EPA/540/G-87/OQ4
                                            (OSWER Directive 9355.0-7B)
                                                          March 1987
                  DATA QUALITY OBJECTIVES

             FOR REMEDIAL RESPONSE ACTIVITIES

                         Example Scenario

(RI/FS Activities at a Site with Contaminated Soils and Ground Water)
                           Prepared for:

            Office of Emergency and Remedial Response
                               and
               Office of Waste Programs Enforcement
           Office  of Solid Waste and Emergency Response

               U.S. Environmental  Protection Agency
                      Washington, DC  20460
                           Prepared by:

                COM Federal Programs Corporation
                7611  Little River Turnpike, Suite 104
                      Annandale, VA  22003
                   EPA Contract No. 68-01-6939
                           March 1987

-------
                      NOTICE

This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication.  Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.

-------
                                          PREFACE


This Data Quality Objectives For Remedial Response Activities (Example Scenario RI/FS Activities at
a Site with Contaminated Soils and Ground Water) provides an outline of the process for development
of data quality objectives (DQOs) for RI/FS activities under the Comprehensive Environmental
Response, Compensation and Liability Act of 1980 (CERCLA) and the Superfund Amendments and
Reauthorization Act of 1986 (SARA).  DQOs are qualitative and quantitative statements specified to
ensure that data of known quality are obtained during remedial response activities to support an
Agency decision.  This document is intended to demonstrate the development of data quality
objectives (DQOs)  for an example RI/FS activity. The example presented herein should be utilized in
conjunction with the companion manual, Data Quality Objectives For Remedial Response Activities
(Development Process) in developing DQOs  for site  specific applications.

This example and the companion guidance manual have been  prepared under the direction of the Office
of Solid Waste and Emergency  Response (OSWER). The documents were prepared in accordance with the
National  Oil and Hazardous Substance Pollution Control Contingency Plan (NCP) final rule, published
in the Federal Register November 20, 1985 and effective February 18,  1986. These documents will be
updated in the near future to be consistant with SARA and the new NCP.  These documents are part of
a series of documents which includes the following titles:

     •   Guidance on Remedial Investigations  Under CERCLA (EPA 540/G-85/002)

     •   Guidance on Feasibility Studies Under CERCLA (EPA 540/G-85/003)

     •   Superfund Remedial Design and  Remedial  Action Guidance (OSWER Directive 9355.0-4A)

     •   Compendium of Field Operations Methods (planned June 1987)

     •   Superfund Public Health Evaluation Manual (OSWER Directive 9285.4-1)

     •   Superfund Exposure Assessment  Manual (OSWER Directive 9285.5-1)

Collectively, these documents provide guidance for the development and performance of technically
sound and cost-effective remedial response activities which will support the program goals of both
the Office of Emergency and Remedial Response (OERR) and the Office of Waste Programs Enforcement
(OWPE). These documents are also available for use by state agencies and private parties conducting
remedial  response activities to ensure that their activities are consistent with the intent of
CERCLA.
                                             Hi

-------

-------
                            TABLE OF CONTENTS






SECTION                                                             PAGE




1.0   INTRODUCTION                                                    1-1




     1.1   DQO STAGES                                                 1-1




     1.2   RELATIONSHIP OF DQOs TO RI/FS EXECUTION                     1-3




     1.3   FORMAT AND PURPOSE OF DOCUMENT                           1-3




2.0   SUMMARY OF DQO DEVELOPMENT EXAMPLE                           2-1




     2.1   STAGE 1 - IDENTIFY DECISION TYPES                            2-1




     2.2   STAGE 2 - IDENTIFY DATA USES/NEEDS                           2-1




     2.3   DESIGN DATA COLLECTION PROGRAM                            2-4




3.0   DQO STAGE 1 - RI/FS SCOPING PROCESS                                3-1




     3.1   IDENTIFY DECISION TYPES                                     3-1




     3.2   IDENTIFY AND INVOLVE DATA USERS                            3-1




     3.3   EVALUATE AVAILABLE INFORMATION                            3-3




          3.3.1   DESCRIBE CURRENT SITUATION                          3-4




          3.3.2   REVIEW EXISTING DATA                                3-3




          3.3.3   ASSESS ADEQUACY OF DATA                            3-5




     3.4   DEVELOP CONCEPTUAL SITE MODEL                            3-5




     3.5   SPECIFY RI/FS OBJECTIVES                                     3-10




     3.6   DETERMINE NEED FOR ADDITIONAL DATA                      3-10




4.0   DQO STAGE 2 - RI/FS DEVELOPMENT                                 4-1




     4.1   DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: OVERALL RI/FS    4-1




     4.2   REMEDIAL ALTERNATIVES                                      4-4




     4.3   IDENTIFY DATA TYPES                                         4-5




     4.4   IDENTIFY DATA QUALITY/QUANTITY NEEDS                      4-5




     4.5   EVALUATE SAMPLING/ANALYSIS OPTIONS                        4-5

-------
                            TABLE OF CONTENTS
                                  (continued)
SECTION                                                             PAGE

     4.6  REVIEW PARCC PARAMETERS                                   4-7

5.0   DQO DEVELOPMENT - PHASE I REMEDIAL INVESTIGATIONS              5-1

     5.1  DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: RI PHASE IA -
          SAMPLING OF EXISTING WELLS                                 5-1

          5.1.1   IDENTIFY DATA USES: RI PHASE IA - SAMPLING OF
                  EXISTING WELLS                                      5-1
          5.1.2   IDENTIFY DATA TYPES: RI PHASE IA - SAMPLING OF
                  EXISTING WELLS                                      5-1
          5.1.3   IDENTIFY DATA QUALITY NEEDS:  RI PHASE IA - SAMPLING
                  OF EXISTING WELLS                                   5-4
          5.1.4   IDENTIFY DATA QUANTITY NEEDS: RI PHASE IA -
                 SAMPLING OF EXISTING WELLS                          5-4
          5.1.5   EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE IA -
                  SAMPLING OF EXISTING WELLS                         5-4
          5.1.6   REVIEW PARCC PARAMETERS: RI PHASE IA - SAMPLING
                  OF EXISTING WELLS                                   5-6

     5.2  DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: RI PHASE IB -
           SOIL GAS INVESTIGATIONS                                    5-9

          5.2.1   IDENTIFY DATA USES: RI PHASE IB - SOIL GAS
                  INVESTIGATIONS                                      5-9
          5.2.2   IDENTIFY DATA TYPES: RI PHASE IB - SOIL GAS
                  INVESTIGATIONS                                      5-9
          5.2.3   IDENTIFY DATA QUALITY NEEDS:  RI PHASE IB - SOIL
                  GAS INVESTIGATION                                  5-10
          5.2.4   IDENTIFY DATA QUANTITY NEEDS: RI PHASE IB - SOIL
                  GAS INVESTIGATION                                  5-10
          5.2.5   EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE IB -
                  SOIL GAS INVESTIGATION                              5-12
          5.2.6   REVIEW PARCC PARAMETERS: RI PHASE IB - SOIL GAS
                  INVESTIGATION                                      5-13

     5.3  DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: PHASE 1C -
           SURFACE SOIL INVESTIGATION                                5-13
          5.3.1   IDENTIFY DATA USES: RI PHASE 1C - SURFACE SOIL
                  INVESTIGATIONS                                     5-13
          5.3.2   IDENTIFY DATA TYPES: RI PHASE 1C - SURFACE SOIL
                  INVESTIGATIONS                                     5-15
          5.3.3   IDENTIFY DATA QUALITY NEEDS:  RI PHASE 1C - SURFACE
                  SOIL INVESTIGATIONS                                5-15
          5.3.4   IDENTIFY DATA QUANTITY NEEDS: RI PHASE 1C -
                  SURFACE SOIL INVESTIGATIONS                        5-15

-------
                            TABLE OF CONTENTS
                                  (continued)
SECTION
          5.3.5   EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE 1C -
                  SURFACE SOIL INVESTIGATIONS
       ,   5.3.6   REVIEW PARCC PARAMETERS: RI PHASE 1C - SURFACE
                  SOIL INVESTIGATIONS.

     5.4   DQO STAGE 3 - DESIGN DATA COLLECTION PROGRAM: PHASE I
           REMEDIAL INVESTIGATIONS

          5.4.1   ASSEMBLE DATA COLLECTION COMPONENTS: PHASE I
                  REMEDIAL INVESTIGATIONS
          5.4.2   DEVELOP DATA COLLECTION DOCUMENTATION: PHASE I
                  REMEDIAL INVESTIGATIONS

     5.5   DQO STAGE 1 - COLLECT AND EVALUATE DATA: PHASE I REMEDIAL
           INVESTIGATIONS

          5.5.1

          5.5.2

          5.5.3
ANALYSIS OF RESULTS: RI PHASE IA - EXISTING WELL
 SAMPLING
ANALYSIS OF RESULTS: RI PHASE IB - SOIL GAS
 SAMPLING
ANALYSIS OF RESULTS: RI PHASE 1C - SURFACE SOIL
 SAMPLING
5.5.3.1  CALIBRATION OF X-MET (PRECISION AND
          ACCURACY ACHIEVED FOR METALS ANALYSIS
5.5.3.2    GEOSTATISTICAL ANALYSIS OF SURFACE SOIL
          SAMPLING  RESULTS
6.0   DQO DEVELOPMENT - PHASE II REMEDIAL INVESTIGATIONS

     6.1   DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: RI PHASE HA -
           GROUND WATER INVESTIGATIONS

          6.1.1   IDENTIFY DATA USES: RI PHASE HA - GROUND WATER
                  INVESTIGATIONS
          6.1.2   IDENTIFY DATA TYPES: RI PHASE IIA - GROUND WATER
                  INVESTIGATIONS
          6.1.3   IDENTIFY DATA QUALITY NEEDS: RI PHASE IIA - GROUND
                  WATER INVESTIGATIONS
          6.1.4   IDENTIFY DATA QUANTITY NEEDS: RI PHASE IIA -
                  GROUND WATER INVESTIGATIONS
          6.1.5   EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE II A -
                  GROUND WATER INVESTIGATIONS
          6.1.6   REVIEW PARCC PARAMETERS:  RI PHASE IIA GROUND
                  WATER INVESTIGATIONS
PAGE


5-18

5-22


5-22


5-22

5-22


5-27


5-27

5-27

5-31

5-31

5-34

 6-1


 6-1


 6-1

 6-1

 6-1

 6-3

 6-5

 6-5
                                    Vll

-------
                            TABLE OF CONTENTS
                                  (continued)
 SECTION                                                             PAGE

      6.2   DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS: PHASE IIB -
           SUBSURFACE SOIL INVESTIGATIONS                            6-7

           6.2.1    IDENTIFY DATA USES: RI PHASE IIB - SUBSURFACE
                  SOIL INVESTIGATIONS                                 6-7
           6.2.2    IDENTIFY DATA TYPES: RI PHASE IIB - SUBSURFACE
                  SOIL INVESTIGATIONS                                 6-8
           6.2.3    IDENTIFY DATA QUALITY NEEDS: RI PHASE IIB -
                  SUBSURFACE SOIL INVESTIGATIONS                      6-8
           6.2.4    IDENTIFY DATA QUANTITY NEEDS: RI PHASE IIB -
                  SUBSURFACE INVESTIGATIONS                          6-9
           6.2.5    EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE II B -
                  SUBSURFACE SOIL INVESTIGATIONS                     6-10
           6.2.6    REVIEW PARCC PARAMETERS: RI PHASE II B - SUBSURFACE
                  SOIL INVESTIGATIONS                                6-12

      6.3   DQO STAGE 3 - DESIGN DATA COLLECTION PROGRAM: PHASE II
           REMEDIAL INVESTIGATIONS                                  6-14

           6.3.1    ASSEMBLE DATA COLLECTION COMPONENTS: PHASE II
                  REMEDIAL INVESTIGATIONS                           6-14
           6.3.2    DEVELOP DATA COLLECTION DOCUMENTATION: PHASE II
                  REMEDIAL INVESTIGATIONS                           6-14

      6.4   DQO STAGE 1 - COLLECT AND EVALUATE DATA: PHASE II REMEDIAL
           INVESTIGATIONS                                            6-14

           6.4.1    ANALYSIS OF RESULTS: RI PHASE HA - GROUND WATER
                  INVESTIGATIONS                                     6-14
           6.4.2  ,  ANALYSIS OF RESULTS: RI PHASE IIB - SUBSURFACE
                  SOIL INVESTIGATIONS                                6-18

      6.5   EXTENSION OF THE DQO PROCESS TO THE REMEDIAL DESIGN (RD)
           AND REMEDIAL (RA) OF UNCONTROLLED HAZARDOUS WASTE
           REMEDIAL RESPONSE ACTIVITIES                              6-20
7.0   CONCLUSION

APPENDIX A  HISTORICAL PRECISION AND ACCURACY DATA
APPENDIX B  CONTRACT - REQUIRED DETECTION LIMITS FOR HSL ANALYSIS
             USING CLP IFB PROCEDURES
7-1
                                    Vlll

-------
                             LIST OF FIGURES
                               I
FIGURE                         .                                   PAGE

1-1       DQO THREE STAGE PROCESS                                    1-2
1-2       PHASED RI/FS APPROACH AND DQO PROCESS                       1-4
2-1       DQO PROCESS WITHIN PHASED WORK FLOW FOR EXAMPLE RI/FS      2-2
2-2       EXAMPLE SITE LOCATION MAP                                  2-3
3-1       DQO STAGE 1 ELEMENTS                                       3-2
3-2       SITE PLAN                                                   3-4
3-3       FIT SITE INVESTIGATION SAMPLING LOCATIONS                    3-6
3-4       CONCEPTUAL SITE MODEL                                     3-8
4-1       DQO STAGE 2 ELEMENTS                                       4-2
5-1       RESIDENTIAL AND ON-SITE WELL LOCATIONS                      5-5
5-2       INITIAL SOIL GAS SAMPLING GRID                              5-11
5-3       HYBRID GRID FOR SAMPLING THE DEPRESSED AREA               5-17
5-4       LOCATION OF CALIBRATION SAMPLES                           5-20
5-5       X-MET QC SAMPLES                                          5-21
5-6       DQO STAGE 3 ELEMENTS-DESIGN DATA COLLECTION PROGRAM      5-24
5-7       PHASE I REMEDIAL INVESTIGATION SCHEDULE                    5-29
5-8       RESULTS OF SOIL GAS SAMPLING                               5-30
5-9       LEAD CONCENTRATION CONTOURS                             5-36
5-10      HISTOGRAM OF LEAD CONCENTRATION                         5-38
5-11      VAR1OGRAM OF LEAD CONCENTRATIONS                        5-40
6-1       LOCATION OF MONITORING WELLS                              6-4
6-2       SOIL SAMPLING LOCATIONS                                   6-11
6-3       PHASE II REMEDIAL INVESTIGATION SCHEDULE                   6-16
6-4       SOIL SAMPLING RESULTS  (DEPTH 8FT) AND KRIGED
         CONTOUR LINE (4 ppm)                                       6-19
6-5       KRIGED PROBABILITY CONTOURS                               6-21
                             LIST OF TABLES

TABLE                                                             PAGE

2-1       DATA QUALITY SUMMARY                                     2-5
2-2       DATA COLLECTION PLAN SUMMARY - PHASE I                     2-7
2-3       DATA COLLECTION PLAN SUMMARY - PHASE II                    2-9
3-1       EXAMPLE SITE FIT SITE INVESTIGATION DATA                     3-7
4-1       DATA USES                                                  4-3
4-2       RI/FlS DATA TYPES                                            4-6
5-1       DATA USES                                                  5-2
5-2       DQO SUMMARY FORM                                         5-3
5-3       DQO SUMMARY FORM                                        5-14
5-4       DQO SUMMARY FORM                                        5-23
5-5       DATA COLLECTION COMPONENTS - PHASE I                      5-26
5-6       EXAMPLE SITE QUALITY ASSURANCE PROJECT PLAN ELEMENTS      5-28
5-7       RESULTS OF REPLICATE ANALYSIS FOR LEAD                     5-32
5-8       ACCURACY AND PRECISION OF THE X-MET                       5-33
5-9       PROBABILITY TABLE                                         5-35
6-1       DATA USES                                                  6-2
6-2       DQO SUMMARY FORM                                         6-6
6-3       DQO SUMMARY FORM                                        6-13
6-4       DATA COLLECTION COMPONENTS - PHASE II                     6-15

-------
                                   LIST  OF ACRONYMS

ARAR     Applicable or Relevant and Appropriate Requirements
ATSDR    Agency for Toxic Substances and Disease Registry
CERCLA   Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (Superfund)
CDC       Center for Disease Control
CLP       Contract Laboratory Program
DQO      Data Quality Objective
EMSL-LV  Environmental Monitoring and Support Laboratory - Las Vegas
BSD       Environmental Services Division (of EPA)
FIT        Field Investigation Team
FS         Feasibility Study
GC/MS    Gas Chromatograph/Mass Spectrograph
HSL       Hazardous Substance List
LFI        Limited Field Investigation
MDL      Method Detection Limit
NBS       National Bureau of Standards
NCP       National Oil and Hazardous Substance Pollution Contingency Plan
NEIC      National Enforcement Investigation Center
NPL       National Priorities List
ORC       Office of Regional Counsel
PARCC    Precision, Accuracy, Representativeness, Completeness, Comparability
PRP       Potentially Responsible Party
QAMS     Quality Assurance Management Staff
QAPP     Quality Assurance Program Plan
QAPjP     Quality Assurance Project Plan
QA/QC    Quality Assurance/Quality Control
RA        Remedial Action
RAS       Routine Analytical Service
RD        Remedial Design
RI         Remedial Investigation
ROD       Record of Decision
RPM       Remedial Project Manager
RSCC      Regional Sample Control Center
S&A       Sampling and Analysis
SAS        Special Analytical Service
SMO       Sample Management Office
SRM       Standard Reference Materials
TAG       Technical Advisory Committee
TAT       Technical Assistance Team
TIC        Tentatively Identified Compounds
TSCA      Toxic Substances Control Act
VOC       Volatile Organic Compounds

-------
                  :                ACKNOWLEDGMENTS

This document was developed for the Office of Solid Waste and Emergency Response (OSWER) by a
task force composed of the following individuals:

           Randall! Kaltreider  (Hazardous Site Control Division, OERR)
           Linda Boornazian (CERCLA Enforcement Division, OWPE)
           Andrew Szilagyi (CDM Federal Programs Corporation)
           Jeffery Sullivan (Camp Dresser & McKee Inc.)
           Rosemary Ellersick (CDM Federal Programs Corporation)
           Tom Pedersen (Camp Dresser & McKee Inc.)
           James Occhialini (Camp Dresser & McKee Inc.)
           Dennis Gagne (Region 1, Waste Management Division)
           Bill Coakley (Region 2, Environmental Services Division)
           Edward Shoener (Region 3, Hazardous Waste Management Division)
           Diane Moshman (Region 5, Waste Management Division)
           Steve Lemons (Region 6, Environmental Services Division)
           Bill Bunn (Region  7, Environmental Services Division)
           Mike Carter (Hazardous Response Support Division, OERR)
           Duane Geuder (Hazardous Response Support Division, OERR)
           Michael Kosakowski  (CERCLA Enforcement Division, OWPE)
           Dennisse Beauchamp ( CERCLA Enforcement Division, OWPE)
           Gary Liberson (Lloyd Associates)
           Craig Zamuda (Policy Analysis Staff, OERR)
           John Warren (Statistical Policy Branch, OPPE)
           Wendy ;Sydow (CDM Federal Programs Corporation)
           Paul Clay (NUS Corporation)

Helpful suggestions and comments on the draft document were provided by the following as well
as other EPA and contractor staff.
                  i
           David F,. Doyle (Camp Dresser & McKee Inc.)
           Dean Neptune (QAMS)
           Gene Btantly (RTI)
           Daniel Michael (RTI)
                                           XI

-------

-------
          DQO STAGES
  RELATIONSHIP OF DQOs
     TO RI/FS EXECUTION
FORMAT AND PURPOSE OF
            DOCUMENT

-------

-------
                                      1.0  INTRODUCTION
Data quality objectives (DQOs) are qualitative and quantitative statements specified to ensure that
data of known and appropriate quality are obtained during remedial response activities.  To ensure
that the data generated during the remedial response activities are adequate to support Agency
decisions, a clear definition of the objectives and the method by which decisions will be made must
be established early in the project planning process.  These determinations are facilitated through
the development of DQOs.

Data quality objectives are specified for each data collection activity associated with a remedial
response. The majority of data collection activities will be undertaken during the remedial
investigation (RI) and additional data needs may be identified during the feasibility study (FS),
remedial design (RD), and remedial action (RA).

The intent of this document is to illustrate, through a  case study scenario, how development of DQOs
is incorporated into RI/FS  planning activities.  The example describes RI/FS planning activities as
the context for DQO development. However, the site and analytical values are hypothetical and,
importantly, the example is not to be considered a complete description on how to develop a full
RI/FS.  Detailed guidance on RI/FS activities can be found in EPA policy and technical guidance
documents,  technical and scientific literature, and through experienced EPA and remedial  contractor
staff.  The guidance document provides references for additional information at the end of each
section.
                   I
In actual practice to date, RI/FS projects conducted under CERCLA have complied with the intent of
the DQO process.  DQOs  have been informally incorporated as parts of sampling and analysis plans,
quality assurance project plans, or work plans.  The purpose of this example and companion manual
(Data Quality Objectives For Remedial Response Activities - Development Process, EPA 1987)  is to
present a formal approach  to DQO development and documentation.  The DQO process outlined in  this
document serves as the basis for development of this example.

 1.1    DQO STAGES

DQOs are developed using the following three-stage process:

      «   Stage 1 - Identify decision  types

      «   Stage 2 - Identify data uses and needs
                   I
      «   Stage 3 - Design data collection program

 Figure 1-1  identifiejs the major components in each of the DQO stages. These stages should be
 undertaken  in an interactive  and iterative manner whereby all the elements of the DQO process are
 continually reviewed and applied during execution of  data collection activities. As such, DQOs are
 developed at the onset of a project and revised or expanded as needed based  upon the results of each
 data collection activity. During the implementation of the DQO process, these stages occur in a
 natural progression  and flow together  without a formal stage delineation.
                                                1-1

-------
              STAGE  1
      IDENTIFY DECISION TYPES
     • IDENTIFY & INVOLVE DATA USERS
     • EVALUATE AVAILABLE DATA
     • DEVELOP CONCEPTUAL MODEL
     • SPECIFY OBJECTIVES/DECISIONS
           STAGE 2
  IDENTIFY DATA USES/NEEDS
  • IDENTIFY DATA USES
  • IDENTIFY DATA TYPES
  • IDENTIFY DATA QUALITY NEEDS
  • IDENTIFY DATA QUANTITY NEEDS
  • EVALUATE SAMPLING/ANALYSIS OPTIONS
  • REVIEW PARCC PARAMETERS
             STAGE 3
 DESIGN DATA COLLECTION PROGRAM
 • ASSEMBLE DATA COLLECTION COMPONENTS
 • DEVELOP DATA COLLECTION DOCUMENTATION
          FIGURE  1-1
DQO THREE-STAGE PROCESS
               1-2

-------
1.2    RELATIONSHIP OF DQOs TO RI/FS EXECUTION

The overall objective of the RI is to determine the nature and extent of the threat posed by the
release or threat of release of hazardous substances and to evaluate proposed remedies.  The
ultimate goal of the FS is to select the most cost-effective remedial alternative which mitigates
threats to and provides protection of public health, welfare, and the environment, consistent with
theNCP.

The quality and amount of data required to identify sources of contamination and delineate the
extent  of contamination with adequate certainty to select a remedial alternative will  vary by site.
In most situations it may not be possible to identify all data needs during the initial scoping
activities.  Rather, data needs will become more apparent as additional data are obtained and
evaluated.  Phasing of RI/FS projects is undertaken to accommodate this iterative process. By
separating the RI into distinct phases, data can be collected and evaluated sequentially with a
refinement and/or redefinition of data collection  needs at the completion of each phase.

Figure 1-2 shows the relationship between  the phased RI/FS approach and the DQO process.  The DQO
process is applied during scoping and following  each data collection activity.  Through the
application of the DQO process, decisions regarding the need for additional data can be made and
subsequent data Collection activities designed.

It is important to realize that DQOs are an integrated set of thought processes which define data
quality requirements based on the end use of the data.  At no time during the RI/FS, RD or RA is a
DQO deliverable required.  Also, it is not required that the procedures for selecting  an analysis or
sampling option be discussed in the detail shown in this example.  The example merely shows the type
of analysis which must be performed to correctly select an option. The rationale for selection and
the actual DQO will be documented in the sampling and analysis (S&A) plan in accordance with
regional requirements.

1.3    FORMAT AND PURPOSE OF DOCUMENT

This document is  intended to provide an example of the process  of DQO development. The example is
based  upon an  RI/FS for a hypothetical uncontrolled hazardous waste site with  known soil and  ground
water contamination.  Conditions and requirements will vary from site to site but the process
remains the same.

This example document is organized in  the following manner.  Section 2.0 presents a brief summary of
the example.  Section 3.0 describes Stage  1 activities for the RI/FS scoping process.  Section  4.0
describes DQO Stage 2 activities for the overall  RI/FS.  Section 5.0 describes Stage  2 and 3
activities for  the first phase of the RI, while Section 6.0 describes Stage 2 and 3 for the second
phase  of the  RI.  Section 7.0 presents a brief overview and conclusion of the DQO development for
the example  site.
                                                1-3

-------
INmATIONOF
  RI/FS
                            FIGURE  1-2
         PHASED RI/FS  APPROACH AND THE DQO PROCESS
                                  I-4

-------
STAGE 1 - IDENTIFY DECISION
                  TYPES

   STAGE 2 - IDENTIFY DATA
             USES/NEEDS

    STAGE 3 - DESIGN DATA
    COLLECTION PROGRAM

-------

-------
                    2.0  SUMMARY OF DQO  DEVELOPMENT  EXAMPLE
This section provides a brief summary of the example to give readers a perspective on the overall
process.  Quite sjmply, what is shown in the example is the process of (1) identifying the objectives of
the overall RI/FS and each of its components, (2) identifying the specific uses for which data must be
collected and the [data quality required for each use, and (3) developing a sampling and analytical plan
to meet the RI/FS objectives in the most efficient and effective manner possible.  To perform each  of the
above steps, the three-stage DQO development process is applied during the planning phase of the  RI/FS.
Figure 2-1 illustrates integration of the DQO process into the planning for the phased RI/FS  in this
example.

This summary is organized according to each of the DQO stages.  The detailed discussion of the example is
organized according to the RI/FS phases, showing how the DQO stages fit in to the normal sequence of
events for an RI/FS.

2.1  STAGE 1 - IDENTIFY DECISION TYPES

Stage 1 of the DQO process takes  place as part of RI/FS scoping.  Through interaction with data users and
evaluation  of existing information, a  conceptual model of the site is developed and objectives are set
for further data  collection and evaluation efforts (if needed) to meet remedial program goals.  Stage 1
activities are resumed at the completion of each RI phase to evaluate new data, refine or revise the
conceptual model as appropriate, and to set objectives for the subsequent phase.  Stage !  for the example
site is discussed in detail in Section 3.0.

At the completion of Stage 1 activities, a conceptual model of a site has been developed showing clear
evidence of contaminated soil and ground water. Potential contamination of private wells, which are
screened within  an unconfined aquifer,  presents a health threat to nearby residents who rely on the wells
for drinking water.  The general layout of the site is shown  in Figure 2-2.   The soil depression, where
discolored soils  show visual evidence of contamination, is approximately 200 ft by 200 ft.  Contaminants
of concern include TCE in the ground water and volatile organics and metals in the soil.  Contaminated
surface soils present a direct-contact threat.

The RI/FS will  assess the threat posed by the site.  FIT data are insufficient to determine fhe extent of
contamination at the site.  A phased  approach will  be used to first determine the boundaries of
contaminated soil and ground  water in Phase I and then to collect more extensive data through a
well-directed investigation in Phase II.  Phase I entails sampling of existing  wells  and soil gas to
determine  the boundaries  of the ground water plume and sampling of surface soils to determine the areal
extent of soil contamination.  Phase II activities will include installation and testing of additional
ground water wells and sampling of subsurface soils.

2.2  STAGE 2 - IDENTIFY DATA  USES/NEEDS

Stage 2 activities entail defining the quality and quantity of data that will be required to meet the
objectives set in Stage 1.  Definition of specific uses for data and attendant data quality requirements
lays the groundwork for a sound and efficient data collection program.  Data are required for risk
assessment, site characterization, evaluation of alternatives, and engineering design.
                                               2-1

-------
                           PHASE I Rl
                                                                                       PHASE II Rl
to
                     STAGE2
                     STAGES
                                   EBSTOO WELLS, SOL GAS.SURFACE SOLS
    •IDENTIFY DATA USES
   • IDENTIFY DATATYPES
 •DETERMNEDATAQUALITYNEEDS
• DETERMNE DATA QUANTITY NEEDS
  • EVALUATE SAMPLING AND
    ANALYTICAL OPTIONS
  • REVIEW PARCC PARAMETERS
   DESIGN DATACCUECnCN
       PROGRAM
                                       COLLECT ANOTABULATE DATA
                                                                                  STAGE 2
                                                                                  STAGES
                                                                                                  GROUND WATER, SUBSURFACE SOLS
    •IDENTIFY DATA USES
    • BENTIFY DATATYPES
• DETERMINE DATA OUALFTY NEEDS
• DETERMHE DATA QUANTITY NEEDS
  • EVALUATE SAMPLING AND
    ANALYTICAL OPTIONS
  • REVIEW PARCC PARAMETERS
   DESIGN DATACCLLECnON
       PROGRAM
                                                                                                    COLLECT AND TABULATE DATA
                                                                                                                                           STAGE 1
RERNEOR
REVISE
CONCEPTUAL
MODEL


HI
REPORT
                                                                      FIGURE  2-1
                             DQO  PROCESS WITHIN  PHASED  WORK FLOW FOR  EXAMPLE RI/FS

-------
                                               LAKE
 GLACIATED
 LOWLANDS
   SOIL
DEPRESSION
                         /Co)
         SITE BOUNDARY'
                                      RESIDENTIAL
                                       HOMES
      UPLAND
       AREA
NOT TO SCALE
                       FIGURE  2-2
              EXAMPLE SITE  LOCATION MAP
                          2-3

-------
Table 2-1 provides a summary of the results of Stage 2 activities for the example site.  Detailed
discussion of Stage 2 activities is included in the following sections:

     •   Section 4.0 - Overall RI/FS

     •   Section 5.1 - Phase I Ground Water

     •   Section 5.2 - Phase I Soil Gas

     •   Section 5.3 - Phase I Surface Soils

     «   Section 6.1 - Phase II Ground Water

     •   Section 6.2 - Phase II Subsurface Soils

2.3  STAGE 3 - DESIGN DATA COLLECTION PROGRAM

For Stage 3, a data collection program is designed to meet the requirements identified  in Stage 2.  Data
collection activities must be designed for each  media component of the phases. Tables 2-2 and 2-3
summarize plans for data collection activities.  Stage 3 activities are discussed in  detail in the
following sections:

     •   Section 5.4 - Phase I

     •   Section 6.3 - Phase II
                                               2-4

-------
                                             TABLE 2-1  DATA QUALITY SUMMARY
                                                                PHASE I
   Activity
to
Ul
   Objective
Prioritized Data Use(s)
    Appropriate  Analytical
    Levels
    Contaminants  of  Concern

    Level  of Concern



    Required Detection Limit

    Critical Samples
                            Sample Existing Wells
                                                                Soil Gas
                                                                                              Surface Soil
                            Samples from existing
                            wells will be used to'
                            determine if contaminants
                            are present in residential
                            wells and to obtain
                            information on the levels
                            of contaminants in on-site
                            monitoring wells.
Risk assessment
Site characterization
                             Risk  Assess.:   Ill,  IV, V
                             Site  Charac.:   I,  II,  III
                             TCE,  Arsenic,  Chromium,  Lead

                             5 ppb TCE/50 ppb metals



                             2 ppb TCE

                             Residential wells
                                Soil gas samples will be
                                taken and analyzed to
                                indicate the extent of
                                volatile orgariics in the
                                ground water.
Site characterization
Evaluation of
alternatives
                                 Site Char.:   II,  III,  IV
                                 Eval. Alt.:   II,  III,  IV
                                 TCE

                                 Not applicable



                                 5-10 ppb

                                 Two consecutive clean
                                 samples indicating the
                                 outer boundary of the
                                 plume
Surface soil samples will
be taken to assess the
ingestion threat
presented by lead,
arsenic, and chromium.
Samples will also be
taken to measure the
horizontal extent of
contaminants.

Risk assessment
Evaluation of
alternatives
Engineering Design

Risk Assess.:   Ill,  IV,  V
Eval. Alt.:  II,  III,  IV
Eng. Design:   II,  III,  IV

Arsenic,  chromium,  lead

Arsenic - 25 - 35 mg/kg
Lead  -  450 - 550  mg/kg
Chromium - 90  - 110 mg/kg

Low mg/kg range metals

Clean  samples  at  outer
boundary of contaminated
area

-------
                                               TABLE 2-1  DATA QUALITY SUMMARY
                                                         (continued)
                                                           PHASE II
                Activity
Ground Water
Subsurface Soils
                Objective
Kl
                Prioritized Data (Use(s)
                Appropriate Analytical
                Levels
                Contaminants of Concern
                Level of Concern
                Required Detection Limit
                Critical Samples
Ground water data are
required to evaluate the
extent of contamination,
develop a risk
assessment, and assess
potential remedial
alternatives.
Risk assessment
Evaluation of
alternatives
Risk Assess: III, IV, V
Eval. Alt.: II, III, IV
TCE, arsenic, chromium,
lead

5 ppb TCE/50 ppb metals
2 ppb TCE
Wells MW1 and MW2
Soil samples will be taken
and analyzed for VOAs and
metals to determine the
horizontal and vertical
extent of contaminants,
provide input to a risk
analysis, and provide
information necessary to
evaluate remedial
alternatives.

Risk assessment
Evaluation of
alternatives
Engineering Design

Risk Assess.: Ill, IV, V
Eval. Alt.: II, III, IV
Eng. Design: II, III, IV

TCE, arsenic, chromium,
lead

4-40 mg/kg TCE
Arsenic - 25 - 35 mg/kg
Lead - 450 - 550 mg/kg
Chromium - 90 - 110 mg/kg

2 mg/kg TCE
Low mg/kg range metals

Clean samples at boundaries
of contaminated area

-------
                                 TABLE 2-2  DATA COLLECTION PLAN SUMMARY PHASE I
   Activity
^  Sample Type

   Number of Samples
   QA/QC Samples
   Background Samples
                                                             PHASE I
Sample Existing Wells
Soil Gas
Grab

3 private wells
2 on-site wells

4 replicates (private)
4 matrix spike (private)
1 duplicate (on-site)
1 spike (on-site) •

Well OW2
Soil gas

49


Not applicable
3 samples 0.5 mi
south of depression
Surface Soil
Staff Requirements
Data Types
Field technicians
Chemi st
VOA
Metal s
Field technicians
Chemi st
VOA
Field technicians
Chemist
Metal s
Grab

89


60
See Section 5.3.5
for detail


4 (minimum)
   Sampling Procedures
Private wells sampled at
tap; on-site wells
sampled by bailer
Withdraw soil gas from
a hand-dug hole;
inject into detector
Obtain sample
from 0-2 in.
depth intervals

-------
                                  TABLE 2-2  DATA COLLECTION PLAN SUMMARY PHASE I
                                                    (continued)
    Activity
Sample Existing Wells
                                                              PHASE I
Soil Gas
Surface Soil
    Analytical  Methods/Equip.
    Level  I  Field  Screening
    Level  II Field Analysis
ts)
00   Level  III  Non-CLP
    Laboratory Methods
    Level  IV CLP  RAS  Methods
    Level  V  Nonstandard
    Methods
     GC/MS
     EPA Method 624 (VOA)
     Metals RAS
     Method 601/602 (VOA)
                                PID
                                Field GC with PID
                            X-ray fluorescence
                            AA, FAA,  ICAP
                            Metals RAS

-------
                                    TABLE 2-3  DATA COLLECTION PLAN SUMMARY  PHASE  II
                                                         PHASE  II
                   Activity
                             Ground Water
                             Subsurface Soils
to
Staff Requirements
Data Types
Hydrogeologist
Drillers
Chemist
pH
Conductivity
Geologist
Drillers
Chemist
VOA
Metals
Sample Type
Number of Samples
QA/QC Samples

Background Samples
Sampling Procedures

Analytical Methods/Equip.
Level I Field Screening
Level II Field Analysts
                   Level  III Non-CLP
                   Laboratory Methods
                   Level  IV CLP RAS Methods
Basic water quality
parameters
Permeability
Hydraulic head
Grab
5
1 duplicate
1 spike
1 background well
Well installation SOPs
                                                GC/PID - yolatiles only
                             GC/MS, FAA, ICAP
                             VOA
                             Metals RAS
Grab
72
18
(replicates sent to CLP)
2 per event
Standard split-spoon
sampling procedures
                             GC for volatiles
                             X-met for metals
                             GC/MS, AA, ICAP
                             RAS
                   Level V Nonstandard
                   Methods
                             Method 601/602 (VOA)

-------

-------
    IDENTIFY DECISION TYPES

 IDENTIFY AND INVOLVE DATA
                   USERS

       EVALUATE AVAILABLE
             INFORMATION
DESCRIBE CURRENT SITUATION
      REVIEW EXISTING DATA
  ASSESS ADEQUACY OF DATA

  DEVELOP CONCEPTUAL SITE
                   MODEL
        EVALUATION MODEL
          COMPUTER MODEL

   SPECIFY RI/FS OBJECTIVES

       DETERMINE NEED FOR
          ADDITIONAL DATA

-------

-------
                      3.0    DQO  STAGE 1 - RI/FS SCOPING PROCESS
3.1  IDENTIFY DECISION TYPES

Stage 1  of the DQO sequence is an inherent component of the RI/FS project scoping process and is
shown in Figure 3-1. As shown in Figure 1-2, Stage 1 is initiated during the RI/FS scoping process.

As the DQO (and RI/FS) process continues, the scoping of the project will become focused.  Stage 1
will be initiated whenever new data are evaluated or objectives/decisions must be redefined.
Subsequent to the initial RI/FS  scoping process (e.g., after the completion of Phase I), Stage 1 of
the DQO sequence is abbreviated in scope, and is focused mostly on the evaluation of newly acquired
data.  In cases where the field investigations have revealed a situation requiring a redefinition of
the objectives, the entire Stage 1 process may have to be repeated.

Stage 1  of the DQO process is undertaken to identify the decision makers and data users and to
involve them in the process of identifying the data requirements and decision types which will  have
to be made during the RI/FS.  This section outlines the process for performance of Stage 1 through
an example situation.  Detailed descriptions of the Stage 1 process are contained in the companion
document, Data Quality Objectives For Remedial Response Activities  - Development Process (Section
3.0).
                   i
For the example site,  the data available from previous investigations performed by EPA's Field
Investigation Team (FIT) contractor serve as the basis for scoping the RI/FS. The DQO process is
initiated upon receipt  of a work assignment which, in this case, will be undertaken as a federal-
lead  RI/FS.

3.2  IDENTIFY AND INVOLVE DATA  USERS

The  list of potential data users must be developed  at the outset of the DQO process.  The primary
data  users are those individuals involved in ongoing RI/FS activities.  For this site,  primary data
users are the EPA Remedial Project Manager (RPM), and the contractor's site manager and staff. The
site manager has the primary responsibility for incorporating DQOs into the planning and
implementation activities.  The  RPM and the site manager will work in a parallel fashion and be
continually involved with the technical staff through the course of the project.

The  initial list of decision makers and data users that will be involved in the example site are as
follows:

      »   Decision Maker:

             EPA Remedial Project Manager (RPM)

      •   Primary D^ata User:

         -  EPA RPM
             Contractor site manager
             Contractor personnel (hydrogeologist,
             analytical chemist, chemical engineer, water treatment
             engineer, and others)

Secondary data users include all individuals (or parties) that rely on RI/FS outputs to support
their programmatic activities.  Secondary data  users provide input to the  decision maker  (and
                                               3-1

-------
                           IDENTIFY & INVOLVE DATA USERS
   EVALUATE
AVAILABLE DATA
DEVELOP CONCEPTUAL MODEL

- CONTAMINANT SOURCES
- MIGRATION PATHWAYS
- POTENTIAL RECEPTORS
- CONTAMINANTS OF CONCERN
                           SPECIFY  OBJECTIVES/DECISIONS
                                   FIGURE  3-1
                            DQO STAGE 1 ELEMENTS

-------
primary data users) during the DQO development process through generic data needs and, on occasion,
site-specific data needs.  Secondary data users that may be included in this example site are listed
below:
                 L
     •   Secondary Data Users:

             EPA Enforcement Personnel (PRP determination)
             State Agency Personnel (remedy concurrence)
             Agency for Toxic Substances and  Disease Registry (ATSDR)
             (health assessment)
         -   Corps of Engineers  (RD/RA)

Other groups which may be involved in the RI/FS process include the following:

     •   Support group:

             BSD personnel (QA integrity)
             Office of Regional Counsel personnel  (compliance with
             policy)
             EPA HQ Personnel

Primary data users  will  attend an  RI/FS scoping meeting which will include a review of available
data and identification of data users'  needs.  Secondary data users are brought into the scoping
process as necessary.

3.3  EVALUATE AVAILABLE INFORMATION

In this step of the DQO process,  the existing information and available data for the example site
are compiled and Devaluated.  In addition, a reconnaissance level site visit is performed by the site
manager and appropriate staff to evaluate and confirm the available data, and thus develop an
objective assessment of current site conditions.

3.3.1  DESCRIBE  CURRENT SITUATION
                 t
The example site ;is located in a low-lying area with land forms created by glacial activities. A
number of homes in the vicinity of the site obtain water supplies from an unconfmed aquifer. A
plan of the site showing the location of the residences and general site configuration is presented
on Figure 3-2.  The site has been used for disposal of hazardous wastes.  During site reconnaissance
discolored soils were noted within a depression  measuring approximately 200 ft by 200 ft. Other
areas of the site did not show any evidence of disturbance, and the vegetative ground cover outside
the area of the depression did not display any sign of stress.  Review of time-sequential aerial
photographs and other information obtained  from EPA and state files confirmed the depression as  the
only area of waste disposal.  Based on interviews with witnesses, the depressed area has been the
site  of many disposal events.  It appears that material was  not disposed of in one particular
location; rather, material was disposed of randomly throughout the depressed area.  Based on this
information, several highly contaminated zones  are  expected  to be scattered throughout the depressed
area.  The site is presently unsecured and has unlimited access.

3.3.2     REVIEW EXISTING DATA

As part of the site investigation performed by the FIT contractor, a thorough field reconnaissance
of the entire site was performed.   A number of samples were obtained and submitted to the CLP for a
full  scale Hazardous  Substance List and metals  (HSL-1,2) analysis consisting of volatile and
semi-volatile organics, base/neutral and acid extractables, PCB/pesticides and inorganics (metals).

                 '                              3-3

-------
                                SITE BOUNDARY
     N
NOT TO SCALE
                         FIGURE  3-2
                          SITE  PLAN
                              3-4

-------
In addition, a photoionization detector (PID) was used to monitor the air for organic vapors over
the entire site. Samples taken include three near surface (12 inches deep) soil samples taken from
the discolored soil area; a ground water sample from each of two on-site monitoring wells; and three
samples from off-site private wells.  Contaminants of concern include TCE in ground water and
volatile organics and metals in the soils.  Figure 3-3 shows the FIT sampling locations and the site
boundary.

The samples were Analyzed by the CLP using RAS analytical methods and detection limits. Table 3-1
summarizes the data available for the example site. Soils contained metals and organics plus  a high
pH (about  10).  PID  screening did  not indicate any above-background organic concentrations in the
air with the exception of small (2 ppm) deflections inside the surface soil sample boreholes. The
data are insufficient to characterize  the site in  terms of the degree and extent (both horizontal
and vertical) of contamination and thus to support any potential remedial alternatives in the FS.
Air and surface water have not been identified as potential sources or pathways for contaminant
migration in the FIT  investigation.

3.3.3  ASSESS ADEQUACY OF DATA

An essential step in the evaluation of available information is determining the reliability and
acceptability of Unavailable data.  The data available for the example site were reviewed in terms
of methods of collection and analytical techniques. The documentation of sample collection
techniques is drawn from the site investigation report. Based on this review, the site manager
concludes that the site data are  both reliable and acceptable, but are insufficient to adequately
characterize the site.  Although no  contamination was reported for the residential well samples,
they were analyzed by CLP RAS methods which have a detection  limit (5 ppb) equal to the drinking
water level of concern for TCE and, therefore, the data may not be indicative of the actual health
risks.

3.4  DEVELOP CONCEPTUAL SITE MODEL

Based on available Information, a conceptual model is developed to provide an understanding  of the
sources of contaminants, the migration pathways of contaminants, and potential  receptors.   The
conceptual model is presented schematically at the initial meeting of the data users.  Conceptual
models can  include components from computer models, analytical models, graphic models, and/or other
techniques.   Conceptual models are discussed  in Section 3.3 of the DQO Development Process document.

The conceptual model of the example site is presented in Figure 3-4.  The conceptual model is
relatively simplistic; however, if additional data collection activities identify any complex
geologic features, these would be reflected in a more  complex conceptual model.  If necessary, a
series of models could be developed by media to identify contaminant migration pathways.

Under day-to-day circumstances the potential for contaminant release into the atmosphere  is
considered minimal.  A review  of meteorological  information obtained from a NOAA weather station
southwest of the site has shown the winds to be predominantly towards the west, away from the
residential areas.  Since there have  been  no documented odor complaints and the only elevated
organic vapor readings  measured by the site inspection team were at the discolored soil area,  the
air contaminant pathway is not considered significant  for either  organics or metals (dusts).

More information is required concerning the potential threat from direct contact with - and the
potential ingestion of -- surface soils contaminated with lead, arsenic and chromium.  The potential
for direct contact with (or exposure to) organics is assessed to be low since organics (detected on
site) tend to volatilize rapidly from surface soils: however, the potential for direct contact with
on-site soils contaminated with metals must be evaluated.
                                               3-5

-------
                                     SITE BOUNDARY
      N
  Legend

• Soil Sampling Locations
® Monitoring Well Location
i Private Well Locations
NOT TO SCALE

Note:  Distance from depression to private wells is approximately 1 mile.
                                FIGURE  3-3
         FIT  SITE  INVESTIGATION SAMPLING  LOCATIONS
                                    3-6

-------
                                        TABLE  3-1

                                      EXAMPLE  SITE
                               FIT SITE INVESTIGATION  DATA


VOLATILE ORGANICS (ppb)
Benzene
Trichloroethene (TCE)
Tetrachloroethene (PCE)
Toluene
Xylene
METALS (ppra)
Arsenic
Chromium
Lead

S-l

ND
47
ND
ND
ND

ND
ND
ND
SOILS
S-2

5
350
23
12
10

75
5,000
1,000

S-3

ND
ND
ND
ND
ND

ND
ND
ND
                                                     MONITORING
                                                        WELL
                                                     OW1   OW2
           PRIVATE WELLS
ND
52
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
                                                       ND
                                                       ND
                                                       ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
NOTES:  ND - Not detected
        Base Neutrals,  Acid Extractables  and  PCB/Pesticides were not
        detected in any sample.

-------
                   WOODED AREA
LAKE-,
                                                 REGIONAL
                                             GEOLOGIC CROSS SECTION
                                                      WOODED AREA
                     DEPRESSION
                                FILL MATERIAL
       (DIRECT CONTACT)
                       LACUSTRINE DEPOSITS
                    'GLACIAL TILL
                  SHALE BEDROCK
POTENTAL
CONTAMINANT
MIGRATION
PATHWAY
                                             SITE CROSS SECTION,
                         FIGURE  3-4
                 CONCEPTUAL SITE MODEL
                       (Looking South)
                              3-8

-------
Surface runoff is not considered a complete contaminant pathway since the area of discolored soil is
located within a depression and precipitation is low while infiltration is high.  The low topography
of the source area results in all runoff flowing into the soil depression.

The site  is located in an area which has been influenced by glacial activities.  The lake northeast
of the site at one time covered much of the  area.  The fine sediments laid down during periods of
glacial activity (lacustrine deposits) have low permeabilities and limit the rate at which water
percolates and migrates in these materials.  Directly beneath the site, however, glacial till
deposits are encountered (as shown in Figure 3-4).  These deposits are permeable and serve as a
source of potable water for homes in the area. Shale bedrock underlies the entire region.  The shale
contains  brackish ground water and is not a suitable source of potable water.

Regional ground water flow is towards the east. Water level measurements from the two on-site wells
indicate that the hydraulic gradient within the shallow unconfined  aquifer may cause ground water to
move in  an easterly  direction; however, the exact direction of the hydraulic gradient is unknown
until additional wells are installed.  The depth to the water table averages 15 ft. There is no
available information on the vertical hydraulic gradient.  Private residences are located
approximately 1 mile east of the discolored  source area.  Residents obtain drinking water from wells
screened in the  unconfined aquifer. No contamination has been found in these wells to date;
however, the potential  migration of contaminants to these wells poses a threat to human  health.

The soils in the depression are known to  contain elevated levels of heavy metals, specifically lead,
arsenic and chromium.  Based on the concentrations of metals detected during the FIT investigation,
and the soil characteristics (i.e., pH of 10 which will keep the metals  bound to the soils),  the
metals do not pose a major risk via the ground water pathway. The presence of volatile organics,
however, does present problems since migration to water supplies may occur.  The  presence of the
volatiles  in the monitoring well (OW-1) on site indicates that they are in the ground water
underlying the site and may migrate off site to the private water supply wells.
                  i
Based on the results of the FIT investigation, TCE (which occurs at the highest concentration  of the
organics) and the three metals will be considered contaminants of concern. Sampling during the
initial phase of the RI  will be used to confirm the presence of these contaminants  and indicate if
other potential contaminants are present.

The major pathway'for migration of contaminants from the site is the  unconfined  glacial till located
beneath the site. This glacial till also serves as the source for water supply wells  (receptors)
off site.  A secondary exposure pathway  is through direct contact with and ingestion of on-site
soils.

The site-specific conceptual model  identifies the following components:

     •    The  contaminated soil area is  a potential source of contaminants

     •    The  unconfined aquifer is the primary contaminant pathway

     •    The  private wells east of the site are potential receptors

     e    The  surface soils present a potential direct-contact pathway

Because of the limited amount of data available for the example site, the site manager and RPM
determined that a computer simulation model should not be developed at this  time.
                                                3-9

-------
3.5    SPECIFY RI/FS OBJECTIVES

The objective of an RI/FS is to determine the nature and extent of the threat posed by the release
or potential release of hazardous substances and to evaluate remedial alternatives to support Agency
decisions on the remedial action for the site. Achieving this broad objective requires that several
complicated and interrelated activities be performed, each having objectives, acceptable levels of
uncertainty, and attendant data quality requirements.  The expression of these objectives in clear
precise statements is the first step toward development of a cost-effective program for collection
of sufficient data for decision making.

In general, the objectives for this example site are the following:

     •  Determine the extent and concentration of soil and ground water contamination

     •  Determine if human receptors are at risk from the ingestion of contaminants

     •  Determine and evaluate feasible remedial alternatives

3.6    DETERMINE NEED FOR ADDITIONAL DATA

The available data for the example site has identified potential source materials on site,
contaminant  migration pathways, and potential receptors. The available data are not adequate
to complete the RI/FS or to support an RI/FS decision regarding site remediation.  Therefore,
collection of additional data is warranted and Stage 2 of the DQO process should be
initiated.

Based  on the information obtained from EPA and state files, the results of the FIT
investigations, and the conceptual model, the site manager and the RPM have decided that a
phased approach will be used for the collection of additional data.
                                               3-10

-------
  DQO STAGE 2 - IDENTIFY DATA
    USES AND NEEDS: OVERALL
                      RI/FS

      REMEDIAL ALTERNATIVES

      IDENTIFY DATA QUALITY /
             QUANTITY NEEDS

EVALUATE SAMPLING / ANALYSIS
                   OPTIONS

   REVIEW PARCC PARAMETERS

-------

-------
                        4.0   DQO  STAGE 2  -  RI/FS  DEVELOPMENT
In Stage I  for the example site, the basic decision making process for the RI/FS was identified.
The need for additional data to support the RI/FS decision was identified.  The conceptual model
developed in Stage I will serve as the basis for completion of the Stage 2 elements for the example
site.  In Stage 2 of the DQO process,  the information required for the example site will be
identified, the data quality and quantity required to support the RI/FS will be specified, and
appropriate sampling and analytical methods will be chosen.

Stage 2 is initially undertaken for the overall RI/FS. Once data uses and attendant data quality
needs are established for the overall site, the process will be  refined for the components of
individual phases.  At the completion of individual tasks, results are integrated into the
conceptual  model and data base for the entire site.  In this manner,  the iterative and interactive
DQO process  is incorporated in the RI/FS work flow.
                   I
The major  DQO Stage 2 elements are identified in Figure 4-1.  Although the elements shown on Figure
4-1 appear as  discrete units,  in practice they are part of an integrated thought process with a
feedback loop  operating to continuously refine each element.

4.1    DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  OVERALL RI/FS


Data developed during the RI will be used  for:

     •  Risk assessment

     9  Site characterization

     e  Screening and evaluation of remedial alternatives

     •  Remedial design

Table 4-1 summarizes  the overall RI/FS  data uses and needs.

Discolored soils in the depressed portion of the site indicate areas of contamination.  The  organic
contaminants in the|soil are suspected  to be leaching into the underlying unconfmed  aquifer. Thus,
the contaminants in the soil  may affect the  private wells east of the site.  In addition, high
levels of metals have been detected in the soils of the depressed area.

The potential for direct contact with contaminated soils exists.  The extent and magnitude  of soil
contamination  and the potential risks associated with direct contact and ingestion must be
addressed.   To adequately assess the risk presented by the soils, the  total area of contaminated
soils must be determined. This value will  be used in conjunction with the  action level determined
during the  risk assessment to determine an appropriate remedial action for  the site.

Ground water  is the'major pathway for migration of contaminants from the suspected sources to the
receptors.  Information on the movement and contaminant concentration  of the ground water is
therefore required. l
                                                4-1

-------
                      IDENTIFY
                     DATA USES
                    IDENTIFY DATA
                       TYPES
  IDENTIFY
DATA QUALITY
   NEEDS
   IDENTIFY
DATA QUANTITY
    NEEDS
                     EVALUATE
                SAMPLING/ANALYTICAL
                      OPTIONS
                    REVIEW PARCC
                    PARAMETERS
                  FIGURE  4-1
          DQO STAGE 2 ELEMENTS
                        4-2

-------
SITE
NAME.
LOCATION.
NUMBER—
PHASE	
                                                  TABLE  4-1
                                                  DATA USES
EPA REGION
      RI1  RI2 RI3 ERA FS  RD  RA
DATE	
CONTRACTOR—
SITE MANAGER ,
^^^ DATA USE
MEDIA ^S**v.
SOURCE SAMPLING
TYPE
SOIL SAMPLING
GROUND WATER SAMPLING
SURFACE WATER/SEDIMENT
SAMPLING
AIR SAMPLING
BIOLOGICAL SAMPLING
OTHER

SITE
CHARACTERIZATION
(INCLUDING
HEALTH &
SAFETY)







RISK
ASSESSMENT







EVALUATION OF
ALTERNATIVES







ENGNEERNG
DESIGN Of
ALTERNATIVES







MONITORING
DURING
REMEDIAL ACTION







PRP
DETERMINATION







OTHER








NOTE: CHECK APPROPRIATE BOX (ES)
                                                                                                         COM SFDQ01.001

-------
Consistent with the objectives of the RJ/FS as defined in Stage 1, data required to address the
overall RI/FS include:

     •   Data on the extent and magnitude of contaminants in the ground water and soils

     •   Data concerning the potential migration and timing of migration

     •   Data on the health  and environmental risk of ingestion of contaminated ground water and soils

     •   Data on the physical constraints associated with ground water/soil extraction and treatment

     •   Data 011 the physical and chemical properties of ground water and soil

     •   Data related to any residual or sidestream disposal requirements associated with ground water
         treatment and on-site soil remediation or removal/treatment

4.2   REMEDIAL ALTERNATIVES

The following potential remedial alternatives will be evaluated for the ground water as part of the
RI/FS:

     •   No action

     •   In situ treatment

     •   Hydraulic containment

     •   Physical containment

     •   Ground water extraction and treatment

     •   Alternate water supplies

For the ground  water extraction option, a number of treated effluent discharge alternatives will be
evaluated, including discharge to municipal sewer, deep well injection,  or discharge to infiltration
basins on site.  If contaminants are found above  levels of concern in drinking water wells,
alternate water supplies may be provided as an expedited response.

In addition  to the ground water pathway,  the direct contact pathway for contaminated soils will be
assessed. The potential remedial alternatives for the soils which will be evaluated as part of the
RI/FS include:

     •    No  action

     •    Excavation and on-site treatment

     •    Excavation and off-site treatment/disposal

     •    Cap (may also require a barrier such as a slurry wall)

     •    Enhanced volatilization

     •    Incineration
                                                 4-4

-------
4.3   IDENTIFY DATA TYPES
                  I
Data types required for site evaluation, risk assessment, and evaluation of the remedial
alternatives include both chemical and physical characteristics as well as the extent of
contamination. Table 4-2 summarizes the data types required to assess remedial alternatives.

The physical properties of the aquifer are important in evaluation of remedial alternatives which
involve ground water extraction or containment.  The physical properties of the aquifer and the
spatial data will be jutilized in determining the volume of the contaminated plume. Parameters which
influence the volume of contaminated ground water are the horizontal and vertical extent of
contaminants (i.e.,;a three-dimensional outline of the contaminant plume) and the porosity of the
aquifer.  In any remedial action involving pumping, the volume of water removed is expected to be at
least an order of magnitude greater than  the volume marked by the boundaries of the plume.
Effective porosity, grain size, and permeability data will also be obtained for the evaluation of
enhanced volatilization procedures.

The water quality parameters and the contaminants analysis (VOA and metals) obtained from  both the
private and newly installed monitoring wells will be used to determine the extent of ground water
contamination and jto evaluate the applicability of various treatability options.  Physical and
chemical data (typejs) will also be obtained for soils, and are required  for evaluating treatment and
disposal options.

4.4   IDENTIFY DATA QUALITY/QUANTITY NEEDS
The various tasks and phases of this remedial investigation will require different levels of data
quality/quantity.  The data quality/quantity needs for each specific task/phase will be discussed in
the following sections.

Data quality will be summarized for each medium within each phase in the following format:
       Prioritized Data Uses

       Appropriate Analytical Levels

       Contaminants of Concern

       Levels of Concern

       Required Detection Limit

       Critical Samples

 The Development Process manual provides a thorough description of these parameters in Section 4.0.
 Although not always addressed quantitatively,  precision and accuracy values for analytical methods
 are also used to asjsess data quality.

 4.5   EVALUATE SAMPLING/ANALYSIS OPTIONS
                  I

 Sampling and Analysis Components

 There are several options available for investigating potential ground water and soils contamination
 at the site.  The options are based on combinations of the following tasks:
                                                 4-5

-------
   TABLE  4-2
RI/FS DATA TYPES
DATA TYPES
A} PHYSICAL PARAMETER
PERMEABILITY
POROSITY
HYDRAULIC HEAD
GRAIN SIZE
STANDARD PENETRATION TEST
PARTICLE SIZE DISTRIBUTION
% ORGANIC CARBON
BTU CONTENT
B) WATER QUALITY PARAMETERS
Fa
Mn
PH
TDS
TOG
COD
TOX
HARDNESS
ALKALINITY
ORGANIC COLOR
FILTERED METALS
UNFILTERED METALS
C) CONTAMINANTS
VOLATILE ORGANICS
METALS
ORGANICS SCREENING
METALS SCREENING
D) SPATIAL DATA
HORIZONTAL EXTENT
VERTICAL EXTENT
GROUND WATER


xX





xX
xX-
xX
\s
xX
\s
xX
\s
^
\s
\s
\s
%/
\s


v/
vX
SOILS
^
\s

xX
xX
xX
xX
xX












^x
^/
^x
%x
^x
v^
       4-6

-------
      •  Existing well sampling

      •  Soil gas sampling

      •  Soil sampling

      •  Installing and sampling monitoring wells .

The two major types of contaminants of concern are volatile organics and metals.  Existing wells in
the vicinity of the site will be sampled to determine if contaminants are present.  If contaminated,
consideration must be given to implementation of an expedited response (i.e., supply an alternate
source of water).   :

Soil gas sampling can assist in delineating the boundaries of the ground water plume. The soil gas
evaluation will be conducted as a continuous field activity and completed prior to the installation
of any monitoring  \yells.

Monitoring wells will be installed based on the results of the soil gas sampling program. These
wells  will be used  tp evaluate the extent of ground water contamination and will serve as an early
warning system of contaminant migration towards the private wells.

To determine the range and extent of metals contamination, a surface soil sampling  program will be
initiated.  Data obtained from this analysis will be used to determine the extent of contamination
and to assist  in determining the level of more detailed vertical and horizontal  sampling.

Sampling and Analysis Approach

The RI at this example site is planned to proceed in a phased approach with the following tasks:

      PHASE IA -   Ground water sampling and analysis from  the five existing wells (three
                    residential and two on-site wells)

      PHASE IB - |  Soil gas sampling and field screening for VOAs

      PHASE 1C - !  Surface soil sampling and field analyses for metals

      PHASE HA  -   Monitoring well installation with soil and ground water sampling and analysis

      PHASE IIB  -   Subsurface soil sampling and analysis (may or may not be performed based on the
                     results of Phase I investigations)

Resource Requirements

Performance of theTield program will  require, at a minimum, a drilling crew, a geologist, and an
analytical chemist. \ The site manager must plan to have these personnel available throughout the
soil sampling phase.  Analytical equipment required includes a field GC, an X-ray fluorescense
metals analyzer, and analytical support from the CLP and/or other established laboratories.

4.6   REVIEW PARCC PARAMETERS

The PARCC (precison, accuracy,  representativeness, completeness and comparability) parameters are
overall indicators  of data quality and are defined in Section 4.0 of the Development Process manual.
As with data quality and quantity,  the PARCC parameters are specified at the phase and task level
                                                4-7

-------
and are not specified for the overall RI/FS.  Furthermore, PARCC parameters, specifically precision
and accuracy (where they are available), are compound, media, and method-specific.

The historical precision and accuracy achieved by different analytical techniques will be reviewed
for each task to allow a comparison of the analytical techniques.  In addition, representativeness,
completeness and comparability will also be reviewed and addressed.
                                               4-8

-------
         STAGE 2 - HI PHASE SA -
    SAMPLING ©F EXISTING WELLS
                   DATA USES
                   DATA TYPES
            DATA QUALITY NEEDS
          DATA QUANTITY NEEDS
    SAMPLING /ANALYSIS OPTIONS
            PARCC PARAMETERS

  STAGE 2 - RS PHASE IB - SOIL GAS
               INVESTIGATIONS
                   DATA USES
                   DATA TYPES
            DATA QUALITY NEEDS
          DATA QUANTITY NEEDS
    SAMPLING /ANALYSIS OPTIONS
            PARCC PARAMETERS

    STAGE 2 - PHASE 1C - SURFACE
           SOIL INVESTIGATIONS
                   DATA USES
                   DATA TYPES
            DATA QUALITY NEEDS
          DATA QUANTITY NEEDS
    SAMPLING /ANALYSIS OPTIONS
            PARCC PARAMETERS

         STAGE 3 - DESIGN DATA
  COLLECTION PROGRAM: PHASE I
       REMEDIAL INVESTIGATIONS
  DATA COLLECTION COMPONENTS
DATA COLLECTION DOCUMENTATION

         STAGE 1 - COLLECT AND
               EVALUATE DATA
     RI PHASE IA - EXISTING WELLS
           RI PHASE IB - SOIL GAS
       RI PHASE 1C - SURFACE SOIL

-------

-------
           5.0  iDQO DEVELOPMENT PHASE I REMEDIAL INVESTIGATIONS
5.1   DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  RI PHASE IA - SAMPLING
      EXISTING WELLS

5.1.1 IDENTIFY DATA USES:  RI PHASE IA - SAMPLING OF EXISTING WELLS

Residents living in the immediate vicinity of the site currently obtain potable water from the unconfined
aquifer.  Analysis! of ground water samples obtained from three private wells sampled during the FIT site
investigation indicated that no contamination was present in the private water supply wells.  However,
organics were detected in the on-site monitoring well, and the conceptual model for the site indicates
that a potential route of migration of contaminants is through the soil to the private wells tapping the
surficial aquifer,  'in addition, while metals were not detected in any of the wells, their presence in
the soil  samples indicates they should be a concern.  While the FIT samples were analyzed for the full
scale Hazardous Substance List and metals (HSL-1,2), only  one sample  (S-2) showed significant
contamination, with only organics and metals detected.  To determine if the private wells are
contaminated, samples will be obtained from each of the three homes located immediately east of the site.
In addition, the on-site monitoring wells will be sampled to confirm the results of the FIT  sampling. As
such, samples will be analyzed for the full scale HSL-1,2 compounds.  However for the sake of brevity,
the DQO process Iwill only be carried through  for organics and metals as described in the  remainder of
Section  5.0.      ;

High quality  samples from the private wells are critical because the risk  of making  a wrong decision
concerning continued use of the water supply has significant public health implications.

The  data needs for RI Phase IA have been identified as information concerning the presence or absence and
concentration of contaminants in the drinking water wells of nearby residents.  This information, will be
used to  perform a risk assessment by comparing the existing concentration to established action levels or
standards.

Data Use Categories

Ground water is the major pathway  for migration of contaminants from the suspected source to the
receptors.  Information about the movement and contaminant concentration of the ground water (at the
identified receptoO is therefore required.
                 |
As shown on Tables 5-1  and 5-2, data obtained from Phase IA ground water investigations will be used to
determine the presence and concentration of organics and metals contaminants in the three private and two
on-site monitoring wells.  This information can then be used to perform a risk analysis to  determine if a
health risk exists  Idue to ingestion of ground water.

5.1.2  IDENTIFY DATA TYPES:  RI PHASE IA - SAMPLING OF EXISTING WELLS

The data type required to evaluate the potential hazards associated with the ingestion of (potentially)
contaminated ground water is the presence and concentration of contaminants (i.e.  VOAs and metals).
                                               5-1.

-------
     SITE
     NAME	
     LOCATION.
     NUMBER-
     PHASE	
                                                       TABLE 5-1
                                                       DATA USES
EPA REGION
           RI1  RI2 RI3 ERA FS  RD  RA
DATE	
CONTRACTOR-
SITE MANAGER.
^V. DATA USE
MEDIA ^*s.
SOURCE SAMPLING
TYPE

SOIL SAMPLING
GROUND WATER SAMPLING
SURFACE WATER/SEDIMENT
SAMPLING
AIR SAMPLING
BIOLOGICAL SAMPLING
OTHER

SITE
CHARACTERIZATION
(INCLUDING
HEALTH &
SAFETY)







RISK
ASSESSMENT







EVALUATION OF
ALTERNATIVES







ENGNEERNG
DESIGN OF
ALTERNATIVES







MONITORING
DURNG
REMEDIAL ACTION







PRP
DETERMINATION







OTHER









-------
                                         TABLE  5-2
                                   DQO SUMMARY FORM
 1.  SITE
                                                                EPA
                                                                REGION
     LOCATION
     NUMBER
                                                                    RI2 RI3 ERA FS RD  RA
                                                                       (C1RCLEONE)
 2. MEDIA

    (COKiEONE)
                       SOL
                                              SW/SEO
                                                            AIR
                                                                                      OTHER
   USE
   (CIRCLEALL THAT
   APPLY)
                                          EVAL
                                          ALTS.
                           ENOG
                          DESIGN
 PHP
DETER
MONITORING
 REMEDIAL
  ACTION
OTHER
 4. OBJECTIVE
\A0JL5
                                                              TD
                                                                                        IF
  ARD "IP
                                               OAJ TJ-fg US\J&US
 5.  SITE INFORMATION
      GROLW WATERUSE
      SB^mVE RECEPTORS .
 6. DATA TYPES (CIRCLE APPROPRIATE DATA TYPES)
            A. ANALYTICAL DATA
     pH             PESTICIDES
     coNDUcnvrrY    peg
                     CYANIDE
       TOX
       TOC
       BTX
       COD
      TCLP
                                                                   B. PHYSICAL DATA

                                                           PERfceABILFTY      HYDRAULIC HEAD
                                                           POROSITY         PENETRATION TEST
                                                           GRAM SIZE        HARDNESS
                                                           BULKOENSfTY	
 7. SAMPLING METHOD (CIRCLEMETHOO(S) TOBEUSED)
                           ORE)
      SOUBCE
                                                          NON- INTRUSIVE
                                                         —r"=?--^
                                                          IMTRUSJVE,
                                                                              PHASED
 8. ANALYTICAL LEVELS (INDICATELEVEL(S) AND EQUIPMENTS METHODS)

     LEVEL 1  FIELD 3CHE&llhg3 - EQUIPMENT

     LEVEL 2  FIELD ANALYSIS - EQUIPMENT ___™__™____™__
    LEVELS  NON-CLP LABORATORY-METHODS

    LEVEL 4  CLP/RA3 - METHODS rASTP^S

    LEVEL NS
                                                jl SA
-------
5.1.3    IDENTIFY DATA QUALITY NEEDS:  RI PHASE IA - SAMPLING EXISTING WELLS

Primary Data Quality Factors
     Prioritized Data Uses:
     Appropriate Analytical Levels:
     Primary Contaminant of Concern:

     Level of Concern:

     Required Detection Level:

     Critical Samples:
Risk Assessment
Site Characterization

Risk Assessment: III, IV, V
Site Characterization: I, II, III

TCE

5 ppb TCE/50 ppb Lead, Chromium, Arsenic

2 ppb TCE

Residential Wells
The private wells were sampled during the FIT investigation and analyzed using CLP RAS procedures.  CLP
RAS methods have a detection  limit of 5 ppb for most volatile organic analytes (see Appendix B).
However, the level of concern  associated with TCE in drinking water is 5 ppb, which is equal to the CLP
RAS detection limit.  Since analytical procedures perform poorly near their detection limits, CLP RAS
methods will not be acceptable for samples obtained from the private wells. CLP SAS methods with lower
detection limits will be used.

The 5 ppb level of concern for TCE has been obtained from the proposed drinldng water standard under the
Safe Drinking Water Act as (proposed) maximum contaminant levels (MCLs). MCLs represent the allowable
lifetime exposure to the contaminant of concern for a 70 kg adult who is assumed to ingest two liters of
water per day.

For this phase of the RI, the three private wells are considered critical data points.

5.1.4    IDENTIFY DATA QUANTITY NEEDS:  RI PHASE IA - SAMPLING OF
         EXISTING WELLS

The existing on-site wells and the three private wells east of the site will be sampled once during Phase
IA. The locations  of these wells  are shown  in Figure 5-1.

5.1.5    EVALUATE SAMPLING/ANALYSIS OPTIONS:  RI PHASE IA - SAMPLING EXISTING WELLS

Sampling and Analysis Components

The three residential wells in the vicinity of  the site and the two on-site monitoring wells will be
sampled as part of  the Phase I  investigation.  If the three residential wells are contaminated,
consideration must be given to implementation of an expedited response  (i.e.,  supply an alternate source
of water).

Sampling  and Analysis Approach

Water from the private wells will be sampled directly from the tap. after allowing the water to run for
at least 5 minutes.  Samples will  be taken  from  the tap because there is no other access to these wells.
Any filtration or aeration devices  will be removed or bypassed before sampling.  Samples from the on-site
monitoring wells will be obtained with dedicated (stainless steel) bailers.  A minimum of 3-5 well
volumes of water will be evacuated from the  wells prior to obtaining samples.

                                             5-4

-------
                                 SITE BOUNDARY
                                                 Legend

                                              ® Soil Sampling Locations
                                              © Monitoring Well Location
                                              A Private Well Locations
NOT TO SCALE
                               FIGURE  5-1
           RESIDENTIAL AND ON-SITE WELL  LOCATIONS
                                 5-5

-------
The sample from the on-site monitoring wells (OW1  and OW2) will be analyzed for volatile organics via
method 624 because the associated detection limit of 5 ppb is sufficient and turnaround time is not an
issue for these samples.

The analytical method chosen for analyzing the samples  from the residential wells is method 601/602.
This method has a detection limit of less than 1  ppb  for the volatile compounds of concern.  Method
601/602 can be obtained through the CLP via the special analytical services (SAS) feature or through a
non-CLP laboratory.

Since residents are currently drinking well water, laboratory results are required within a short
turnaround time.  If the water is contaminated,  an alternate water supply  may be necessary.  For these
reasons, the IFB for the analyses of private well samples was issued with a turnaround time requirement
of two-weeks. A CLP laboratory was located where it could meet the provisions of this IFB under an SAS
request.  Samples from the private wells will  therefore be sent to this  laboratory.

While the FIT sampling program did not identify detectable concentrations of metals in any of the well
samples, the presence of lead, chromium and arsenic in  the soils justifies testing of all wells for
metals using CLP RAS methods.

The CLP RAS Contract Required Detection Limit (CRDL) of 5 ug/I,  10 ug/1 and  10 ug/1 (for lead, chromium
and arsenic) are deemed sufficient for these analyses  based on the MCL values of 50 ug/1 for the three
metals.

5.1.6    REVIEW PARCC PARAMETERS:  RI PHASE IA -  SAMPLING OF EXISTING WELLS

Because two analytical methods will  be used for the ground water samples,  two statements of the PARCC
goals are required.

Precision - Well OW1 and OW2 Samples

The samples obtained from OW1 and OW2 will be analyzed using CLP RAS methods.

The historical precision of CLP RAS analytical  methods  for the contaminants found in  soil sample S-2 (see
Table 3-1)  is shown below.  These historical  values are  known  from blind performance evaluation samples
(see Appendix A) and are presented in percent relative standard deviation  (%RSD). The method of
calculating %RSD is shown in Appendix A.
      Contaminant

      TCE
      PCE
      Benzene
      Toluene
      Lead
      Arsenic
      Chromium
Precision (% RSD)

         17
         13
         12
         14
         32
          9.4
          9.8
The methods by which these values were calculated are shown in Appendix A. Precision indicates the
average percent error  likely in a replicate measurement.  A numerical example demonstrating the use of
these precision values  is given in Section 6.4.1.  QC samples will be examined to determine the precision
which is actually achieved.
                                             5-6

-------
Accuracy - Well OW1  and OW2 Samples

The samples obtained from OW1 and OW2 will be analyzed for metals and volatile organics using CLP RAS
methods.  The accuracy of these methods is known from blind evaluation sample data (see Appendix A).  The
historical accuracy of the selected methods for the contaminants found in the FIT investigations is shown
below:
       Contaminant

       TCE
       PCE
       Benzene
       Toluene
       Lead
       Chromium
       Arsenic    ;
                                           Accuracy (% Bias)

                                                  -22.8
                                                  -42.5
                                                   -3.3
                                                  -23.3
                                                   -0.7
                                                   -2.6
                                                   -8.3
Accuracy, as expressed in percent bias, indicates the systematic error in an analytical method.  Negative
values indicate underestimation while positive values indicate overestimation.  The values reported for
TCE, for example, will be on average 22.8 percent less than the actual values.

QC samples will be analyzed to determine the actual accuracy  achieved.

Representativeness - Three to five well volumes will be purged before sampling the observation wells to
ensure, that standing water is removed from the wells and that  the samples are representative of ground
water quality.                                                                                 .

Completeness - The historical completeness achieved for CLP  RAS analyses is 80-85 percent. This
completeness range is acceptable because the observation well  samples are not critical samples.
                  i
Comparability - The use of standard, published sampling and analytical methods plus the use of the QC
samples described above will ensure data of known quality.  This data set can then be compared with any
other data of known quality.

Precision - Private Well Samples

The precision of method 601/602 is unknown and must be estimated. At the outset the final precision of
the method is unknown; however, sufficient QC samples will be collected to determine precision.

Precision will be  measured from replicate samples. The following formula will be used for precision as
defined by relative [percent difference (RPD):
                                            -  X
                                                    x 100
                            RPD =
                                        Xi  +X2
    where
              X1  is the concentration of replicate #1
              X2  is the concentration of replicate #2

              RPD = /2 RSD

Determining precision from a single pair of replicate samples is very inaccurate. To improve the
efficiency of determining precision, more than one pair of replicate samples should be taken and the
precision measures! should be averaged.  As the number of replicate samples increases, the certainty in

                                              5-7

-------
the estimated precision measure increases.  Since no information on the analytical method is available,
it is impossible to state the precision of the method before the samples are analyzed. Although the
certainty in the precision of the method cannot be stated ahead of time, it is known from basic
statistics that as the number of measurements (n) increases, the uncertainty  surrounding the average of
the measurements decreases as 1/n.  Given this relationship between  uncertainty and the number of
replicate samples, the reduction in uncertainty obtained from an additional replicate sample can be
determined.
            Number of Replicates
Reduction in Uncertainty
                      2

                      3
                      4

                      5
                      6
               50%
               3.4%

               24%

               20%
               17%
The preceding table indicates that for more than four replicates there are diminishing returns in the
reduction in uncertainty.  For this reason it is cost effective to analyze four replicate samples.

To obtain four replicate samples from the three private wells, one well will be sampled in triplicate
while the remaining two wells will be sampled in duplicate.

Accuracy - Private Well Samples

The private well samples will be analyzed via method 601/602.  Performance evaluation sample data is
unavailable for this method so the accuracy of the method is unknown.   To estimate the accuracy of the
method,  spiked samples must be analyzed by the laboratory.

A spiked sample (as discussed in Appendix B of the guidance manual) contains a known amount of an
analyte.  If the laboratory method consistently overestimates or underestimates the concentration of
spiked samples, the method contains a systematic error or, in statistical terms, the method is biased.
The accuracy of the method is a measure of this bias.

A measure of the accuracy (% bias) of the method is given by:

                                    Accuracy  =  R-S  x 100%

                                                   S
    where     S is the known concentration, and

               R is the value reported by the lab.
For this definition of accuracy, as the absolute value of the accuracy measure approaches zero, accuracy
increases.

To efficiently determine accuracy, several spiked samples must be submitted for analyses.  The accuracy
measure will be calculated on each of the spikes, and the average of the accuracy measures from each of
the spikes will be used as the accuracy of the method.  As shown in the section on precision when an

                                              5-8

-------
                f
average is used als an estimator, diminishing returns in uncertainty occur after four data points. For
this reason four spikes will be required to efficiently determine accuracy.
Ideally, samples would be spiked in the field. However, field spiking is very difficult, so the spiking
procedure will be performed in the laboratory. Spiking should be performed so that the spiked
concentration is 5 ppb.  This spiked value is specified so that the accuracy of the method can be
estimated near the level of concern.

Representativeness - Private Well Samples
             '   I
To ensure that samples are representative of the water consumed by the residents, samples will be taken
from kitchen taps|.  Taps  will be run for 5 minutes or until three well volumes have been removed prior to
sampling so that 'the sample is representative of the overall quality of the wel!. During sampling the
tap flow rate will be reduced so that the potential for volatilization is reduced.  Also, any filtration
or aeration devices will be bypassed or removed prior to sampling.

Completeness - Private Well Samples
                L
Because these samples are critical, validated analyses must be obtained from each of the private wells.
If validated analyses are not obtained from any of the three wells, an analytical chemist will be
consulted immediately to determine why validated analyses were not obtained.  Based on consultations with
the chemist the analytical method will be modified or an alternative method will be suggested and the
wells will  be resampled.

Comparability - private Well Samples

The  use of standard sampling, analytical and quality control procedures and the QC samples described
above will ensure known data quality and therefore comparability of the results with other data of known
quality.

5.2    DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  RI PHASE IB - SOIL GAS
       INVESTIGATIONS

5.2.1    IDENTIFY DATA USES:  RI PHASE IB SOIL GAS INVESTIGATIONS
                     _

To determine the approximate extent of the  ground water plume and soil gas concentrations, soil gas
sampling and analysis will be performed. The soil gas evaluation will aid in the selection of ground
water monitoring well locations. These wells will be installed in Phase II of the RI based on the
results of soil gafc sampling.

Data Use  Categories

Site characterization  is the major data use category for information derived from this phase.  A minor
secondary data  use category is engineering  screening of alternatives.
                I
5.2.2    IDENTIFY DATA TYPES: RI PHASE IB - SOIL GAS INVESTIGATIONS
Data types required to estimate the extent and concentration ranges of the contaminated ground water
plume, (and to a lesser extent to allow engineering evaluation of alternatives) include the chemical
analysis of the soil gas.  This analysis will in turn reflect the volatile organic constituents and
concentrations present in the soil and ground water (plume).  Soil gas analyses will be conducted for
benzene, TCE, PCE and 1,2 trans dichloroethane (DCE).  DCE is a degradation product of TCE.
                                               5-9

-------
5.2.3    IDENTIFY DATA QUALITY NEEDS:  RI PHASE IB - SOIL GAS INVESTIGATIONS

Data Quality Factors
     Prioritized Data Uses:
     Appropriate Analytical Level:
     Contaminants of Concern:

     Level of Concern:

     Required Detection Limit:

     Critical Samples:
Site Characterization
Evaluation of Alternatives

Site Characterization:  I, II, III
Evaluation of Alternatives:  II, III, IV

TCE

Not applicable

5 to 10 ppb

Two consecutive clean (i.e., representative of background
conditions) samples indicating the outer boundary of plume.
Soil gas sampling and analysis results will indicate volatile organic concentrations in the soil pore
spaces in the vadose (unsaturated) zone.  Since health effects associated with the ingestion of
contaminated ground water can occur when organics are in the low ppb range (see Section 5.1.3), analysis
of soil gases will be required with a detection limit in the low ppb range.

Critical data points for the soil gas sampling task are the samples taken at the outer  boundary of the
plume (as defined by two consecutive clean  samples), since these samples will define the extent of
contamination.

5.2.4    IDENTIFY  DATA QUANTITY NEEDS:  RI PHASE IB - SOIL GAS INVESTIGATIONS

Factors

Soil gas samples will be obtained in the field at the locations identified in  Figure 5-2. Sampling  will
begin at the suspected source  and  continue in a direction moving away from the source.  Thus, the primary
factor influencing the  number of soil gas samples taken is  the areal extent of the soil gas plume.  Since
the extent of the plume is unknown at this time, it is impossible to predict the number of required soil
gas samples.

Number of Samples

Soil gas samples will be taken  on  a regular  grid to maximize the representativeness of the samples.  A
sampling grid will be  used to  provide coverage over the entire 200-ft-by-200-ft discolored area. A
number of important factors are considered  in determining the grid spacing and hence the number of
samples.   These factors include the technical objectives, schedule, costs, the size of the  site, and the
conceptual model.

The grid size will be chosen based on these factors and the goals of the soil gas sampling task. The
soil gas plume is related to the extent of the ground water  plume.  It is known from the FIT team
sampling that TCE is  present  in the ground water at  approximately 50 ug/1 near well OW1.  Based on the
vapor pressure of TCE and the detection limits of the proposed analytical  methods, the detectable soil
gas plume will extend beyond well OWL Well OW1 is approximately 300 ft  from the center of the depressed
area,  so the soil gas plume is expected to extend a minimum of 300 ft east of well OW1.
                                              5-10

-------
    •    •
                                            •    •   •
    •    •
                                            •    •   •
    •  !  •
    •  I  •
                                               WELLOW1
                                                            N
                                                     \
                          LIMIT OF
                       DEPRESSED AREA
                                                        SOIL GAS
                                                     SAMPLING LOCATION
SCALE (FT)
           200
                          FIGURE   5-2
           INITIAL  SOIL-GAS  SAMPLING  GRID
                           5-11

-------
The previous analysis provides a guide for determining a grid size.  East of the depressed area soil gas
contamination is likely to be found; therefore, only a coarse grid is  required between the depressed area
and well OWL  East of well OW1 a tighter grid is necessary to accurately determine the extent of the
soil gas plume.  Between the center of the depressed area and well OW1, a grid with 150-ft spacing in the
east-west direction is required. This spacing will provide several samples within the depressed area
which can be used to confirm the type and concentration of contaminants found in the soil-gas plume.
East of well OW1 a 50-ft grid will be used.  This grid size provides a maximum error of 50 ft in the
determination of the soil gas plume.  This error is deemed acceptable for this study.

North, south, and west of the depressed area, the extent of ground water contamination is unknown, so
specific site information cannot be used to set the grid size.  It is assumed that soil-gas contamination
will be found beneath the depressed area, hence it is not necessary to sample soil gas in the depressed
area.  Beyond the depressed area,  a 50-ft grid will be used to provide definition of the plume boundary.
The initial soil gas grid is shown in Figure 5-2.

To determine the maximum soil gas plume extent, samples will be taken following the established grid
space system and extending outwards at 50-ft intervals until two consecutive clean samples are obtained.
For purposes of the Phase  IB investigations,  two clean samples will define the extent of contamination.

In addition to the locations listed above, three samples  (not shown on Figure 5-2) will be taken 1/2 mile
south of the discolored  area to determine background soil gas conditions.  Additional sample locations
will be determined based on the results obtained at these initial locations.

5.2.5     EVALUATE SAMPLING/ANALYSIS OPTIONS:  RI PHASE IB - SOIL GAS
          INVESTIGATIONS

Sampling and Analysis  Approach

Two readily available analytical methods for volatile organic determination are suggested: an HNu 101,
which is a photoionization  detector (FID); and an HNu 301, which is a field gas chromatograph (GC).  Each
method has advantages  and disadvantages.  The HNu 101 is a hand-held, direct read-out, field instrument
calibrated to detect benzene in the high ppb to low ppm range.  Results are qualitative and provide a
determination of total organic volatiles. The HNu 101 cannot be used to distinguish between individual
organic fractions.  Analyses are, however, easy to obtain on a real-time basis.

The HNu 301  is a field GC and combines the capabilities of a (PID) and a GC. This unit can provide a
qualitative and  quantitive measure  of the contaminants present, with a detection limit of appoximately 5
ppb (for benzene).  The HNu 301 utilizes the PID to "detect" organic compounds and  the GC to separate the
individual organic fractions.  A trained operator is required to accurately operate the GC.  Based on the
overall objectives of the Phase IB effort, a quantitative evaluation of the contaminants is required. In
order to characterize the composition of the soil gas plume, a GC will be used to analyze all soil gas
samples.

This system will provide Level II analytical support.  Prior to use of the GC in the field, appropriate
columns, detectors, temperatures,  and flow rates will be selected to  ensure adequate component
separation.  Standard curves will be prepared for the soil gas in an analytical  laboratory prior to
initiating field work.  The  curves will be developed by spiking known quantities of the target volatile
compounds in non-contaminated background samples representative of the physical nature of the soil
on-site.  The spiked samples will be run to develop volatile gas concentration  chromatograms.

At each sampling point, a soil bucket auger will be used to excavate to a depth of 2 ft.  A polyethylene
pipe will be used as a sample probe and placed into the auger hole.  The pipe will be connected to a
portable air sampling pump by means of tygon tubing.  The pump  will  be used to purge gases from the soil
at a rate of 2 liters-per-minute for  a period of 5 minutes to allow for equilibration of soil and tubing

                                               5-12

-------
gas concentrations.  The sample (5 ml) will then be withdrawn from the sample probe by syringe and
injected into the GC;  The amount of sample injected, column temperature, sample location,  identification
number, and point of injection will be recorded on  the strip chart.  The GC will be allowed to run long
enough for the complete sample to elute from the system.  Soil gas sampling will continue until the
extent of the contaminant plume has been determined.

The DQO process for this phase is summarized in Table 5-3.

5.2.6    REVIEW  PARCC PARAMETERS:  RI PHASE IB - SOIL GAS INVESTIGATIONS

Level II analysis of the contaminants of concern using a field GC has been selected.  Almost no
historical data on the precision and accuracy achieved by this analytical technique exist. Fortunately,
the numerical precision  and accuracy of each measurement are not a serious concern for this site
characterization effort since the analyses will be used solely to  determine the presence or absence of
contaminants.

Precision - No specific requirement.

Accuracy - No specific requirement.

Note:  Since the objectives of Phase IB do not require that quantitative information be obtained, no
specific QA/QC samples will be analyzed to determine precision and accuracy.  However,  the field GC will
be calibrated  daily to known standards to ensure proper instrument  operation.

Representativeness - ;A sampling grid has been designed to obtain a representative picture of the soil gas
plume.
                                                                          •
Completeness - Completeness  of 100 percent will be achieved for the critical samples i.e.,  the two
consecutive clean samples at each  edge of the plume. This  will be accomplished by analyzing any clean
samples in duplicate!

Comparability - The use of standard operating procedures should ensure comparability of the results.

5.3   DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  RI PHASE 1C - SURFACE
      SOIL INVESTIGATION
                   i
5.3.1    IDENTIFY DATA USES: RI PHASE 1C - SURFACE SOIL INVESTIGATIONS

Analytical results frojm the FIT investigations indicated that the soils were contaminated with heavy
metals (sludges) -- specifically lead, chromium and arsenic.  Because the site is not secure, nearby
residents (especially phildren)  could have direct contact with these soils. The most important exposure
route is through direct ingestion of soils.

Data from this sampling phase will be  used to determine the expected and  likely worst-case exposures from
ingestion of the surface soils in the depressed area.  The exposures  can be input to a simple exposure
assessment model to determine the magnitude of the direct contact threat.
The data from this p
lase will also be used to determine the lateral extent of surface soil contamination.
This information will be used to design a subsurface soil sampling program (if required based on Phase I
information).  Ultimately the information gathered in this phase will be combined with the subsurface
information if gathered in a later phase to determine the total volume of soils contaminated by heavy
metals.
                                             5-13

-------
                             TABLE  5-3
                        DQO SUMMARY FORM
1. SITE
inriATWj
KIMWH

2. MEDIA gQ^ QW sw
(CRCLEONE)
3. USE r^sTfi~\ RISK /'EVALX
(CIRCLEALLTHAT sgHARAC.^ ASSESS. ( ALTsJ
APPLY) (H&sTl ^— r-'
4. OBJECTIVE -S>0\U G./VS Sh'CYxV'USS IOVU
fO Ip^ t>VCATlS TVrt£^ SXrnST^T
EPA
REGION
PHA?E

(ST?) Rl 2 Rl 3 ERA FS RD RA
(CIRCLEONE)
/SED AIR BIO
ENGG PRP MONJ
DESIGN DETER REK
AC

TC»WG OJHH,
•EDIAL
TION 	
-"%& TAKSI^ AND P»K>(X<_VZ-eD
OT= VouATiuer c>ite
AtCllCS INi
TrtPT <Of^C> WATEI^

5. SITE INFORMATION
GROUND WATER USE P^lT^V^^t* ^^^F^:,
SOI TYPES (?L$*£ \pcL_ T\<— U- DEPTH O-
SFMBim/PBFCPPmRsR E5 UDeWTS LlU/Ate t

i. DATA VtPES (CIRCLE APPROPRIATE DATA TYPES)
A. ANALYTICAL DATA
pH PESTICIDES TOX
CONDUCTIVITY PCS TOO
CSgi£} METALS BTX
ABN CYANIDE COD
TCLP

7. SAMPLING METHOD (CIRCLE METHODfS) TO BE USED)
<^§WIRONMENTA£> BIASED CGRAB^?
SOURCE ^GTO) COMPOSITE
-30 -S-V ; <5\\^ue -PeTTW
C^ L \^i£. CA-ST (3P1 "TVs

"*)O- > KiOW-
e. •SITS

B. PHYSICAL DATA
PERMEABILITY HYDRAULIC HEAD
POROSITY PENETRATONTEST
GRAIN SIZE HARDNESS
BULK DENSITY



NON- INTRUSIVE
PHASED

8. ANALYTICAL LEVELS (INDICATE LEVEL(S) AND EQUIPMENT S METHODS)
LEVB_ 1 RELD SCREBJM3 - EQUIPfUENT
LEVFL? RH D ANAi,yais - EQUIPMENT T^l.^L-D &C- — i^AL-t R£./X"TSTO E^tift. *-\ \fc>A-S
LEVELS NON-CLP LABORATORY -METHODS


LEVEL 4 CLP/RAS - METHODS 	
LB/H_ MS NON STANDARD,.. 	 	 	 	





6. SAMPLING PROCEDURES
BACKGROUND -2 PER EVENT OH 'Z, "^t^SUSSy /2- WWL^" SOUT)4 dP ^ePRKS^OAJ
CRITICAL (LIST) 2 CL&P&3 SWNN?!!^2^ / K)^\r ATI f\) b TPL.Urii6-
VtoLlS"

10. QUALITY CONTROL SAMPLES (CONFIRM OR SET STANDARt
A. RELD ^-v a LA
rni i nnATpra^s%y>B 	 _ 	 	 	 QPA
REPLICATE- ^S?pDR ......... 	 	 	 F1EPI
FIFinRIANK-VS%OH N/^ "AT
TRIP BLANK - 1 PER DAY OH I^Pi OIH
11. BUDGET REQUIREMENTS
BUDGET 3| \ i~J^O srwqxjLE ifr-k

VO\TH oN6 'PeW-'botO ^>&o&ciMt,  (A-ETA-t
COMTRACTOR , PRIME
SITE MANAGER

V
BORATORY
3ENT BLANK - 1 PER ANALYSIS BATC
LICATE - 1 PER ANALYSIS BATCf
RIX SPIKE - 1 PER ANALYSIS BATC
ERtAtuy c»TF

t^A
jnn tO A
HOR ]0 A-
^mi?L-&-S

•rTj^^Oi^fYVrrrf \K J

eV-TOfC



FOR DETAILS SEE SAMPLING & ANALYSIS PLAN
                                     5-14
                                                                   CDM SF DQO 1.002

-------
5.3.2    IDENTIFY DATA TYPES: RI PHASE 1C - SURFACE SOIL INVESTIGATIONS

In this phase, surface contamination  is the prime focus; therefore, only contaminants which are stable
near the surface are of interest.  Since the previous analyses failed to detect pesticides or other
non-volatile contaminants, only heavy metals will be considered.  As discussed in the conceptual model
(Section 3.4.1) organic contaminants are not a source of concern for direct contact exposure as they tend
to volatilize or migrate downward rapidly from surface soils.
                                        '

To assess the magnitude of the potential  threat associated with direct ingestion of (metals) contaminated
soils, surface soil samples will be collected within the depressed area and analyzed for heavy metal
concentrations—specifically arsenic, lead, and chromium since only these three metals were detected
during previous FIT sampling.
5.3.3    IDENTIFY DATA QUALITY NEEDS:
         INVESTIGATIONS
   RI PHASE 1C - SURFACE SOIL
Data Quality Factofs

     Prioritized Data Uses:
     Appropriate Analytical Levels:
     Contaminants of Concern:
     Levels of Concern:
     Required Detection Limits:
     Critical Samples:
Risk Assessment
Evaluation of Alternatives
Engineering Design

Risk Assessment: III, IV, V
Evaluation of Alternatives:  II, III, IV
Engineering Design: II, III, IV

As, Cr, Pb

As - 25 to 35 mg/kg
Pb - 450 to 550 mg/kg
Cr - 90 to 110 mg/kg

Given the high cleanup levels anticipated, detection limits in
the  low mg/kg range will be acceptable.

Clean samples at outer boundary of contaminated area.
The levels of concern shown above are typical of those used at past sites.  These values are based on the
health effects associated with ingestion of contaminated soil.  Since the metal contaminants at this site
are not expected to migrate to the ground water, ingestion is the only major route of exposure.  Thus,
the cleanup levels shown are representative of the actual cleanup level which will be determined after
the data have been collected and an exposure assessment has  been  performed.

To assess the direct contact threat posed by soils contaminated by heavy metals, quantitative information
on the concentrations of metals  present must be obtained.
5.3.4    IDENTIFY DATA QUANTITY NEEDS:
         INVESTIGATIONS
    RI PHASE 1C - SURFACE SOIL
To provide a representative and unbiased measure of the surface metals concentrations within the
depressed area, samples will be taken from the depth interval 0-2 in.  The depth interval 0-2 in. is
chosen because a child is most likely to  ingest soil from this interval.  Samples  will be located on  a
regular two-dimensional grid. As discussed in Appendix C of the Development Process manual, sampling on a
regular grid will provide representative samples and  will minimize bias.  The grid size is directly
                                              5-15

-------
related to the number of samples so the grid size must be based on the acceptable uncertainty in the
results, the spatial variability of the contaminants, and the cost of acquiring and analyzing a sample.

The spatial variability of contaminants is a measurement of how contaminants vary as a function of
location.  If contaminant concentrations vary radically over  short distances, spatial  variability is
high.  If contaminant concentrations do not vary radically throughout the site,, spatial variability is
low.  When spatial variability is high, unexpectedly large concentrations of contaminants can occur in
regions where contaminant levels are otherwise low. In areas with high spatial  variability, a large
number of samples are required to ensure, with reasonable confidence, that the majority of the highly
contaminated zones are sampled.  If the grid spacing is sufficiently small to adequately measure the
spatial variability of the contaminants, it will also be sufficient to determine the  uncertainty in the
results.

The depressed area has been the site of many disposal events.  Based on interviews with witnesses to the
disposal it appears that material was dumped not in one particular location, but  randomly throughout the
depressed area.  Based on this information, several  highly contaminated zones are expected to be
scattered throughout the depressed  area.  In other terms, the spatial variability of the contaminants is
expected to be high.  This information  indicates that a tight grid will be required to assess the spatial
variability  and the risk.

To determine an adequate grid size a meeting was held between contractor and EPA personnel. The various
expected properties of the contaminants were discussed.  Major topics were the  spatial variability of the
contaminants, the threat posed by the contaminants, and the cost of obtaining the samples.  Because
spatial variability is expected to be  high, a large number of samples were planned.  The grid spacing was
chosen based on perceived threat.  It was determined that the largest highly contaminated area  which
would be acceptable to miss was  100 sq ft.  Based on this assessment a 10-ft grid was chosen (see
Appendix A of the Development Process manual).  This grid spacing yields a total of approximately 400
samples over the 200-ft-by-200-ft depressed area.

Based on the conceptual model and the assumed threat, a large number (400) of samples  have  been proposed
to assess surface soil  contamination.  The large number of proposed samples is  based on an assumed
conceptual model.  If the conceptual model is incorrect and contaminants are actually found in large
continuous zones, far fewer than 400 samples will be required.  To validate the conceptual model, surface
soil data will be evaluated after approximately 90 samples have been collected and analyzed.

To assess the validity of the existing conceptual model, data must be collected on a  tight grid (10 ft);
however, at this point there is no justification for sampling the entire depressed  area at this density.
One method for assessing the validity of the conceptual model without sampling  the entire site on a 10-ft
grid is to sample on a hybrid grid.  The hybrid grid consists of 25  samples located on a 50-ft grid and
64 samples located on a  10-ft grid. The 64 closely spaced samples are  split into 16 groups of 4 samples.
One group of four is  then located at the center of each of the 50-ft cells defined by the data located on
the 50-ft grid.  The hybrid grid is shown  in Figure 5-3.

The two components of the hybrid  grid will provide different information concerning the spatial
variability of the contaminants.  The samples from the 50-ft grid will provide information  over  the
entire depressed area. This information will be used to determine the mean concentration of surface
contaminants within the depressed area and to assess the variability (correlation) of the data  at large
distances.  The samples  from  the 10-ft grid will be used to assess the validity of the conceptual model
and to determine the variability of the data at short distances.

Based on the evaluation of this initial data set, the conceptual model might be revised.  This revised
conceptual model will  be used in determining additional data requirements.
                                               5-16

-------
                                          LIMITS OF THE
                                         DEPRESSED AREA
                                                       N
                            50
                   FIGURE  5-3
HYBRID GRID FOR SAMPLING THE DEPRESSED AREA
       \        (89 DATA TOTAL)
                        5-17

-------
As part of the data analysis the uncertainty surrounding the estimate of the mean contaminant
concentration within the depressed area will be determined. Determining this uncertainty requires
knowledge of the spatial variability of the data as a function of distance (see Appendix A Development
Process manual).  To model spatial variability, data separated by small and large distances are required;
thus, the  hybrid grid is ideal for determining spatial variability as a function of distance.
5.3.5       EVALUATE SAMPLING/ANALYSIS OPTIONS:
            INVESTIGATIONS
                                    RI PHASE 1C - SURFACE SOIL
Analysis options include CLP,  local laboratory, and on-site analysis.  Each type of analyses has certain
properties which are presented below.
Analytical Method
      CLP/RAS
Turnaround Time
    6 weeks
Cost per Sample


        $60b
Relative Accuracy
And Precision
        High
      Local Lab/SASe
    2-7 days
        $80°
        High
      On-Site Analysis
    2-24 hours
        $8°
        Unknown
     a.  Time includes data validation.
     b.  Cost is for paperwork and shipping only.  No lab cost is included.
     c.  Atomic absorption, acid digestion analysis cost.  Includes paperwork cost.
     d.  Cost is for sample preparation and analysis labor only.
     e.  Costs for SAS are similar to b above.

Because future phases of this study depend on the results of this phase, the turnaround time of an
analytical method is a critical issue.  If the CLP is used to analyze these samples, project delays may
be unavoidable. Both the local lab and on-site analysis provide adequate turnaround times, however,
on-site analysis is  10 times less expensive than the local lab.  Thus, on-site analysis would allow 10
times more samples to be analyzed at the same cost as  local lab analysis.

Based on known site history and the conceptual model  (see Section 5.3.4)  a large number of samples will
be required to characterize the extent of contamination. The only available analytical method which can
be used to  analyze a large number of samples for an acceptable cost is on-site analysis.  The on-site
analysis method of choice is X-ray fluorescence using a Columbia Scientific X-Met 840 (X-Met) or similar
instrument.

Ordinarily on-site  (Level II) analysis would not be suitable for risk assessment uses.  However, in this
case, a rigorous field calibration procedure with off-site laboratory verification of the calibration
standards will be used.  Also, a large number of QC samples will be analyzed to estimate precision and
accuracy.  The resulting data will be statistically reviewed  and. if  the field  data are judged
unreliable, the  soil samples will be sent to an off-site laboratory for analysis.  The use of these
procedures makes  this field analysis more like a Level  III  analysis and, therefore, suitable for risk
assessment uses.

Experience with the X-Met at previous sites indicates that the detection limit of the X-Met ranges from 2
to 200 mg/kg.  Based on these values the X-Met might not provide adequate detection  limits; however,
consultations with  experts on the method indicate that there is high likelihood  that the X-Met will
provide detection limits less than 20 mg/kg.   To allow for  the possibility that the detection limits of
                                               5-18

-------
the X-Met will be ajbove the action levels, 8 ounces of soil will be retained from each sample.  The
retained soil will, be) sent to a local laboratory if unacceptably high X-Met detection limits are found.

Because the accuracy and  precision of this instrument are unknown,  sufficient QC samples must be
analyzed to assess accuracy and precision.  The procedure for assessing the accuracy and precision of the
X-Met and the number  of QC samples required are discussed below.
                   I
The X-Met will be calibrated using the lead, arsenic, and chromium concentrations from four on-site
surficial soil samples.  The four calibration samples will be taken along a radial line stretching from
the center to the edge of the discolored area as shown in Figure 5-4. The four calibration  samples will
be equally spaced along this line to ensure that these  samples span the range of metal concentrations
occurring within the discolored area.  The actual metal concentrations will be determined by laboratory
analyses.           j
                                              ,
The four calibration! samples will contain 40 ounces of soil each. The soil will be homogenized and then
split into seven samples. Two of the splits will be sent to the CLP, four of the splits will be sent to
a local lab and the remaining sample will be retained. The method of splitting the calibration samples
is shown in Figure  5-5. The rationale for this replicate analysis procedure is presented below.

Sixteen samples (four replicates of each of the four calibration samples) will be sent to a local lab
with an in-place QA/QC program.  The results for these samples will be obtained within 1  week of
submission to the lab. The  concentration values for each-set of four replicate samples will  be averaged
to obtain the estimated concentration of each calibration sample.  The X-Met will then be calibrated
using these four average values.
                             '
The average value, of four replicate samples is used to calibrate the X-Met because there will be high
confidence in the ayerage of these four values. The exact degree of confidence is unknown until the data
values are obtained.  It  is  known however, that uncertainty surrounding the average of four replicate
measurements is four times  less than the uncertainty  surrounding a single measurement. This four times
reduction in uncertainty is deemed acceptable for this study by the site manager with approval by the
RPM.

After  the X-Met is calibrated using the four calibration samples, analysis of actual samples  can proceed.
During analysis of actual samples, each of the four calibration samples must be run daily using the
X-Met. Each calibration sample must be run a minimum of 15 times during this RI  phase to estimate the
distribution of error? for each calibration sample.  Calibration samples will be analyzed at various
times  during the day to  assess any temporal variability in the measurements.  A minimum of 15 samples has
been chosen becausp work at previous sites has shown that the X-Met can produce unreliable results.  The
reliability of the X-Met  will  be assessed by examining the distribution of the errors for each of the
calibration samples.
termed reliable and
 If these error distributions are symmetric and well behaved, the X-Met will be
the accuracy and precision of the procedure will be defined.  If however, the
distribution of errors are highly skewed or otherwise ill behaved, the X-Met will be considered
unreliable and the numerical values given by the X-Met cannot be used.  In the event that the X-Met is
judged unreliable, s
-------
                                       LIMITS OF THE
                                      DEPRESSED AREA
      CALIBRATION SAMPLE
              #1
                      .#2
N
                                         #4
             SCALE (FT)
              FIGURE 5-4
LOCATION  OF CALIBRATION SAMPLES
                5-20

-------
o

I
u
UJ
oe
u
UJ
Q


g


a
O
CALIBRATION
: SAMPLE

   #3
                                                   2 - 8 OZ SAMPLES TO CLP
                                                   4 - 4 oz SAMPLES TO LOCAL LAB
                                                   1 -8 oz SAMPLE RETAINED
                                                  2 - 8 OZ SAMPLES TO CLP
                                        4 - 4 oz SAMPLES TO LOCAL LAB
                                                  1 -8 OZ SAMPLE RETAINED
                                         2 - 8 OZ SAMPLES TO CLP
                                                   4 - 4 OZ SAMPLES TO LOCAL LAB
                                                   1 -8 oz SAMPLE RETAINED
                                                   2 - 8 oz SAMPLES TO CLP
                                                   4 - 4 oz SAMPLES TO LOCAL LAB
                                                   1 -8 OZ SAMPLE RETAINED
  SAM RLE ALIQUOT
                      FIGURE  5-5

                X-MET QC  SAMPLES
                               5-21

-------
be used to assess any matrix effects which might cause the site-specific accuracy and precision to be
different from the historical accuracy and precision of CLP labs.  The average of these two replicates
will be considered the best estimate of the concentration of each calibration sample.  Thus, the CLP
analyses will  serve to confirm the X-Met analyses.  Because the CLP and the local lab independently
measure the concentration of the calibration sample and the analyses are expected to follow normal
distributions, the t-test can be used to state the confidence surrounding a conclusion  that the CLP data
do or do not support the X-Met calibration data.

The DQO process for this phase is summarized in Table 5-4.

5.3.6    REVIEW  PARCC PARAMETER:  RI PHASE 1C - SURFACE SOIL
         INVESTIGATIONS

On site analysis of arsenic, chromium and lead using an X-Met x-ray fluorescence monitor has been
selected. A rigorous field calibration procedure with off-site laboratory verification of the
calibration standards will be used with this Level  II procedure.

Precision - No historical  precision data is available for this analytical technique.  Replicate analysis
of the calibration standards will allow estimation of the achieved precision.

Accuracy - A minimal amount of data for lead analysis is presented in Appendix A.  The accuracy achieved
will be calculated from the X-Met analysis of the CLP laboratory-verified calibration standards.

Representativeness - A sampling grid has been designed to obtain  a representative picture of the metals
contamination area.

Completeness - Since this is a field technique, 100 percent completeness can be achieved.

Comparability - The use of standard soil sampling procedures and a recognized field analytical procedure
should make the resulting data comparable with other data of the same type.

5.4    STAGE 3 - DESIGN DATA COLLECTION PROGRAM:  PHASE I REMEDIAL
       INVESTIGATIONS

Stage 3 of the DQO process is undertaken to integrate the detailed data collection program developed in
Stage 2 and required to meet the RI/FS objectives.  Figure 5-6 presents the elements necessary to design
the data collection program. Through the  process of addressing the elements identified in Stages 1 and
2, all the necessary components required for completion of Stage  3 should be available.

A phased RI/FS approach has been  identified as the appropriate manner in which to collect and evaluate
data  for the example site.  All the details required to identify the specific sampling components of the
second  phase of the RI would not be available during the initial scoping of the RI/FS.

The  development of the S&A plan for Phase II would, therefore,  be undertaken following completion of
Phase I data  collection and evaluation activities.   The example provided herein identifies the general
manner in which data collection documentation is developed. Work plans and S&A plans must comply with
EPA requirements.
5.4.1    ASSEMBLE DATA COLLECTION COMPONENTS:
         INVESTIGATION
PHASE I REMEDIAL
The intent of Stage 3 is to compile the information and DQOs developed for specific tasks into a
comprehensive data collection program.  This will allow  the site manager and the RPM to identify field
                                              5-22

-------
                                TABLE 5-4
                            DQO SUMMARY FORM
1. SITE
| nf!aj|r»! 	 _.._.
tUFMppp, M_ 	

2. MEDIA
(CIRCLEONE)
3.. USE
(CIRCLE ALL THAT
APPLY)
4. OBJECTIVE 5OR

EPA
REGION 	
fV\0^5TRATCDi^ D^B
^T) Rl ? Rl 3 FRA FS- RP RA


(g£) GW SW/SED AIR . BIO O7"511

SITE /^FMSKN /^EVAL\ /^S^\ PHP MONITORING OTHEJ1
CHARAC. VASSESSj V^ALTsJ ^DESIGN ) DETER. RBffiDIAL
(H&S) -»— ' ' '• 	 ' ACIION 	 ""
Fi^cg" 2o\LS/\^PLes w\uLBe •w-^e't^ TD Assess "tix^
TrtftEAT ftp L0\D , AR'SBt^lC, &N>D C^ROmmm
LovTV\ifO TW3 PEPftETSStDM

3. SITE INFORMATION
4HP4200JM- )C J£X>W- J>ePf?65S/o*0 DEPTHTOGFmiNnwATi?H \^> %efc
GROUND WATER USE PSVC^K t f^fc WATBR
^ -rvpp, g[j&
SENSITIVE HECEFT
8. DATA TYPES (CIRC,
A. ANAL^
PH
CONDUCTIVITY
VOA
ABN
TCLP :
C\fti---TLLJ-- OgpTlA (O--2)*^"^ ] "^V^PrU^ '-TifiPTV'! •f3o-^ lOO-?-V"
DBS ASSlDg'^TS 1 WMug E^tST t>P S>CT£"

LEAPPROPHM7EMM TYPES)
rTICAL DATA B. PHYSICAL DATA
PESTICIDES TOX PERMEABILITY HYDRAULIC HEAD
PCS TOC POROSITY PENETRATION TEST
 BTX GRAWSEE HARDNESS
CYANIDE COD BULKDFJNSITY 	


7. SAMPLING METHOD (CIRCLE METHOD(S) TO BE USED)
ENVIRONMENTAL BIASED (^GRABD NON- INTRUSIVE PHASED
(jgjRCE^ C@filD3 COMPOSITE (^JNTRUSIVEj)
8. ANALYTICAL LEVELS (MDICATELEVEL(S) AND EQUIPMENT & METHODS)
LEVEL 1 FIELD SCREENING - EQUIPMBMT
LEVEL 2 FIELD A
LEVELS NON-CL
LEVEL 4 CLP/RA
LEVEL NS NONST
wai vsis. POUIPMFMT X~, '^VST JL> mA- LjEU^T— "^ 7 M (jOl^l^l R-KYl ArTI O/O
P LABORATORY - METHODS
S- METHODS
Awrwktjn

9. SAMPLING PROCEDURES
BACKGROUND -2 PER EVENT OR "^ "'jtrt'YVP'L-.g^
CRITICAL (LIST)
PROCEDURES ^
^.CCg^tJ SAi!N\?L^S l^ BACfeV 3>ifte:(^rL^/O
^rTT\P/jg" f*i -tt> 5" bF??rTH TNjlTSV2-vJ P^ I 	

10. QUALITY CONTRO
A. FIELD
COLLOCATED - S
REPLICATE- S<
FIELD BLANK - Sc
TRIP BLANK - 1 1
L SAMPLES (CONFIRM OR SET STANDARD)
B. LABORATORY
!40H , 	 np4f5FNTRi4KlK -1 PFRANAIYSIf?RATCH«R
(dOR .. __.,,,.. 	 	 BEPLICATE - 1 PER ANAI YRIS BATCH nR 	 — .
feOR , MATRIX SPIKP . 1 PPP ANAI VSIS HATCH OH
JFPnAVOR 	 	 OTHFfl&C'PPiOCEpv^M^'RrtfOS'Sfen l!\1 TKf.'T"

11. BUDGET REQUIREMENTS ,
B,irepT 4 Ti;/v^ e~^.c2 ij6^v i^crwi-ST" ' •

CONTRACTOR
SITE MANAGER
PRIME CONTRACTOR
n«Tc

FOR DETAILS SEE SAMPLING & ANALYSIS PLAN
                                       5-23
                                                                 CDM SF DQO 1.002

-------
               ASSEMBLE
            DATA COLLECTION
              COMPONENTS
  DEVELOP DATA COLLECTION DOCUMENTATION

      •  WORK PLAN
      •   SAMPLING & ANALYSIS PLAN
           Include QAPjP Elements
      -  WORK PLAN
             FIGURE  5-6
         STAGE 3 ELEMENTS
DESIGN DATA COLLECTION PROGRAM
                5-24

-------
investigation tasks which could be undertaken simultaneously and thereby reduce costs associated with the
RI/FS.

The data collection program should be developed to account for all sampling tasks and phases. During
this process a detailed list of all samples to be obtained should  be assembled as well as a schedule for
all sampling activities.

5.4.2    DEVELOP DATA COLLECTION DOCUMENTATION:  PHASE I REMEDIAL
         INVESTIGATIONS

The result of applying the DQO process is a well defined sampling and analysis plan with summary
information provided in the work plan.  Quality assurance project plan (QAPjP) elements should be
included in the S&A plan and the work plan.

Sampling and Analysis Plan

Separate S&A  plans will be prepared for each of the two phases of the remedial investigation.

For Phase I of the RI, S&A components should be written for  each individual activity including the
following:

     @   Existing well sampling

     9   Soil gas sampling

     ®   Soil sampling (metals)

The following  information  will be provided in the Phase I S&A plan for the example site:

     a   Number of samples to be obtained from the existing wells, soil gas sampling
         and the soil (metals) sampling

     ©   Number of QA/QC samples including field blanks, trip blanks collocated
         samples;  method blanks, laboratory replicates and matrix spikes

     ©   Identification of sampling locations and numbering system

     »   Prioritized listing of the sequence in which samples are to be taken from the
         existing wells, etc.

     ®   List of critical samples  for each  media

     ®   List of analyses which will be performed

     »   Chain of custody for samples transported off-site

     ©   Instrument calibration and maintenance procedures

The standard sections of a  quality assurance project plan (QAPjP), are listed in Table 5-5.  Details on
preparation of QAPjPs are contained in Interim Guidelines .and Specification for Preparing QAPjPs (EPA
1980).  The required information should be addressed in the S&A Plan.
                                              5-25

-------
                                      TABLE  5-5
                  DATA COLLECTION COMPONENTS  -  PHASE  I
   IA
                 MEDIA
GROUND WATER
                    SAMPLE TYPE
 GRAB
   IB      GROUND WATER/SOIL      SOIL GAS
    NUMBER OF
     SAMPLES

3 RESIDENTIAL WELLS
                                                  2ON-SITEWELLS
                                            52'
NUMBER OF QA/QC
    SAMPLES

r 4 REPLICATES
1 4 MATRIX SPIKE

  1 DUPLICATE
  1 SPIKE

      NA
   1C
    SOIL
 GRAB
                                                        89
                                                                  60
   2A
GROUNDWATER
 GRAB
                                                         5*
  2B
    SOIL
AUGER
                                                       100*
 NA - Not applicable - calibration standards will be run daily

1  Forty-nine initial soil gas samples will be taken, on a regular grid pattern.
   This grid will also be extended at 50 ft. intervals until 2 clean sample are obtained.
   In addition, 3 background samples will be obtained.

2  See Section 5.3.5 for details

* Estimates of  number of samples included for costing purposes. May be
   revised after evaluation of phase I data.
                                          5-26

-------
Work Plan

Work plans define the scope of services, level of effort, costs, and schedule for performing the RI/FS.
The work plan provides a general description of how all tasks and activities will be undertaken.
However, it would not contain the detailed description of how each sample is obtained or how the analysis
is performed, which is presented in the sampling and analysis plan.  Table 5-6 provides a summary of
data collection components for the example site and Figure 5-7 provides a schedule for RI activities.

The level of detail in the work plan for the example RI site is dutlined below:

     9   A brief description of the level of personnel protection to be used in the field. For a
         detailed;description of health and safety concerns,  the health and safety plan is referenced.

     •   Number of individuals to be involved in each field  sampling task and estimated duration in
         days,  including time for mobilization and demobilization.

     e   Approximate locations of soil  sampling, existing and new wells will be provided, since  costs
         associated with obtaining samples can vary with different sampling locations.  Costs for
         drilling will also vary depending on location.

     9   How data will be validated, compiled and evaluated. Data validation efforts require 2 to 3
         hours per sample for complete HSL packages.

5.5   DQO STAGE 1 - COLLECT AND EVALUATE DATA:  PHASE I REMEDIAL
      INVESTIGATIONS

This section presents a general review of the data collected during Phase I of the RI.  The data
collection (i.e., actual field investigations) and  evaluation  steps (DQO Stage 1) take place at the
conclusion of Stage 3 of the DQO process.  In order to simplify the discussion, the elements of  Stage 1
will be presented in an abbreviated form.  The DQO Stage 1 process must be repeated (usually in an
abbreviated form) whenever significant amounts of new data  are collected.

5.5.1    ANALYSIS OF RESULTS: RI PHASE IA - EXISTING WELL SAMPLING

The analysis of water from the private wells during Phase IA of the RI confirms that the wells are not
contaminated.  These data support and confirm the analytical results obtained  during the FIT
investigation.  The data was validated by the analytical chemists and accepted by the RPM.  As a result
of this data evaluation, the RPM and the contractor's site manager have determined that no alternate
water supply is required for the protection of public health and welfare.

Analyses of ground water from the two on-site wells further  confirmed the results of the FIT
investigation.  Contaminants were not detected  in the background upgradient well (OW2) and only TCE was
detected in OW1. TCE was detected in comparable levels to that reported by  the FIT analyses.  A
monitoring program will be implemented to ensure the continued potability of the residential drinking
water. This program will consist of quarterly testing of each private well.  Further, monitoring  wells
installed  as part of the Phase II effort will also be tested quarterly.

5.5.2 ANALYSIS OF RESULTS:  RI PHASE IB -  SOIL GAS SAMPLING

The soil  gas analysis confirmed the presence of a volatile organic plume originating in the depression
and migrating toward the east.  These data are presented graphically in Figure 5-8. Specifically, the
results of the GC analyses have shown  that while all compounds were consistently  detected within the
actual source areas (as defined by discolored soil), all compounds showed a rapid decrease in
concentration as a function of distance from the center  of the source area.

                                              5-27

-------
                                     TABLE 5-6

                                   EXAMPLE  SITE
                  QUALITY ASSURANCE  PROJECT PLAN ELEMENTS
                  QAPJP SECTIONS
EXAMPLE SITE INFORMATION
       PROVIDED IN
1 -
2 -
3 -
4 -
5 -
Title Page
Introduction
Table of Contents
Project Description
Project Organization and Responsibility
Quality Assurance Objectives for Data
S&A Plan
S&A Plan
Work Plan
Work Plan
S&A Plan
         Measurement

  6  -   Sampling Procedures


  7  -   Sample and Document Custody Procedures

  8  -   Calibration Procedures and Frequency

  9  -   Analytical Procedures


 10  -   Data Reduction, Validation, and
         Reporting

 11  -   Internal Quality Control Checks

 12  -   Performance and System Audits

 13  -   Preventive Maintenance

 14  -   Data Measurement Assessment Procedures

 15  -   Corrective Action

 16  -   Quality Assurance Reports to Management


*Agency and contractor quality assurance program plan
      S&A Plan
      Referenced SOPs

      Referenced SOPs

      Referenced SOPs

      CLP IFB and
      S&A Plan

      Referenced SOPs
      and QAPP*

      QAPP

      QAPP

      QAPP and SOPs

      CLP IFB and SOPs

      QAPP

      QAPP
                                         5-28

-------
                                                TIME (WEEKS)
    TASK
                              02468
                                                           10     12     14    16
    PHASE J

    MOBILIZATION
   PHASE IA
   • EXISTING WELL SAMPLING
    (INCLUDING ANALYSIS)
   PHASE !B
   • SOIL GAS SURVEY
   PHASE 1C
   • SOIL (METALS)
   DATA EVALUATION
* May be longer based on
  the extent of contamination
                                       FIGURE 5-7
                     PHASE I  REMEDIAL INVESTIGATION SHEDULE

-------
                                          X    X   X   X

                                          O  WELLOW1
                                          X    X   X    X  X
                      LIMIT OF
                    DEPRESSED AREA
                                   LIMITS OF THE SOIL GAS
                                         PLUME
SCALE (FT)
  100
            200
X  Volatile Organics detected
    in soil gas

0  No Volatile Organics detected
                             FIGURE   5-8
             RESULTS  OF SOIL  GAS  SAMPLING
                                5-30

-------
Outside the boundary of the source area, TCE was the only compound detected in appreciable
concentrations.  The presence of volatile organics in the soil gas outside the bounds of the soil
depression may be indicative of the movement of the ground water plume  in an easterly direction.

The results of Phase I soil gas sampling indicate a need to obtain additional soil samples (at depth) in
order to determine the extent of soil contamination,  Samples of ground water encountered within the area
delineated by the soil gas plume should also be obtained to determine if the soil gas plume data can be
correlated to the ground water contaminant levels.

5.5.3  ANALYSIS OF RESULTS:  RI PHASE 1C - SURFACE SOIL SAMPLING

5.5.3.1    Calibration of X-Met (Precision and Accuracy achieved for metals analysis)

To calibrate the X-Met, four calibration samples were taken along a radial line from the center of the
depressed area.  The sample locations were shown in Figure 5-4.  Each of the four samples was split into
seven replicate samples  as shown  in Figure 5-5.  Four replicates from each sample or 16 samples were sent
to a local lab with an in-place QA/QC program and were analyzed for lead, chromium, and arsenic. Only
the results for the lead samples are discussed here since the analysis performed for the other elements
is analogous.  Table 5-7 summarizes the results for lead.

The average of the four replicate analyses was taken as the actual value for each of the four calibration
samples and the X-Met was calibrated using these values.  During analyses of actual samples, each of the
calibration samples were run 15 times.  Based on the X-Met analyses of the replicates, the accuracy and
precision can be expressed as  a function of concentration.  Accuracy will  be expressed in terms of bias
where bias is expressed as:
Bias =
X-A
Where:

            X is the mean of the 15 replicates, and

            A is the concentration determined from samples sent to the local lab.

Precision will be expressed as the standard deviation of the 15 replicates. The accuracy and precision
of the X-Met are presented in Table 5-8.

Table 5-8 shows that the X-Met has accuracy values which are within j+10 percent over the entire range of
concentration.  This is an acceptable accuracy value and indicates that the X-Met should, on average,
accurately reproduce the contaminant levels throughout the site.

Given the accuracy and precision of the X-Met analyses, the detection limit for the method can be
determined.  When the X-Met results are reported, it is extremely unlikely that the reported values will
be exactly equal to the actual value.  This analytical  error is expected and acceptable; however, it is
generally not acceptable to report a positive concentration for a compound when, in fact, the compound is
not present in the sample.   The use of a detection limit lowers the risk of this occurrence to an
acceptable level.   For X-Met analyses (lead in this case), the detection limit will be set so that when a
value is reported above the detection limit, there will be greater than a 99 percent chance that lead is
actually present in the sample.
                                               5-31

-------
                             TABLE 5-7
 RESULTS OF REPLICATE ANALYSES FOR LEAD (CALIBFiATSON SAMPLES)
SAMPLE #
     1
178
171
                              REPLICATE*
192
183
                                     MEAN
181
S.D.

 8.8
     2     811

     3     263

     4       5
          777

          287

            4
         820
         242
         840
         277
         812
         267
                             5.8
         26.3

         19.3

          1.7
    ALL UNITS IN mg/kg

    S.D. - STANDARD DEVIATION
                              5-32

-------
                              TABLE  5-8
             ACCURACY AND PRECISIOM OF THE X-MET

                (Results  of Lead  Analysis-  mg/kg)

CALIBRATION  PB CONCENTRATION   MEAN X-MET
 SAMPLE #     (LOCAL LAB)      CONCENTRATION   ACCURACY   PRECISION   PRECISION/MEAN
                5.8
                              5.5
                                          -.05
          2.7
                                                                 .46
                181
                              162
                                          -.10
          6.2
                                                                 .03
                267
                              278
.04'
                                                     7.2
                                                                 .03
                812
                              800
                                         -.02
          14.0
                                                                 .02
                                 5-33

-------
The detection limit will be based on the distribution of analytical errors.  In this example, the
distribution of analytical errors is the distribution of errors for calibration sample #4.  This sample
was chosen since it has the lowest concentration of lead and is therefore most representative of the
performance of the X-Met at low concentrations.

The distribution of the 15 replicates of calibration sample #4 is normal, with a  mean of 5.5 mg/kg and a
standard deviation of 2.4 mg/kg. The actual concentration of sample #4  is 5.0  mg/kg.  Thus the average
error is 0.5 mg/kg and the distribution of errors is normal, with a mean of 0.5 mg/kg and a standard
deviation of 2.4.

Based on the above assumption, the detection limit can be determined as:

                                             Pr (Z < D)_> 99%

                                             where  Z is an error
                                                    D is the detection limit

Since the errors are normally distributed, a normal probability table can be used to determine the
detection limit D. The standard normal variable corresponding to 99% probability is 2.33 (see Table
5-9).  The detection limit is then:

                                             D - m   = 2.33
                                             where s is the standard deviation, and

                                             m is the average error


                                             D - (-.3)  =  2.33

                                               2.17
                                             D
5.99 mg/kg

6.0 mg/kg
 So, if the X-Met reports greater than 6.0 mg/kg lead there is at least a 99 percent chance that lead is
 present in the sample.  If the X-Met reports less than 6 mg/kg, a value of 3 mg/kg will be used as an
 estimate of the concentration.  A non-zero concentration is reported when lead is below the detection
 limit because lead is present to some degree in all surface soils. The value 3 mg/kg is attributed to
 soils with non-detectable lead concentrations because this value is thought to adequately represent the
 background lead concentration  in the site area.

 5.5.3.2     Geostatistical Analysis of Surface Soil Sampling Results

 Samples were collected and analyzed at each of the 89 locations on the hybrid grid.  Samples were
 analyzed for lead, arsenic, and chromium. Only the results for lead are discussed here.  The lead
 concentrations found at each sample location are shown in Figure  5-9.  Contours of the data indicate that
 the proposed conceptual model for this site is  incorrect.  Contamination does not occur in small isolated
 pockets; rather, there are two large contaminated zones. The two  contaminated  zones are bounded by zones
 of undetectable lead'contamination, so the horizontal extent of the  contamination is  known.

                                                5-34

-------
                                         TABLE 5-9



                                    PROBABILITY TABLE








          Table of the Cumulative Distribution of a Standard Normal Random Variable
k
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
12
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2J
2,9
2.4
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
0.00
0.50000
0.53981
0.57926
0.61791
0.65542
0.69146
0.72575
0.75804
0.78814
0.81594
0.84134
0.86433
0.88493
0.90320
0.91924
0.93319
0.94520
0.95543
0.96407
0.97128
0.97725
0.98214
0.98610
0.98928
0.99180
0.99379
0.99534
0.99653
.0.99744
0.99813
0.99865
0.99903
0.99931
0.999S2
0.99966
0.9?977
0.99984
0.99989
0.99993
0.9999S
0.99S97
0.0!
0.50399
0.54380
0.58317
0.62172
0.65910
0,69497
0.72907
0.76115
0.79103
0.81859
0.84375
0.86650
0.88686
0.90490
0.92073
0.9344S
0.94630
0.95637
0.96485
0.97193
0.97778
0.98257
0.98645
0,98956^
0.99202
0.99396
0.99S47
0.99664
0.99752
0.99S19
0.99869
0.99906
0.99934
0.99953
0.9f$SS
0.99978
0.99985
0.99990
0.99993
0.99995
0.9S997
0.02
0.50798
0.5477$
0.58706
0.62552
0.66276
0.69847
0.73237
0.76424
0.79389
0.82121
0.84614
0.86864
0.88877
0.90658
0.92220
0.93574
0.94738
0.95728
0.96562
0.97257
0.97831
0.98300
0.98679
09898*
0.03
0.51197
0.55172
0.59095
0.62930
0.66640
0.70194
0.7356S
0.76730
0.79673
0.82381
0.24849
0.87076
0.89065
0.90824
0.92364
0.93699
0.94845
0.95818
0.96638
0.97320
0.97882
0.98341
0.98713
o
-------
SCALE (FT)
  25
        50
                    FIGURE 5-9
          LEAD CONCENTRATJON CONTOURS
                (VALUES  IN  mg/kg)

                      5-36

-------
The first step in the analysis of these data is to determine the mean lead concentration within the
depressed area. This mean value can be used in an exposure assessment as the expected concentration of
lead which will be encountered.  A simple averaging of the 89 data points gives a mean value of 207 mg/kg
lead. This mean value is not, however, representative of the true mean lead concentration.

The data within the depressed zone are collected according to a hybrid grid. This grid contains data
points separated by 10 ft and data points separated by 50 ft.  The data points located on the 50-ft grid
provide much more valuable information concerning the mean than the data points on the 10-ft grid due to
the spatial correlation of the data.  Because the data are spatially correlated, samples taken very close
together are expected to have similar values.  Two samples, taken close together, yield  little additional
information over one sample concerning overall site properties. Since two closely located samples
provide little more information than one sample, such samples should not be counted as two separate
samples when calculating the  mean.  Thus, closely located samples must be down weighted when calculating
a mean.

In this case the data points located on the 10-ft grid have much stronger correlation than the data
points located on the 50-ft grid.  Or, in other words, the 10-ft data points are less valuable and must
be down weighted.

The undue influence of the 10-ft data is removed by averaging the four data within each of the 16 groups
of samples separated by 10 ft to produce 16 new values which are located at the center  of each of the
50-ft cells.  This procedure yields 41 data located on a regular 71-ft grid.' Since these  41 data are
located  on a regular grid, they have equal correlation and provide equally valuable information for
estimating the mean.  The average of these  41 values is 174 mg/kg.  This value is the best estimate of
the mean concentration of lead within the depressed area.

To obtain a reasonable worst case value for the risk assessment, the value which represents a 10 percent
chance  of exposure will be determined.  This value will be determined by examining the  histogram of the
41 data which provide equal site coverage.

The histogram provides an experimental measure of the likelihood of a.sample concentration falling
between any two concentrations of interest.  The histogram of lead data is shown as Figure 5-10.  The
histogram shows, for example, that 7 percent of the data fall between 200 and 300 mg/kg. In addition,
the histogram indicates that approximately 10 percent of the data and  equivalently  10 percent of the soil
in the depressed area has  lead concentrations  in excess  of 735 mg/kg. Therefore there is  a 10 percent
chance  that an exposure of at  least 735 mg/kg will be received. This value can be used in an exposure
assessment as a reasonable worst-case exposure.

Given an estimate of the mean lead concentration within the depressed area, the next quantity of interest
is the uncertainty surrounding this estimate.  As discussed in the guidance document, the uncertainty
associated with an estimate of concentration can  be determined once a model of the site-specific spatial
correlation is available.  This  model will be obtained by modeling the variogram of the  available data.

The variogram model describes the spatial variability of the data as a  function of distance. It is
determined by modeling the experimental variogram of the data which is calculated by grouping pairs of
data into distance classes and calculating a type of variance measure for each pair  of data. For
instance at this site, 64 pairs of data are separated  by 10 ft.   For each pair of these data,  the
following variability measure is calculated

                                 g= 1/2  (Xi  -X2)2

      where   g is the variability measure,  and
               X  and X   are data separated  by 10 ft
                                               5-37

-------
60-
50-
40-
30-
20-
10-
          nnnn  n  n  n     n
                                            LEAD
      100  200  300  400  500  600  700  800  900  CONCENTRATION
                                           (mg/kg)
 Number of data - 41
 Mean - 174.4
 Variance - 68012.3
 Coefficient of variation - 1.50
                     FIGURE 5-10
          HISTOGRAM OF LEAD CONCENTRATION
                       (mg/kg)
                     5-38

-------
This measure is then calculated for all 63 other pairs of data and the average of the 64 variability
measures is obtained.  This average value is the average spatial variability or correlation of data
separated by 10 ft.  Similarly, the spatial variability (variogram) of data separated by 20 ft, 30 ft, 40
ft, and so on can be calculated.  Plotting the experimental variogram versus  distance gives a pictorial
representation of the spatial variability as a function of distance (see Figure 5-11).

In the majority of cases  an experimental variogram shows three features:   (1) at zero distance the
variogram is non zero, (2) as distance increases the variogram increases linearly, and (3)  at some
distance  the variogram will level off and remain constant.  The first feature of the variogram  is
essentially the variance of data separated by very small distances.  This variance comprises the
variability of the analytical method, the  sampling method and the intrinsic variability of contaminants
and is typically many times larger than  the analytical precision.

The linear increase of the variogram is  due to the  fact that when data are separated by  greater and
greater distances,  the spatial variability of the data increases or, in  other words, the correlation
decreases. This relationship is expected since it is intuitive that data separated by small distances
will have more similar values than data separated by large distances. However, spatial variability does
not continue to increase for all distances.  Beyond some site specific distance variability will cease
increasing.  This distance is known as the range of correlation. If a pair of data are  separated by a
distance  greater than the range of correlation,  the  data are  not  correlat  This average value is the
average spatial variability or correlation of data separated by 10 ft.  Similarly, the spatial
variability (variogram) of data separated by 20 ft, 30 ft, 40 ft, and so on can be calculated. Plotting
the experimental variogram versus distance gives a pictorial representation of the spatial variability as
a function of distance (see Figure 5-11).

In the majority of cases  an experimental variogram shows three features:   (1) at zero distance the
variogram is non zero, (2) as distance increases the variogram increases linearly, and (3)  at some
distance  the variogram will level off and remain constant.  The first feature of the variogram  is
essentially the variance of data separated by very small distances.  This variance comprises the
variability of the analytical method, the  sampling method and the intrinsic variability of contaminants
and is typically many times larger than  the analytical precision.

The linear increase of the variogram is  due to the  fact that when data are separated by  greater and
greater distances,  the spatial variability of the data  increases or, in  other words, the correlation
decreases. This relationship is expected since it is intuitive that data separated by small distances
will have more similar values than data separated by large distances. However, spatial variability does
not continue to increase for all distances.  Beyond some site specific distance variability will cease
increasing.  This distance is known as the range of correlation. If a pair of data are  separated by a
distance  greater than the range of correlation,  the  data are  not  correlated.

The experimental  variogram for lead shows  the three general features of the  variogram. At zero distance
a non zero variability is  observed. Variability increases  until a distance of 75 ft is reached.  After
this distance, variability  ceases increasing. A  model which  incorporates these three features has been
fit to the experimental variogram  (Figure 5-11). This  model is known as a spherical model.  The model
fits well  for distances less than 50 ft, however, the fit is  not good at greater distances since the
experimental data  show a slight periodicity.  This periodicity can be modeled; however, the complexity of
the model would be increased significantly and the increased usefulness of the model would be slight.  In
short, a  more complex variogram model would increase  the costs of performing this  study but would not
influence the results.  For this reason, the spherical model fit to the experimental values will  be
acceptable.

Once the variogram model  is known,  the uncertainty associated with the estimate of the mean can be
determined.  This uncertainty is expressed as a variance and is termed the estimation variance. The
variogram model can also be used to determine uncertainty as a function of the number of data. Hence  the

                                                5-39

-------
                                              MEAN = 207 mg/kg  ,
                                              VARIANCE = .73X1CT
   1 x 10
O
cc
   .5 x 10
                                                        Model
                                                     Experimental Points
                       50
100
                                                150
                         200  DISTANCE
                             FIGURE 5-11
              VARIOGRAM OF LEAD CONCENTRATIONS
                                  5-40

-------
variogram model can be used to determine how many data are required to lower the uncertainty to an
acceptable level. The techniques used to determine the estimation variance are beyond the scope of this
document but are described briefly in the guidance manual.

The estimation variance associated with the estimate of mean lead concentration at the site is 350
(nig/kg)2 .  This estimation variance was determined using the vfiriognim model.  Given this variance,
confidence  limits can be set.  Assuming a normal distribution of the means, the 95 percent confidence
limits on the mean site lead concentration are:
Z.025  <
                  m < Z
                        .975
       L,  - X  <  m  <  L,  - X
         s~~              s
       z.
  il5, Z  025  are known from a standard normal table (Table 5-8)
       X  is the estimated mean lead concentration = 174

       S is the square root of the estimation variance = 18.3

       L± and L2  are the confidence limits


        .025  = —1 "
       Ll =Z.025S+X
       L   = (- 1 .96) (18.3) +  174
           = 138
Similarly

       L2  = 210

Based on this analysis 95 percent of the time the true mean will fall between 138 and 210 mg/kg. This
statement is a numerical data quality statement which for this example is the output of the DQO process.

The confidence interval can be used to assess whether the available data are sufficient for the data
uses.  The primary use for this data is as input into a risk assessment model.   The confidence  limits are
approximately _+  20 percent ((1.96)(18.3)/174) of the estimated mean. Given  the data uses this confidence
level is deemed acceptable.  Additional data are not required.
                                               5-41

-------

-------
       STAGE 2 - DATA USES AND
         NEEDS: GROUND WATER
               INVESTIGATIONS
                   DATA USES
                   DATA TYPES
           DATA QUALITY NEEDS
          DATA QUANTITY NEEDS
    SAMPLING /ANALYSIS OPTIONS
            PARCC PARAMETERS

       STAGE 2 - DATA USES AND
       NEEDS: SUBSURFACE SOIL
               INVESTIGATIONS
                   DATA USES
                   DATA TYPES
           DATA QUALITY NEEDS
          DATA QUANTITY NEEDS
    SAMPLING /ANALYSIS OPTIONS
            PARCC PARAMETERS

     STAGE 3 - DATA COLLECTION
            PROGRAM: PHASE II
   DATA COLLECTION COMPONENTS
DATA COLLECTION DOCUMENTATION

         STAGE 1 - COLLECT AND
      EVALUATE DATA: PHASES II
     . REMEDIAL INVESTIGATIONS
     Rl PHASE IIA - GROUND WATER
   RI PHASE IIB - SUBSURFACE SOIL

-------

-------
           6.0  DQO  DEVELOPMENT  -  PHASE  II  REMEDIAL INVESTIGATION
6.1    DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  RI PHASE IIA - GROUND
       WATER INVESTIGATION

6.1.1    IDENTIFY DATA USES:  RI PHASE IIA - GROUND WATER INVESTIGATION

During Phase I of the RI, ground water samples were obtained from the on-site wells and three private
water supply wells located east of the site.  The samples obtained from the private wells did not contain
detectable levels of volatile organics or other compounds which would require that alternative water
supplies be provided.  The soil gas analyses, however, indicate that volatile organics are present in the
ground water.  The ground water plume appears to be migrating eastward and could potentially affect the
private wells.       '

Ground water data are required in this phase to evaluate the extent of contamination,  to develop a risk
assessment,  and to evaluate the potential remedial alternatives for the site.

Data uses and the DQO summary for this phase are provided in Table 6-1.
                   i
6.1.2    IDENTIFY DATA TYPES:  RI PHASE IIA - GROUND WATER INVESTIGATIONS

The types of data required in this phase are varied.  To satisfy the data uses, ground water monitoring
wells will be installed.  All wells will be screened in the unconfined aquifer. No wells are planned in
deeper aquifers since these aquifers are brackish and therefore non-potable.  The types of data which
will be obtained from testing the monitoring wells are contaminant concentrations, general water quality
information, hydraulic head, and horizontal hydraulic conductivity.  Additional data types will be
measured during monitoring well installation.   Soil samples can be taken to measure soil contaminant
levels, porosity, vertical  permeability, and other general soil properties  including particle size
distribution, density, and standard penetration  tests.

In this example document, it is not feasible to discuss all the considerations associated with each of
the identified data types.  For this reason the document will focus on the concentration of contaminants
within  the ground water as this type of information (used for site characterization,  risk assessment, and
evaluation of alternatives) is the most commonly collected and discussed type of ground water data.

Soil gas analyses for volatile organics indicate  that TCE is the primary contaminant of concern outside
of the depressed area. Ground water sampling and analysis will be undertaken to identify and confirm the
concentration of TCE, other volatile organics,  and metals and to obtain general water quality information
in the shallow aquifer.  This information will be used in conjunction with data on aquifer
characteristics to develop the risk assessment.

6.1.3     IDENTIFY DATA QUALITY NEEDS:  RI PHASE IIA - GROUND WATER
         INVESTIGATION
Data Quality Factors

     Prioritized Data Uses:


     Appropriate Analytical Levels:


     Contaminants of Concern:
Risk Assessment
Evaluation of Alternatives

Risk Assessment:  Level III, IV, V
Evaluation of Alternatives:  II, HI, IV

TCE, Arsenic, Chromium,  Lead

 6-1

-------
     SITE
LOCATION.
NUfcBER_
PHASE	
                2\R13 ERA FS  RD  RA
                                                       TABLE  6-1
                                                       DATA USES
                                                        EPA REGION
DATE	
CONTRACTOR-
SfTE MANAGER
^V. DATA USE
MEDIA ^X,.
SOURCE SAMPLWG
TYPE
SOtLSAMPUNG
GROUND WATER SAMPUNG
SURFACE WATER/SEDMENT
SAMPUNG
AIR SAMPLING
BIOLOGICAL SAMPUNG
OTHER

SITE
CHARACTERIZATION
(NCLUOiNG
HEALTHS
SAFETY)







RISK
ASSESSMENT

X
X




EVALUATION OF
ALTERNATIVES

X
X




ENQNEEHNG
DESIGN OF
ALTERNATIVES

X





MONITORING
DURNG
REMEDIAL ACTION







PRP
DETERMINATION







OTHER








ON
     NOTE: CHECK APPROPRIATE BOX (ES)
                                                                                                             COM SFDQ01.001

-------
     Levels of Concern:
     Required Detection Limit:

     Critical Samples:
5 ppb TCE/50 ppb Metals

2 ppb TCE

Wells MWI, MW2
Monitoring wells will be installed following procedures which will be outlined in Section 6.1.6.  Well
locations will be chosen to accomplish the following specific goals: (I) measure contaminant
concentrations within the ground water plume and (2) serve as an early warning system to detect the
migration of contaminants towards the residential wells.  Because of the differing goals of the
monitoring wells, different data quality, as expressed by analytical level, will be required.

Analysis of water from the early warning wells will be conducted by Method 601/602 to allow detection of
volatile organic contaminants near 1 ppb.  Since the potential level of concern for volatile organics
tends to be in the low ppb range,  the detection limits for this method are acceptable to meet the
objectives of Phase HA.  Method  601/602 can be performed by CLP SAS.  Analysis of water from the
remaining wells can  be accomplished by Method 624, which has 5-ppb detection level for TCE.  Since
turnaround time is not a major issue, these samples will be sent to the CLP (RAS) for analysis.

Ground water samples from the newly installed wells will also be analyzed for lead, chromium and arsenic.
The CLP RAS Contract Required  Detection Limit (CRDL) of 5 ug/1, 10 ug/I  and 10 ug/1 respectively will be
required for this phase. Based on the MCL values of 50 ug/1 for arsenic, lead, and chromium, CLP RAS
detection limits of 5 to 10 ug/1 are sufficiently low (see Appendix B).

Methods available to accomplish these analyses are the Inductively Coupled Argon Plasma (ICAP) method for
chromium, and the Furnace Atomic Absorbtion (FAA) method for arsenic and lead.
6.1.4    IDENTIFY DATA QUANTITY NEEDS:
         INVESTIGATION
    RI PHASE HA - GROUND WATER
The factors influencing the choice of monitoring well locations are the size and shape of the soil gas
plume (primary factors),  schedule, and budget (secondary factors).  The plume determined by soil gas
sampling is shown in Figure 5-8.

Wells must be installed to answer specific questions concerning the distribution of contaminants in the
ground water. Three wells will be installed in addition to the two existing wells on site.  For this
example site, five wells will be adequate to: (1) confirm sources, (2) confirm the extent of the soil
gas plume, and (3) determine the magnitude of contamination within known plumes.  The locations of these
monitoring wells are shown on Figure 6-1.

To determine the extent of the ground water contamination plume, wells MWI and MW2 will be installed at
locations shown in Figure 6-1.  These well locations are beyond the edge of the soil gas plume and hence
should encounter uncontaminated ground water.  These wells provide early warning by indicating if
contaminants are moving closer to the private  water supply wells. Well MW3 and existing well OW1 are
located within the soil gas plume to indicate the magnitude of ground water contamination.  Only one new
well is required for this purpose and will provide a measure of the range of contaminants present in the
plume.  To determine background water quality, existing well OW2 will be sampled.

After the results from these wells are obtained, an assessment of the data will be made and any data gaps
will be indicated.  Additional wells will be added if necessary to fill data gaps.
                                              6-3

-------
                                   SITE BOUNDARY
                           EXTENT
                           OF SOIL
                         GAS PLUME
      N
NOT TO SCALE
                                                 LEGEND

                                                 Monitoring Well Locations

                                                 Existing well
                               FIGURE  6-1
                  LOCATION OF MONITORING  WELLS
                                   6-4

-------
 6.1.5    EVALUATE SAMPLING/ANALYSIS OPTIONS: RI PHASE I GROUND WATER INVESTIGATION

 The installation of monitoring wells will be undertaken by use of a drill rig equipped with a hollow stem
•auger.  This drill rig will also be used to obtain soil samples at the site.  Ground water and soil        !
 samples will be collected at thte locations specified previously following standard operation procedures.

 The DQO process for Phase HA is summarized in Table 6-2.

 6.1.6    REVIEW PARCC PARAMETERS: RI PHASE IIA - GROUND WATER INVESTIGATION

 Ground water samples from the early warning wells (MWI, MW2) will be analyzed for volatile organics using
 Method 601/602.  Ground water samples from the other wells will be analyzed for volatiles using Method
 624.  All well samples will be analyzed for metals using CLP RAS procedures.  PARCC parameters are
 reviewed separately below for wells MWI and MW2, and for the other wells.

 PARCC Parameters for Wells MWI and MW2

 Precision

 Precision data on Method 601/602 for volatile organics are not readily available.  Replicate samples will
 be collected and analyzed to estimate the precision actually achieved on volatile organics analysis.
 Historical precision data for CLP RAS metals analysis are listed below;
                  Contaminant

                  , Lead
                   Arsenic
                   Chromium
Precision (% RSD)

    32
     9.4
     9.8
Accuracy - Historical accuracy data for analysis of volatile organic compounds by Method 601/602 are not
readily available.  Matrix spike samples will be analyzed to determine the accuracy achieved on these
samples.  Historical accuracy data for the metals, analyses using CLP RAS procedures are listed below:
                  Contaminant

                   Lead
                   Chromium
                   Arsenic .
Accuracy (% Bias)

      -0.7
      -2.6
      -8.3
Representativeness -[ Three to five well volumes will be purged before sampling the observation wells to
ensure that standing water is removed and that the samples are representative of the ground water.

Completeness - These samples have been defined as critical samples; therefore,  100 percent completeness
is required.  If valid results are not obtained for any sample, a new sample aliquot will be analyzed or
the well will be resampled.

Comparability - The use of standard, published sampling and analytical method will ensure the
comparability of the data.

PARCC PARAMETERS FOR WELL MW3, OW1  AND OW2

Precision - The CLP RAS historical precision data for the intended analytes are outlined below:
                                             6-5

-------
                                 TABLE 6-2
                            DQO SUMMARY FORM
1. SITE



2. MEDIA
(CIRCLEONE)
3. USE
(CIRCLE ALL THAT
4 OBJECTIVE S^£
EPA
REGION 	
	 	 	 	 RM /fuv) 019 FHA FS HD RA
— — ( CIRCLE CINF)

SOIL (ON) SW/SED AIR BO OTHER
SITE RISK EVAL. ENGQ PRP MONTrOHNG OTHER
CHARAC. ASSESS. ALTS. DESIGN DETER. REMEDIAL
(HSS) ACTION 	
>oNyoiofccE<2- wvcA Aftfr "Reov)L\e€:iD to FMAVJGATTE' TWET
Sx.TFKi"T OF ^to-rAiAAiNATi&M, DEVELOP A- •RVS^ AfjsessmeNrrr
Avt^O A65£55
4v^ 1 J*" ^^I I LJk ( PJ^lV^ T"V^\ r\ 1— ~ l^V- V_rt\ IV\/*'C"!L\ Ut— -^^

S. SITE INFORMATION
GROUND WATER L
SOILTYPES.£iL£
SENSmVERECEP
BE ORiVo¥U(M£> \\JATFSR
>if 1 CtL T\ LI - D6PTV4 ft -^O -f -V • Sl^V^ -'DeflU 3D - > lOO^
TOHS 'ftC'SiDEwr^ 1 V\ML_ET PA"^T of -rue: SITS

8. DATA TYPES (CIRCLE APPROPRIATE DATA TYPES)
A. ANALYTICAL DATA B. PHYSICAL DATA
(&
ccoNjjucnnvrrYL,
cyoo
ABN
TCLP
7. SAMPLING METHC
ENVIRONMENTAL
SOURCE
PFRTICinFS TOX CeEHMEABIUTY^ CSYOBAUJC HEAD^>
) PCB TOC POROSITY PENETRATONTEST
ClM£TALS^> BTX GRAIN SIZE HARDNESS
l^PiTgR QUA* 'TV VA u^eui— "D^SWCC-A-TfoiO SC^Ps

10. QUALITY CONTRC
A. FIELD
COLLOCATED -i
REPLICATE- S
FIELD BLANK -S
TRIP BLANK- 1
)L SAMPLES (CONFIRM OR SET STANDARD)
B. LABORATORY
140H HFARFMTRI Af*< -1 PFRANAIYR IS BATCHER 	 	 ,.
%OR _ 	 REPLICATE 1 PFR ANAI YSIR RATOH OR , 	 	 ; 	 	 	 -
V.Cia 	 	 MATRIX BPIKF . 1 PPP AMAI VSIS BATCH OH 	 	 	 	
PFRnAYOH „ CTIVBH

11. BUDGET REQUIREMENTS
STAFF tLvc\Y~f:>o^r->UySi5>i~ t^vil \Pv S > LVK^wusf^
r
CONTRACTOR
SITE MANAGER

1111 r- ,) -J
PRIME CONTRACTOR
r-flT=

FOR DETAILS SEE SAMPLING S ANALYSIS PLAN
                                                                  CDM SF DQO 1.002
                                       6-6

-------
            Contaminant
Precision (% RSD)
              TCE
              PCE
              Benzene
              Toluene
              Lead
              Arsenic
              Chromium
        17
        13
        12
        14
        32
         9.4
         9.8
 QC samples will be analyzed to determine the precision achieved on the well samples.

 Accuracy - The CLP RAS historical accuracy for the intended analytes are:

            Contaminant                                        Accuracy (% Bias)
              TCE
              PCE
              Benzene
              Toluene
              Lead
              Arsenic
              Chromium
        -22.8
        -42.5
         -3.3
        -23.3
         -0.7
         -8.3
         -2.6
 Representativeness - Three to five well volumes of water will be purged to ensure 'that well samples are
 representative of the ground  water quality near the well.

 Completeness - The historical completeness achieved for CLP RAS analyses is 80-85 percent. This
.completeness range is acceptable given the project goals.  However, if validated results are not obtained
 for each well sample, the well will be resampled.

 Comparability - The use of standard published sampling and analytical methods will ensure data
 comparability.

 6.2    DQO STAGE 2 - IDENTIFY DATA USES AND NEEDS:  PHASE IIB - SUBSURFACE
       SOILS INVESTIGATION

 Soil sampling in Phase I of the RI provided information on the nature, extent, and magnitude of surface
 metals contamination.  In  Phase  I only surface soil samples were obtained so the concentrations of metals
 and, more importantly, volatile organics  are not  known at depth.

 The purpose of this phase of the RI is to obtain information on the nature, extent, and magnitude of
 volatile and heavy metal contamination at depth.

 Soil contamination is a major concern at the site and adequate data must be obtained to allow for an
 accurate estimate of the areal extent and  total volume of contaminated soil present.   This information
 must be determined for detailed cost estimates to be developed.

 6.2.1     IDENTIFY DATA USES:  RI PHASE .IIB - SUBSURFACE SOIL INVESTIGATION

 Soil sampling data will be used to assess the magnitude and distribution of subsurface soil
 contamination.  This information will be used in a risk assessment, in evaluating remedial alternatives,
                                              6-7

-------
and in designing a remedial action.  Soil sampling will also provide information on the physical
properties of the soils which can be used in assessing remedial alternatives.

6.2.2    IDENTIFY DATA TYPES:  RI PHASE IIB - SUBSURFACE SOIL INVESTIGATION

The types of data to be collected are concentrations of volatile organics and metals in the soils and the
porosity and  permeability (horizontal and vertical) of the soil.

6.2.3    IDENTIFY DATA QUALITY NEEDS: RI PHASE IIB - SUBSURFACE SOIL
         INVESTIGATION
Daily Quality Factors

     Prioritized Data Uses:
     Appropriate Analytical Levels:
     Primary Contaminants
     of Concern:

     Levels of Concern:
     Required Detection Limit:
     Critical Samples:
Risk Assessment
Evaluation of Alternatives
Engineering Design

Risk Assessment: Levels III, IV, V
Evaluation of Alternatives:  Levels II, III, IV
Engineering Design: Levels II, III, IV
TCE, Arsenic, Chromium, Lead

TCE: 4 - 40 nig/kg"
As:  25 - 35 mg/kg
Cr:  90 - 110 mg/kg
Pb:  450 -  550 mg/kg

TCE: 2 mg/kg (Given the high cleanup level anticipated for the
metals,  detection limits in the .low mg/kg range will be
acceptable.)

Clean samples at boundaries of contaminated area.
The level of concern for soil TCE concentration shown above (4 mg/kg) is actually the lowest value from a
range of potential levels of concern (4 to 40 mg/kg).  This range of values was obtained by applying a
simple transport model which predicts the concentration of TCE in the soil which will result in TCE
concentrations at the drinking water wells which exceed 5 ppb (proposed MCL).   Inputs to this process
include soil organic carbon content, TCE solubility, net inflow, permeability, dispersion, retardation,
biodegradation, and hydraulic .gradient.  Uncertainty in these values causes the large range in the
possible levels of concern.  After completion of the RI, the level of concern for soil TCE will be
refined  and an action level will be chosen. The level of concern  is presented here only to ensure that
an analytical technique with appropriate detection limits is selected.

To assess remedial alternatives and provide input into a risk assessment, quantitative data concerning
the magnitude,  nature and distribution of contaminants are required.  Metals are not expected to  migrate
downward  from  the surface to any great extent due to the relatively high pH of the soils and waste
material. As such,  metals are not expected to impact ground water resources,  or  cause a human health
concern via this pathway.  For this reason Level II data (X-Met)  will  be sufficient for metals analyses.

Volatile organic contaminants are expected to be present in significant quantities at depth.
Potentially, analytical methods from Levels II,  III or IV could be used. The proposed methods from each
analytical level are shown below:

                                              6-8

-------
            Level

             II

             HI

             IV
Method

Head Space/GC/PID .

Purge & Trap/GC/MS

CLP RAS (purge & trap/GC/MS)
Level HI and IV data provide quantitative information on the concentration of organics in the soil.
general level of the expected accuracy and precision of these methods is available from historical
performance data (see Appendix A).
                                                       The
Level II analyses (field GC) can be used to obtain numerical concentration values for a small set (5) of
important VOA compounds.  The precision and accuracy values for this procedure are unknown.
6.2.4    IDENTIFY DATA QUANTITY NEEDS:
         INVESTIGATION
            RI PHASE IIB - SUBSURFACE SOIL
Sufficient data must be collected to define the vertical and horizontal extent of the contaminants.  This
objective can be cost effectively accomplished by sampling according to a regular three-dimensional grid.
The chosen grid size will directly influence the number of samples taken.

Based on  previous site investigations and the conceptual model, the depth to ground water is 15 ft. No
information concerning the variation in contaminants with depth is available. Thus, the choice of the
vertical distance between samples must be based on assumed variation in contaminant concentration with
depth and the goals of the study.

The goal of this phase is to determine the horizontal and vertical extent of contaminants.  Any available
or assumed information on these quantities will aid in choosing the necessary grid size.

The horizontal extent  of contaminants  is expected to increase with depth due to the dispersion of the
contaminants during downward migration.   For this reason, dispersion within the unsaturated zone should
be greatest at the water table, so one sample must be taken at or just above the water table.  Samples at
the water  table become critical samples near the boundaries  of contamination.  In addition to the samples
taken at the water table, one sample at a depth of 7-9 ft will be taken from each  boring.  The
information obtained from this sample will be used in conjunction with the information from the deeper
samples to assess any vertical trends in contaminant levels and to determine  the total quantity of
organic contaminants  in the soil.  Based on the above rationale, two samples (at depths of 13-15 and  7-9
ft) will be taken from  each soil boring.

A major factor influencing the choice  of horizontal grid spacing is the maximum likely horizontal extent
of soil contamination.  This information will indicate the areal coverage required for the soil samples.
Given the 200 ft square surface contamination area and a 15 ft depth to  the water table, vadose zone
contamination is expected to be contained within a 300 ft square area which  includes the depressed area.
This is only a preliminary estimate of the area of contamination which may be modified based  on initial
Level II soil analyses.

The horizontal grid size chosen is based on the assumed spatial variability of the contaminants.
Information on the spatial  variability of organic  contaminants is not available for samples taken at
depth;  however, information on surface metals contamination is available (see Section  5.5.3).  Although
inorganic  contaminants do not behave  identically to organic contaminants in the subsurface environment,
                                              6-9

-------
 both the organic and inorganic contaminants have an identical site genesis.  For this reason, the spatial
 variability of the organic contaminants will be assumed to be similar to that of inorganic contaminants.
 This assumption suggests a process for setting the horizontal grid size.

 In Appendix C of the Development Process manual, the relationship between spatial variability and the .
 represeutivity of the grid size is discussed, and it is shown that representivity of samples is linked to
 the range of correlation of the variogram model. Ideally, samples should be taken on a grid which is
 approximately one half of the range of correlation.  A geostatistical analysis of surface soil lead
 concentration (Section 5.5.3.2) indicates that the range of correlation for lead surface contamination is
 75 ft.  Based on this analysis, a representative grid size would be 40 ft,  assuming lead and organic
 contaminants behave identically.  However, since organic contaminants do not behave identically to lead,
 a slightly larger grid is chosen to avoid oversampling.  A 50-ft grid  will provide the required sampling
 density.

 Based on the previous analysis, 36 soil borings will be installed and sampled at two depths for a total
 of 72 samples over a 300-ft-by-300-ft area. The selected grid  is  shown  in Figure 6-2.  The number of
 samples chosen for this phase is slightly less than the number of samples estimated during the project
 scoping and included in the work plan (100).  The reduction in sample requirement is a result of
 information gathered in Phase 1C.

 The choice of the sampling grid spacing and sampling depths was based on  the assumed spatial distribution
 of the organic contaminants.  Phase 1C of the RI identified surface soil metal contamination within the
 0-2 in. depth interval.  Soil sampling will be performed, in this phase, to investigate the vertical
 extent of metal contamination.

 Metal contamination is considered separately from organic contamination because the vertical distribution
 of these two types of contaminants are expected to be very different.  The sludges  containing the metals
 were relatively caustic so metals are expected to be immobile.

 For this reason, the downward migration of metals, will be limited.  Based on this model of the vertical
 distribution of metals the vertical grid spacing must be denser for the metals than  for the organics.

 Based on the conceptual model of metals migration and experience at other sites,  the maximum depth of
 migration is estimated to be 3 ft.  Based on this assumed depth of migration samples will be taken at
 6-in. depth intervals.  To ensure that these samples are representative  of the contamination at depth
•over the most highly contaminated portion of the depressed area, four  of the soil borings (see Figure
 6-2) located within this zone will be sampled at 6-in. intervals from  the surface to a depth of 3 ft.

 This sampling procedure yields 24 soil samples which will be analyzed on-site for arsenic lead, and
 chromium content  using the X-Met.  If on-site analysis indicates that metals contamination extends below
 a depth of 3 ft, additional samples will be taken until 1 ft of soil with no detectable metals
 concentrations is located.
 6.2.5    EVALUATE SAMPLING/ANALYSIS OPTIONS:
          INVESTIGATIONS
RI PHASE IIB - SUBSURFACE SOIL
.The turnaround times, analysis costs, precision and accuracy vary with the chosen level of the analytical
 method.  Each of the analytical methods.has its strengths.  A comparison of the methods is given below:
                                               6-10

-------
   LIMIT OF
THE DEPRESSION
    AREA
                                                                              N
  SCALE (FT)


 0   25   50
                                                         t
BORING LOCATION
 TWO SAMPLES
  PER BORING
     Borings sampled at 6" intervals to 3 feet
                                 FIGURE  6-2
                     SOIL  SAMPLING  LOCATIONS
                                     6-11

-------
Level
Turnaround Time
Cost of Analysis
Precision & Accuracy
   II        10 minutes to 2 hrs.
   Ill       1  day to I week

   IV       6  weeks
                                      $40

                                      $300

                                      $ 801
                                   Low

                                   High
                                   High
 Labor associated with sample shipment.  Does not include laboratory costs.


Based on the above table, the primary drawback of Level II data is the low accuracy and precision of the
method.  The low quality of these data is offset by the fast turnaround and low cost of the analyses.
These factors suggest that Level II data should be used for  screening and Level III or IV data should be
used for confirmation or to adjust the Level II analyses.

The proposed sampling plan contains 72 samples located on 2 vertical levels.  All 72 samples will be
analyzed on-site using Level II analytical procedures.  A subset of these samples will be sent to the
CLP for confirmatory analysis.  These confirmatory  samples will be analyzed for volatile organics only,
because the Level II metals analyses (X-Met)  were confirmed during the surface  sampling phase of the RI.

Confirmation of Level II organic analyses is of greatest importance when the reported concentration is
near the level of concern.  To accurately assess the performance of the Level II method near the level of
concern, at least six samples with measured TCE concentrations near the level of concern range will be
sent to the  CLP.  Six samples were chosen to  provide an accurate estimate of the potential errors of the
Level II procedure at the level of concern.

In addition to the previous 6 confirmation samples, at least 12 more confirmation samples are required to
assess the effectiveness of the field GC over the likely range of concentration. To ensure that the
field GC is not reporting false negatives, four  samples containing no detectable organics will be sent to
the lab. Also, eight of the samples with reported concentrations which  are larger than the action level
range will be sent to the lab.  This set of 18 confirmatory samples (25 percent of the total) will
provide information on the effectiveness of the field GC  over the range of concentrations which are
encountered.

The DQO process for Phase IIB is summarized in Table 6-3.

6.2.6    REVIEW PARCC PARAMETERS:  RI PHASE IIB - SUBSURFACE SOIL
         INVESTIGATIONS

The achievable precision and accuracy of Level II methods are  unknown. For CLP RAS confirmatory samples
the following information is available.

Precision - The CLP RAS historical precision for analysis of soils for TCE is unavailable.  QC samples
will be analyzed to determine the precision achieved.  However, historical precision for similar organic
compounds ranges from 10 to 30 percent RSD (see Appendix A).

Accuracy - The CLP RAS historical accuracy  for analysis of soils for TCE is unavailable.  However,
historical accuracy for similar organic compounds ranges from  -12 to +13  percent bias. QC samples will
be analyzed to determine the accuracy achieved (see  Appendix A).
                                              6-12

-------
                                TABLE  6-3
                            DQO SUMMARY FORM
1. SITE
MAUP Don D^mofosri1
inrijpM
M aocp

2. MEDIA (y^
(CIRCLEONE)
3. USE SITE
(CIRCLE ALL THAT CHARAC.
APPLY) (H&S)
4. OBJECTIVE SeSU5^«fVPt-g
fft&TPrUS -IT) •pRfgfiiM
EPA
REGION 	 	
Rl 1 fWg^ Rl 3 ERA FS RD RA


GW SW/SED AIR BO OWBl
/RISK\ /EviuN XENQG\ PRP MONiTOR£ •AK>fcUY'Z Pbft \)CAs AihiO
IfVJg Tftf5 ftbftt2_fttOi-TVVU. Af^ft Xfg^TVCAU ^XTE'KiT
ftF toteT&MlIOATl'C'AJ

9. SITE INFORMATION
AQCA^CO-W X 2OD.Pt DEI
5R66S(t>AJ DEPTHTOGRoiRinwA-n=H I1:? -T«5e'"t
GROUND WATER USE C'PV!^>f^"v^-> V^Af®^
SOB TVPfiS 61AO frL-Tl 1
SENSmVE RECEPTORS P>££ I
-U--DeT>lV\. £>-2>O ^-t-j 5Wf\L^-DSPTH 3n.->loo-H-
DgMH3 \ miLE 0ft5T O1? 2>ilgl.

8. DATA TYPES (CIRCLE APPROPRIt
A. ANALYTICAL DATA
pH PESTICIDE
CONDUCTWrfY PCS
(SgS> CMETAL8J>
ABN CYANIDE
TCLP
%TECMM TYPES;
B. PHYSICAL DATA
3 TOX PERMEABILITY HYDRAULIC HEAD
TOO POROSITY PENETRATION TEST
BTX GRAM SIZE HARDNESS
COO BULKDENSrrY — __ _ ,_


7. SAMPLING METHOD (CIRCLE METHOOfS) TO BE USED)
ENVIRONMENTAL BIASED ^gJRAEp NON- INTRUSIVE PHASED
(^SOURCE> C§FWD COMPOSfTE (^INTRUSIVE)
8. ANALYTICAL LEVELS (INDICATE LEVELfS) AND EQUIPMENTS METHODS)
LEVEL 1 RELD SCREENING -EQUIPMENT
LEVEL 2 FIELD ANALYSIS - ECU
LEVELS NON-CLP LABORATOf
IIPUPMT IYl^T7\l-S - XVYiET , VOAs GrC.
»Y- METHODS
LEVEL 4 CLP/RAS - METHODS COT^Pl R YYl A-Tl O rJ £>1p \J O A S
LEVEL MS NON STANDARD


9. SAMPLING PROCEDURES
BACKGROUND -2 PER EVB^T OR ^> ^AXY\P(_.SS
CRITICAL (LIST) T\>0 Ci C(-gp\ QPC^ Dlft-S^TtO/O
PROCEDURES 5"FL lT~
^yPdrSl/O SFvmA-JNifi-

10. QUALITY CONTROL SAMPLES (CONFIRM OR SET STANDARD)
A. FIELD ^~. B. LABORATORY
COIiOCATED ^jyOH _ PBARPMT B] IWtC - 1 DCQ AKIAI VOIO OATOU OQ
REPLICATE jSitoR 	 ...
FIELD BLANK fsgftR
TRIP BLANK /b<5-/5T
REPLICATE - 1 PFR ANAI YSIS BATCH OR
	 . MATRIX SPIKE . 1 PFB AWAI VBK BATCH on 	 	
	 OTHER

 & i t-t-E^fi-S \

CONTRACTOR
SITE MANAGER

PRIME CONTRACTOR
PATE

FOR DETAILS SEE SAMPLING & ANALYSIS PLAN
                                                                 CDM SF DQO 1.002
                                      6-13

-------
Representativeness - A sampling grid has been defined to ensure representativeness of the soil samples.

Completeness - The RAS CLP historical completeness is approximately 80-85 percent. If valid analytical
results are not obtained for the clean samples, a new sample aliquot will be analyzed.

Comparability'- The use of standard soil sampling procedures and recognized field and laboratory
techniques should make the resulting data comparable with other similiar measurements on similiar
samples.

6.3    DQO STAGE 3 - DESIGN DATA COLLECTION PROGRAM:  PHASE II REMEDIAL
       INVESTIGATION

As the data collection  documentation (i.e., the work plan,  S&A plan and QAPjP) was developed prior to the
initiation  of Phase I, the discussion at this point will focus on Phase II elements.  Whereas the work
plan completed at the  start of the RI generally discussed the anticipated Phase II tasks, the S&A plan
was specific for Phase I elements.  The development of the S&A plan for Phase II is undertaken following
evaluation of Phase I data.

6.3.1    ASSEMBLE DATA COLLECTION COMPONENTS: PHASE II REMEDIAL
         INVESTIGATIONS

The S&A plan developed for Phase II will account for all sampling tasks and phases. Table 6-4 provides  a
summary of Phase II data collection components for the example site.  The schedule for  these activities
is shown  in Figure 6-3.

6.3.2    DEVELOP DATA COLLECTION DOCUMENTATION: PHASE II REMEDIAL
         INVESTIGATION

For Phase II of the RI, S&A components  will be prepared for each individual activity including:

     •   New well installation and sampling

     •   Soils sampling

The detailed information to be provided is similar  to that discussed for Phase I elements and will not be
repeated here.

6.4    DQO STAGE  1 - COLLECT AND EVALUATE DATA:  PHASE II REMEDIAL
       INVESTIGATIONS

This section presents a general review of the data  collected during Phase II of the RI.  In addition,
these results will be taken together with Phase I results to form an overall evaluation of the site.

6.4.1    ANALYSIS OF RESULTS:  RI PHASE HA - GROUND WATER INVESTIGATIONS

Five wells were sampled and analyzed for volatile  organics and metals during Phase II.  The well
locations  are shown in Figure 6-1.  The results obtained are shown below:
                                            6-14

-------
                             TABLE  6-4
            DATA COLLECTION COMPONENTS - PHASE  I!
SAMPLE
Rl PHASE
   2A
   2B
   MEDIA
                               TYPE
              NUMBER OF
              SAMPLES
GROUND WATER
SOIL
GRAB
                               AUGER
QA/QC SAMPLES


 1 DUPLICATE
 1 SPIKE
                                  72 (ORGANICS)   18
                                  24 (METALS)
                                6-15

-------
                                           TIME (WEEKS)
 TASK
                                               8    10
12
14     16
PHASE  II
PHASE  HA

 •  INSTALL/SAMPLE
  MONITORING WELLS
  (INCLUDING ANALYSIS)
 PHASE  IIB

  • SOIL SAMPLING
 DATA EVALUATION
                                     FIGURE  6-3
                   PHASE 51 REMEDIAL INVESTIGATION SCHEDULE

-------
            Well
            OWl

            OW2

            MWI

            MW2

            MW3
     Metals

        ND

        ND

        ND

        ND

        ND
Volatile Organics

TCE - 62 ug/1

ND

ND

ND

TCE - 47 ug/1
ND - under metals indicates no significant concentrations detected.
ND - under volatile organics indicates not detected above method detection limit

In wells OW1 and MW3 volatiles other than TCE were not detected.

No volatile contaminants were detected in monitoring wells MWI or MW2 so these wells will be used to warn
of the encroachment of volatile organics toward the residential wells.  These two wells also mark the
eastern extent of the  ground water plume.  Finally, these wells verify that soil gas correctly indicates
the extent of the ground water plume.  Thus, the north,  south and western limits of the ground water
plume can be extrapolated from the measured soil gas plume.

Wells OW1 and MW3 are located within the soil gas plume.  These two wells show a small difference in TCE
concentration.  The observed difference in TCE concentration between these two wells  (15 ug/1) can be
attributed  to analytical error.  Based on  the small difference observed between these concentrations,
there is no indication that TCE varies erratically within the plume.

Well OWl has been sampled by the FIT team and twice during this investigation.  The analyses obtained for
well OW1 are:
     Sample
           1
           2
           3
Obtained During
    FIT
    RI Phase 1
    RI Phase 2
TCE Concentration (ppb)
            52

            68
            62
These data suggest an increase in concentration over time. By examining the precision of the
analytical  method (EPA method 624) and comparing this value with the fluctuations in
concentration observed in well OWl  it is possible to state whether the increase in
concentration seen in well OWl  is significant.

The historical precision for method 624 is 17 percent RSD.  The precision observed in the
three samples from OWl is calculated below.

The definition of percent RSD is:

      % RSD  = [2 | X1  - X2 | / (Xt +X2)] (IOO//2)

                             Where Xj,  is measurement #1 of a replicate

                                     X2  is measurement #2 of a replicate
                                              6-17

-------
For the three analyses from OW1 percent RSD is:
                  Sample #'s
                      1, 2
                      2,3
                      1,3
                   %RSD

                   18.8
                   6.5
                   12.4
The average of these three values is 12.6 percent. This value is less than the historical precision of
the analytical method (17 percent).  Since the observed variation in the samples from well OW1 is less
than the expected variation in the analytical  method,  the observed increase in concentration in well OW1
over time may simply be due to analytical variability rather than an actual increase in TCE concentration
in the well.
6.4.2    ANALYSIS OF RESULTS:
         RESULTS
RI PHASE IIB - SUBSURFACE SOIL SAMPLING
Soil samples from the depth intervals 7-9 ft and 13-15 ft were obtained from 36 boreholes.  In addition,
the four boreholes identified in Figure 6-2 were sampled at 6-in.  intervals to a depth of 3 ft. The
results obtained are summarized  below for lead and TCE.  The results for other contaminants of concern
are analogous  and, for sake of brevity, will not be discussed here.

                         MEAN CONCENTRATION (mg/kg)

                                   Lead                    TCE

                                   432                      NA
                                   189                      NA
                                    67                      NA
                                    10                      NA
                                    ND                     NA              .
                                    ND                     NA
                                    ND                     3.9
                                    ND       '               .87
ND - No concentrations detected above background
NA - No analysis performed

The results indicate that lead has not migrated below a depth of 2 ft and the bulk of TCE contamination
has not migrated to a depth of 14 ft.  In the horizontal plane, TCE contamination is much more widespread
at a depth of 8 ft than at a depth of 14 ft.  Since contamination found at 8 ft will be of greater
importance in the assessment of remedial alternatives, the discussion of results will center on the 8-ft
depth interval. The data analysis procedures demonstrated on the data from the 8-ft depth interval are
equally applicable on the data from the 14-ft depth interval.

The analyses of samples taken at a depth of 8 ft are shown in Figure 6-4.  The contour  line shown on this
figure is the best estimate of the line separating soil containing less than and greater than 4 nig/kg
TCE.  Due to analytical and sampling errors and  the intrinsic variability  of the contaminants, the exact
location of this line is uncertain.  Uncertainty in the location of this line  indicates that there is
                                              6-18

-------
      .08       .06
      »
      1.1
      »
      1.9
      :80
      o
      .05
             3.3
      .05      .08
•
.07
                     .07
               •
               .05
       .06      .07
                                4.0
                         4.0.
                     3.8
•
3.1
                                           1.1
                                           .90
                                           .07
                                           .05
                                    N
                                    t
                                   .05      .05
      SCALE (FT)
     BBE-^
     L^—jBm
     0 . •  25   50
                 FIGURE 6-4
SOIL SAMPLING  RESULTS  (DEPTH  8 FEET)
    AND  KRIGED  CONTOUR LINE (4ppm)
                       6-19

-------
some chance that material located outside the 4 ppm TCE contour line (and therefore assumed to contain
less than 4 ppm TCE) might, in fact, contain more than 4 mg/kg TCE.

To assess the likelihood that soil which is assumed to contain less than 4 mg/kg actually contains more
than 4 mg/kg advanced geostatistical methods can  be used (see Appendix A of the Development Process
document).  This procedure directly determines the probability that soil  at a particular location exceeds
4  ppm.  This probability value is based on all  measured site specific uncertainties and quantifies the
total uncertainty surrounding a particular measurement.  These probability values provide valuable
information for assessing the volume of material which must be removed from a  site.

Using advanced kriging, thevprobability that TCE  concentration exceeds 4 ppm at each location within the
depressed area can be determined. Once determined these probability levels can be contoured.  One such
contour corresponding to a 15 percent chance of exceeding 4 ppm is shown in Figure 6-5.  Soil within  this
contour line has greater tKan a 15 percent chance of exceeding 4 ppm TCE while soil outside this of
contour line has less than a 15 percent of exceeding 4 ppm. Thus, if soil is removed to the 15 percent
probability contour line, the remaining soil has, at most, a 15 percent chance of containing more than 4
ppm TCE.

The previous statistical analysis provides a measure of uncertainty surrounding the volume of material
which should be removed as  part of a remedial action.  This uncertainty is represented in Figure 6-5 by
the distance between the 4 ppm contour line and the 15 percent probability of exceeding the 4 ppm line.
Based on the uncertainty represented in such a figure, the decision maker can assess whether the
available data are sufficient to reach a decision. If the data are insufficient to reach a decision, an
uncertainty  contour map can  be used to indicate the area of greatest uncertainty and can define the
location of samples which must be taken during the remedial design phase to reduce uncertainty to an
acceptable level.  In  this example, any additional samples would be located between the 4 ppm and  15
percent probability lines since samples in this area would produce the greatest reduction in  uncertainty.

The uncertainty represented by the distance between the 15  percent contour line  and the 4 ppm contour
line is qualitative, however, it can be converted into a quantitative volumetric uncertainty.  In Figure
6-5 a single probability line (.15 percent) is shown.  There are however a family of probability lines
corresponding to  25  percent, 35 percent, 45 percent etc. probability of exceeding 4 ppm. If all of these
contours were plotted, the volume between the 35 and 45 percent contour lines has, on average,  a 40
percent chance of exceeding 4 ppm.  Thus, 40 percent of the volume between these contour lines can be
expected to  exceed 4 ppm.

6.5    EXTENSION OF THE DQO PROCESS TO THE REMEDIAL  DESIGN (RD)
       AND REMEDIAL ACTION (RA) OF UNCONTROLLED HAZARDOUS WASTE
       REMEDIAL RESFONSEACTIVITIES

At the conclusion of the RI/FS and after the preparation of the Record of Decision by EPA, the design
phase at the site would commence.  At this stage in the analysis the planned remediation for the site has
been selected, costs have been estimated and design has started to a limited degree. During the RD/RA
phase the preliminary design and cost estimates prepared  during  the RI/FS will be refined which will
allow the design contractor to effectively implement the proposed remedy.

To plan  the required work involved with tfie RA phases of work at the site, the DQO process will be used
to (1) identify data gaps that  may exist (2) plan additional sampling activities, and (3) collect the
appropriate level  of data needed to design and implement  the selected  remedy.

The DQO process that will be used during the RD/RA phase of work will be essentially identical to the
process described for the  RI/FS evaluation.  The  only difference in the  execution of the RD/RA DQO will
be in setting the objectives for these phases.
                                               6-20

-------
CONTOUR OF POINTS
•WITH 15% CHANCE
OF EXCEEDING 4ppm
   ' 4.0ppm
 •CONTOUR UNE
 AS DETERMINED
   BYKRIGING
  SCALE (FT)
 0   25    50'
N
                 DATA POINTS
                                 FIGURE   6-5
                  KRIGED  PROBABILITY  CONTOURS
                                   6-21

-------
A critical part of the design development of remedial measure would include defining the operating and
monitoring parameters to insure that: (1) the full extent of the media contaminated will be treated. (2)
the remediation performs in accordance with  design specifications; and (3) the remediation is not
resulting in the release of contaminants into the environment.  The contract documents, plans and
specifications, will have to define, through the,use of the DQO process, the methods,  equipment, action
levels and detection limits to monitor the extent of remediation required, the effectiveness of the
remediation in terms of the cleanup of the affected media and the potential for the release of
contaminants to the environment associated with the remediation.

The key part of the RD/RA DQO will be the evaluation of existing data.  Since the data collection program
for the RI/FS should consider the data needs of the RD/RA phase (at least in  a general manner), at many
sites the data collected during the RI/FS will  of sufficient detail and quality to be used for the RD/RA.
However, at the initiation of the RD/RA phases, the data requirements  will be reviewed to insure that all
design cost estimates can be developed to the accuracy  required.

The analysis of the data obtained during this  two-phased remedial investigation can be used to assess
potential remedial alternatives and are sufficient to develop cost estimates which are within +50 percent
and -30 percent of the actual cost of implementation.  The process by which potential  remedial
alternatives are evaluated and a viable remedial action selected, is beyond the scope of this document.
However, the DQO process does  continue beyond the RI/FS and into the remedial design and remedial action.

For example, if soil removal was  chosen in the Record of Decision, one question of interest would be
whether it is necessary or desirable to treat the zones of metals and organic contamination separately.
Additional sampling might be required during the RD phase to lower the uncertainty surrounding the volume
of metals or volatile organics which must be  removed to a level consistent with the + 15 percent and -10
percent cost uncertainty associated with  the RD phase.  The analysis of uncertainty, including the
location and number of required additional samples, can be performed  using geostatistical methods such as
advanced kriging.
                                               6-22

-------

-------

-------
                                      7.0  CONCLUSIONS
This example demonstrated the application of DQOs to an RI at a fictitious hazardous waste site.  The use
of DQOs required that the uses and needs for each data type be specified at the project outset and be
consistent with  project objectives.  Once data uses were specified, the quality and quantity of data
required were determined. The DQO process, incorporated with development of the S&A plan, QAPjP, and
work plan, ensured that data of sufficient quality to meet project objectives were obtained.

The tangible results of applying DQOs appear in  cost savings.  Sampling costs are reduced by using the
conceptual model as a guide in determining the number of samples  required. The conceptual  model is
refined continually .as information is gathered during an investigation.  Thus, data quantity needs are
also continually refined.  The use of a sampling methodology which conforms with the conceptual  model can
significantly reduce the number of samples obtained.

Analytical costs are reduced when DQOs are applied since the chosen analytical method will be the least
expensive option which meets  all project objectives. An analysis of the possible analytical options
together with specified data uses will ensure that appropriate data quality (as defined by the level of
analyses) is obtained for each  specified data use.

The DQO is not a' separate deliverable. The analysis of sampling and analytical options provided in this
example document will not appear explicitly in either the work plan or sampling and analysis plan.
However, it is envisioned  that the analysis presented in this example will occur during meetings and
phone conversations between primary data users and the rationale behind the selection of a particular
sampling and analysis option will appear in meeting minutes or internal inemos which will become part of
the project file.  The  result of the DQO process will be a well thought out sampling and analysis plan
which details the chosen sampling and analysis option.
                                               7-1

-------

-------
           APPENDIX A

HISTORICAL PRECISION AND ACCURACY
     DATA CLASSIFIED BY MEDIA
       BY ANALYTICAL LEVEL

-------

-------
                    APPENDIX A CONTENTS
       HISTORICAL PRECISION AND ACCURACY  TABLES
Introduction

Table A-l-C
Table A-l-D

Table A-2-A
Table A-2-B
Table A-2-C
Table A-2-D

Table A-3-A
Table A-3-B
Table A-3-C

Table A-4-C
Water: Level III
Water: Level IV

Soil:  Level I
Soil:  Level II
Soil:  Level III
Soil:  Level IV

Air:  Level I
Air:  Level II
Air:  Level III

Other Media: Level III
                              A-l

-------
                                       INTRODUCTION

The data in this Appendix have been compiled to assist the reader in selecting an analytical
method appropriate for each data use.  The methods are classified by media and by analytical
levels defined as follows:

     •    Level I - field screening or analysis using portable instruments.  Results
           are often not compound specific and not quantitative but results are
           available in real-time.

     •    Level II - field analysis using more sophisticated portable analytical
           instruments; in some cases, the instruments may be set up in a mobile  or
           onsite laboratory. There is a wide range in the quality of data that can  be
           generated.  Quality depends on the use of suitable calibration standards,
           reference materials, and sample preparation equipment; and the training of
           the operator.  Results are available in real-time or several hours.

     •    Level III - all analyses performed in an offsite analytical laboratory using
           standard, documented procedures.  The laboratory may or may not be a CLP
           laboratory.

     •    Level IV - CLP routine analytical services (RAS).  All analyses are performed
           in an offsite CLP analytical laboratory following CLP protocols.

Precision and accuracy data are presented in tabular fashion.  Footnotes to each table cite
the sources of the data and the concentration or concentration range at which the precision
and accuracy were determined.  When no concentration is cited no concentration information
was available in the source material.

Precision is a measure of the variability in repeated measurements of the same sample
compared to the average value.  Precision is reported as  % Relative Standard Deviation (RSD).
The lower the .% RSD, the more precise the data.

RSD is calculated for a pair of replicates using the following formula:

                                %RSD  = PX-X/CJ+X)] (100//2)
                               where Xx  is measurement #1 of a replicate

                               X2 is measurement #2 of a replicate


 Accuracy is reported as % Bias; as % Bias approaches zero, accuracy increases.  Bias is
 calculated by the following formula:

                                % Bias =   X-Y (100)
                                             Y

                           where Y  is the known concentration or true value

                           X is the reported concentration


 Bias measures the systematic error within an analytical technique.
                                               A-2

-------
                         TABLE A-1 -C: HISTORICAL PRECISION AND ACCURACY DATA/WATER a
             LEVEL III ANALYTICAL TECHNIQUES - METHODS OTHER THAN CLP RAS METHODS
            ANALYTES
            BRCM3DICHLOROMETHANE
T
OJ
            BROMOFORM
METHOD C
(TECHNIQUE)
624
(GC/MS)
8240
(GC/MS)
624
(GC/MS)
501.1
(PURGE & TRAP GC/MS)
501.2
(EXTRACTION GC/MS)
624
(GC/MS)
501.1
(PURGE & TRAP GC/MS)
501.2
(EXTRACTION GC/MS)
CONCENTRATION
RANGE
11 ug/l
480 ug/l
5-100 ug/l

8 ug/l
480 ug/l
0.9 ug/l
550 ug/l
1.8 ug/l
170 ug/l
9 ug/l
400 ug/l
4.8 ug/l
550 ug/l
6 ug/l
170 ug/l
PRECISION
%RSD
16
21
21

28
18
66
34
61
23
32
30
44
41
14
15
ACCURACY
% BIAS
0
-16
12

-8.8
-6.7
0
-3.8
33
-19
-23
10
-27
7.5
-23
1.8

-------
TABLE A-1-C: HISTORICAL PRECISION AND ACCURACY DATA/WATER a
                       (continued)
ANALYTES
CHLOROFORM


DIBFOO>ILQROMETHANE


DIOXIN
METHOD CC
(TECHNIQUE)
624
(GC/MS)
501.1
(PURGE & TRAP GC/MS)
501.2
(EXTRACTION GC/MS)
624
(GC/MS)
501.1
(PURGE & TRAP GC/MS)
501.2
(EXTRACTION GC/MS)
613
(GC/MS)
3NCENTRAT10N
RANGE
4.5 ug/l
300 ug/l
0.9 ug/l
550 ug/l
1.8 ug/l
170 ug/l
8.1 ug/l
360 ug/l
0.8 ug/l
550 ug/l
1.8 ug/l
170 ug/l
21 ng/l
202 ng/l
PRECISION
%RSD
31
14
64
14
68
26
13
19
35
36
37
13
25
21
ACCURACY
% BIAS
2.2
-0.6
44
-0.02
-39
-1.2
-3.1
10
-12.5
4.7
0
0.02
N.A.
N.A.

-------
TABLE A-1-C: HISTORICAL PRECISION AND ACCURACY DATA/WATER a
(continued)
LEVEL III ANALYTICAL TECHNIQUES - METHODS OTHER THAN CLP RAS METHODS
ANALYTES METHOD
(TECHNIQUE!
METHYLENE CHLORIDE . 624
._..._ 	 (GC/MS)
TOLUENE 624
(GC/MS)
8240
(GC/MS)
TRICHLOROETHENE 624
!f (GC/MS)
Ul
8240
(GC/MS)
1£A£ 200.7
(ICP)
239.1
(FLAME AA)
239.2
(FURNACE AA)
CONCENTRATION
RANGE
7.2 ug/l
480 ug/l
13.5 ug/l
600 ug/l
25 ug/l
75 ug/l
5.4 ug/l
360 ug/l
25 ug/l
75 ug/l
42 ug/l
47.7 ug/l
12 ug/l
105 ug/l
10 ug/l
234 ug/l
PRECISDN
% RSD
78
52
19
31
19
48
39
24
34
5
5.9
6.7
53
19
ACCURANCY
% BIAS
-17
-25
15
-14
-10
44
-2.3
5
31
4.4
17
-1.9
-22
-3.1
a.  Source: Draft Compendium of Information and Performance Data on Routinely Used Measurement Methods (RUMM) - Pilot Phase,
           RTI/3087/03, prepared for EPA Quality Assurance Management Staff, January 1986.  This document should be
           consulted for more information on individual analytes.

-------
                        TABLE A-l-C:  HISTORICAL PRECISION AND ACCURACY DATA/WATER
                                                (Continued)
                                         LEVEL III SW-846 METHODS



T"
ON


Method
Number
ORGANICS:
8010
8020
8030
8040
8060
8080
Method Name
•
•
Halogenated. Volatile Organics
Aromatic Volatile Orariics
Acrolein, Acrylonitrile,
Acetonitrile
Phenols
Esters
Ofganochlorine Pesticides
Data
Source
SW 846
SW 846
SW 846
SW 846
EPA 606
SW 846
Range of
Recovery (%)
75.1
77.0
96 -
41 -
82 -
86 -
- 106.1
- 120
107
86
94
97
Precision
(%)
2.0
9.4
5.6
7.9
1.3
1.3
- 25.1
-- 27.7
- 11.6
- 16.5
- 6.5
- 6.5
MDL
(mg/L)
0.03 -
0.2 - 0
0.5 - 0
058 - 2
0.29 -
0.29 -
0.52
.4
.6
.2
3.0
3.0
8090

8100

8120
8140
8150
8240
8250

8040
and PCBs
Nitroaromatics and Cyclic      SW 846
Ketones
Polynuclear Aromatic
Hydrocarbons
Chlorinated Hydrocarbons       SW 846
Organophosphorous Pesticides   SW 846
Chlorinated Herbicides         SW 846
Volatile Organics              SW 846
GC/MS Semivolatiles (Packed
Column)
GC/MS Semivolatiles
(Capillary)
63 - 71

NAb

76-99
56.5 - 120.7
NA
95 - 107
41 - 143

NA
3.1 - 5.9

NA

10 - 25
5.3 - 19.9
NA
9 - 28
20 -145

NA
0.06/ND

NA

0.03 - 1.34
0.1 - 5.0
0.1 - 200
1.6 - 6.9
0.9 - 44

NA

-------
                                TABLE A-l-C:  HISTORICAL PRECISION AND ACCURACY DATA/WATER
                                                        (Continued)
                                                 LEVEL III SW-846 METHODS
Method
Number
Method Name
Data
Source
Range of
Recovery (%)
Precision
.(%)
MDL
(mg/1)
T
8310         Polynuclear Arc-malic           SW 846       78 - 1-16
             Hydrocarbons (HPLC)
             (Capillary)
INORGANICS;  Metals (ICAP)                  EPA 200.7    NA
             Metals (FLAME) 7000 Series     EPA 200      NA
7000 Series  Metals (FLAME LESS/GF)         EPA 200      NA
7470         Metals (MERCURY)               EPA 245.2    87 - 125
9010         Cyanides                       EPA 335.2    85 - 102
9030         Sulfides                       EPA 376.1    NA
                                                                                 7.3 - 12.9
3 - 21.9 (RSD)
NA
NA
0.9 - 4.0
0.2 - 15.2
NA
                 0.03 - 2.3
1.3 - 75 Mg/1
0.01 - 5
0.001 - 0.2 Mg/1
0.0002
0.02 Mg/1
1 Mg/1
        a.   For water only
        b.   NA Not Available
        NOTES:  Method Detection Limit (MDL)  as listed on this table is the minimum concentration of a substance
                that can be measured and reported with 99% confidence that the value is above zero.
                Accuracy, presented as an average percent recovery,  was determined from replicate (10-25)  analyses
                of water and wastewater samples fortified with known concentrations of the analyte of interest at
                or near the detection limit.   In most cases this was less than 10 times the MDL.
                Precision data are used to measure the Variability of these repetitive analyses reported as a
                single standard deviation or, as a percentage of the recovery measurements.  For presentation
                purposes accuracy, precision and MDL information is presented as an average range of individual
                values for every analyte covered by the procedure.  If specific information on a particular
                compound is required, the specific analytical method cited should be consulted.

-------
                                    TABLE A-l-D:  HISTORICAL PRECISION AND ACCURACY DATA/WATER3

LEVEL IV ANALYTICAL TECHNIQUES - CLP HAS METHODS
T
oo
ANALYTES

Volatilesb
   Methylene chloride
   1,1-Dichloroethene
   1,1-Dichloroethane
   Trans-1,2-Dichloroethene
   Chloroform
   1,2-Dichloroethane
   1,1,1-Trichloroethane
   Carbon Tetrachloride
   1,1,2,2-Tetrachloroe thane
   Bromodichloroiae thane
   1,2-Dichloropropane
   Trans-1,3-Dichloropropene
   Trichlproethene
   Dibromochloromethane
   1,1,2-Trichloroethane
   Benzene
   Cis-1, 3-Dichloropropene
   Bromoform
   Tetrachloroethene
   Toluene
   Chlorobenzene
   Ethyl Benzene

Semivolatiles
   bis(2-Chloroethyl)ether
   2-Chlorophenol
   1 -, 3-Dichlorobenzene
   1,4-Dichlorobenzene
   1,2-Dichlorobenzene
   2-Methylphenol
   bis(2-Chloroisopropyl)ether
                                  TECHNIQUE

                                  Purge & Trap GC/MS
CONCENTRATION
   RANGE

    N.A.C
                                  GC/MS
    N.A.
PRECISION
 % RSD
                                   56
                                   20
                                   13
                                   31
                                   12
                                   13
                                   19
                                   12
                                   11
                                   19
                                   18
                                   31
                                   17
                                   14
                                   11
                                   12
                                   22
                                   1.6
                                   13
                                   14
                                   14
                                    4
                                                                                            24
                                                                                            29
                                                                                            24
                                                                                            21
                                                                                            29
                                                                                            29
                                                                                            25
ACCURACY
% Bias
                   +36.6
                   -26.3
                   -46.4
                   -21.7
                   -21.1
                     +2.4
                   -41.
                   -32.
                     -5.8
                   -13.0
                   -12.9
                   -41,2
                   """A*** * O
                     -3.3
                     -7.0
                     -3.3
                   -35.5,
                     +6.5
                   -42.5
                   -23.3
                                                                                                               .0
                                                                                                               .1
                                                                                                            -15.9
                                                     -16
                                                     -21
                                                     -48
                                                     -25
                                                     -28
                                                     -30
                                                     -22

-------
                                     TABLE A-l-D:  HISTORICAL PRECISION AND ACCURACY DATA/WATER

  LEVEL IV  ANALYTICAL TECHNIQUES - CLP RAS METHODS
T
ANALYTES

Semivolatiles
   4-Methylphenol
   N-Nitroso-di-n-propylamine
   Nitrobenzene
   Isophorone
   2-Nitrophenol
   bis( 2-Chloroethoxy) me thane
   2,4-Dichlorophenol
   1,2,4-Trichlorobenzene
   Naphthalene
   4-Chloro-3-methylphenol
   2,4,6-Trichlorophenol
   2-Chloronapthalene
   Acenapthene
   2,4-Dinitrophenol
   2,4-Dinitcbtoluene
   2,6-Dinitrotoluene
   4-Chlorophenyl-phenylether
   Fluorene
   4,6-Dini tro-2-methylphenol
   4-Bromophenyl^phenylether
   Hexachlorobenzene
   Pentachlorophenol
   Phenanthrene
   Fluoranthene
   Benzo(b)fluoranthene
   Benzo(a)pyrene
                                   TECHNIQUE

                                   GC/MS
CONCENTRATION
   RANGE

      N.A.C
PRECISION
  % RSD
                               33
                               31
                               32
                               23
                               30
                               34
                               29
                               30
                               44
                               26
                               25
                               24
                               28
                               24
                               34
                               25
                               34
                               25
                               30
                               32
                               36
                               31
                               21
                               42
                               39
                               42
ACCURACY
 % Bias
                       -36
                      +0.3
                       -23
                        -8
                       -21
                      -2.6
                       -20
                       -47
                       -38
                       -32
                       -17
                      +3.4
                       -12
                       -23
                       -33
                       -48
                       +12
                       -24
                       -13
                      -0.1
                       -42
                       -24
                       -28
                       -15
                       -10
                       -29

-------
                                     TABLE A-l-D: HISTORICAL PRECISIOi AND ACCURACY DATA/WOER
                                                            (continued)
LEVEL  IV ANALYTICAL TECHNIQUES - CLP HAS METHODS
                                                           OCNCENIRATION
ANALYTES
Metals6
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
> Cower
1 *-**cTr**i
i- iron
0 Lead
Magnesium
Manganese
Mercury
Nickel
Potassium
Selenium
Sodium
Thallium
Tin
Vanadium
Zinc
TECHNIQUE

ICP
ICP
Furnace AA
ICP
ICP
ICP
ICP
ICP
ICP
ICP
ICP
Furnace AA
ICP
ICP
Cold Vapor
ICP
ICP
Furnace AA
ICP
Furnace AA
ICP
ICP
ICP
RANGE

1000-3000 ug/1
180-600 ug/1
50-150
800-1500
30-45
25-50
1000-30000
50-150
200-1000
125-250
200-800
30
10000-40000
30-150
5-20
160
10000-20000
50
10000-45000
80-100
160
60-200
50-800
PRECISION
  % RSD
                                                                                      9.1
                                                                                       11
                                                                                      9.4
                                                                                      6.8
                                                                                       15
                                                                                       12
                                                                                      6.0
                                                                                      9.8
                                                                                      6.7
                                                                                      6.7
                                                                                     10.4
                                                                                       32
                                                                                      6.6
                                                                                      6.2
                                                                                     18.8
                                                                                      9.0
                                                                                     16.2
                                                                                      8.7
                                                                                      8.7
                                                                                     7.6
                                                                                     9.1
ACCURACY
 % Bias
                          -4.3
                          -9.2
                          -8.3
                          -3.9
                          +3.7
                          -3.3
                          -1.6
                          -2.6
                          -2.9
                          -1.1
                          +6.5
                          -0.7
                          -2.5
                          -1.0
                        -14.4
                          -2.5
                        -12.1
                          -5.7
                          -2.8
                          -4,2
                          -2.5
                        -0.46
                          +3.0
a.  Source:  Quality Control in Remedial Site Investigation:  Hazardous and Industrial Solid Wast* testing, Fifth Volume,
    ASTM SIP 925, C.L. Perket, Bd., American Society for Testing Materials, Philadelphia, 1986.

b.  Volatile precision and accuracy data fsrcsa 26-34 laboratories' results on quarterly Mind performance evaluation
    samples; 29-152 data points for each conpound.

c.  N.A. * Not Available.

d.  Semi volatile precision and accuracy data from 1985 preaward program data; 22-227 data points for each compound.

e.  Netals precision and accuracy data is based on performance evaluation sample results from 3.8 laboratories; number
    of data points is not given.

-------
                          TABLE A-2-A:  HISTORICAL PRECISION AND ACCURACY DATA/SOILS
LEVEL I FIELD SCREENING TECHNIQUES
MEASUREMENT
RESISTIVITY
TERRAIN
CONDUCTANCE
TERRAIN
OM3OCTANCE
Magnetic Field
Intensity
Subsurface
Lithology
Changes
Subsurface
Lithology
Changes
INSTRUMENT
(TECHNIQUE)
Bison 2390 T/R
(Resistivity meter)
EM 31
(conductivity)
EM 34-3
(conductivity)
EDA - Omni IV
(Magnetometer)
SIR-8
(Ground Penetrating
Radar)
EG+G 1225
(Seismograph)
INSTRUMENT
RANGE
0-1999
millivolts
0-1000
millimhos/meter
0-300
millimhos/meter
18000-110000
gassnas
1-81 dielectric
constant
0-2000
milliseconds
INSTRUMENT .
PRECISION
at 1% range setting,
0-5% of full scale
2% of full scale
2% of full scale
0.02 gaifflna
N/Ad
N/&d
INSTRUMENT
ACCURACY C
/
2% of measured
value
5% at 20 millimhos/meter
5% at 20 millimhos/meter
1 ganma at 50000 gananas
at 23oC
N/Ad
0.01%

-------
T
i—•
ho
                              TABLE A-2-A:
    LEVEL I FIELD SCREENING TECHNIQUES
HISTORICAL PRECISION AND ACCURACY DATA/SOIL3
          (continued)
INSTRUMENT FIELD SCREENING
CLP
MEASUREMENT (TECHNIQUE) RESULTS in ppm (X) RESULTS in Don m
TOTAL PHOTO VAC
VOLATILE (GC/Photoionization)
ORGANICS







11.4
22.0
56.0
139
70.0
24.9
60.0
6.6
12.1
8.7
26.9
32.8
129.7
228.0 & 258.0
126.7
2823.0
53.3
0.056
0.032
0.024
ACCURACY6
(% Bias)
-57.6
-32.9
-56.8
-42.8
-44.8
+99.1
-5-12.6
+116.9
+377.1
+361.5
   a.  Source:   Manufacturers' manuals unless otherwise cited.   Mention of specific models does not constitute
       and endorsement of these instrument.

   b.  Precision refers to reproducibility of meter or instrument reading as  cited in instrument specifications.

   c.  Accuracy refers to instrussnt specifications unless otherwise cited.

   d.  N.A. » not available.

   e.  Accuracy of PhotoVac field screening results calculated by assuming that CLP results on the same samples
       were completely accurate.    % Bias - 100  (X-Y)..  Source of these data  is CDM project files.
                                                  Y

-------
                                       TABLE A-2-B: HISTORICAL PRECISION AND ACCURACY DATA/SOIL
             LEVEL IIFIFID TECHNIQUES
T
ANALYTES INSTRUMENT FIELD R&ULTS
(TECHNIQUE) IN ppm (x)
PCBs HNU301 6.0
(GC/ELECTRON 6.0
CAPTURE) 6.0
9.0
13.0
14.0
14.0
21.0
35.0
41.0
48.0
50.0
65.0
87.0
92.0
95.0
11
202
269
286
1215
1 647
3054
CLP RESULTS
IN ppm (y)
22.0
6.1
510.0
3.9
3.0
3.1
23.5
8.1
7.7
2.1
11.0
460.0
23.1
18.7
75.0
30.0
12.3
99.0
370.0
80.5
640,0
1040.0
9,300
ACCURACY b
% BIAS
-72.7
-1.6
-98.8
+56.7
+333.3
+351.6
-40.4
+159.3
354.5
+1,852
+336.3
-89.1
+181.4
+258.3
22.7
+216.7
-10.6
+104.0
-27.3
+255.3
+90.0
+58.4
-67.2.
            a. Source: COM Project files.




            b. Source:  Accuracy calculated by assuming that CLP. results on the same samples were completely accurate. % Bias = 100

-------
                              TABLE A-2-C: HISTORICAL PRECISION AND ACCURACY DATA/SOIL
               LEVEL III ANALYTICAL TECHNIQUES - METHODS OTHER THAN CLP RAS METHODS
              ANALYTE
              DIOXINS
      METHOD
   . (TECHNIQUE)

        8280
     (HPLC/LRMS)

JAR EXTRACTION GC/MS
1
1-1
.£»
CONCENTRATION
    RANGE

     5ppb
   125 ppb

     1ppb
    10 ppb
PRECISION
 %RSD

  6-30
  3-10

    20
    10
ACCURACY
 % BIAS

    N.A.
    N.A.

      0
    -18
              a. Source:  Draft Compendium of Information and Performance Data on Routinely Used Measurement Methods (RUMM) - Pilot Phase,
                        RTI/3087/03, prepared for EPA Quality Assurance Management Staff, January 1986. This document should be
                        consulted for more information on individual analytes.

-------
                                    TABLE A-2-D:  HISTORICAL PRECISION AND ACCURACY DATVSOILS

LEVEL IV ANALYTICAL TECHNIQUES - CLP RAS METHODS
  ANALYTES

  Volatiiesb
     Chloroform
     1,2-Dichloroethane
     Dibronbchlorome thane
     Benzene
     Bromoform
     2-Hexanone
     Toluene
     Chlorobenzene

> Semivolatiles
,L    1,4-Di chlorobenzene
01    Nitrobenzene
     Isophorone
     2-Nitrophenol
     2,4-Dichlorophenol
     1,2,4-Trichlorobenzene
     Penta Chlorophenol
     Pyrene
     2-ffethylnaphthalene
     bis-(2-Ethylhexyl)phthalate
     Phenol
     Acenaphthylene
     Diethyphthalate
                                  TECHNIQUE

                                  Purge & Trap GC/MS
CONCENTRATION
   RANGE

       N.A.C
                                  GC/HS
        N.A.
PRECISION
 % RSD
                                                                                       8.0
                                                                                      13.1
                                                                                      35.0
                                                                                      32.1
                                                                                      16.6
                                                                                      16.6
                                                                                      13.8
                                                                                      21.2
Dioxin
    !73,7,8-TCCD
1-10    ugAg
                                                                                        27
                                                                                        21
                                                                                        24
                                                                                        35
                                                                                        31
                                                                                        28
                                                                                        17
                                                                                        25
                                                                                        26
                                                                                        33
                                                                                        38
                                                                                        26*
                                                                                        16
       15
ACCURACY
% Bias
                                                -0.1
                                               +11.1
                                               -12.0
                                               -10.3
                                               -12.1
                                               -45.5
                                               +13.7
                                               +13.2
                                                 -51
                                                 -48
                                                 -47
                                                 -36
                                                 -59
                                                 -43
                                                 -48
                                                 -15
                                                 -42
                                                  -2
                                                 -27
                                                 -27
                                                 -20
   -11.5

-------
                                    TABLE A-2-D:  HISTORICAL PRECISION AND ACCURACY DATA/SOILS
                                                            (continued)
LEVEL IV ANALYTICAL TECHNIQUES - CLP RAS METHODS
                                                         CONCENTRATION
ANALYTES
Metals*3
Aluminum
Cadmium
Calcium
Chromium.
Copper
Iron
Lead
> Magnesium
>L Manganese
^ Mercury
Nickel
Tin
Zinc
TECHNIQUE

ICP
ICP
ICP
ICP
ICP
ICP
Furnace AA
ICP
ICP
Cold Vapor
ICP
ICP
ICP
RANGE (ug/kg)

2-22600
5.5-20
2664-29000
8.5-29600
33-109
5028-113000
11.5-714
2428-7799
73.5-785
1.1-26.5
44-67
N.A.
19-1720
PRECISION
  % R3D
                                                                                           14.4
                                                                                           33.1
                                                                                          N.A.
                                                                                            7.8
                                                                                           11.
                                                                                           10.
                                                                                            9.2
                                                                                            7.5
                                                                                            9.4
                                                                                           25.0
                                                                                           15.0
                                                                                           44.1
       .2
       .7
                                                                                            5.8
ACCURACY
% Bias
   -78.8
    +2.9
    -4.2
    -6.1
    -2.5
   -27.0
    -2.2
   -10.6
   -15.1
    -9.1
   -17.g
   N.A.
    -6.2
a.  Source:  Quality Control  in Remedial Site Investigation:  Hazardous and Industrial Solid Waste Testing,  Fifth Volume,
    ASTM STP 925, C.L.  Perket, Ed., American Society for Testing Materials, Philadelphia, 1986.

b.  Volatiles precision and accuracy data is based on 1985 preaward analysis results from laboratories awarded
    contracts; 6-14  data points for each compound.

c.  N.A. =* Not Available.

d.  Semivolatiles precision and accuracy data is based on 1985 preaward analysis results; 9-20 data points
    for each compound.                                                                         .

e.  Dioxin precision and accuracy data is based on results of four performance evaluation samples including
    120 data points.

f.  Metals precision and accuracy data is based on performance evaluation sample results from 18 laboratories;
    number of data points  is  not given.

-------
                      TABLE A-3-A:  HISTORICAL PRECISION AND ACCURACY DATA/AIR'
                                                                              ,a
LEVEL I FIELD SCREENING TECHNIQUES



1
I--
-~l
ANALYTES
Organics
Organics
Organics
Organics
INSTRUMENT
(TECHNIQUE)
Century OVA-128
(Flame lonization)
HNu PI-101
( Photoionization)
AID - 710
(Flame lonization)
PhotoVac
(GC-Photoion-
ization)
INSTRUMENT
RANGE
0.1 - 1000 ppm
Methane
0.1 - 2000 ppm
Benzene
0.1 - 2000 ppm
Methane
N.A.
INSTRUMENT
SENSITIVITY
0.1 ppm Methane
0.1 ppm Benzene
0.1 ppm Methane
0.001 ppm
Benzene
INSTRUMENT
PRECISION
N.A.d
± 1% of full scale
deflection
N.A.d
N.A.d
a.  Source:  Manufacturers' manuals unless otherwise cited.  Mention of specific models
    does not constitute an endorsement of these instruments.

b.  It is difficult to differentiate between Level I and Level II techniques and
    instrumentation.  Several instruments may be used at both levels.

c.  Sensitivity and precision refer to instrument specifications.

d.  N.A. = Not Available.

-------
                             TABLE A-S-B:  HISTORICAL PRECISION AND ACCURACY DATA/MR
       LEVEL II FIELD TECHNIQUES
T
I—>
00
ANALYTES
Organics
Compound-
Specific
Organics,
Compound-
Specific
Organics,
Compound-
Specific
Organics,
Compound-
Specific
Mercury
(TECHNIQUE)
Mi ran IB
(Infrared)
Century OVA-128
(GC/Flame
lonization)
PhotoVac
(GC-Photo-
ionization)
SCENTOR
(Argon lonization
or Electron Capture)
Gold film Mercury
Analyzer
INSTRUMENT
RANGE
Compound Dependent,
0-2000 ppm
1-1000 ppm
Methane
N.A.
N.A.
N.A.
INSTRUMENT
SENSITIVTTYC
N.A.d
N.A.
0.001 ppm
Benzene
0.001 ppm
Benzene
less than
0.01 ppm
                                                                                        INSTRUMENT

                                                                                        PRECISION^



                                                                                        N.A.d
                                                                                        N.A.
                                                                                        N.A.
N.A.
                                                                                        N.A.
       a.  Source:  Manufacturers' manuals.  Mention of specific models does not constitute an

           endorsement of these instruments.


       b.  It is difficult to differentiate between Level I and Level  II  techniques and

           instrumentation.  Several instruments may be used at both levels.


       c.  Sensitivity and precision refer to instrument specifications.


       d.  N.A. = Not Available.

-------
                          TABLE A-3-C: HISTORICAL PRECISION AND ACCURACY DATA/AIR
         LEVEL III ANALYTICAL TECHNIQUES - METHODS OTHER THAN CLP RAS METHODS
i—1
vD
          ANALYTES
      METHOD
    (TECHNIQUE)

 CRYOGENIC TRAP/GC
                                       TENAX GC/MS
         TOLUENE
CONCENTRATION
    RANGE

   3.9 ppb
    93 ppb

   7.8  ug/m3
   4.5  ug/m3


  10.8 ppb
PRECISION
 %RSD

   4.0
   5.1

    11
    21
                                            5.11
ACCURACY
 % BIAS

    N.A.
    N.A.

    N.A.
    N.A.
                   N.A.
         TRICHLORQETHENE
                         3.5 ppb
                          84 ppb
                       4.1
                       3.7
                   N.A.
                   N.A.
         VINYL CHLORIDE
                         7.8 ppb
                      6.37
                   N.A.
         LEAD
40 CFR 50, APP G
   (FLAME AA)
   0.6  ug/m3
  8.01  ug/m3
   8.6
   3.9
      0
   -3.6
        a. Source:  Draft Compendium of Information and Performance Data on Routinely Used Measurement Methods (RUMM) - Pilot Phase,
                   RTI/3087/03, prepared for EPA Quality Assurance Management Staff, January 1986.  This document should be
                   consulted for more information on individual analytes.

-------
                             TABLE A-4-C: HISTORICAL PRECISION AND ACCURACY DATA/OTHER MEDIA*
        LEVEL III ANALYTICAL TECHNIQUES - METHODS OTHER THAN CLP RAS METHODS

        ANALYTE
       LEAD
  METHOD
(TECHNIQUE)

    6010
    (ICP)
N>
O
  MEDIUM

 OIL WASTE


SOLID WASTE
CONCENTRATION
    RANGE
PRECISDN
 % RSD
          1.0 mg/kg.    3.1
         -2.5 mg/kg     22
                                                50 mg/kg
                                                75 mg/kg
                       10
                       3.7
ACCURACY
 % BIAS

    -10
    -20

    3.4
   -0.8
                                       SOLID
                       SLUDGE
                            5 mg/kg      2
                           20 mg/kg     11
                                       0
                                      55
        a.  Source:  Draft Compendium of Information and Performance Data on Routinely Used Measurement Methods (RUMM) - Pilot Phase,
                  RTI/3087/03, prepared for EPA Quality Assurance Management Staff, January 1986. This document should be
                  consulted for more information on individual analytes.

-------
         APPENDIX B

CONTRACT REQUIRED DETECTION
   LIMITS FOR HSL ANALYSES
  USING CLP IFB PROCEDURES

-------

-------
                                    TABLE B-l


                            CLP VOLATILE ORGANIC CRDL
Taraet compound name
Ehioromethane
Bromomethane
Vinyl Chloride
Chi oroethane
Methyl ent Chloridt
Acetone
Carbon D1su1f1de
1 ,l-D1eh1oroethene
1.1-01 ehl oroethane
Trans- 1 .2-01 ehl oroethene
Chi orof orm
1,2-01 chl oroathane
2-Butanone
1 , 1 ,1-TH ehl oroethane
Carbon Tetrachloride
Vinyl Acetate
Bromodl chl ororoethane
1 ,1 ,2,2-Tetrachloroethane
1,2-01 chl oroprop«ne
Trans-1 ,3-01 ehl oropropene
Trlchloroethene
01 bromochl oromethane
1 , 1 , 2-Tr 1 chl oroethane
Benzene •
C1 s-1 ,3-01 chl oropropene
2-Chloroethyl Vinyl Ether
Bromoforra
4-Methyl -2-pcntanone
2-Hexanone
Tetrachl oroethene
Toluene
Ch1orobenzen«
Ethyl Benzene
Styrene
Total Xylenes
SPCC&
CCCC
SPCC

ccc




ccc
SPCC

ccc






SPCC
ccc







SPCC



ccc
SPCC
ccc


CRDl.
uQ/kg
10
10
10
10
5
10
5
5
5
5
5
5
10
5
5
10
5
5
5
5
5
S
5
S
S
10
5
10
10
5
5
5
S
5
5
Low water
CRDL,
uQ/L
10
10
10
10
5
10
5
S
5
5
5
S
10
5
5
10
5
5
S
5
5
5
5
5
5
10
S
10
10
5
S
S
S
S
S
CAS number
74-87-3
74-83-9
75-01-4
75-00-3
75-09-2
67-64-1
75-15-0
75-35-4
75-35-3
156-60-5
67-66-3
107-06-2
78-93-3
71-55-6
56-23-5
108-05-4
75-27-4
79-34-5
78-87-5
10061-02-6
79-01-6
124-48-1
79-00-5
71-43-2
10061-01-5
110-75-8
75-25-2
108-10-1
591-78-6
127-18-4
108-88-3
108-90-7
100-41-4
100-42-5
N.A.
          values  obtained from the IFB WA85-J6S4 1.7J.
    bSystea Performance  Check Compounds (SPCC) are used to check compound
     Instability  and  degradation 1n the GC/MS tnd to Insure minimum average
     response  factors are met prior to the use of the calibration curve.
    cColumn Check Compounds  (CCC) are used to cheek the validity of the
     Initial calibration.
     Note:   Medium soil  and  water CROLs are 100 times the low level CRDls.
SOURCE:   Flotard, R.D. et al 1986
                                        B-l

-------
                                TABLE B-2


                       CLP INORGANIC COMPOUND CRDL,
                 INSTRUMENT DETECTION LEVEL AND WAVELENGTH
Element
Al
Sb
As
Ba
Be
Cd
Ca
Cr
Co
Cu
Fe
Pb
Mg
Mn
Hg
N1
K
Se
Ag
Na
n
. Sn
V
Zn
CRDL
ZOO
60
10
200
5
5
5000
10
50
25
100
5
5000
15
0.2
40
5000
5
10
5000
10
40
50
20
Method
ICP
ICP
FAA
ICP
ICP
ICP
ICP
ICP
ICP
ICP
ICP
ICP
ICP
ICP
CV
ICP
ICP
FAA
ICP
ICP
ICP
ICP
ICP
ICP
N
7
5
18
5
10
5
7
9
11
11
10
12
11
10
12
9
8
18
10
9
18
7
10
0
IDL
Mean
70.7
42.3
4.6
22.1
2.3
4.0
529
5.8
11.4
9.7
27.4
2.3
385
5.2
0.2
17.8
668
2.8
5.4
756
4.3
23.8
13.1
8.4
KDL
Std Dev
59.3
11.3
2.3
31.7
1.7
1.1
472
2.9
8.5
6.5
20.9
1.2
449
4.6
0.1
10.1
444
1.3
2.7
864
2.4
8.4
10.0
6.3
Wave-
length (nm)
3d9.3
217.6
198.7
493.4
312.0
228.8
317.9
267.7
228.6
324.5
259.9
283.3
279.6
257.6
253.7
232.0
766.5
196.0
328.1
589.0
276.8
190.0
292.5
213.9
 IDL - Instrument Detection Limit Ug/L).
   N - Number of laboratories using the most common wavelength.
CRDL - Contract Required Detection Limit (ug/L).
SOURCE:  Aleckson,  K.A.  et al  1986.

-------
               TABLE B-3




CLP SEMI-VOLATILE HSL COMPOUNDS AND CRDL
Compound name
Phenol
b1 s( 2-Chl orotthyl ) ether
2-Chl orophenol
1 ,3-Dlehlorobenztne
1 ,4-01 chl orobenzene
Benzyl alcohol
1 , 2-01 chl orobenzene
2-Methyl phenol
b1 s(2-Chl orol sopropyl ) ether
4-Methyl phenol
N-N1 tro$o-d1 -n-propyl ami ne
Hexaehl oroe thane
Nitrobenzene
I sophoron®
2-N1 trophenol
2, 4-01 nee thy 1 phenol
Benzole add
bl sC 2-Chl oroethoxy )rae thane
2,4-D1ehl©rophtnol
1 ,2, 4-Trl chl orobenzene
Naphthalene
4-Chloroan1l1ne
Hexaehl orobutadl ene
4-Chl oro-3-nstthy 1 phenol
2-Methyl naphthal ene
Hexaehl oroeyel open tad lent
2, 4, 6-Trl chl orophenol
2,4, 5-Trl chl orophenol
2-Chl oronaphthal ene
2-N1troan1l1ne
Dimethylphthalate
Acenaphthylene
3-N1troan1l1iie
Acenaph thene
2,4-OlRltr9phtnol
4-Nltrophtnol
Dlbenzofuran
2 84-Oi nl trotol u®n@
2, 6-01 nltro toluene
Dlethylph thai ate
4-Chl orophenyl -phenyl ether
Fl uorene
4-N1trot«1l1n@
4, 6-D1n1tro-i-»i thy! phenol
or CCCfe C
CCC



CCC





SPCC



cce







CCC
CCC

SPCC
CCC






cce
SPCC
SPCC








a^S
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
330
1,600
330
330
330
330
330
330
330
330
330
330
1,600
330
1,600
33©
330
1,600
333
1,600
1,600
330
330
330
330
330
330
1.60©
1,600
LOW water
V, C*!Dlf uq/L
hr —
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
50
10
10
10
10
10
10
10
10
10
10
50
10
50
10
10
50
10
50
50
10
10
10
10
10
10
50
SO
CAS , "-
Number
108"9S'-Z"
111-44-4
95-57-8
541-73-1
106-46-7
100-51-6
95-50-1
95-48-7
39638-32-9
106-44-5
621-64-7
67-72-1
98-95-3
78-59-1
88-75-5
105-67-9
65-85-0
111-91-1
120-83-2
120-82-1
91-20-3
106-47-8
87-68-3
, 59-50-7
91-57-6
77-47-4
88-06-2
95-95-4
91-58-7
88-74-4
131-11-3
208-96-8
99-09-2
' 83-32-9
51-28-5
100-02-7
132-64-9
121-14-2
60S-20-2
84-66-2
7005-72-3
86-73-7
100-01-6
534-51-1
                    B-3

-------
                                TABLE B-3

                 CLP SEMI-VOLATILE HSL COMPOUNDS AND CRDL
                               (continued)
Compound name or CCCB
N-N1trosod1 phenyl ami ne CCC
4-Bromophenyl -phenyl ether
Hexachl orobenzene
Pentachl orophenol CCC
Phenanthrene
Anthracene
D1 -n-butyl phthal ate
Fl uoranthene CCC
Pyrene
Butyl benzyl phthal ate
3 , 3 ' -01 chl orobenzl dl ne
Benzol a j (anthracene
b1 s( 2-Ethyl hekyl } phthal ate
Chrysene
01 -n-octyl phthal ate CCC
BenzoC b ) fl uoran thent
Benzol k ) fl uoranthene
Benzo(a)pyrene CCC
Indeno(l,2,3-cd)pyrene
D ibenzt a, h) anthracene
Benzo(g,h,1 )perylen«
aCCC-tall brail on Check Compound
Low Soil
CRDL. tig/kg
330
330
330
1.600
330
330
330
330
330
330
660
330
330
330
330
330
330
330
330
330
330

CRDL, ug/L
10
10
10
50
10
10
10
10
10
10
20
10
10
10
10
10
10
10
10
10
10

CAS
Number
86-3(5-6
101-55-3
118-74-1
87-86-5
85-01-8
120-12-7
84-74-2
A4h«» M 6 A
206-44-0
129-00-0
85-68-7
91-94-1
56-55-3
117-81-7
M 4 A #% * f^
218-01-9
117-84-0
a*f\» fkffk A
205-99-2
207-08-9
50-32-3
193-39-S
53-70-3
191-24-2

bSPCC-System Performance Check Compound
 Note:   Medium soil/sediment contract required
         times the Individual low soil/sediment
         contract required dettctlon limits  art
         low water CRDL.
detection limits are 60
CRDL and medium water
100 times the Individual
   t,U.S. GOVERNMENT PRINTING OFFICEd 9 87 -748 -121/ 67043
SOURCE:  Wolf, J.S. et al 1986.
                                      B-4

-------