = 0.1 rag/kg


EDL = 5.3 mg/kg


MDL = 0.1 mg/kg

CRDL = 1.0 mg/kg


MDL » o.S mg/kg


MDL = 0.02 mg/kg


EDL * 0.03 mg/kg


CRDL » 1.0 mg/kg


MDL = 0.5 mg/kg


MDL - 0.01 mg/kg

-------
                                                             APPENDIX III
                                                               TABLE II
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                      SOIL/SEDIMENT MATRICES
 ANALYTE/
COMMON NAME
CAS NUMBER
CVfrnmni
7440439
Chromium, Total
7440473
METHOD REFERENCE/ TITLE OF METHOD
SW846 METHOD 6010 'Inductively Coupled Plasma Atomic Emission
Spectroscopy"
CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
Multi-Media, Multi-Concentration*
INSTRUMENT-
ATION
ICP
GFAA-ICP-
FLAME
QUANTITATION/
DETECTION LIMIT
EDL = 0.4 rag/kg
CRDL = 2.0 mg/kg
                     MCAWW METHOD 218.1/SW846 Method 7190 •Chromium (Atomic
                     Absorption, Direct Aspiration)"

                     MCAWW METHOD 218.2/SW846 Method 7191 •Chromium (Atomic
                     Absorption, Furnace Technique)"

                     SW846 METHOD 6010  'Inductively Coupled Plasma Atomic Emission
                     Spectroscopy"

 Chromium, Hexavalent   SW846 METHOD 7195  'Chromium Hexavalent (Coprecipitation) for EP
                                                                       FLAME
                                                                       GFAA
                                                                       ICP
                 MDL = 5.0 mg/kg


                 MDL = 0.1 mg/kg


                 EDL - 0.7 mg/kg
 7440473
Cyanide. Total
57-12-5
Extracts'

SW846 METHOD 7196 'Chromium Hexavalent (Colorimetric) for EP Extracts'

SW846 METHOD 7197 'Chromium Hexavalent (Chelation/Extraction) for EP
Extracts'

SW846 METHOD 7198 'Chromium Hexavalent (Differential Pulse Polarography)
for EP Extracts'

CLP SOW for Inorganic Analysis-Multi-Media, High Concentration

SMEWW Method 4500 CN, C, D, E, F. Total Cyanide after Distillation
                                                                       FLAME-GFAA    MDL = 100 rag/kg
Colorimeter

FLAME


Polarograph


Colorimeter

Colorimeter-
Titrimetric-
lon-Selective
Electrode
MDL = 10 mg/kg

MDL = 20 mg/kg


MDL = 20 mg/kg


CRDL= 1.0 mg/kg

EDL «  2.0 mg/kg
EDL =  5.0 mg/kg

-------
                                                         APPENDIX III
                                                          TABLE II
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                  SOIL/SEDIMENT MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER METHOD REFERENCE/ TITLE OF METHOD
Cyanide,
Total &
Amenable to
Chlorination
Lead
7439921



Mercury
7439976


SW846 Method 9010, "Total and Amendable Cyanide (Colonroetnc, manual)'
CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
Multi-Media, Multi-Concentration'
MCAWW METHOD 239.1/SW846 Method 7420 "Lead (Atomic Absorption,
Direct Aspiration)"
MCAWW METHOD 239.2/SW846 Method 7421 'Lead (Atomic Absorption,
Furnace Technique)*
SW846 METHOD 6010 "Inductively Coupled Plasma Atomic Emission
Spectroscopy*
CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
Multi-Media, Multi-Concentration*
MCAWW METHOD 245.5 "Mercury in Sediment (Manual Cold Vapor
Technique)"
SW846 METHOD 7471 "Mercury in Solid or Semisolid Waste (Manual Cold-
INSTRUMENT-
ATION
Colorimeter
GFAA-FLAME-
ICP
FLAME
GFAA
ICP
CVAA
CVAA
CVAA
QUANTITATION/
DETECTION LIMIT
CRDL= l.Omg/kg
CRDL = 0.6 mg/kg
MDL = 10 mg/kg
MDL = 0. 1 mg/kg
EDL = 4.2 mg/kg
CRDL = 0.1 mg/kg
MDL = 0.2 mg/kg
MDL = 0.1 mg/kg
                   Vapor Technique)"

ORGANOCHLQRINE PESTICIDES/AROCLORS
Aroclorl260
(PCB-1260)
11096825
CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*


CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media,
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques"
                                                                                       GC-ECD
                                                                                       GC-ECD
CRQL = 33 ug/kg
CRQL = 33 ug/kg

-------
                                                            APPENDIX HI
                                                              TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                     SOIL/SEDIMENT MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
     METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
Chtofdane
57749
Diddrin
60571
Heptachlor
76448
Undine
58899
CLP SOW METHOD ORG "Statement of Work for Organic; Analysis • Multi-
Media, Multi-Concentration*

CLP SOW METHOD QTM (Alpha and Gamma) 'Chemical Analytical Services
for Multi-Media, Multi-Concentration Samples for Organic Analysis by Quick
Turnaround Gas Chromatography Techniques* (CRQL is for Gamma Chlordane)

SW846 METHOD 8080 •Organochlorine Pesticides and PCBs*

CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration*

CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques*

SW846 METHOD 8080 *Organochlorine Pesticides and PCBs"

CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*

CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques*

SW846 METHOD 8080 "Organochlorine Pesticides and PCBs*

CLP SOW METHOD ORG  'Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*

CLP SOW METHOD QTM •Chemical Analytical Services for Multi-Media.
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques*
  GC-ECD         CRQL = 1.7 ug/kg


  GC-ECD         CRQL = 3.3 ug/kg



  GC-ECD         PQL = 9.0 ug/kg

  GC-ECD         CRQL - 3.3 ug/kg


  GC-ECD         CRQL = 3.3 ug/kg



  GCECD         PQL =1.3 ug/kg

  GC(-ECD         CRQL = 1.7 ug/kg


  GC-ECD         CRQL = 3.3 ug/kg



 GC-ECD         PQL - 2.0 ug/kg

 GC-ECD         CRQL = 1.7 ug/kg


 GC-ECD          CRQL « 3.3 ug/kg

-------
                                                            APPENDIX in
                                                              TABLE D
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                     SOIL/SEDIMENT MATRICES
 ANALYTE/
COMMON NAME
CAS NUMBER METHOD REFERENCED TITLE OF METHOD
y
p.p'-DDE
72559


p.p'-DDT
50293
CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration'
CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques'
SW846 METHOD 8080 •Organochlorine Pesticides and PCBs'
CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*
INSTRUMENT-
ATION
GC-ECD
GC-ECD
GC-ECD
GC-ECD
QUANTTTATION/
DETECTION LIMIT
CRQL
CRQL
PQL =
CRQL
= 3.3 ug/kg
= 3.3 ug/kg
2.7 ug/kg
= 3.3 ug/kg
                     CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,
                     Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
                     Chromatography Techniques*

                     SW846 METHOD 8080 •Organochlorine Pesticides and PCBs'

SEMIVOLATILE COMPOUNDS
3.5,5-trimethyl-
2-cyclohexen-l-one
(Isopborooe)
78591
Benzo  pyreoe
50328
CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*
SW846 METHOD 8270 'Gas Chromatography-Mass Spectrometry for
Semivolatile Organics: Capillary Column Technique"

CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration"

CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media.
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
      tnonnhv Techniaues"
                                                                       GC-ECD
                                                                       GC-ECD
GC-MS
                                                                                          GC-MS
GC-MS
                                                                                          GC-FID
CRQL = 3.3 ug/kg



PQL = 8.0 ug/kg



CRQL = 330 ug/kg



PQL = 660 ug/kg


CRQL = 330 ug/kg


CRQL « 330 ug/kg

-------
                                                           APPENDIX in
                                                             TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                    SOIL/SEDIMENT MATRICES
 ANALYTE/
COMMON NAME
CAS NUMBER
Benxo  pyrene
50328

Bis-{2-Dichloroemyl)
ether
111444
Bis-(2-ethyIhexyI)
phthalate
117817
METHOD REFERENCE/ TITLE OF METHOD
SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for
Semivolatile Organics: Capillary Column Technique*
SW846 METHOD 8310 "Polynuclear Aromatic Hydrocarbons"
CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration'
CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*
SW846 METHOD 8060 "Phthalate Esters*
SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for
INSTRUMENT-
ATION
GC-MS
HPLC
GC-MS
GC-MS
GC-ECD
GC-MS
QUANTITATION/
DETECTION LIMIT
PQL = 660 ug/kg
PQL - 15 ug/kg
CRQL - 330 ug/kg
CRQL = 330 ug/kg
PQL = 1340 ug/kg
PQL = 660 us/kg
N-nitrosodi-
pbenylamine
86306
Semivolatile Organics: Capillary Column Technique"

CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration"

SW846 Method 8270 'Gas Chromatography-Mass Spectrometry for SemivoUtile
Organics: Capillary Column Technique"
VOLATILE COMPOUNDS
1,1-dichloroethane
75343
CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration"

SW846 METHOD 8010 "Halogenated Volatile Organics*

SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics*
GC-MS


GC-MS



GC-MS


GC-ELCD

GC-MS
CRQL = 330 ug/kg


PQL = 660 ug/kg




CRQL = 10 ug/kg


PQL m 0.7 ug/kg

PQL = 5.0 ug/kg

-------
                                                            APPENDIX III
                                                              TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                      SOIL/SEDIMENT MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
     METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITAT10N/
DETECTION LIMIT
 1.1-dichloroethane
 75343
i.l-dichloroethene
75354
1.1,2-trichloroethane
79005
1.1.2.2-
tetnchloroethane
79345
 CLP SOW METHOD QTM  •Chemical Analytical Services for Multi-Media,
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatognphy Techniques*

 CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
 Media. Multi-Concentration*

 CLP SOW METHOD QTM  'Chemical Analytical Services for Multi-Media.
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatognphy Techniques*

 SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile
 Organics*

 CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
 Media. Multi-Concentration'

 SW846 METHOD 8010 'Halogenated Volatile Organics'

 SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
 Organics*

 CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-
 Media. Multi-Concentration*

 CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media.
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatognphy Techniques*

 SW846 METHOD 8010 "Halogenated Volatile Organics'

SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics*
  GC-PID



  GC-MS


 -GC-PID



  GC-MS


  GC-MS


  GC-ELCD

  GC-MS


  GC-MS


  GC-ECD



  GC-ELCD

  GC-MS
CRQL = 40 ug/kg



CRQL = 10 ug/kg


CRQL = 40 ug/kg



PQL = 5.0 ug/kg


CRQL = 10 ug/kg


PQL = 0.2 ug/kg

PQL = 5.0 ug/kg


CRQL = 10 ug/kg


CRQL = 40 ug/kg



PQL =  0.3 ug/kg

PQL =  5.0 ug/kg

-------
                                                               APPENDIX III
                                                                 TABLE II
                METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                         SOIL/SEDIMENT MATRICES
    ANALYTE/
    COMMON NAME
    CAS NUMBER
     METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-
ATION
 QUANTITATION/
 DETECTION LIMIT
S
    1.2-dichIoroethane
    107062
5  1,2-dichloropropuie
    78875
    1,4-dichlorobenzene
    106467
    Benzene
    71432
CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration'  '

CLP SOW METHOD QTM -Chemical Analytical Services for Multi-Media.
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques"

SW846 METHOD 8010 "Halogenated Volatile Organic*'

SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile
Organics"

CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration"

SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics"

SW846 METHOD 8010 "Halogenated Volatile Organics"

SW846 METHOD 8010 "Halogenated Volatile Organics"

SW846 METHOD 8020 "Aromatic Volatile Organics"

SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for
Semivolatile Organics: Capillary Column Technique"

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis • Multi-
Media, Multi-Concentration"

CLP SOW METHOD QTM  •Chemical Analytical Services for Multi-Media,
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques*
  GC-MS


  GC-PID



  GC-ELCD

  GC-MS


  GC-MS


  GC-MS


  GC-ELCD

  GC-ELCD

  GC-PID

  GC-MS


 GC-MS


 GC-PID
 CRQL = 10 ug/kg


 CRQL = 40 ug/kg



 PQL = 0.3 ug/kg

 PQL =5.0 ug/kg


 CRQL = 10 ug/kg


 PQL = 5.0 ug/kg


 PQL = 0.4 ug/kg

 PQL = 2.4 ug/kg

 PQL = 3.0 ug/kg

 PQL = 660 ug/kg


CRQL = 10 ug/kg


CRQL = 40 ug/kg

-------
                                                          APPENDIX III
                                                            TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                    SOIL/SEDIMENT MATRICES
 ANALYTE7
 COMMON NAME
 CAS NUMBER
     METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-     QUANTITATION/
AT10N           DETECTION LIMIT
Benzene
71432
Chloroethene
(Vinyl Chloride)
75014


SW846 METHOD 8020 "Aromatic Volatile Organics"
SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics"
CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*
CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media.
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques*
SW846 METHOD 8010 'Halogenated Volatile Organics'
SW846 METHOD 8240 'Gas Chromatograpby-Mass Spectrometry for Volatile
GC-PID
GC-MS
GC-MS
GC-PID
GC-ELCD
GC-MS
PQL » 2.0 ug/kg
PQL = 5.0 ug/kg
CRQL = 10 ug/kg
CRQL = 40 ug/kg
PQL =* 1.8 ug/kg
PQL = 10 ug/kg
Dichloromethane
(Methyleoe Chloride)
75092
Etheoyl Benzene
(Stymie)
100425
Tetrachloroetbene
(TetncbJoroetbylene)
127184
Organics*

CLP SOW METHOD ORG  'Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration*

SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile
Organics*

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*

SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics*

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration*
 GC-MS
                                                                                        GC-MS
 GC-MS
                                                                                        GC-MS
 GC-MS
CRQL = 10 ug/kg


PQL =« 5.0 ug/kg


CRQL = 10 ug/kg


PQL « 5.0 ug/kg


CRQL * 10 ug/kg

-------
                                                            APPENDIX III
                                                             TABLE II
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                     SOIL/SEDIMENT MATRICES
 ANALYTE/
 COMMON NAME
 CAS NUMBER
METHOD REFERENCE/ TITLE OF METHOD
INSTRUMENT-
ATION
                                                                                       QUANTITAT10N/
                                                                                       DETECTION LIMIT
 TetracfaloroedKne
 (Tetrachloroethylene)
 127184
Tetnchloromethane
(Carbon Tetrachloride)
56235
TricbJoromedune
(Chloroform)
67663
 CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media.
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatography Techniques"

 SW846 METHOD 8010 "Halogenated Volatile Organics"

 SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
 Organics"

 CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
 Media. Multi-Concentration"

 CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media.
 Multi-Concentration Samples for Organic Arjysis by Quick Turnaround Gas
 Chromatography Techniques"

 SW846 METHOD 8010 'Halogenated Volatile Organics'

 SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile
 Organics" .

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration*
                    CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,
                    Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
                    Chromatography Techniques"

                    SW846 METHOD 8010 "Halogenated Volatile Organics'

                    SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
                    Organics" •
                                                                 GC-PID



                                                                 GC-ELCD

                                                                 GC-MS


                                                                 GC-MS


                                                                 GC-PID



                                                                 GC-ELCD

                                                                 GC-MS


                                                                 GC-MS


                                                                 GC-PID



                                                                GC-ELCD

                                                                GC-MS
                  CRQL = 40 ug/kg



                  PQL = 0.3 ug/kg

                  PQL = 5.0 ug/kg


                  CRQL = 10 ug/kg


                  CRQL = 40 ug/kg



                 PQL = 1.2 ug/kg

                 PQL = 5.0 ug/kg


                 CRQL = 10 ug/kg


                 CRQL = 40 ug/kg



                 PQL = 0.5 ug/kg

                 PQL = 5.0 ug/kg

-------
                                                            APPENDIX III
                                                              TABLE III
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
  QUANTITATION/
  DETECTION LIMIT
 INORGANICS

 Arsenk
 7440382
Beryllium
7440417
 CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
 Multi-Media, Multi-Concentration*

 MCAWW METHOD 200.7/SW846 Method 6010/SMEWW Method 3120B
 "Inductively Coupled Plasma-Atomic Emission Spectrometric Method for Trace
 Element Analysis of Water and Wastes*

 MCAWW METHOD 206.2/SW846 Method 7060/SMEWW Method 3113B
 "Arsenk (Atomic Absorption. Furnace Technique)*

 MCAWW METHOD 206.3/SW846 Method 7061/SMEWW Method 3114B
 "Arsenic (Atomic Absorption-Gaseous Hydride)* Use method 206.5 for sample
 preparation

 MCAWW METHOD 206.4 "Arsenic (Spectrophotometric-SDDC)* Use method
 206.5 for sample preparation

 SMEWW METHOD 3500AS C 'Silver Diethyldithiocarbamate Method*

 CLP SOW METHOD INORG "Statement of Work for Inorganics Analysis -
 Multi-Media, Multi-Concentration*

 MCAWW  METHOD 200.7/SW846 Method 6010/SMEWW Method 3120B
 "Inductively Coupled Plasma-Atomic Emission Spectrometric Method for Trace
 Element Analysis of Water and Wastes*

MCAWW  METHOD 210.1 'Beryllium (Atomic Absorption,  Direct Aspiration)'

MCAWW METHOD 210.2/SW846 Method 7091/SMEWW Method 3113B
'Beryllium (Atomic Absorption, Furnace Technique)*
 GFAA-ICP
                                                                                           ICP
                                                                                           GFAA
                                                                                           HYDAA
                                                                                           Colcnmeter
Colorimeter

GFAA-FLAME-
ICP

ICP
                                                                                           FLAME

                                                                                           GFAA
 CRDL = 10 ug/L
                 MDL = 53 ug/L, 53 ug/L
                 EDL=50 ug/L
                MDL = 1.0 ug/L. 1.0 ug/L
                EDL =1.0 ug/L

                MDL = 2.0 ug/L. 2.0 ug/L
                EDL=  1.0 ug/L
MDL = 10 ug/L


EDL = 28.6 ug/L

CRDL = 5.0 ug/L


EDL = 0.3 ug/L



MDL = 5.0 ug/L

MDL = 0.2 ug/L, 0.2 ug/L
EDL=0.2 ug/L

-------
                                                            APPENDIX III
                                                              TABLE III
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALVTES OF CONCERN TO RISK ASSESSMENT

                                                        AQUEOUS MATRICES
 ANALYTE/
COMMON NAME
CAS NUMBER METHOD REFERENCE/TITLE OF METHOD
Beryllium
7440417


Cadmium
7440439
SMEWW METHOD 311 ID/SW846 Method 7090 -Direct Nitrous Oxide-
Acetylene Flame Method*
SMEWW METHOD 311 1 E 'Extraction/Nitrous Oxide-Acetylene Flame Method*
SMEWW METHOD 3SOOBE D *Aluminon Method*
CLP SOW METHOD INORG •Statement of Work for Inorganics Analysis -
Multi-Media. Multi-Concentration*
INSTRUMENT-
ATION
FLAME
FLAME
Colorimeter
GFAA-FLAME-
ICP
QUANTITATION/
DETECTION LIMIT
EDL= 5.0 ug/L. 5.0
MDL=5.0 ug/L
EDL = 5.0 ug/L
EDL - 5.0 ug/L
CRDL = 5.0 ug/L
ug/L



Chromium, Total
7440473
 MCAWW METHOD 200.7/SW846 Method 6010/SMEWW Method 3I20B
 "Inductively Coupled Plasma-Atomic Emission Spectrometric Method for Trace
 Element Analysis of Water and Wastes*

 MCAWW METHOD 213.1/SW846 Method 7130/SMEWW Method 311 IB
 'Cadmium (Atomic Absorption, Direct Aspiration)*

 MCAWW METHOD 213.2/SW846 Method 7131/SMEWW Method 3113B
 'Cadmium (Atomic Absorption, Furnace Technique)*

 SMEWW METHOD 3111C  'Extraction/Air-Acetylene Flame Method*

 SMEWW METHOD 3500CD D 'Dithizone Method*

 CLP SOW METHOD INORG  'Statement of Work for Inorganics Analysis -
 Multi-Media, Multi-Concentration*

 MCAWW METHOD 200.7/SW846 Method 6010/SMEWW Method 3120B
 "Inductively Coupled Plasma-Atomic Emission Spectrometric Method for Trace
Element Analysis of Water and Wastes*

MCAWW METHOD 218.1/SW846 Method 7I90/SMEWW Method 311 IB
"Chromium (Atomic Absorption. Direct Aspiration)"
                                                                                           ICP             EDL = 4.0 ug/L
 FLAME         MDL = 5.0 ug/L, 5.0 ug/L
                IDL=2.0 ug/L

 GFAA           MDL = 0.1 ug/L, 0.1 ug/L
                EDL=0.1 ug/L

 FLAME         NA

Colorimeter       EDL =  20 ug/ml

GFAA-ICP-       CRDL = 10 ug/L
FLAME

ICP             EDL =  7.0 ug/L
                                                                                          FLAME          MDL = 50 ug/L, 50 ug/L
                                                                                                          EDL « 20 ug/L

-------
                                                            APPENDIX III
                                                              TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                        AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
Chromium. Total       MCAWW M ETHOD 218.2 /SW846 Method 7191 /SM EWW Method 3113B
7440473              'Chromium (Atomic Absorption. Furnace Technique)'

                     MCAWW METHOD 218.3 •Chromium (Atomic Absorption, Chelation-
                     Extraction)*

Chromium. Hexavalent  MCAWW METHOD 218.4/SW846 Method 7197 •Chromium. Hexavalent
                     (Atomic Absorption, delation-Extraction)*

                     MCAWW METHOD 218.5 •Chromium. Dissolved Hexavalent (Atomic
                     Absorption, Furnace Technique)*

                     SMEWW METHOD 3111C "Extraction/Air-Acetylene Flame Method'

                     SW846 METHOD 7195 'Chromium. Hexavalent (Coprecipitation)*

                     SW846 METHOD 7196/SMEWW Method  3500CR D •Chromium, Hexavalent
                     (Colorimetric)'

                     SW846 METHOD 7198 •Chromium. Hexavalent (Differential Pulse
                     Polarography)'
Cyanide. Total
57-12-5
CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
Multi-Media, Multi-Concentration*

SMEWW Method 4500-CN, C. D, E, F 'Total Cyanide after Distillation'
                    MCAWW Method 335.2  'Cyanide. Total, Titrimetric Spectrophotornetric)'
                                                                        GFAA
                                                                        FLAME
                                                                        FLAME
                                                                        GFAA
                 MDL = l.Oug/L, l.Oug/L
                 EDL = 2.0 ug/L

                 MDL= l.Oug/L
                 MDL= 10 ug/L, l.Oug/L
                 MDL= l.Oug/L
                                                                        FLAME          NA

                                                                        FLAME, GFAA    MDL - 5.0 ug/L

                                                                        Colorimeter       MDL - 500 ug/L, NA
                                                                       Polarograph
Colorimeter/
Titrimetric

Colorimeter/
Titrimetric/
Ion-Selective
Electrode

Colorimeter/
Titrimetric
MDL = 10 ug/L


CRDL = 10 ug/L
                                                                                                            EDL - 20 ug/L
                                                                                                            EDL = 50 ug/L
                                                                                        EDL - 20 ug/L

-------
                                                         APPENDIX III
                                                           TABLE III
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                      AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER METHOD REFERENCE/TITLE OF METHOD
Cyanide. Total and SW846 METHOD 9010A. "Total and Amenable Cyanide (Colorimetric, Manual)
Amenable to
Chlorinatioa
SWS46 METHOD 9012 "Total and Amenable Cyanide (Colorimetric, Automated
UV)"
Cyanide, Amenable to SMEWW METHOD 4500-CN.G "Cyanide Amenable to Chlorination after
Chlorination Distillation"
MCAWW METHOD 335. 1 "Cyanide. Amenable to Chlorination"
Cyanide. Weak and SMEWW METHOD 4500-CN, I, D, E, F "Weak and Dissociable Cyanide'
Dissociable
Lead CLP SOW METHOD INORG 'Statement of Work for Inorganics Analysis -
7439921 Multi-Media. Multi-Concentration"
INSTRUMENT-
ATION
Colorimeter/
Titrimetric
Colorimeter/
Titrimetric
Colorimeter/
Titrimetric/
Ion-Selective
Elecrode
Colorimeter/
Titrimetric
Colorimeter/
Titrimetric/
Ion-Selective
Elecrode
GF.VA-FLAME-
ICP
QUANTITATION/
DETECTION LIMIT
EDL » 20 ug/L
EDL = 20 ug/L
EDL - 20 ug/L
EDL = 50 ug/L
EDL = 20 ug/L
EDL = 20 ug/L
EDL = 50 ug/L
CRDL = 3.0 ug/L
                   MCAWW METHOD 200.7/SW846 Method 6010/SMEWW Method 3120B             ICP
                   "Inductively Coupled Plasma-Atomic Emission Spectrometric Method for Trace
                   Element Analysis of Water and Wastes'

                   MCAWW METHOD 239.1/SW846 Method 7420/SMEWW Method 311 IB  'Lead       FLAME
                   (Atomic Absorption, Direct Aspiration)"

                   MCAWW METHOD 239.2/SW846 Method 7421 /SMEWW Method 3113B  "Lead       GFAA
                   (Atomic Absorption. Furnace Technique)*

                   SMEWW METHOD 3111C "Extraction/Air-Acetylene Flaac Method"                FLAME

                   SMEWW METHOD 3500PB D "Dithizone Method'                             Colorimeter
EDL = 42 ug/L. 42 ug/L,
40 ug/L
MDL= 100 ug/L, 100 ug/L
EDL=50ug/L

MDL= 1.0 ug/L, 100 ug/L
EDL= 1.0 ug/L

NA

EDL - 100 ug/L

-------
                                                            APPENDIX III
                                                              TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                        AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
Mercury
7439976
METHOD REFERENCE/TITLE OF METHOD
V
CLP SOW METHOD INORG/MCAWW Method 245. 1 and 245.2
"Statement of Work for Inorganics Analysis - Multi-Media. Multi-Concentration,
Mercury Manual ; Mercury Automated Cold Vapor Technique"
INSTRUMENT-
ATION
CVAA
QUANTITATION/
DETECTION LIMIT
CRDL = 0.2 ug/L
MDL=0.2 ug/L.0.2 ug/L
                     SMEWW METHOD 3112B/SW846 Method 7470 "Cold-Vapor Atomic
                     Absorption Spectrometric Method*

                     SMEWW METHOD 3500HG C
                     "Dithizone Method"
ORGANOCHLOR1NE PESTlCtDES/AROCLORS
Aroclorl260
(PCB-1260)
11096825
CLP SOW METHOD LC-ORG  "Chemical Analytical Services for the Analysis of
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
Capture (GC-ECD) Techniques"

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration"

CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media.
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques"

EPA METHOD 608  "Organochlorine Pesticides and PCBs"

EPA METHOD 625  "Base/Neutrals and Acids"

EPA DW METHOD 505 "Analysis of Organohalide Pesticides and Aroclors in
Water by Microextraction and Chromatography*

EPA DW METHOD 508 "Determination of Chlorinated Pesticides in Water by
Gas Chromatography with an Electron Capture Detector"
                                                                        CVAA
                                                                        Colorimeter
GC-ECD
                                                                                            GC-ECD


                                                                                            GC-ECD



                                                                                            GC-ECD

                                                                                            GC-MS

                                                                                            GC-ECD


                                                                                            GC-ECD
                 EDL= 1.0 ug/L
                 MDL=0.2 ug/L

                 EDL = 2.0 ug/L
CRQL = 0.20 ug/L




CRQL = 1.0 ug/L


CRQL » 1.0 ug/L



NA

NA

MDL = 0.189 ug/L


NA

-------
                                                              APPENDIX III
                                                                TABLE III
              METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
   ANALYTE/
   COMMON NAME
   CAS NUMBER
                                                           AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
  QUANTJTAT10N/
  DETECTION LIMIT
   Aroclor 1260
   (PCB-I260)
   11096825
   Chlordane
   57749
00
oo
 SMEWW METHOD 64)OB "Liquid-Liquid Extraction Gas Chromalographic-
 Mass Spectmmetric Method*

 SMEWW METHOD 6630B "Liquid-Liquid Extraction Gas Chromatographic
 Method I"

 SMEWW METHOD 6630C "Liquid-Liquid Extraction Gas Chromatographic
 Method II"

 CLP SOW METHOD LC-ORG (CRQL is for alpha and gamma Chlordane)
 'Chemical Analytical Services for the Analysis of Low Concentration Water
 Samples for Organic Compounds by Gas Chromatography-Mass Spectrometry
 (GC-MS) and Gas Chromatogrephy-Electron Capture (GC-ECD) Techniques*

 CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-
 Media. Multi-Concentration"

 EPA METHOD 608/SW846 Method 8080 "Organochlorine Pesticides and PCBs'

 EPA METHOD 625 "Base/Neutrals and Acids"

 EPA DW METHOD 505  "Analysis of Organohalide Pesticides and Aroclors in
 Water by Microextfaction and Chromatography"

 EPA DW METHOD 508  "Determination of Chlorinated Pesticides in Water by
Gas Chromatography with an Electron Capture Detec'nr"

SMEWW METHOD 64IOB  "Liquid-Liquid Extraction Gas Chromatographic -
Mass Spectrometric Method"

SMEWW METHOD 6630B  "Liquid-Liquid Extraction Gas Chromatographic
Method I"
 GC-MS
                                                                                              GC-MS
                                                                                             GC-ECD
GC-ECD
                                                                                             GC-ED


                                                                                             GC-ECD

                                                                                             GC-MS

                                                                                             GC-ECD


                                                                                             GC-ECD


                                                                                             GC-MS


                                                                                             GC-MS
 NA


 NA


 NA


 CRQL = 0.01 ug/L




 CRQL = 0.05 ug/L


 MDL  = 0.014 ug/L

 NA

 MDL  = 0.14 ug/L


 NA


NA


MDL = 0.014 ug/L

-------
                                                             APPENDIX III
                                                              TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                         AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-     QUANTITATION/
ATION            DETECTION LIMIT
Dieldrin
60571
 SMEWW METHOD 6630C  'Liquid-Liquid Extraction Gas Chromatographic           GC-ECD
 Method II'

 CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of       GC-ECD
 Low Concentration Water Samples for Organic Compounds by Gas
 Chronutography-Mtss Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques"

 CLP SOW METHOD ORG "Statement of Work for Orgamcs Analysis - Multi-          GC-ECD
 Media, Multi-Concentration*

 CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatography Techniques"

 EPA METHOD 608/SW846 Method 8080 "Organochlorine Pesticides and PCBs"        GC-ECD

 EPA METHOD 625 "Base/Neutrals and Acids"                                  GC-MS

 EPA DW METHOD 505 "Analysis of Organohalide Pesticides and Aroclors in          GC-ECD
 Water by Microex tract ion and Chromatography"

 EPA DW METHOD 508 "Determination of Chlorinated Pesticides in Water by          GC-ECD
 Gas Chromatography with an Electron Capture Detector"

 SMEWW METHOD 64IOB "Liquid-Liquid Extraction Gas Chromatographic-           GC-MS
 Mass Spectrometric Method"

SMEWW METHOD 6630B "Liquid-Liquid Extraction Gas Chromatographic            GC-MS
Method I"

SMEWW METHOD 6630C "Liquid-Liquid Extraction Gas Chromatographic            GC-ECD
Method II"
                 MDL = 0.014 ug/L


                 CRQL = 0.02 ug/L




                 CRQL = 0.1 ug/L


                 CRQL = 0.1 ug/L



                 MDL = 0.002 ug/L

                 MDL = 2.5 ug/L

                 MDL = 0.012 ug/L


                 EDL = 0.02 ug/L


                 MDL = 2.5 ug/L


                 MDL = 0.002 ug/L


                 MDL = 0.002 ug/L

-------
                                                            APPENDIX III
                                                              TABLE III
            METHODS AND DETECTION/QUANTITATICN LIMITS FOR SPECIFIED ANALVTES OF CONCERN TO RISK ASSESSMENT
ANALVTE/
COMMON NAME
CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
HepUchlor
76448
 CLP SOW METHOD LC-ORG! •Chemical Analytical Services for the Analysis of        GC-ECD
 Low Concentration Water Samples for Organic Compounds by Gas
 Chnimatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques"

 CLP SOW METHOD ORG •Statement of Work for Organics Analysis - Multi-          GC-ECD
 Media. Multi-Concentration*

 CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media.            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatognphy Techniques"

 EPA METHOD 608/SW846 Method 8080 'Organochlorine Pesticides and PCBs"         GC-ECD

 EPA METHOD 625 "Base/Neutrals and Acids*                                   GC-MS

 EPA DW METHOD SOS  "Analysis of Organohalide Pesticides and Aroclors in           GC-ECD
 Water by Microextraction and Chromatography"

 EPA DW METHOD SOS  "Determination of Chlorinated Pesticides in Water by           GC-ECD
 Gas Chromatography with an Electron Capture Detector*

 EPA DW METHOD 525 "Determination of Organic Compounds in Drinking            GC-MS
 Water by Liquid-Solid Extraction and Capillary Column Gas Chromatography-
 Mass Spectrometry*

 SMEWW METHOD 6410B "Liquid-Liquid Extraction Gas Chromatographic-            GC-MS
 Mass Spectrometric Method"

SMEWW METHOD 6630B "Liquid-Liquid Extraction Gas Chromatographic           GC-MS
Method I"

SMEWW METHOD 6630C 'Liquid-Liquid Extraction Gas Chromatographic           GC-ECD
Method II*
                 CRQL = 0.01 ug/L




                 CRQL = 0.05 ug/L


                 CRQL = 0.1 ue/L



                 MDL = 0.003 ug/L

                 MDL = 1.9 ug/L

                 MDL = 0.003 ug/L


                 EDL = 0.01 ug/L


                 MDL = 0.04 ug/L



                MDL = 1.9 ug/L


                MDL = 0.003 ug/L


                MDL = 0.003 ug/L

-------
                                                            APPENDIX III
                                                              TABLE HI
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                        AQUEOUS MATRICES
METHOD REFERENCE/TITLE OF METHOD
                                                                       INSTRUMENT-     QUANTITATION/
                                                                       AT10N            DETECTION LIMIT
 Lindane
 58899
p,p'-DDE
72559
 CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-ECD
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
 Capture (GC-ECD) Techniques"

 CLP SOW METHOD "Statement of Work for Organics Analysis - Multi-Media,          GC-ED
 Multi-Concentration"

 CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatography Techniques"

 EPA METHOD 608/SW846 Method 8080 "Organochlorine Pesticides and PCBs"         GC-ECD


 EPA METHOD 625 "Base/Neutrals and Acids"                                   GC-MS

 EPA DW METHOD 505  "Analysis of Organohalide Pesticides and Aroclors in           GC-ECD
 Water by Microextraction and Chromatography"

 EPA DW METHOD 508  "Determination of Chlorinated Pesticides in Water by           GC-ECD
 Gas Chromatography with an Electron Capture Detector"

 EPA DW METHOD 525  "Determination of Organic Compounds in Drinking            GC-MS
 Water by Liquid-Solid Extraction and Capillary Column Gas Chromatography-
 Mass Spectrometry"

 CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of         GC-ECD
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
Capture (GC-ECD) Techniques"

CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-          GC-ECD
Media, Multi-Concentration"
                                                                                   CRQL = 0.01 ug/L




                                                                                   CRQL = 0.5 ug/L


                                                                                   CRQL = 0.1 ug/L



                                                                                   MDL = 0.009 ug/L,
                                                                                   0.004 ug/L

                                                                                   MDL = 3.1 ug/L

                                                                                   MDL = 0.003 ug/L


                                                                                   EDL = 0.015 ug/L


                                                                                   MDL = 0.1 ug/L



                                                                                   CRQL = 0.02 ug/L




                                                                                   CRQL = 0.1 ug/L

-------
                                                            APPENDIX III
                                                             TABLE HI
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-    QUANTTTATION/
ATION           DETECTION LIMIT
 p.p'-DDE
 72559
p,p'-DDT
50293
 CLP SOW METHOD QTM "Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromalognphy Techniques"

 EPA METHOD 608/SW846 Method 8080 "Organochlorine Pesticides and PCBs"        GC-ED

 EPA METHOD 625 "Base/Neutrals and Acids"                                  GC-MS

 EPA DW METHOD 508 'Determination of Chlorinated Pesticides in Water by          GC-ECD
 Gas Chromatography with an Electron Capture Detector"

 SMEWW METHOD 6410B "Liquid-Liquid Extraction Gas Chromatographic-           GC-MS
 Mass Spectrometric Method"

 SMEWW METHOD 6630B "Liquid-Liquid Extraction Gas Chromatographic            GC-MS
 Method I"

 SMEWW METHOD 6630C "Liquid-Liquid Extraction Gas Chromatographic            GC-ECD
 Method II"

 CLP SOW METHOD LC-ORG  "Chemical Analytical Services for the Analysis of        GC-ECD
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques"

 CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-         GC-ECD
 Media, Multi-Concentration"

CLP SOW METHOD QTM "Cnemical Analytical Services for Multi-Media.            GC-ECD
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques"
                 CRQL = O.lug/L



                 MDL =  0.004 ug/L

                 MDL =  5.6 ug/L

                 EDL = 0.01 ug/L


                 MDL = 5.6 ug/L


                MDL = 0.004 ug/L


                MDL = 0.004 ug/L


                CRQL =  0.02 ug/L




                CRQL =  0.10 ug/L


                CRQL =  0.1 ug/L
                    EPA METHOD 608/SW846 Method 8080 "Organochlorine Pesticides and PCBs"
                                                                      GC-ECD
               MDL = 0.012 ug/L

-------
                                                           APPENDIX III
                                                            TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                       AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
 QUANTITATION/
 DETECTION LIMIT
p,p--DDT
50293
 EPA METHOD 625 "Base/Neutrals and Acids"

 EPA DW METHOD 508  'Determination of Chlorinated Pesticides in Water by
 Gas Chromatography with an Electron Capture Detector"

 SMEWW METHOD 641 OB  "Liquid-Liquid Extraction Gas Chromatographic-
 Mass Spectrometric Method"

 SMEWW METHOD 6630B  "Liquid-Liquid Extraction Gas Chromatographic
 Method I"

 SMEWW METHOD 6630C  "Liquid-Liquid Extraction Gas Chromatographic
 Method II"
SEM1VOLATILE COMPOUNDS
3,5,5-trimethyl-2-
cyclohexene-
1-one (Isophorone)
78591
CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
Capture (GC-ECD) Techniques*

CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media, Multi-Concentration*

EPA METHOD 609  "Nitroaromatics and Isphorone*

EPA METHOD 609  "Nitroaromatics and Isphorone*

EPA METHOD 625  "Base/Neutrals and Acids"

SMEWW METHOD 6410B 'Liquid-Liquid Extraction Gas Chromatographic-
Mass Spectrometric Method"

SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for
Semivolatile Organics: Capillary Column Technique"
 GC-MS

 GC-ECD


 GC-MS


 GC-MS


 GC-ECD
GC-MS
                                                                                         GC-MS


                                                                                         GC-FID

                                                                                         GC-ECD

                                                                                         GC-MS

                                                                                         GC-MS


                                                                                         GC-MS
 MDL = 4.7 ug/L

 EDL = 0.06 ug/L


 MDL = 4.7 ug/L


 MDL = 0.012 ug/L


 MDL « 0.012 ug/L




 CRQL = 5.0 ug/L




 CRQL = 10 ug/L


 MDL = 5.7 ug/L

 MDL = 15.7 ug/L

MDL = 2.2 ug/L

MDL = 2.2 ug/L


PQL = 10 ug/L

-------
                                                           APPENDIX HI
                                                            TABLE HI
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                       AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTTTATION/
DETECTION LIMIT
Benzo <•> pyrene
50328
 CLP SOW METHOD LC-ORG  •Chemical Analytical Services for the Analysis of       GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques'

 CLP SOW METHOD ORG  "Statement of Work for Organic* Analysis - Multi-         GC-MS
 Media, Multi-Concentration*

 EPA METHOD 610/SW846 Method 8100 "Polynuclear Aromatic Hydrocarbons*        GC-FID

 EPA METHOD 625 "Base/Neutrals and Acids'                                 GC-MS

 EPA DW METHOD 525  "Determination of Organic Compounds in Drinking           GC-MS
 Water by Liquid-Solid Extraction and Capillary Column Gas Chromatography-
 Mass Spectrometry"

 SMEWW METHOD 6410B  'Liquid-Liquid Extraction Gas Chromatographic-           GC-MS
 Mass Spectrometric Method*

 CLP SOW METHOD QTM  'Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatograpby Techniques*

 SMEWW METHOD 6440B  "Liquid-Liquid Extraction Chromatographic Method*        GC-MS

SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for                GC-MS
Semivolatile Organics: Capillary Column Technique"
                 CRQL - 5.0 ug/L




                 CRQL - 10 ug/L


                 MDL =  0.023 ug/L

                 MDL -  2.5 ug/L

                 MDL = 0.04 ug/L



                 MDL - 2.5 ug/L


                CRQL = 20 ug/L



                MDL = 0.023 ug/L

                PQL = 10 ug/L
                   SW846 METHOD 8310 "Polynuclear Aromatic Hydrocarbons*
                                                                      HPLC
                MDL - 0.023 ug/L

-------
                                                              APPENDIX III
                                                               TABLE III
              METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
  ANALYTE/
  COMMON NAME
  CAS NUMBER
                                                          AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
   Bis-(2-Chloroethyl)
   ether
   111444
VI
  Bis (2-ethylhexyl)
  phthalate
  117817
 CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatognphy-Electron
 Capture (GC-ECD) Techniques*

 CLP SOW METHOD ORG  'Statement of Work for Organic* Analysis - Multi-          GC-MS
 Media, Multi-Concentration"

 EPA METHOD 625 -Base/Neutrals and Acids*                                  GC-MS

 SMEWW METHOD 6040B  •Closed-Loop Stripping, Gas-Chromatographic-Mass-        GC-MS
 Spectrometric Analysis*

 SMEWW METHOD 6410B  'Liquid-Liquid Extraction Gas Chromatographic-           GC-MS
 Mass Spectrometric Method"

 SW846 METHOD 8250  'Gas Chromatography-Mass Spectrometry for                 GC-MS
 Semivolatile Organics: Packed Column Technique'

 CLP SOW METHOD LC-ORG 'Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and GA: Chromatograpby-Electron
 Capture (GC-ECD) Techniques*

 CLP SOW METHOD ORG 'Statement of Work for Organics Analysis - Multi-          GC-MS
 Media, Multi-Concentration*

 EPA METHOD 606  'Phthalate Ester'                                         GC-ECD

 EPA METHOD 625  'Base/Neutrals and Acids'                                   GC-MS

 EPA DW METHOD 525 "Determination of Organic Compounds in Drinking            GC-MS
Water by Liquid-Solid Extraction and Capillary Column Gas Chromatography-
Mass Spectrometry"
                 CRQL - 5.0 ug/L




                 CRQL - 10 ug/L


                 MDL =  5.7 ug/L

                 EDL = 0.001 ng/L


                 MDL =  5.7 ug/L


                 MDL =  5.7 ug/L


                 CRQL = 5.0 ug/L




                CRQL= 10 ug/L


                MDL = 2.0 ug/L

                MDL - 2.5 ug/L

                MDL - 0.8 ug/L

-------
                                                            APPENDIX III
                                                             TABLE III
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
 INSTRUMENT-
 ATION
  QUANTTTATION/
  DETECTION LIMIT
 Bis (2-ethylhexyl)
 phthalate
 117817
 N-nitrosodi-
 phenylunine
 86306
 SMEWW METHOD 6410B 'Liquid-Liquid Extraction Gas Chromatographic-
 Mass Spectrometric Method*

 SW846 METHOD 8060 'Phthalate Esters'

 SW846 METHOD 8270 'Gas Chromatography-Mass Spectrometry for
 Semivolatile Organics: Capillary Column Technique"

 SW846 METHOD 8250 "Gas Chromatography-Mass Spectrometry for Semi-
 Violatile Organics: Packed Column Technique"

 CLP SOW METHOD LC-ORG  •Chemical Analytical Services for the Analysis of
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
 Capture (GC-ECD) Techniques'

 CLP SOW METHOD ORG  'Statement of Work for Organics Analysis - Multi-
 Media, Multi-Concentration'

 EPA METHOD 607 'Nitrosamines'

 EPA METHOD 625 'Base/Neutrals and Acids'

 SMEWW METHOD 6410B  'Liquid-Liquid Extraction Gas Cnromatognphic-
 Mass Spectrometric (GC-MS) Method*

 SW846 METHOD 8270 'Gas Chromatognphy-Mass Spectrometry for
 Semivolatile Organics: Capillary Column Technique*
VOLATILE COMPOUNDS
1,1-dichloroethane
75343
CLP SOW METHOD LC-ORG 'Chemical Analytical Services for the Analysis of
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
Capture (GC-ECD) Techniques"
 GC-MS


 GC-ECD

 GC-MS


 GC-MS


 GC-MS
                                                                                          GC-MS


                                                                                          GC-ELCD

                                                                                          GC-MS

                                                                                          GC-MS


                                                                                          GC-MS
GC-MS
 MDL = 2.5 ug/L


 MDL - 2.0 ug/L

 PQL = 10 ug/L


 MDL = 2.5 ug/L


 CRQL = 5.0 ug/L




 CRQL = 10 ug/L


 MDL = 0.81 ug/L

 MDL- 1.9 ug/L

 MDL = 1.9 ug/L


 PQL = 10 ug/L




CRQL- 1.0 ug/L

-------
                                                           APPENDIX III
                                                            TABLE UI
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                       AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-    QUANTITATION/
ATION           DETECTION LIMIT
1,1-dichlorocthane
75343
 CLP SOW METHOD QTM  'Chemical Antlytiod Services for Multi-Media,           GC-PID
 Multi-Concentration Samples for Organic Analysts by Quick Turnaround Gas
 Chromatography Techniques"

 CLP SOW METHOD ORG •Statement of Work for Organics Analysis - Multi-          GC-MS
 Media. Multi-Concentration*

 EPA METHOD 601/SW846 Method 8010/SMEWW Method 6230B "Purgeable         GC-ELCD
 Halocarbons"

 EPA METHOD 624 "Purgeables"                                            GC-MS

 EPA DW METHOD 502.1 "Volatile Halogenated Organic Compounds in Water          GC-ELCD
 by Purge and Trap Gas Chromatography"

 EPA DW METHOD 502.2  "Volatile Organic Compounds in Water by Purge and        GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series*

 EPA DW METHOD 524.1/SMEWW Method 6210B (Method I)/SMEWW              GC-MS
 Method 6210C (Method II) "Measurement of Purgeable Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry"

 EPA DW METHOD 524.2/SMEWW Method 6210D "Measurement of Purgeable        GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry*

 SMEWW METHOD 6230C "Purge and Trap Packed-Column Gas                    GC-MS
 Chromatographic Method II"

SMEWW METHOD 6230D "Purge and Trap Capillary-Column Gas                  GC-ECD
Chromatographic Method"
                 CRQL = 20 ug/L



                 CRQL= 10 ug/L


                 MDL -  0.07 ug/L


                 MDL «  4.7 ug/L

                 MDL »  0.003 ug/L


                 MDL =  0.07 ug/L
                                                                                                         MDL = 0.2 ug/L
                                                                                                         MDL « 4.7 ug/L
                                                                                                         MDL « 0.04 ug/L



                                                                                                         MDL = 0.07 ug/L


                                                                                                         NA

-------
                                                           APPENDIX III
                                                             TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
1.1-dichloroethane
75343

1. l-dichlofoetbene
75354
 SW846 METHOD 8240 "Gas Chromatography-Mass Spectroroelry for Volatile          GC-MS
 Organics'

 CLP SOW METHOD LC-ORG •Chemical Analytical Services for the Analysis of       GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques'

 CLP SOW METHOD ORG "Statement of Work for Organics Analysis-Multi-          GC-MS
 Media, Multi-Concentration*

 CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatograpoy Techniques'

 EPA METHOD 624 'Purgeables'                                            GC-MS

 EPA METHOD 601/SMEWW Method 6230B 'Purgeable Hydrocarbons'              GC-ELCD

 EPA DW METHOD 502.1 'Volatile Halogenated Organic Compounds in Water          GC-ELCD
 by Purge and Trap Gas Chroraatography*

 EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-PID
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series'

 EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-ELCD
Trap Ct^illary Column Gas Chromatography with Photoionization and Electrolytic
Conductivity Detectors in Series'

EPA DW METHOD 524.1/SMEWW Method 6210B (Method 0/SMEWW              GC-MS
Method 6210C (Method II) "Measurement of Purgeable Organic Compounds in
Water by Packed Column Gas Chromatography-Mass Spectrometry'
                 PQL - 5.0 ug/L


                 CRQL- 1.0 ug/L




                 CRQL » 10 ug/L


                 CRQL»20ug/L



                MDL« 2.8 ug/L

                MDL = 0.13 ug/L

                MDL =• 0.003 ug/L


                NA



                MDL » 0.07 ug/L
                                                                                                         MDL - 0.2 ug/L
                                                                                                         MDL =« 2.8 ug/L. 2.8 ug/L

-------
                                                                APPENDIX HI
                                                                 TABLE III
                METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                            AQUEOUS MATRICES
     ANALYTE/
     COMMON NAME
     CAS NUMBER
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
     l.l-dichloroethene
     75354

-------
g
                                                                APPENDIX III
                                                                 TABLE HI
                METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                            AQUEOUS MATRICES
    ANALYTE/
    COMMON NAME
    CAS NUMBER
    l.l.2>trichloroeth*ne
    79005
     METHOD REFERENCE/TITLE OF METHOD
                          •4-
    1.1.2.2-
    tetrachloroethane
    79345
INSTRUMENT-
ATION
 EPA DW METHOD 502.2  'Volatile Organic Compounds in Water by Purge and        GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series*

 EPA DW METHOD 524.2  'Measurement of Purgeabte Organic Compounds in          GC-MS
 Water by Capillary Column Gas Chromatography-Mass Spectrometry"

 SMEWW METHOD 6040B  'Closed-Loop Stripping. Gas-Chromatographic-Mass        GC-MS
 Spectrometric Analysis'

 SMEWW METHOD 62JOB  'Purge and Trap Packed-Column Gas                    GC-MS
 Chromatographic-Mass Spectrometric Method I"

 SMEWW METHOD 6230C  'Purge and Trap Packed-Column Gas                    GC-MS
 Chromatographic Method H*

 SMEWW METHOD 6230D  'Purge and Trap Capillary-Column Gas                  GC-ECD
 Chromatogrephic Method"

 SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
 Organics'

 CLP SOW METHOD LC-ORG  'Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques'

 CLP SOW METHOD ORG,  "Statement of Work for Organics Analysis - Multi-          GC-MS
 Media, Multi-Concentration"

CLP SOW METHOD QTM  'Chemical Analytical Services for Multi-Media,            GC-PID
Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
Chromatography Techniques"
QUANTITATION/
DETECTION LIMIT
                 NA



                 MDL = 0.1 ug/L


                 EDL = 0.002 ug/L


                 MDL = 5.0 ug/L


                 MDL = 0.02 ug/L


                 NA


                 POL = 5.0 ug/L


                CRQL = 1.0 ug/L




                CRQL = 10 ug/L


                CRQL = 20 ug/L

-------
                                                            APPENDIX HI
                                                              TABLE III
            METHODS AND DETECTION/QUANT1TATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
 ANALYTE/
 COMMON NAME
 CAS NUMBER
                                                        AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-     QUANTITAT1ON/
ATION            DETECTION LIMIT
 1.1,2.2-
 tetrachloroeUuuie
 79345
1,2-dichloroethane
107062
 EPA METHOD 601/SW846 Method 8010/SMEWW Method 6230B 'Purgeable          CC-ELCD
 Halocarbons*

 EPA METHOD 624 'Purgeables"                                             GC-MS

 EPA DW METHOD 502.1 'Volatile Halogenated Organic Compounds in Water          GC-ELCD
 by Purge and Trap Gas Chronutography"

 EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series*

 EPA DW METHOD 524.1/SMEWW Method 6210B 'Measurement of Purgeable         GC-MS
 Organic Compounds in Water by Packed Column Gas Chromatography-Mass
 Spectrometry"

 EPA DW METHOD 524.2/SMEWW Method 6210D 'Measurement of Purgeable         GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry"

 SMEWW METHOD 6040B  'Closed-Loop Stripping, Gas-Chromatographic-Mass-        GC-MS
 Spectrometric Analysis"

 SMEWW METHOD 6230D  'Purge and Trap Capillary-Column Gas                  GC-P1D
 Chromatographic Method"

 SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
 Organics"

CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
                 MDL = 0.03 ug/L


                 MDL = 6.9 ug/L

                 MDL = 0.01 ug/L


                 MDL = 0.08 ug/L
                                                                                                            MDL = 0.4 ug/L
                                                                                                            MDL - 6.9 ug/L
                                                                                                            MDL = 0.04 ug/L
                                                                                                            MDL = 1.11 ug/L
                EDL = 50 ug/L


                MDL = 0.03 ug/L


                POL = 5.0 ug/L


                CRQL = 1.0 ug/L

-------
                                                           APPENDIX HI
                                                            TABLE III
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                       AQUEOl'S MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
1,2-dichloroethane
107062
 CLP SOW METHOD ORG  'Statement of Work for Organic* Analysis - Multi-         GC-MS
 Media, Multi-Concentration"

 CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media,           GC-EC
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatography Techniques"

 EPA METHOD 60I/SW846 Method 8010/SMEWW Method 6230B "Purgeabie         GC-ELCD
 Halocarbons"

 EPA METHOD 624 "Purgeables*                                            GC-MS

 EPA DW METHOD 502.1  "Volatile Halogenaled Organic Compounds in Water        GC-ELCD
 by Purge and Trap Gas Chromatography"

 EPA DW METHOD 502.2  "Volatile Organic Compounds in Water by Purge and        GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series"

 EPA DW METHOD 524.1/SMEWW Method 6210B (Method I)/SMEWW             GC-MS
 Method 6210 C (Method II) "Measurement of Purgeabie Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry"

 EPA DW METHOD 524.2  "Measurement of Purgeabie Organic Compounds in          GC-MS
 Water by Capillary Column Gas Chromatography-Mass Spectrometry*

 SMEWW METHOD 6230C "Purge and Trap Packed Column Gas                    GC-MS
Chromatographic Method II *

SMEWW METHOD 6230D "Purge and Trap Capillary Column Gas                  GC-ECD
Chromatographic Method"
                 CRQL =  10 ug/L


                 CRQL = 20 ug/L



                 MDL = 0.03 ug/L


                 MDL = 2.8 ug/L

                 MDL = 0.002 ug/L


                 MDL = 0.03 ug/L
                                                                                                         MDL = 0.2 ug/L, 2.8 ug/L.
                                                                                                         MDL = 2.8 ug/L
                                                                                                         MDL = 0.06 ug/L
                                                                                                         MDL = 0.03 ug/L
                                                                                                         NA

-------
                                                            APPENDIX III
                                                             TABLE HI
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                        AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
    METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
1.2-dichloroethane
107062

1,2-dichloropropane
78875
SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
Organics"

CLP SOW METHOD LC-ORG  'Chemical Analytical Services for the Analysis of       GC-MS
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
Capture (GC-ECD) Techniques*

CLP SOW METHOD ORG  'Statement of Work for Organics Analysis - Multi-          GC-MS
Media, Multi-Concentration"

EPA METHOD 601 /SW846 Method 8010/SMEWW Method 6230B                  GC-ELCD
'Purgeable Halocarbons*

EPA METHOD 624  'Purgeables'                                            GC-MS

EPA DW METHOD 502.1  'Volatile Halogenated Organic Compounds in Water         GC-ELCD
by Purge and Trap Gas Chromatography"
                                                                         (
EPA DW METHOD 502.2  'Volatile Organic Compounds in Water by Purge and        GC-PID
Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
Conductivity Detectors in Series"

EPA DW METHOD 502.2  'Volatile Organic Compounds in Water by Purge and        GC-ELCD
Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
Conductivity Detectors in Series'

EPA DW METHOD 524.1/SMEWW Method 6210B/SMEWW Method 6210C          GC-MS
"Measurement of Purgeable Organic Compounds in Water by Packed Column Gas
Chromatography-Mass Spectrometry"

EPA DW METHOD 524.2/SMEWW Method 62IOD "Measurement of Purgeable        GC-MS
Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
Spectrometry"
                 PQL = 5.0 ug/L


                 CRQL= 1.0 ug/L




                 CROL = 10 ug/L


                 MDL = 0.04 ug/L


                 MDL = 6.0 ug/L

                 NA


                 NA



                 MDL = 0.01 ug/L
                                                                                                           MDL = 0.2 ug/L
                                                                                                           MDL = 6.0 ug/L. 6.0 ug/L
                                                                                                           MDL = 0.04 ug/L

-------
                                                        APPENDIX III
                                                          TABLE III
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                     AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
1 ,2-dichIoropropane
78875


1 ,4-dichlorohenzene
106467







METHOD REFERENCE/TITLE OF METHOD
EPA SMEWW METHOD 6230C "Purge and Trap Packed-Column Gas
Chromatographic Method II"
SMEWW METHOD 6230D "Purge and Trap Capillary-Column Gas
Chromatographic Method*
SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organics"
CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
Capture (GC-ECD) Techniques"
CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-
Media. Multi-Concentration"
EPA METHOD 60I/SW846 Method 8010/SMEWW Method 6230B " Purgeable
Halocarbons*
EPA METHOD 602/SW846 Method 8020/SMEWW Method 6220B " Purgeable
Aromatics"
EPA METHOD 612 "Chlorinated Hydrocarbons"
EPA METHOD 624 "Purgeables"
EPA METHOD 625 "Base/Neutrals and Acids"
EPA DW METHOD 502. 1 "Volatile Halogenated Organic Compounds in Water
INSTRUMENT-
ATION
GC-MS
GC-ECD
GC-MS
GC-MS
GC-MS
GC-ELCD
GC-P1D
GC-ED
GC-MS
GC-MS
GC-ELCD
QUANTITATION/
DETECTION LIMIT
MDL » 0.04 ug/L
NA
POL = 5.0 ug/L
CRQL = 1.0 ug/L
CRQL = 10 ug/L
MDL = 0.24 ug/L
MDL = 0.3 ug/L
MDL= 1.34 ug/L
NA
MDL » 4.4 ug/L
NA
                  by Purge and Trap Gas Chromatography"

                  EPA DW METHOD 502.2 "Volatile Organic Compounds in Water by Purge and
                  Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
                  Conductivity Detectors in Series"
GC-PID
MDL - 0.01 ug/L

-------
                                                            APPENDIX III
                                                             TABLE HI
             METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALVTES OF CONCERN TO RISK ASSESSMENT
                                                        AQUEOUS MATRICES
 ANALYTE/
 COMMON NAME
 CAS NUMBER
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
 1,4-dichlorobenzene
 106467
Benzene
71432
 EPA DW METHOD 502.2  "Volatile Organic Compounds in Water by Purge and        GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity
 Detectors in Series"

 EPA DW METHOD 503.1  "Volatile Aromatic and Unsaturated Organic               GC-PID
 Compounds in Water by Purge and Trap Gas Chromatography*

 EPA DW METHOD 524.1/SMEWW Method 6210B (Method D/SMEWW              GC-MS
 Method 62IOC (Method II) "Measurement of Purgeable Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry"

 EPA DW METHOD 524.2/SMEWW Method 62 JOD •Measurement of Purgeable        GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry"

 SMEWW METHOD 6230C  "Purge and Trap Packed-Column Gas                    GC-MS
 Chromatographic Method II"

 SMEWW METHOD 6230D  "Purge and Trap Capillary-Column Gas                  GC-PID/
 Chromatographic Method"                                                   GC-ECD

 SMEWW METHOD 64IOB "Liquid-Liquid Extraction Gas Chromatographic-           GC-MS
 Mass Spectrometric Method"

 SW846 METHOD 8270 "Gas Chromatography-Mass Spectrometry for                GC-MS
 Semivolatile Organics: Capillary Column Technique"

CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-MS
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
Capture (GC-ECD) Techniques"
                 MDL = 0.01 ug/L




                 MDL » 0.006 ug/L


                 MDL = 2.0 ug/L



                 MDL « 0.03 ug/L



                 MDL = 0.24 ug/L


                NA


                MDL = 4.4 ug/L


                PQL  = 10 ug/L


                CRQL= 1.0 ug/L

-------
                                                          APPENDIX III
                                                           TABLE III
           METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                      AQUEOUS MATRICES

ANALYTE/
COMMON NAME                                                                         INSTRUMENT-    QUANTITATION/
CAS NUMBER            METHOD REFERENCE/TITLE OF METHOD                             ATION           DETECTION LIMIT

Benzene              CUP SOW METHOD ORG  "Statement of Work for Orguiics Analysis • Multi-          GC-MS           CRQL - 5.0 ug/L
71432               Media, Multi-Concentration"

                    CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media,            GC-ECD         CRQL - 20 ug/L
                    Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
                    Chromatography Techniques*

                    EPA METHOD 602/SW846  Method 8020/SMEWW Method 6220B "Purgeable          GC-PID           MDL = 0.2 ug/L
                    Aromatics*

                    EPA METHOD 624 "Purgeables"                                            GC-MS           MDL - 4.4 ug/L

                    EPA DW METHOD 502.2 "Volatile Organic Compounds in Water by Purge and         GC-PID           MDL - 0.01 ug/L
                    Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
                    Conductivity Detectors in Series"

                    EPA DW METHOD 503.1 "Volatile Aromatic and Unsaturated Organic               GC-PID          MDL - 0.02 ug/L
                    Compounds in Water by Purge and Trap Gas Chromatography"
                                                                                          <
                    EPA DW METHOD 524.1/SMEWW Method 6210B (Method D/SMEWW             GC-MS          MDL - 0.1 ug/L, 4.4 ug/L
                    Method 6210C (Method II) "Measurement of Purgeable Organii Compounds in                         MDL = 4.4 ug/L
                    Water by Packed Column Gas Chromatography-Mass Spectrometry"

                    EPA DW METHOD 524.2/SMEWW Method 6210D "Measurement of Purgeable        GC-MS          MDL = 0.04 ug/L
                    Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
                    Spectrometry"

                    SMEWW METHOD 6220C  "Purge and Trap Gas Chromatographic Method II"         GC-MS          MDL = 0.2 ug/L

                    SMEWW METHOD 6230D  "Purge and Trap Capillary-Column Gas                  GC-ECD         NA
                    Chromatographic Method"

-------
                                                APPENDIX III
                                                  TABLE III
METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                            AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
Bcnarne
71432
Chlonethene
(Vinyl Chloride)
75014
INSTRUMENT- QUANTTTATION/
METHOD REFERENCE/TITLE OF METHOD ATION DETECTION LIMIT
SW846 METHOD 8240 'Gas Chnputography-Miss Spectrometry for Volatile GC-MS
Organics' '
CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of GC-MS
Low Concentration Water Samples for Organic Compounds by Gas
Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Eleclroa
Capture (GC-ECD) Techniques*
POL - 5.0 ug/L
CRQL = 1 .0 ug/L
         CLP SOW METHOD ORG •Statement of Work for Organics Analysis - Multi-          GC-MS
         Media. Multi-Concentration*

         CLP SOW METHOD QTM 'Chemical Analytical Services for Multi-Media,            GC-ECD
         Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
         Chromatography Techniques*

         EPA METHOD 601/SW846 Method 8010/SMEWW Method 6230 "Purgeable           GC-ELCD
         Halocarbons"

         EPA METHOD 624 'Purgeables*                                             GC-MS

         EPA DW METHOD 502.1 "Volatile Halogenated Organic Compounds in Water         GC-ELCD
         by Purge and Trap Gas Chromatography*

         EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-PID
         Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
         Conductivity Detectors in Series"

         EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-ELCD
         Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
         Conductivity Detectors in Series"

         EPA DW METHOD 524.1/SMEWW Method 6210B (Method D/SMEWW              GC-MS
         Method 62 IOC (Method II) 'Measurement of Purgeable Organic Compounds in
         Water by Packed Column Gas Chromatography-Mass Spectrometry*
 CRQL = 10 ug/L


 CRQL - 20 ug/L



 MDL =  0.18ug/L


 NA

 MDL =  0.01 ug/L


 MDL =  0.02 ug/L



MDL =  0.04 ug/L



MDL -  0.3 ug/L

-------
                                                           APPENDIX HI
                                                            TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                       AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITAT1ON/
DETECTION LIMIT
Chloroethene
(Vinyl Chloride)
75014
Dichloromethane
(Methylene Chloride)
75092
 EPA DW METHOD S24.2/SMEWW Method 6210D 'Measurement of Purgeable       GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry"

 SMEWW METHOD 6230C "Purge and Trap Packed-Column Gas                   GC-MS
 Chromatcgraphic Method II"

 SMEWW METHOD 6230D "Purge and Trap Capillary Column Gas                 GC-PID/
 Chromatographic Method"                                                 GC-ECD

 SW846 METHOD 8010 "Halogenated Volatile Organics"                          GC-ELCD

 SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
 Organics"

 CLP SOW METHOD LC-ORG  "Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques"

 CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-          GC-MS
 Media, Multi-Concentration"

 EPA METHOD 601/SMEWW Method 6230B "PurgeaWe Halocarbons"                GC-ELCD

 EPA METHOD 624 "Purgeables"                                            GC-MS

 EPA DW METHOD 502.1 "Volatile Halogenated Organic Compounds in Water          GC-ELCD
by Purge and Trap Gas Chromatogrephy"

EPA DW METHOD 502.2 "Volatile Organic Compounds in Water by Purge and         GC-ELCD
Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
Conductivity Detectors in Series"
                 MDL = 0.17 ug/L



                 MDL = 0.18 ug/L


                 NA


                MDL = 0.18 ug/L

                PQL  -  10 ug/L


                CRQL = 2.0 ug/L




                CRQL = 10 ug/L


                MDL  -  0.25 ug/L

                MDL  = 2.8 ug/L

                NA


                MDL  = 0.02 ug/L

-------
                                                           APPENDIX III
                                                            TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                       AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTJTATJON/
DETECTION LIMIT
 Dichloromethane
 (Methylera Chloride)
 75092
Ethenyl Benzene
(Styrene)
100425
 EPA DW METHOD 524.1/SMEWW Method 6210B (Method I)/SMEWW             GC-MS
 Method 62 IOC (Method II) •Measurement of Purgeible Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry*

 EPA DW METHOD 524.2 /SMEWW Method 62IOD 'Measurement of Purgeable       GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry*

 SMEWW METHOD 6230C "Purge and Trap Packed-Column Gas                   GC-MS
 Chromatographic Method II*

 SMEWW METHOD 6230D "Purge and Trap Capillary-Column Gas                 GC-ECD
 Chromatographic Method"

 SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
 Organict"

 CLP SOW METHOD LC-ORG 'Chemical Analytical Services for the Analysis of       GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
 Capture (GC-ECD) Techniques*

 CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-          GC-MS
 Media, Multi-Concentration*

 EPA METHOD 602 'Purgeable Aromatics"                                    GC-PID

 EPA DW METHOD 502.2  "Volatile Organic Compounds in Water by Puige and        GC-PID
Trap Capillary Column Gas Chromatography with Photoionization  and Electrolytic
Conductivity Detectors in Series"

EPA DW METHOD 503.1  "Volatile Aromatic and Unsaturated Organic                GC-PID
Compounds in Water by Purge and Trap Gas Chromatography"
                 MDL - l.Oug/L
                 MDL -2.8 ug/L
                 MDL « 0.03 ug/L



                 MDL « 0.25 ug/L


                 NA


                 PQL -  5.0 ug/L


                CRQL=« 1.0 ug/L




                CRQL « 10 ug/L


                MDL - 0.20 ug/L

                MDL = 0.01 ug/L



                MDL » 0.008 ug/L

-------
                                                            APPENDIX III
                                                              TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT

                                                        AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
Ethenyl Benzene
(Stymie)
100425
METHOD REFERENCE/TITLE OF METHOD
EPA DW METHOD 524. 1/SMgWW Method 62 IOC -Measurement of Purgeable
Organic Compounds in Water by Packed Column Gas Chromatography-Mass
Spectrometry"
INSTRUMENT-
ATION
GC-MS
QUANTTTATION/
DETECTION LIMIT
MDL - 0.2 ug/L
M Tetnchloroethene
5 (Tetnchloroethylene)
   127184
                     EPA DW METHOD 524.2/SMEWW Method 6210D 'Measurement of Purgeable       GC-MS
                     Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
                     Spectrometry"

                     SW846 METHOD 8240 'Gas Chromatography-Mass Spectrometry for Volatile          GC-MS
                     Organics*

                     CLP SOW METHOD ORG  "Statement of Work for Organics Analysis - Multi-          GC-MS
                     Media, Multi-Concentration'

                     CLPSOWLC-ORG "Chemical Analytical Services for Analysis of Low                GC-MS
                     Concentration Water Samples for Organic Compounds by Gas Chromatography-
                     Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron Capture (GC-
                     ECD) Technique"

                    CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media,             GC-ECD
                    Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
                    Chromatography Techniques'

                    EPA METHOD 601/SW846  Method 8010/SMEWW Method 6230B  "Purgeable          GC-ELCD
                    Halocarboas"

                    EPA METHOD 624  "Purgeables'                                             GC-MS

                    EPA DW METHOD 502.1 'Volatile Halogenated Organic Compounds in Water         GC-ELCD
                    by Purge and Trap Gas Chromatography*

                    EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and         GC-PID
                    Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
                    Conductivity Detectors in Series"
 MDL - 0.04 ug/L



 PQL - 5.0 ug/L


 CRQL - 10 ug/L


 CRQL- 1.0 ug/L




 CRQL - 20 ug/L



 MDL - 0.03 ug/L


 MDL = 4.1 ug/L

MDL = 0.001 ug/L


MDL - 0.05 ug/L

-------
                                                              APPENDIX III
                                                               TABLE III
              METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                          AQUEOUS MATRICES
   ANALYTE/
   COMMON NAME
   CAS NUMBER
                         METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITAT1ON/
DETECTION LIMIT
   Tetrachloroethene
   (Tetnchloroethylene)
   127184
K>
                     EPA DW METHOD 502.2 'Volatile Organic Compounds in Water by Purge and        GC-ELCD
                     Trap Capillary Column Gas Chromatography with Photoionizatioo and Electrolytic
                     Conductivity Detectors in Series*

                     EPA DW METHOD 503.1 "Volatile Aromatic and Unsaturated Organic               GC-PID
                     Compounds in Water by Purge and Trap Gas Chromatography"

                     EPA DW METHOD 524.1/SMEWW Method 621 OB (Method D/SMEWW              GC-MS
                     Method 6210C (Method II) 'Measurement of Purgeable Organic Compounds in
                     Water by Packed Column Gas Chromatography-Mass Spectrometry"

                     EPA DW METHOD 524.2/SMEWW Method 6210D "Measurement of Purgeable        GC-MS
                     Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
                     Spectrometry"

                     SMEWW METHOD 6040B  "Closed-Loop Stripping. Gas-Chronutographic-Mass-        GC-MC
                     Spectrometric Analysis"

                     SMEWW METHOD 6230C  Purge and Trap Packed-Column Gas                     GC-MS
                     Chromatographic Method II"

                     SMEWW METHOD 6230D  "Purge and Trap Capillary-Column Gas                  GC-PID/
                     Chromatographic Method"                                                  GC-ECD

                     SW846 METHOD 8240G "Gas Chromatography-Mass Spectrometry for Volatile         GC-MS
                     Organics"

Tetnchloromethane      CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-MS
(Carbon Tetrachloride)   Low Concentration Water Samples for Organic Compounds by Gas
56235                Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-Electron
                     Capture (GC-ECD) Techniques"

                     CLP SOW METHOD ORG "Statement of Work for Organics Analysis - Multi-          GC-MS
                     Media, Multi-Concentration"
                 MDL « 0.04 ug/L
                                                                                                             MDL - 0.01 ug/L
                                                                                                             MDL = 0.3 ug/L, 4.1 ug/L
                                                                                                             MDL » 4.1 ug/L
                MDL = 0.14 ug/L



                EDL = 0.10 ug/L


                MDL = 0.03 ug/L


                NA


                PQL = 5.0 ug/L


                CRQL = 1.0 ug/L




                CRQL = 10 ug/L

-------
                                                              APPENDIX III
                                                                TABLE III
               METHOOS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
                                                          AQUEOUS MATRICES
ANALYTE/
COMMON NAME
CAS NUMBER
                            METHOD REFERENCE/TITLE OF METHOD
                                                                      INSTRUMENT-
                                                                      ATION
                 QUANTTTATION/
                 DETECTION LIMIT
                 M     CLP SOW METHOD QTM  "Chemical Analytical Services for Multi-Media.
    (Carbon TetracUoride)   Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
    56235
10
10
 Quomatognphy Techniques"

 EPA METHOD 601/SW846 Method 8010/SMEWW Method 6230B "Purgeable
 Halocarbons"

 EPA METHOD 624 "Purgeables"

 EPA DW METHOD 502.1 "Volatile Halogenated Organic Compounds in Water
 by Purge and Trap Gas Chromatography"

 EPA DW METHOD 502.2 -Volatile Organic Compounds in Water by Purge and
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series"

 EPA DW METHOD 524. 1/SMEWW Method 6210B (Method Q/SMEWW
 Method 6210C (Method II) "Measurement of Purgeable Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry"

 EPA DW METHOD 524.2/SMEWW Method 6210D "Measurement of Purgeable
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
 Spectrometry"

 SMEWW METHOD 6230C "Purge and Trap Packed-Column Gas
 Chromaiographic Method IT

SMEWW METHOD 6230D "Purge and Trap Capillary-Column Gas
Chromatographic Method"

SW846 METHOD 8240 "Gas Chromatography-Mass Spectrometry for Volatile
Organicc"
GC-ECD         CRQL - 20 ng/L



GC-ELCD        MOL » 0.12 ug/L


GC-MS          MDL-2.8 ug/L

GC-ELCD        MDL - 0.003 ug/L


GC-ELCD        MDL • 0.01 ng/L
                                                                                            GC-MS          MDL = 0.3 ug/L, 2.8 ug/L
                                                                                                            MDL • 2.8 ug/L
                                                                                            GC-MS          MDL - 0.21 ug/L
                                                                                            GC-MS          MDL-0.12 ug/L
                                                                                            GC-ECD         NA
                                                                                            GC-MS          PQL - 5.0 ug/L

-------
                                                           APPENDIX 111
                                                             TABLE III
            METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                       AQUEOUS MATRICES
     METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
QUANTITATION/
DETECTION LIMIT
Trichloromethane
(Chloroform)
67663
 CLP SOW METHOD LC-ORG "Chemical Analytical Services for the Analysis of        GC-MS
 Low Concentration Water Samples for Organic Compounds by Gas
 Chromatography-Mass Spectrometry (GC-MS) and Gas Chromatography-EIectron
 Capture (GC-ECD) Techniques'

 CLP SOW METHOD ORG •Statement of Work for Organics Analysis - Multi-          GC-MS
 Media, Multi-Concentration*

 CLP SOW METHOD QTM •Chemical Analytical Services for Multi-Media,            GC-ECD
 Multi-Concentration Samples for Organic Analysis by Quick Turnaround Gas
 Chromatography Techniques"

 EPA METHOD 601/SW846 Method 8010/SMEWW Method 6230B 'Purgeable         GC-ELCD
 Halocarbons*

 EPA METHOD 624 -purgeables-                                            GC-MS

 EPA DW METHOD 502.1 "Volatile Halogenated Organic Compounds in Water         GC-KLCD
 by Purge and Trap Gas Chromatography*                                           <

 EPA DW METHOD 502.2 "Volatile Organic Compounds in Water by Purge and         GC-ELCD
 Trap Capillary Column Gas Chromatography with Photoionization and Electrolytic
 Conductivity Detectors in Series"

 EPA DW METHOD 524.1/SMEWW Method 6210B (Method I)/SMEWW              GC-MS
 Method 6210C (Method II) "Measurement of Purgeable Organic Compounds in
 Water by Packed Column Gas Chromatography-Mass Spectrometry*

 EPA DW METHOD 524.2/SMEWW Method 6210D "Measurement .of Purgeable         GC-MS
 Organic Compounds in Water by Capillary Column Gas Chromatography-Mass
Spectrometry"

SMEWW METHOD 6230C  "Purge and Trap Packed-Column Gas                    GC-MS
Chromatographic Method II"
                 CRQL = l.Oug/L




                 CRQL = 10 ug/L


                 CRQL =• 20 ug/L



                 MDL « 0.05 ug/L


                 MDL= 1.6 ug/L

                 NA


                MDL = 0.02 ug/L
                                                                                                          MDL = 0.2 ug/L, 1.6 ug/L
                                                                                                          MDL= 1.6 ug/L
                                                                                                         MDL = 0.03 ug/L
                                                                                                         MDL » 0.05 ug/L

-------
                                                     APPENDIX III
                                                       TABLE III
          METHODS AND DETECTION/QUANTITATION LIMITS FOR SPECIFIED ANALYTES OF CONCERN TO RISK ASSESSMENT
ANALYTE/
COMMON NAME
CAS NUMBER
                                                  AQUEOUS MATRICES
    METHOD REFERENCE/TITLE OF METHOD
INSTRUMENT-
ATION
               QUANTITATION/
               DETECTION LIMIT
Trichloromethane
(Chloroform)
67663
SMEWW METHOD 6230D "Purge and Trap Capillary-Column Gas
Chromatognphic Method*

SW846 METHOD 8240 "Gas Chromatography-Mass Spectromctry for Volatile
Organics'
GC-ECD
                                                                                 GC-MS
               NA
              PQL = 5.0 ug/L

-------
   METHOD REFERENCE
                                                                  APPENDIX HI
                                                                   TABLE IV
                                                      METHOD TITLES AND APPLICATIONS
                                   TITLE OF METHOD
                                                              APPLICATION OF METHOD
   *CLP SOW

   METHOD INORG


   METHOD LC-ORG
   METHOD ORG
10
METHOD QTM
    IEA

   METHOD 601



   METHOD 602
                       "Sutement of Work for Inorganics Analysis - Multi-Media,
                       Multi-Concentration,* Doc No. ILM02.0

                       "Chemical Analytical Services for the Analysis of Low
                       Concentration Water Samples for Organic Compounds by Gas
                       Chromatography-Mass Spectrometry (GC-MS) and Gas
                       Chromatography-Electron Capture (GC-ECD) Techniques,*
                       6/91 Draft

                       'Statement of Work for Organics Analysis - Multi-Media,
                       Multi-Concentration, * Doc No. OLM01.8  (8/91)
                          'Chemical Analytical Services for Multi-Media, Multi-
                          Concentration Samples for Organic Analysis by Quick
                          Turnaround Gas Chromatography Techniques,' Draft 7/91
•Purgeable Halocarbons*
'Purgeable Aromatics"
                                                       This method is for the analysis of 23 metals and cyanide. Sample matrices
                                                       compatible with this method include water and soil/sediment.

                                                       This method consists of three separate methods. These methods are for
                                                       the analysis of 40 volatile compounds, 60 semivolatile compounds and 28
                                                       organochlorine pesticides and Aroclors.  Sample matrices compatible with
                                                       this method include drinking water, surface water and groundwater.
                                                       This method consists of three separate methods. These methods are for
                                                       the analysis of 34 volatile compounds, 65 semivolatile compounds and 27
                                                       organochlorine pesticides and Aroclors.  Sample matrices compatible with
                                                       these methods include water and soil/sediment.

                                                       This method consists of five separate methods. These methods are for the
                                                       analysis of 21 volatile compounds, 16 polynuclear aromatic hydrocarbons,
                                                       16 phenols. 19 pesticides and 8 Aroclors plus toxaphene.  Sample matrices
                                                       compatible with this method include water and soil/sediment.
                                                                             This method is for the analysis of 29 purgeable halocarbons. Sample
                                                                             matrices compatible with this method include municipal and industrial
                                                                             discharges.

                                                                             This method is for the analysis of seven purgeable aromatic compounds.
                                                                             Sample matrices compatible with this method include municipal and
                                                                             industrial discharges.
   1
   CLP SOW      CONTRACT LABORATORY PROGRAM (CLP) STATEMENT OF WORK. OFFICE OF EMERGENCY AND REMEDIAL RESPONSE
   'EPA
               GUIDELINES ESTABLISHING TEST PROCEDURES FOR THE ANALYSIS OF POLLUTANTS UNDER THE CLEAN WATER ACT FINAL
               RULE AND INTERIM FINAL RULE AND PROPOSED RULE, 10/84, 40 CFR PART 136

-------
   METHOD REFERENCE
                                                                     APPENDIX III
                                                                      TABLE IV
                                                        METHOD TITLES AND APPLICATIONS
                                     TITLE OF METHOD
                                                                                            APPLICATION OF METHOD
K>
EPA

METHOD 606



METHOD 607


METHOD 608



METHOD 609



METHOD 610



METHOD 612



METHOD 624



METHOD 625
                           "Phthalate Ester'
                           'Nitrosamines*
                           •Organochlorine Pesticides and PCBs*
                           •Nitroaromatics and Isophorone*
                           •Polynuclear Aromatic Hydrocarbons'
                           'Chlorinated Hydrocarbons'
                           "Purgeables*
                           'Base/Neutrals and Acids'
This method is for the analysis of six phthalate ester compounds. Sample
matrices compatible with this method include municipal and industrial
discharges.

This method is for the analysis of three nitrosamines. Sample matrices
compatible with this meCwd include municipal and industrial discharges.

This method is for the analysis of 27 organochlorine pesticides and
Aroclors. Sample matrices compatible with this method include municipal
and industrial discharges.

This method is for the analysis of four nitroaromatics and isophorone.
Sample matrices compatible with this method include municipal and
industrial discharges.

This method is for the analysis of 16 polynuclear aromatic hydrocarbons.
Sample matrices compatible with this method include municipal and
industrial discharges.

This method is for the analysis of nine chlorinated hydrocarbons. Sample
matrices compatible with this method include municipal and industrial
discharges.

This method is for the analysis of 30-33 purgeable organic compounds.
Sample matrices compatible with this method include municipal and
industrial discharges.

This method is for the analysis of 80-84 semivolatile compounds. Sample
matrices compatible with this method include municipal and industrial
discharges.

-------
                                                                   APPENDIX III
                                                                     TABLE IV
                                                       METHOD TITLES AND APPLICATIONS
   METHOD REFERENCE
            TITLE OF METHOD
    'EPA AIR
   METHOD TO-1
   METHOD TO-I4
   METHOD TO-2
5 METHOD TO-3
   METHOD TO-4
'Method for the Determination £f Volatile Organic
Compounds in Ambient Air Using Tenax Adsorption and Gas
Chromatography-Mass Spectrometry (GC-MS)"

"The Determination of Volatile Organic Compounds (VOCs)
in Ambient Air Using Sununa Passivated Canister Sampling
and Gas Chromatographic Analysis"

'Method for the Determination of Volatile Organic
Compounds in Ambient Air by Carbon Molecular Sieve
Adsorption and Gas Chromatography-Mass Spectrometry
(GC-MS)"

"Method for the Determination of Volatile Organic
Compounds in Ambient Air Using Cryogenic
Preconcentration Techniques and Gas Chromatography with
Flame lonization and Electron Capture Detection"

"Method for the Determination of Organochlorine Pesticides
and Polychlorinated Biphenyls in Ambient Air"
       APPLICATION OF METHOD
This method is for the analysis of 18 nonpolar volatile compounds with
boiling points between 80 and 200 degrees °C. Samples are collected on
pre-cleaned tenax cartridges.

This method is for the analysis of 40 volatile organic compounds. Samples
are collected on cleaned and certified SUMMA canisters.
This method is for the analysis of 11 volatile organic compounds with
boiling points between -\5 and 120 degrees *C. Samples are collected on
pre-cleaned carbon molecular sieves.
This method is for the analysis of eight volatile organic compounds with
boiling points between -10 and 200 degrees °C.
This method is for the analysis of 11 Organochlorine pesticides and
Aroclors.  Samples are collected on polyurethane foam filters. Samples
are prepared using a Soxhlet extraction. Analysis is performed by GC-
ECD.
   3EPA AIR     COMPENDIUM OF METHODS FOR THE DETERMINATION OF TOXIC ORGANIC COMPOUNDS IN AMBIENT AIR, 5/88,
                 ENVIRONMENTAL MONITORING SYSTEMS LABORATORY/RTP, EPA 600/4-84-041

-------
                METHOD REFERENCE
                                    TITLE OF METHOD
                                                                                 APPENDIX III
                                                                                   TABLE IV
                                                                     METHOD TITLES AND APPLICATIONS
                                                                                                        APPLICATION OF METHOD
r
 EEAJ2W

METHOD 502.1



METHOD 502.2



METHOD 503.1



METHOD 505



METHOD 508



METHOD 524.1



METHOD 524.2
                                        •Volatile Halogenated Organic Compounds in Water by Purge
                                        and Trap Gas Chromatography"
                                        •Volatile Organic Compounds in Water by Purge and Trap
                                        Capillary Column Gas Chromatography with Photoionization
                                        and Electrolytic Conductivity Detectors in Series'

                                        •Volatile Aromatic and Unsatureted Organic Compounds in
                                        Water by Purge and Trap Gas Chromatography*
                                        •Analysis of Organohalide Pesticides and Aroclors in Water
                                        by Microextnction and Chromatography'
                                        •Determination of Chlorinated Pesticides in Water by Gas
                                        Chromatography with an Electron Capture Detector'
                                        •Measurement of Purgeable Organic Compounds in Water by
                                        Packed Column Gas Chromatography-Mass Spectrometry'
                                        •Measurement of Purgeable Organic Compounds in Water by
                                        Capillary Column Gas Chrom»tography-Mass Spectrometry"
This method is for the analysis of 40 halogenated volatile organic
compounds. Sample matrices compatible with this method include
drinking water, source water and water being treated for potability.

This method is for the analysis of 60 volatile organic compounds.  Sample
matrices compatible with this method include drinking water, source water
and water being treated for potability.

This method is for the analysis of 28 aromatic and unsaturated organic
compounds. Sample matrices compatible with this method include
drinking water, source water and water being treated for potability.

This method is for the analysis of 25 organohalide pesticides and
Aroclors. Sample matrices compatible with this method include drinking
water, source water and water being treated for potability.

This method is for the analysis of 34 chlorinated pesticides and Aroclors.
Sample matrices compatible with this method include groundwater and
drinking water.

This method is for analysis of 48 volatile compounds.  Sample matrices
compatible with this method include drinking water, source water and
water being treated for potability.

This method is for the analysts of 60 volatile organic compounds.  Sample
matrices compatible with this method include drinking  water, source water
and water being tested for potability.
                4EPA DW      METHODS FOR THE DETERMINATION OF ORGANIC COMPOUNDS IN DRINKING WATER, 12/88. ENVIRONMENTAL MONITORING
                               SYSTEMS LABORATORY/CINN, EPA 600/4-88/039

-------
     METHOD REFERENCE
                                                                      APPENDIX HI
                                                                       TABLE IV
                                                         METHOD TITLES AND APPLICATIONS
                                    TITLE OF METHOD
       APPLICATION OF METHOD
jo
VO
EPADW

METHOD 525



SMCAWW

METHOD 200.7


METHOD 206.2


METHOD 206.3



METHOD 206.4



METHOD 206.5



METHOD 210.1
                             "Determination of Organic Compounds in Drinking Water by
                             Liquid-Solid Extraction and Capillary Column Gas
                             Chromatography-Mass Spectrometry"
                             "Inductively Coupled Plasma-Atomic Emission Spectrometric
                             Method for Trace Element Analysis of Water and Wastes"

                             "Arsenic (Atomic Absorption, Furnace Technique)"
                             "Arsenic (Atomic Absorption-Gaseous Hydride)"
                             "Arsenic (Spectrophotometric-SDDC)"
                             "Arsenic (Sample Digestion prior to Total Arsenic Analysis
                             by Silver Diethyldithiocarbamate or Hydride Procedures)"
                             "Beryllium (Atomic Absorption, Direct Aspiration)"
This method is for the analysis of 35 organic compounds.  Sample
matrices compatible with this method include drinking water, source water
and water being treated for potability.
This method is for the analysis of 30 metals.  Sample matrices compatible
with this method include drinking water, surface water and wastewater.

Sample matrices compatible with this method include drinking water,
surface water, saline water, waste, sludge and soil/sediment.

This method is for the analysis of inorganic arsenic. Sample matrices
compatible with this method include drinking water, fresh water and saline
water.

This method is for the analysis of inorganic arsenic. Sample matrices
compatible with this method include drinking water, surface water,
groundwater and wastes.

This method is a preparation procedure for the conversion of organic
arsenic to inorganic arsenic. Sample matrices compatible with this method
include drinking water, surface water and waste.

Sample matrices compatible with this method include drinking water,
surface water, groundwater, waste, sludge and soil/sediment.
      MCAWW     METHOD FOR CHEMICAL ANALYSIS OF WATER AND WASTES, 3/83, ENVIRONMENTAL MONITORING SYSTEMS
                    LABORATORY/CINN. EPA 600/4-79/020

-------
                                                                   APPENDIX III
                                                                    TABLE IV
                                                      METHOD TITLES AND APPLICATIONS
 METHOD REFERENCE
             TITLE OF METHOD
        APPLICATION OF METHOD
 MCAWW

 METHOD 210.2


 METHOD 213.1


 METHOD 213.2


 METHOD 218.1


 METHOD 218.2


 METHOD 218.3


 METHOD 218.4


 METHOD 218.5


 METHOD 239.1


 METHOD 239.2


METHOD 245.1
 "Beryllium (Atomic Absorption, Furnace Technique)"
 "Cadmium (Atomic Absorption, Direct Aspiration)"
 "Cadmium (Atomic Absorption. Furnace Technique)"
 "Chromium (Atomic Absorption, Direct Aspiration)*
"Chromium (Atomic Absorption, Furnace Technique)"
"Chromium (Atomic Absorption, dictation- Extraction)"
"Chromium, Hexavalent (Atomic Absorption, Chelation-
Extraction)"

"Chromium, Dissolved Hexavalent (Atomic Absorption,
Furnace Technique)"

"Lead (Atomic Absorption, Direct Aspiration)"
"Lead (Atomic Absorption, Furnace Technique)"
"Mercury (Manual Cold Vapor Technique)"
 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater and waste.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater and waste.

 Sample matrices compatible with this method include drinking water,
 surface water and certain filtered wastes.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

 Sample matrices compatible with this method include drinking water,
 surface water, groundwater, waste, sludge and soil/sediment.

Sample matrices compatible with this method include drinking water,
surface water and saline water.

-------
     .„„>
                                                                 APPENDIX HI
                                                                  TABLE IV
                                                    METHOD TITLES AND APPLICATIONS
METHOD REFERENCE
            TITLE OF METHOD
        APPLICATION OF METHOD
MCAWW

METHOD 245.2


METHOD 245.5


METHOD 335.1



METHOD 335.2


*SMEWW

METHOD 311 IB


METHOD 3111C



METHOD 311 ID


METHOD 31 ME
'Mercury (Automated Cold Vapor Technique)"
"Mercury in Sediment (Manual Cold Vapor Technique)"
"Cyanide, Amendable to Chlorination*
"Cyanide, Total (Titrimetric, Spectrophotometric)"
•Direct Air-Acetylene Flame Method"
"Extraction/Air-Acetylene Flame Method"
•Direct Nitrous Oxide-Acetylene Flame Method"
"Extraction/Nitrous Oxide-Acetylene Flame Method"
Sample matrices compatible with this method include surface water, waste
water and effluent.

Sample matrices compatible with this method include bottom deposits,
sludge and soil/sediment.

This method is applicable to the determination of cyanide amenable to
chlorination in drinking, surface and saline waters and domestic and
industrial waste*.

This method is applicable to the determination of cyanide in drinking,
surface and saline waters and domestic and industrial wastes.
This method is for the analysis of 27 metals.  Sample matrices compatible
with this method include surface water, groundwater and drinking water.

This method is for the analysis of 10 metals at low concentrations.
Sample matrices compatible with this method include surface water,
groundwater and drinking water.

This method is for the analysis of 10 metals.  Sample matrices compatible
with this method include groundwater, surface water and drinking water.

This method is for the analysis of aluminum and beryllium. Sample
matrices compatible with this analysis include groundwater, surface water
and drinking water.
6SMEWW    STANDARD METHODS FOR THE EXAMINATION OF WATER AND WASTEWATER. 17TH EDITION. 1989

-------
    METHOD REFERENCE
TITLE OF METHOD
                                                                     APPENDIX III
                                                                       TABLE IV
                                                         METHOD TITLES AND APPLICATIONS
                                                                                            APPLICATION OF METHOD
SMEWW

METHOD 3112B


METHOD 3113B



METHOD 3114B



METHOD 3120B
                            "Cold Vapor Atomic Absorption Spectrometric Method*
                            "Electrothermal Atomic Absorption Spectrometric Method*
                            'Manual Hydride Generation/Atomic Absorption
                            Spectrometric Method*
                            'Inductively Coupled Plasma (ICP) Method'
N   METHOD 3500AS C*    "Silver Diethyldithiocarbamate Method"
    METHOD 3500BE D*    "Aluminon Method*
    METHOD 3500CD D*    "Dithizone Method*
    METHOD 3500CR D*    'Colorimetric Method*
                                            This method is for the analysis of mercury. Sample matrices compatible
                                            with this method include groundwater, surface water and drinking water.

                                            This method is for the analysis of 17 metals in microquantities.  Sample
                                            matrices compatible with this method include groundwater. surface water
                                            and drinking water.

                                            This method is for the analysis of arsenic and selenium. Sample matrices
                                            compatible with this method include groundwater, surface water and
                                            drinking water.

                                            This method is for the analysis of 27 metals.  Sample matrices compatible
                                            with this method include groundwater,  surface water and drinking water.

                                            This method is for the analysis of arsenic. Sample matrices compatible
                                            with this method include groundwater,  surface water and drinking water.

                                            This method is for the analysis of beryllium. Sample matrices compatible
                                            with this method include groundwater,  surface water and drinking water.

                                            This method is for the analysis of cadmium. Sample matrices compatible
                                            with this method include groundwater,  surface water and drinking water.

                                            This method is for the analysis of chromium. Sample matrices compatible
                                            with this method include groundwater,  surface water and drinking water.
                   The first two letters after the number represent the element name and the third letter is the method code.

-------
METHOD REFERENCE
                                                                  APPENDIX HI
                                                                   TABLE IV
                                                     METHOD TITLES AND APPLICATIONS
             TITLE OF METHOD
        APPLICATION OF METHOD
SMEWW

METHOD 3500HG C*    'Dithizone Method'


METHOD 3500PB D*    'Dithizone Method'


METHOD 4500 CN      'Cyanide'
METHOD 6040B
METHOD 6210B
METHOD 6210D
METHOD 6220B
METHOD 6220C
'Closed-Loop Stripping, Gas Chromatographic-Mass
Spectrometric Analysis'
"Purge and Trap Packed-Column Gas Chromatographic-Mass
Spectrometric Method I*
'Purge and Trap Capillary-Column Gas Chromatographic- %
Mass Spectrometric Method'
"Purge and Trap Gas Chromatographic Method I"
'Purge and Trap Gas Chromatographic Method II*
This method is for the analysis of mercury. Sample matrices compatible
with this method include groundwater, surface water and drinking water.

This method is for the analysis of lead. Sample matrices compatible with
this method include groundwater, surface water and drinking water.

This method is used for the analysis for cyanide in aqueous and solid
matrices.  It includes total cyanide, cyanide amenable to chlorination, and
weak and dissociable cyanides.

This method is for the analysis of volatile organic compounds of
intermediate weight.  Sample matrices compatible with this method
include groundwater,  surface water and drinking water.

This method is for the analysis of 31 volatile organic compounds. Sample
matrices compatible with this method include groundwater, surface water
and drinking water.

This method is for the analysis of 62 purgeable organic compounds.
Sample matrices compatible with this method include drinking water, raw
source water and water being treated for potability.

This method is for the analysis of seven aromatic volatile compounds.
Sample matrices compatible with this method include groundwater,
surface water and drinking water.

This method is for the analysts of 28 purgeable aromatic and unsaturated
compounds. Sample matrices compatible with this method include
drinking water, raw source water, and water being treated for potability.
       The first two letters after the number represent the element name and the third letter is the method code.

-------
 METHOD REFERENCE
                                                                 APPENDIX III
                                                                   TABLE IV
                                                     METHOD TITLES AND APPLICATIONS
                      TITLE OF METHOD
                                                                APPLICATION OF METHOD
 SMEWW

 METHOD 6230B



 METHOD 6230C



 METHOD 6230D



 METHOD 6410B



 METHOD 6440B



 METHOD 6630B


 METHOD 6630C



8SW846

METHOD 6010
         "Purge ••* Trap Packed Column Gis Chromatographic
         Method r
         "Purge ami Trap Packed Column Gas Chromatographic
         Method! IT
         'Purge aoiTrap Capillary-Column Gas Chromatographic
         MethodT
         'Liijuhriiqaiil Extraction Gas Chromatographic-Mass
         Spectromrtric Method*
         "Liquidl-lJqpid Extraction Chromatographic Method'
              idUJqpnd Extraction Gas Chromatographic Method I*
              idUJvrid Extraction Gas Chromatographic Method II*
"Inductiwrfy Coupled Plasma Atomic Emission Spectroscopy"
                                                         This method is for the analysis of 29 purgeable halocarbons. Sample
                                                         matrices compatible with this method include municipal and industrial
                                                         discharges.

                                                         This method is for the analysis of 39 purgeable halocarbons. Sample
                                                         matrices compatible with this method include drinking water, raw source
                                                         water and water being treated for potability.

                                                         This method is for the analysis of 60 purgeable halocarbons. Sample  '
                                                         matrices compatible with this method include drinking water, raw source
                                                         water and water being treated for potability.

                                                         This method is for the analysis of 91 semi volatile organic compounds.
                                                         Sample matrices compatible with this method include groundwater,
                                                         surface water and drinking water.

                                                         This method is for the analysis of 16 polynuclear aromatic hydrocarbons.
                                                         Sample matrices compatible with this method include municipal and
                                                         industrial discharges.

                                                         This method is for the analysis of 18 organochlorine pesticides. Sample
                                                         matrices compatible with this method include agricultural discharges.

                                                         This method is for the analysis of 25 organochlorine pesticides. Sample
                                                         matrices compatible with this method include groundwater. surface water
                                                         and drinking water.
                                                                 This method is for the analysis of 26 metals. Sample matrices compatible
                                                                 with this method include groundwater, soils and wastes.
8
 'SW846
TEST METHODS FOR EVALUATING SOLID WASTE. THIRD EDITION, 11/86. OFFICE OF SOLID WASTE AND EMERGENCY RESPONSE.

-------
                                                                 APPENDIX III
                                                                   TABLE IV
                                                     METHOD TITLES AND APPLICATIONS
METHOD REFERENCE
                                     TITLE OF METHOD
                                                                 APPLICATION OF METHOD
 SW846

 METHOD 7060


 METHOD 7061


 METHOD 7090

 METHOD 7091

 METHOD 7130


 METHOD 7131


 METHOD 7190


 METHOD 7191


 METHOD 7195


METHOD 7196


METHOD 7197
 "Arsenic (Atomic Absorption, Furnace Technique)*


 "Arsenic (Atomic Absorption, Gaseous Hydride)"


 "Beryllium (Atomic Absorption, Direct Aspiration)*

 "Beryllium (Atomic Absorption, Furnace Technique)*

 "Cadmium (Atomic Absorption, Direct Aspiration)"


 "Cadmium (Atomic Absorption, Furnace Technique)*


 "Chromium (Atomic Absorption, Direct Aspiration)"


 "Chromium (Atomic Absorption, Furnace Technique)*


 "Chromium, Hexavalent (Coprecipitation)"


 "Chromium. Hexavalent (Colorimetric)"


"Chromium, Hexavalent (Chelation/Ex tract ion)"
                                                                                 Sample matrices compatible with this method include groundwater, soils,
                                                                                 extracts and wastes.

                                                                                 Sample matrices compatible with this method include groundwater, soils,
                                                                                 extracts and wastes.

                                                                                 Sample matrices compatible with this method include water and wastes.

                                                                                 Sample matrices compatible with this method include water and wastes.

                                                                                 Sample matrices compatible with this method include water, waste and
                                                                                 sludge.

                                                                                 Sample matrices compatible with this method include water, soil and
                                                                                 waste.

                                                                                 Sample matrices compatible with this method include water, soil and
                                                                                 waste.

                                                                                 Sample matrices compatible with this method include water, soil and
                                                                                waste.

                                                                                This method is for the analysis of dissolved hexavalent chromium in
                                                                                extraction procedure (EP) toxicity extracts and groundwater.

                                                                                This method is for the analysis of dissolved hexavalent chromium in
                                                                                extraction procedure (EP) toxicity characteristic extracts and groundwater.
                                                                                               *
                                                                                This method is for the analysis of dissolved hexavalent chromium in
                                                                                extraction procedure (EP) toxicity extracts and groundwater.

-------
                                                                   APPENDIX III
                                                                     TABLE IV
                                                      METHOD TITLES AND APPLICATIONS
 METHOD REFERENCE
              TITLE OF METHOD
        APPLICATION OF METHOD
SW846

METHOD 7198


METHOD 9010A



METHOD 9012



METHOD 7420


METHOD 7421


METHOD 7470


METHOD 7471


METHOD 8010



METHOD 8020
 'Chromium, Hexwalent (Differential Pulse Polarography)"
 "Total and Amenable Cyanide*
 "Total and Amenable Cyanide (Colorimetric, Automated
 UV)-
"Lead (Atomic Absorption, Direct Aspiration)"
"Lead (Atomic Absorption, Furnace Technique)"
"Mercury in Liquid Waste (Manual Cold-Vapor Technique)"
"Mercury in Solid or Semisolid Waste (Manual Cold-Vapor
Technique)"

"Halogenated Volatile Organics"
"Aromatic Volatile Organics*
 This method is for the analysis of dissolved hexavalent chromium in
 extraction procedure (EP) toxicity extracts, natural water and waste water.

 This method is for the analysis of inorganic cyanide (total and amendable
 to chlorination) in waste and leachate. The method detects inorganic
 cyanides that are present as either soluble salts or complexes.

 This method is for the analysis of inorganic cyanide (total and amendable
 to chlorination) in waste and leachate. The method detects inorganic
 cyanides that are present as either soluble salts or complexes.

 Sample matrices compatible with this method include water, waste and
 sludge.

 Sample matrices compatible with this method include water, waste and
 soils.

 Sample matrices compatible with this method include groundwater,
 aqueous waste and mobility procedure extracts.

 This method is for the analysis of inorganic and organic mercury. Sample
 matrices compatible with this method include soil, sludge and sediment.

 This method is for the analysis of 34 halogenated volatile organic
 compounds. Sample matrices compatible with this method include
 soil/sludge, groundwater, liquid waste and water immiscible waste.

 This method is for the analysis of seven aromatic volatile organic
compounds. Sample matrices compatible with this method include
soil/sludge, groundwater. liquid waste and water immiscible waste.

-------
                                                                  APPENDIX III
                                                                    TABLE IV
                                                      METHOD TITLES AND APPLICATIONS
 METHOD REFERENCE
             TITLE OF METHOD
        APPLICATION OF METHOD
 METHOD 8060



 METHOD 8080



 METHOD 8100



 METHOD 8240


 METHOD 8250



METHOD 8270


METHOD 8310
 "Phthalate Esters"
 "OrganocbJorine Pesticides and PCBs"
 "Polynuclear Aromatic Hydrocarbons"
"Gas Chromatography-Mass Spectrometry for Volatile
Organics Packed Column Technique"

"Gas Chromatography-Mass Spectrometry for Semivolatile
Organics: Packed Column Technique"
"Gas Chromatography-Mass Spectrometry for Semivolatile
Organics: Capillary Column Technique"

"Polynuclear Aromatic Hydrocarbons"
This method is for the analysis of six phthalate esters. Sample matrices
compatible with this method include water, soil, sludge and water
immiscible waste.

This method is for the analysis of 26 organochlorine pesticides and
Aroclors.  Sample matrices compatible with this method include water,
soil, sludge and water immiscible waste.

This method is for the analysis of 24 polynuclear aromatic hydrocarbons.
Sample matrices compatible with this method include groundwater,
surface water, drinking water and soil/sediment.

This method is for the analysis of 73 volatile organic compounds. Sample
matrices include groundwater, caustic or acid liquors, and soil/sediment.

This method is for the analysis of 113 Semivolatile organic compounds.
Sample matrices compatible with this method include solid waste, soil and
groundwater.

This method is for the analysis of 131 Semivolatile compounds. Sample
matrices compatible with this method include groundwater, waste and soil.

This method is for the analysis of 16 polynuclear aromatic hydrocarbons.
Sample matrices compatible with this method include waters, soil, waste
and sludge.

-------
                      APPENDIX III

                       Table V- A
SUMMARY OF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
                  ORGANIC COMPOUNDS
Drinking Water (USEPA, Office of Water)
EPA
Comoound Class ^lethad No
Acrolein and Acrylonitrile
Base/Neutrals, Acids and
Pesticides
Jenzidines
Carbamates and Urea
Pesticides
Chlorinated Acids
Chlorinated Hydrocarbons
Chlorinated Pesticides
1 ,2-Oibromoethane and
1 ,2-Dibromo-3-Chloropropane
Dithiocarbamate Pesticide*
Extractable Organics
Haloethers
Nitroaromatics and Isophorone
Nitrogen and Phosphorous
Containing Pesticide
^itrosamines
N-Methylcarbamates and
N-Methylcarbamoyloxime*
Organohalide Pesticides and
PCBs
Organophosphate Pesticides
Organophosphate Pesticides
Perchlorination Screening of
PCBs
Pesticide and PCS*

Pesticides and PCBs
Organochlorlne
Phenols
Phthalate Ester*
Purgeable Aromatics
603
625'
605
632
515.1
612
508
504
630
525'
611
609
507
607
531.1 '
617
614
622
S08A
505'

60S*

604
606
602*
Analytical Sy$ter^
GC-FID
GC-MS
HPLC/Electrochem
HPLCAJV
ECO
Capillary Column
GC-ECD
ECO
Capillary Column
GC-ECD
Jolorimetric
GC-MS
Capillary Column
GC-ELCD
GC-FID * ECO
NPO
Capillary Column
GC-NPD
HPLC
Fluorescence Detector
GC-ECD
GC-FPDorNPD
GC-FPD
ECD/ELCD Packed or
Capillary Column
GC-ECD
Capillary Column

GC-ECD

GC-FID
GC-ECD
GC-PID
Sample
Introduction/
Preparation
P&T
XTN
XTN
XTN
XTN
XTN
XTN
XTN
CS2 Liberation
XTN
XTN
XTN
XTN
XTN
Dl
XTN
XTN
XTN
XTN
XTN

XTN

XTN
XTN
P&T
Detection Limit/
Range foob^
0.5-0.6
0.09-44.0
0.08-0.13
0.003-11.1
EOL, 0.1-1.0
0.03-1.34
EDU 0.01 -0.5 (most
0.01
1.9-15.3
0.1-1.0
0.3-3.9
0.01-15.7
EOL (Estimated D.L)
0.1-5.0 (most <1.0)
0.15-0.81
0.5-4.0
0.002-0.176
0.012-0.015
0.1-5.0
0.1-0.3
Variable
Pesticide 0.005-1.0
Herbicide 02-7.0
PCBs 0.1-0.5
0.002-024

0.14-16.0
029-3.0
0.2-0.4
• Frequently requested method
                          228

-------
                       APPENDIX III

                        Table V-A
SUMMAHY OF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
              ORGANIC COMPOUNDS (continued)
Industrial and Municipal Waste Water (USEPA, Office of Research and Development)
Compound Class
Purgeable Halocarbons
Purgeable Organics
Purgeable Organics
Purgeables
Volatile Aromatics and
Unsaturated Compounds
Volatile Halocarbons
Volatile Halocarbons
2.3.7.8-Tetrachlorodibenzo-p
dioxin
Triazine Pesticides
Sample
EPA Introduction/
Method No. Analytical System Preparation
601* GC-ELCD P&T
524.1 GC-MS P&T
Capillary Column
524.2* GC-MS P&T
Capillary Column
624* GC-MS P&T
503.1 GC-PID P&T
502.1 GC-ECD P&T
Packed Column
502.2* GC-ELCD/PID P&T
Capillary Column
613 GC-MS XTN
619 GC-NPD XTN
Detection Limit/
Range (ppb)
0.02-1.81
0.1-1.0
0.02-0.2
1.6-7.2
0.002-0.03
0.001-0.01
0.01-0.10
0.002
0.03-0.07
Aqueous and Solid Matrices (USEPA, Office of Water)
Sample
EPA Introduction/
Compound Class Method No. Analytical System Preparation
Semivolatile Organics 1625
Tetra- through octa- 1613
chlorinated dioxins
and furans
.Volatile Organics 1624
Isotope Dilution by XTN
GC-MS (Capillary
Column)
isotope Dilution by XTN
high resolution
GC-high resolution MS
Isotope Dilution by P&T
GC-MS (Capillary
Column)
Detection
Range (ppb)
most 20-1 00 ppb
(dependent on
% solids)
10- 100 parts per
quadrillion in water
1-10 parts per trillion
in soil
5-100 ppb
(dependent
on % solids)
* Frequently requested method.
                           229

-------
                       APPENDIX III

                        Table V- A
SUMMARY OF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
              ORGANIC COMPOUNDS (continued)
Solid Matrices (USEPA, Test Methods for Evaluating Solid
Waste, SW846, November, 1986.)
EPA
Compound ClaSS Method Nn AnnMiml Suetam
Acrolein, Acrylonitrite,
Acetonitrile
Aromatic Volatile Organics
Chlorinated Herbicides
Chlorinated Hydrocarbons
Nitroaromatics and Cyclic
Ketones
Organophosphorus Pesticides
Organochlorine Pesticides and
PCBs
Phenols
Phthalate Esters
Polynudear Aromatic
Hydrocarbons
-wtb a.
Polynudear Aromatic
Hydrocarbons
Purgeable Halogenated Volatile
Organics
Purgeable Non-Hatogenated
Volatile Organics
Semlvotatile Organics
Volatile Organics
8030
8020*
8150
8120
8090
8140
8080*
8040
8060
8100
8310
8010
8015
8270*
8240*
GC-FID
GC-FID
GC-ECDorELCD
GC-ECD
GC-FID or ECO
GC-FPDorNPD
GC-ECD
GC-FID
GC-ECD
GC-FID
HPLC/UV and Fluor
GC-ELCD
GC-FID
GC-MS
Capillary Column
GC-MS
Sample
Introduction/
Preparation
5030
5030
3550
3550
3550
3550
3550
3550
3550
3550
3550
5030
5030
3550
5030
Detection Limit/
Range (ppb)
0.5-0.6
0.2-0.4
0.1-200
0.03-1.3
0.06-5.0
0.1-5.0
70-1000
0.14-16
0.29-31
Not Reported
0.013-2.3
0.03-0.52
Not Reported
Not Reported
1.6-7.2
* Frequently requested method.
                         230

-------
                     APPENDIX III
                     TABLE V-B
SUMMARY OF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
                 INORGANIC ANALYTES

Analyta
Total/Dissolved Metals
Total/Dissolved Metals
Total/Dissolved Metals
Aluminum
Antimony
Antimony
Antimony
Barium
Barium
Beryllium
Beryllium
Boron
Calcium
Calcium
Cobalt
Cobalt
Copper
Copper
Cyanide

Cyanide


Cyanide



Cyanide,
Amenable to
Chtorinatton,
without
distillation
Cyanide

Gold

Gold

Iron
Iron
EPA
Sample Detection Limit
fjlethod No. Analytical System
1620
6010
7000
7020
204.2 CLP
7040
7041
7080
7081
7090
7091
212.3
215.2
7140
7200
7201
7210
7211
335.2

335.2


355.1



4500-CN-H
Standard Method
for the Examin-
ation of Water
and Wastewater
1989
335.3

231.1

231.2

7380
7381
ICP
ICP
AA
AA
GFAA
AA
GFAA
AA
GFAA
AA
GFAA
Spectrophotometric
T&rimetric
AA
AA
GFAA
AA
GFAA
Total. (Titrimetric,
Spectrophotometric)
Midi (Distillation,
Total, Cotorimetric.
Automated UV)
Amenable to
Chtori nation
(Titrimetric.
Spectrophotometric)
Spectrophotometric




Total, Spec-
trophoto-
metric
AA

GFAA

AA
GFAA
Preparation
3005,3010
3005,3010
3005.3010
3005,3010
*
3005,3010
3005,3010,3020
3005.3010
Nitric acid, reflux
3005,3010
3020
Hydrochloric add
*
3005,3010
3005-3010
3020
3005,3010
Nitric acid, reflux
*«*

•*•


**••



pH>12




***

Nitric acid, Aqua
Regla
Nitric acid, Aqua
Regla
3005,3010
Nitric add, reflux
Range (ppb)



1,000
4300-5700

70
20
30
2.0
50-200
1.0-30
200
100,000
4800-5200
3400-4600
50
3700-4300
1.0
10

5.0


10



20




10

100

1.0

4400-5600
1.0
                          231

-------
                     APPENDIX III
                      TABLE V-B
SUMMARY CF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
                  INORGANIC ANALYTES
                       (continued)

Analyte
Iridium

Irkjium

Magnesium
Manganese
Manganese
Molybdenum
Molybdenum
Molybdenum
Molybdenum
Nickel
Osmium

Osmium
Osmium
Palladium
Palladium
Platinum
Platinum
Potassium
Rhenium
Rhenium
Rhodium

Rhodium
Ruthenium
Ruthenium
Selenium
Selenium
Selenium
Silver
Silver
Sodium
Thallium
Thallium
Tin
Tin
Titanium
Titanium
Vanadium
Vanadium
Zinc .
Zinc
EPA
Method No.
235.1

235.2

7450
7460
7461
246.1
246.2
7480
7481
7520
252.1

252.2
7550
253.1
253.2
255.1
255.2
7610
264.1
264.2
265.1

265.2
267.1
267.2
270.3
7740
7741
7760
7761
7770
7840
7841
282.1
282.2
283.1
283.2
7910
7911
7950
7951
Sample Detection Limit/
Analytical System
AA

GFAA

AA
AA
GFAA
AA
GFAA
AA
GFAA
AA
AA

GFAA
AA
AA
GFAA
AA
GFAA
AA
AA
GFAA
AA

GFAA
AA
GFAA
AA-Hydride
GFAA
AA Hydride
AA
GFAA
AA
AA
GFAA
AA
GFAA
AA
GFAA
AA
GFAA
AA
GFAA
Preparation
Nitric acid, Aqua
Regia
Nitric acid, Aqua
Regia
3005,3010
3005,3010
Nitric acid, reflux
*
*
3005,3010
3020
3005,3010
Nitrtc.sulfuric
acids
Nitric acid
3005,3010
Nitric acid
Nitric acid
**
**
3005,3010
Nitric acid
Nitric acid
Nitric acid
Regia
Nitric acid
Hydrochloric acid
Hydrochloric acid
*•
3020
3005,3010
3005,3010
Nitric acid, reflux
3005,3010
3005,3010
3020
••
**
**
**
3005,3010
3020
3005,3010
Nitric acid, reflux
Range (ppb)
3000

30

970-1030
10
0.2
100
1.0
10,000
-
4900-5100
300

20
-
100
5.0
1000
20
1000-2200
5000
200
50

5.0
200
20

3.0-5.0
5.0
1200-2800
0.2
4800-5200

1.0-10
800
5.0
400
10
49400-50600
50
5.0
0.05
                            232

-------
                                           APPENDIX III
                                            TABLE V-B
               SUMMARY OF ROUTINE METHODS BY PROGRAM AND COMPOUND CLASS
                                        INORGANIC ANALYTES
                                             (continued)
Sample Preparation Methods
3005   Acid Digestion of Waters for Total Recoverable Dissolved Metals for Analysis by Flame Atomic Absorption
       Spectroscopy or Inductively Coupled Plasma Spectroscopy.
3010   Acid Digestion of Aqueous Samples and Extracts for Total Metals for Analysis by Flame Atomic Absorption
       Spectroscopy or Inductively Coupled Plasma Spectroscopy.
3020   Acid Digestion of Aqueous Samples and Extracts for Total Metals for Analysis by Furnace Atomic
       Absorption Spectroscopy.
       CLP preparation methods are categorized by water/soil, ICP, AA, .and GFAA instrumentation.
     • CLP methods are based on the 200 series Methods for Chemical Analysis of Water and Wastes. U.S.
       Environmental Monitoring Systems Laboratory. .Cincinnati, Ohio. March. 1983.
     • Water sample preparation for GFAA uses nitric acid, hydrogen peroxide and mild heat. SOW 788,0-5.
     • Water sample preparation.for ICP and AA uses nitric acid, hydrochloric acid and mild heat. SOW 788,0-5.
     • So!) sample preparation for ICP, AA, GFAA uses nitric acid, hydrogen peroxide and mild heat.
     • Hydrochloric add Is used as the final reflux acid for several analytes.  SOW 788, D-5.6.
       Nitric and hydrochloric acids are used for digestion.
  '     Total cyanide is determined by a reflux-distillation procedure using a sodium hydroxide scrubber.
  "    Cyanide amenable to chtorination is chlorinated at pH greater than 11.
                                                   233

-------
                                          APPENDIX IV
                  CALCULATION FORMULAS FOR STATISTICAL EVALUATION


       Appendix IV provides calculation formulas to enable responsible risk assessment personnel to determine the
minimum number of samples necessary to meet statistical performance objectives. This appendix also provides
statistical guidelines on the probability that a given sampling plan will identify a hot spot, and the probability that no
hot spot exists given none was found after sampling.
                       Calculation Formulas to Determine the Number of
                     Samples Required Given Coefficient of Variation and
                               Statistical Performance Objectives
   The minimum number of samples, n. required to achieve a specified precision and confidence level at a
   defined minimum detectable relative difference may be estimated by the following equation:


   For one-sided, one-sample t-test

   For one-sided, two-sample t-test

   where: Z. is a percentile of the standard normal distribution such that P(Z i 2L) - a, Z, is slmflarfy defined,
   and D • MDRD/CV. where MDRD is the minimum detectable relative difference and CV is the coefficient of
   variation.  NOTE: Data must be transformed (ZJ, for example:
                Confidence Level
                1-o    o      2
                0.80
                0.85
                0.90
                0.95
                0.99
0.20
0.15
0.10
0.05
0.01
0.842
1.039
1.282
1.645
2.326
     Power
l-fl    p     Z,

0.80   2.00   0.842
0.85   0.15   1.039
0.90   0.10   1.282
0.95   0.05   1.645
0.99   0.01   2.326
   As an example of applying the equation above, assume CV - 30%, Confidence Level - 80%, Power. 95%,
   and Minimum Detectable Relative Difference » 20%. For Infinite degrees of freedom (t distribution becomes a
   normal one), Z.- 0.842 and Z,«1.645. From the data assumed. D - 20% /30%. Therefore,

          n 21(0.842 + 1.645)/(20/30)P 1- 0.5 (0.842)*

          n 213.917 * 0.354 -14.269

          n 215 samples required (round up)
   Source: Adapted from EPA 1989C.
                                                 235

-------
                                           APPENDIX IV
                                            (continued)

                                Calculation Formulas For Tht
                                 Statistical Evaluation Of The
                                    Detection Of Hot Spots
Hot Spot Will Be Identified: Example * 1

These formulas are useful In evaluating the probability that a particular sampling plan wHI tientty • hot spot.
Let R represent the radius of a hot spot and D be the distance between adjacent grid points where sample*
will be collected. The probability that a grid point will fall on a hot  spot to easily obtained from • geometrical
argument since at least one grid point must fall in any square of area D 2 centered at the center of the hot
spot. From this concept, It follows that the probability of sampling a hot spot P(H/E) Is given by.

                                                                         URiD/2

                               • 2 arc cos (D/(2R))) + (D/4)V(4R2. D^J/D2      | on < R < Of VJT
IRiD/
                                                                                VT
 where the angle D/(2R) Is expressed In radian measure, H Is the case that a hot spot is found, and E Is the
 case that a hot spot exists.

 An example Is H the grid spacing Is D - 2R , then the probability of a hit Is *I4 • 0.785, which
 Implies that the probability that this grid spacing would not hit a hot spot If ft exists is 0.21 5.

 No Hot Spot Exists: Example f 2

 This set of formulas addresses the probability that no hot spot exists (given that none were found). This
 argument requires the use of a subjective probability, P(E) (where P(E) is the probability that • not spot
 exists), based on historical and perhaps geophysical evidence Then, if E is the case that there are no hot
 spots at the study site and If H is the case that no hot spot is found In the sample, Bayes formula gives:


                P(E I H) • P(H I E) P(E) / (P(H I E) P(E) * P(H I E) P(E))

                              - P(H I E) P(E) / (P(H I E) P(E) +P(E)).

  For the case where D - 2R, It was found from Example 1 that P(HIE) -0.21 5. Therefore, V one Is given that
  the chance P(E) of a hot spot is thought to be 0.25 prior to the Investigation, the probability of a hot spot
  existing if the study does not find one is:

                  P(E I no hit) - 0.21 5 (0.25) / [0.21 5 (0.25) + 0.75] - 0.067.

  Hence, the probability that  no hot spot exists Is (1-0.067) • 0.933.
  Source: Adapted from EPA 1989c.
                                                   236

-------
r
                          Appendix IV (continued)
Number of Samples Required in a One-Sided One-Sample t-Test to Achieve a Mini-
 mum Detectable Relative Difference at Confidence Level (1-a) and Power of (1-p)
Coefficient
of Variation
(%)
10
15
20
Power
(%)
95



90



80



95


90
80
95
90
80
Confidence
Level
(%)
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
80
Minimum Detectable
Relative Difference (%)
5
66
45
36
26
55
36
28
19
43
27
19
12
145
99
78
57
120
79
60
41
94
58
42
26
256
175
138
100
211
139
107
73
164
101
73
46
10
19
13
10
7
16
10
8
5
13
8
6
4
39
26
21
15
32
21
16
11
26
16
11
7
66
45
36
26
55
36
28
19
43
27
19
12
20
7
5
3
2
6
4
-3
2
6
3
2
2
12
8
6
4
11
7
5
3
9
5
4
2
19
13
10
7
16
10
8
5
13
6
e
4
30
5
3
2
2
5
3
2
1
4
3
2
1
7
5
3
2
6
4
3
2
6
3
2
2
10
9
5
4
9
6
4
3
8
5
3
2
40
4
3
2
1
4
2
2
1
4
2
2
1
5
3
3
2
5
3
2
1
5
3
2
1
7
5
3
2
6
4
3
2
6
3
2
2
              Source: EPA1989C
                                               237

-------
                          Appendix IV (continued)
Number of Samples Required In a One-Sided One-Sample t-Test to Achieve a Mini-
 mum Detectable Relative Difference at Confidence Level (1-a) and Power of (1-p)
                                (continued)
Coefficient
of Variation
(%)



35
Power
(%)
95
90



60

95
90
80
95
90
60
Confidence
Level
(%)
99
95
90
80
99
95
90
80
99
95
90
80
99
95
90
60
99
95
90
80
99
95
90
80
99
95~
90
60
99
95
90
80
99
95
90
60
Minimum Detectable Relative Difference
{%)
5
397
272
216
155
329
272
166
114
254
156
114
72
571
391
310
223
472
310
238
163
364
224
164
103
775
532
421
304
641
421
323
222
495
305
222
140
10
102
69
55
40
65
70
42
29
66
41
30
19
145
99
76
57
120
79
61
41
84
58
42
26
196
134
106
77
163
107
82
56
126
78
57
36
20
28
19
15
11
24
19
12
8
19
12
8
5
39
26
21
15
32
21
16
11
26
16
11
7
42
3* '
28
20
43
28
21
• 15
34
21
15
10
30
14
9
7
5
12
9
6
4
10
6
4
3
19
13
10
i
16
10
8
5
13
8
6
4
25
\i
13
9
21
14
10
7
17
10
7
5
40
9
6
5
3
8
6
4
3
7
4
3
2
12
8
6
4
11
7
5
3
9
5
4
2
15
10
8
6
13
8
6
4
11
7
5
3
Sourer EPA 1B89C
                                    238

-------
                                       APPENDIX V
                        DATA QUALIFIER SOURCE AND MEANING1
       Appendix V lists the parameters and criteria that produce a "J"  flag in accordance with the
National Functional Guidelines for Organic Data Review (EPA 1991e) and Laboratory Data Validation
Functional  Guidelines for Inorganics Analyses (EPA 1988e) as applied to data  from the Contract
Laboratory  Program.  The appendix also indicates the likely implication of this flag on the  associated
result(s).

       The criteria listed in this guidance  should be used to flag CLP data as  "J," or "estimated
concentration" (the associated numerical value is an estimate of the amount actually present  in the
sample).  With proper interpretation, the results of analytes which are flagged "J" can often be used in
making decisions.

       Data flagged  with "UJ" indicates that the value is undetected and quantitation limit may be
imprecise.  Data flagged with "NJ" indicates that the value is tentatively identified and confirmation is
needed in future sampling efforts.
 PARAMETER    CRITERIA
                       ACTION
                               LIKELY
                               IMPLICATION1
 ANALYSIS:  Organic (3/90) VOA & BNA
 Holding times      14 < VOA < 30 days   Associated samples
                   7 <  BNA <  22 days    (+ results)
                                                      Low
 Mass Calibration
 Ion Abundance
Several data elements
in expanded window
All associated data
No generalization


Precision
 Calibrations

 - initial


 - continuing
Average RRF <  .05
%RSD >  30%

RRF < .05

%D between initial
and continuing
calibration >  25%
Compound specific (+ results)     Low
Compound specific (+ results)

Compound specific (+ results)     Precision

Compound specific (+ results)
 Blanks
If associated result is
between detection limit
and CRQL
Compound specific
High
                                              239

-------
                              APPENDIX V (CONTINUED)
PARAMETER     CRITERIA
                      ACTION
                                LIKELY
                                IMPLICATION*
Surrogates
If surrogate            Fraction specific (+ results)
recoveries are low but   (negative results are flagged
> 10%               w/sample quantitation limit as
                      estimated (UJ))
                                Low
                  Any surrogate in a
                  fraction shows
                  < 10% recovery

                  If surrogate
                  recoveries are high
                      Fraction specific (+ results)
                      Fraction specific (+ results)
Internal standards  If an IS area count is    Associated compounds
                  outside -50% or        (+ results) (non-detects flagged
                  +100% of the         w/sample quantitation limit - UJ)
                  associated standard
                                Low



                                High



                                No generalization
TICs
None
All TIC results - (NJ)
No generalization
 ANALYSIS: Pesticides (2/88)
 Holding Times      7 < PEST < 22
                   days
 Instrument
 Performance
 DDT breakdown
 > 20%
                   Endrin breakdown
                   >20%
                      Associated positive results
                      (negative results - UJ)
                                Low
Associated positive DDT           Low
results (J)
Results for DDD and/or
DDE (NJ)

Associated positive Endrin results   Low
(J); Results for Endrin Ketone (J)
                                            240

-------
                              APPENDIX V (CONTINUED)
PARAMETER     CRITERIA
                                       ACTION
Calibrations

- initial

~ continuing
                 If criteria for linearity   Associated positive results
                 not met
Surrogates
Compound
Quanritation and
Detection Limits
                  If %D between
                  calibration factors
                  > 15% (20% for
                  compounds being
                  confirmed)
                  If low surrogate
                  recoveries obtained
                                       Associated positive results
Associated results
                  Quantltation limits      Estimated quantitation limit (UJ)
                  affected by large, off-
                  scale peaks
ANALYSIS:  Inorganic (3/90)
Holding Times/
Preservation
Calibrations
                  Exceeded
Associated samples >  IDL
[ IDL
                   < 0.995               [
120%
                                       Associated samples
                                       Associated samples >  IDL
                                       Associated samples > IDL
                               LIKELY
                               IMPLICATION*
                               No generalization


                               No generalization
                                                    Low
                                No generalization
Low


No generalization


Precision


Low/High
                                High
                                           241

-------
                             APPENDIX V (CONTINUED)
PARAMETER     CRITERIA
                    ACTION
                               LIKELY
                               IMPLICATION*
                  If 1CS recovery falls
                  between 50-79%
                    Associated samples > IDL
                    [ IDL
                    1  2xCRDL
                  and 10% reported
                  concentration of the
                  affected element
                     Associated samples
                               High
 LCS (Aqueous)
Recovery within
range 50-79% or
> 120%
Associated samples > IDL
1  IDL
Low/High
                   Recovery lower than   Associated samples (< IDL (UJ)J    Low
                   control limits
 Duplicate
Outside control limits   Associated samples of same
                     matrix > IDL
                                Precision
 Matrix Spike
 Sample
  AA Post
  Digestion Spike
 Recovery >  125% or   Associated samples > IDL
 < 75%
 Recovery within
 range 30-74%

 Duplicate Injection
 outside 4- 20%
 RSD (or CV) and
 sample not rerun once
                                Low/High


 Associated samples [ < IDL (UJ)]    Low
 Associated data > IDL
 Precision
                                            242

-------
PARAMETER     CRITERIA
                              APPENDIX V (CONTINUED)
                    ACTION
                              LIKELY
                              IMPLICATION*
                  Rerun sample does
                  not agree within
                  + 20% RSD (CV)
                    Associated data >  IDL
                              Precision
                  Post digestion spike
                  recovery < 40%
                  even after rerun
                    Associated data >  IDL
                              Low
                  Post digestion spike     Associated data [ < IDL (UJ)]
                  recovery >  115% or
                  <  85%
                                                    High/Low
                  If sample absorbance   Associated samples > IDL
                  is < 50% of post      [ IDL
                     Associated data > IDL
                               Precision


                               No generalization
                  MSA correlation
                  coefficient < 0.995
                     Associated data > IDL
                               No generalization
 ICP Serial
 Dilution
Criteria not met
Associated data > IDL
Precision
                                           243

-------
                              APPENDIX V (CONTINUED)
1   Selected Acronym Key



   BNA   -     Base/neutral/acid or semivolatile



   CRDL -     Contract required detection limit (inorganics)



   CRQL --     Contract required quantitation limit (organics)



   CV    ~     Coefficient of variation



   ICS    -     Interference check sample



   ICV    -     Initial  calibration verification



   IDL    -     Instrument detection limit



   IS     -     Internal standard



   PEST  --     Pesticide



   RRF   -     Relative response factor



   RSD   -     Relative standard deviation



   TIC    -    Tentatively identified compound



   VGA  -    Volatile
 2  Implication Key



   Low:  The associated result may underestimate the true value.



   High: The associated result may overestimate the true value.



   Precision:  The associated result may be of poor precision (high variability).



   No generalization:  No generalization can be made as to the likely implication.
                                             244

-------
                                      APPENDIX VI
                    "R" DATA QUALIFIER SOURCE AND MEANING1
       Appendix VI lists the parameters and criteria that produce an "R" flag in accordance with the
National Functional Guidelines for Organic Data Review (EPA 1991e) and Laboratory Data Validation
Functional Guidelines for Inorganics Analyses (EPA  1988e) as applied to data from the Contract
Laboratory Program.  The appendix also indicates the likely implication of this flag on the associated
result(s).

       The criteria listed in this guidance should be used to flag CLP data as "R," or "unuseable." If
the flagged analytes are of interest, then resampling or reanalysis is necessary.
 PARAMETER
CRITERIA
ACTION
LIKELY
IMPLICATIONS1
 ANALYSIS: Organic (3/90) VOA & BNA
 Holding times
Grossly exceeded
Professional judgment   Low
(non-detects)
  Mass Calibration
In error
Associated samples      Unuseable
  Ion Abundance
Outside expanded
windows
Associated samples     Unuseable
  Calibrations
Mean RRF or
RRF < 0.05
Compound specific
(non-detects)
Low
  Blanks
Gross contamination     Compound specific      High
(saturated peaks)    .   (associated samples)
  Surrogates
 <  10% Recovery
Entire fraction
(negative results)
 Low
  Internal Standards
 Extremely low area
 counts; Major abrupt
 drop off
 Associated compounds   Low
 (non-detects)
  TICs
 Suspected artifacts      Professional judgment   Unuseable
                                             245

-------
                           APPENDIX VI (CONTINUED)
PARAMETER
CRITERIA
ACTION
LIKELY
IMPLICATION1
ANALYSIS:  Pesticides (2/88)
Holding Times
Grossly exceeded
Professional judgment
(non-detects)
Low
Instrument
Performance
       DDT
       Retention
       Time
Inadequate separation    Affected compounds
                        Unuseable
       RT
Peaks of concern
outside windows
Professional judgment
(positive results and
quantitation limits)
Unuseable
       DDT/Endrin   Not detected and
       Degradation   breakdown
                     concentrations
                     positive
                      Samples following last
                      in-control standard
                      (quantitation limit • DDT
                      and Endrin)
                        Low
       Retention
       Time Check
DBC > 2.0%
(packed)
 > 0.3% (narrow-
bore)
 > 1.5% (wide-bore)
Professional judgment     Unuseable
 Surrogates
 Not present
Suggested (negative
results)
Low
 Compound
 Quantitation and
 Detection Limits
 Large off-scale peaks   Quantitation limits
                         Unuseable
                                            246

-------
                                          APPENDIX VI (CONTINUED)
                PARAMETER
CRITERIA
                ANALYSIS: Inorganic (3/90)
                Holding Times
                Calibrations
                - ICV or CCV
                ICS (for ICP)
                LCS (Aqueous)
Grossly exceeded
Minimum number of
standards not used;
Not calibrated daily
or each time
instrument set up


%R outside of 75-
125% (CN, 70-130;
Hg, 65- 135%)


Al, Ca, Fe or Mg in
samples :
50% of levels found
in ICS, and estimated
interferences due to
Al, Ca, Fe or Mg
> 90%


Recovery < 50%
                 Matrix Spike Sample   Recovery < 30%
                 AA Post Digestion     Recovery < 10%
                 Spike
ACTION
Professional judgment
(Results < IDL)


Professional judgment
(associated samples)
Associated samples




Affected analytes



Affected analytes
Affected analytes
LIKELY
IMPLICATION*
Low
Precision
Low/High




High



High
Low
                     Affected samples (results   Low
                     < IDL)

                     Affected samples (results   Low
                     < IDL)
fr.
                                                         247

-------
                              APPENDIX VI (CONTINUED)
1  Selected Acronym Key
  AA   ~   Atomic absorption
  BNA -   Base/neutral/acid or semivoiatile
  CCV -   Continuing calibration verification
  DBC -   Dibutyl chlorendate
  ICP  -   Inductively coupled plasma
  ICS  -   Interference check sample *
  ICV  -   Initial calibration verification
  IDL  -   Instrument detection limit
   LCS -   Laboratory control sample
   RRF -   Relative  response factor
   RT   -   Retention time
   TIC  -   Tentatively identified compound
   VGA -    Volatile
 1  Implication Key
   Low: The associated result may underestimate the true value.
   High:  The associated result may overestimate the true value.
   Precision:  The associated result may be of poor precision (high variability).
   No generalization: No generalization can be made as to the likely implication.
   .Unuseable:  Data are probably unuseable without resampling and  reanalysls.
                                             248

-------
                                      APPENDIX VII
      SUMMARY OF COMMON LABORATORY CONTAMINANTS, CONCENTRATION
                REQUIREMENTS, AND RISK ASSESSMENT IMPLICATIONS

     Appendix VII lists common organic laboratory contaminants that may appear in blanks.
The purpose of this appendix is to inform the reader of chemicals that may appear in analyses
but may not be present at the site.  Analytes with values above instrument detection  limits are
reported by laboratories. Some sample concentrations may not be reported through the review
process, as explained below, but if they are reported, possibilities of false positives exist. The
implications for risk assessment are included.
 Common Laboratory
   Contaminants
Targtt Compound

Methylene Chloride
 Acetone
 Toluene
 Concentration Requirements
Risk Assessment
Implications
Sample concentrations less than
I Ox that detected in method
blanks will be reported as
undetected (or flagged B).
Sample concentrations less than
lOx that detected in method
blanks will be reported as
undetected (or flagged B).
 Sample concentrations less than
 lOx that detected in method
 blanks will be reported as
 undetected (or flagged B).
  Include analyte if
  concentration is greater
  than  I Ox blank.

  Include analyte if
  concentration is less than
  lOx greater than  blank
  concentration and multiple
  chlorinated volatile analytes
  are detected.
  Exclude analyte in all other
  situations.

  Include analyte if
  concentration is greater
  than lOx blank.

  Include analyte if
  concentration is less than
  lOx greater than blank
  concentration and multiple
  ketones are detected.

  Exclude analyte in all other
  situations.

  Include analyte if
  concentration is greater
  than I Ox blank.

  Include analyte if
  concentration is  less than
  lOx blank concentration
  and multiple aromatic or
  fuel hydrocarbons are
  detected.

  Exclude analyte  in all other
  situations.
                                              249

-------
                                 APPENDIX VII (CONTINUED)
 Common Laboratory
   Contaminants
 Concentration Requirements
 Risk Assessment
 Implications
2-Butanone (methyl
ethylketone)
Sample concentrations less than
lOx that detected in method
blanks will be reported as
undetected (or flagged B).
Phthalates (i.e., dimethyl
phthalate, diethyl
phthalate, di-n-butyl
phthalate, butylbenzyl
phthalate, bis(2-
ethylhexyl) phthalate, di-
n-octyl phthalate)

Tentatively Identified
     Compound*

Carbon dioxide
 Diethyl ether
Sample concentrations less than
I Ox that detected in method
blanks will be reported as
undetected (or flagged B).
Not reported if present in the
method blank.

Not reported if present in the
method blank.
 Hexanes
 Not reported if present in the
 method blank.
   Include analyte if
   concentration is greater
   than lOx blank.

   Include analyte if
   concentration is less than
   lOx blank concentration
   and multiple ketones are
   detected.

   Include analyte if
   concentration is greater
   than lOx blank.

   Exclude analyte in all other
   situations.
o  Exclude analyte in all
   situations.

o  Include analyte if
   concentration is greater
   than lOx blank.

o  Exclude analyte in all other
   situations.

o  Exclude if analyte
   concentration is not lOx
   method blank.

o  Exclude if analyte
   concentration is not lOx
   field blank (EPA
   definition).

o  Exclude if sample is not
   analyzed within seven days.
                                               250

-------
                                 APPENDIX VII (CONTINUED)
 Common Laboratory
   Contaminants
 Concentration Requirements
Risk Assessment
Implications
Freons (e.g., 1.1,2-
trichloro-1,2,2-
trifluoroethane, fluorotri-
chloromethane)
Not reported if present in the
method blank.
Solvent preservative
artifacts (e.g.,
cyclohexanone,
cyclohexenone,
cyclohexanol,
cyclohexenol,
chlorocyclohexene,
chlorocyclohexanol)
 Aldol reaction products of
 acetone (e.f., 4-hydroxy-
 4-methyl-2-pentanone, 4-
 methyl-penten-2-one,
 5,5-dimeihyl-2(5H)-
 furanone)
Not reported if present in the
method blank.
 Not reported if present in the
 method blank.
  Exclude if analyte
  concentration is not I Ox
  method blank.

  Exclude if analyte
  concentration is not I Ox
  field blank (EPA
  definition).

  Exclude if sample is not
  analyzed within seven days.

  Exclude if artifact
  concentration is not lOx
  method  blank.

  Exclude if artifact
  concentration is not lOx
  field blank (EPA
  definition).

  Exclude if sample is not
  analyzed within seven days.

  Include  analyte if
  concentration is greater
  than lOx blank.

  Include analyte if
  concentration is less than
   lOx greater than  blank
  concentration and multiple
   ketones are detected.

   Exclude analyte in  all other
   situations.
                                                231

-------
                                   APPENDIX VIII
                         CLP METHODS SHORT SHEETS

TITLE:        USEPA CONTRACT LABORATORY PROGRAM
              STATEMENT OF WORK FOR ORGANIC ANALYSIS
              MULTI-MEDIA, MULTI CONCENTRATION
DOCUMENT NUMBER:
DOCUMENT DATE:
EFFECTIVE DATES:
CONCENTRATION:
DATA TURNAROUND:
MATRICES:
OLM01.0
Not Applicable
September 28, 1990 through February 1994
Low to Medium
'14 Days or 35 Days
Aqueous/Soil/Scdimcnt*
SIGNIFICANT FEATURES
•   The compounds include volatiles, scmivolaiilcs, and pcsiiddc/PCBs.
•   Volatiles and semivolatiles are analyzed by GC/MS; pestlddes/PCBs are analyzed by GC/ECD.
•   Major Tentatively Identified Compounds (TICs) are reported for GC/MS analyses.
•   Second column confirmation by GC/ECD is required for all pesticides/PCBs. Pesticides/PCBs which
    are identified by GC/ECD at concentrations above 10 ng/uL art confirmed by GC/MS analysis.

REVISIONS/MODIFICATIONS
        The following is a list of the significant changes from the 2/88 SOW that are incorporated in the
OLM01.0SOW:

•   Selected volatile CRQLs have been raised; pesticide/PCB low soil CRQLs have been lowered; and
    selected pesticide/PCB aqueous CRQLs have been changed.
•   Target Compound List (TCL) changes include the elimination of vinyl acetate from the volatile TCL,
    the elimination of benzyl alcohol and benzole acid from the semivolaiile TCL, the addition of
    carbazole to the semi volatile TCL, and the addition of endrin aldehyde to the pesticide TCL. The
    cemivolatile TCL compound bis(2-chloroisopropyl)ether was renamed 2,2'oxybis(l-chloropropane).
•   A new method for analysis of pesticides/PCBs is used. Changes include the use of wide bore capillary
    columns, new surrogates, and new calibration techniques.
•   Pesticide/PCB quantitation is performed using both the primary and secondary columns.  The lower
    value is reported by the laboratory.
        The only significant change in the OLM01.1  (December. 1990) and OLMOU.l (February. 1991)
revisions to the OLM01.1 through OLM01.0 SOW was the lowering of selected scmlvolatilc CRQLs. The
significant changes in the OLM01.* through OLM01.7 revisions to the OLM01.0 SOW were the lowering
of (elected semivolaiile CRQLs and options for either a 14 day or 35 day data turnaround.

RECOMMENDED USES
        This Routine Analytical Services (RAS) method is recommended for broad spectrum analysis to
define the nature and extent of potential site contamination during SSI, LSI, and RI/FS activities. This
method Is suitable when a 14 day or 35 day turnaround for results is adequate. It is recommended for
samples from known or suspected hazardous waste sites where potential contamination may be present it
significant risk levels.

        • Sediment samples with high moisture content should be solicited as RAS + SAS (Special
Analytical Service) in order to achieve the CRQLs.

COMPOUNDS AND CRQLs
        The Target Compound List compounds Included in the analysis and their Contract Required
QuantluUon Limits (CRQLs) ire listed In Attachment 1.
                                           253

-------
TITLE:         USEPA CONTRACT LABORATORY PROGRAM
               STATEMENT OF WORK FOR ORGANIC ANALYSIS
               MULTI-MEDIA, HIGH CONCENTRATION
DOCUMENT NUMBER:
DOCUMENT DATE:
EFFECTIVE DATES:
CONCENTRATION:
DATA TURNAROUND:
MATRICES:
Not Applicable
September 1988
June 7. 1989 through December 26, 1991
High: Greater than 20 ppm
35 Days
Uquid/SoIid/MuUi-phase
 SIGNIFICANT FEATURES
 •   No holding time* are designated for high concentration samples.

 •   The analyses are suitable for highly contaminated samples (>20 mg/Kg).

 •   The analyses arc acceptable Tor liquid, solid, or multiphase samples. Multi-phase samples are
    separated into water misciblc liquid, water immiscible liquid, or solid phases. Each phase is analyzed
    separately.

 •   Volatile, cxtractablc (scmivolatilcs and pesticides), and mullicomponcnt cxtractable (Aroclors and
    Toxaphcnc) compounds arc included.

 •   Volatiles and extractablcs are analyzed by GC/MS; Aroclors and Toxapbene are analyzed by GC/ECD.

 •   Second column confirmation by GC/ECD is required for Aroclors and Toxaphene.

 •   Major Tentatively Identified Compounds (TICs) are  reported for GC/MS analyses.

 REVISIONS/MODIFICATIONS
         The 1/89 and 4/89 revisions to the 9/88 SOW do not significantly affect data uscability.

 RECOMMENDED USES
         This Routine Analytical Services (RAS) method U recommended for pre-rcmedlal. remedial, or
 removal projects where high concentrations of organic contaminants (greater than 20 mg/Kg) are suspected
 and a 35 day turnaround for results is adequate.  It U recommended for samples obtained from drummed
 material, waste pits or lagoons, waste piles, tanker trucks, onsite tanks, and apparent contaminated soil
 areas. The waste material may be industrial process waste, byproducts, raw materials, intermediates and
 contaminated products. Samples may be spent oil, spent solvents, paint wastes, metal treatment wastes,
 and polymer formulations.

         The method U suitable for solids, liquids, or multiphase samples, a phase being either water
 mlsciblc liquid, water immiscible liquid, or solid.  Various methods of phase separation may be utilized
 depending on (he number and types of phases in a sample.

 COMPOUNDS AND CRQU
         The Target Compound List compounds included In the analysis and their Contract Required
 Quantilallon Limits (CRQLs) are listed In Attachment I.
                                             254

-------
TITLE:        USEPA CONTRACT LABORATORY PROGRAM
              STATEMENT OF WORK FOR INORGANIC ANALYSIS
              MULTI-MEDIA, MULTI CONCENTRATION
DOCUMENT NUMBERt
DOCUMENT DATE:
EFFECTIVE DATES:
CONCENTRATION:
DATA TURNAROUND:
MATRICES:
IIA10I.O
Not Applicable
September 7, 1990 through September 26, 1993
Low to Medium
35 Days
Aqucous/Soil/Scdimcnt*
SIGNIFICANT FEATURES
•   The analyses arc suitable for aqueous, soil, or sediment samples at low to medium concentration levels.

•   This Statement of Work includes the midi distillation Tor cyanide analysis and the microwave digestion
    for GFAA and ICP analyses. These two sample preparation procedures require less sample volume
    than the traditional Statement of Work sample preparation procedures.
 REVISIONS/MODIFICATIONS


        None to date

 RECOMMENDED USES


        This Routine Analytical Service (RAS) method is recommended for broad spectrum analysis to
 define the nature and extent of potential site contamination during SSI, LSI, and Rl/FS activities. This
 method is suitable when a 35 day turnaround for results is adequate. It is recommended for samples from
 known or suspected hazardous waste sites when: potential contamination may be present at significant risk
 levels.

         • Sediment samples with high moisture content should be solicited as RAS + SAS (Special
 Analytical Service) in order to achieve the CRQ1J.
 ANALYTES ANDCRQLs
        The Target Analytc List analylcs included in the analysis and their Contract Required
 Quantitation Limits (CRQLs) are listed in Attachment 2.
                                           255

-------
TITLE-         USEPA CONTRACT LABORATORY PROGRAM
               STATEMENT OF WORK FOR INORGANIC ANALYSIS
               MULTI-MEDIA, HIGH CONCENTRATION
DOCUMENT NUMBER:
DOCUMENT DATE:
EFFECTIVE DATES:
CONCENTRATION:
DATA TURNAROUND:
MATRICES:
IHCOI.2
Not Applicable
May 15, 1991 through November 30, 1993
High
35 Days
Liquid/SoHd/Multi-phase
SIGNIFICANT: FEATURES
•   The analyses are suitable for highly contaminated samples.
•   The analyses are acceptable for liquid, solid, or multi-phase samples. Multi-phase samples are
    separated into water miscible liquid, water immiscible liquid, or solid phases. Each phase is analyzed
    separately.
•   The analyses include conductivity and pi I; potassium is not included.

REVISIONS/MODIFICATIONS


        The IHC01.1 and IHC01.2 revisions to the IHC01.0 SOW do not significantly affect data
useability.

RECOMMENDED USES


        This routine Analytical Service (RAS) method is recommended for prc-rcmedial. remedial, or
removal projects where high concentrations of inorganic contaminants are suspected and a 35 day
turnaround for results is adequate. It is recommended for samples obtained from drummed material, waste
pits or lagoons, waste piles, tanker trucks, onsite tanks, and apparent contaminated soil areas. The waste
material may be industrial process waste, byproducts, raw materials, intermediates, and contaminated
products.  Samples may be spent oil. spent solvents, paint wastes, metal treatment wastes, and polymer
 formulations.

        The method is suitable for solids, liquids, or multiphase samples, a phase being either water
 miscible liquid, water immiscible liquid, or solid. A phase separation step is applied prior to digestion.
 Each phase is analyzed and reported as a separate sample.

 ANALYTES AND CRQLs


        The Target Analyte List analytcs included in the analysis arid their Contract Required
 QuanlitaUon Limits (CRQLs) arc listed in Attachment 2.
                                             256

-------
USEPA Contract Laboratory Program
Statement ol Work lor Organic Analysis
Multi-Media, Low to Medium and High Concentration
                               Attachment 1 (Cont'd)
                 Target Compound List and Associated CRQLs
Swn/-Vo(if/to*
Compound
Acensphthalene '
2.4-Dinitrophenol
4-Nirophtnol
Ifcenzoluran
2,4-Dinitrotolu«n«
Diethylphlhalate
4-Chkxophenyl-phenylether
'luwene
4-Niroaniline
4.frDinitrc-2-melriylphtnol
N-ntrosodphtnylamine
4-Bromoprwnyl-prwfyMrMr
Hexachbrobentene
Pertachbrophenol
Phenarthrene
Anthractne
Carbaiole
Di-n-buiylphihaliit
Fkioranthtne
V*°*
Butybeniylphlhilate
J.J.OcNwobtnjidin*
Benzo(a)anlriracene
Chrysene
bis(2-Ethylhexvl)phthalate
Oi-rvoctytphthalMe
Benzo(b)lluoranlhene
Bewo(k)lluorsnthe
-------
USEPA Contract Laboratory Program
Statement ot Work lor Organic Analysis
Muli-Media. Low to Medium and High Concentration
                                  Attachment 1 (Cont'd)
                     Target Compound List and Associated CRQLs
Stml-Volttlltt
Compound
Acenaphthalene


Dlbenzoluran
2.4-Olnltrotoluene
Dtothylphthalatt
4-Cnloroprienvl-prMnytafher

4-NKroanllln*
4.6-Olnltro-2-methylphenol
N-nllrosodlphenylainlne
4-Brornoprwnyl-phenylelhar
Hexachlorobaniene
Pentachlorophenol
Phenanthrene
Anthracene
Carbuole
Dl-n-butylphirialate
Ruoranthene
Pyrene
Buryfbenjytphihalale
3.3'-Otchkxot»rukJio«
Benio(a)anitvacene
Chrysene
W*(2-Ethytfieicyl}phthalate
Ol-n-octylphthalale
Beruo(b)«uoranlhene
BenzoCOnuoranlhene
B«nzo(a)pyrene
lndeno(1,2.3-cd>pyrene
Dlbeiuo{a,h)anlnracene
Benzo(g.h.lX>erylene
S»mf-Votatn»t(1j)
Low to U»
-------
vr
                                  USEPA Contract Laboratory Program
                                  Statement ol Work lor Organic Analysi*
                                  Multi-Media. Low to Medum and High Concentration

                                                              Attachment 1  (Cont'd)
                                               Target Compound List and Associated CRQLs

Compound
Phenol
bi*(2-Chloroe!hyJ)e»>er
2-Chlorophenol
1.3-DfchlorobenMne
1.4-Ofchlorobenzene
1.2-Dlchloroberaene
2-Metiytphenol
2.7-oxyt>ii(t.Chloropropane)
4-Methytphenol
•J-nitro»o-(»-n-dpfopylamine


Nitrobenzene
laophorone
2-Nltrephenol
2,4-Oime«iytphenol
bis(2-Chloroethoxy)meihane
2.4-Dichlorophenol
1 ,2.4-Trichlorobenzene
Naphtialene
4-ChloroaniKne
HexachlorobutacSene
4-Chkxo-3-methylphenol
2-Me*iylnaphthalene
Hexachloroocydopentadene
2,4,6-Trichlorophenol
2.4.6-Trtehlorophenol
2-CH<-«onaph»ialene
2-NHreaniline
Oimethytphthalate
Aeenaphtialene
2.6-Oini»otoluene
3-Nitroanitine
S*nt-VotMlt»(M)
Low to MeoVum
Aqutout
COOL
(up/(..ppb)
10
10
10
10
10
10
10
to
10
10
10
10
10
10
to
10
10
10
10
10
10
10
10
10
10
25-
10
25'
10
10
10
25
Low Soil
coot
("9*0. PP*>)
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
eoo-
330
aoo*
330
330
330
MO*
High Conctntntlon
(».«;
JJqu**S3#<*MuW-
PhttfCROL
(mp/kg. pom)
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
too
20
100
20
20
20
100
                                      •  CKXa previously 5 upland 5 uiykg In 2/M SOW
                                     1  The tample-tpecHle CRQLt lor toil tamplet will be adju (ted tor percent moisture and wM be higher Vian
                                        tho*el«ted above.
                                     2  Medium level tolCHOL- 1000 x AqueoutCROJ. reported In ug/kg.
                                     3  Al CRQLt are bated on wet weight and apply to tolld and Iquld tamplet.
                                     4  Retultt tor botitoKd and iquld tamplet are reported at mg/kg. wet weight
                                                                                                                        21-002-079.221-002-0?e.i
                                                                              259

-------
                             Attachment 1 (Cont'd)
               Target Compound List and Associated CRQLs
Siml-VoltWit
Compound
alpha-BHC
beta-BHC
delta-BHC
gamma-BHC (Lindane)
Heptachlor
AWrin
Heptachlor epoxide
Endosulf an I
Dieldrin
4.4'-DDE
Endrin
EndosuHan II
4.4--DDD
Endosulfan sulfate
4,4'-DDT
Methoxychlor
Endrin katone
Endrin aldehyde
alpha-Chlordane
gamma-Chlordane
Sfml-VolaU/09
Low to M»dlum
Aquvoua
CRQL
(ug/L.ppb)
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.5
0.10
0.10
0.05'
0.05*
Low So//"
CRQL
(ug*9, ppb)
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
3.3
3.3
3.3
3.3
3.3
3.3
3.3
17.0
3.3
3.3
1.7
1.7
Extr*clabl»t(1,2)
High Concentration
LJquid/SoliaVMulti-Phaso
CRQL (mg/kg, ppm)
20
20
20
20
20
2O
20
20
20
20
20
20
20
20
20
20
20
•-
20
20
Note:
  1 AR CRQLs are based on wet weight and apply to solid and Equid samples.

  2 Results for both solid and Gqu'd samples are reported as mg/kg, wet weighL

    Aqueous CRQLs changed from 2/88 SOW to the following:

  * Aqueous CRQLs (ug/L) • alpha- and gamma-Chlordane from 05 to 0.05.

    An low sofl CRQLs changed from 2/88 SOW to the following:

  •• Low Soil CRQLs (ug/kg):      alpha-BHC through Endosulfan I from 8.0 to 1.7;
                             Dieldrin through 4.4'-DDT and Endrin ketone from 16.0 to 3.3;
                             Methoxychlor from 80.0 to 17.0;
                             alpha- and gamma-Chlordane from 80.0 to 1.7.
                                                                                   21-002-079.3
                                         260

-------
I,
                                 Attachment 1 (Cont'd)
                    Target Compound List and Associated CRQLs

Compound
Butyl alcohol
Benzole acid
Monochtorobiphenyl
Dlchlorobiphenyl
Trichloroblphenyl
Tetrachloroblphenyl
Hexachloroblphenyl
Pantachloroblphenyl
Octachlorobiphenyl
Nonachtoroblphenyl
Oecachloroblphenyl
Heptachloroblphenyl
Toxaphene
Aroctor-1016
Aroctor-1221
Aroctor-1232
Aroclor-1242
Aroclor-1248
AroctoM254 """
Aroclor-1260
Scml-Volatllei
Low to Medium
Aqueous
CRQL
(ugA., ppb)
-
—
~
"
—
-
'•
-
•'
"
~
-
5.0*
1.0'
2.0*
1.0*
1.0*
1.0*
1.0
1.0
Low Soil"
CRQL
(ugftg. ppb)
••
••
••
••
-
--
-
-
-
-
-
-
170.0
33.0
67.0
33.0
33.0
33.0
33.0
33.0
Extractables (1,2)
High Concentration
Liquid/Solid/Multi-Phase
CRQL. (mg/kg, ppm)
20
100
100
100
100
100
100
100
200
200
200
100
50
10
10
10
10
10
10
10
Note:
  1 All CRQLs are based on wet weight and apply to solid and liquid samples.

  2 ResuRs for both solid and liquid samples are reported as mg/kg, wet weight.

    Aqueous CRQLs changed from 2/88 SOW to the following:
                   • Aqueous CRQLs (ug/L) -
                               Toxaphene from 1.0 to 5.0;
                               Aroclors-1016.1232.1242. and 1248 from 0.5 to 1.0;
                               Aroclor-1221 from 0.5 to 2.0.
                     All low soil CRQLs changed from 2/888 SOW to the following:

                  - Low Soil CRQLs (ug/kg):
                               Toxaphene from 160.0 to 170.0;
                               Aroclor-1016,1232.1242, and 1248 from 80.0 to 33.0;
                               Aroclor-1221 from 80.0 to 67.0;
                               Aroclor-1254 and 1260 from 160.0 to 33.0.TCL Ex
                                             261

-------
USEPA Contract Laboratory Program
Statement of Work for Organic Analysis
Multi-Media. Multi-Concentration and High Concentration
                                         Attachment 2
                        Target Analyte List and Associated CRQLs

Analylo
Aluminum
Antimony
Arsenic
terium
ieryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
ron
Lead
Magnesium
Manganese
Mercury
Nickel
Potassium
Selenium
Silver
Sodium
Thallium
Vanadium
Zinc
Cyanide
pH
Conductivity
Mutti-Conctntntlon (I)
Aqueous
CRQL
(ugfl. ppb)
200
60
10
200
5
5
5000
10
SO
25
100
3
5000
15
0.2
40
5000
5
10
5000
10
50
20
10
-
-
Low Soil
CRQL
(ugfcg. ppb)
40
12
2
40
1
1
1000
2
10
5
20
0.6
1000
3
0.1
6
1000
1
2
1000
2
10
4
2
-
-
High Concentration (2,3)
Liquid/Solid/Multi-Phase
CF1OL (mg/kg, ppm)
eo
20
5
BO
5
10
80
10
20
40
20
10
80
10
0.3
20
-
5
10
80
20
20
10
1.5
N/A
3.0 (umhos/cm)
 Note:
    1  The sample-specific CRQLs for soil samples wilt be adjusted for percent moisture and will be higher than those listed
      above.

    2  Medium level soil CRQL- 120 x Aqueous CRQL reported In ug/Xg.

    3  Results for both solid and liquid samples are reported as mg/kg, wet weight
                                                                                                  21-002-079.5
                                                   262

-------
                                  APPENDIX IX
     EXAMPLE DIAGRAM FOR A CONCEPTUAL MODEL FOR RISK ASSESSMENT
     This appendix provides a schematic example of a conceptual site model. This example is
a copy of Figure 2-2 of Guidance for Conducting  Remedial Investigations and Feasibility
Studies Under CERCLA (EPA 1989i).
                    ll
•
•
•
•
!
!




§
J
•
•
•
•
It

                                               i
                                                        3
                                          263

-------
                                           Glossary


Accuracy. The degree of agreement of a measured value with the true or expected value of the quantity of concern.

Analytq.  The chemical for which a sample is analyzed.

Analvte Speciation. The ability of an analyte to exist in, or change between, chemically different forms (e.g.,
valence state, complexation state) depending on ambient conditions.

Anthropogenic Background Levels. Concentrations of chemicals that are present in the environment due to human-
made, non-site sources (e.g., industry, automobiles).

Audit Sample. A sample of known composition provided by EPA for contractor analysis to evaluate contractor
performance.
                                                                                            4
Average. The sum of a set of observations divided by the number of observations. Other measures of central
tendency are median, mode, or geometric mean.

Background Sample. A sample taken from a location where chemicals present in the ambient medium are assumed
due to natural sources.

Bias. A systematic error inherent in a method or caused by some artifact or idiosyncrasy of the measurement
system.

Biased Sampling. A sampling plan in which the data obtained may be systematically different from the true mean.
Biased sampling protocols are appropriate for certain objectives (e.g., clustering of samples to search for hot spots).

Biota. The plants and animals of the study area.

Blank. A clean sample that has not been exposed to the analyzed sample stream in order to monitor contamination
during sampling, transport, storage, or analysis.

Broad Spectrum Analysis. An analytical procedure capable of providing identification and quantitation of a wide
variety of chemicals.

Calibration. The comparison of a measurement standard or instrument with another standard or instrument to report
or eliminate, by adjustment, any variation (deviation) in accuracy of the item being compared. The levels of
calibration standards should bracket the range of levels for which actual measurements are to be made.

Cancer Slope Factor. A plausible, upper-bound estimate of the probability of cancer response in an exposed
 individual, per unit intake over a lifetime exposure period.

Chaln-of-Custodv Records. Records that contain information about the sample from sample collection to final
analysis. Such documentation includes labeling to prevent mix-up, container seals to detect unauthorized tamperinj
 with contents and to secure custody, and the necessary records to support potential litigation.

Chemical of Potential Concern.  A chemical initially identified or suspected to be present at a site that may be
hazardous to human health.

 cia«dcal Model.  A statistical description of experimental data that assumes normality and independence.
                                                   265

-------
Confidence. Statistically, a measure of the probability of taking action when action is required or that an observed
value is correct. A confidence limit is a value above or below a measured parameter that is likely to be observed at a
specified level of confidence.

Contract Laboratory Program (CLP). Analytical program developed for analysis of Superfund site samples to
provide analytical results of known quality, supported by a high level of quality assurance and documentation.

Contract Required Quantitntion Limit (CROP. The chemical-specific quantitation levels that the CLP requires to
be routinely and reliably quantitated in specified sample matrices.

Data Assessment. The determination of the quantity and quality of data and their useability for risk assessment.

Datn Quality Indicator (DQ1>. A performance measure for sampling and analytical procedures.

Data Quality Objectives (DQOs). Qualitative and quantitative statements that specify the quality of the data
required to support decisions. DQOs are determined based on the end use of the data to be collected.

Data Review. The evaluation process that determines the quality of reported analytical results. It involves
examination of raw data (e.g., instrument output) and quality control and method parameters by a professional with
knowledge of the tests performed.

 Data llseahilirv. The ability or appropriateness of data to meet their intended use.

 Data Validation.  CLP-specific evaluation process that examines adherence to performance-based acceptance criteria
 as outlined in National Functional Guidelines for Organic (or Inorganic) Data Review (EPA 1991e, EPA 1988e).

 Detection Limit.  The minimum concentration or weight of an analyte that can be detected by a single measurement
 above instrumental background noise.

 Dilution. Adding solvent to a sample, with an analyte concentration higher than the standard calibration curve, to
 bring the analyte concentration into a quantifiahly measurable range.

 Dissolved Metals. Metals present in solution rather than sorbed on suspended particles.

 Domain. A mappable subset of the total area containing the populations, after which distinct statistical properties
 can be described.

 Dnse-Response Evaluation. The process of quantitatively evaluating toxicity infoimation and characterizing the
 relationship between the dose of a contaminant administered or received and the incidence of adverse health effects
 in the exposed populations.

 Duplicate. A second sample taken from the same source at the same time and analyzed under identical conditions to
 assist in (he evaluation of sample variance.

  Exposure Area.  The area of a site over which a receptor is likely to contact a chemical of potential concern.

  Ffrpnsnre Assessment. The determination or estimation (qualitative or quantitative) of the magnitude, frequency,
  duration, and route of exposure.

  Ptpnsnre Pathway. The course of a chemical or physical agent from a source to a receptor. Each exposure pathway
  includes a release from a source, an exposure point, and an exposure route.
                                                     266

-------
Extraction. The process of releasing compounds from a sample matrix prior to analysis.

False Negative ftvpe 11 or hetn errorV  A statement that a condition doe* not exist when it actually does.

False Positive (type I or alpha errort. A statement that a condition does exist when it actually does not.

Field Analyses. Analyses performed in the Held using sophisticated portable instruments or instruments set up in a
mobile laboratory on site. Results are available in real time or in several hours and may be quantitative or
qualitative.

Field Portable.  An instrument that is sufficiently" rugged and not of excessive weight that can be curried and used by
an individual in the field.

Field Screening. Analyses performed in the field using portable instruments. The results are available in real time
but are often not compound-specific or quantitative.

Flrtd Laboratory Analyses. Analyses performed in an off-site analytical laboratory.

Frequency of Occurrence.  The ratio of occurrence of a chemical existing at a site compared to occurrence at all sites
or compared to the frequency at which the chemical was tested for.

Geographical Information System ffilSV A computerized database designed to overlay multiple information
elements such as maps, annotations, drawings, digital photos, and estimated concentrations.

Geostatistical Model.  A statistical or mathematical description of experimental data with special attention to spatial
covariancc or temporal variation.

fieostatLstica. A theory of statistics that recognizes observed concentrations as dependent on one another and
governed by physical processes.  Gcostatistical methods consider the location of data and the size of the site for
calculations.

Heterogeneous Distribution. Sample property that is unevenly distributed in the population.

Historical Data. Data collected before the remedial investigation.

Holding Time.  The length of time from the date of sampling to the date of analysis. CLP designates the holding
time ax the date from laboratory receipt of sample until date of analysis.

Homogeneous Distribution.  A sample property that is evenly distributed over the population.

Hm Spot. Location of a substantially higher concentration of a chemical of concern than in surrounding areas of a
Kite.

Hydrocarbon. An organic compound composed of carbon and hydrogen.

Identification.  Confirmation of the presence of a specific compound or annlytc in a sample.

Instrument Detection Limit HD1A  The lowest amount of a substance that can be detected by an instrument without
correction for the effects of sample matrix, handling and preparation.
                                                    267

-------
      . A measure of exposure expressed as the mass of a substance in contact with the exchange boundary per unit
body weight and unit time.

Integrated Risk Information System (TR1SV  An EPA database containing verified RfDs, RfCs, slope factors, up-to-
date health risks and EPA regulatory information for numerous chemicals. IRIS is EPA's preferred source for
toxicity information for Superfund.

Internal Standard. A compound added to organic samples and blanks at a known concentration prior to analysis.  It
is used as the basts for quantitation of target compounds.

Ititlgmgntal/Purposive Sampling. The process of locating sampling points based on the investigator's best judgment
from historical data of where the sample should be taken.

Kriging. A procedure utilizing a spatial covariance function and known values at sampling locations to estimate
unknown values at unsampled locations. For each estimate, an error of estimate is generated.

Limit of Detection (LQDl The concentration of a chemical that has a 99% probability of producing an analytical
result above background "noise" using a specific method.

Limit of Quamiiation (LQO>. The concentration of a chemical that has a 99% probability of producing an analytical
result above the LOD.  Results below LOQ are not quantitative.

 Linearity.  The agreement between an actual instrument reading and the reading predicted by a straight line drawn
 between calibration points that bracket the reading.

 I .nwest-Qbservable- Ad vcrse-Ef fect-I .evel (I /) AELV In dose experiments, the lowest exposure level at which there
 are statistically or biologically significant increases in frequency or severity of adverse effects between the exposed
 population and its apparent control group.

 Mass Spectrum.  A characteristic pattern of ion fragments of different masses resulting from analysis that can be
 compared wit! a mass spectral library for annlyte identification.

 Mntrix/Mgdium. The predominant material comprising the sample to be analyzed (e.g., drinking water, sludge, air).

 Measurement Error. The difference between the true sample value and the observed measured value.

 Measurement Variability. The difference between an observed measurement and the unknown true value of the
 property being measured.

 %fiia Variability. Variability attributed to matrix effects.

  Method Blank Performance. A measure that defines the level of laboratory background and reagent contamination.
  It Is determined by analyzing a method blank consisting of all reagents, internal standards, and surrogate standards
  that are carried through the entire analytical procedure.
         Detertinn T \m\t fMDIA The detection limit that takes into account the reagents, sample matrix, and
  preparation steps applied to a sample in specific analytical methods.

  Minimum Detectable Relative Difference. Percent difference between two concentration levels that can be detected
  in analyses.

                                                    268

-------
          A mathematical description of an experimental data set.

       Variation. Variation in values or properties of a parameter that are primarily determined by natural forces or
conditions (e.g., variation in background levels of a chemical of potential concern in soils at a site).

Normal Distribution. A probability density function that approximates the distribution of many random variables
and has the form generally called the "bell-shaped curve."
     Hypothesis.  For risk assessment, statistical hypothesis that states on-site chemical concentrations are not
higher than background.

Paniculate. Solid material suspended in a fluid medium (air or water).

performance Evaluation Sample. A sample of known composition provided for laboratory analysis to monitor
laboratory and method performance.

Performance Objectives. Statements of the type and content of deliverables and results that are necessary to assess
the useability of data for risk assessment. For example, documentation (cbain-of-custody records) must be available
to relate all sample results to geographic locations.

Population Variability. The variation in true pollution levels from one population unit to the next Some factors that
cause this variation are distance, direction, and elevation.

 Power.  A parameter used in statistics that measures the probability that the result from a specified sampling or
 analytical process correctly indicates that no further action is required.

 Practical Ouantitation Limit (POL). The lowest level that can be reliably achieved within specified limits of
 precision and accuracy during routine laboratory operating conditions.

 Precision. A measure of the agreement among individual measurements of the same property, under prescribed
 similar conditions.

 Preliminary Remediation finals .PRGsV Initial clean-up goals that 1) are protective of human health and the
 environment and 2) comply with ARARs. They are developed early in the process based on readily available
 information and are modified to reflect results of the baseline risk assessment They also are used during analysis of
 remedial alternatives in the remedial investigation/feasibility study (RI/FS)

 Preservation. Treatment of a sample to maintain representative sample properties.

 Qualifier. A code appended to an analytical result that indicates possible qualitative or quantitative uncertainty in
 the result

 Qualitative. An analysis thai identifies an analyte in a sample without numerical certainty.

 Quality Assurance Protect Plan fQAPiPl. An orderly assembly of detailed and specific procedures which delineate:
 bow data of known and accepted quality is produced for a specific project.
               hnit. The lowest experimentally measurable signal obtained for the actual analyte using a particular
  procedure.
                                                     269

-------
Quantitative  An analysis that gives a numerical level or certainty to the concentration of an analyte in a sample.

Random Sampling. The process of locating sample points randomly within a sampling area.

Ranye of Linearity. The concentration range over which the analytical curve remains linear. The limit within which
response is linearly related to concentration.

Reasonable Maximum Exposure (RMHV The maximum exposure that could reasonably be expected to occur for a
given exposure pathway at a site.  The RME is intended to account for both variability in exposure parameters and
uncertainty in the chemical concentration.

Receptor. An individual organism or species, or a segment of the population of the organism or species, that is
exposed to a chemical.

Recovery.  A determination of the accuracy of the analytical procedure made by comparing measured values for a
spiked sample against the known spike values.

Reference Concentration (RfTV An estimate, with uncertainty spanning an order of magnitude, of continuous
exposure to the human population (including sensitive subgroups) through inhalation that is likely to be without
appreciable risk of deleterious effect during a lifetime.

Reference Dose (RfD). An estimate (with uncertainty spanning an order of magnitude or more) of a daily exposure
level for a human population, including sensitive subpopulations, that is likely to be without an appreciable risk of
adverse health effects over the period of exposure.

Relative Percent Difference fRPD).  A measure of precision which is based on the mean of two values from related
analyses and is reported as an absolute value.

 Relative Response Faeijr (RRR. A measure of the relative mass spectral response of an analyte compared to its
 internal standard. RRFs are determined by the analysis of standards and are used in the calculation of concentration
 of analytes in samples.

 Remedial Investigation (RI>.  A process for collecting data to characterize site and waste and for conducting
 treatability testing as necessary to evaluate the performance and cost of the treatment technologies and support the
 design of selected remedies.

 Representativeness.  The degree to which the data collected accurately reflect the actual concentration or
 distribution.

 Retention  Time. The length of time that a compound is retained on an analytical column (common in GC, HPLC,
 and 1C).

 Risk* Assistant  A software developed for EPA which provides analytical tools and databases to assist exposure and
 risk assessments of chemically contaminated sites.

 Rut rharacigrirfltinn. The process of integrating the results of the exposure and toxicily assessments (i.e.,
 comparing estimates of intake with appropriate toxicological values to determine the likelihood of adverse effects in
 potentially exposed populations).

 Routine Method. A method issued by an organization with appropriate responsibility. A routine method has been
 validated and published and contains information on minimum performance characteristics.

                                                    270

-------
                Sample Integrity. The maintenance of the sample in the same condition as when sampled.

                Snmple Quantitntion Limit (SQL). The detection limit that accounts for sample characteristics, sample preparation
                and analytical adjustments, such as dilution.

                Sampling and Analysis Plan (SAP). A document consisting of a quality assurance project plan, and the field
                sampling plan, which provides guidance for all field sampling and analytical activities that will be performed.

                ftampling Variability. The variability attributed to various sampling schemes, such as judgmental sampling and
                systematic sampling.

:                Sensitivity. The capability of methodology or instrumentation to discriminate between measurement responses for
!                quantitative differences in a parameter of interest.

;               Simple Random Sampling. A sampling scheme where positions, times, or intervals are based on a randomized
\               selection.
i
i               Slope Factor. A plausible upper-bound estimate of the probability of a response per unit intake of a chemical over a
!               lifetime. The slope factor is used to estimate an upper-bound probability of an individual developing cancer as a
\               result of a lifetime exposure to a particular level of a potential carcinogen.

 •               Solvent. A liquid used to dissolve and separate analytes from the matrix of origin.
 i
                Spatial Vnriation. The manner in which contaminants vary within a defined area. The magnitude of difference in
                contaminant concentrations in samples separated by a known distance is a measure of spatial variability.

                Spike. A known amount of a chemical added to a sample for the purpose of determining efficiency of recovery; a
                type of quality control sample.

                Split. A single sample divided for the same measurement by two processes for the purpose of monitoring precision,
                accuracy or comparability of two analyses.

                Standard Deviation? "The most common measure of the dispersion of observed values or results expressed as the
                magnitude of the square root of the variance.

                Standard Operating Procedures (SOPs). A written document which details an operation, analysis, or action whose
                mechanisms are thoroughly prescribed.

                 Srmtified Random Sampling. A sampling scheme where the target population is divided into a certain number of
                 non-overlapping parts for the purpose of achieving a better estimate of the population parameter.

                 Stratified Systematic Sampling. A sampling scheme where a consistent pattern is apportioned to various subareas or
 I                domains.
 j
 >                Stratify. To divide a physical volume or area into discrete units (strata) which are assumed to have different
                 characteristics; a numeric procedure to subdivide a set or sets of data.

                 Surrogate Standard.  A standard of known concentration added to environmental samples for quality control
                 purposes. A surrogate standard is not likely to be found in an environmental sample, but has similar analytical
                 properties to one or more analytes of interest.

                                                                   271

-------
Surrogate Technique. The use of surrogate analytcs to assess the effectiveness of an analytical process (i.e., the
ability to recover analytes from a complex environmental matrix).

Systematic Random fflrid) Sampling. A random sampling plan utilizing points predefined by a geometric pattern.

Target Compound/ Analvte. The compound/analyte of interest in a specific method. The term also has been used in
the Federal Register to denote compounds/analytes of regulatory significance.

Temporal Variation. Variation observed in chemical concentrations that is dependent on time.

Tentatively Identified Compound CTIO.  Organic compounds detected in a sample that are not target compounds,
internal standards or surrogates.                          .                -
           *
Toxicitv Assessment. The toxicity assessment considers the following: 1) the types of adverse health effects
 associated with chemical exposures; 2) The relationship between magnitude of exposure and adverse effects; and 3)
 related uncertainties such as the weight of evidence of a particular chemical's carcinogenicity in humans.

 Toxicolnpical Threshold. The concentration at which a compound exhibits toxic effects.

 Turnaround Time.  The time from laboratory receipt of samples to receipt of a data package by the client.

 Uncertainty. The variability in a process that may consist of contributions from sampling, analysis, review, and
 random error.
      Upper Confidence Limit HJCL). A value that, when calculated repeatedly for different, randomly drawn
 subsets of site data, equals or exceeds the true mean 95% of the time.

 Useful Range. That portion of the calibration curve thai will produce the most accurate and precise results.

 Variance. A measure of dispersion.  It is the sum of the squares of the differences between the individual values and
 the arithmetic mean of the set, divided by one less than the number of values.

 Viscosity.  The physical property of a fluid that offers a continued resistance to flow.

  Volatile Organic*. The solid or liquid compounds that may undergo spontaneous phase change to a gaseous state at
  standard temperature and pressure.

  Wavelength. The linear distance between successive maxima or minima of a wave form.

  WHyht.of.EvMj.npft Classification.  An EPA classification system for characterizing the extent to which available
  data Indicate that an agent is a human carcinogen. Recently, EPA has developed welght-of-evidence systems for
  other kinds of toxic effects, such as developmental effects.
                                                     272

-------
                                        References
Aitchison, J. and Brown, J. A.C. 1957. The Lognormal Distribution with Special Reference to its Uses in
Economics, Cambridge University Press.

American Society for Testing and Materials (ASTM). 1979.  Sampling and Analysis of Toxic Organics in the
Atmosphere.  ASTM Symposium. American Society for Testing and Materials. Philadelphia, PA.

Baudo, R., Glesy, J., and Muntan, H., eds. 1990. Sediments: Chemistry and Toxicity ofln-Place Pollutants. Lewis
Publishers, Inc. Ann Arbor, MI.

Caulcutt, Roland. 1983. Statistics for Analytical Chemists. Chapman and Hall. New York.

Clesceri, et al, eds. 1989. Standard Methods for the Examination of Water and Wastewater. 17lh Edition.
American Public Health Association. Washington, DC.

Dragun, J. 1988. The Soil Chemistry of Hazardous Materials. Hazardous Materials Control Research Institute.
Silver Spring, MD.

Eckel, William P., Fisk, Joan F., and Jacob, Thomas A.  1989. Use of a Retention Index System to Better Identify
Non-Target Compounds. Hazardous Materials Control Research Institute 1989, Proceedings of the 10th National
Conference,  pp. 86-90.

Environmental Protection Agency (EPA). 1983. Methods for Chemical Analysis of Water and Wastes (EPA 200
and 300 Methods).  Environmental Monitoring Systems Laboratory. Las Vegas, NV. EPA/600/4-83/020.

Environmental Protection Agency (EPA). 1984. Methods for Organic Chemical Analysis of Municipal and
Industrial Wastewater (EPA 600 Methods) as presented in 40 CFR Part 136. Guidelines Establishing Test
Procedures for the Analysis of Pollutants under the Clean Water Act.

Environmental Protection Agency (EPA). 1985. Methodology fo/ Characterization of Uncertainty in Exposure
Assessment.  Office of Research and Development. EPA/600/8-85/009.

Environmental Protection Agency (EPA). 1986a. Guidelines for Carcinogenic Risk Assessment. 51 Federal
Register 33992 (September 24,1986).

Environmental Protection Agency (EPA). 1986b. Test Methods for Evaluating Solid Waste (SW846): Physical/
Chemical Methods. Third Edition. Office of Solid Waste.

Environmental Protection Agency (EPA). 1987a. Data Quality Objectives for Remedial Response Activities:
Development Process.  EPA/540/G-87/003 (NTIS 9B88-131370).

Environmental Protection Agency (EPA). 1987b. Field Screening Methods Catalog.  Office of Emergency and
Remedial Response.

Environmental Protection Agency (EPA). 1987c. A Compendium of Supcrfltnd Field Operations Methods. Office
of Emergency and Remedial Response. HPA /540/P-87/001. (OSWER Directive 9355.0-14).

Environmental Protection Agency (EPA). 1988a. Review of Ecological Risk Assessment Methods. Office of Policy
Analysis.  EPA/230/10-88/041.
                                                 273

-------
Environmental Protection Agency (EPA).  1988b. Superfund Exposure Assessment Manual. Office of Emergency
Response. EPA/540/1-88/001. (OSWER Directive 9285.5-1).

Environmental Protection Agency (EPA).  1988c. Geostatistical EnvironmenUil Assessment Software (GEOEAS)
(database).

Environmental Protection Agency (EPA).  1988d. Methods for the Determination of Organic Compounds in
Drinking Water (EPA 500 Methods). Environmental Monitoring Systems Laboratory. Las Vegas, NV. EPA/600/4-
88/039.

Environmental Protection Agency (EPA).  1988e. Laboratory Data Validation: Functional Guidelines for
Evaluating Inorganics Analysis.  Office of Emergency and Remedial Response.

Environmental Protection Agency (EPA).  I989a. Risk Assessment Guidance for Superfund, Volume I:  Human
Health Evaluation Manual, Pan A.  Office of Solid Waste and Emergency Response. EPA/540/1-89/002. (OSWER
Directive 9285.7-01 A).

Environmental Protection Agency (EPA).  1989b.  Risk Assessment Guidance for Superfund, Volume II:
Environmental Evaluation Manual.  Office of Solid Waste and Emergency Response. EPA/540/1-89/001.

Environmental Protection Agency (EPA). 1989c.  E^olo^ical Assessment of Hazardous Waste Sites: A Field and
Laboratory Reference. Environmental Research Laboratory. EPA/600/3-89/013.

 Environmental Protection Agency (EPA). 1989d. Integrated Risk Information System (IRIS) (database). Office of
 Research and Development.

 Environmental Protection Agency (EPA). 1989e. Methods for Evaluating the Attainment of Cleanup Standards.
 Volume 1: Soils and Solid Media. Office of Policy, Planning and Evaluation. EPA/230/2-89/042.

 Environmental Protection Agency (EPA). 1989f. Soil Sampling Quality Assurance User's Guide. Environmental
 Monitoring Systems Laboratory. Las Vegas. NV. EPA/600/8-89/046.

 Environmental Protection Agency (EPA). 1989g. Office of Water Regulations and Standards/Industrial
 Technology Division (ITD) Methods (EPA 1600 Methods). Office of Water.

 Environmental Protection Agency (EPA). 1989h. Data Use Categories for the Field Analytical Support Project. In
 Draft. Office of Solid Waste and Emergency Response.

 Environmental Protection Agency (EPA). 1989L Guidance for Conducting Remedial Investigations and Feasibility
 Studies under CERCLA, Interim Final. Office of Solid Waste and Emergency Response. EPA/540/G-89/004.
 (OSWER Directive 9355.3-01).

 Environmental Protection Agency (EPA). 1990a. Health Effects Assessment Summary Tables. First and Second
 Quarters FY 1990. Office of Research and Development. (OERR 9200.6-303).

 Environmental Protection Agency (EPA). 1990b. Gcostatlstics for Wnstc Management (GEOPACK) (database).

 Environmental Protection Agency (EPA). 1990c. A Rationale for the Assessment of Errors in the Sampling of Soils.
 Office of Research and Development. EPA/600/4-90/013.
                                                  274

-------
                 Environmental Protection Agency (EPA). 1990d. Contract Laboratory Program Statement of Work for Inorganic
    '•             Analysis: Multi-Media, Multi-Concentration. Document No. ILMO 1.0. Office of Emergency and Remedial
                 Response.

    ;             Environmental Protection Agency (EPA). 1990e. Contract Laboratory Program Statement of Work for Organic
    •             Analysis: Multi-Media. Multi-Concentration. Document No. OLM01.0. Office of Emergency and Remedial
    ;             Response.
                                                                                   .x
                 Environmental Protection Agency (EPA). 1991a. ECO Update. Office of Emergency and Remedial Response.
                 Publication No. 9345.0-051.

                 Environmental Protection Agency (EPA). 1991b. Risk Assessment Guidance for Superfund. Volume 1: Human
                 Health Evaluation Manual, Pan B.  Office of Solid Waste and Emergency Response. EPA/540/1-89/002.
                 (OSWER Directive 9285.7-01 A).

                 Environmental Protection Agency (EPA). 199 Ic. Role of Baseline Risk Assessment in Superfund Remedy Selection
                 Decision. Office of Solid Waste and Emergency Response. (OSWER Directive 9355.0-30).

                 Environmental Protection Agency (EPA). 199 Id. Human Health Evaluation Manual Supplemental Guidance:
                 Standard Default Exposure Factors. Office of Solid Waste and Emergency Response. (OSWER Directive 9285.6-
                 03).

                 Environmental Protection Agency (EPA). 199 le. National Functional Guidelines for Organic Data Review.
             \    Office of Emergency and Remedial Response.

                 Hnkel, A.M. 1990. Confronting Uncertainty in Risk Management: A Guide for Decision-Makers. Center for Risk
                 Management. Washington, DC.

                 Gilbert, R.0.1987. Statistical Methods for Environmental Pollution Monitoring. Van Nostrand. New York, NY.

                  Keith, L.H. 1987. Principles of Environmental Sampling.  American Chemical Society. Washington, DC.

                  Keith, L.H. 1990a. Environmental Sampling and Analysis.  In Print. American Chemical Society. Washington,
                  DC.

                  Keith, L.H. 1990b. Environmental  Sampling: A Summary. Environmental Science and Technology. 24:610-615.

                  Koch, George S. and Link, Richard F.  1971. Statistical Analysis of Geological Data. Dover Publications. 0-486-
                  64040.X.

                  Krigc, D.O. 1978. Lognormatde Wysian Geostatisticsfor Ore Evaluation. South Africa Institute of Mining and
                  Metallurgy Monograph Scries.

                 Manahan, S.E.  1975.  Environmental Chemistry. Willard Grant Press. Boston, MA.

                 Neptune, D.E., Brantly, E.P., Messner, M., and Michael, D.I.  1990. Quantitative Decision Making in Superfund.
                 Hazardous Materials Control, pp.  18-27.

                 National Research Council (NRC).  1983. Risk Assessment in the Federal Government: Managing the Process,
.^^'         •>   National Academy Press.  Washington, DC.
V
I                                                                 275

-------
Seichel, H.S.  1956.  The Estimation of Means and Associated Confidence Limits for Small Samples from Lognormal
Populations. A Symposium on Mathematical Statistics and Computer Applications in Ore Valuation. South Africa
Institute of Mining and Metallurgy, pp. 106-122.

Taylor, J.H.  1987. Quality Assurance of Chemical Measurements. Lewis Publishers, Inc. Ann Arbor, MI.

Thistle Publishing 1991. Risk*Assistant (software).  Hampshire Research lastitute. Alexandria, VA.
                                                    276

-------
                                             Index
Accuracy See Data quality indicators
Analytical
  base/neutral/acid (BNA)  39
  iron  1,52,53
  oil and hydrocarbons  45. 51.52.84.106.119
  polycyclic aromatic hydrocarbons  45.119
  phlhalaies and non-pesticide chlorinated com-
    pounds  52
  volatile organics (VOAs)  55.57,78.80.113
Analytical methods  13.21.22.25.26.29. 30, 33.
    41.45,47,57,59,63,64.78,83,89,99. 100.
     117..118.120
  atomic absorption (AA)  47, 55,58
  gas chromatography-mass spectrometry (GC-
    MS)   41,45.46,52.53
  gel permeation chromatography (GPC)  39
  inductively coupled plasma (ICP)  52. 53. 55. 58.
     101
  X-ray fluorescence (XRF)  57
Analytical services  3. 21, 28.29. 83
  field analyses   2.21.28. 29. 57. 58. 84. 88. 89
  fixed laboratory analyses  21. 29, 54. 57. 58. 84.
     89,100
  quick turnaround method   28
  special analytical services (SAS)  29
Automated data review   35

B
Background sampling   29,50,75,119
   anthropogenic  2.75,119.120
   sampling  29,50
 Baseline human health risk assessment   1.3,4.7
 Biota sampling  39,83
 Chain-of-custody  29,101
 Chemical intake  14,15
 Chemicals of potential concern  1,4,25,26,29,30,
     35.40,41,46,47.50,52.53.55,63-65,72-74,
     77.78,80.83.84,87.88,117-120
 Comparability See Data quality indicators (DQIs)
 Completeness See Data quality indicators (DQIs)
 Concentration of concern  10,33.34,47,48,83
 Conceptual model  11,18,22,28
 Contract Laboratory Program (CLP)   2.29,41.49,
     58,83.84.87,100.103.105,106,113
 Contract required detection limit (CRDL)   49
 Contract required quantltatlon limit (CRQL) 49
 Corrective action  4,22.36.88,95,97,100.101.
      106
D
Data
  assessment  11.21.22.95. 100-103,105,107.
     109.111.113.114.116
  collection  1-4.7.11.18.20.25.29-31.33.34.
     36. 37.50.51.63.81.101.106-112.116
  qualifiers  4.100,113
  review  2. 3.4.20.22.23.25.29. 34. 35.89.99-
     103.105.107.117-119
  sources  1.2,3,26,28,29,99,101.111
Data quality indicators (DQIs)  3,29. 31,76. 103,
     121
  accuracy  25.29. 31.33,34,39.49.51,55, 58.
     99.101.102. 105-107.112,113.116-118
  comparability 33,57.76,78,80,99,105,107,
     108.112.114,116,121
  completeness  76-78,99,100,102,105-107.114.
     116-118.120,121
  precision  29. 34.49.99-102.105-107,109,111-
     113.116-118
  representativeness  76.99,105,107-109.114.
     116.117.121
Data quality objectives (DQOs) 2,11,13,31.34,
     63.100,110.111
Data useability criteria  3,25.26.99,117,121
Design decisions  81.89
Detection limits 2. 25.28.30,33,37,45-48.54, 55.
     77,83.84.87.89,117-120
  contract required detection limit (CRDL)  113
  contract required quantitation limit (CRQL)   113
   instrument detection limit (IDL)   47,48
   limit of quantilation (LOQ)  49,50
   method detection limit (MDL)  2.47.48.49, 50,
     102.113
   practical quantitation limit (PQL)  49
   sample quantitation limit (SQL)   2.22,23,48,49,
     50,84
 Exposure  95.97.101.105.107.108.112
   area  4.11.13.18.20.25.26.33.54.55.63.65.
     72,74,77.78.80.89,120.121
   assessment  4,7,13.14.15.17.18.101.102.108
   pathway  11,13-15,17.18,33.58.63.80.89.
      117,120.121
 False negatives  11.13.18,25.35.40.41.47.48.
      50.58,64,75.76,101,105,108.113.116-118
                                                  277

-------
False positives  11,13,25,30,35.41,45,47,48.
    50,53,64,76,101,105.113.117-119
Field analyses See Analytical Services
Field records  29
Fixed laboratory analyses See Analytical Services

G
Geographical Information System (CIS)   18.72

H
Hazard Ranking System (MRS)  13,26
Health Effects Assessment Summary Tables
     (HEAST)   15
Historical data   11,18,26,28.41.45.52.73.74.
     78.119
Hot spots  13,33. 51,54,57,66, 73-76,78.89

I
Integrated Risk Information System (IRIS)  15

L
Laboratory performance   25,33,58,59. 88,107.
     Ill
Land use alternatives  78
Linearity
   limit of linearity (LOL) 50
   range of linearity   47

1VI
 Measurement error  33.37,38,50,76,109,111
 Media variability  51,74

 N
 National Priorities List (NPL)   50
 Natural variation  38
 Performance evaluation  33,39.58,87,88,116
 Performance measures   63,76,80,88,110
 Performance objectives  4,25,29,33,50,97,105,
      111
 Precision See Data quality indicators (DQIs)
 Preliminary remediation goals (PRGs)  2,48

 Q
 Qualified data  2,23.105,106.113
 Qualitative/quantitative analysis  57
 Quality assurance (QA)  2.18,20,22,29,39,58,
      76.100.101
 Quality assurance project plan (QAPjP)  2,20,29,
      33
 Quality control (QO  2.29,33.34.37.50,58,59.
    88,100-103. 105.107,108, 111,113.116.118.
    119
Reasonable maximum exposure (RME)   13.14, 17,
    55.66.105.107,109,116
Reference concentrations (RI'Cs)   15,17
Reference doses (RIDs)  15.17
Remedial investigation (RI)   1-4. 11, 18.20.21,25,
    26,28.29,63,65. 81.95. 100. 105
Remedial project manager (RPM)  1,4,11. 18,20-
    23.25. 29.30. 34-37. 39.41.45^7.51.53. 58.
    59.63-65. 72. 77.78.80,81. 84,85.8T-89.95.
    113
Representativeness See Data quality indicators
    (DQIs)
Resource issues   88
Risk Assessment Guidance for Supcrfund
    (RAGS)  1-3.7..13-15,  17, 18. 102, 114. 119
Risk assessor  1-4.7.14.15.18.20-23.25,50.52-
    55.58.63-65,77,78,80,81,84.87-89,95,97.
     100-102. 106-108, 110, 111. 113, 114, 116

s
Sample
   preparation   47,49.51.54,55,77,88.108.112
   preservation  76.108,116
Sampling and analysis plan (SAP)   1,20-22,25,63,
     74.88. 97. 100. 107. 110. 112.113
Sampling design methods
   classical model  65,72.78.88
   gcosiatistical model  65.66. 73-75,78.88
   judgmental/purposive model  65.73,74
   systematic grid sampling   65,66.72,75,78,88
Sampling Design  Selection Worksheet  63,65,72.
     80, 83.89
Sampling variability   64,65,74,77.108,109,110,
     116
Scheduling   21
 Scoping  11.25.28,29.41.88,105
 Site
   concentrations  11,13,25.63-66,72-77,80,95,
     101,107.109.116,119,120
   inspections  3,18,26,73
 Soil   4.37.38,41,50,51.55,119.120
   data collection  1,4.63.77.78.80.81.117.119.
     121
   locution of hot spots  66,73-75,78
   sampling depth  78,80
   characierislics 11,80
 Soil Depth Sampling Worksheet  37,63
                                                  278

-------
Standard operating procedures (SOPs)  29, 31,100.
    101
Target compound list  4
Tentatively identified compounds (TICs)  41,45, 52
Toxicity assessment  4,7,15,17.22
Turnaround time  2,29, 54. 58.83.84. 87,89

u
Uncertainty  1-4,7.10,11.14.15.17. 18.25. 33,
    37,38,50.51. 55.63.76.80.81, 89.95. 97.
    102,105,107,111.114.117,121
  analytical  7,10,14.15.17,18. 80
  sampling  63.76.77, 80.81.89.118
  total  76.77
                                                     U.S.  Environmental Protection Agency
                                                     Region 5, Library (PL-12J)         ^^
                                                     77 West Jackson Boulevard,  12u» MOOT
                                                     Chicago,  IL   60604-3590
  •  U.S.  r,.P.M.j1992-311-B93:60679
                                                279

-------