DRC-2024-006918

Click here for
DISCLAIMER
Document starts on next page
EPA
United States
Environmental Protection
Agency
Office of
Waste Programs Enforcement
Office of Solid Waste and
Emergency Response
September 1986
Solid Waste
RCRA
Ground-Water
Monitoring Technical
Enforcement
Guidance Document
OSWER-9950.1
RCRA GROUND-WATER MONITORING
TECHNICAL ENFORCEMENT GUIDANCE DOCUMENT
(TEGD)
SEPTEMBER 1986
OSWER-9950.1
OVERVIEW
This publication, entitled the RCRA Ground-Water Monitoring Techni-
cal Enforcement Guidance Document (TEGD), describes in detail what the
United States Environmental Protection Agency deems to be the essential
components of a ground-water monitoring system that meets the goals of
the Resource Conservation and Recovery Act. This guidance is intended
to be used by enforcement officials. permit writers, field inspectors
and attorneys at the federal and state levels to assist them in making
informed decisions regarding the adequacy of existing or proposed
ground-water monitoring systems or modifications thereto. It is not a
regulation and should not be used as such. The TEGD is divided into six
chapters which contain discussions on the following:
• Characterization of site hydrogeology;
• Location and number of ground-water monitoring wells;
• Design, construction and development of ground-water monitoring
wells:
• Content and implementation of the sampling and analysis plan;
• Statistical analysis of ground-water monitoring data: and
• The content and implementation of the assessment plan.
The document is mainly directed towards interim status facilities.
Much of the purely technical content, especially regarding site charac-
terization, well design and construction, and assessment of contamination
of ground water, is germane to permitted facilities as well as non-RCRA
programs. Clearly, the spectrum of hydrogeologic regimes is great, and
no single document could provide detailed, step-by-step instructions for
monitoring each one. The writers of the TEGD concur and have developed a
framework within which a dynamic decision-making process may be applied
using a combination of national opinion and site-specific considerations.
ii
In August 1985, the RCRA Ground-Water Monitoring Compliance Order --
Guide was published. It is the companion document to the TEGD and
contains guidance on the use and formulation of compliance orders. It is
the hope of U.S. EPA that these guidance documents will further the goal
of the regulators and regulated community alike to protect human health
and the environment.
The U.S. EPA fully recognizes the dynamic nature of the RCRA program.
The TEGD. as it is presented, documents current Policy and direction for
enforcement and compliance. The TEGD can be used by technical reviewers
and the regulated community toward attaining the mandate of protection of
human health and the environment.
iii
OSWER-9950.1
TABLE OF CONTENTS
Page
CHAPTER ONE. CHARACTERIZATION OF SITE HYDROGEOLCGY 1
1.1 Investigatory Tasks for Hydrogeologic Assessments ........ 2
1.2 Characterization of Geology Beneath the Site .................. 5
1.2.1 Site Characterization Boring Program ................... 6
1.2.2 Interpretation of Geology Beneath the Site ............. 18
1.2.3 Presentation of Geologic Data .......................... 19
1.3 Identification of Ground-Water Flow Paths ..................... 22
1.3.1 Determining Ground-Water Flow Directions ............... 22
1.3.1.1 Ground-Water Level Measurements ...................... 24
1.3.1.2 Interpretation of Ground-Water Level Measurements .... 25
1.3.1.3 Establishing Vertical Components of Ground-Water
Flow ................................................. 26
1.3.1.4 Interpretation of Flow Direction and Flow Rates ...... 30
1.3.2 Seasonal and Temporal Factors: Ground-Water Flow ...... 30
1.3.3 Determining Hydraulic Conductivities ................... 31
1.4 Identification of the Uppermost Aquifer . . . . . . . . ... .. ....... 34
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
CHAPTER TWO. PLACEMENT OF DETECTION MONITORING WELLS . . . . . . . . . . . . . . 45
2.1 Placement of Downgradient Detection Monitoring Wells . . . . . . . . . . 47
2.1.1 Location of Wells Relative to Waste Management Areas .. 47
2.1.2 Horizontal Placement of Downgradient Monitoring
Wells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.1.3 Vertical Placement and Screen Lengths . . . . . . . . . . . . . . . . . . 51
2.1.4 Examples of Detection Well Placement in Three Common
Geologic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.2 Placement of Upgradient (Background) Monitoring Wells ........ 66
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
CHAPTER THREE. MONITORING WELL DESIGN AND CONSTRUCTION ........... 71
3.1 Drilling Methods .............................................. 71
3.1.1 Hollow-Stem Continuous-Flight Auger .................... 73
3.1.2 Solid-Stem Continuous-Flight Auger ..................... 74
3.1.3 Cable Tool ............................................. 74
3.1.4 Air Rotary ............................................. 75
3.1.5 Water Rotary ........................................... 76
3.1.6 Mud Rotary ............................................. 77
iv
TABLE OF CONTENTS
(Continued)
3.2 Monitoring Well Construction Materials . . . . . . . . . . . . . . . . . . . . . . . .
3.2.1 Well Casings and Well Screen ..........................
3.2.2 Monitoring Well Filter Pack and Annular Sealant ........
3.3 Well Intake Design ........................ . . . . . . . . . . ........
3.4 Well Development .......................... . . . . . . . . . . . . . . . . . . .
3.5 Documentation of Well Design and Construction . . . . . . . . . . . . . . . . . 88
3.6 Specialized Well Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.7 Evaluation of Existing Wells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
CHAPTER FOUR. SAMPLING AND ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.1 Elements of Sampling and Analysis Plans . . . . . . . . . . . . . . . . . . . . . . . 98
4.2 Sample Collection .............................................
4.2.1 Measurement of Static Water Level Elevation ............
4.2.2 Detection of Immiscible Layers .........................
4.2.3 Well Evacuation ........................................
4.2.4 Sample Withdrawal ......................................
4.2.5 In-Situ or Field Analyses ..............................
4.3 Sample Preservation and Handling .............................. 108
4.3.1 Sample Containers ..................................... 109
4.3.2 Sample Preservation .................................... 110
4.3.3 Special Handling Considerations ........................ 110
4.4 Chain of Custody .............................................. 114
4.4.1 Sample Labels .......................................... 115
4.4.2 Sample Seal ............................................ 115
4.4.3 Field Logbook .......................................... 116
4.4.4 Chain-of-Custody Record ................................ 116
4.4.5 Sample Analysis Request Sheet .......................... 117
4.4.6 Laboratory Logbook ..................................... 117
4.5 Analytical Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
4.6 Field and Laboratory Quality Assurance/Quality Control ........ 118
4.6.1 Field QA/QC Program .................................... 118
4.5.2 Laboratory QA/QC Program ............................... 120
v
Page
77
78
83
86
87
99
99
100
102
104
107
OSWER-9950.1
TABLE OF CONTENTS
(Continued)
Page
4.7 Evaluation of the Quality of Ground-Water Data ................ 120
4.7.1 Reporting of Low and Zero Concentration Values ......... 121
4.7.2 Missing Data Values .................................... 123
4.7.3 Outliers ............................................... 125
4.7.4 Units of Measure ....................................... 126
References ........................................................ 127
CHAPTER FIVE. STATISTICAL ANALYSIS OF DETECTION MONITORING DATA . . . 129
5.1 Methods for Presenting Detection Monitoring Data . . . . . . . . . . . . . . 129
5.2 Introductory Topics: Available t-Tests, Definition of Terms,
Components of Variability, Validity of the t-Test Assumptions,
False Positives Versus False Negatives, and the Transition to
Permitting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.2.1 Available t-Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 130
5.2.2 Definition of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.2.3 Components of Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.2.4 Validity of the t-Test Assumptions . . . . . . . . . . . . . . . . . . . . . 133
5.2.5 False Positives Versus False Negatives . . . . . . . . . . . . . . . . . 134
5.2.6 The Transition to Permitting . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.3 Statistical Analysis of the Background Data . . . . . . . . . . . . . . . . . . . 136
5.4 Statistical Analysis of Detection Monitoring Data After the
First Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.4.1 Comparison of Background Data with Upgradient Data
Collected on Subsequent Sampling Events . . . . . . . . . . . . . . . . 138
5.4.2 Comparison of Background Data with Downgradient Data . . . 139
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
CHAPTER SIX. ASSESSMENT MONITORING . . . . . . . . . . . . . . . . ............... 143
6.1 Relationship of Assessment Monitoring to Ground-Water
Responsibilities Under the Permit Application Regulations
(Part 270) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.2 Contents of a Part 265 Assessment Monitoring Plan . . . . . . . . . . . . . 145
6.3 Description of Hydrogeologic Conditions . . . . . . . . . . . . . . . . . . . . . . . 147
6.4 Description of Detection Monitoring System . . . . . . . . . . . . . . . . . . . . 148
vi
TABLE OF CONTENTS
(Continued)
6.5 Description of Approach for Making First Determination -
False Positives ...............................................
6.6 Description of Approach for Conducting Assessment .............
6.6.1 Use of Direct Methods ..................................
6.6.2 Use of Indirect Methods ................................
6.6.3 Mathematical Modeling of Contaminant Movement ..........
6.7 Description of Sampling Number, Location, and Depth ...........
6.7.1 Collection of Additional Site Information ..............
6.7.2 Sampling Density .......................................
6.7.3 Sampling Depths ........................................
6.8 Description of Monitoring Well Design and Construction ........
6.9 Description of Sampling and Analysis Procedures ................
6.10 Procedures for Evaluating Assessment Monitoring Data ..........
6.10.1 Listing of the Data ...................................
6.10.2 Summary Statistics Tables .............................
6.10.3 Data Simplification ...................................
6.10.4 Graphic Displays of Data ..............................
6.11 Rate of Migration .............................................
6.12 Reviewing Schedule of Implementation ..........................
GLOSSARY ...........................................................
INDEX ..............................................................
APPENDICES
A. Evaluation Worksheets
B. A Statistical Procedure for Analyzing Interim Status Detection
Monitoring Data: Methodology and Application
Page
148
151
152
154
155
160
161
162
164
165
165
168
171
174
178
179
181
188
191
207
C. Description of Selected Geophysical Methods and Organic Vapor Analysis
vii
OSWER-9950.1
LIST OF TABLES
1-1.
1-2.
1-3.
1-4.
2-1.
2-2.
3-1.
4-1.
6-1.
6-2.
6-3.
6-4.
6-5.
Hydrogeologic Investigatory Techniques ....................... 3
Factors Influencing Density of Initial Boreholes ............. 7
Field Boring Log Information ................................. 16
Suggested Laboratory Methods for Sediment/Rock Samples ....... 17
Factors Influencing the Intervals Between Individual
Monitoring Wells Within a Potential Migration Pathway ........ 50
Factors Affecting Number of Wells Per Location (Clusters) .... 56
Drilling Methods for Various Types of Geologic Settings ...... 72
Sampling and Preservation Procedures for Detection
Monitoring ................................................... 111
An Example of How Assessment Monitoring Data Should be
Listed ....................................................... 173
An Example of How Data Should be Summarized by GWCC .......... 175
An Example of How Data Should be Summarized by GWCC/Well
Combination .................................................. 176
An Example of How Data Should Be Summarized by GWCC/Well/
Date Combination ............................................. 177
An Example of How Ranks of the Mean Concentrations for Each
GWCC/Well Combination Can Be Used to Simplify and Present
Concentration Data Collected for a Variety of GWCCs in a
Number of Monitoring Wells ................................... 180
viii
LIST OF FIGURES
Page
l-l.
1-2.
1-3.
1-4.
1-5.
1-6.
1-1.
1-8.
1-9.
1-10.
1-11.
1-12.
1-13.
2-1.
2-2.
2-3.
2-4.
2-5.
2-6.
2-7.
3-1.
3-2.
3-3.
3-4.
3-5.
3-6.
Possible Borehole Configuration for a Small Surface
Impoundment ..................................................
Subsequent Iteration of Borehole Program at a Small Surface
Impoundment from Figure 1-1A .................................
Example of a Contaminant That May Affect the Quality of a
Confining Layer ..............................................
Data Points Used to Generate a Geologic Fence Diagram ........
Example of an Acceptable Geologic Cross Section Showing
Gamma and Resistivity Logs ...................................
Example of a Topographic Map (2-Foot Contour Interval) .......
Potentiometric Surface Map ...................................
An Example of a Flow Net Derived from Piezometer Data ........
Example of Hydraulic Communication Between Water-Bearing
Units ........................................................
An Example of Hydraulic Communication Caused by Faulting .....
Perched Water Zones as Part of the Uppermost Aquifer .........
An Example of an Undetected Structure in the Uppermost
Aquifer ......................................................
An Example of an Undetected Portion of the Uppermost
Aquifer Due to an Improperly Screened Borehole ...............
Dowgradient Wells Immediately Adjacent to Hazardous Waste
Management Units .............................................
Illustration of Multiple Ground-Water Flow Paths in the
Uppermost Aquifer Due to Hydrogeologic Heterogeneity .........
Monitoring Well Placement and Screen Lengths in a Glacial
Terrain ......................................................
Plan View of Figure 2-3 Showing Lines of Equipotential in
the Upper (A) and Lower (B) Sand Units .......................
Monitoring Well Placement and Screen Lengths in an Alluvial
Setting ......................................................
Monitoring Well Placement and Screen Lengths in a Mature
Karst Terrain/Fractured Bedrock Setting ......................
Placement of Background Wells ................................
General Monitoring Well - Cross Section ......................
General Stainless Steel Monitoring Well - Cross Section ......
Composite Well Construction (Inert Construction Materials
in Saturated Zone) ...........................................
Decision Chart for Turbid Ground-Water Samples ...............
Monitoring Well Cross Section--Dedicated Positive Gas
Displacement Bladder Pump System .............................
Monitoring Well Cross-Section--Dedicated Purge Pump and
Sample Withdrawal Pump. Well Screened in a High Yielding
Aquifer ......................................................
10
13
14
20
21
23
27
29
37
39
40
41
42
48
59
60
61
63
65
68
79
80
82
84
90
32
(Continued)
ix
OSWER-9950.1
LIST OF FIGURES
(Continued)
Page
6-1. Procedure for Evaluating False Positive Claims by
Owner/Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
6-2. Example of Using Soil Gas Analysis to Define the Probable
Location of Ground-Water Plume Containing Volatile Organics . . 153
6-3. Example of Assessment Monitoring Well Placement . . . . . . . . . . . . . . 166
6-4. Selection of Plume Characterization Parameters for Units
Subject to Part 265 and Part 270 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
6-5. Plot of Chromium Concentrations Over Time (Well 9A) . . . . . . . . . . 182
6-6. Chromium and Lead Concentrations Over Time (Well 9A) . . . . . . . . . 183
6-7. General Schematic of Multiphase Contamination in Sand . . . . . . . . 187
X
OSWER-9950.1
CHAPTER ONE
CHARACTERIZATION OF SITE HYDROGEOLOGY
The adequacy of an owner/operator's ground-water monitoring program
hinges, in large part, on the quality and quantity of the hydrogeologic
data the owner/operator used in designing the program. Technical
reviewers and permit/closure plan reviewers (hereafter permit writers),
therefore, should evaluate the adequacy of an owner/operator's
hydrogeologic assessment as a first step towards ascertaining the overall
adequacy of the detection and/or assessment monitoring network. Clearly,
if the design of the well system is based upon poor data, the system
cannot fulfill its intended purpose. Because of the complexity of
ground-water monitoring systems, owner/operators should discuss the
intended approach initially with the State or EPA.
In performing this evaluation, technical reviewers should ask
themselves two questions.
• Has the owner/operator collected enough information to:
(1) identify and characterize the uppermost aquifer and
potential contaminant pathways, and (2) support the place-
ment of wells-capable of determining the impact of the
facility on the uppermost aquifer?
• Did the owner/operator use appropriate techniques to collect
and interpret the information used to support the placement
of wells?
The answer to each question will, of course, depend on site-specific
factors. For example, sites with more heterogeneous subsurfaces require
more hydrogeologic information to determine placement of wells that will
intercept contaminant migration. Likewise, investigatory techniques that
may be appropriate in one setting, given certain waste characteristics
and geologic features, may be inappropriate in another.
This chapter is designed to help technical reviewers answer the
above questions. It identifies various investigatory tasks that enable
-1-
an owner/operator to characterize a site, and explores the factors that
technical reviewers should consider when evaluating whether the
particular investigatory program an owner/operator used was appropriate
in a given case. Technical reviewers should also find this chapter
useful when constructing compliance orders that include hydrogeologic
investigations.
1.1 Investigatory Tasks for Hydrogeologic Assessments
An owner/operator should accomplish two tasks in conducting a
hydrogeologic investigatory program:
1. Define the geology beneath the site area: and
2. Identify ground-water flow paths and rates.
A variety of investigatory techniques are available to achieve these
goals, and technical reviewers must evaluate the success of the
combination of techniques used by the owner/operator, given the site-
specific factors at the facility.
There are certain investigatory techniques that all owner/operators,
at a minimum, should have used to characterize their sites. Table 1-1
illustrates a number of techniques that an owner/operator may use to
perform hydrogeologic investigations. Those techniques that the
owner/operator, at a minimum, should have used to define the geology or
identify ground-water flow paths are identified with check marks.
Table 1-1 also presents preferred methods for presentation of the
data generated from a hydrogeologic assessment. An owner/operator who
has performed the level of site characterization necessary to design a
RCRA ground-water monitoring program will be able to supply any of the
outputs (cross sections, maps, etc.) listed in the last column of
Table l-l.
The owner/operator should have reviewed the available literature on
the hydrogeology of the site area prior to conducting the site-specific
-2-
OSWER-9950.1
TABLE l-l
HYDROGEOLOGIC INVESTIGATORY TECHNIQUES
INVESTIGATORY INVESTIGATORY DATA PRESENTATION FORMATS/
TASKS TECHNIQUES ASSESSMENT OUTPUTS
Definition of Subsurface
Materials [geology]
Survey of existing geologic J
information
J
Soil borings
J
Rock corings
J
Material tests (grain size
analyses, standard penetration a
tests, etc.)
Geophysical well logs (point 0
and lateral resistivity and/or
electromagnetic conductance,
gamma ray, gamma density, J
calipher, etc.>
Surface geophysical surveys
(D.C. resistivity, E.M., seismic)
Hydraulic conductivity measure-
ments of cores (unsaturated
zone>
Aerial photography (fracture
trace analysis)
Detailed lithologic/structural
mapping of outcrops and trenches
Narrative description of geology
Geologic cross sections
Geologic or soil map (1" = 200')
Boring logs or coring logs
Structure contour maps of aquifer
and confining layer (plan view>
Raw data and interpretive analysis
of geophysical studies
Raw data and interpretive analysis
of material tests
(Continued)
-3-
-__-~ ___..~
INVESTIGATORY
TASKS
__..
Idcntitication of Ground-
W<itrr tlow Paths [hydrology]
G\-ound-water flow
directions (including
vertical and horizontal
components of flow)
Hydraulic conductivities
TABLE l-l (Continued>
HYOROGEOLOGIC INVESTIGATORY TECHNIQUES
~~_l___l..-~_-~--^. ~~~_.. __
INVESTIGATORY
TECHNIQUES
~~-____ --____
J Installation of piezometers;
water level measurements at
different depths and locations
J Slug tests and/or pump tests
---~~ -____---.- ~. .-
OATA PRESENTATION FORMATS/
ASSESSMENT OUTPUTS
~~-___-_.-.
J Narrative description of ground
water with flow patterns
J Water table or potentiometric
maps (plan view> with flow lines
(1" = 200')
l Tracer studies
J Hydrologic cross sections
l Estimates based on
sieve analyses
l Raw data and interpretive analysis
of slug tests, pump tests, and
tracer studies
J Minimum techniques and corresponding outputs that should be used to define site hydrogeological conditions.
-4-
OSWER-9950.1
investigation. Such a review provides a preliminary understanding of the
distribution of sediments and rock, general surface water drainage, and
ground-water flow that serves to guide the site-specific investigation.
The owner/operator's site-specific investigatory program should have
included direct (e.g., borings, piezometers, geochemical analysis of soil
samples) methods of determining the site hydrogeology. Indirect methods
(e.g., aerial photography, ground penetrating radar, resistivity), espe-
cially geophysical studies, may provide valuable sources of information
that can be used to interpolate geologic data between points where
measurements with direct methods were made. Information gathered by
indirect methods alone, however, generally would not have provided the
detailed information necessary. The owner/operator should have combined
the use of direct and indirect techniques in the investigatory program to
produce an efficient and complete characterization of the facility,
including an identification of:
• The geology below the owner/operator's hazardous waste facility;
• The vertical and horizontal components of flow in the uppermost
aquifer below the owner/operator's site:
• The hydraulic conductivity(ies) of the uppermost aquifer:
• The vertical extent of the uppermost aquifer; and
• The pertinent physical/chemical properties of the confining
unit/layer relative to hazardous wastes present.
The following sections outline the basic steps an owner/operator should
have followed to implement a site hydrogeologic study, and detail the
methods that the owner/operator should have used to collect and present
site hydrogeologic data.
1.2 Characterization of Geology Beneath the Site
In order to detail the geology beneath the site and therefore be
able to identify potential pathways of contamination, the owner/operator
-5-
should have collected direct information identifying the lithology and
structural characteristics of the subsurface. Indirect methods of
geologic investigation such as geophysical studies may be used to augment
the evidence gathered by direct field methods, but should not be used as
a substitute for them. Surface geophysical studies, such as resistivity,
electromagnetic conductivity. seismic reflection, and seismic refraction,
and borehole methods like electromagnetic conductivity, resistivity, and
gamma ray may yield valuable information on the depth to the confining
unit, the types of unconsolidated material(s) present, the presence of
fracture zones or structural discontinuities, and the depth to the
potentiometric surface. Additionally, geophysical methods may have their
greatest utility in correlating the continuity of formations or strata
between boreholes. The result is the efficient compilation of extensive
site data without drilling an excessive number of boreholes. Geophysical
methods, however, should have been used primarily to supplement infor-
mation obtained from direct sources. In order to characterize the
lithology, depositional environment, and geologic characteristics of the
area beneath the site, the owner/operator should have used direct means.
The limitations of geophysical methods should also be recognized. For
instance, electrical borehole logging cannot be performed when the hollow
stem auger drilling method is used.
1.2.1 Site Characterization Boring Program
The technical reviewer should determine whether an owner/operator,
through the soil/rock boring program, gathered the information necessary
to characterize the geology beneath the site and consequently to identify
potential contaminant migration pathways. Such a program should have
entailed the following:
• Initial boreholes should be installed at a density based on
criteria described in Table 1-2 and sufficient to provide initial
information upon which to determine the scope of a more detailed
evaluation of geology and potential pathways of contaminant
migration.
-6-
OSLER-9950.1
FACTORS
TABLE l-2
INFLUENCING DENSITY OF INITIAL BOREHOL ES
FACTORS THAT MAY SUBSTANTIATE REDUCED DENSITY OF FACTORS THAT MAY SUBSTANTIATE INCREASED DENSITY OF
BOREHOLES BOREHOLES
-
l Simple geology (i.e., horizontal, thick,
homogeneous geologic strata that are con-
tinuous across site that are unfractured
and are substantiated by regional geologic
information)
l Use of geophysical data to correlate well
log data. Preferred methods: DC resistivity,
seismic reflection or seismic refraction,
geophysical well logging
. Fracture zones encountered during drilling
. Suspected pinchout zones (i.e., discontinuous
units across the site)
l Geologic formations that are tilted or folded
l Suspected zones of high permeability that woul!l
not be defined by drilling at large intervals
. Laterally transitional geologic units with
irregular permeability (e.g., sedimentary
facies changes)
-_ - -.--.- ___.-
-7-
l Initial boreholes should have been drilled into the first
confining layer beneath the uppermost aquifer. The portion of
the borehole extending into the confining layer should have been
plugged properly after a sample was taken.
l Additional boreholes should be installed in numbers and locations
sufficient to characterize the geology beneath the site, The
number and locations of additional boreholes should have been
based on data from initial borings and indirect investigation.
l Collection of samples of every significant stratigraphic contact
and formation, especially the confining layer, should have been
taken. Continuous cores should have been taken initially to
ascertain the presence and distribution of small- and large-scale
permeable layers. Once stratigraphic control was established,
samples taken at regular, e.g., five-foot intervals, could have
been substituted for continuous cores.
l Boreholes in which permanent wells were not constructed should
have been sealed with material at least an order of magnitude
less permeable than the surrounding soil/sedimentirock in order
to reduce the number of potential contaminant pathways.
l Samples should have been logged in the field by a qualified
professional in geology.
l Sufficient laboratory analysis should have been performed to
provide information concerning petrologic variation, sorting (for
unconsolidated sedimentary units), cementation (for consolidated
sedimentary units), moisture content, and hydraulic conductivity
of each significant geologic unit or soil zone above the
confining layer/unit.
. Sufficient laboratory analysis should have been performed to
describe the mineralogy (X-ray diffraction). degree of compac-
tion, moisture content, and other pertinent characteristics of
any clays or other fine-grained sediments held to be the
confining unit/layer. Coupled with the examination of clay
mineralogy and structural characteristics should have been a
preliminary analysis of the reactivity of the confining layer
in the presence of the wastes present.
At many sites a site characterization has already been done and
monitoring wells installed. In evaluating the design of such systems,
the technical reviewer should utilize, irhere appropriate, data already
- “, -
OSXER-9950.1
gathered by the owner/operator. Because of the quality of existing data,
it is possible that site characterization may be complete or ma;r only
need to be supplemented by a few additional boreholes, piezometers, or
monitoring wells. Some facilities, including closed facilities, may need
to undertake a site characterization from the first phase.
The borehole program to elucidate site hydrogeology generally
requires more than one iteration. A benefit to this technique 1s that
data and observations derived from previous boreholes may be used to
guide the placement of future ones.
It is imperative that the owner/operator research local hydrogeology
before initiating a borehole program. Existing reports, maps, and
research papers gathered from a variety of sources can be used to
understand, in a broad sense, the hydrogeological regime in which the
facility is located. Thus, such information as local stratigraphy,
depositional environment, and tectonic history serves to provide an
estimate of the distribution and types of geologic materials likely to be
encountered. Similarly, knowledge of regional ground-water flow rate,
depth, quality, and direction, local pumping, evapotranspiration rates,
and surface water hydrology represents an effective first approximation
of site-specific ground-water characteristics. The next phase should
have been the progressive placement of boreholes based, at first, on
research and, subsequently, on previous boreholes and data from research.
The number of initial boreholes should have been sufficient to
provide initial information upon which to determine the scope of a more
detailed evaluation of geology and potential pathways of contaminant
migration. An example of a simple case is illustrated in Figure I-1.
The objective of the initial boreholes is to begin to reconcile the
broad, conceptual model derived from research data with the true site-
specific hydrogeologic regime. In other words, the borehole program is
necessary to establish the small-scale geology of the area beneath the
facility and place it in the context of the geology of the region or
locale.
A
IMPOUNDMENT
LEGEND
# BOREHOLE
A PIEZOMETER
- FENCE DIAGRAM LINES
0 100 200’
I 1 J
f
l
A
A A’
LEGEND
l BOREHOLE
A PIEZOMETER
A- A’ CROSS SECTION LINE I
FIGURE 1-1. POSSIBLE BOREHOLE CONFIGURATION FOR A
SMALL SURFACE IMPOUNDMENT
-IQ-
3SXER-9950.1
The distance between these initial boreholes should be varied based
on site-specific criteria, yet shoilid have been close enough so that
cross sections woclld have accurately sortrayed stratigraphy with minimai
reliance on inference (see Table l-2). in this way, a suitably restricted
configuration of a iimited number of in:tial boreholes, in combination
with indirect investigative techniques and research data, will serve to
guide efficiently the placement of additional Soreholes where needed to
characterize potential Tathways for contaminant migration. A parallel
program using 2iezometers should also be undertaken. iithoiogic data
should uitimately correlate with hydraulic parameters (e.g., clean, well
sorted, unconsolidated sands should exhibit high hydraulic conductillity).
If they do not, further hydraulic testing should be done.
During the completion of the borings, the owner/operator should
check drill logs for:
l Correlation of stratigraphic units between soii/rock borings;
0 Identification of zones of potentially high hydraulic
conductivity;
0 Identification of the confining formation/layer;
0 Indication of unusual or unpredicted geologic features such as
fault zones, fracture traces, facies changes, solution channels,
buried stream deposits. cross cutting structures, pinch out
zones, etc.: and
l Continuity of petrographic features such as sorting, grain size
distribution, cementation, etc., in significant formations.
If the owner/operator is unable to define such structural anomalies, or
zones of potentially high canduc:ivltli, or to correlate petrographic
features and/or stratigraphy betsteen any two adjacent boreholes, then
additional intermediate boreholes should be drilled and ancillary
investigative techniques employed to describe potentiai contaminant
migration.
On the other hand, if the ne~essa:-;' <harac:srizat;on is largely
achieved at the initial placement, f3.,~91- additiona; boreholes and less
additional indirect investigation :~rould b? necessar:: zo iescrlbe pathways.
. _--
F i rgu L- e 1 - 2 illustrates how subsequent boreholes and indirect supple-
mentary techniques can be added to the initial borehole configuration to
characc erize potential pathways for contaminant migration. In most cases,
additionai boreholes wili be necessary to compiete the characterization
because the majority of hydrogeologic settings are complex.
It is vitally important that the owner/operator consider the thick-
ness and potential reactivity of confi ning ciays or other fine-grained
sediments in the presence of si:e-specific waste types. Marl, for
instance, is chemically attacked by low pH wastes because of its high
carbonate content. Smectites and, to a lesser extent, illitic clays are
ineffeztive impediments to the migration of various organic chemicals
(e.g., xylene). In contaminated areas, a chemically degraded confining
layer may lead to hydraulic communication unanticipated by literature
L-e-J1 ews of atratigraphy. An example is shown in Figure l-3. In pristine
areas, the possible future chemicai degradation of a confining layer
should be of concern during any assessment monitoring or corrective
action necessary at the facility.
All samples should have been logged in the field by a qualified
FL-ofessional in geology (see glossary). These samples should have been
:,qi!ectec! with a Shelby tube, split barrel sampler, or rock corer, and
1‘ e 7: 1 + j + r. t the significant formations and stratigraphic contacts.
yo*. ?- : ,‘.... (J' .- IA ..a :ores should ha*ie been taken initially to obtain stratigraphic
;:kJ;-Ltl'~3L. Sar;.ples could ha-re been taken at regular intervals, depending
cm s: :--s;ec;f1c conditions once stratigraphic controi was established.
1:~:Ll:r.r; logs and field records should have been prepared detailing the
1 :: . . +::I ::c: Lnformat:on:
. C.-0;s 2etrograuhy {e.g., soil classification or rock type) of
e.:~-k geologic unit, including the confining unit;
. .: .'~jS.~ ;;truc+llra; -. interpretation of each geologic unit and
.--ructilrai features (e.g., fractures, fault gouge, solution
.:.~:-seis, buried s-~reams or *valleys), bioturbatlon zones,
;:,:rology, 3nd 2:scontinuities:
OSWER-9950.1
0 100 200'
c # I
LEGEND
3 INfTlAL BOREHOLE
0 NEW BOREHOLE
A INfTfAL PIEZOMETER
A NEW PIEZOMETEA
GEOPHYSICAL TRAVERSE
- (SURFACE AND/OR BOREHOLE)
- CROSS SECTION LINE A A’
FIGURE 1-2 SUBSEQUENT ITERATION OF BOREHOLE PROGRAM AT A
SMALL SURFACE IMPOUNDMENT FROM FIGURE l-1A.
-13-
L
PlEZOb’ETER
CLUSTER
A
300’ -
275’ -
GROUNDWATER FLOI
TlME A
250’ -
225’ -
200’ -
0 100 200 FEET
PIEZOMETEA
CLUSTER
0 PIEZOMETEA
-- ----‘-3 A8 C
300’ -8
275’ -
GROUNDWATER FL0
TIME 0
250’ -
225’ -
16 1
0 100 200 FEET
l SOME CLAYS SUCH AS MONTMORILLONITE AND ILLITE
ARE SUSCEPTIBLE TO CHEMICAL ATTACK BY SOLVENT.
BASED LEACHATE
WELL AND SCREEN
POTENTIOMETRIC
FIGURE 13 EXAMPLE OF A CONTAMINANT THAT MAY AFFECT THE QUALITY
OF A CONFINING LAYER
-li-
OSWER-9950.1
l Development of soil zones and vertical extent and field
description of soil type (pritir to any necessary laboratory
analysis);
l Depth of water-bearing unit(s) and vertical extent of each;
a Depth and reason for termination of borehole;
l Depth, location, and identification of any contamination
encountered in borehole; and
l Blow counts, colors. and grain-size distributions(s).
Table l-3 identifies the minimum required information that should have
been included in a drilling log. These items are marked with asterisks.
In addition to field descriptions as described above, the owner/
operator should have provided, where necessary, a laboratory analysis of
each significant geologic unit and soil zone. These analyses should
contain the following information:
l Mineralogy and mineralogic variation of the confining layer and
confining units/layers, especially cla.,s (e.g., microscopic
analysis and other methods such as X-ray diffraction as
necessary);
l Petrology and petrologic variation of the confining layer and
each unit above the confining unit/layer {e.g., petrographic
analysis, other laboratory methods for unconsolidated materials
as deemed necessary) to determine among other things:
- degree of crystallinity and cementation of matrix
- degree of sorting, size fraction, and textural variation
- existence of small-scale structures that may affect fluid flow
l !?oisture content and moisture variation of each significant soil
zone and geologic unit; and
l Hydraulic conductivity and variation of each significant soil
zone and type and geologic unit in the unsaturated zone.
Some laboratory analysis methods available to investigate these
laboratory parameters are shown in Table 1-4.
TABLE l-3
FIELD BORING LOG INFORMATI!lN
General
. ProJect name
'0 tiole name/number
*o Date started and finished
** Geologist's name
** Driller's name
. Sheet number
'. Hole location; map and
elevat7on
** Rig type
bit size/auger size
'0 Petrologic :itholog'c
classification scheme used
(Wentdorth, unified sol1
ciassificatlon system)
Informdt~on Columns
** Depth l * Percent sample recovery
-0 Sample location~n~moer '0 Narrative description
. Blow counts and advance rate '. 3eDtCI :O S?tJTdtlOl
Narrative Oescriotio_l!
l Geologic Observations:
*- soil/rock type
*- color and stain
*- gross petrology
- friability
l - moisture content
l - degree of
weathering
*- presence of
carbonate
. Drilling Observations:
loss of circulation
advance rates
rig chatter
dater - eve1 s
amount of air
used. air pressure
orilling
difficulties
*-
fractures *. depositional
solution cavities structtires
bedding "- organic content
d'scontinuities: -- odor
e.g., foliation *- suspected
water-bearing zones ccntaminant
formational strike
and OlD
~c3SSllS
changes in drilling *.- amounts and tyoes
method or equipment of any liquids
-eadings from used
%ztectyve equipment, *- running sands
if any *.. caving/hole
smoun: of dater stability
yield or loss during
drqlling at different
depths
l Other Remarks
- equipment failures
*- possible sontaminat-07
*. deviationj From dri:l.ng p‘an
*- +!eather
'Indicates items that ?he o,tiner/operatcr ;nculd -ecord. it 4 min‘mum.
OSWER-9950.1
TABLE 1-4
SUGGESTED LABORPlTORY METHODS FOR SEDIMENT/ROCK SAMPLES
Sample Origin Parameter Laboratory Method Used to Determine
Geologic fOrmdtlOn.
unconsolidated
sediments. consoli-
dated sediments.
solum
Contaminated samples
(e.g., soils pro-
ducing higher than
background organic
vapor readings)
Hydraulic conductivity
Size fraction
Sorting
Specific yield
Specific retention
Petrofogy/Pedology
Mineralogy
Bedding
Lamination
Atterberg Limits
Appropriate subset
of Appendii VIII
parameters (5261;
Falling head, static
head test
Hydraulic conductivity
Sieving (ASTM)
Settlirg measurements
(ASTM)
"etrographic analysis
Column drawings
Centrifuge tests
Petrographic analysis
X-ray diffraction
confirming clay
mineralogy/chemistry
Petrographic analysis
Petrographic analysis
ASTM
SW-845
Hydraulic conductivity
Hydraulic conductivity
Porosity
Porosity
Soil type, rock type
Geochemistry, poten-
tial flow oaths
Soil cohesiveness
Identity of
contaminants
-
'Owners and operators mlqht also want to consider performing this test while they are obtaining
the other types of :nformation listed on this table.
-;7-
1.2.2 Interpretation of Geology Beneath the Site --
The technical reviewer should review the owner/operator's geologic
characterization and verify:
l The completeness of the narrative and the accuracy of the
owner/operator's interpretation, and
l That the geologic assessment addresses or provides means to
resolve any information gaps which may be suggested by the
geologic data.
In order to assess the completeness and accuracy of the owner/
operator's interpretation, the technical reviewer should:
a Examine and evaluate the raw data;
l Compare his own interpretation, based on the raw data, with that
of the owner/operator;
l Compare with other studies and information; and
l Identify any information gaps that relate to incomplete data
and/or to narrative presentation.
The technical reviewer should independently conduct the following
tasks to support and develop his interpretation of the site geology:
l Review drilling logs to identify major rock or soil types and
establish their horizontal and vertical variability;
l Construct representative cross sections from well log data;
l Identify zones of suspected high permeability, or structures
likely to in luence contaminant migration through the unsaturated
and saturated zones:
l Review laboratory data, determine whether laboratory data
corroborate field data and that both are sufficient to define
petrology; and
l Review mineraloqic identification of confining clays and the
owner/operator's assessment of general geochemistry and determine
corroboration between analytic and field data.
-13-
CSNER-9950.1
After the technical reviewer has interpre,ted the geologic data, these
results should be compared to the results developed by the owner/operator.
The technical reviewer should:
l Identify information gaps between narrative and data.
l Determine whether resolution requires collection of additional
data or reassessment of existing data; and
l Identify any information gaps that will affect the owner/
operator's ability to have located his/her RCRA monitoring well
system.
1.2.3 Presentation of Geologic Data
In addition to the generation and interpretation of site-specific
geologic data, the technical reviewer should review the owner/operator's
presentation of data in geoiogic cross sections, topographic maps, and
aerial photographs.
An adequate number of cross sections should be presented by an
owner/operator to depict significant geologic or structural trends and
reflect geologic/structural features in relation to local and regional
ground-water flow. Figure l-4 illustrates an example of a waste disposal
unit that is traversed by an adequate number of cross-section lines from
which a fence diagram may be created.
On each cross section, the owner/operator should have identified:
petrography of significant formations/strata, significant structural
features, stratigraphic contacts between significant formations/strata,
zones of high permeability or fracture, the location of each borehole,
depth of termination, depth to the zone of saturation, and depiction of
any geophysical logs. If the owner/operator is unable to supply such
details, the site characterization nay be inadequate. Figure l-5
illustrates an example of a geologic cross section. Vertical exaggera-
tion in cross sections should be minimized.
Additionally, surficial features may affect ground-water h;Tdro-
geology. An owner/operator should have provided a surface topographic
GROUND-WATER
FLOW
C
D’
D
SCALE
t I ’ ’ ’ 1
400 0 loo’ 200’ 300’ 400
LEGEND
0 BORING LOCATION
A A‘ FENCE DIAGRAM LINE
FIGURE l-4 DATA POINTS USED TO GENERATE A GEOLOGIC FENCE DIAGRAM
500
I- 400
300
200’
100
ELEVATION
NGVD A’
-600’ BOREHOLE l3OREHOLE
1
_......................
+ 3RD SILT LAYER +
C K = 4.0 x 10-ae~/%?c - POTENTIDM
SURFACE
.a GROUND WATER
loo’ 50’ 0’ 50’ loo \:: ,,‘.I,‘:“,~.:‘-,;-,,.,’
FIGURE l-5 EXAMPLE OF AN ACCEPTABLE GEOLOGIC CROSSSECTION
SHOWING GAMMA AND RESISTIVITY LOGS
IETRIC
1 PLAN VIEW
GROUNDWATER
N
63 BOREHOLE 1
l ROREHOLE NOT SHOWN
0 100’ 200’
ON CROSS StCTlON I
-21-
map and aerial photograph of the site. The topographic map should have
been constructed under the supervision of a licensed surveyor and should
provide contours at a two-foot contour interval, locations and illustra-
tions of man-made features (e.g.. parking lots, factory buildings,
drainage ditches, storm drains, pipelines, etc.), descriptions of nearby
water bodies and/or off-site wells, site boundaries, individual RCRA
units, delineation of the waste management areas, solid waste management
areas, and well and boring locations. An example of a site map is
depicted in Figure 1-6. An aerial photograph of the site should depict
the site and adjacent off-site features. This photograph should have the
site clearly delineated and labeled. In addition, adjacent surface water
bodies, municipalities and residences should be labeled.
1.3 Identification of Ground-Water Flow Paths
In addition to evaluating the owner/operator's characterization of
geology, technical reviewers must determine whether owner/operators have
identified ground-water flow paths. The characterization must have
included:
• The direction(s) of ground-water flow (including both horizontal
and vertical components of flow);
• The seasonal/temporal, naturally and artificially induced (i.e.,
off-site production well pumping, agricultural use) variations in
ground-water flow; and
• The hydraulic conductivities of the significant hydrogeologic
units underlying their site.
In addition, technical reviewers must ensure that owner/operators used
appropriate methods for obtaining the above information,
1.3.1 Determining Ground-Water Flow Directions
To locate wells so as to provide upgradient and downgradient well
samples, owner/operators should have a thorough understanding of how
ground water flows beneath their facility. Of particular importance is
-22-
OSWER-9950.1
FIGURE 16 EXAMPLE OF A TOPOGRAPHIC MAP (2-FOOT CONTOUR INTERVAL)
-23-
the direction of ground-water flow and the impact that external factors
(intermittent well pumping, temporal variations in recharge patterns,
etc.) may have on ground-water patterns. In order for an owner/operator
to have assessed these factors, a program should have been developed and
implemented for precise water level monitoring. This program should have
been structured to provide precise water level measurements in a
sufficient number of piezometers and at a sufficient frequency to gauge
both seasonal average flow directions and to account for seasonal or
temporal fluctuation of flow directions.
In addition to considering the components of flow in the horizontal
direction, a program should have been undertaken by the owner/operator to
accurately and directly assess the vertical components of ground-water
flow. Ground-water flow information must be based at least in part on
empirical data from borings and piezometers. Technical reviewers should
review independently an owner/operator's methodology for obtaining
information on ground-water flow and account for factors that may
influence that flow at the facility. The following sections provide
acceptable methods by which an owner/operator should have assessed the
vertical and horizontal components of flow at the site.
1.3.1.1 Ground-water level measurements
In order for the owner/operator to have initially determined the
elevation of the potentiometric surface in any monitoring well or
piezometer , several criteria should have been considered by :he
owner/operator.
l The casing height should have been measured by a licensed
surveyor zo an accuracy of 0.01 feet. This may have required the
placement of a topographic benchmark on the facility property.
l Generally, water level measurements from boreholes, piezometers,
or monitoring wells used to construct a single potentiometric
surface should have been collected within a 24-hour period. This
Fractice is adequate if the magnitude of change 1s small over
OSWER-9950.1
that period of time. There are other situations, however, which
necessitate that all measurements be taken within a short time
interval :
- tidally influenced aquifers;
- aquifers affected by river stage, impoundments, and/or unlined
ditches:
- aquifers stressed by intermittent pumping of production wells;
and
- aquifers being actively recharged due to a precipitation event.
l The method used to measure water levels should have been adequate
to attain an accuracy of 0.01 feet.
l A survey mark should be placed on the casing for use as a
measuring point. Many times the lip of the riser pipe is not
flat. Another measuring reference should be located on the grout
apron.
l Piezometers should be re-surveyed periodically to determine the
extent of subsidence or rise in ground surface.
l Water levels in piezometers should have been allowed to stabilize
for a minimum of 24 hours after well construction and develop-
ment, prior to measurement. In low yield situations, recovery
may take longer.
If an owner/operator cannot produce accurate documentation or
provide assurance that these criteria were met during the coilection of
water level measurements, this may indicate that the generated
information may be inadequate.
In cases where immiscible contamination is found during the
characterization, Water level measurements should be adjusted to reflect
its true elevation.
1.3.1.2 Interpretation of ground-water level measurements
After the technical reviewer has assured that the water level data
are valid, he should proceed to independently interpret the information.
The technical reviewer should:
-25-
l Use the owner/operator's raw data to construct a potentiometric
surface map (see Figure l-7). The data used to develop the
potentiometric map should be data from piezometors/wells screened
at equivalent stratigraphic horizons;
l Compare these data with that of the owner/operator and deter-
mine whether the owner/operator has accurately presented the
information, and ascertain if the information is sufficient to
describe ground-water flow trends: and
l Identify any information gaps.
In reviewing this information, the technical reviewer should now have
an approximate idea of the general flow direction; however, In order to
have properly located monitoring wells, the owner/operator should have
established hydrauiic gradient (fiow direction) in both the hor;zon:al and
vertical directions.
1.3.1.3 Establishing vertical components of ground-water flow
In order for the owner/operator to have determined the direction of
flow, vertical components of flow must have been directly determined.
This will have required the installation of piezometers in clusters.
A piezometer cluster is a closely spaced group of wells screened at
different depths to measure TJertical variations in hydraulic head. TO
obtain reliable measurements, the follow:ng criteria should be considerrd
in the placement of piezometer clusters:
. information obtained from multipie piezometer placement in s:ngle
boreholes may generate erroneous data. Dlacenent of vertically
nested piezometers in closeiy spaced separate bore:hoLes is the
preferred method.
l Piezometer measurements should have been collected at least
within a 24-hour period, and w:thin shorter interJals I;nder
certain conditions, if measurements are to be used in an-;
correlative presentation of data.
0' Piezometer measurements should have been determined along a
minimum of two vertical profiles across the site. These profiles
should be cross sections roughly parailel to the direction of
ground-water flow indicated by the potentiometric surface.
OSWER-9950.1
GROUND-WATER ! A
WATER - LEVEL
PIEZOMETER ELEVATION
Pl 423.10
P2 421.76
P3 427.13
P4 426.96
P5 427.05
P6 422.36
P7 417.55
P8 416.32
P9 417.10
PlO 413.26
Pll 416.35
P12 419.10
P13 419.01
P14 420.63
P15 426.73
P16 431.48
FLOW DIRECTION ; I
I I’ I , ,’ 11’
DATE \
RECORDED
312185
312185
312185
3/2/85
312185
312185
312185 ,30
312185 ------
312185
312185 A
312185
312185 “‘:--
3/2/85 H N
3/2/85 426
3/2/85 ----
312105 111’ __--
\
LEGEND
I
PIEZOMETER LOCATION
P3
B1\11x437 -1 nl-.Is.L1.‘1 “I0
I 420 POTENTIOMETRIC SURFACE (NGVD) --
FIGURE 1-7 POTENTIOMETRIC SURFACE MAP
-27-
When reviewing plezome ter information obtaineC from multiple
placement of piezometers in single boreholes, the technical reviewer
should closely scrutinize the construction deta:1s for the well. It is
extremely difficult to adequately seal several piezometers at discrete
depths within a single borehole, and special design considerations should
have been considered by the owner/operator. If detailed information for
the design 1s not availabie, it may in dicate that adequate construction
considerations have not been used. Placement of piezometers in closely
spaced well clusters, where piezometers have been screened at different,
discrete depth intervals, is more likely to Produce accurate
information. Additionally, multiple well clusters sampie a greater
proportion of the aquifer, and thus may provide a greater degree of
accuracy for considerations of vertical potentiometric head in the
aquifer as a whole.
The information obtained from the piezometer readings should have
been used by the owner/operator to construct flow nets (see Figure l-8).
These flow nets should include information as to piezometer depth and
length of screening. The flow net in Figure l-8 was developed from
information obtained from piezometer clusters screened at different,
discrete intervals. The technical reviewer should be able to verify the
accuracy of the owner/operator’s presentation and calculations by either
constructing a flow net independently from the owner/operator’s data or
spot- checking the owner/operator’s presentation. It 1s also important to
verify that t:?e screened interval is accurately portrayed and to
determine whether the piezometer is actually monitoring the water level
of the desired water-bearing unit.
If there is reasonable conc’Jrrence between the information presented
by the owner/operator and the technical reviewer’s interpretation, the
technical reviewer should next interpret the flow directions from the
waste management area.
-23-
OSWER-9950.1
ELEVATI
NGVD
960
900
880
860 I’-
LEGEND
I I I 1 1
100’ 50’ 0 50’ 100’
WELL AND SCREEN
< - - FLOW LINE
V - - - - POTENTIOMETRIC SURFACE
..*....**... EQUIPOTENTIAL LINE
FIGURE ‘I-8. AN EXAMPLE OF A FLOW NET DERIVED FROM PIEZOMETER DATA
-29-
1.3.1.4 Interpretation of fiow direction and flow rates
In considering flow directions establ 'ished by the owner/operator,
the technical reviewer should have first established:
l That the potentiometric surface measurements are valid; that is
the distributions of hydraulic head and hydraulic conductivity
are known, and that the total porosities as approximations of
effective porosities (determination of effective porosity can be
time consuming) of significant strata are known to permit
estimation of flow rate; and
l That the vertical components of flow have been accurately
depicted and are based on *Jalid data.
At this point, general direction(s) and rate(s) of ground-water flow
may be estimated. The technical reviewer should construct vertical
intercepts with the potentiometric contours for both the Gotentiometric
surface map and flow nets. 3nce the vertical and horizontal directions
of flow are established (from points of higher to lower hydraulic head),
it is possible to estimate where monitoring wells will most likely
intercept contaminant flow in the .Jertical slane. To consider the
placement that wll 1 most effectively intercept contaminant flow,
hydraulic conductivity(ies) must be calculated.
1.3.2 Seasonal and Te%oral Factors: Ground-Water Flow - -.- -.. -
It is important to note i f the owner/operator 1has identified and
assessed factors that may r esult in short-term or long-term vrariations in
ground-water level and flow patterns. Such factors that may influence
ground-water conditions include:
l Off-site well. pumping, recharges, and discharges;
l Tidal processes or other intermittent natsxal variations (e.g.,
river stage, etc.);
l On-site well pumping;
l Off-site, on-site construct
l Deep well injection; and
0 Waste disposal practices.
con or chang ing land use patterns;
gff-site or on-site wei pumping 3ay affect both the rate and
direction of ground-water flow. Xunicipal, industrial, or agricultural
OSWER-9950.1
ground-water use may significantly change ground-water flow patterns and
levels over time. Pumpage may be seasonal or dependent upon complex
water use patterns. The effects of pumpage thus may reflect continuous
or discontinuous patterns. Water level measurements in piezometers must
have been frequent enough to detect such water use patterns.
Natural processes such as riverine, estuarine, or marine tidal move-
ment may result in =rariations of well water levels and/or ground-water
quality. An owner/operator should have documented the effects of such
patterns. Seasonal patterns have a significant effect on hydraulic head
and ground-water flow. Short-term recharge patterns may affect ground-
water flow patterns that are markedly different from ground-water flow
patterns determined by seasonal averages. An owner/operator should have
gauged such transitional patterns.
Additionally, an owner/operator should have implemented means for
gauging long-term effects on water movement that may result from on-site
or off-site construction or changes in land-use patterns. Development
may affect ground-water flow by altering recharge or discharge patterns.
Exampies of such changes might include the paving of recharge areas or
damming of waterways.
In reviewing the owner/operator's assessment of ground-water flow
patterns, the technical reviewer should consider whether the owner/
operator's program was sensitive to such seasonal or temporal variations.
An owner/operator should have. in effect, determined not oniy the location
of water resources, but the sources and source patterns that contribute
to or affect ground-water patterns below the regulated site.
1.3.3 Determining Hydraulic Conductivities _. -. --.--_ _
In addition to defining vertical and horizontal gradients and
sources of spatial and temporal variation, the owner/operator must
identify the distribution hydraulic conductivity (K) values within each
significant formation. Variations in the hydraulic conductivity within
or between formations or strata can create irregularities in ground-water
flow paths. Strata/formations of high hydraulic conductivity represent
areas of greater ground-water flow and therefore zones of potential
migration. Furt:ler, anisotropy within strata or formations affects the
magnitude and direction of ground-water flow. Thus, information on
hydraulic conductivities is necessary before owner/operators can make
reasoned decisions regarding well placements.
Technical reviewers should review the owner/operator's hydrogeo-
logic assessment to ensure that it contains data on the hydraulic
conductivities of the significant formations underlying the site.
In addition, technical reviewers should review the method the owner/
operator used to derive the conductivity values. It may be beneficial to
use analogous or laboratory methods to augment results of field tests:
however, field methods provide the best definition of the hydraulic
conductivity in most cases.
Hydraulic conductivity can be determined in the field using either
single or multiple well tests. Single well tests, more commonly referred
to as slug tests, are performed by suddenly adding or removing a slug
(known volume) of water from a well and observing the recovery of the
water surface to its original level. Similar results can be achieved by
pressurizing the well casing, depressing the water level, and suddenly
releasing the pressure to simulate removal of water from the well. One
recommended method, which will be proposed for inclusion in SW-846 (Test
Methods for Evaluating Solid Waste, U.S. EPA, July 1982), is Method 9100,
which is also recommended for use in determining aquifer vulnerability.
When reviewing information obtained from single well tests, the
technical reviewer should consider several criteria. First, they are run
on one well and, as such, the information is limited in scope to the
geologic area directly adjacent to the screen. Second, the vertical
extent of screening will control the part of the geologic formation that
is being tested dur:ng the test. That part of the column above or below
the screened interval that has not been tested may also have to be tested
for hydraulic conductivity. Third, the methods that the owner/operator
-32-
OSWER-9950.1
used to collect the information obtained from single well tests should be
adequate to measure accurately parameters such as changing static water
{prior to initiation, during, and following completion of the test), the
amount of water added to, or removed from, the well, and the elapsed time
of recovery. This is especially important in highly permeable formations
where pressure transducers and high speed recording equipment may need to
be used. The owner/operator's interpretation of the single well test
data should be consistent with the existing geologic information (boring
log data). The well screen and filter pack adjacent to the interval
under examination should have been properly developed to ensure the
removal of fines or correct deleterious drilling effects. It is.
therefore, important that reviewers examine the owner/operator's program
of single well testing to ensure that enough tests were run to provide
representative measures of hydraulic conductivity and to document lateral
variations of hydraulic conductivity at various depths in the subsurface.
Multiple well tests, more commonly referred to as pumping tests, are
performed by pumping water from one well and observing the resulting
drawdown in nearby wells. Tests conducted with wells screened in the
same water-bearing formation provide hydraulic conductivity data. Tests
conducted with wells screened in different water-bearing zones furnish
information concerning hydraulic communication. Fultiple well tests for
hydraulic conductivity are advantageous because they characterize a
greater proportion of the subsurface and thus provide a greater amount of
detail. Multiple well tests are subject to similar constraints to those
listed above for single well tests. Some additional problems that should
have been considered by the owner/operator conducting a multiple well
test include: (1) storage of potentially contaminated water pumped from
the well system and (2) potential effects of ground-water pumping on
existing waste plumes. The technical reviewer should consider the
geologic constraints that the owner/operator has used to interpret the
pumping test results. Incorrect assumptions regarding geology may
translate into incorrect estimations of hydraulic conductivity.
-33-
In reviewing the owner/operator’s hydraulic conductivity measure-
ments, the technical reviewer should use the following criteria to
determine the accuracy or completeness of information.
• Values of hydraulic conductivity between wells in similar
lithologies should not exceed one order of magnitude difference.
If values exceed this difference, the owner/operator may not have
provided enough information to sufficiently define a potential
flow path, or there is a mistake in the logs.
• Hydraulic conductivity determinations based upon multiple well
tests are preferred. Multiple well tests provide more complete
information because they characterize a greater portion of the
subsurface.
• Use of single well tests will require that more individual tests
be conducted at different locations to sufficiently define
hydraulic conductivity variation across the site.
• Hydraulic conductivity information generally provides average
values for the entire area across a well screen. For more depth
discrete information, well screens will have to be shorter. If
the average hydraulic conductivity for a formation is required,
entire formations may have to be screened, or data taken from
overlapping clusters.
It is important that measurements define hydraulic conductivity both
vertically and horizontally across an owner/operator’s regulated site.
Laboratory tests may be necessary to ascertain vertical hydraulic
conductivity in saturated formations or strata. In assessing the
completeness of an owner/operator’s hydraulic conductivity measurements,
the technical reviewer should also consider results from the boring
program used to characterize the site geology. Zones of high permeability
or fractures identified from drilling logs should have been considered in
the determination of hydraulic conductivity. Additionally, information
from boring logs can be used to refine the data generated by single well
or pumping tests.
1.4 Identification of the Uppermost Aquifer
The owner/operator is required under 40 CFR § 265 Subpart F to monitor
the uppermost aquifer beneath the facility in order to immediately detect
-34-
OSWER-9950.1
a release. Proper identification of the uppermost aquifer is therefore
essential to the establishment of a compliant ground-water monitoring
system. EPA has defined the uppermost aquifer as the geologic formation,
group of formations, or part of a formation that is the aquifer nearest
to the ground surface and is capable of yielding a significant amount of
ground water to wells or springs (40 CFR 5260.10) and may include fill
material that is saturated. The identification of the confining layer
or lower boundary is an essential facet of the definition of uppermost
aquifer. There should be very limited interconnection, based upon
pumping tests, between the uppermost aquifer and lower aquifers.* If
zones of saturation capable of yielding significant amounts of water are
interconnected, they all comprise the uppermost aquifer. Quality and use
of ground water are not factors in the definition. Even though a
saturated formation may not be presently in use, or may contain water not
suitable for human consumption, it may deserve protection because contami-
nating it may threaten human health or the environment. Identification
of formations capable of "significant yield" must be made on a case-by-
case basis.
There are saturated zones, such as low permeability clay, that do
not yield a significant amount of water, yet act as pathways for
contamination that can migrate horizontally for some distance before
reaching a zone which yields a significant amount of water. If there is
reason to believe that a potential exists for contamination to escape
along such pathways, the technical reviewer may invoke enforcement and
permitting authorities other than § 265.91 to require such zones to be
monitored. These authorities include 3008(h) for interim status
*Some hydrogeologic settings (e.g., basin and range provinces, alluvial
depositional environments) do not offer a clear confining layer. In
such cases, the technical reviewer should note the situation and
concentrate on the placement of wells in the uppermost aquifer to
immediately detect potential releases of contaminants.
-35-
corrective action, 3004(u) for corrective action for permitting, the
omnibus condition authority under 3005(c) which mandates permit
conditions to protect human health and the environment, and 3013
authority which permits broad investigations. Of course, if a release
has been detected the plume should be characterized in such sa%urated
zones regardless of yield.
In all cases, the obligation to assess any hydraulic communication
and the proper definition of the uppermost aquifer rests with the
owner/operator. The owner/operator should be able to prove that the
confining unit is of sufficiently low permeability as to minimize the
passage of contaminants to saturated, stratigraphically lower units.
The following examples illustrate geologic settings wherein hydrau-
lic communication must be demonstrated before proper identification of
the uppermost aquifer can be made. The examples are not intended to be
exhaustive in the situations they portray; rather, they are meant to
provide a sampie of geoiogic settings that depict hydraulic communication.
Figure l-9 illustrates a site where preliminary drill logs indicated
a confining layer of unfractured, continuous clay beneath the site.
(Note: the actual geologic conditions are pictured for purposes of
clarity in the figure.) In order to confirm whether the clay layer is
continuous or discontinuous, the owner/operator conducted a pumping
test. A well at drill point No. 2 was screened at the uppermost par: of
the potentiometric surface. Another well at drill point No. 3 was
located close by and screened below the clay layer. Measurable
drawdown was observed :n the upper well when the well below the confining
layer was pumped. This indicated that the confining unit is not of
sufficient impermeability to serve as a significant boundary to
contaminant flow. In this case, the water-bearing unit below the clay
layer and the formation above the clay layer are both part of the
uppermost aquifer.
-36-
ELEVATION
NGVD
PI EZOMETER PIEZOMETER PIEZOMETER
3 2 1
,
,GR&i’JD-WATER FLOW
‘,“j 1 , ;.. .:. ‘..: #’ ‘1. “‘SAND ,, -,.. .“. : ;,--
., .:; >;, ,: ‘, ~ ;. :
.” .,.,(‘ ,1-,. i a : : : ‘;, :Y ‘.. ,. K = 7.0 x 10-3cm/sec ‘. ,
; .::. ,: ;’ “. 5,. : ,, /‘., : “. ‘. ; ,.I ,. ‘.
: : ; i : : ‘.I _. . .-.. ,’ ,,‘,. ____---------------
---L-z---- -=_-=-=======-===-_-==--- ------ __------ ------FL- - - - - - - --_-_--_-------- ------. -----_---------------- -- -_---------------- ---I--------,-- - - - - - ----------- ----_-__------------- --
-.-------- -- ------- -=====-==---==--==- _ --------- - __ ---- ----e-w ---------------------- _--. -L-z - - - - - - - - -----_-__------------ --- __----- ---I=,Y-~-~ - - - - ---_-__-_---------v-e --. __-----____-_----------A--
---~r--~-~-~-- -L--.7------------. ____--------- -----. ----.---- ------
-.c------ -- -:=T------------I CLAY -=====-=-~-~--==-=. __----- ----7-Y-z - - - - ---_--_--_--------e-m--. ------------------------ -- ------- ----. ------ - - - - --- ----- ------- ___------ -~-L.-~-~ - - - - - - -- K = 3.5 x lo-“cm/set -~-~-~-~--------1
--L-I--- -l-.----- -- -- --------. =-----=----_-__----=----- - - - - -- -- __-------------------- -~-~--7-~---- - - ------. __-_---------_------ ------
-L-Y-- -- -- -- ------ -..T.-Y--Y------------- - --- - --------- - - -- -----. ------- ------- -LT..---- - - - - - - ---------------k-d-----. ___-__-__--_--------- --- ___---------------- ---~-~-~---- - - ----------- __-_---------------- __--.--
---l-I-- -- ----------- --- =-1-------------- - - - --- - - - z- -__--. __------ --. ------------ ------ -===_-==--=--====-=-- _- - - --_-- __--------- ---L-L=-.. - - - - - - --------------------. --------- --_------------
I ,,“.I , I
100’ 50’ 0 50 100 F
I
v POTENTIOMETRIC
--- SURFACE
I! WELL AND SCREEN
FIGURE l-9 EXAMPLE OF HYDRAULIC COMMUNICATION BETWEEN
WATER -BEARING UNITS
-37-
In Figure l-10, the owner/operator drilled test borings throush sand
and limestone formations Into a sandstone unit. 11-. the Initial cores, CO
indication of fracturing 0, F the limestone unit was observed. The owner/
operator initially assumed that the limestone unit dips at a m0derat.e
slope due to differing levels of contact. However, as illustrated by the
figure, actual conditions involve faulting and post-depositional eros:on
of the limestone formacior. (additional corings and geophysical studies
detected fracture zones). These fractures represent hydraul:c corxn..xl:ca-
t10r. between the upper unconsolidated sand layer and the sandstone
formation below the limestone unit. The uppermost aquifer, therefore,
includes the unconsoiidated sand formation, t!-.e limestone formation, and
the sandstone formation.
Figure l-11 illustrates a situation where perched water zones Lie
above the potentiometric surface. The containment pathway includes the
perched water zones and that part of the sand formation from the top of
the potentiometric surface to the top of the granitic basement.
In Figure L-12, initial test borings indicated that horizontal sand
units are underlain by a consolidated. well-cemented, limestone unit.
Initial borings did not indicate the presence of the anticline. The
owner/operator incorrectly assumed that the sandstone unit was a confining
layer that extended across :he subsurface below the site. A doiomite
unit, ir. contact w:th the unconsolidated sandy silts and dlrec-l:/ below
the waste unit, is fracture? and highly permeable. Additional in’~es:ica-
tion including ?ump tests. borings, and/or geophysical anai::sis betts-
defined the subsurface. The uppermost aquifer, in this case, includes
the anticlinal formations,
In Figure 1-13. unconsolidated units are underlain by a consolidated
series of variabie, near-shore, shallow marine sediments. The cxner/
operator has installed three borings near th e waste management ur.15 to
identify the uppermost aquifer. Interpretation of these bor:nqs :nSicat;Js
that the unconsolidated units are underlain by a xell-cemanc.ed l:mesror:2
OSWER-a950 1 J .
ELEVATION
NGVD
300
250’
200
150
100
PlEZOMETER PlEZOMETER PIEZOMETER PIEZOMETER
1 4 3 2
-LIMESTONE
T
WASTE DISPOSAL
)i>‘,-. .: . ,.;Y.,. p. ..:. . . ,_‘. ” &‘$ FAULT LINE
I I,, 1
;K=3.5x10-5cm/sec I I h [
I I , , , 1 I 1 \I [ HYDRAULIC - I I T CONDUCTIVITY
I I 1 I I I I J
100’ 50’ 0 50 100’
POTENTIOMETRIC
SURFACE
CALCIUM
BENTONITE
’ PLUGGED
BOREHOLE
FIGURE l-10 AN EXAMPLE OF HYDRAULIC COMMUNKATION CAUSED BY FAULTING
-39-
LANDFILL
SAND
PERCHEO WATER ZONES
poTENTlOMETRlC
-J-2zcE UPPERMOST
AQUIFER
.--b-o.- ~
---nn --divw- -w--Ie
jAN0 “.’
I
I. ,
FIGURE I-11 PERCHED WATER ZONES AS PART OF THE UPPERMOST AQUIFER
-4o-
3SXER-9950.1
BOREHOLE BOREHOLE
1 2
WASTE UNIT
, :;. “Z,.
“‘,.., ‘*‘,j.‘, ‘,: __. .-: K = 7.0 x IO-~,,/,,, ‘;:, :::.,‘. .:,;
I 1 I I I I,, , , , , , , , ,
I
LCALCl”MBENTONlTE,‘,‘,‘,1L’ME~ToNEl’I’,’I’,’I’I’I’,’ ’ ’ ’
IZPLUGGED~OREH~LE ’ I ’ 1 ’ 1 1 1 1 & I i I I I
‘. < K = 10-5cm/sec
I I ’ I ’ I ’ I ’
3.5 x
- ;-.,.r ,- *,.;:. .,, ..;;. . . . . .
. .* ..;.,.,.“. 1...:;. SANDSTONE
.; ,..,,. .,;,r ,.,x ,,
. ,... &..~
,,,. . . . ..+&.Y..
’ .r:.::., ,;.p:,,:: ::, .;A;.“..:; : . .:*. ., ‘.; , _” ,,.,, ~,:d K = 3.ox ,o-lo cm,sec
I 1 I I 1
1Of 50 0 50 100
I LEGEND
B WELL AND SCREEN
POTENTIOMETRIC
- x -SURFACE
FIGURE 1-12 AN EXAMPLE OF AN UNDETECTED STRUCTURE IN THE UPPERMOST
AQUIFER (VERTICAL SCALE IS EXAGGERATED).
-31-
BOREHOLE BOREHOLE BOREHOLE
1 2 3
FINE SAND’
. .
I-- _. K = 5.0 x 10-4cmisec -_ _ .
POTENTIOMETRIC
SURFACE
FIGURE 1-13 AN EXAMPLE OF AN UNDETECTED PORTION OF THE UPPERMOST AQUIFER
DUE TO AN IMPROPERLY SCREENED BOREHOLE (VERTICAL SCALE IS
EXAGGERATED)
-42-
OSXER-9950.1
of very low permeability. However , an undetected sandstone unit, which
is laterally continuous with the limestone unit, is highly permeable and
saturated and represents an undetected portion of the uppermost aquifer.
Interpretation of the depositional environment of the limestone unit,
coupled with a knowledge of the local or regional geology, should have
been used in addition to other investigatory techniques to establish the
presence of the transitional lateral structural feature and thus properly
define the uppermost aquifer.
A special case that should be considered by the technical reviewer
is the possibility that existing wells may provide avenues for hydraulic
communication between hydrogeologic units. This is of special importance
when considering a site where a contaminant plume may have migrated down-
gradient to the extent that the plume approaches off-site wells. Such
wells may not have been construct ,ed in a manner sensitive to problems of
cross-contamination between aquifers (see Chapter Four).
The goal of the site charact erization is the identification of
potential pathways for contaminant migration in the uppermost aquifer.
The next step is to complete the installation of monitoring wells and
piezometers in those pathways and upgradient, which will comprise the
detection monitoring network.
REFERENCES
Anderson, D.C. and S.G. Jones. 1983. Clay Barrier-Leachate Interaction.
Proceedings of the National Conference on Management of Uncontrolled
Hazardous Waste Sites, pp. 154-160.
ASTM D2434-68. Reapproved 1974. Standard Test Method for Permeability
of Granular Soils (Constant Head). Annual Book of ASTM Standards:
Part 19 - Natural Building Stones; Soil and Rock. 7 PP.
Brown, K.W., J.C. Thomas, and J.W. Green. 1984. Permeability of
*Compacted Soils to Solvent Mixtures and Petroleum Products. Land
Disposal of Hazardous Waste. 10th Ann. Res. Symp., pp. 124-137.
Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Prentice-Hall, Inc.
Heath, Ralph C. 1953. Basic Ground-Water Hydrology. United States
Geological Survey, Water Supply Paper 2220.
Pollack C.R., G.A. Robbins, and C.C. Mathewson. 1983. Groundwater
Monitoring in Clay-Rich Strata--Techniques, Difficulties, and
Potential Solutions. 3rd National Symp. of Aquifer and Groundwater
Monitoring, pp. 347-354.
U. S. Army Corps of Engineers. 1970. Falling-Head Permeability Test with
Permeameter Cylinder. Appendix VII, Section 4, Laboratory Soils
Testing, Engineering Manual 1110-2-1906, pp. VII-13 to VII-16.
U. S. Army Corps of Engineers. 1970. Permeability Tests with
Consolidometer. Appendix VII, Section 8, Laboratory Soils Testing,
Engineering Manual 1110-2-1906, pp. VII-22 to VII-24.
U. S. Environmental Protection Agency. 1983. RCRA Draft Permit Writer's
Ground-Water Protection, 40 CFR Part 264, Subpart F. U.S. Environ-
mental Protection Agency Contract No. 68-01-6464.
U. S. Environmental Protection Protection Agency. 1983. Ground-Water Monitoring
Guidance for Owners and Operators of Interim Status Facilities.
National Technical Information Service. PB83-209445.
U. S. Environmental Protection Agency. September 1985. Protection of
Public Water Supplies from Ground-Water Contamination. EPA/625/4-85/
016.
U. S. Department of Interior, Bureau of Reclamation. 1974. Designation
E-15, One-Dimensional Consolidation of Soils. Earth Manual, 2nd
Edition, pp. 509-521.
-44-
OSWER-9950.1
CHAPTER TWO
PLACEMENT CF DETECTION MONITORING WELLS
The purpose of this chapter is to examine criteria the technical
reviewer should use in deciding if the owner/operator has made proper
decisions regarding the number and location of detection monitoring
wells. In evaluating the design of an owner/operator's detection
monitoring system, the technical reviewer should examine the placement of
upgradient and downgradient monitoring wells relative to hazardous waste
management units, and review the placement and screening of detection
monitoring wells for their interception of predicted pathways of
migration. The minimum number of monitoring wells an owner/operator may
install in a detection monitoring system under the regulations is
four--one upgradient well and three downgradient wells. Typically, site
hydrogeology is too complex or the hazardous waste unit is too large for
the regulatory minimum number of wells to prove adequate in achieving the
performance objectives of a detection monitoring system.
A fundamental concept that will be emphasized throughout this chapter
is that the placement and screening of wells in the detection monitoring
network will be based on the results of a thorough site characterization.
The basic goals of the site characterization process as described in
Chapter One are the description of the hydrogeological regime and the
identification of the uppermost aquifer and potential pathways for
contaminant migration. This information is the foundation for the entire
ground-water monitoring program and crucial to the placement of detection
monitoring wells in particular. It is likely that the technical reviewer
may encounter situations where the owner/operator has collected little or
no site hydrogeologic information or has relied exclusively on regional
data to design a monitoring system. In this situation, the technical
reviewer should carefully examine the decisions the owner/operator has
made regarding well placement and screen depths, and it may be necessary
to require the owner/operator to collect additional site information.
-45-
Upgradient monitoring wells are to provide background ground-water
quality data in the uppermost aquifer. Upgradient wells must be
(1) located beyond the upgradient extent of potential contamination from
the hazardous waste management unit to provide samples representative of
background water quality, (2) screened at the same stratigraphic
horizon(s) as the downgradient wells to ensure comparability of data, and
(3) of sufficient number to account for heterogeneity in background
ground-water quality.
It is important to recognize that potential pathways for contaminant
migration are three dimensional. Consequently, the design of a detection
monitoring network that intercepts these potential pathways requires a
three-dimensional approach. Downgradient monitoring wells must be
located at the edge of hazardous waste management units to satisfy the
regulatory requirements for immediate detection. The placement of
detection monitoring wells along the downgradient perimeter of hazardous
waste management units must be based upon the abundance, extent, and the
physical/chemical characteristics of the potential contaminant pathways.
The depths at which contaminants may be located and at which downgradient
wells must be screened are functions of (1) geologic factors influencing
the potential contaminant pathways of migration to the uppermost aquifer,
(2) chemical characteristics of the hazardous waste controlling its
likely movement and distribution in the aquifer, and (3) hydrologic
factors likely to have an impact on contaminant movement (and
detection). The consideration of these factors in evaluating the design
of detection monitoring systems is described in Section 2.1.3.
A sufficient number of detection monitoring wells screened at the
proper depths must be installed by the owner/operator to ensure that the
ground-water monitoring system provides prompt detection of contaminant
releases. A detection monitoring system should be judged against site-
specific conditions: however, there are a number of criteria that
-46-
OSWER-9950.1
technical reviewers can apply to ensure that detection monitoring systems
satisfy the RCRA regulatory requirements. This chapter describes those
criteria and provides examples on how technical reviewers can evaluate
detection monitoring systems, in various hydrologic situations. This
chapter also examines three common geologic environments: alluvial,
karst, and a glacial till. The rationale for well placement and vertical
sampling intervals within each geologic environment is discussed.
2.1 Placement of Downgradient Detection Monitoring Wells
The criteria for evaluating the location of downgradient wells
relative to waste management areas are described in Section 2.1.1.
Section 2.1.2 contains the criteria for evaluating horizontal placement
of downgradient detection wells. Section 2.1.3 details the rationale for
selection of the vertical placement and sampling intervals of detection
monitoring wells. Discussed in Section 2.1.4 are three geologic settings
that have been encountered at hazardous waste sites and the rationale for
detection well placement at each site.
2.1.1 Location of Wells Relative - to Waste Management Areas
In order to immediately detect releases as required by the
regulations, the owner/operator must install downgradient detection
monitoring wells adjacent to hazardous waste management units. In a
practical sense, this means the owner/operator must install detection
monitoring wells as close as physically possible to the edge of hazardous
waste management unit(s). The two drawings in Figure 2-1 (A and B)
illustrate the concept of the placement of wells immediately adjacent to
hazardous waste management unit(s). Note: the placement of wells
relative to the units shifts as a function of the direction of
ground-water flow.
Geologic environments with discrete solution channels such as Karst
formations must have detection monitoring wells located in those solution
channels likely to serve as conduits for contamination migration.
-47-
FIGURE 2-1 DOWNGRADIENT WELLS IMMEDIATELY ADJACENT TO
HAZARDOUS WASTE MANAGEMENT LIMITS
-48-
OSWER-9950.1
At sites underlain by interbedded, unconsolidated sands, silts, and
clays (e.g., alluvial facies) where the potentlometric surface is
deep-seated, the lateral component of contaminant migration may carry
contaminants beyond the ground-water monitoring system before they reach
ground water, and therefore beyond detection. The owner/operators could
institute a program of vadose zone monitoring as a supplement to the
ground-water monitoring program in such cases. to provide immediate
detection of any release(s) from th e hazardous waste management area.
Volatile organics that escape to the vadose zone, for instance, may be
detected and characterized through soil gas analysis.
2.1.2 Horizontal Placement of Downgradient Monitoring Weils
The horizontal placement of detection monitoring wells along the
downgradient perimeter of hazardous waste management units should be
predicated on the interception of Potential pathways for contaminant
migration. The majority of hazardous waste sites will have identifiable
pathways for potential contaminant migration. Some potential pathways
for contaminant migration are: zones with relatively high intrinsic
(matrix) hydraulic conductivities, fractured/faulted zones, solution
channels, and zones suspected to be incompatible with the waste{s)
present. Sites located in heterogeneous geologic settings can have
numerous, discrete zones of potential migration. Each zone of potential
migration must be identified and monitored.
Within a potential migration pathway, the horizontai distance
between wells should be based upon site-specific factors such as those
described in Table 2-1 should be considered by technical reviewers when
evaluating the horizontai distance between detection wells. These
factors cover a varie ty of physical and operational aspects relating to
the facility, including hydrogeologic setting, dispersivity, seepage
velocity, facility ciesign, and waste characteristics.
-49-
TABLE 2-l
FACTORS INFLUENCING THE INTERVALS BETWEEN XNDI'dIOUAL MONITORING WELLS
WITHIN A POTENTIAL MIGRATION PATHWAY
WELL UTERVALS MAY BE CLOSER IF THE SITE: WELL INTERVALS MAY BE WIDER IF THE SITE:
. Manages or has managed liquid waste
. Is very small
. Has fill materla
management units
flow might occur
1 near the waste
(where preferen:ia!
. 4as bur-ied p'pes. utility trenches. etc..
,where a point-source leak miqht occur
. Has complicated geology
- closely spaced 'ractures
- faults
tight folds
- solution channels
- discontinuous structures
. Has heterogeneous conditions
- variable iydraulic conductivity
- variable lithology
. Is located in or near a recharge zone
. Has a steep or variable hydraul-c
gradient
. Is characterized by low dispersivity
potentiai
. Has a high seepage velocity
l Has simple geology
- ro fractures
no faults
- no folds
- no solution channels
- continuous structures
. Has homogeneous conditions
uniform hydraulrc conductivity
- uniform lithology
. Has a low (flat) and constant hydraulic
gradient
l Is characterized by high dispersivity
potential
. Has a low seepage velocity
-5u-
;r. the less common homogeneous geologic settir.g where no preferred
pathways are identified, a more regular well placement pattern can be
utilized based on formational characteristics (e.g.. dispersivity,
hydraulic conductivity, and other factors listed in Table 2-i).
2. :. 3 Vertical Placement and Screen Lengths -.
This d0c.lrrer.t addresses separately the ‘horizontal placement and the
T?2 rtlcal sampling inter.Jals of ‘detection monitoring wells. These two
parameters, 3.owe~~er, should be e.Jaiuated together in the design of the
ground-water detection monitoring system. Proper selection of the
vertical sampilng interval provides the third dimension to the detection
monitoring of potential contaminant pathways to the uppermos: ag%Aifer.
Site-specific hydrogeologic data obtained by the owner/operator during
the site characterization are essential for the determination of the
horizontal placement of detection wells, and for the selection of the
vertical sampling interval(s). Proper design of a detection monitoring
system enables the owner/operator to select the vertical sampling
interval capable of immediately detecting a release from the hazardous
waste management area. It is essential, therefore, that the
owner/operator’s decisions regarding vertical sampling inter:lals are
based upon a full site charac:erization, which defines both the depth and
thickness of the stratigraphic horizon(s) that could ser-Je as contaminant
pathways, There are several guidelines or criteria that the tezhr:ical
reviewer should follow in evaluating owner/operator decisions. A
discussion of these guidelines follows in the examples in Section 2.1.4.
The owner/operator should have determined from the site character:-
zation which stratigraphic horizons renresent potential pathways for
contaminant migration, and should screen monitoring wells at the
appropriate horizon(s) to provide immediate detection of a release. It
is extremei:; important to screen upgradient and downgradient wells in the
-;- -_-
same stratlgrapk.ic horizor.( s) to obta:n comparabie ground-water quality
data. <as long as the str3ta are not dipping too strongly. The owner/
operator should have ensured and demonstrated that the upgradient and
ior~rc+-2iient well s-re3n.s ‘,LL L Q - intercepted the same uppermost aquifer. The
Iet?rr::-.ation of +I-YE “.e?t:i to a potential contaminant migration pathway
may be iXi:ie f fern 50111 rock cores, supplemented by geophysical and
a ‘“‘a 1 1 at 1 $3 res:onal/!o;al :l;ldrogeologlcal data.
Ar.oths:- factor IO Se Tonsidered in selecting the depth at which
ve ! 1 s s’noui.: Se placze (3::d the seiection of well screen lengths) is the
physxa! ichsr?:ral cha:-3r-‘teL _ r*stics of the hazardous waste or hazardous
*era s t e constlruenrs Zontrol ling the movement and distribution of contamina-
tion in The aqalfer. The technical rerriewer should consider the mobility
of tkr !-2zar~~.~ous warjZ2, 1:s pocentlal reaction products, and the potential
for chemlca’. degradation of clays. Different transport processes control
cont;m:nan: mo’:erent jepsndrng on P:hether the contaminant dissolves in
water or :s ;mmis::C13. .$ _ _ I-mic,- lb;e contaminants may vary from extremely
1 igh: .=oLat:irs to drns? organic liquids whose migration is governed
largely by ?.3:13i:y 3::s *.riscosity. Lighter than water phases spread
-ap:dly ;n the cap::l3:-y zone 2~1s: above the potentiometric surface,
Altsrnatlve:., I ’ ‘the ~,lgr3: ion of 2cr,se organic liquids is largely
ar:co~pTed from the h;~<raulic grad lent that drives advective transport and
.30v:?::e~c; ma-: :?a:,? 3 Icrr::!.3r.t vertlcai component even in horizontall;J
c - _ lowli:y ;ii~:~l~~~rs” :.‘.!2 :K;iy, et 3.i. , 1985).
: r-t .addition to :.;?e normai flow of ground water (advection), the
c:?en::ca: St-ocesses sf 2ispe:-slor. and sorption (retardation) greatly
inf lILe2ce t!?+ potent ia: rnigratlon pathways of contaminants within an
aquifer. 21sper;ion 1s tke spread of contaminants resulting from
mol+cL;lar Sif - f~:s:on 3i:d aech3::lcal mixing and “may result in the arrival
of .3eteztable cont+::;nari: concei-.tratioCs at a given iocation significantly
before Lh~ >--ri-.-al tin:e ::;at is -1s_ -- expected solely on the basis of the
3 ‘,’ 2 L‘ 3 ,’ tl- .I : ::r.d-wa:lr flow : .:+” (\iac:iay, et a;. , 1985). The moblii::; of
GSWER-9950.1
different leachate constituents will vary depending upon the extent to
which each constituent is adsorbed to solid surfaces (sorption processes).
Some nonreactive ionic species (e.g., chioride ion) and low molecular
weight organics of relatively high water solubility (e.g., trichloro-
ethylene) can be quite mobile. Heavy metals (e.g., lead) and organics
with high molecular weights and relatively low solubilities in water
(e.g., chlorinated benzenes) tend to be the least mobile in natural
conditions of near neutral pH and Eh.
All of these processes are important in choosing the depth of the
screened interval and locating monitoring wells, because contaminants may
be confined to and move within narrow zones. For instance, to monitor
for heavy metals the screened interval should be just above the confining
layer--for light organics, at the potentiometric surface/capillary zone
interface. The local lithological variation can influence the rate,
quantity, and degree of sorption of particular contaminants and is
important in the proper location of monitoring wells.
Studies have shown that certain organic liquids can cause desiccation
cracks in clay which can lead to significant increases in permeability.
When organic chemicals and strongly acidic wastes are present, the com-
patibility of these wastes and chemicals with any potentially confining
clay layer(s) should be confirmed.
Determination of the appropriate thickness of the vertical sampiing
interval(s) is a natural extension of the depth selection. The owner/
operator should have made the decision on the basis of site characteriza-
tion data. Sources of information that can be used in determining the
thickness of potential contaminant pathways can include isopach maps of
highly permeable strata, coring data, sieve analysis, and fracture traces.
The lengths of well screens used in ground-water monitoring wells
can be a significant factor in the detection of releases of contaminants.
The complexity of the hydrogeology at a site is an important consideration
-53-
when selecting the lengths of well screens. Most hydrogeologic settings
are complex (heterogeneous, anisotropic) and the permeability is variable
with depth due to interbedded sediments. Highly variable formations
require shorter well screens, which allow sampling of discrete portions
of the formation. Longer well screens that span more than a single flow
zone can result in excessive dilution of a contaminant present in one
zone by uncontaminated ground water in another zone. This dilution can
make contaminant detection difficult or impossible, since contaminant
concentrations may be reduced to levels below the detection limits for
the prescribed analyticai methods.
Even in hydrologically simple (homogeneous) formations or within a
potential pathway for contaminant migration, the use of shorter well
screens may be required to detect contaminants concentrated at a
particular depth. A contaminant may be concentrated at a particular
depth because of its physical/chemical properties and/or hydroiogic
factors. In this situation, a longer well screen (length of well screen
>> thickness of the contamination zone) can permit excessive amounts of
uncontaminated formation water to dilute the contaminated ground water
entering the well. This resultant dilution may prevent the detection of
statistically significant changes in indicator parameters (pH changes)
and, In extreme cases, the diluted concentration of contaminants may be
below detection limits of the laboratory method being used.
The use of shorter we:1 screens helps to maintain chemical resolution
by reducing excessive dilution and, when placed at depths of predicted
preferential flow, such screens can monitor the aquifer or portion of the
aquifer of concern. The importance of determining these preferential
fiow paths in the ground-water monitoring process confirms the need for
a complete hycirogeologic site investigation prior to the design and
placement of detection wells.
-54-
CSWER-9950. i
Monitoring wells can be used to confirm or detect changes in ground-
water flow directions (determined during the site characterization) by
comparisons of potentiometric levels in neighboring wells. In hetero-
geneous geologic settings, however, longer well screens can intercept
stratigraphic horizons with different (contrasting) ground-water flow
directions. In this situation, the potentiometric surface will not
provide the depth discrete head measurements required for accurate
ground-water flow direction determination.
Certain hydrogeologic settings necessitate the use of longer well
screens for detection monitoring. Hydrogeologic settings with widely
fluctuating potentiometric surfaces are better monitored with longer
screens that continuously intercept the water surface and provide moni-
toring for the presence of contaminants less dense than water. Formations
with low hydraulic conductivities can also necessitate the use of longer
well screens to allow sufficient amounts of formation water to enter the
well for sampling.
Note: The vertical sampling interval is not necessarily synonymous
with aquifer thickness. In other words, the owner/operator may select an
interval which represents a portion of the thickness of t‘he uppermost
aquifer. When a single well cannot adequately intercept and monitor the
vertical extent of a potential pathway of contaminant migration at each
sampling location, the owner/operator should have installed a well
cluster. A well cluster is a number of wells grouped closely together
but not in the same borehole and often screened at different stratigraphic
horizons. The greater the need for stratified sampling, the more wells
the owner/operator should place in a cluster. The use of well clusters
is illustrated in the examples in Section 2.1.4.
There are situations where the owner/operator should have multiple
wells at a sampling location and others where typically one well is
sufficient. They are summarized in Table 2-2. The potential for
-55-
TABLE 2-2
FACTORS AFFECTIKC NUi?BER OF WELLS PER LOCATION (CLUSTERS)
One Veil Per Sampling Location
l No "sinkers" or "floaters"
(immiscible liquid phases;
see glossary for more detail)
0 Thin flow zone (relative to
screen length)
l Homogeneous uppermost aquifer;
simple geology
More Than One Well Per Sampling
l Presence of sinkers or
floaters
l Heterogeneous uppermost aquifer;
complicated geology
- multiple, interconnected
aquifers
- variable lithology
- perched water zone
- discontinuous structures
l Discrete fracture zones
OSWER-9950.1
immiscibles in a thick, complex saturated zone of the uppermost aquifer
should prompt the owner/operator to use well clusters. Conversely, in
situations where ground water is contaminated by a single contaminant,
and geologically there is a thin saturated zone within the uppermost
aquifer or homogeneous hydrologic properties are prevalent in the
uppermost aquifer, the need for multiple wells at each sampling iocation
is reduced. The number of wells screened at specific depths that should
be installed at each sampling location increases with site complexity.
Each potential contaminant pathway must be screened to ensure prompt
detection of a hazardous waste or hazardous waste constituent release.
2 .1.4 Examples of Detection Well Placemen: in Three Common Cmeoaic -
Environments
The following examples are based on actual geologic environments
encountered during hydrogeologic investigations. The three geologic
settings presented--a Karst, an alluvial, and a glacial till--are not
intended to be inclusive of all hydrogeologic factors; however, they are
illustrative of the technique used in the design of a minimum detection
monitoring system. The basic steps in the development of a detectlon
monitoring network include: (1) a review of existing information to
determine the regional geologic regime and regional ground-water flow
rates and direction: (2) a hydroqeologic investigation of the site to
determine the depth to and the extent of the uppermost aquifer; the
presence and extent of any confining layers/units; the abundance,
location(s), and extent of any potential pathways for contaminant
migration: and the direction and flow rates of the ground water; (3) a
review of the waste analysis plan to determine the chemical/physical
properties that may affect the distribution of a contaminant in the
aquifer; (4) the installation of detection wells in order to intercept
and completely monitor the potential pathways of contaminant migration;
(5) the selection of well screen lengths to provide resolute ground-water
samples; and (6) the placement/screening of upgradient monitoring wells
to provide representative background samples.
-57-
Figures 2-2, 2-3, and 2-4 depict a block diagram, a cross section,
and plan views of two lined waste impoundments located in a glacial till
envirorment. This heterogeneous glacial terrain is encountered in many
parts of the country, especially northern states. A review of the
published regional geologic data aided the subsequent and thorough site-
specific hydrogeologic investigation that made it possible to identify
three lirhoiogic units in The upper 100 feet of sediments overlying a
granite with low hydraulic conductivity. These units were identified by
geologic and geophysical analysis. Color, grain size, and texture were
also used to characterize each unit. Two sand units are separated by an
undulating glacial till varying between 10 and 50 feet thick. Pumping/
slug tests were conducted to determine the hydraulic conductivities of
each unit. These tests in conjunction with piezometer (not shown in
Figure 2-3) readings identified hydraulic intercommunication between the
two sand units. This -Jertical flow from the upper sand unit to the lower
sand unit is predominantly a function of the thickness and continuity of
the till unit. In locations where the till is thinnest, vertical flow is
most prevalent. Borings show that the granite confining unit extends
laterally across the entire site. Therefore, the uppermost aquifer
includes the two sand units and the till.
Flow in the upper sand unit is southerly, towards a nearby river,
and has a moderate hydraulic gradient of 0.01. Flow in the iower sand is
representative of regional ground-water flow generally to the south-
east. This lower outwash sand has a low hydraulic gradient of .004.
Figure 2-4 contains two plan views showing the equipotentiai lines in the
upper and lower sand units. These equipotential lines rere drawn using
information from the well/piezometric data tabulated on Figure 2-4. The
block diagram in Figure 2-2 illustrates :he multiple ground-water flow
paths present in this glacial terrain. The southern and eastern
perimeters of the waste lagoons are downgradient and therefore require
monitoring. The cross section in Figure 2-3 depicts the well placement
-5e-
OSXER-9950.1
DIRECTION OF
GRCkUND-WATER FLOW
IN UPPER SAND AQUIFER
DIRECTION OF
GROUND-WATER
IN LOWER SAND
: FLOW
AQUIFER
I LEGEND I
0 UPGRADIENT MONITORING WELL ~~ SAN,,
l DOWNGRADIENT MONITORING WELL GLACIAL TILL
a MONITORING WELL CLUSTER GRANITE
FlGURE 2-2 ILLUSTRATION OF MULTIPLE GROUND-WATER FLOW PATHS IN THE
UPPERMOST AQUIFER DUE TO HYDROGEOLOGIC HETEROGENElTY
-59--
LEGEND
Ki SODIUM BENTONITE PLUGGED BOREHOLE
0 UPGRADIENT MONITORING WELL IMWJ
l DOWNGRADIENT MONITORING WELL (MW)
0 PIEZOMETER (PI
A SURFACE IMPOUNDMENT
(PROJECTED 90 FEET)
~.
I
WELL 5 WELL 6 I
WELL 9
WELL7 /
(PROJECTED
(PROJECTED 30 FT) WELL 9 ,PROJECTED
I El WELL AND SCREEN
1
_ (PRFyFCTED ~PROJECTED
,O(J FT) GAMMA RESISTIVITY
3BA - (C.P.S.) ohm - FT
760
100 200 300 400 600 700 800 900 1.m (LOG OF WELL 101 1,100 (FEETJ
OF UPPER
OF LOWER
FIGURE 2-3 MONITORING WELL PLACEMENT AND SCREEN LENGTHS IN A GLACIAL TERRAIN
-6O-
OSWER-9950.1
820
815
810
t
. . . . . h4 0’. . . .$ . . . 0:(-J;....
13 I
. . .
6
0 6 I
765
763
761 -- H@
HYDRAULIC GRADIENT r .OO4
0 300 FT.
LOWER SAND
0
It
MONITORING WELL/PfEZOMETER TUPOF CASING ELEVATIONS
IIN FEET RELATIVE ~0 NATIONAL GEODETIC VERTICAL DATUM)
11
Pl
PZ
P3
P4
P5
P6
815.35 14/l/85) w1 765.04
819.81 (4/l/651 MW2 765.10
763.19 14/Z/85) W6 761.6O
761.48 (4/Z/851 MW 7A 811.30
819.50 14/l/85) Mw 76 761.55
811.26 (4/l/85) MW 9A 811.45
Mw 11 761.59
Mw 12 815.67
Mw 13 763.24
(4121851
(4121851
14/2/65)
(4/l/85)
14/l/85)
(4/l/85)
(4/Z/85)
14/l/85)
1412185)
I’ I(
FIGURE 24 A & B. PLAN VIEW OF FIGURE 2-3 SHOWING LINES OF EQUIPOTENTIAL
IN THE UPPER (A) AND LOWER (I31 SAND UNITS
-61-
and screen lengths for the detection monirorlng network along the
southern perimeter of the impoundment. Along the southern perimeter, the
upper sand unit requires more stringent monitoring than the lower sand
unit because of the higher ground-water velocity and steeper gradient in
the upper zone. Any release must seep through the upper sand before it
reaches the till. The hydraulic head resulting from the depth of liquid
in the lagoons, and an inventory of wastes and byproducts, indicate the
potential for "sinkers and floaters." The decision regarding horizontal
well placement was also based upon the likeiy size of a leak, the
distance from a leak source to the downgradient perimeter, dispersion,
and seepage velocity. Well placement in the lower sand unit along the
southern perimeter reflects the easterly component of ground-water flow
in the lower sand, that is, wells screened in the lower sand are located
toward the eastern end of the lagoons. It is important to note the care
that must he taken to properly grout the boreholes (wells) penetrating
the less permeable till to avoid increasing the (or cause a) hydraulic
communication between the sand units.
Figure 2-5 illustrates a cross section and plan view of a landfill
that may occur in an alluvial setting. A review of the regional and
local geology indicated that the area was possibly underlain by
interbedded sand and clay units. Spiit spoon samples collected during
the site-specific characterization revealed a massive clay unit extending
across the entire area at a depth of approximately 100 feet. Borehole
samples and interpretation of geophysical logs suggested that two sand
units overlie the massive clay, separated by a clay layer of variable
thickness. The upper sand contains several clay lens, each averaging
approximately 20 feet thick, beneath the disposal area. Pumping tests
within the sand units provided hydraulic conductivity values for the sand
units. Laboratory tests were used to determine hydraulic conductivity
values for the clay. Further analysis of clay samples identified an
illitic clay. Pumping tests across c_he inter.Jen:rq clay estab?lshed
hydraulic communication between the sand units with downward flow.
OSWER-9950.1
LEGEND
mm SUMMER DOWNGRAOIENT AREA 0 UPGf?ADlENT/BACKGROUND WELL
m WINTER DOWNGRADIENT AREA 0 MONITORING WELL
-. P . - POTENTlOMETRlC SURFACE OF 03 WELL CLUSTER
OF UPPER UNIT
9- POTENTIDMETRIC SURFACE OF
LOWER UNIT
lANDFILL r lPROJECTED 100 TO 225 FEETI
I
A WELL 4
WELL 2 WELL 3 ~PRDJECTED 251 (PROJECTED 50 FEET)
CD WELL 5 WELL 6 (PROJECTED 300 FEET)
tmt WELL7 P
1, “.. K - 2.0 x lOA cm/see “’ I
H.‘.‘. : .,, .;::.::. :GR~U~~~-WATER FLow .’ :. . :
.: -..:: ..,’ - .‘$e . . . . .;.. :I..,., . . .; ._ :_ ” ,.. : ----
GAMMA SP
(C.P.S.1 (MILLIVOLTS)
160
106 200 300 400 500 6W 700 800 900 1,&O l.iDO 1.2~ (LOG OF WELL Bc) (FEETI
FIGURE 2-5 MONlTORlNG WELL PLACEMENT AND SCREEN LENGTHS IN AN ALLUVIAL SETTING
-63-
It is determined through research and substantiated by piezometers
that the direction of ground-water flow 1s predominantly east northeast
(out of the page). This direction fluctuates seasonally, however, due to
the influence of the river. In the summer, flow is toward the east; in
the winter, it shifts to the northeast. The potentiometric surface in
the upper sand varies by approximately six feet during the year. Dense
phase immiscible wastes are known to be disposed of at the site.
The resultant horizontal and vertical placement of wells (and screen
lengths) reflects all of the waste management practices and hydrogeologic
factors at the site. The potential pathways for contaminant migration
are the two sand units. A greater number of wells are established in the
overlapping east-northeast flow zone, because ground-water flow there is
continuous and not seasonal. Wells are also placed in the area of
intermittent flow. Generally, the lengths of well screens installed at
the site reflect the vertical extent of the potential contaminant pathway
at the desired sampling location. However , shorter well screens (not
fully penetrating the depth of the sand unit) are employed in the thick
sand units where dilution effects may impair potential contaminant
detection. Several wells are screened at the sand/clay interfaces where
high specific gravity (dense) immiscibles may be expected to accumulate.
Also, those screens that intercept the potentiometric surface in the
upper sand are at least long enough to accommodate seasonal fluctuations
in ground-water elevations.
Figure 2-6 illustrates a cross-sectional and plan view of a waste
landfili situated in a mature Karst environment. This setting is charac-
teristic of carbonate environments encountered in various parts of the
country, but especially in the southeastern states. An assessment of the
geologic conditions at the site, through the use of borings. geophysical
surveys, aerial photography, tracer studies, and other geological
investigatory techniques, made it possible to identify a mature Karst
geologic formation characzerlzed by well-defined sinkholes, solution
-64-
OS’rlER-9950.1
LEGEND
0 EACUGROUND/UPGRAOlENT UELL
a DOWNGRADIENT MONtTORlNG WELL
1
WELL SCREEN
-P- POTENTIOMETRIC SURFACE
4-c FRACTURE0 TRACE
- - - OUTLINE OF CAVERN (PLAN VIEW)
PLUGGED EOREHOLE
WELL 2 WELL4 WELL6
(PROJECETD 100 FEET) IPROJECTED 155 FEET) IPROJECTED 60 FEET)
A
1.120
1,100
1,080
1.060
1,040
LANDFILL
I
\iPROJECfED 5 -_ 50 FEETI +
I
WELL 1 WELL 3 WELL 6
(FEET1 100 150 200 250 300 350 400 450 500 550 600
t
FOSSIL HASH
K = 4.5 I lo- ’ cm/we
FIGURE 2-6 MONITORING WELL PLACEMENT AND SCREEN LENGTHS IN A MATURE KARST TERRAIN/FRACTURED BEDROCK SETTING
-65-
channels, and extensive vertical and horizontal fracturing in an
interbedded limestone/dolomite. Using potentiometric data, ground-water
flow direction was found to be to the east. Solution channels are formed
by the flow of water through the fractures. The chemical reaction
between the carbonate rock and the ground water in the fractures produces
voids. These voids are referred to as solution channels. Through time,
these solution channels are enlarged to the point where the weight of the
overlaying rock (overburden) may be too great to provide support, thereby
causing a "roof" collapse and the formation of a sinkhole. The location
of these solution channels dictates the placement of detection monitoring
wells. Note in the plan view the placement of well No. 2 is offset
50 feet from the perimeter of the landfill. The horizontal placement of
well No. 2, although not Immediately adjacent to the landfill, is
necessary in order to monitor all potential contaminant pathways. The
discrete nature of these solution channels dictates that each potential
pathway be monitored.
The distance between the "floor" and "ceiling" (vertical extent)
(height) of the solution channels ranges from three to six feet directly
beneath the sinkhole to one foot under the landfill except for the
40-foot deep cavern. This limited vertical distance of the cavities
allows for a full screened interval in the solution channels. (Note the
change in orientation of solution channels due to the presence of the
shell hash layer.)
2.2 Placement of Upgradient (Background) Monitoring Wells
The downgradient wells must be designed and installed to immediately
detect releases of hazardous waste or hazardous waste constituents to the
uppermost aquifer. The upgradient wells must be located and constructed
to provide representative samples of ground water in the same portion of
the aquifer monitored by the downgradient wells to permit a comparison of
ground-water quality (40 CFR 265, Subpart F, 265.92(a)(1)).
-66-
OSWER-9950.1
There are at least three main questions that the technical reviewer
should ask when reviewing the decisions the owner/operator has made
regarding the placement of the background monitoring wells:
• Are the background wells far enough away from waste management
areas to prevent contamination from the hazardous waste
management units?
• Are enough wells installed and screened at appropriate depths to
adequately account for spatial variability in background water
quality?
• Are well clusters used at sampling locations to permit
comparisons of background ground-water data with downgradient
ground-water data obtained from the same hydrologic unit?
By regulation, the owner/operator must install as a minimum one
background well. However, a facility that uses only one well for
sampling background water quality may not be able to account for spatial
variability. It is, in fact, a very unusual circumstance in which only
one background well will fully characterize background ground-water
quality. The owner/operator who makes comparisons of background and
downgradient monitoring well results with data from only one background
well increases the risk of a false indication of contaminant release. In
most cases, the owner/operator should install multiple background
monitoring wells in the uppermost aquifer to account for spatial
variability in background water quality data.
The owner/operator should also install enough background monitoring
wells to allow for depth-discrete comparisons of water quality. This
means simply that for downgradient wells completed in a particular
geologic formation, the owner/operator should install upgradient well(s)
in the same portion of the aquifer, so that the data can be compared on a
depth-discrete basis (Figure 2-7).
Owner/operators should avoid installing background monitoring wells
that are screened over the entire thickness of the uppermost aquifer.
-67-
ELEVATION
NGVO
300’
UPGAAOIENT MONlTORlNG
BACKGROUND WELL
WELL CLUSTER
CLUSTER
1C 18 1A
,,.; I, :.:
.’
‘, : :.: “‘,,(,
. . . . ,,
GROUNPWATER FLOW I- ‘..’ ,’
‘. .:
., L:, ” :?.
,. ”
.‘,., , I’ ,‘,I
.:
\,‘.’
<).
I _ ‘,.:;. ,’ ,. ,: !
,’ : ,‘.. . . .’ ,...
. : .‘,,‘. .‘: , ;’
:,‘, ,_
‘.
: ; .‘, ‘.‘. ‘.I,.,. .:
-----
r----2-L~-~---~--T-~-~~ cuy c--_--_-_-- _- ---- --------------------------- ---_-_- _-.-.--___v-_---------
.--:--c---r.. _ _ - - I - - - - - - - - - - - - - - - - - - - - -
:------ _ - - --- - _ --_-e-----v- - - ----.-=;c= __-_-____--_--------I
c--L_--__-- ------ ----__---------
_--__-___--___e- ----------- ----a
) 1 1
loo’ 0 100
‘,I , ‘_’
.._. ‘. :._’
: ,, - : ::
:
.‘: ‘, :’
,‘, . -,
,_:
._
_;.
..-..
.- ..:._ .; “.‘.,
: --G-c *--&,.A-.
: ‘. .;_ . . .;_
. . :,_‘, ,.
_‘. .I ..; :. __
iAN K * 1.0 x 10-3cmhc
,a.. . ‘. _. . ‘. , ..‘.’ ,
.,.’
; . ,. .
,P..’ :
:“.. ; ;::,..
,.-.: ,.:
_ . ‘. :. .
SAN0 K - 7.0 x IU-%mhc
. _ _
..: . .- ,,,_ ““‘. ..1. ,
‘, . .
LEGEND
I! WELL AN0 SCREEN
I
- 2 -
POTENTIOMETRIC SUFIFACE
FOR UPPERMOST SAN0
7,’ POTENTIOMETRhZ SURFACE
-- FOR LOWER SAND
f IGURE 2-7 PLACEMENT OF BACKGROUND WELLS
da-
OSWER-9950.1
Screening the entire thickness of th e uppermost aquifer will not allow
the owner/operator to obtain depth-discrete water quality data. Instead,
the owner/operator should use shorter well screens in order to obtain
depth-discrete water quality data.
In order to establish background ground-water quality, it is
necessary to properly identify the ground-water flow direction and place
wells hydraulically upgradient to the waste management area. Usuaily,
this is accompiished by locating the background wells far enough from
waste management units to avoid contamination by the hazardous waste
management units. There are geologic and hydrologic situations for which
determination of the hydraulically upgradient location is often
difficult. These cases require further site-specific examination to
properly position or place background wells. Examples of such cases
include the following:
0 Waste management areas above ground-water mounds;
s Waste management areas located above aquifers in which
ground-water flow directions change seasonally:
l Waste management areas located close to a property boundary that
is in the upgradient direction;
l Waste facilities containing significant amotints of immiscible
contaminants with densities greater than or less than water:
0 Waste management facilities located in areas where nearby surface
water can influence ground-water levels (e.g., river floodplains);
. Waste management facilities located near intermittently or
continuously used production wells; and
. Waste management facilities located in Karst areas or faulted
areas where fault zones may modify flow.
-59-
REFERENCES
Electric Power Research Institute. November 1981. Groundwater Quality
Monitoring at Coal-fired Power Plants: Status and Review. Encon
Associates, Inc., Research Project 1457. CS-2126.
Geraghty and Miller, Inc. 1980. The Fundamentals of Ground-Water Quality
Protection, Seminar Handbook, Geraghty and Miller, Inc. American
Ecology Services, Inc.
MacKay, D.M., P.V. Roberts, and J.A. Cherry. 1985. Transport of Organic
Contaminants in Groundwater, Engineering Science and Technology,
Vol. 19. No. 5, pp. 284-392.
Scalf, M.R., et al. 1981. Manual of Ground-Water Quality Sampling
Procedures. National Technical Information Service PB-82-103-045.
Shepard, W.D. 1983. Practical Geohydrological Aspects of Groundwater
Contamination. 3rd National Symp. of Aquifer and Groundwater
Monitoring, pp. 365-372.
U.S. Environmental Protection Agency. August 1977. Procedures Manual
for Ground-Water Monitoring at Solid Waste Disposal Facilities,
EPA/530/SW-611.
U.S. Environmental Protection Agency. 1983. RCRA Draft Permit Writer's
Ground-Water Protection, 40 CFR Part 264, Subpart F. U.S. Environ-
mental Protection Agency Contract No. 68-01-6464.
U.S. Environmental Protection Agency. 1983. Ground-Water Monitoring
Guidance for Owners and Operators of Interim Status Facilities.
National Technical Information Service. PB83-209445.
-70-
OSWER-9950.1
CHAPTER THREE
MONITORING WELL DESIGN AND CONSTRUCTION
The purpose of this chapter is to examine important aspects of RCRA
monitoring well design and construction. Included in this chapter are
discussions on the following topics:
• Drilling methods for installing wells (Section 3.1);
• Monitoring well construction materials (Section 3.2);
• Design of well intakes (Section 3.3);
• Development of wells (Section 3.4);
• documentation of well construction activity (Section 3.5):
• Specialized well design (Section 3.6); and
• Replacement of existing wells (Section 3.7).
In order to better understand proper ground-water monitoring
procedure. a differentiation between monitoring wells and piezometer
wells should be made. Monitoring wells provide for the measurement of
total well depth, the collection of representative ground-water samples,
the detection of light- and dense-phase organics, and, under certain
circumstances, the collection of samples of light- and dense-phase
organics. Piezometer wells are used to determine static water level, in
addition to establishing horizontal and vertical ground-water flow
directions.
3.1 Drilling Methods
A variety of well-drilling methods can be used in the installation
of ground-water monitoring wells. It is important that the drilling
method or methods used minimize disturbance of subsurface materials and
not contaminate the subsurface and ground water (40 CFR 265.91(c)).
Table 3-1 lists the drilling methods that are most commonly used to
install wells. The selection of the actual drilling method is, of course,
-71-
TABLE 3-1
DRILLING METHODS FOR
VARIOUS TYPES OF GEOLOGIC SETTINGS
Geologic Environment
Drilling Methods
Air** Water/Mud Cable Hollow-Stem Solid-Stem
Rotary Rotary Tool Continuous Continuous
Auger Auger"
Glaciated or unconsolidated
materials less than 150 feet
deep
• • • • •
Glaciated or unconsolidated
materials more than 150 feet
deep
Consolidated rock formations
less than 500 feet deep (minimal
or no fractured formations)
Consolidated rock formations
less than 500 feet deep (highly
fractured formations)
Consolidated rock formations
more than 500 feet deep (minimal
formations)
Consolidated rock formations
more than 500 feet deep (highly
fractured formations)
•
•
•
•
•
•
•
•
•
•
• •
•
•
•
* Above potentiometric surface.
** Includes conventional and wireline core drilling.
NOTE:
Although several methods are suggested as appropriate for similar conditions, one method
may be more suitable than the others.
-72-
OSWER-9950.1
a function of site-specific geologic conditions. Table 3-1 provides an
interpretation of how geologic conditions may influence the choice of
drilling method. The following sections discuss each drilling method and
its applicability to the installation of RCRA monitoring wells. It is
important to note that regardless of the drilling method selected, the
owner/operator is responsible for the drilling equipment and for having it
decontaminated. This procedure should be followed before use and between
borehole locations to prevent cross contamination of weils where contamin-
ation has been detected or is suspected from the site characterization
work that precedes the well instaliation work. In addition to selecting
the proper drilling techniques, other precautions to prevent distribution
of any existing contaminants throughout a borehole should be taken.
3.1.1 Hollow-Stem Continuous-Flight Auger
The hollow-stem continuous-fLight auger is among the most frequently
employed tools used in drilling monitoring wells in unconsolidated
materials. The drill rigs used for this drilling method are usually
mobile, fast, and relatively inexpensive to operate. Drilling fluids
normally are not used. and disturbance to the aquifers of concern is
minimal. Auger drilling is usually limited to unconsolidated materials
and to depths of approximately 150 feet. In formations where the borehole
will not stand open, the well is constructed inside the hollow-stem auger
prior to the auger's removal from the ground. Hollow-stem augers with
inside diameters of six inches or six and one-quarter inches are readily
available for this purpose. Generally, the diameter of the well that can
be constructed with this type of drill rig is limited to four inches or
less, although firms now manufacture eight and one-quarter inch inside
diameter hollow-stem augers and are experimenting with ten and one-quarter
inch inside diameter hollow-stem augers. The differential between the
inner diameter of the auger and the outer diameter of the well casing
should ideally be at least three to five inches to permit effective
placement of filter pack and annular sealant.
-:3-
The use of hollow-stem auger drilling in heaving sand environments
also presents some difficulties. However, with care and the use of proper
drilling procedures, this difficulty can be overcome. For example, a
positive pressure head within the auger st2m can be developed by filling
the aug2r with clean water. The heaving sands are thus displaced when a
knock-out plug (which is part of the auger) is removed. If casing is
driven, the added outer diameter of the drive shoe must be considered in
the calculation of sealant and filter pack yrolume.
3 - .A.2 Solid-Stem Continuous-Flight Auqe_r -__---
The use of solid-stem continuous-flight auger drilling techniques
for monitoring well construction is limited to fine-grained unconsoli-
dated materials that vi11 maintain an open borehole or in consolidated
sediments. The method is similar to the hollow-stem continuous-flight
augers except that the augers must be removed from the ground to allow
insertion of the well casing and screen. This method is also limited to
a depth of approximately 150 feet. In areas characterized by less
competent sediments or soils (i.e., unstable, unable to retain the
sphericity of the borehole during drilling operations), solid-stem auger
drilling can be utilized to limited depths. Caving of the borehole,
howevler, is an imposing problem. Another limitation of the solid-stem
auger is its use be1o.J the potentiometric surface. Maintaining the
int2grity of the borehole in the saturated zone is also difficult at
times, especiaily in poorly consolidated sediments. Solid-stem auger
drilling is not used for in-place well construction, whereas hollow-stem
auger drilling is. Collection of soil or formation samples is
impractical , and ther2fore, accurate depicrion of sit2 stratigraphy is
difficult. Solid-stem augers ha.:2 very limited utility in the boring
program for site characterization.
3.1.3 Cable Tool .~-.-. .- -
Cable tool drilling 1s reLati,Jely siow but off2rs many advantages
for monitoring well construct:on in relati-Jely shallow consolidated
fo:miat:ons and unconsoiidated formations. The method allows for the
OSWER-3350 .l
collection of excellent formation samples and detection of even relatively
fine-grained permeable zones. The installation of a steel casing as
drilling progresses also provides an excellent temporary host for the
construction of a monitoring well once the desired depth is reached.
Small amounts of water must be added to the hole as drilling
progresses until the potentiometric surface is encountered. The
owner/operator should only use wat2r that cannot itself contaminate
formation water. A minim*um six-inch diameter drive pi22 should be used to
facilitate the placement of the well casing, screen, and gravel pack, and
a minimum five-foot long seal should be mad2 prior to beginning the
removal of the drive pipe. The drive pipe should 52 pulled while the
sealant is still fluid and capable of flowing outward to fill the annular
space vacated by the drive pipe and shoe. The drive pip2 also should be
pulled in sections and additional sealant added to ensure that a
satisfactory seal is obtained. Cable tool rigs have generally been
replaced by rotary rigs for water well construction in most areas of the
United States. Therefore, cable tool rigs may not be readily available in
many regions.
3.1.4 Air Rotary
Rotary drilling involves the use of circulating fluids, i.e., mud,
water, or air, to remove the drill cuttings and maintain an open hole as
drilling progresses. The different types of rotary drilling methods are
named according to the type of fluid and the direction of fluid flow.
Air rotary drilling forces air down the drill pipe and back up the bore
hole to remove the drill cuttings. The use of air rotary drilling
techniques is best suited for use in hard-rock formations. In soft
unconsolidated formations, casing is driven to keep the formations from
caving.
Air rotary drilling can be used without affecting the quality of
ground water from monitoring wells in hard rock formations with minimum
unconsolidated overburden. The successful construction of monitcring
-:5-
wells using this drilling technique hinges on the bore hole remaining
open after the air circulation ceases. It is an inappropriate method in
areas where the upper soil horizons are contaminated and sloughing of
sidewalls would likely result in contamination of the well. The air from
the compressor on the rig should be filtered to ensure that oil from the
compressor is not introduced into the ground-water system to be monitored.
Foam or joint compounds for the drill rods should not be used with air
rotary drilling because of the potential for introduction of contaminants
into the hydrogeologic environment. Caution should be taken in using air
rotary drilling techniques in highly polluted or hazardous environments.
Contaminated solids and water that are blown out of the hole are difficult
to contain and may adversely affect the drill crew and observers. When
air rotary is used, shrouds, canopies, bluooey lines, or directional
Dipes should be used to contain and direct the drill cuttings away from
the drill crew. Any contaminated materials (soil and/or water) should be
collected and disposed of in an approved waste disposal facility. On the
other hand, air rotary drilling techniques have actually improved safety
condi t ions.
? . 5 . - . - 7ater Rotary
.‘!a’i$?r- rotar.1 drilling involves the introduction of water into the
i-0 1 :-eiic 1 % through the drill pipe and subsequent circulation of water back
1:. 'i..- I:oIe to remove drill cputtings. Great care must be taken to ensure
t'* . ,.-. ';3t+r 2se2 in the dr illing process does not contain contaminants.
-+. ::Ls droller uses . . water rotary drilling to install wells, drilling
I..3 t e ! should be analyzed to ensure that it is contaminant-free.
:;ene:-ally, except when core drilling in hard rock units, the water
becomes middy after a few circulations.
There are problems associated with the use of water rotary drill-
ing . The recognition of water-bearing zones is hampered by the addition
of water :nto the system. Also, in poorly consolidated sediments, the
-76-
OSWER-9950.1
drillers may have a problem with caving of the borehole prior to instal-
lation of the screen and casing. In highly fractured terrains, it may
also be hard to maintain water circulation.
3.1.6 Mud Rotary
Mud rotary drilling techniques involve the use of various types of
drilling muds as the fluid that is introduced into the borehole. The mud
circulates back up the hole during drilling, carrying away drill cuttings
in the same manner as the air and water rotary drilling methods. Muds
provide the additional benefit of stabilizing the hole.
There are several types of muds available at present, primarily
bentonite, barium sulfate, organic polymers, cellulose polymers, and
polyacrylamides. The owner/operator should provide any chemical data
regarding potential impacts on water quality. While there are
hydrogeologic conditions under which mud rotary drilling is the best
option, the technical reviewer should make certain that the mud(s)
utilized do not affect the chemistry of ground-water samples, samples
from the borehole, or the operation of the well. The latter may
adversely affect the assessment of aquifer characteristics, for example:
• Bentonite muds reduce the effective perosity of the formation
around the well, thereby compromising estimates of well recovery.
Bentonite may also affect local ground-water pH. Additives to
modulate viscosity and density may also introduce contaminants to
the system or force large, irrecoverable quantities of mud into
the formation.
• Some organic polymers and compounds provide an environment for
bacterial growth which, in turn, reduces the reliability of
sampling results.
3.2 Monitoring Well Construction Materials
The technical reviewer must ensure that the owner/operator used well
construction materials that are durable enough to resist chemical and
physical degradation and do not interfere with the quality of ground-water
samples. Specific well components that are of concern include well
-77-
casings, well screens, filter packs, and annular seals or backfills.
Figure 3-1 is a drawing of a typical ground-water monitoring well. The
following sections describe various acceptable materials the owner/
operator should have used in constructing the well as depicted in
Figure 3-1.
3.2.1 Well Casings and Well Screen
A variety of construction materials have been used for the casings
and well screens, including virgin fluorocarbon resins (i.e., fluorinated
ethylene propylene (FEP), polytetrafluoroethylene (PTFE), Teflon*),
stainless steel (304, 316, or 2205), cast iron, galvanized steel,
polyvinyl chloride (PVC), polyethylene, epoxy biphenol, and polypropylene.
Many of these materials, however, may affect the quality of ground-water
samples and may not have the long-term structural characteristics required
of RCRA monitoring Wells. For example, steel casing deteriorates in
corrosive environments; PVC deteriorates when in contact with ketones,
esters, and aromatic hydrocarbons; polyethylene deteriorates in contact
with aromatic and halogenated hydrocarbons; and polypropylene deteriorates
in contact with oxidizing acids, aliphatic hydrocarbons, and aromatic
hydrocarbons. In addition, steel, PVC, polyethylene, and polypropylene
may adsorb and leach constituents that may affect the quality of
ground-water samples.
The selection of well casing and screen materials should have been
made with due consideration to geochemistry, anticipated lifetime of the
monitoring program, well depth, chemical parameters to be monitored and
other site-specific factors. Fluorocarbon resins or stainless steel
should be specified for use in the saturated zone when volatile organics
are to be determined, or may be tested, during a 30-year period. In such
cases, and where high corrosion potential exists or is anticipated,
fluorocarbon resins are preferable to stainless steel. An example of a
stainless steel monitoring well is provided in Figure 3-2. National
Sanitation Foundation (NSF) or ASTM-approved polyvinylchloride (PVC) well
casing and screens may be appropriate if only trace metals or nonvolatile
-78-
OSWER-9950.1
FIGURE 3-1. GENERAL MONITORING WELL - CROSS SECTION
-79-
- LOCKING WELL CAP
GAS VENT TUBE
“f
- STAINLESS STEEL WELL CASING
/
SURVEYOR’S PIN (FLUSH MOUNT1
CONCRETE WELL APRON
(MINIMUM RADIUS OF 3 FEET
AND WELL APRON (EXPANDING CEME
CEMENT AND SODIUM
BENTONITE MIXTURE
c- WELL DIAMETER = 4”
BOREHOLE DIAMETER = 10” TO 12”
INOMtNAL DIMENSION)
+ ANNULAR SEALANT
FILTER PACK (2 FEET OR
LESS ABOVE SCREEN)
POTENTIOMETRIC SURFACE
SCREENER INTERVAL
NT)
FIGURE 3-2 GENERAL STAINLESS STEEL MONITORING WELL - CROSS SECTION
-8@-
OSWER-9950.1
organics are the contaminants anticipated. As research demonstrates the
appropriateness of other materials for screens or casing in the saturated
or vadose zones, they may be utilized on a site-specific basis.
Stainless steel, fluorocarbon resins, or PVC are appropriate casing
materials in the unsaturated zone.
Figure 3-3 illustrates the concept of a composite well. Xany
combinations of materials may be employed in a manner consistent with
this guidance. One combination that should be avoided is the use of
dissimilar metals, such as stainless steel and galvanized steel, without
an electrically isolating (dielectric) bushing. If such dissimilar
metals are in direct contact in the soil, a potential difference is
created and leads to accelerated corrosion of the galvanized steel (in
this example). More generically, in the Galvanic series the less noble
metal becomes the anode to the more noble metal and is corroded at an
accelerated rate. In well construction, this acceleration in corrosion
at the point of connection will lead to failure of the construction
materials and loss of a RCRA monitoring well. Theoretically, a potential
difference is created in one type of metal penetrating heterogeneous
strata, but the difference in potentials would not be as great. In
conclusion, a dielectric coupling should be used for connecting
dissimilar metals in either the saturated or vadose zone.
There are two reasons why owners/operators should have selected
appropriate well screen and casing materials:
0 Long term structural integrity, i.e., 30 or more years, is
essential to the collection of unbiased ground-water samples over
the active life of the facility and post-closure period.
l Owner/operators of facilities whose Part B or post-closure per-
mit application has been called are required under 270,14(c)(4)
to analyze any plume(s) for Appendix VIII constituents (see the
RCRA Ground-Water Monitoring Compliance Order Guide, August
1985). The remainder of facilities must monitor for Appendix VII
constituents. Well construction materials should not bias the
collection and analysis of low concentrations of hazardous
constituents by reacting with the ground-water samples.
-81-
GROUND SURFACE
PVC OR OTHER NONINERT MATERIAL
ABOVE SATURATED ZONE
I I POTENTIOMETRIC SURFACE
>
INERT MATERIALS IN SATURATED
ZONE (CASING AND WELL SCREENJ
, CONFINING LAYER
FIGURE 3-3. COMPOSITE WELL CONSTRUCTION
(INERT CONSTRUCTION MATERIALS IN SATURATED ZONE)
-32-
OSWER-9950.1
Plastic pipe sections must be flush threaded or have the ability to
be connected by another mechanical method that does not introduce
contaminants such as glue or solvents into the well. Also, monitoring
wells must be structurally sound in order to withstand vigorous well
development procedures. Well casings and screens should be steam cleaned
prior to emplacement to ensure that all oils, greases, and waxes have been
removed. Because of the softness of casings and screens made of
fluorocarbon resins, these materials should be detergent-washed, not
steam-cleaned, prior to installation.
The owner/operator should normally use well casing with either a
two-inch or four-inch inside diameter. Larger casing diameters, however,
may be necessary where dedicated purging or sampling equipment is used or
where the well is screened in a deep formation.
The installation of a sump (sampling cup device) at the bottom of
a monitoring well (Figure 3-1) is recommended. The sump will aid in
collecting fine-grain sediments and result in prolonging the operating
life of the screen. An extra benefit of using a sump is its ability to
capture intermittent dense-phase contaminants for analysis, In zones
composed of fine-grained material (clays and silts) where turbidity may be
problematic, the decision flow chart (Figure 3-4) for turbid ground-water
samples should be consulted to evaluate well construction and development.
3.2.2 Monitoring Well Filter Pack and Annular Sealant
The materials used to construct the filter pack should be chemically
inert (e.g., clean quartz sand, silica, or glass beads), well rounded, and
dimensionally stable (see Section 3.3 for more detail on well intake
design). Fabric filters should not be used as filter pack materials.
Natural gravel packs are acceptable, provided that the owner/operator
conducts a sieve analysis to establish the appropriate well screen slot
size and determine chemical inertness of the filter pack materials in
anticipated environments.
-83-
TURBID GROUNDWATER
SAMPLE
1 .
ANALYZE THE SAMPLE
WITH A TURBIDMETER
NO YES
I
- SAMPLE IS ACCEPTABLE
REPURGE WELL (4.2.4)
REANALYZE WITH TURBIDMETER
ANALYZE SAMPLE USING
X - RAY DIFFRACTION
ANALYZE
FOR ORGANICS
NO
,
YES PRlMARlLY METALLiC COMPOUNOS;
I
RETAIN WELL NETWORK
4
SAMPLE IS ACCEPTABLE:
WELL NETWORK IS USEABLE
. . NO WELL HAS BEEN IMPROPERLY
7 CONSTRUCTED AND/OR DEVELOPED;
DO NOT USE SAMPLES
\
FIGURE 3-4 DECISION CHART FOR TURBID GROUND-WATER SAMPLES
-34-
OSWER-9950.1
The materials used to seal the annular space must prevent the
migration of contaminants to the sampling zone from the surface or
intermediate zones and prevent cross contamination between strata. The
materials should be chemically compatible with the anticipated waste to
ensure seal integrity during the life of the monitoring well and
chemicaliy inert SO they do not affect the quality of the ground-water
samples. The permeability of the sealants should be one to two orders of
magnitude less than the surrounding formation. Figure 3-l illustrates an
appropriate distribution of annular sealants. An example of an
appropriate use of annular sealant material is using a minimum of two
feet of certified sodium bentonite pellets immediately over the filter
pack when in a saturated zone. The pellets are most appropriate in a
saturated zone because they will penetrate the column of water to create
an effective seal. Coarse grit sodium bentonite is likely to hydrate and
bridge before reaching the filter pack. A cement and bentonite mixture,
bentonite chips, or antishrink cement mixtures should be used as the
annular sealant in the unsaturated zone above the certified-bentonite
pellet seal and below the frost line. Again, the appropriate clay must
be selected on the basis of the environment in which it is to be used.
In most cases, sodium bentonite is appropriate. The addition of
bentonite to the cement admixture should generally be in the amount of 2
to 5 percent by weight of cement content. This will aid in reducing
shrinkage and control time of setting. Calcium bentonite may be more
appropriate in calcic sediments/soils due to reduced cation exchange
potential. Clays should be pure, i.e., free of additives that may affect
ground-water quality. From below the frost line, the cap should be
composed of concrete blending into a four-inch thick apron extending
three feet or more from the outer edge of the borehole.
The untreated sodium bentonite seal should be placed around the
casing either by dropping it directly down the borehole or, if a hollow-
stem auger is used, putting the bentonite between the casing and the
inside of the auger stem. Both of these methods present a potential for
-85-
bridging. In shallow monitoring wells, a tamping device should be used
to reduce this potential. In deeper wells, it may be necessary to pour
a small amount of formation water down the casing to wash the bentonite
down the hole. In either case, a spacing differential of 3 to 5 inches
should exist between the outer diameter of the casing and the inner
diameter of the auger or the surface of the borehole to facilitate
emplacement of filter pack and annular sealants. Moreover, the precise
volume of filter pack and sealant required should be calculated to
establish their correct subsurface distribution. The actual volume of
materials used should be determined during well construction.
Discrepancies between calculated volumes and volumes used require
explanation.
The cement-bentonite mixture should be prepared using clean water
and placed in the borehole using a tremie pipe. The tremie method
ensures good sealing of the borehole from the bottom.
The remaining annular space should be sealed with expanding cement
to provide for security and an adequate surface seals. Locating the
interface between the. cement and bentonite-cement mixture below the frost
line serves to protect the well from damage due to frost heaving. The
cement should be placed in the borehole using the tremie method.
Upon completion of the well, installation of a suitable threaded or
flanged cap or compression seal should be placed or locked in properly to
prevent either tampering with the well or the entrance of foreign
material into it (Figure 3-2). A one-quarter inch vent hole pipe
provides an avenue for the escape of gas. Placement of concrete or steel
bumper guards around the well will prevent external damage by a vehicular
collision with the exposed casing.
3.3 Well Intake Design
The owner/operator should have designed and constructed the intake
of the monitoring wells to (1) allow sufficient ground-water flow to the
well for sampling; (2) minimize the passage of formation materials
-86-
OSWER-9950.1
(turbidity) into the well; and (3) ensure sufficient structural integrity
to prevent the collapse of the intake structure.
For wells completed in unconsolidated materials, the intake of a
monitoring well should consist of a screen or slotted casing with
openings sized to ensure that formational material is prohibited from
passing through the well during development. Extraneous fine-grained
material (clays and silts) that has been dislodged during drilling may be
left on the screen and the water in the well. These fines should be
removed from the screen and filter pack during development of the well.
The owner/operator should use commercially manufactured screens or
slotted casings. Field slotting of screens should not be allowed.
The annular space between the face of the formation and the screen
or slotted casing should be filled to minimize passage of formation
materials into the well. The driller should therefore install a filter
pack in each monitoring well that is constructed on site. Furthermore, in
order to ensure discrete sample horizons, the filter pack should extend
no more than two feet above the well screen as illustrated in Figure 3-1.
3.4 Well Development
After the owner/operator completed constructing monitoring wells,
natural hydraulic conductivity of the formation should have been restored
and all foreign sediment removed to ensure turbid-free ground-water
samples.
A variety of techniques are available for developing a well. To be
effective, they require reversals or surges in flow to avoid bridging by
particles, which is common when flow is continuous in one direction.
These reversals or surges can be created by using surge blocks, bailers,
or pumps. Formation water should be used for surging the well. In low-
yielding water-bearing formations, an outside source of water may
sometimes be introduced into the well to facilitate development. In
-87-
these cases, this water should be chemically analyzed to evaluate its
potential impact on in-situ water quality. The driller should not have
used air to develop the wells. All developing equipment should have been
decontaminated prior to use as should have the materials of construction.
The owner/operator should have developed wells to be clay- and
silt-free. If, after development of the well is complete, it continues
to yield turbid ground-water samples, the owner/operator should follow
the procedure described in Figure 3-4. The recommended acceptance/
rejection value of five nephelometric turbidity units (N.T.U.) is based
on the need to minimize biochemical activity and possible interference
with ground-water sample quality. The same criteria applies to turbidity
measurements expressed in other units such as the formazin turbidity unit
(F.T.U.) or Jackson turbidity unit (J.T.U.).
One should determine the relative hydraulic conductivity of
different layers within the aquifer in which the screen is placed (the
transmissivity/pumping test method is recommended). Using this
information along with pH, temperature measurements and mean seasonal
flow rates, one should evaluate the initial performance of the well and
use these values for periodic redevelopment and maintenance assessments.
3.5 Documentation of Well Design and Construction
In the context of a compliance order, the technical reviewer should
require the owner/operator to compile information on the design and
construction of wells. Such information may include:
• Date/time of construction
• Drilling method and drilling fluid used
• Well location (± 0.5 ft.)
• Bore hole diameter and well casing diameter
• Well depth (± 0.1 ft.)
• Drilling and lithologic logs
• Casing materials
-88-
OSWER-9950.1
• Screen materials and design
• Casing and screen joint type
• Screen slot size/length
• Filter pack material/size, grain analysis (D10)
• Filter pack volume calculations
• Filter pack placement method
• Sealant materials (percent bentonite)
• Sealant volume (lbs/gallon of cement)
• Sealant placement method
• Surface seal design/construction
• Well development procedure
• Type of protective well cap
• Ground surface elevation (± 0.01 ft.)
• Surveyor's pin elevation (± 0.01 ft.) on concrete apron
• Top of monitoring well casing elevation (± 0.01 ft.)
• Top of protective steel casing elevation (± 0.01 ft.)
• Detailed drawing of well (include dimensions)
3.6 Specialized Well Designs
There are two cases where owners/operators should use special
monitoring well designs:
• Where the owner/operator has chosen to use dedicated pumps to
draw ground-water samples; and
• Where light and/or dense-phase immiscibles may be present.
If the owner/operator elected to use a dedicated system, it should
be a fluorocarbon resin or stainless steel bailer, or a dedicated positive
gas displacement bladder pump composed of the same two materials. As
other sampling devices that can perform at least equivalently become
available, they may be employed as well.
The introduction of this pump, however, necessitates certain changes
in the well cross section depicted in Figure 3-1. Figure 3-5 represents
-89-
FIGURE 3-5 MONITORING WELL CROSS-SECTION - DEDICATED POSITIVE GAS
DISPLACEMENT BLADDER PUMP SYSTEM.
-90-
GSWER-9950.1
an appropriate cross section of a well that uses a dedicated positive gas
displacement bladder pump as the sampling device/well evacuation device.
The principal change is the addition of a two-inch diameter pump with
fluorocarbon resin outlet tubing to the well. A four-inch interior
diameter outer well casing should easily accommodate this additional
equipment. However, should a larger pump (e.g., three inches in
diameter) be required because of greater well depth or yield, a larger
outer casing may prove necessary (six-inch inside diameter). The pump
should be positioned midway along the screened interval, and the top of
its outlet pipe should extend into the well cap as depicted in Figure 3-5.
If light and dense-phase immiscible layers are presumed to be
present, the owner/ operator must obtain discrete samples of them. The
well system should have been designed to allow sampling of both light and
dense phases by using a well screen that extends from above the
potentiometric surface to the lower confining layer. Where well clusters
are employed, one well in the cluster may be screened at horizons where
floaters are expected (e.g., potentiometric surface, Figure 3-5), another
at horizons where dense phases are expected (e.g., aquiferlaquiclude
interface, Figure 3-6), and others within other portions of the uppermost
aquifer.
A periodic check of the dedicated sampling system should be
exercised to prevent damage and maximize efficiency. This inspection
should include removal of samples for verification of proper function.
The design of the dedicated sampling system should also allow access for
regular testing of aquifer characteristics. It is also recommended that
the well be periodically resurveyed using the protective casing and apron
(constructed to specific dimensions, Figure 3-1) as points of reference.
An option that can be exercised in constructing a monitoring well (e.g.,
dedicated sampler) is the use of fine sand at the top of the filter pack
to reduce or minimize invasion.
-31-
PRESSURE INLET \ , SAMPLE OUTLET (SEE ENLARGEMENT)
LOCKING
WELL CAP
“P&Y “Cl” I I “DC --
l II I /SURVEYOR’S PIN (FLUSH MOUNT)
CONCRETE WELL APRON
(MINIMUM RADIUS OF 3 FEET
AND FOUR INCHES THIC
CONTINUOUS POUR CONCRETE CAP
WELL APRON (EXPANDING CEMEN
PIPE (FLUOROCARBON
RESIN TUBING)
- CEMENT AND SODIUM
BENTONITE MIXTURE
WELL DIAMETER = 4” - 6” [OR AS
REQUJRED BY PUMPING DEVICE1
BOREHOLE DIAMETER = 10” TO 12”
(NOMINAL DIMENSION)
POTENTIOMETRIC SURFACE
w ANNULAR SEALANT
3 - INCH PURGE PUMP
FILTER PACK (2 FEET OR LESS
ABOVE SCREEN1
SCREENED INTERVAL
DEDICATED POSITIVE GAS
DISPLACEMENT BLADDER PUMP
2 - INCH SAMPLE WITHDRAWAL PUMP
Kl
T)
FIGURE 3-6 MONITORING WELL CROSSSECTION - DEDICATED PURGE PUMP AND SAMPLE
WITHDRAWAL PUMP. WELL SCREENED IN A HIGH YIELDING AQUIFER.
-92-
OSWER-9950.1
3.7 Evaluation of Existing Wells
The technical reviewer must decide whether wells--as designed and
constructed--allow for the collection of representative ground-water
samples. There are two situations the technical reviewer may encounter:
(1) where existing wells produce consistently turbid samples, i.e.,
greater than 5 N.T.U. (F.T.U. or J.T.U. depending on the method used),
and (2) where the owner/operator can produce little or no documentation
on how the wells were designed and installed.
Wells with turbidity or lack of information on well design and con-
struction may prompt the technical reviewer to order the owner/operator
to replace monitoring wells. In other, less obvious, cases the technical
reviewer must use best judgment in deciding when to order an owner/operator
to replace wells. The technical reviewer must decide whether the owner/
operator's wells--as built--allow the sampler to collect representative
ground-water samples (40 CFR 265.91(a)). This may not be an easy judgment
to make. In cases where it is not clear whether the wells can produce
representative ground-water samples, the technical reviewer may consider
requiring the owner/operator to conduct a field demonstration. This
demonstration would involve the installation of new well(s) near existing
wells. The owner/operator would sample and analyze for the same set of
parameters in both wells. If parameter values are comparable, the
technical reviewer should assume the owner/operator's existing wells are
producing representative samples. The field demonstration for existing
and new wells will be extremely difficult to evaluate in practice.
Differences in construction may or may not manifest themselves during the
field test. The results may lead to false conclusions in view of the
normal variabilities inherent in water quality parameters or sampling
which may be attributed to differences between old and new wells.
Similarly, differences in well construction, development, etc., that can
never be duplicated may also result in negative or positive biases due to
-93-
causes other than well construction. When such situations arise, the
wells should be decommissioned, sealed, and replaced. Where the only
question is whether or not the well casing material is negatively
affecting the chemical quality of the ground-water samples, a side-by-side
comparison at selected wells should be undertaken using stair-less steel or
one of the fluorocarbon resins. If analysis results are comparable, then
it is likely that chemical bias is not a major issue at the time of the
test.
Once wells have been properly designed and constructed, an appro-
priate sampling and analysis plan must be developed and implemented.
These procedures are discussed in Chapter Four.
-94-
OSWER-9950.1
REFERENCES
Barcelona, M.T., J.P. Gibb, and R.A. Miller. August 1983. A Guide to
the Selection of Materials for Monitoring Well Construction and
Ground-Water Sampling. U.S. Environmental Protection Agency. EPA
600/52-84-024.
Campbell, M.D. and J.H. Lehr. 1973. Water Well Technology, McGraw-Hill
Book Company.
Clark, J. II., R.D. Mutch, Jr., and M.R. Brother. 1983. Design of Cost-
Effective Chemical Monitoring Program for Land Disposal Facilities.
3rd National Symp. of Aquifer and Groundwater Monitoring, pp. 201-204.
Koehring Company. Date Unknown. Well Drilling Manual, National Water
Well Association.
U.S. Department of Army/Air Force. 1965. Well Drilling Operation.
Reprinted by National Water Well Association (No. 48).
U.S. Environmental Protection Agency. 1983. RCRA Draft Permit Writer's
Ground-Water Protection, 40 CFR Part 264, Subpart F. U.S. Environ-
mental Protection Agency Contract No. 68-01-6464.
U.S. Environmental Protection Agency. 1977. Manual of Water Well
Construction Practices. EPA 570/9-75/001.
Code of Federal Regulations. Title 40. Part 265, Environmental
Protection Agency Interim Status Standards for Owners and Operators of
Hazardous Waste Facilities, Subpart F, Ground-Water Monitoring.
-95-
OSWER-9950.1
CHAPTER FOUR
SAMPLING AND ANALYSIS
Federal regulation 40 CFR Part 265, Subpart F, Section 265.92,
requires the owner/operator to prepare and implement a written
ground-water sampling and analysis (S&A) plan. This plan must include
procedures and techniques for sample collection, sample preservation and
shipment, analytical procedures, and chain-of-custody control. The plan
is an important document. It allows the technical reviewer to thoroughly
review how the owner/operator has structured the S&A program. Also,
comparison of the written plan to field activities will allow the
technical reviewer to ensure the owner/operator is, in fact, following
his plan while collecting and analyzing ground-water samples. The
purpose of this chapter is to describe important elements of written S&A
plans and to discuss the level of detail that owner/operators should
include in their plans.
EPA has observed a number of problems in the way in which owner/
operators prepare their S&A plans or implement their S&A programs. Some
of the more common problems are listed below.
• Owner/operators have not prepared S&A plans or do not keep plans
on site.
• Plans contain very little information or do not adequately
describe the S&A program that the owner/operator is employing at
his facility.
• Field sampling personnel are not following the written plan or
are not even aware that it exists.
• Improper well evacuation techniques are used.
• Sampling equipment is used that may alter chemical constituents
in ground water.
• Sampling techniques are used that may alter chemical composition
of samples, particularly in regard to stripping of volatile
organic compounds in samples.
-97-
• Facility personnel are not using field blanks, chemical
standards, and chemically spiked samples to identify changes in
sample quality after collection.
• Field personnel do not properly clean nondedicated sampling
equipment after use.
• Field personnel are placing sampling equipment (rope, bailer,
tubing) on the ground where it can become contaminated prior to
use.
• Field personnel do not document their field activities adequately
(e.g., keep sampling logs).
• Field personnel are not following proper chain-of-custody
procedures.
• Little attention is paid to data reporting errors or anomalies.
• QA/QC protocol is inadequate (field and/or laboratory).
This chapter describes important elements in S&A plans (Section 4.1).
and then discusses the level of detail the owner/operator should include
(Sections 4.2 through 4.6). Furthermore, this chapter describes important
aspects of evaluating the field implementation of S&A plans (Sections 4.2
through 4.6). Section 4.7 describes how technical reviewers may examine
ground-water data to identify problems in the way owner/operators
acquire, process, and evaluate data.
4.1 Elements of Sampling and Analysis Plans
The owner/operator's S&A plan should, at a minimum, address a number
of elements. Specifically, the S&A plan should include information on:
• Sample collection (Section 4.2);
• Sample preservation and handling (Section 4.3);
• Chain-of-custody control (Section 4.4);
• Analytical procedures (Section 4.5); and
• Field and laboratory quality assurance/quality control
(Section 4.6).
-98-
OSWER-9950.1
4.2 Sample Collection
4.2.1 Measurement of Static Water Level Elevation
The sampling and analysis plan should include provisions for
measurement of static water elevations in each well prior to each
sampling event. Collection of water elevation on a continuing basis is
important to determine if horizontal and vertical flow gradients have
changed since initial site characterization. A change in hydrologic
conditions may necessitate modification to the design of the owner/
operator's ground-water monitoring system. The S&A plan should specify
the device to be used for water level measurements, as well as the
procedure for measuring water levels.
The owner/operator's field measurements should include depth to
standing water and total depth of the well to the bottom of the intake
screen structure. This information is required to calculate the volume
of stagnant water in the well and provide a check on the integrity of the
well (e.g., identify siltation problems). The measurements should be
taken to 0.01 foot. Each well should have a permanent, easily identified
reference point from which its water level measurement is taken. The
reference points should be established by a licensed surveyor and
typically located and marked at the top of the well casing with locking
cap removed or on the apron, and, where applicable, the protective
casing. The references points should be established in relation to an
established National Geodetic Vertical Datum (NGVD). In remote areas, a
temporary benchmark should be established to facilitate resurveying. The
reference point should be established in relation to an established NGVD,
and the survey should also note the well location coordinates and the
coordinates of any temporary benchmarks. The device used to detect the
water level surface must be sufficiently sensitive so that a measurement
to ± 0.01 foot can be obtained reliably. A steel tape will usually
suffice; however, it is recommended that an electronic device (e.g.,
-99-
M-Scope) be used to measure depth to the surface of the ground water or
light phase immiscibles. Whenever nondedicated equipment is used,
procedures need to be instituted to ensure that the sample is not
contaminated. Equipment should be constructed of inert materials and
decontaminated prior to use at another well.
4.2.2 Detection of Immiscible Layers
The S&A plan should include provisions for detecting immiscible
contaminants (i.e., "floaters" and "sinkers") where they would not be
detected in an aqueous phase if the owner/operator manages wastes of this
type at his facility. "Floaters" are those relatively insoluble organic
liquids that are less dense than water and which spread across the
potentiometric surface. "Sinkers" are those relatively insoluble organic
liquids that are more dense than water and tend to migrate vertically
through the sand and gravel aquifers to the underlying confining layer.
The detection of these immiscible layers requires specialized equipment
that must be used before the well is evacuated for conventional
sampling. The S&A plan should specify the device to be used to detect
light phases and dense phases, as well as the procedures to be used for
detecting and sampling these contaminants.
Owner/operators should follow the procedures below for detecting the
presence of light and/or dense phase immiscible organic layers. These
procedures should be undertaken before the well is evacuated for
conventional sampling:
1. Remove the locking and protective caps.
2. Sample the air in the well head for organic vapors using either
a photoionization analyzer or an organic vapor analyzer, and
record measurements.
3. Determine the static liquid level using a manometer and record
the depth.
4. Lower an interface probe into the well to determine the
existence of any immiscible layer(s), light and/or dense.
-100-
OSWER-9950.1
The air above the well head should be monitored in order to determine
the potential for fire, explosion, and/or toxic effects on workers. This
test also serves as a first indication of the presence of light phase
immiscible organics. A manometer or acoustical sounder (for very shallow
wells) will provide an accurate reading of the depth to the surface of
the liquid in the well, but neither are capable of differentiating
between the potentiometric surface and the surface of an immiscible
layer. Nonetheless. it is very useful to determine that surface depth
first to guide the lowering of the interface probe. The interface probe
serves two related purposes. First, as it is lowered into the well, the
probe registers when it is exposed to an organic liquid and thus
identifies the presence of immiscible layers. Careful recording of the
depths of the air/floater and floater/water interfaces establishes a
measurement of the thickness of the light phase immiscible layer.
Secondly, after passing through the light phase immiscible layer, the
probe indicates the depth to the water level. The presence of floaters
precludes the exclusive use of sounders to make a determination of static
water level. Dense phase immiscible layers are detected by lowering the
device to the bottom of the well where, again, the interface probe
registers the presence of organic liquids.
The approach to collecting light phase immiscibles is dependent on
the depth to the surface of the floating layer and the thickness of that
layer. The immiscible phase must be collected prior to any purging
activities. If the thickness of this phase is 2 feet or greater, a
bottom valve bailer is the equipment of choice. The bailer should be
lowered slowly until contact is made with the surface of the immiscible
phase, and lowered to a depth less than that of the immiscible/water
interface depth as determined by preliminary measure with the interface
probe.
When the thickness of the floating layer is less than 2 feet, but
the depth to the Surface of the floating layer is less than 25 feet, a
peristaltic pump can be used to "vacuum" a sample.
-lOl-
When the thickness of the floating layer is less than 2 feet and the
depth to the surface of the floating layer is beyond the effective
"reach" of a peristaltic pump (greater than 25 feet), a bailer must be
modified to allow filling only from the top. Sampling personnel should
disassemble the bottom check valve of the bailer and insert a piece of
2-inch diameter fluorocarbon resin sheet between the ball and ball seat.
This will seal off the bottom valve. The ball from the top check valve
should be removed to aliow the sample to enter from the top. The
buoyancy that occurs when the bailer is lowered into the floater can be
overcome by placing a length of l-inch stainless steel pipe (304, 316,
2205) on the retrieval line above the bailer (this pipe may have to be
notched to allow sample entry if the pipe remains within the top of the
bailer). The device should be lowered carefully, measuring the depth to
the surface of the floating layer, until the top of the bailer is level
with the top of the floating layer. The bailer should be lowered an
additional one-half thickness of the floating layer and the sample
collected. This technique is the most effective method of collection if
the floating phase is only a few inches thick.
The best method for collecting dense phase immiscibles is to use a
double check valve bailer. The key to sample collection is controlled,
slow lowering (and raising) of the bailer to the bottom of the well. The
dense phase must be collected prior to any purging activities.
4.2.3 Well Evacuation
The water standing in a well prior to sampling may not be
representative of in-situ ground-water quality. Therefore, the
owner/operator should remove the standing water in the well and filter
pack so that formation water can replace the stagnant water. The
owner/operator's S&A plan should include detailed, step-by-step
procedures for evacuating wells. The equipment the owner/operator plans
to use to evacuate wells should also be described.
-102-
CSWER-9950.1
The owner/operator's evacuation procedure should ensure that all
stagnant water is replaced by fresh formation water upon completion of
the process. The owner/operator's approach should allow drawing the
water down from above the screen in the uppermost part of the water
column in high yield formations to ensure that fresh water from the
formation will move upward in the screen. In low-yield formations, water
should be purged so that it is removed from the bottom of the screened
interval.
The procedure the owner/operator should use for well evacuation
depends on the hydraulic yield characteristics of the well. When
evacuating low-yield wells (wells that are incapable of yielding three
casing volumes), the owner/operator should evacuate wells to dryness
once. As soon as the well recovers sufficiently, the first sample should
be tested for pH, temperature, and specific conductance. Samples should
then be collected and containerized in the order of the parameters'
volatilization sensitivity. The well should be retested for pH,
temperature, and specific conductance after sampling as a measure of
purging efficiency and as a check on the stability of the water samples
over time. Whenever full recovery exceeds two hours, the owner/operator
should extract the sample as soon as sufficient volume is available for a
sample for each parameter. At no time should an owner/operator pump a
well to dryness if the recharge rate causes the formation water to
vigorously cascade down the sides of the screen and cause an accelerated
loss of volatiles. The owner/operator should anticipate this problem and
purge three casing volumes from the well at a rate that does not cause
recharge water to be excessively agitated. For higher yielding wells,
the owner/operator should evacuate three casing volumes prior to sampling.
In order to minimize the introduction of contamination into the
well positive-gas-displacement, fluorocarbon resin bladder pumps are
recommended for purging wells. Fluorocarbon resin or stainless steel
bailers are also recommended purqinq equipment. Where these devices
-103-
cannot be used, peristaltic pumps, gas-lift pumps, centrifugal pumps, and
venturi pumps may be used. Some of these pumps cause volatilization and
produce high pressure differentials, which result in variability in the
analysis of pH, specific conductance, metals, and volatile organic
samples. They are, however, acceptable for purging the wells if
sufficient time is allowed to let the water stabilize prior to sampling.
When purging equipment must be reused, it should be decontaminated,
following the same procedures required for the sampling equipment. Clean
gloves should be worn by the sampling personnel. Measures should be
taken to prevent surface soils from coming in contact with the purging
equipment and lines, which in turn could introduce contaminants to the
well. Purged water should be collected and screened with photoionization
or organic vapor analyzers, pH, temperature, and conductivity meters. If
these parameters and facility background data suggest that the water is
hazardous. it should be drummed and disposed of properly.
4.2.4 Sample Withdrawal
The technique used to withdraw a ground-water sample from a well
should be selected based on a consideration of the parameters to be
analyzed in the sample. To ensure the ground-water sample is represen-
tative of the formation, it is important to minimize physically altering
or chemically contaminating the sample during the withdrawal process. In
order to minimize the possibility of sample contamination, the
owner/operator should:
l Use only fluorocarbon resin or stainless steel sampling devices,
and
l Use dedicated samplers for each well. (If a dedicated sampler is
not availabie for each well, the owner/operator should thoroughly
clean the sampler between sampling events, and should take blanks
and analyze them to ensure cross-contamination has not occurred.)
The S&A plan should specify the order in which samples are to be
collected. Samples should be collected and containerized in the order of
-134-
OSWER-9950.1
the volatilization sensitivity of the parameters. A preferred collection
order for some common ground-water parameters follows:
l
l
l
l
l
0
l
l
l
l
l
l
l
l
Volatile organics (VOA)
Purgeable organic carbon (POC)
Purgeable organic halogens (POX)
Total organic halogens (TOX)
Total organic carbon (TOC)
Extractable organics
Total metals
Dissolved metals
Phenols
Cyanide
Sulfate and chloride
Turbidity
Nitrate and ammonia
Radionuclides
Temperature, pH, and specific conductance measurements should be
made in the field before and after sample collection as a check on the
stability of the water sampled over time. The S&A plan should also
specify in detail the devices the owner/operator will use for sample
withdrawal. The plan should state that devices are either dedicated to
a specific well or are capable of being fully disassembled and cleaned
between sampling events. Procedures for cleaning the sampling equipment
should be included in the plan. Any special sampling procedures that the
owner/operator must use to obtain samples for a particular constituent
(e.g., TOX or TCC) should also be described in the plan.
Equipment and procedures that minimize sample agitation and
reduce/eliminate contact with the atmosphere during sample transfer must
be used. When used properly, the following are acceptable sampling
devices for all parameters:
-105-
l Gas-operated, fluorocarbon resin or stainless steel squeeze pump
(also referred to as a bladder pump with adjustable flow control);
l Bailer (fluorocarbon resin or stainless steel), provided it is
equipped with double check valves and bottom emptying device;
l Syringe bailer (stainless steel or fluorocarbon resin); and
l Single check valve fluorocarbon resin or stainless steel bailer.
Sampling equipment should be constructed of inert material. Equipment
with neoprene fittings, PVC bailers, tygon tubing, silicon rubber
bladders, neoprene impellers, polyethylene, and viton is not acceptable.
If the owner/operator is using bailers, an inert cable/chain (e.g.,
fluorocarbon resin-coated wire, single strand stainless steel wire)
should be used to raise and lower the bailer.
While in the field, the technical reviewer should observe the
owner/operator's sampling technique to ensure that the owner/operator
satisfies the following:
l Positive gas displacement bladder pumps should be operated in a
continuous manner so that they do not produce pulsating samples
that are aerated in the return tube or upon discharge.
l Check valves should be designed and inspected to assure that
fouling problems do not reduce delivery capabilities or result in
aeration of the sample.
l Sampling equipment (e.g., especially bailers) should never be
dropped into the well, because this will cause degassing of the
water upon impact.
l The contents should be transferred to a sample container in a way
that will minimize agitation and aeration.
l Clean sampling equipment should not be placed directly on the
ground or other contaminated surfaces prior to insertion into the
well.
When dedicated equipment is not used for sampling (or well
evacuation), the owner/operator's sampling plan should include procedures
-106-
OSWER-9950.1
for disassembly and cleaning of equipment aefore each use. If the
constituents of interest are inorganic, the equipment should be cleaned
with a nonphosphate detergent/soap mixture. The first rinse should be a
dilute (0.1 N) hydrochloric acid or nitric acid, followed by a rinse of
tap water and finally Type II reagent grade water. Dilute hydrochloric
acid is generally preferred to nitric acid when cleaning stainless steel
because nitric acid may oxidize stainless steel. When organics are the
constituents of concern, the owner/operator should wash equipment with a
nonphosphate detergent and rinse with tap water, distilled water,
acetone, and pesticide-quality hexane, in that order. The sampling
equipment should be thoroughly dried before use to ensure that the
residual cleaning agents (e.g., HCl) are not carried over to the sample.
The owner/operator should sample background wells first and then proceed
to downgradient wells.
When collecting samples where volatile constituents or gases are of
interest using a positive gas displacement bladder pump, pumping rates
should not exceed 100 milliliters/minute. Higher rates can increase the
loss of volatile constituents and can cause fluctuation in pH and pH-
sensitive analytes. Once the portions of the sample reserved for the
analysis of volatile components have been collected, the owner/operator
may use higher pumping rate, particularly if a large sample volume must
be collected. The sampling flow rate should not exceed the flow rate
used while purging.
4.2.5 In-Situ or Field Analyses .-
Several constituents of %he parameters being evaluated are
physically or chemically unstable and must be tested either in the
borehole using a probe (in-situ) or immediately after collection using a
field test k-t. Examples of unstable elements or properties include pH,
redox potential, chlorine, dissolved oxygen, and temperature. Although
specific conductivity (analogous to electrical resistance) of a substance
-io7-
is relatively stable, it is recommended that this characteristic be
determined in the field. Most conductivity instruments require
temperature compensation; therefore, the temperature of the samples
should be measured at the time conductivity is determined. If the
owner/operator uses probes (pH electrode, specific ion electrode,
thermistor) to measure any of the above properties, it is important that
this is done on water samples taken after well evacuation and after any
samples for chemical analysis have been collected, so that the potential
for probe(s) to contaminate a sample designated for laboratory analysis
is minimized. Monitoring probes should not be placed in shipping
containers containing ground-water samples for laboratory analysis.
The owner/operator should complete the calibration of any in-situ
monitoring equipment or field-test probes and kits at the beginning of
each use, according to the manufacturers' specifications and consistent
with Test Methods for Evaluating Solid Waste - Physical/Chemical Methods
(SW-846), 2nd Edition, 1982.
4.3 Sample Preservation and Handling
Many of the chemical constituents and physiochemical parameters that
are to be measured or evaluated in ground-water monitoring programs are
not chemically stable, and therefore sample preservation is required.
Test Methods for Evaluating Solid Waste - Physical/Chemical Methods
(SW-846) includes a discussion by analyte of the appropriate sample
preservation procedures. In addition, SW-846 specifies the sample
containers that the owner/operator should use for each constituent or
common set of parameters. The owner/operator should identify in the S&A
plan what preservation methods and sample containers will be employed.
Each sampling and analysis plan should also detail all procedures and
techniques for transferring the samples to either a field or off-site
laboratory.
-108-
OSWER-9950.1
Improper sample handling may alter the analytical results of the
sample. Samples should be transferred in the field from the sampling
equipment directly into the container that has been specifically prepared
for that analysis or set of compatible parameters. It is not an
acceptable practice for samples to be composited in a common container in
the field and then split in the laboratory, or poured first into a wide
mouth container and then transferred into smaller containers. The S&A
plan should specify how the samples for volatiles will be transferred
from the sample collection device to the sample container in order to
minimize loss through agitation/volatilization.
4.3.1 Sample Containers
The owner/operator's S&A plan should identify the type of sample
containers to be used to collect samples, as well as the procedures the
owner/operator will use to ensure that sample containers are free of
contaminants prior to use.
When metals are the analytes of interest, fluorocarbon resin or
polyethylene containers with polypropylene caps should be used. When
organics are the analytes of interest, glass bottles with fluorocarbon
resin-lined caps should be used. The plan should refer to the specific
analytical method (in SW-846) that designates an acceptable container.
Containers should be cleaned based on the analyte of interest. When
samples are to be analyzed for metals, the sample containers as well as
the laboratory glassware should be thoroughly washed with nonphosphate
detergent and tap water, and rinsed with (1:1) nitric acid, tap water,
(1:1) hydrochloric acid, tap water, and finally Type II water, in that
order.
Similarly, an EPA-approved procedure is available for cleaning
containers used to store samples for organics analysis. The sampling
container should be emptied of any residual materials, followed by
washing with a nonphosphate detergent in hot water. It should then be
-109-
rinsed with tap water, distilled water, acetone, and finally with
pesticide-quality hexane. Dirty or contaminated glassware does not form
a very thin sheet of water on its surface and may require treatment with
chromic acid and/or baking in a muffle furnace at 400°C for 15 to
30 minutes to ensure that the glass is clean. Chromic acid may be useful
to remove organic deposits from glassware; however, the analyst should be
cautioned that the glassware must be thoroughly rinsed with water to
remove the last traces of chromium. The use of chromic acid can cause a
contamination problem and must be avoided if chromium is an analyte of
interest.
Glassware should be sealed and stored in a clean environment
immediately after drying or cooling to prevent any accumulation of dust
or other contaminants. It should be stored capped with aluminum foil and
inverted.
The cleanliness of a batch of precleaned bottles should be verified
in the laboratory. The residue analysis should be available prior to
sampling in the field.
4.3.2 Sample Preservation
The owner/operator's S&A plan should identify sample preservation
methods that the owner/operator plans to use. Methods of sample
preservation are relatively limited and are generally intended to
(1) retard biological action, (2) retard hydrolysis, and (3) reduce
sorption effects. Preservation methods are generally limited to pH,
control, chemical addition, refrigeration, and protection from light.
The owner/operator should refer to the specific preservation method in
SW-846 that will be used for the constituent in the sample. A summary
list of appropriate sample container types and sample preservation
measures is presented in Table 4-1.
4.3.3 Special Handling Considerations
Samples requiring analysis for organics should not be filtered.
Samples should not be transferred from one container to another, because
-llO-
OSWER-9950.1
TABLE 4-1
SAMPLING AND PRESERVATION PROCEOURES FOR DETECTION MONITORINGa
-l--l__ __-------_- -._----._-___ -
Parameter
Recormnended
Containerb
-
Maximum Minimum Volume
Preservative Required for
Holding Time Analysis
.-~__- ---------___
Chloride
Iron
Manganese
Sodium
Phenols
Sulfate
Indicators of Ground-Water ContaminatmnC
PH T. P. G Field determined
Specific conductance T. P, G Field determined
TUC G, amber. T-lined coo1 40C,d
cape HCl to pH t2
TOX G, amber, T-lined Cool 4OC, add 1 ml of
septa or caps l.lM sodium sulfite
Ground-Water Oualitv Characteristlu
T. P. G
r. p
4oc
Field acidified
to pH t2 with HNO,
G
T, P, G
40C/H2S04 to pH t.?
Cool. 4oc
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
Fluoride
None 25 ml
None 100 ml
25 days 4 x 15 ml
7 days 4 x 1.5 ml
28 days
6 months
50 ml
200 ml
28 days
28 days
500 ml
50 ml
EPA Interim Drinkina Water Characteristics
T. f~
Dark Bottle
T. P
T. f's G
Total Metals
Field acidified to
pH t2 with HN05
Dissolved Metal2
1 . Field filtration
(0.45 micron)
2. Acidify to pH ~2
with HN03
Cool I dOC
?~C/Jl~SO4 to OH tz
1Lontinued)
6 months
5 months
28 days
:4 days
1,000 ml
l.GOO ml
300 ml
:.ooo ml
-:11-
TABLE 4-1 (Continued)
SAMPLING AND PRESERVATION PROCEDURES FOR DETECTION MONITORING
Parameter
Endrin
Lindane
Methoxychlor
Toxaphene
2.4 0
2.4,s TP Silvex
Recomnended
Container4
T. G
Preservative
Cool. 40c
Maximum Minimum Volume
Required for
Holding Time Analysis
7 days 2.000 ml
Radium
Gross Alpha
Gross Beta
P, G Field acidified to
pi-i t2 with HN01
Coliform bacteria PP. G (sterilized) Cool, 4OC
6 months 1 gallon
6 hours 200 ml
Other Ground-Water Characteristics of Interest
Cyanide P, G Cool, 4Y. NaOH to
pti >12. 0.6 g
ascorbic acidf
14 daysg 500 ml
Oil and Grease G only Cocl. 4OC H2S04 to
pH t2
28 days 100 ml
Semivolatile. T. G cool. 4QC 14 days 60 ml
nonvolatile organics
Volatiles i. T-lined Cool. 4oc 14 days 60 m1
aReferences: w Methoas for Evaluatlna Solid Waste - PhvsicaliChemical Methods. SW-846
(2nd editicn. 1982).
Methods for Chemical Analvsis of Water and Wastel, EPA-600/4-79-D20.
c Jtandard Methods for the Examination of Water md 'Wastewate r, '6th edition (198s).
bContainer Types:
P = Plastic (polyethylene)
G = Glass
T = Fluorocarbon resins (PTFE, Teflon", FEP. PFA, etc.)
PP = Polypropylene
(Continued)
-112-
OSWER-9950.1
TABLE 4-l (Continued)
SAMPLING AND PRESERVATION PROCEDURES FOR DETECTION MONITORING
Clased on the requirements for detection monitoring ($265.93). the owner/operator must
collect a sufficient volume of ground water to allow for the analysis of four separate
replicates.
dShtpping containers (cooling chest with ice or ice pack) should be certified as to the 4PC
temperature at time of sample placement into these containers. Preservation of samples
requires that the temperature of collected samples be adjusted to the 4PC imnediately after
collection. Shipping coolers must be at 4Y and maintained at 4Y upon placement of sample
and during shipment. Maximum-minimum thermometers are to be placed into the shipping chest
to record temperature history. Chain-of-custody forms will have Shipping/Receiving and
In-transit (max/min) temperature boxes for recording data and verification.
eOo not allow any head space in the container.
FUse ascorbic acid only in the presence of oxidizing agents.
gHaximum holding time is 24 hours when sulfide is present. Optionally, all samples may be
tested with lead acetate paper before the pH adjustment in order to determine if sulfide is
present. If sulfide is present, it can be removed by addition of cadmium nitrate powder
until a negative spot test is obtained. The sample is filtered and then NaOH is added to
ptl 12.
-113-
losses of organic material onto the walls of the container or aeration
may occur. Total organic halogens (TOX) and total organic carbon (TOG)
samples should be handled and analyzed as materials containing volatile
organics. No headspace should exist in the sample containers to minimize
the possibility of volatilization of organics. Field logs and laboratory
analysis reports should note the headspace in the sample container(s) at
the time of receipt by the laboratory, as well as at the time the sample
was first transferred to the sample container at the wellhead.
Metallic ions that migrate through the unsaturated (vadose) and
saturated zones and arrive at a ground-water monitoring well may be
present in the well. Particles (e.g., silt, clay), which may be present
in the well even after well evacuation procedures, may absorb or adsorb
various ionic species to effectively lower the dissolved metal content in
the well water. Ground-water samples on which metals analysis will be
conducted should be split into two portions. One portion should be
filtered through a 0.45-micron membrane filter, transferred to a bottle,
preserved with nitric acid to a pH less than 2 (Table 4-1), and analyzed
for dissolved metals. The remaining portion should be transferred to a
bottle, preserved with nitric acid, and analyzed for total metals. Any
difference in concentration between the total and dissolved fractions may
be attributed to the original metallic ion content of the particles and
any sorption of ions to the particles.
4.4 Chain-of-Clstody
The owner/operator must describe a chain-of-custody program in the
S&A plan. An adequate chain-of-custody program will allow for the
tracing of possession and handling of individual samples from the time of
field collection through laboratory analysis. An owner/operator's chain-
of-custody program should include:
• Sample labels, which prevent misidentification of samples:
• Sample seals to preserve the integrity of the sample from the
time it is collected until it is opened in the laboratory:
OSWER-9950.1
• Field logbook to record information about each sample collection
during the ground-water monitoring program:
• Chain-of-custody record to establish the documentation necessary
to trace sample possession from the time of collection to
analysis;
• Sample analysis request sheets, which serve as official
communication to the laboratory of the particular analysis(es)
required for each sample and provide further evidence that the
chain of custody is complete; and
• Laboratory logbook and analysis notebooks. which are maintained
at the laboratory and record all pertinent information about the
sample.
4.4.1 Sample Labels
To prevent misidentification of samples, the owner/operator should
affix legible labels to each sample container. The labels should be
sufficiently durable to remain legible even when wet and should contain
the following types of information:
• Sample identification number
• Name of collector
• Date and time, of collection
• Place of collection
• Parameter(s) requested (if space permits)
• Internal temperature of shipping container at time sample was
placed
• Internal temperature of shipping container upon opening at
laboratory
• Maximum and minimum temperature range that occurred during
shipment
4.4.2 Sample Seal
In cases where samples may leave the owner/operator's immediate
control, such as shipment to a laboratory by a common carrier (e.g., air
freight), a seal should be provided on the shipping container or
individual sample bottles to ensure that the samples have not been
disturbed during transportation.
-115-
4.4.3 Field Logbook
An owner/operator or the individual ,designated to perform ground-
water monitoring operations should keep an up-to-date field logbook that
documents the following:
l
l
l
l
l
l
l
l
l
l
l
l
l
Identification of well
Well depth
Static water level depth and measurement technique
Presence of immiscible layers and detection method
Well yield - high or low
Purge volume and pumping rate
Time well purged
Collection method for immiscible layers and sample identification
numbers
Well evacuation procedure/equipment
Sample withdrawal procedure/equipment
Date and time of collection
Well sampling sequence
Types of sample containers used and sample identification numbers
Preservative(s) used
Parameters requested for analysis
Field analysis data and method(s)
Sample distribution and transporter
Field observations on sampling event
Name of collector
Climatic conditions including air temperature
Internal temperature of field and shipping (refrigerated)
containers
4.4.4 Chain-of-Custodyecord -- --
To establish the docunentation necessary to trace sample possession
from time of collection, a chain-of-custody record should be filled out
and should accompany every sample. The record should contain the
following types of information:
Sample number
Signature of collector
Date and time of collection
Sample type (e.g., ground water, immiscible layer)
Identification of well
Number of containers
Parameters requested for analysis
Signature of person(s) involved in the chain of possession
Inclusive dates of possession
-116-
OSWER-9950.1
• Internal temperature of shipping (refrigerated) container (chest)
when samples were sealed into the shipping container
• Maximum temperature recorded during shipment
• Minimum temperature recorded during shipment
• Internal temperature of shipping (refrigerated) container upon
opening in the laboratory
4.4.5 Sample Analysis Request Sheet
This document should accompany the sample(s) on delivery to the
laboratory and clearly identify which sample containers have been
designated (e.g., use of preservatives) for each requested parameter.
The record should include the following types of information:
• Name of person receiving the sample
• Laboratory sample number (if different from field number)
• Date of sample receipt
• Analyses to be performed
• Internal temperature of shipping {refrigerated) container upon
opening in the laboratory
4.4.6 Laboratory Logbook
Once the sample has been received in the laboratory, the sample
custodian and/or laboratory personnel should clearly document the
processing steps that are applied to the sample. All sample preparation
techniques (e.g., extraction) and instrumental methods must be identified
in the logbook. Experimental conditions, such as the use of specific
reagents (e.g., solvents, acids), temperatures, reaction times, and
instrument settings, should be noted. The results of the analysis of all
quality control samples should be identified specific to each batch of
ground-water samples analyzed. The laboratory logbook should include the
time, date, and name of the person who performed each processing step.
4.5 Analytical Procedures
The S&A plan should describe in detail the analytical procedures
that will be used to determine the concentrations of constituents or
parameters of interest. These procedures should include suitable
analytical methods as well as proper quality assurance and quality
-117-
control protocols. The required precision, accuracy, detection limits,
and percent recovery (if applicable) specifications should be clearly
identified in the plan.
The S&A plan should identify one method that will be used for each
specific parameter or constituent. The plan should specify a method in
SW-846 or an EPA-approved method, and clearly indicate if there are going
to be any deviations from the stated method and the reasons for these
deviations.
Records of ground-water analyses should include the methods used,
extraction date, and date of actual analysis. Data from samples that are
not analyzed within recommended holding times should be considered
suspect. Any deviation from an EPA-approved method (SW-846) should be
adequately tested to ensure that the quality of the results meets the
performance specifications (e.g., detection limit. sensitivity,
precision, accuracy) of the reference method.
4.6 Field and Laboratory Quality Assurance/Quality Control
One of the fundamental responsibilities of the owner/operator is
the establishment of continuing programs to ensure the reliability and
validity of field and analytical laboratory data gathered as part of the
overall ground-water monitoring program.
The owner/operator's S&A plan must explicitly describe the QA/QC
program that will be used in the field and laboratory. Many owner/
operators use commercial laboratories to conduct analyses of ground-water
samples. In these cases, it is the owner/operator's responsibility to
ensure that the laboratory of choice is exercising a proper QA/QC
program. The QA/QC program described in the owner/operator's S&A plan
must be used by the laboratory analyzing samples for the owner/operator.
4.6.1 Field QA/Qc Program
The owner/operator's S&A plan should provide for the routine
collection and analysis of two types of QC blanks: trip blanks and
-118-
OSWER-9950.1
equipment blanks. Each time a group of bottles is prepared for use in
the field, one bottle of each type (e.g., glass, fluorocarbon resin,
polyethylene) should be selected from the batch and filled with deionized
water. The bottles filled with the blank should be transported to the
sampling location and returned to the laboratory in a manner identical to
the handling procedure used for the samples. These trip blanks should be
subjected to the same analysis as the ground water. Any contaminants
found in the trip blanks could be attributed to (1) interaction between
the sample and the container, (2) contaminated rinse water, or (3) a
handling procedure that alters the sample analysis results. The
concentration levels of any contaminants found in the trip blank should
not be used to correct the ground-water data. The contaminant levels
should be noted, and if the levels are within an order of magnitude when
compared to the field sample results, the owner/operator should resample
the ground water.
Various types of field blanks should be used to verify that the
sample collection and handling process has not affected the quality of
the samples. The owner/operator should prepare each of the following
field blanks and analyze them for all of the required monitoring
parameters:
Trip Blank - Fill one of each type of sample bottle with Type II
reagent grade water, transport to the site, handle like a sample.
and return to the laboratory for analysis. One trip blank per
sampling event is recommended.
Equipment Blank - To ensure that the nondedicated sampling device
has been effectively cleaned (in the laboratory or field), fill the
device with Type II reagent grade water or pump Type II reagent
grade water through the device, transfer to sample bottle(s), and
return to the laboratory for analysis. A minimum of one equipment
blank for each day that ground-water monitoring wells are sampled is
recommended.
The results of the analysis of the blanks should not be used to
correct the ground-water data. If contaminants are found in the blanks
the source of the contamination should be identified and corrective
action, including resampling, should be initiated.
All field equipment that the owner/operator will use should be
calibrated prior to field use and recalibrated in the field before
measuring each sample. The owner/operator's S&A plan should describe a
program for ensuring proper calibration of field equipment. Other QA/Qc
practices such as sampling equipment decontamination procedures and
chain-of-custody procedures should also be described in the
owner/operator's S&A plan.
4.6.2 Laboratory QA/QC Program
The owner/operator's S&A plan should provide for the use of
standards. laboratory blanks, duplicates, and spiked samples for
calibration and identification of potential matrix interferences. The
owner/operator should use adequate statistical procedures (e.g., QC
charts) to monitor and document performance and implement an effective
program to resolve testing problems (e.g., instrument maintenance,
operator training). Data from QC samples (e.g., blanks, spiked samples)
should be used as a measure of performance or as an indicator of
potential sources of cross-contamination, but should not be used to alter
or correct analytical data. These data should be submitted to the Agency
with the ground-water monitoring sample results.
4.7 Evaluation of the Quality of Ground-Water Data
A ground-water sampling and analysis program produces a variety of
hydrogeological, geophysical, and ground-water chemical constituent
(GWCC) data. This section pertains primarily to the evaluation of GWCC
data because these data are specifically required by the regulations, are
evaluated in the statistical tests, provide the fundamental evidence used
to determine whether the facility is contaminating the ground water, and
are used to determine the extent of plume migration during assessment
monitoring. Also, details regarding how to obtain and identify quality
-120-
OSWER-9950.1
hydrogeological and geophysical data have been discussed earlier. The
GWCC data may be initially presented by the laboratory (by electronic
transmittal or) on reporting sheets: these data then must be compiled and
analyzed by the owner/operator prior to submission to the state or EPA in
order to evaluate the degree of ground-water contamination.
It is essential for owner/operators to make sure that, during
chemical analysis, laboratory reporting, computer automation, and report
preparation, data are generated and processed to avoid mistakes, and that
data are complete and fully documented. Data must be reported correctly
to have accurate analyses and valid results. If data errors do occur,
statistical analyses cannot discover, correct, or ameliorate these errors.
The following discussion considers aspects of data quality that may
indicate to the technical reviewer that the data acquisition, processing,
and evaluation were executed poorly or incorrectly.
The specific areas that are addressed include:
• Reporting of low and zero concentration values:
• Missing data values:
• Outliers; and
• Units of measure.
4.7.1 Reporting of Low and Zero Concentration Values
A critical concern is the interpretation, reporting, and analysis of
GWCCs that are measured at less than (LT) a limit of detection. LT limit
of detection values presently result from a variety of laboratory
conventions and protocols. Technical reviewers, during the review of
data submissions, may confront a variety of codes indicating that GWCC
concentrations are below a value which the laboratory designates as the
detection limit.
Values that are LT a limit of detection can result when:
• GWCCs are present at low concentrations;
• An insensitive analytical technique has been used; and
• The chemical matrix of the ground water interferes with the
analytical technique.
-121-
The following guidelines should help the technical reviewer identify
problems associated with the reporting of LT detection limit values,
analyze the data sets that contain LT detection limit values, and
prescribe remedies for future owner/operator submissions.
GiiCC should be given close attention if the LT detection limit
values appear to increase over time. Increasing detection limits may be
used to conceal an increasing concentration trend. Similarly, if back-
ground data are reported without a LT designation at low concentrations
and comparison downgradient data are presented at higher concentrations
with a LT designation, then it is possible that LT detection limit values
are being used to conceal larger downgradient concentrations. It is
unacceptable to report only qualitative information for values that were
measured below a limit of detection. The technical reviewer must ensure
that numerical values accompany the LT designation, so that data are
available for analysis. LT detection limit values that are high or that
vary should be reduced in future work by laboratory procedures that
remove or control interfering constituents.
The owner/operator must explain and follow a specific laboratory
protocol for determining and reporting low concentration values.
Technical reviewers should not allow the use of highly variable reporting
formats. An appropriate protocol for determining and reporting GKC data
at low concentrations is described in Appendix B of 40 CFR 5136, titled
"Definition and Procedure for the Determination of the Method Detection
Limit - Revision 1.11." Other methods are offered by the American
Chemical Society and the International Union of Pure and. Applied
Chemistry.
LT values should not be deleted from the analysis. Instead, when
data sets consist of a mixture of values that are LT a limit of detection
and actual concentration measurements, LT values may be analyzed at half .-____
their r-i-ted value. This technique is simple to use and has been
presented for use in the following references:
-122-
3SWER-9950.1
Gilbert, R.O. and Kinnison, R.R. 1981. Statistical Methods for
Estimating the Mean and Variance from Radionuclide Data Sets
Containing Negative, Unreported, or Less than Values. Health
Physics 40:377-390.
Nehls, G.J. and Akland G.G. 1973. Procedures for Handling
Aerometric Data. Journal of the Air Pollution Control Association
23:180-184.
LT values may also be analyzed using Cohen's Method This method is also ---- ----'
simple to use and has been described by:
Cohen C. 1961. Tables for Maximum Likelihood Estimates from Singly
Truncated and Singly Censored Samples. Technometrics 3:535-541.
Finally, a variety of other techniques, which are slightly more
complicated, are described in the following references:
Gilliom, R.J. and Helsel, D.R. 1986. Estimation of Distributional
Parameters for Censored Trace Level Water Quality Data. 1. Esti-
mation Techniques. Water Resources Research 22:135-146.
Helsel, D.R. and Gilliom, R.J. 1986. Estimation of Distributional
Parameters for Censored Trace Level Water Quality Data. 2. Verifi-
cation and Applications. Water Resources Research 22:147-155.
In some cases, the technical reviewer will be confronted with a
situation where a the italues for a chemical constituent in the back-
ground well system are LT a limit of detection. In this case, no data
are available to estimate the background variance, and the background
mean will be biased higher than its actual value, which is some value iT
the limit of detection. In this case, the technical reviewer should
ensure that laboratory protocols and data which are used to establish the
detection limit values are provided. In addition, it is recommended
that, especially in this case, the laboratory should ensure that any
values, which are reported above a limit of detection, are quantifiable.
The American Chemical Society's LCQ or the upper confidence limit of
EPA'S MDL may be used to establish a threshold criteria.
4.7.2 'Missing Data Values __--
Qwner/operators incur a substantial risk of missing an extreme
environmental event and measurement of particularly large or small values
if they fail to coilect all of the data required for a monitoring program.
This may result in an incomplete measure of environmental variability and
an increased likelihood of falsely detecting contamination. Also, if
assessment monitoring data are missing, there is a danger that the full
extent of contamination may not be characterized. Owner/operators must
take extreme care to ensure that concentration measurements result from
all samples taken. Nevertheless, the technical reviewer is likely to
confront situations where complete detection monitoring data have not
been collected. The technical reviewer should have the owner/operator
perform the t-test despite incomplete data collection, provided that the
following criteria have been met:
l If there are data from one upgradient well and one downgradient
weil, statistical comparisons should still be made. If data
exist for three quarters at a well, statistical comparisons
should be made after applying the rule described in the next
bullet.
0 If only one quarter of data is missing, values should be assigned
for the missing quarter by averaging the values obtained during
the other three quarters.
l If there are missing replicate measurements from a sampling
event, then average the replicate(s) that are available for that
sampling event.
These guidelines have been described previously in the November 1983 EPA
memorandum on statistical analyses of indicator parameter data. The
intent of this methodology is to encourage use of the t-test, despite
prior noncompliance with the data collection requirements in the
regulations, so that a determination can be made as to whether assessment
monitoring should begin. Regardless of whether there are sufficient data
for performing the t-test, the technical reviewer should consider taking
enforcement action to compel additional sampling on an accelerated
schedule. The technical reviewer must minimize delays in the evaluation
of a facility's ground water because of prior incomplete data collection.
-124-
OSWER-9950.1
4.7.3 Outliers
A GWCC value that is much different from most other values in a data
set for the same GWCC can be referred to as an "outlier." The reasons
for outliers can be:
l A catastrophic unnatural occurrence such as a spill;
l Inconsistent sampling or analytical chemistry methodology;
l Errors in the transcription of data values or decimal points; and
l True but extreme GWCC concentration measurements.
The technical reviewer should attempt to have owner/operators
correct outlying values if the cause of the problem can be documented and
corrected by the owner/operator without delay. The data should be
corrected if outliers are caused by incorrect transcription and the
correct values can be obtained and documented from valid owner/operator
records. Also, if a catastrophic event or a problem in methodology
occurred that can be documented, then data values should be from
calculations with clear reference to this deletion at all relevant
stages. Documentation and validation of the cause of outliers must
accompany any attempt to correct or delete data values, because true but
extreme values must not be altered. The technical reviewer should not
accept the mere presence of an extreme value in data or the effect of an
extreme value on the statistical analysis as a valid reason for the
continuation of detection monitoring.
Ground-water contaminant concentrations when influenced by a
hazardous waste management facility do not necessarily vary gradually.
Instead, it is not uncommon for contamination (e.g., halogenated organic)
to be reflected in a series of data collected over time with the followirg
trend. Measurements remain below a limit of detection and then, in a
single or several sampling event(s), concentrations rise to measurable
levels and soon return to concentrations which are LT a limit of detection
-125-
in subsequent sampling Periods. In general, technical reviewers should
not accept the contention that contaminant concentrations measured in
wells immediately downgradient or in the vicinity of hazardous waste
management areas increase only gradually. Rapidly increasing and
decreasing concentrations can occur in ground waters subjected to con-
tamination: the high concentrations in these cases would be true extreme
values but not outliers.
4.7.4 Units of Measure
Associated with each GWCC! value is a unit of measure that must be
reported accurately. Mistakes in the reporting of the units of measure
can result in gross errors in the apparent concentrations of GWCCs. For
example, a lead value of 30.2 might have a unit of measure of parts per
billion (ppb). Alternatively, the same lead value of 30.2 might have
been incorrectly reported with a unit of measure in parts per million
(ppm). The reported value would transform to a concentration with the
units of measure in ppb as 30,200 ppb of lead or three orders of
magnitude larger than it was measured.
The following guidelines should help the technical reviewers
ensure that units of measure associated with data values are reported
consistently and unambiguously:
l The units of measure should accompany each chemical parameter
name. Laboratory data sheets that include a statement "values
are reported in ppm unless otherwise noted" should generally be
discouraged but at least reviewed in detail by the technical
reviewer. It is common to find errors in reporting the units of
measure on this type of data reporting sheet especially when
these reporting sheets have been prepared manually.
l The units of measure for a given chemical parameter must be
consistent throughout the report.
l Finally, reporting forms for detection monitoring, as specified
in the EPA November 1983 memorandum, and the data presentation
methods described in Chapter Five should help to reduce problems
associated with the reporting of units of measure.
-126-
OSWER-9950.1
REFERENCES
American Public Health Association, American Water Works Association,
Water Pollution Control Federation. 1985. Standard Methods for the
Examination of Water and Wastewater, 16th Edition.
Barcelona, M.J., J.A. Helfrich, and E.E. Garske. February 1985.
Sampling Tubing Effects on Groundwater Samples, Analytical
Chemistry, 57(2). pp. 460-464.
Clayton, C.A.. et al. 1985. Demonstration of a Technique for Estimating
Detection Limits with Specified Assurance Probability. Research
Triangle Institute, Research Triangle Park, North Carolina.
TRI/2757/05-OID, EPA Contract 68-01-6826. DRAFT.
Currie, L.A. (1968) Limits for Qualitative Detection and Quantitative
Determination, Analytical Chemistry. 40(3): 5860
Gibb, J.P., R.M. Schuller, and R.A. Griffin. 1981. Procedures for the
Collection of Representative Water Quality Data for Monitoring
Wells. Illinois State Water Survey. Cooperative Groundwater Report
7.
Gillham, R.W.. M.J.L. Robin, J.F. Barker, and J.A. Cherry. 1983.
Groundwater Monitoring and Sample Bias. Environmental Affairs
Department, American Petroleum Institute.
Scalf, M.R., et al. 1981. Manual of Ground-Water Quality Sampling
U.S.
U.S.
U.S.
U.S.
U.S.
Procedures. National Technical Information Service PB-82-103-045.
Environmental Protection Agency. August 1977. Procedures Manual
for Ground-Water Monitoring at Solid Waste Disposal Facilities.
EPA/530/SW-611.
Environmental Protection Agency. 1979. Handbook for Analytical
Quality Control in Water and Wastewater Laboratories. EPA 600/4-79/
019.
Environmental Protection Agency, 1983. Ground-Water Monitoring
Guidance for Owners and Operators of Interim Status Facilities.
National Technical Information Service. PB83-209445.
Environmental Protection Agency. March 1983. Methods for Chemical
Analysis of Water and Wastes. EPA-600/4-79/020.
Environmental Protection Agency. August 1983. Handbook for Sampling
and Sample Preservation of Water and Wastewater. EPA-600/4-82/329.
-127-
U.S. Environmental Protection Agency. April 1984. Test Methods for
Evaluating Solid Waste - Physical/Chemical Methods, Second Edition
(Revised), SW-846.
-128-
OSWER-9950.1
CHAPTER FIVE
STATISTICAL ANALYSIS OF DETECTION MONITORING DATA
Owner/operators of hazardous waste facilities must implement a
ground-water monitoring program capable of determining if a facility has
had an effect on the quality of the ground water. This determination is
based on the results of a statistical test. This chapter discusses the
data that should be collected to perform the statistical test while
facilities are operating under interim status detection monitoring, and
what actions should be taken based on the results of the statistical
test. A general description of a recommended statistical procedure is
described below. A more specific description, which includes the
computational details and an example, appears in Appendix B.
5.1 Methods for Presenting Detection Monitoring Data
Data reporting sheets such as those presented in the November 30,
1983, EPA memorandum titled "Guidance on Implementation of Subpart F
Requirements for Statistically Significant Increases in Indicator
Parameter Values" should be used when owner/operators present data as
required by § 265.94(a). The technical reviewer should make sure that
owner/operators are aware of and use standardized data reporting forms.
The technical reviewer should have in the file all of the ground-
water data that have been collected to date from the facility. An
explicit presentation of the statistical test methodology should also
be in the file for the facility.
5.2 Introductory Topics: Available t-Tests, Definition of Terms,
Components of Variability, Validity of the t-Test Assumptions,
False Positives Versus False Negatives, and the Transition to
Permitting
Several introductory topics pertaining to the statistical analysis
of detection monitoring data are discussed in this section. First, the
statistical tests that the owner/operator can use to analyze detection
-129-
monitoring data are examined. Then, definitions of the terms background,
upgradient, and downgradient are presented. The measurement of environ-
mental variability and its relationship to the number of upgradient wells,
analytical replicates, and the statistical test that should be used is
reviewed. In the next section, the t-test assumptions, including the
importance of independent and normally distributed data, are discussed
and methods for correcting nonconformance with the assumptions are
offered. Also, included is a discussion emphasizing the importance of
controlling and evaluating the false positive and false negative rates
associated with the statistical procedures. The final section describes
broad categories of alternative statistical procedures that may be
explored for future application during the permit.
5.2.1 Available t-Tests
The interim status regulations specify that a Student's t-test be
used to determine whether there has been a statistically significant
increase in any ground-water contamination indicator parameter (IP) in
any well. The § 265 regulations do not, however, require a specific
Student's t-test. The owner/operator has the latitude within the
regulations to choose a t-test that will accommodate the data collected.
One reason that interim status facilities frequently adopt the Cochran's
Approximation to the Behrens-Fisher (CABF) t-test is that the Part 264
permit regulations require the use of the CABF t-test, unless an
equivalent statistical test is accepted by the Regional Administrator.
Other more appropriate t-tests are available for owner/operators to use
in the analysis of their interim status detection monitoring data.
One alternative t-test, which has been recommended for use, is
referred to as the averaged replicate (AR) t-test. The AR t-test is a
preferred test for owner/operators to apply to their interim status
detection monitoring data because it helps to reduce statistically-caused
false positives. Although special situations demanding alternative
-130-
OSWER-9950.1
t-test procedures may arise, this document generally recommends the use
of the AR t-test for maintaining compliance with the statistical analysis
requirements of 40 CFR §265, Subpart F.
Other t-tests are available for use while facilities are operatinq
under interim status detection monitoring. T-tests designed to control
the false positive rate despite the installation of additional wells,
measurement of additional chemical parameters, and an increased sampling
frequency may be appropriate (Miller, 1981). An owner/operator choosing
to employ a t-test methodology that controls the false positive rate or
overall significance level must evaluate the procedure’s impact on the
false negative rate, that is, the failure to identify contamination when
it has occurred. The false negative problem should be the primary concern
of the technical reviewer. An alternative t-test may be appropriate
during the administration of enforcement cases when, as described below,
accelerated data collection requirements are imposed. In these cases,
background data from the upgradient wells and downgradient data may be
collected simultaneously, and a t-test that accommodates the data
structure resulting from this sort of sampling program may apply. The
owner/operator may perform the t-test of choice, but the results must be
presented and action taken based on the results of only one type of
t-test. The technical review team should acquire the professional
expertise needed to evaluate thoroughly the statistical methodology.
Regardless of the specific procedure, the t-test methodology should
be explicit and include:
l A clear, understandable explanation of the methodology;
0 Presentation of explicit example calculations:
l The inclusion and documentation of all the original data used in
the statistical analysis procedure;
l Literature reference citations documenting alternative t-test
procedures; and
-i31-
l A detailed explanation of how data were manipulated and evaluated
prior to the statistical analysis, including goodness-of-fit
testing, transformations, less than detection limit value
manipulations, and power evaluations.
Also, it should be noted that although owner/operators have latitude
with respect to the statistical test used, there is much less choice with
regard to the data collection requirements. Finally, no matter which
t-test is used, the comparisons that must be made cannot change. Thus
for example, regardless of the t-test used, the owner/operator must
collect a background data set and compare these data to the data from
each well individually each time they are sampled.
5.2.2 Definition of Terms -._____
Three terms used frequently in discussions regarding the interim
status detection monitoring statistical analysis are: background,
upgradient, and downgradient. The terms upgradient and downgradient
describe well locations (e.g., with respect to the ground-water
hydraulics) and performance (e.g., downgradient wells must be able to
immediately detect contamination). The terms upgradient and downgradient
also describe the data collected from those wells. References to
background data, unlike those to upgradient or downgradient data, which
are well specific, concern all data collected from all upgradient wells
during the period when background levels are being established.
Modification of the background data may be required during the life of
the facility; guidance related to the modification of background data is
presented in Section 5.4.1.
5.2.3 Components of Variabiu _---- -
The inclusion and exclusion of various components of variability in
background ground-water data have a substantial impact on the performance
of the statistical test. When a background sampling program includes
data from only one upgradient well, there is no component of spatial
-132-
variability in the background data. Moreover, when the four measurements
from each sample are included in the analysis, the background data set is
influenced heavily by analytical variability. The result of no spatial
contribution to variability and a large contribution by analytical
variability is a background data distribution that typically has iittle
variability. This results in a statistical evaluation procedure that
readily identifies small differences, because the background distribution
of concentration values, which has little variability, tends to be
distinct and not “overlap” with the downgradient distribution of
concentration values.
To alleviate this situation, the background data set should include
a component of spatial variability and not be heavily influenced by the
typically small component of analytical variability. Two recommendations
are provided to help with this problem.
l First, the owner/operator should install additional upgradient
wells to ensure measurement of spatial ;pariatlon in the ground
water ;n the upgradient area.
l Second, the AR t-test, when applied to the data from well systems
with multiple upgradient wells, can be used by owner/operators
to remove the excessive influence of the analytical replicate
variability.
5.2.4 Validity of the t-Test Assumptions --_--.
Frequently, technical reviewers are confronted with the argument
from owner/operators that the t-test is not an appr,opriate methodology
for use, because the collected data are not independent and normally
distributed. Technical reviewers may find that the following discussion
is useful for supporting the need to evaluate the distributional
properties of the background data.
First, the contention that the background data are not normally
distributed should be supported by a goodness-of--fit analysis. A
contention of non-normality without the su~,por:lnq anaL:,-5:s ::: nor. .‘ai:~.
-133-
Second, yoodness-of-fit tests generally require a data set with a
substantial number of values in order to have enough statistical power to
discriminate among distributional types. The background data sets from
interim status facilities are rarely large enough for reasonable
performance of a goodness-of-fit test. A graphical approach evaluating
the cumulative probabilities of the data in comparison with a standard
normal may be useful.
Third, the presence of LT detection limits does not in itself imply
that the data values do not follow a normal distribution. The censoring
of the data values (which is essentially what happens when chemical
concentrations are reported LT a limit of detection) below a level and
the shape of the distribution above the level are not necessarily
related. In short, a data set with LT detection limit values may or may
not have normal distribution properties above the detection limit.
Fourth, in the case where firm evidence indicating that values do
not follow a normal distribution, owner/operators can use mean and
variance estimates from other distributions such as the lognormal. The
validity of any procedure must be documented and validated as a
technically sound approach (see Appendix B for details).
Finally, other non-t-test statistical procedures (e.g., nonparametric),
which are less dependent on distributional assumptions, do not satisfy the
requirements for interim status detection monitoring. The "Transition to
Permitting" section of this chapter describes when alternative non-t-test
procedures may be useful.
5.2.5 False Positives Versus False Negatives
Technical reviewers are frequently called upon to respond to
contentions from owner/operators that the statistically significant
increase, suggested by the statistical tests, has not actually occurred.
This has been referred to as a false positive. There are several points
that should be considered when a technical reviewer confronts a false
-134-
OSWER-9950.1
positive claim. First, false positives are not necessarily the result of
the statistical procedure. Many other factors influence the false positive
rate. These include, for example, poor well construction, improperly
located wells, too few background wells, improper sampling techniques, and
imprecise or inaccurate laboratory analysis. Owner/operators should not
contend that the statistical test resulted in a false positive unless it
can be shown that all the other aspects of the ground-water monitoring
program have been implemented properly. Second, the resampling program is
intended to reduce the false positive rate caused by laboratory error
only. The owner/operator should not make false positive claims until the
immediate resampling is performed. Third, owner/operators have the
latitude within the interim status regulations to use a t-test methodology
designed to control the false positive rate for the entire facility.
Fourth, false positives are only statistical issues. If engineering
information, including construction methods, age of the unit, waste
composition, or geohydraulic properties, indicates that contamination is
occurring, then a false positive claim is probably unwarranted. Fifth,
a false positive claim must be supported by data substantiating the false
positive claim (see Chapter 6 for more details), Finally, and most
important, the technical reviewer must not consider a false positive claim
or the results of the statistical procedure unless the owner/operator has
evaluated the false negative rate associated with the statistical procedure
in the context of facility-specific data. False negatives, that is, a
failure to indicate statistically significant contamination when a release
has occurred, are of more concern than false positive rates. The false
negative rate is rarely evaluated by owner/operators, and is frequently
higher than the false positive rate for even larger, substantial amounts of
contamination.
5.2.6 The Transition to Permitting -
The 40 CFR §265 Subpart F interim status regulations only allow the
use of a t-test for evaluati:lq data. However, the 40 CFR §264 Subpart F
-135-
permit regulations provide greater latitude in designing a statistical
evaluation methodology by allowing the use of an alternative statistical
procedure. Although facilities must continue to perform t-test methods to
maintain compliance with interim status, it is also wise for owner/operators
to begin to explore, test, and compare methods that may be useful under the
permit requirements.
A large array of methods and associated data manipulation procedures
are available. These approaches may include: linear model, tolerance
interval, nonparametric, control chart, or stochastic process methods.
5.3 Statistical Analysis of the Background Data
As described above, owner/operators should have measured the back-
ground concentrations of ground-water parameters in upgradient wells
within one year of the effective date of the interim status Subpart F
regulations. The initial background concentrations of the Appendix III
parameters in § 265.92(b)(1), the ground-water quality parameters in
§ 265.92(b)(2), and the ground-water contamination (or indicator)
parameters in § 265.92(b)(3) should have been established by monitoring
upgradient wells quarterly for a year. Four replicate measurements
should have been established from each well during each sampling episode
for the indicator parameters.
The background mean and variance should have been determined using
all of the data obtained for the § 265.92(b)(3) parameters during the
first year of sampling from the wells that were upgradient of the
facility. These summary statistics, which describe the background
concentrations, form the basis against which all subsequent upgradient
and downgradient concentration measurements will be compared. The
methods used to estimate the background mean
for AR t-test are described in Appendix B.
It is important to recognize that, in many instances, owner/operators
did not obtain background data during the prescribed period of time in
-136-
OSWER-9950.1
properly located and constructed wells, or did not sample and perform
chemical analyses using appropriate methodologies. In these cases, the
data used to establish the background statistics may have to be obtained
under a program accommodating the site-specific circumstances. Recommen-
dations related to modifying the background data to correct a false
positive problem are described below. In the case of incomplete prior
data collection, the technical reviewer should determine, using the
criteria in the missing data section of Chapter Four, when comparisons
can be conducted, using the existing data. Although some data sets may
be limited, it may still be possible to perform the statistical
comparisons of background versus downgradient data which are described
below. If contamination is suggested by the results of a t-test and the
resampling, then the first determination under assessment monitoring may
be compelled, as discussed in Chapter Six.
5.4 Statistical Analysis of Detection Monitoring Data After the
First Year
Detection monitoring data collected after the first year should be
used in a comparison with the background data to determine if there is a
suggestion that contamination may have occurred. A t-test is used to
make this determination. If the mean concentration of any IF in any
downgradient well is larger by a statistically significant amount than
the background concentration. then contamination may have occurred.
(NOTE: In the case of pH, the t-test is conducted such that an increase
or decrease may be detected. Thus, in the case of pH, all future
references to significant statistical increases imply that a significant
statistical change is being evaluated.)
All of the upgradient and downgradient wells must be sampled after
the first year. The ground-water quality parameters in § 265.92(b)(2)
must be measured at least annually, but are not analyzed statistically.
The IPs in § 265.92(b)(3) must be measured at least semiannually using at
least four replicate measurements from each sample from each well in the
detection monitoring network.
-137-
5.4.1 Comparison of Background Data with Upgradient Data Collected
on Subsequent Sampling Events
There is a suggestion that IP concentrations in the upgradient
ground water may be increasing when the t-tests for an upgradient well,
compared with the background data as required by § 265.93(c)(1), show a
significant increase in the concentration of an IP. There are several
reasons why the statistical test may indicate that the upgradient
concentrations have increased. These include:
• Ground-water flow direction was determined incorrectly and
hazardous waste constituents are migrating into the upgradient
wells.
• Ground-water flow direction was determined correctly, but
hazardous waste constituents are moving in a direction that is
opposite the ground-water flow.
• Upgradient wells were located in a mound caused by the facility.
• An inconsistent methodology (e.g., well construction material,
sampling and analysis techniques) was used, resulting in
concentration differences that are unrelated to any change in
the concentration of IPs in the ground water.
• The t-test indicated a difference between the background data and
upgradient data when actually there was no difference.
The cause of the increase in upgradient concentrations will be
important to the technical reviewer if the owner/operator successfully
establishes during the first determination under assessment that no
contaminants have entered the ground water. Prior to reinstating the
detection monitoring program, the owner/operator may request that,
because of the increase in background concentrations identified through
the background versus upgradient comparisons, the historical data are
unrepresentative of background conditions and should be modified.
The following recommendations are presented to help the technical
reviewer decide whether and how the background data set can be corrected.
• The technical reviewer should require that the owner/operator
undertake the following actions prior to modification of the
-138-
OSWER-9950.1
background data. First, it must be explained exactly why the
background data set should be modified. These demonstrations
must be based upon data and considerations that are documented
thoroughly. The owner/operator must also indicate specifically
how the background data set will be modified. Finally, it should
be shown that change in the background data will not delay the
ground-water sampling and analysis program.
l One of the recommended methodologies involves both the use of
more than one year of background data and a set of only the most
recently acquired background data (i.e., a moving average).
These procedures for modifying the background data may be appro-
priate: however, the decision should be based on site-specific
hydrogeological and engineering circumstances. The method used
to modify the background data should never become a routine part
of the statistical analysis methodology (e.g., use of a "moving
window").
l Many data sets will be unusable because of unacceptable
analytical chemistry, hydrogeological considerations, or the
physical construction of the well system, as for example, when
wells have been located in an area affected by the facility.
Modification of the background data set may require installation
and sampling of a new well system. In this case, it may be
necessary to collect background data from upgradient wells on
an accelerated schedule concomitantly with downgradient data.
l The technical reviewer may find it useful and suggest the
routine analysis of specific chemical parameters in addition
to the interim status indicator parameters. This may help the
owner/operator prepare for the ground-water monitoring and
analysis program to be implemented when the facility obtains
a 5264 permit. These parameter-specific data would also be
available for discussions regarding any future false positive
contentions.
5.4.2 Comparison of Background Data with Downgradient Data
The facility may be affecting the ground water when the t-test for a
downgradient well shows a statistically significant increase relative to
the background data. The owner/operator must immediately resample and
collect multiple ground-water samples from those downgradient wells where
a significant increase in concentration was detected, as required by
§265.93(c)(2). The additional ground-water samples are to be split into
duplicates and analyzed. The resampling data are then evaluated using
-139-
the same t-test methodology. The results of this t-test are then used to
Jetermine whether the originally detected significant increase was a
result of a laboratory mistake or a consequence of ground-water contami-
nation. If the initial results are due to laboratory error and no
:;:::ficant increase has occurred, the detection program may continue.
If the additional analyses performed under §265.93(c)(2) confirm the
.s-q:lificant increase, the owner/operator's facility is in interim status
_. - 7b.7r;;zent monitoring and must, without exception, begin immediately to
fulfill the requirements of the first determination of assessment
mono toring. While contamination is not verified during detection
monitoring, such monitoring is used to learn whether contamination may be
occurring. The first determination of assessment monitoring should be
:hs phase of analysis in which the suggestion of contamination revealed
'by the statistical analysis is documented more fully. Ground-water
contamination cannot be evaluated satisfactorily with a continuation of
detection monitoring.
-i40-
OSWER-9950.1
REFERENCES
Chew, V. 1980. Testing Differences Among Means: Correct Interpretation
and Some Alternatives. Hortscience 15:467-470
Cochran, W.G. 1983. Planning and Analysis of Observational Studies.
John Wiley and Sons. New York, New York.
Dixon, W.J. and F.J. Massey. 1969. Introduction to Statistical Analysis,
Third Edition. MacGraw-Hill Book Company.
Hurlbert, S.H. 1984. Pseudoreplication and the Design of Ecological
Field Experiments. Ecological Monographs 54:187-211
JRB Associates. 1983. Evaluation of Statistical Procedures for Ground-
water Monitoring. EPA Contract No. 68-01-6000. Work Assignment
No. 11
Keith, S.J., L.G. Wilson, H.R. Fitch, and D.M. Esposito. 1983. Sources
of Spatial Temporal Variability in Ground-Water Quality Data and
Method of Control. Ground Water Monitoring Review. Spring: 21-32.
Miller, R.G. 1981. Simultaneous Statistical Inference. Springer-Verlag,
New York, New York.
Nelson, J.D. and R.C. Ward. 1981. Statistical Considerations and
Sampling Techniques for Ground-Water Quality Monitoring. Ground
Water 19:617-625.
Nightingale, H.I. and W.C. Bianchi. 1979. Influence of Well Water
Quality Variability on Sampling Decisions and Monitoring. Water
Resources Bulletin 15:1394-1407.
Pettyjohn, W.A. 1976. Monitoring Cyclic Fluctuations in Ground-Water
Quality. Ground Water 14:472-480.
Sgambat, J.P., and J.R. Stedinger. 1981. Confidence in Ground-Water
Monitoring. Ground Water Monitoring Review 1:62-69.
Skinner, J.H. 1983. Guidance on Implementation of Subpart F Requirements
for Statistically Significant Increases in Indicator Parameter
Values. EPA/OSWER Memorandum, November 30, 1983.
Snedecor, G.W., and W.G. Cochran. 1967. Statistical Methods. The Iowa
State University Press. Ames, Iowa.
-141-
OSWER-9950.1
CHAPTER SIX
ASSESSMENT MONITORING
Once contaminant leakage has been detected via detection monitoring
efforts, the owner/operator must undertake a more aggressive ground-water
program called assessment monitoring. Specifically, the owner/operator
must determine the vertical and horizontal concentration profiles of all
the hazardous waste constituents in the plume(s) escaping from waste
management areas. In addition, the owner/operator must establish the
rate and extent of contaminant migration. This information will be used
later by the permit writer (in addition to other information collected
through the permit application process) to evaluate the need for
corrective action at the facility. Alternatively, this information may
form the basis for issuing an enforcement order compelling corrective
action prior to issuance of a permit.
The Agency has observed a number of problems in the way owner/
operators have conducted their assessment monitoring programs. These
include:
• Many owner/operators lack satisfactory knowledge of site hydro-
geologic conditions. As a result they cannot make informed
decisions on how to carry out their assessment programs. The
owner/operator should have conducted a thorough site hydrogeo-
logic investigation prior to the installation of the detection
monitoring system.
• Some owner/operators fail to implement their assessment programs
quickly enough or they implement programs that will take too long
to provide information on the extent and migration of a plume.
• Some owner/operators do not support geophysical investigation
with a sufficient monitoring well network. Geophysical methods
are useful for indicating contamination and for interpolation of
contaminant concentrations between wells: however, well sampling
is required to provide conclusive data.
• Many owner/operators greatly underestimate the level of effort
the regulatory agency expects of them in characterizing plume
migration. In most cases, assessment monitoring is an intensive
-143-
effort that will require the owner/operator to install numerous
monitoring wells. When full plume characterization is not
achieved with the initial round of well installation, additional
wells will be required. The owner/operator must track and
characterize both the horizontal and vertical components of the
plume (i.e., a three-dimensional characterization).
This chapter describes the technical approaches and techniques the
Agency feels are minimally necessary for characterizing a plume of
contamination as required in Part 265 assessment monitoring.
6.1 Relationship of Assessment Monitoring to Ground-Water Responsi-
bilities Under the Permit Application Regulations (Part 270)
Interim status assessment monitoring is just one in a series of
activities that facilities must undertake to prepare adequate permit
applications. The Part 270 permit application regulations require
interim status facilities to describe in their permit application any
plume of contamination (in terms of Appendix VIII sampling) and, based on
the levels of contamination found, to develop engineering plans for the
appropriate Part 264 ground-water program: detection monitoring,
compliance monitoring, or corrective action. Once a facility's permit is
called, either operating or post-closure, the owner/operator's ground-
water obligations expand from assessment monitoring alone to also include
the monitoring and plan development responsibilities imposed by Part 270.
The requirements relevant to facilities subject only to Part 265
assessment monitoring differ from those subject to Part 265 AND Part 270
(by virtue of a permit call-in) in two important ways.
First, the Part 265 assessment program requires monitoring for
hazardous waste constituents (primarily Appendix VII), whereas Part 270
[§ 270.14(c)(4)] requires Appendix VIII monitoring (Note: Appendix VII
is a subset of Appendix VIII--see Section 3.3 of the Compliance Order
Guidance for a further elaboration of this point). Therefore, assessment
plans of facilities subject to permitting should be based on the broader
Appendix VIII monitoring requirements embodied in Part 279 (see
Section 6.7).
-144-
OSWER-9950.1
Second, Part 265 assessment monitoring applies only to facilities
that detected contamination through a significant increase (or pH
decrease) in Part 265 indicator parameters (i.e., those that were
formally triggered under the regulations). The requirement to look for
and describe any plume of contamination in terms of Appendix VIII
constituents (as a condition of the permit application process) applies
to facilities that detected contamination through Part 265 detection
monitoring, as well as to any facility whose Part 265 detection
monitoring system is inadequate to detect a plume, should it occur.
As noted in Chapter 1 of the Compliance Order Guidance (August
1985), facilities with inadequate Part 265 monitoring systems are
required to conduct the Appendix VIII sampling and assessment activities
required by Part 270 (and necessary to make reasoned decisions about what
Part 264 ground-water program to incorporate in the permit) simply
because they have avoided compliance with Part 265 detection monitoring
in the past. Furthermore, such facilities should not be allowed to start
the Part 265 detection sequence over again, thus postponing the time when
the facility will be compelled to sample for actual constituents in
ground water even if they did not formally "trigger" into Part 265
assessment. The RCRA Ground-Water Monitoring Compliance Order Guidance
explains in greater detail the legal and technical bases for advancing
facilities with inadequate Part 265 detection systems into the type of
assessment activities described in this chapter. While the language of
the chapter speaks in terms of Part 265 assessment activities, the
techniques discussed herein are equally applicable to facilities
conducting plume characterization activities as part of the permit
application process.
6.2 Contents of a Part 265 Assessment Monitoring Plan
Owner/operators conducting plume characterization activities as
part of Part 265 assessment monitoring are required to have a written
-145-
assessment monitoring plan. The plan serves as the blueprint for the
activities undertaken to characterize the rate and extent of contaminant
migration. Plans must contain sufficient detail to determine the nature
and extent of the plume. When evaluating facilities in assessment
monitoring, technical reviewers should focus both on (1) scrutinizing the
adequacy of the written assessment plan, and (2) reviewing the owner/
operator's implementation of the plan in the field.
There are a number of elements that owner/operators should include
in their assessment monitoring plans. The remaining sections of this
chapter are organized around the following elements of an adequate
assessment plan:
• Section 6.3 - narrative discussion of the hydrogeologic
conditions at the owner/operator's site; identification of
potential contaminant pathways;
• Section 6.4 - description of the owner/operator's detection
monitoring system;
• Section 6.5 - description of the approach the owner/operator will
use to make the first determination (false positives rationale);
• Section 6.6 - description of the investigatory approach the
owner/operator will use to fully characterize rate and extent of
contaminant migration; identification and discussion of
investigatory phases:
• Section 6.7 - discussion of number, location, and depth of wells
the owner/operator will initially install, as well as strategy
for installing more wells in subsequent investigatory phases;
• Section 6.8 - information on well design and construction;
• Section 6.9 - a description of the sampling and analytical
program the owner/operator will use to obtain and analyze
ground-water monitoring data:
• Section 6.10 - description of data collection and analysis
procedures the owner/operator plans to employ:
-146-
OSWER-9950.1
• Section 6.11 - a discussion of the procedures the owner/operator
will use to determine the rate of constituent migration in ground
water; and
• Section 6.12 - a schedule for the implementation of each phase of
the assessment program.
6.3 Description of Hydrogeologic Conditions
An owner/operator cannot conduct an adequate assessment monitoring
program without a thorough understanding of site hydrogeologic conditions,
Such an understanding, garnered through site characterization activities
(refer to Chapter One), allows the owner/operator to identify likely
contaminant pathways. Identification of these pathways allows the
owner/operator to focus efforts on tracking and characterizing plume
movement. It is important to note that the initial site characterization
carried out by the owner/operator should provide enough hydrogeologic
information to allow the owner/operator not only to design a detection
monitoring system, but also to plan and carry out an assessment monitoring
program.
The owner/operator's assessment plan should describe in detailed
narrative form what hydrogeologic conditions exist at the owner/operator's
site. The plan should describe the potential pathways of constituent
migration at the site, including depth to water in aquifer, aquifer
connections to surface water and/or deeper aquifers, flow rate and
direction, and any structures such as fractures and faults which could
affect migration. The owner/operator's plan should also describe how
hydrogeologic conditions have influenced the type of assessment effort
that will be used to characterize plume migration. This portion of the
owner/operator's assessment plan should recapitulate the hydrogeologic
investigatory program the owner/operator undertook prior to installing a
detection monitoring system (see Chapter One). It should describe the
investigatory approach used by the owner/operator to characterize subsur-
face geology and hydrology, the nature and extent of field investigatory
-147-
activities, and the results of the investigation, as well as provide an
explicit discussion on how those results have guided decisions the
owner/operator has made concerning the planning and implementation of the
assessment monitoring program. As part of the plan, the owner/operator
should append various supporting documentation such as those described in
Table 1-1.
6.4 Description of Detection Monitoring System
The owner/operator's assessment plan should describe the existing
detection monitoring system in place at the owner/operator's facility.
The primary concern is whether the existing well system is capable of
detecting contaminant leakage that may be escaping from the facility. If
the owner/operator's detection monitoring system is deficient, either in
design or operation, plumes may exist unnoticed. This portion of the
owner/operator's assessment plan should describe the physical layout of
the owner/operator's detection monitoring well system (e.g., horizontal
and vertical orientation of individual wells) and identify assumptions
used by the owner/operator in designing the detection monitoring system,
(particularly how hydrogeologic condition affected the decision making
process).
6.5 Description of Approach for Making First Determination -
False Positives
Chapter Five described requirements that owner/operators must meet
in terms of statistical analysis of detection monitoring data. Once the
owner/operator resamples and the statistical test again suggests that an
indicator parameter has increased in a downgradient well(s), the
owner/operator must implement an assessment monitoring program.
Figure 6-1 illustrates the sequence of events that occurs Immediately
before and after the shift to assessment monitoring. Of particular
interest are those situations where the owner/operator believes that
contamination may have been falsely indicated and thus describes in the
OSWER-9950.1
FIGURE 6-1 PROCEDURE FOR EVALUATING FALSE POSITIVE CLAIMS BY OWNER/OPERATORS
-149-
assessment plan a short-term Frogram to substantiate or disprove this
false positive claim (i.e., false positive investigation is focus of
first determination - §265.93(d)(5)). There are a number of facilities
for which the first determination is no longer relevant, e.g., facilities
under 3008(h) enforcement action. See the RCRA Ground-Water Monitoring
Compliance Order Guide for details.
TWhen an owner/operator's detection monitoring system is properly
designed, the first determination under assessment monitoring may focus
on substantiating a false positive claim. If an owner/operator's
detection monitoring system is inadequate, it is difficult to evaluate
whether leakage has occurred. Substantiation of a false positive claim
would be a lengthy process, potentially involving hydrogeologic work, the
installation of a new detection well network, and evaluation of various
additional sampling data. In those cases, officials should reject a
false positive analysis as the focus of the first determination when the
existing system is inadequate, and instead require the owner/operator to
(1) correct deficiencies in the detection monitoring system; and
(2) initiate a program that will consider specific constituents of
concern in the existing wells, and in the new wells as they are installed.
If, however, an owner/operator's detection monitoring system is
adequately designed, the owner/operator may propose, as the first
determination, a short-term sampling program--generally no longer than
30 days--and an analysis of other related data that will permit
investigation of whether the statistical change noted in Part 265
indicator parameters truly represents migration of leachate into the
uppermost aquifer. Such short-term sampling programs, however, do not
allow for the evaluation of seasonal variation. Data gathered over the
short term, therefore, should be analyzed to control for the season in
which the data were collected, in order to establish comparability
-150-
OSWER-9950.1
with previous data. For units subject only to the Part 265 standards,
the short-term sampling program must, at a minimum, confirm that no
hazardous waste constituents (Appendix VII) have migrated into the
uppermost aquifer. For units subject to the Part 270 requirements
(because they are seeking an operating permit or the Agency has called
in their post-closure permit), the owner/operator should include
constituents selected from Appendix VIII in the sampling program.
After conducting the short-term sampling program (constituting the
first determination), the owner/operator must submit to the Regional
Administrator a written report describing the ground-water quality. If
the sampling program confirms that leakage has not occurred, the
owner/operator may continue the detection monitoring program or enter
into a consent agreement with the Agency to follow a revised detection
protocol designed to avoid future false triggers. If, however, the
short-term sampling confirms that leakage has occurred, the
owner/operator must immediately begin implementation of an assessment
program.
6.6 Description of Approach for Conducting Assessment
A variety of investigatory techniques are available for use during
a ground-water quality assessment. They can be broadly categorized as
either direct or indirect methods of investigation.
All assessment programs should be designed around the direct method
of actual collection of a sample with subsequent chemical analysis to
determine actual water quality (i.e., installation of monitoring wells).
Other methods of investigation may be used when appropriate to choose the
locations for well installation. For certain aspects of an assessment,
such as defining plume location, the use of both direct and indirect
methods may be the most efficient approach.
The methods planned for use in an assessment should be clearly
specified and evaluated to ensure that the performance standard
-151-
established for assessments can be met. Evaluating the use of direct
and indirect methods is discussed separately below.
6.6.1 Use of Direct Methods
Ground-water monitoring wells, either existing or newly installed,
are necessary to provide sampling data to establish the concentration of
hazardous constituents released from the hazardous waste management area,
and the rate and extent of their migration. The owner/operator should
construct assessment monitoring wells and conduct sampling and analysis
in a manner that provides reliable data. Chapters Three and Four,
respectively, present guidance in these areas.
At facilities where it is known or suspected that volatile organics
have been released to the uppermost aquifer, organic vapor analysis of
soil gas from shallow holes may provide an initial indication of the
areal extent of the plume (Figure 6-2). To this end, the owner/operator
may use an organic vapor analyzer (OVA) to measure the volatile organic
constituents in shallow hand-augered holes. Alternatively, the
owner/operator may extract a sample of soil gas from a shallow hole and
have it analyzed in the field, using a portable gas chromatograph. These
techniques are limited to situations where volatile organics are
present. Further, the presence of intervening, saturated, low
permeability sediments strongly interferes with the ability to extract a
gas sample. Although it is not necessarily a limitation, optimal gas
chromatography results are obtained when the analyte is matched with the
highest resolution technique (e.g., electron capture/halogenated
species). The owner/operator should attempt to evaluate the
effectiveness of this approach by initial OVA sampling in the vicinity of
wells known to be contaminated.
Descriptions of the direct methods and their limitations that will
be employed during assessment monitoring should be included in the
-152-
. .
.
.
.
CONTAMINANT PLUME
.
.
. .
.
.
.
.
.
$ . .
.
.
I LEGEND
DETECTION MONITORING WELL
iNO CONTAMINANT DETECTION)
I l
SOIL GAS ANALYSIS PROBE POINT
+
DETECTION MONITORING WELL
(CONTAMINANT DETECTED)
- - - EXTENT OF CONTAMINATED SOIL
I EXTENT OF GROUND-WATER CONTAMINATION
- PLUME
FIGURE 6-2 EXAMPLE OF USING SOIL GAS ANALYSIS TO DEFINE PROBABLE LOCATION OF
GROUND-WATER PLUME CONTAINING VOLATILE ORGANICS
-153-
assessment plan. These descriptions should be sufficiently detailed to
allow the method to be evaluated and to ensure that the method will be
properly executed.
Other direct methods that may be used to define the extent of a plume
include sampling of seeps and springs. Seeps and springs occur where the
local potentiometric surface intersects the land surface and results in
ground-water discharge into a stream, rivulet, or other surface water
body. Seeps and springs might be observed near marshes, at road cuts, or
near streams. Discharges from seeps and springs reflect the height of
the potentiometric surface and are likely to be most abundant during a
wet season.
6.6.2 Use of Indirect Methods
A variety of methods are currently available for identifying and, to
a limited extent, characterizing contamination in the uppermost aquifer.
There are several geophysical techniques of potential use to an owner/
operator, including electrical resistivity, electromagnetic conductivity,
ground penetrating radar, and borehole geophysics. Remote sensing and
aerial photography are additional indirect methods an owner/operator may
find useful. These techniques, with the exception of aerial photographic
methods, operate by measuring selected physical parameters in the
subsurface such as electrical conductivity, resistivity, and temperature.
The value of indirect methods is not the provision of detailed,
constituent-specific data for which they presently are clearly limited,
but rather for delineating the general area1 extent of the plume. This
is extremely important to the owner/operator for two reasons:
1. Knowing the general outline of the plume before additional wells
are constructed reduces the need for speculative wells. The
assessment monitoring program, therefore, becomes more
efficient, since well placement is guided by analytical data.
2. As the plume migrates and its margins change, the owner/operator
may track its movement to help locate new wells.
-154-
OSWER-9950.1
There are drawbacks to the exclusive use of geophysical techniques
in assessment monitoring relating to the high level of detail necessary
to characterize the chemical composition of a ground-water plume. For
these methods to be successful, contaminant(s) of interest must induce a
change in the subsurface parameter measured. This change, in turn, must
be distinguishable from ambient conditions. For example, the electrical
properties of organic hazardous constituents are generally attenuated or
masked by subsurface material properties. Unless these constituents are
present in high concentrations, they generally will not register during
resistivity or conductivity surveys. Moreover, nonuniform subsurface
conditions may obscure low levels of certain contaminants in ground
water. Another drawback to the exclusive use of geophysical methods at
present is their inability to measure specific concentrations of
individual constituents or provide good vertical resolution of
constituent concentration. In addition, man-made structures such as
powerline towers, buried pipelines, roads, and parking lots may interfere
with the performance and reliability of many geophysical methods, The
owner/operator should, therefore, only use indirect methods to guide the
installation of an assessment monitoring system and to provide an ongoing
check of the extent of contaminant migration.
6.6.3 Mathematical Modeling of Contaminant Movement
Mathematical and/or computer modeling may provide information useful
to the owner/operator during assessment monitoring and in the design of
corrective actions. The information may prove useful in refining concep-
tualizations of the ground-water regime, defining likely contaminant
pathways, and designing hydrologic corrective actions (e.g., pumping and
treating, etc.).
Since a model is a mathematical representation of a complex physical
system, simplified assumptions must be made about the physical system, so
that it may fit into the more simplistic mathematical framework of the
model. Such assumptions are especially appropriate, since the model
-155-
assumes a detailed knowledge of the relevant input parameters (e.g.,
permeability, porosity, etc.) e>r@ere in the area being modeled. This
is a limitation that must be considered since it would be impossible to
obtain all of the input parameters without disturbing and aitering the
physical system.
Since a model uses assumptions as to both the physical processes
involved and the spatial and temporal variations in field data, the
results produced by the model at best provide a salitative assessment of _____
the extent, nature, and migration of a contaminant plume. Because of the
assumptions made, a large degree of uncertainty is inherent in most
modeling simulations. Therefore, modeling results should not be unduly
relied upon in guiding the placement of assessment monitoring wells or in
designing corrective actions.
Where a model is to be used, site-specific measurements should be
collected and verified. The nature of the parameters required by a model
varies from model to model and is a function of the physical processes
being simulated (i.e., ground-water flow and/or contaminant transport),
as well as the complexity of the model. In simulating ground-water flow,
the hydrogeologic parameters that are usually required include:
hydraulic conductivity (vertical and horizontal); hydraulic gradient;
specific yield (unconfined aquifer) or specific storage (confined
aquifer); water levels in both wells and nearby surface water bodies; and
estimates of infiltration or recharge. In simulating contaminant
transport, the physicai and chemical parameters that are usually required
include: ground-water velocity; dispersivity of the aquifer; adsorptive
characteristics of the aquifer (retardation); degradation characteristics
of the contaminants; and the amount of each contaminant entering the
aquifer (source).
Dispersivity values of the aquifer should be based on site-specific
field test (i.e., tracer test) data or on field dispersivity values
obtained from the literature. Eautlon should be used where laboratory
OSKER-9950.1
dispersivity values are proposed, since such vaiues are often orders of
magnitude lower than field values. Retardation is often expressed as a
functional relationship (isotherm) between mass of contaminants in the
ground water and mass of contaminants adhering to the soil/rock. These
isotherms are based on soil bulk density, effective porosity, and cation
exchange capacity. Retardation may also be determined from the
octanol-water partition coefficient and fractional portion of organic
matter in representative volumes of soil. Degradation of contaminants
depends upon the type of constituents and the probability for chemical
and biological decay. Dispersion, retardation, and degradation tend to
decrease plume concentration and attenuate its travel time. Where these
parameters are not well characterized, use of lower values will produce --
greater conservatism in the results.
Contaminants leaking/leaching from a waste facility may react with
the pre-existing ground-water chemistry, resulting in an increase (or
decrease) in mobility. Background ground-water quality (e.g., indicator
parameters plus Cl-, Fe, Mn, Na+, S04, Ca+, Mg+, N03-, PO4', silicate,
ammonium, alkalinity, or acidity) is important to determine the reactivity
and solubility of hazardous constituents in ground water, and therefore
is useful in predicting constituent mobility under actual site conditions.
The physical and chemical characteristics of the site-specific leachate
(e.g., density, solubility, vapor pressure, viscosity, and octanol-water
partition coefficient) and hazardous waste constituents should also be
known as they affect constituent movement. To fully assess the effect on
contaminant mobility, a water chemistry model may be empioyed as a
component of the overall modeling study. Since this would add a large
degree of complexity to the modeling study, conservative assumptions
(i.e., maximum mobility of constituents) may be appropriate where time
and/or resources are limited.
Mathematical models are comprised of analytical equations by which
the hydraulic head or concectration of a contaminant may be calculated
-157-
for a specified location at a specified time. These models are
categorized into two main categories: those which are simple enough that
governing equations can be solved by analytic techniques ("analytical
models"); and those which are more complex and can only be solved by
computer ("numerical models"). The analytical soiutions to the first
category are often so sufficiently complex that they too can be solved by
computer. The numerical models are usually better suited to simulate the
complex conditions that describe the actual environment. Both types of
models, collectively referred to in this document as computer models,
require the recognition of inherent assumptions, the application of
appropriate boundary conditions, and the selection of a coherent set of
input parameters.
Model input parameters that can be determined directly should be
measured with consideration given to selecting representative samples.
Since the parameters cannot be measured continuously over the entire
region but only at discrete locations, care shouid be taken when
extrapolating over regions where there are no data. These considerations
are especially important where the parameters vary significantly in space
or time. The sensitivity of the model output both to the measured and
assumed input parameters should be determined and incorporated into any
discussion of model results. In addition, the ability of the model to be
adequately calibrated (i.e., the ability of the model to reproduce
current conditions (water levels, contaminant concentrations, etc.)) and
to reproduce past conditions should be carefully evaluated in assessing
reliability of model predictions. Model calibration with observed
physical conditions is critical to any successful ground-water modeling
exercise.
A plethora of ground-water computer models exists I many of which
would be suitable for a given situation. Since EPA is a public agency
and models used by or for EPA may become part of a jud icial action, EPA
-158-
OSWER-9950.1
approval of model use should be restricted to those models that are
publicly available (i.e., those models that are available to the public
for no charge or for a small fee). The subset of ground-water models
that are publicly available is quite large and should be sufficient for
most ground-water applications. Publicly available models include those
models developed by or for government agencies (e.g., EPA, USGS, DOE,
NRC, etc.) and national laboratories (e.g., Sandia, Oak Ridge, Lawrence
Berkeley, etc.), as well as models made publicly available by private
contractors. Any publicly available model chosen should, however, be
widely used, well documented, have its theory published in peer-reviewed
journals, or have some other characteristics reasonably assuring its
credibility. For situations where publicly available computer models are
not appropriate, proprietary models (i.e., models not reasonably
accessible for use or scrutiny by the public) should only be used where
the models have been well documented and have undergone substantial peer
review. Where these minimal requirements have not been met, the model
should not be considered reliable. A partial list of publicly available
computer models includes:
l Modular 3-Dimensional Finite Difference Groundwater Flow Model
(USGS), to evaluate complex hydrologic conditions;
l Computer Model of Two-Dimensional Solute Transport and Dispersion
in Ground Water (USGS), to predict contaminant transport;
l Illinois State Water Survey Random Walk Solute Transport Model
(ISGS), to predict contaminant transport;
l AT123D (Oak Ridge or EPA), to calculate concentrations isopleths
for transient contaminant flow through a simplistic aquifer flow
field in up to three dimensions;
l FEMWATER/FEMWASTE (Oak Ridge), to predict contaminant transport
in both the saturated and unsaturated zones;
l SWIFT (NRC or Sandia), to predict contaminant transport and
complex hydrologic flow conditions in up to three dimensions; and
l SWIP (EPA), similar to SWIFT.
-159-
If an owner/operator plans to use a model to guide an assessment
monitoring program, the owner/operator must be able and willing to
describe how the model works, as well as to explain all assumptions used
in calibrating and applying the model to the site in question. In
addition, the model and all related documentation should be made
available to EPA and its contractors for review and scrutiny.
6.7 Description of Sampling Number, Location, and Depth
The regulations require that the assessment plan specify the number,
location, and depth of wells to be installed as part of the assessment.
As the discussion on assessment methodology provided in Section 6.4 has
indicated, the owner/operator may use other sampling techniques (e.g.,
indirect methods and coring) in addition to the installation of permanent
monitoring wells to augment the data generated by wells. The owner/
operator's assessment plans should, however, specify the number,
location, and depth of wells that will be installed to characterize rate
and extent of migration, and constituent concentrations, and present
explanations for the decisions.
It may not always be possible for the owner/operator to identify at
the outset of an assessment the exact number, location, and depth of all
sampling that will be required to meet the goals of an assessment. Many
times the investigations undertaken to characterize contamination during
an assessment will proceed in phases in which data gained in one round of
sampling will guide the next phase of the investigation. For example,
surface geophysical techniques can be effectively used in tandem with the
installation of monitoring wells as a first phase in the assessment
program to obtain a rough outline of the contaminant plume. Based on
these findings, a sampling program may subsequently be undertaken to more
clearly define the three-dimensional limits of the contaminant plume. In
the third phase, a sampling program to determine the concentrations of
hazardous waste constituents in the interior of the plume may be under-
taken. In this case, a detailed description of the approach that will be
-160-
OSWER-9950.1
used to investigate the site should be included in the assessment plan.
This description should clearly identify the number, location, and depth
of any sampling planned for the initial phase of the investigation. The
outline should also clearly identify what basis will be used to select
subsequent sampling locations, including the geologic strata that are
likely to be sampled and the anticipated frequency of sampling. At a
minimum, several well clusters should be installed concurrently to define
the extent of contamination and concentration of contaminants (see
Section 6.7.2) and to profile the vertical extent of migration (see
Section 6.7.3).
6.7.1 Collection of Additional Site Information
The hydrogeologic site characterization requirements for the
detection monitoring program include:
• The subsurface geology below the owner/operator's hazardous waste
facility;
• The vertical and horizontal components of flow in the uppermost
saturated zone below the owner/operator's site;
• The hydraulic conductivity of the uppermost aquifer; and
• The vertical extent of the uppermost aquifer down to the first
confining layer.
If this characterization does not include all the hydrogeologic infor-
mation necessary to characterize the rate of contaminant movement, the
owner/operator should obtain this information for the assessment phase.
Examples of the additional information that may be needed to determine
the rate of contaminant movement include: mineralogy of the materials in
the migration pathway; ion exchange capacity of the material; organic
carbon content of the materials; background water quality of the pathway
(e.g., major cations and anions): the temperature of ground water in the
migration pathway; and the transmissivity and effective porosity of the
material in the pathway. This information will help define the transport
-161-
mechanisms which are most important at the site. All information
collected during the investigation of the plume (i.e., boring logs, core
analvsir. etc.) should be recorded and the hydrogeologic descriptions of
the site updated when appropriate.
Prior to adding new wells, a good estimation of plume geometry can
be determined from a review of current and past site characterizations.
For example, piezometer readings surrounding a contaminated detection
well can be taken to ascertain the current hydraulic gradient. When
these values are compared to the potentiometric surface map developed
during the site investigation, the general direction of plume migration
can be approximated. Any seasonal or regional fluctuations should be
considered during this comparison. A review of the subsurface geology
may also identify preferential pathways of contaminant migration.
To limit drilling speculative wells, geophysical and modeling
methods can also be employed to yield a rough outline of the plume. This
expedites the assessment monitoring program. Monitoring wells can then
be strategically placed to precisely define the plume geometry.
6.7.2 Sampling
The program of sampling undertaken during the assessment should
clearly identify the full extent of hazardous waste constituent migration
and establish the concentration of individual constituents throughout
the plume. In the initial phase of the assessment program, the owner/
operator's well installation/sampling should concentrate on defining
those areas that have been contaminated by the facility. A series of
well clusters should be installed in and around the plume to define the
extent of contamination and concentration of contaminants in the
horizontal plane. This network of monitoring wells, the number of which
may vary from site-to-site, must thoroughly define the horizontal
boundaries of the plume, and will identify and quantify contaminants.
Nell placement should be performed expediently, but in accordance with a
OSWER-9950.1
carefully thought out and documented assessment monitoring plan. To
obtain accurate plume definition at a particular moment in time it is
necessary to install well clusters concurrently. Surface geophysical
techniques should also be used, where appropriate, to help facilitate
plume definition. An assessment monitoring program that does not
thoroughly characterize the plume may result in higher assessment
monitoring costs, higher corrective action costs, and unnecessary delay.
The density of wells or amount of sampling undertaken to completely
identify the furthest extent of migration should be determined by the
variability in subsurface geology. Formations, such as unconsolidated
deposits with numerous interbedded lenses of varying permeability or
consolidated rock with numerous fractures, will require a more intensive
level of sampling and carefully placed wells to ensure that all contami-
nation is detected.
Assessment monitoring wells should be constructed of inert materials
to minimize chemical interaction between well casing material and
contaminant constituents. Also, the length of the well screen should be
relatively small, since the wells will be used to assess constituent
concentrations at discrete locations in the plume.
Sampling is also required to characterize the interior of any plume
detected at the site. This is important because the migration of many
constituents will be influenced by natural attenuation/transformation
processes. Sampling at the periphery of the plume may not identify all
the constituents from the facility that are reaching ground water, and
the concentration of waste constituents detected at the periphery of the
plume may be significantly less than in the plume's interior. Patterns
of concentration of individual constituents can be established throughout
the plume by sampling along several lines that perpendicularly transect
it. The number of transects and spacing between sampling points should
be based on the size of the plume and variability in geology observed at
the site. When sampling in fractured rock, for example, monitoring wells
-163-
should be located such that the well screens intersect fracture zones
along likely contaminant pathways. Sampling locations should also be
selected so as to identify those areas of maximum contamination within
the plume. In addition to the expected contaminants, the plume may
contain constituent degradation/transformation products, as well as
reaction products.
6.7.3 Sampling Depths
The owner/operator should specify in the assessment plan the depth
at which sampies will be taken at each of the planned sampling locations.
These sampling depths should be sufficient to profile the vertical distri-
bution of hazardous waste constituents at the site. Vertical sampling
should identify the f.Jll extent of vertical constituent migration.
Vertical concentration gradients, including maximum concentration of each
hazardous waste constituent in the subsurface, should similarly be
identified. The amount of vertical sampling required at a specific site
will depend on the thickness of the plume and the vertical variability
observed in the geology of the site. All potential migration pathways
should be sampled. The sampling program should clearly define the
vertical extent of migration by identifying those areas on the periphery
of the plume that have not been contaminated.
In order to establish vertical concentration gradients of hazardous
waste constituents in the plume, the owner/operator must obtain a
continuous sample of the plume, which means well clusters should be
employed. The owner/operator, however, cannot know the vertical extent
of the plume; therefore, the first well in the cluster should be screened
at the horizon where contamination was discovered, bearing in mind that
screen length should be relatively small. Additional wells in the
cluster should be screened, where appropriate, above and below the
initial sampling depth, until the margins of the plume are established.
Basically, several wells should be placed at the fringes of the plume to
define its vertical marqlns, and several wells should be placed with:n
OSWER-9950.1
the plume to identify contaminant constituents and concentrations. Care
must be taken in placing contiguously screened wells close together,
since the drawdown from one may influence the next, and thus change the
horizon from which the samples are drawn. Figure 6-3 shows an example of
assessment monitoring well cluster placement in the same setting as
depicted in Figure 2-5. These figures illustrate the relationship
between detection and assessment monitoring wells and clusters.
The specifications of sampling depths included in assessment plans
should clearly identify the interval over which each sample will be
taken. It is important that these sampling intervals be sufficiently
discrete to permit vertical profiling of constituent concentrations in
ground water at each sampling location. Sampling will only provide
measurements of the average contaminant concentration over the interval
from which that sample is taken. Samples taken from wells screened over
a large interval will be subject to dilution effects from uncontaminated
ground water lying outside the plume limits. Screened intervals should
be kept relatively small, especially where small vertical concentration
gradients are expected.
As part of the progressive assessment monitoring program, the
owner/operator can use geophysical techniques to help verify the adequacy
of the placement of the assessment monitoring network. Adjustments to
the assessment monitoring program may be needed to reflect plume
migration and changes in direction.
6.8 Description of Monitoring Well Design and Construction
The monitoring well design and construction requirements for
assessment monitoring well networks are equivalent to the requirements
presented in Chapter Three for detection wells.
6.9 Description of Sampling and Analysis Procedures
The owner/operator's sampling and analysis plan should be updated to
reflect the different analytical requirements of assessment monitoring.
-165-
FIGURE 6-3 EXAMPLE OF ASSESSMENT MONITORING WELL PLACEMENT
-166-
OSWER-9950.1
Otherwise, the sampling and analysis plan used by the owner/operator in
the detection monitoring program (see Chapter Four) should suffice for
assessment monitoring.
The assessment monitoring plan should identify the parameters to be
monitored by the owner/operator, and describe why these parameters are
suitable for determining the presence and concentration of contaminants
migrating from the facility in the ground water. At a minimum, the owner/
operator's assessment monitoring plan should include monitoring for all
hazardous waste constituents that are in the facility's waste. Hazardous
waste constituents, as defined in s260.10, include all constituents
listed in Appendix VII of Part 261, all constituents included in Table 1
of s261.24, and any constituent listed in Section 261.33.
An important consideration in assessment monitoring is the potential
for degradation/transformation of hazardous waste constituents; that
is, the chemical and/or physical change of a ground-water contaminant
resulting in a different intermediate or final product. The physical and
chemical properties of all hazardous waste constituents in the facility's
waste are an important consideration in evaluating an assessment
monitoring system. Assessment monitoring should aim at detecting all
contaminants, both initial as well as intermediate or final degraded/
transformed products. An example of the degradation/transformation
process is the breakdown of trichloroethylene (TCE) and its various
isomers into vinyl chloride, a highly toxic substance having different
chemical/physical characteristics than TCE. Since vinyl chloride is more
water soluble and less affected by sorption than TCE, the detection of
vinyl chloride in ground water should lead the owner/operator to suspect
the presence of TCE.
Facilities seeking an operating permit also have additional plume
characterization responsibilities pursuant to Part 270. Section
270.14(c)(4) requires permit applicants to expand their monitoring from
-167-
hazardous waste constituents (primarily Appendix VII) to the full
complement of Appendix VIII constituents (Note: Appendix VII is a subset
of Appendix VIII). Therefore, when a unit is subject to the Part 270
requirements (either because it seeks an operating permit or because the
Agency has called in its post-closure permit), the Agency recommends that
an owner/operator's assessment plan include parameters that will satisfy
the requirements of both Part 265 and Part 270.
Figure 6-4 illustrates in greater detail the sampling protocol
recommended by the Agency for units that are subject to both Part 265 and
Part 270. First, the owner/operator should perform an Appendix VIII scan
of samples from triggering detection monitoring wells. This scan will
provide the owner/operator with a list of hazardous constituents in the
wells that may be migrating into the uppermost aquifer. The owner/
operator should then select a limited number of identified constituents
for inclusion in a sampling program to establish geometric dimensions and
the rate of migration of the contaminant plume(s). Once the geometric
dimensions of the contaminant plume(s) have been established, the owner/
operator should sample for the full subset of identified Appendix VIII
constituents to determine vertical and horizontal concentration gradients.
6.10 Procedures for Evaluating Assessment Monitoring Data
The assessment plan must stipulate and document procedures for the
evaluation of assessment monitoring data. These procedures vary in a
site-specific manner, but must all result in determinations of the rate
of migration, extent. and composition of hazardous constituents of the
plume. Where the release is obvious and/or chemically simple, it may be
possible to characterize it readily from a descriptive presentation of
concentrations found in monitoring wells and geophysical measurements.
Where contamination is less obvious or the release is chemically complex,
however, the owner/operator should employ a statistical inference
approach. Owner/operators should plan initially to take a descriptive
-168-
OSWER-9950.1
FIGURE 6-4 SELECTION OF PLUME CHARACTERIZATION PARAMETERS
FOR UNITS SUBJECT TO PART 265 AND PART 270
-169-
approach to data analysis in order to broadly delineate the extent of
contamination. Statistical comparisons of assessment monitoring data
among wells and/or over time may be necessary, should the descriptive
approach provide no clear determination of the rate of migration, extent,
and hazardous constituent composition of the release.
The objective of assessment monitoring is to estimate the rate and
extent of migration and the concentration of constituents in the plume.
Data are therefore collected from a set of assessment monitoring wells
that will allow characterization of the dimensions and concentrations of
ground-water contaminant constituents (GWCCs) in the plume. In addition,
compared to detection monitoring, the number of chemical species analyzed
in assessment increases. Because the amount of data collected in
assessment is more voluminous than detection monitoring, it is extremely
important for the technical reviewer to make sure that the owner/operators
specify in their assessment plans the evaluation procedures for the data
required by §265.93(d)(3)(iii). The methods used to analyze assessment
monitoring data must emphasize organization, data reduction,
simplification, and summary.
Technical reviewers may find it useful and necessary to leave GWCC
data automated to verify the analyses submitted by owner/operators, to
compare recent submissions with historical data submissions, to manipulate
and evaluate the information for their specific purposes, or to support
permitting activities. EPA's data base system for environmental data is
called STORET and is a recommended mechanism for organizing ground-water
data acquired from hazardous waste management facilities. Several
positive features of STORET are:
l STORET has recently been modified to include data fields that
handle well-specific hydrogeological/technical information (e.g.,
well screen length, general lithology of the screened zone) in
conjunction with the GWCC data.
l Most State and EPA regional offices have access to STORET.
-170-
OSKER-9950.1
l STORET is well supported with capacity for efficient storage,
retrieval, and graphical analysis.
Represented below are specific evaluation and reporting procedures
that should be followed by the owner/operator when recording and evaluat-
ing assessment monitoring data. These procedures are used to structure,
analyze, simplify, and present the ground-water monitoring data to help
the technical reviewer evaluate the extent and concentration of ground-
water contaminants. The four evaluation or reporting procedures that
should be described in the assessment plan used to record data in the
on-site archives required by §265.94(b) are:
l Listing of Data;
l Summary Statistics Tables;
b Data Simplification; and
b Plotting of Data.
6.10.1 Listing of the Data
A list of all the dete ction monitoring and the assessment monitoring
data (as well as any data from related State or other EPA programs) that
have been collected should be available to technical reviewers when they
review on-site records. First, data as originally reported and verified
by the analytical laboratory for those measures requiring laboratory
evaluation, or as recorded in the field for those measures collected at
the time of sampling, should be available to the technical reviewer.
These reporting forms should include information indicating that quality
control samples (e.g., field and filter blanks) were obtained in the
field. Also, the laboratory reporting should indicate that the laboratory
has performed and reported standard quality control procedures (e.g.,
recovery analyses, analytical replicates etc.). Finally, the laboratory
reporting should include the data that were used to determine the method
detection limit or limit of detection (see Chapter 4). Explicit reporting
of these quality control data is essential for documenting the precision
and accuracy of owner/operator data submissions.
-17:-
The listing of GWCC concentration data should follow a format
similar to Table 6-1. The variables to be included in the listing are
codes that identify the GWCC, well, date, unit of measure, whether the
value was LT a limit of detection, and the concentration of the GWCC.
Also, the listing may include the results of and codes identifying the
quality control analyses performed. GWCC concentrations measured as LT a
specific method detection limit or limit of detection should be indicated
and, if possible, the GWCC concentration that was measured should be
reported with the LT designation. Otherwise, the value that accompanies
the LT designation should be the accepted detection limit for the method
used. Documentation that describes the meaning of the codes used in the
listing is required to eliminate ambiguity (e.g., Pb = lead, ppm = parts
per million}. The listing of GWCC data should include all measurements
from all wells since sampling began, including measurements obtained
durir,g detection monitoring.
The listing should be organized to allow quick reference to specific
data values. One categorization would be to first group by GWCC, then
well code, and finally the date, as shown in Table 6-1. For example, all
lead measurements are together, followed by all trichloroethylene
measurements, etc. The values for each GWCC from one well should be
grouped and ordered by date, followed by the data from the next well and
so on for all wells in the ground-water monitoring system. Alternate
sortings of the data listing may also be useful to the technical reviewer.
The data listing is not intended to function alone as an analytic
tool, but the technical reviewer can use the data listing to assist in
the review of the GWCC data. First, the ordered list of data will allow
the technical reviewer quick reference to every GWCC concentration
measurement if, for example, a spurious result was found in a supporting
data analysis or report. Also, by requiring a consistent and orderly
data listing, the technical reviewer can encourage the owner/operator to
-172-
OSWER-9950.1
GUCC
LE4D lUG/Ll
LE4O I UG/L t
LE4D IUGr’L I
LEAD 1 UG/L I
LEAD (UG/L I
LEAD I lJG/L J
LEAD LUG/L 1
LEAD I UG/L I
LEAD (UG/L I
LEAD (UG/L)
LEAD I UWL 1
LEAD tUG/Lt
LEAD t UG/L 1
LEAD LUG/L)
LEA0 IL&/L)
LEAD I UG/L t
LEAD t UC/L 1
TRICHLOROETHY LENE LUG/L I
TRICHLOROETHY LENE f UG/L t
TRICHLOROETHYLENE LUG/L 1
TRICHLOROETHYLENE lUG/L I
TRICHLOROETHYLENE LUG/L 1
TRICHLOROETHYLENE LUG/L1
TRICHLOROETHYLENE (UG/L t
TABLE 6-1
AN EXAMPLE ff H&J ASSESSMENT HDNITDRING DATA SHWLD BE LISTED
TRICHLOROETHY LENE LUG/L t
TRICHLOROETHYLENE LUG/L)
TRICHLOROETHYLENE I lJG/L 1
TRICHLOROETHYLENE (UC/L1
TRICHLOROETHYLENE LUG/L)
TRICHLOROETHY LEtlE ( UG/L 1
TRICHLOROETHY LENE I UG/L I
TRICHLOROETHY LEtlE LUG/L t
TRICHLOROETHYLENE (L&/L t
TRICHLOROETHYLENE IUG/Ll
TRICHLOROETHYLENE LUG/L I
TRICHLOROETHYLENE LUG/L t
TRICHLOROETHYLENE I UC/L )
TRICHLOROETHYLENE (UC/L t
TRICHLOROETHYLENE
TRICHLOROETHYLENE
TRICHLOROETHYLENE
TRICHLOROETHYLENE
TRICHLOROETHYLENE
TRICHLOROETHYLENE
1RXCHLOROETHYCEtIE
TRXCHLOROETHYLENE
TRICHLOROETHYLENE
TRXCHLOROETHYLENE
TRICHLOROETHYLEtdE
JRICHLOROETHYLEttE
TRICH LOROETHY LENE
(UG/L t
LUG/L)
IUG/LI
lUG/Ll
lUG/L t
LUG/L)
IUWC)
lUG/Ll
ILK/L1
IUG/Ll
LUG/L)
IUG/Ll
LUG/L)
WELL REPLICATE ALIPUOT
711
7A
7A
7A
7A
PA
9A
9A
9A
9A
90
96
90
90
90
90
9s
1*
1A
iA
1A
1A
1A
IA
IA
1A
1A
lA
1OA
10A
1OA
10A
1OA
IOA
1OA
1OA
IOA
10A
10A
IOA
10A
IOA
10).
1OA
100
1OB
108
108
1OB
108
108
TRICHLOROETHYLEtLE lUG/L) 108
fP:w%wtKE wt 1 181
2
1
2
1
1
1
2
1
2
A
A
B
A
B
A
B
A
A
A
A
B
A
A
A
A
A
A
A
A
A
A
4
A
A
A
A
A
A
B
A
A
A
A
A
A
A
A
A
A
A
0
A
B
A
B
A
A
A
A
A
A
f
DATE LT DETECTION CONCENTRATION UNITS
12JAN85
17FEB85
17FEB85
17FE685
17FEB85
2bAPR84
2bAPRB4
2bAPR84
05tlAY84
05tlAYB4
2bA?RB4
2bAF’R84
2bAPR84
05HAY 84
05UAY84
15JUN84
15~U~84
2bAPR84
OStlAY84
15JUN84
15JUL84
15AUG84
15SEP84
1 bOCT84
18WV84
20DEC84
12JANB5
17FEB85
2bAPR84
tbAPR84
2bAPA84
05tlAY84
05tlAY84
15JUN84
1 SAUG84
15SEP84
lbOCT84
18NOV84
200EC84
12JAN85
17FEB85
17FE685
17FEB85
17FEB85
2bAPR84
2bAPR84
2bAPR84
05MAY84
05t41~84
15JlJN84
15JUL84
15AUG84
w%
29.82 PPB
28.43 PPB
28.29 PPB
28.17 PPB
28.30 PPB
10.00 PPB
10.00 PPB
20.60 PPB
21.20 PPB
21.80 PPB
67.20 PPB
67.80 PPB
64.10 PPB
58.90 PPB
39.60 PPB
57.22 PPB
20.12 PPB
10.00 ??I3
10.00 PPB
10.00 PPB
11.10 PPB
10.00 PPB
10.10 PPB
10.70 PPB
10.00 PPB
10.00 PPB
10.00 PPB
10.00 PPB
17.00 PPB
17.30 PPB
17.60 PPB
21.00 PPB
21.40 PPB
21.20 PPB
22.90 PPB
19.40 PPB
19.60 PPB
30.10 PPB
31 .bO PPB
33.60 PPE
27.80 PPB
27.80 PPB
26.40 PPB
26.50 PPB
65.10 PPB
65.80 PPB
65.40 PPB
84.00 PPB
83.70 PPB
69.00 PPB
68.40 PPB
93.40
ilk598
PPB
FL i3
-173-
correct many of the data quality problems, that occur frequently on “raw”
laboratory reporting sheets. Finally, data can be placed more easily
onto a state or regional computer if the data are organized and reported
consistently in a listing, rather than on laboratory reporting sheets
having only the sample number identification instead of well codes, dates
of sampling, etc. (see the above discussion).
6.10.2 Summary Statistics Tables
The ground-water monitoring data should be surnmarlzed and presented
in tabular formats. Eight summary statistics should be calculated and
used in each of four summary tables. The eight summary statistics are:
Number of LT detection limit values
Total number of values
Mean
Median
Standard deviation
Coefficient of variation
Minimum value
Maximum value
The methodology used to estimate these summary statistics can be found in
many statistical textbooks.
The four zables of summary statistics should include summaries by:
l GKC summary (e.g., Table 6-2)
l GiiCC summary by wel! (e.g., Table 6-3)
l G’VJCC summary by well and date (e.g., Table 6-4)
l Quality control data
The tables should be formatted so that there are from one to three
columns on the left side of each table, which provide data identifying,
where applicable, the GWCC, well, and date. Eight columns, one for each
summa r:; statis:ic, should be to the right of the identifying columns.
-l-7?-
OSWER-9950.1
NUMBER OF COEFFICIENT
SAtlPLE LT DETECTION STANDARD OF
GUCC SIZE LItlIT VALUES HEAN tIEDIAN DEVIATION VARIATION HINIHWI tlAXIWUH
CHROMIUM I UG/L ) 129 14 50.83 32.10 59.43 117 5.00 345.21
tiETHYLENE CHLORIDE (UG/L) 137 37 21.45 14.30 23.17 108 5.00 112.70
LEAD (UG/L t 129 15 50.31 20.43 168.22 334 1.00 1879.23
TRICHLOROETHYLENE (UG/Ll 139 32 31.21 20.40 27.68 88.7 5.00 98.90
-175-
TABLE 6-3
AN EXARPLE OF HDU DATA SWLO BE SUWHRIZED BY GMXMLL COMBINATION
GWCC WELL
CHROtlIUIl (UG/L I 1A
CHROllIWl (UC/L I 1OA
CHROPlIUtl (UG/L 1 108
CHROHIUM (lJG/L ) 11A
CHROnIUtl I UG/L ) 17.A
CHROlIIUIl I UG/L) 13A
CHROrlIUrl (lx/L ) 14A
CHROflIUM (UG/L I 15A
CHROMIUn (UG/L 1 16A
CHROtlIUH (UG/L ) 17A
CHROflIUH (UG/L 1 3A
CHROhIUIl I UG/L 1 7A
CHROflIUn (UG/C 1 9A
CHROHIM (UG/L I 90
HETHYLENE CHLORIDE (UG/L) 1).
nETHY LENE CHLORIOE (lJG/L 1 1OA
nETHYLENE CHLORIDE (UG/L 1 106
IIETHYLENE CHLORIDE IUG/Ll 11A
METHYLENE CHLORIDE (UG/L) 12A
llETHY LENE CHLORIDE (UG/L 1 13A
tlETHYLENE CHLORIDE (UG/L) 14A
WETHY LENE CHLORIDE (UG/L 1 15A
HETHYLENE CHLORIOE (UG/Ll 16A
flETHYLENE CHLORIDE (lJG/L) 17A
nETHI LENE CHLORIDE (UG/L ) 2A
tlETHYLENE CHLORIDE (UG/L) 3A
SAIIPLE
SIZE
9
16
17
2
11.
11
10
9
9
3
9
11
5
7
11
16
17
2
11
11
10
9
9
3
2
II
NUlBER OF
LT DETECTION
LIHIT VALUES
3
0
0
0
0
0
0
0
0
1
7
0
0
3
7
0
0
2
0
0
0
1
0
3
2
10
IlEAN MEDIAN
a.74 10.20
63.57 49.00
89.15 48.92
13.51 13.51
135.74 109.32
27.36 28.09
45.22 48.06
27.76 29.69
54.82 79.47
10.51 12.31
6.29 5.00
59.64 58.71
21.12 15.00
11.40 11.10
7.40 5.00
13.66 12.95
22.91 21.50
5.00 5.00
39.28 23.60
20.73 18.90
86.21 76.95
11.27 11.90
30.40 28.70
5.00 5.00
5.00 5.00
5.60 5.00
-176-
STANDARD
DEVIATION
2.83
38.99
93. lb
1.70
100.49
3.83
7.08
5.13
36.21
4.87
2.58
12.48
9.41
7.05
3.76
2.26
4.20
0.00
30.97
7.36
18.25
2.44
7.75
0.00
0.00
1.99
COEFFICIENT
OF
VARIATION
32.3
61.3
105
12.6
74.0
14.0
15.7
18.5
66.1
46.3
41.1
20.9
44.5
61.9
50.9
16.6
18.3
0.0
78.8
35.5
21.2
21.7
25.5
0.0
0.0
35.5
nINIHun nAxnnm
5.00 11.24
30.90 140.00
10.10 324.00
12.31 14.72
lb.23 345.21
20.86 32.53
32.63 57.03
18.62 32.01
11.89 87.31
5.00 14.23
5.00 11.51
46.91 85.01
13.80 32.10
5.00 21.60
5.00 16.40
11.00 16.90
19.70 34.20
5.00 5.00
14.30 98.40
11.00 28.60
70.10 112.70
5.00 12.90
16.70 40.10
5.01) 5.00
5.00 5.00
5.00 11.60
OSWER-9950.1
GlJCC
CHROMIUH I uG/L 1
CHROMILRI (UG/L t
‘3iF!DtlILRl i U&/L 1
CHROHIUn ( uG/L)
CHROtlIUtl (UG/L I
CHROtl1UI-l (UG/L )
CHROHILtM I UG/L 1
CHROHIUfl I UC/L 1
CHROWJll (UG/Lt
cnRulIul l m/L I
cHROtlIWl I UG/L )
CHROMIl&l I UG/L J
CHROflIUll I UWL 1
CHROHIUN (UC/L )
cHROIIIUH t UG/L 1
tiETHY LENE CHLORIDE
tlETHYLENE CHLORIDE
tIETHY LENE CHLORIDE
HETHYLENE CHLORIOE
tlETHYLENE CHLORIDE
tlElHrLENE CHLORIDE
METHYLEM CHLORIDE
IlElHYCENE CHLORIDE
HETHlLFNE CHIORIOE
HETHYLENE CHLORIDE
RETHYLENE CHLORIDE
TABLE 6-4
AN EXAtPLE OF How OATA Si#NLD l3E ‘i’.MhRIZED By =CMLL/MTE COWlNATION
WELL
3A
7A
?A
?A
?A
7A
7A
?A
?A
9A
9A
90
9B
?B
90
(lJG/Lh 1A
(UG/L) 1A
tLtG/L) IA
(UG/Li 1A
DATE
17FEB85
15JUL84
15AUG84
15SEP84
lbOCT84
lbNoV84
2OOEC84
1 tJAN85
17FEBBI
26APA84
OStlAY84
2bAPR84
OStlAY84
15JUN84
lSJUL84
2bAPR84
05tlAY84
15JLtN84
15JlJL 84
(UG/L) IA 15AUG04
(UC/L1 IA 15SEP84
(UG/Lt IA IbOCT84
lUG/Ll 1A 1 BNOV84
lUC/Lt 1A 200EC84
luc/Lt IA 12JAN85
lUG/Lt IA 17FEB85
TOTAL NUMBER OF
-2~ LT OETECTION
OF VALUES LItlIT VALUES
1 I
I 0
1 0
1 0
1 0
1 0
1 0
1 0
4 0
3 0
2 0
3 1
2 0
1 I
1 1
I 1
1 1
1 0
1 0
1 1
1 I
1 0
1 1
1 0
1 I
1 1
-177-
MEAN
5.00
85.01
73.52
67.50
b4.38
60.01
58.71
58.70
47.05
14.27
31.40
9.47
20.70
5.00
5.00
5.00
5.00
10.00
10.00
5.00
5.00
16.40
5.00
10.00
5.00
5.00
NEDIAN
5.00
85.01
73.52
67.50
64.38
60.01
58.71
58.70
46.95
14.00
31.40
11.10
20.70
5.00
5.00
5.00
5.00
10.00
10.00
5.00
5.00
16.40
5.00
10.00
5.00
5.00
STAMARD
DEVIATION
COEFFICIENT
OF
VARIATION
0.22 0.5
0.64 4.5
0.99 3.2
3.91 41.4
1.27 6.1
.
IlINIHlM
5.00
85.01
73.52
67.50
64.38
60.01
58.71
58.70
46.91
13.80
30.70
5.00
19.80
5.00
5.00
5.00
5.00
10.00
10.00
5.00
5.00
16.40
5.00
10 .oo
5.00
5.00
tl*x1mJtI
5.00
85.01
73.52
67.50
64.38
60.01
58.71
58.70
47.38
15.00
32.10
12.30
21.60
5.00
5.00
5.00
5.00
10.00
10.00
5.00
5.00
lb.40
5.00
10.00
5.00
5.00
There will be one row for each category that is being summarized. A
summary statistics table by GWCC, for example, will have a number of rows
equal to the number of GWCC that have been sampled. The GWCC-well table
will have a number of rows equaling the number of GWCCs measured times
the number of wells in the monitoring system (provided that each GWCC was
measured at least once in each well). The GWCC-well-date table will be
the largest table, and each row should be prefixed with a GWCC, well, and
date code. The statistics in the GWCC-well-date table should summarize
all replicate sampling that was performed for each GWCC, from each well,
during each sampling.
The sample sizes, ranges, minimum, and maximum values will provide a
rapid means for checking whether errors appear in the data. It will also
facilitate rapid evaluation of GWCC concentrations over the entire
ground-water monitoring system. In addition, the summary statistics will
allow evaluation of spatial change in GWCC concentrations, which includes
identifying the rate and extent of migration of the GWCC plume.
The quality control data should be provided whenever assessment
monitoring data are submitted by an owner/operator. The quality control
data can be submitted in the format in which they are received from the
laboratory, provided that all data are clearly documented. The quality
control samples taken in the field (e.g., field and sampling equipment
blanks) may not be identified when the samples are supplied to the
laboratory, but should be identified in assessment monitoring data
submissions. Owner/operators should ensure that the laboratories provide
the quality control data that support and validate the data resulting
from the analysis of their field samples.
6.10.3 Data Simplification
Ranking procedures, which are described in this section, may be
useful for simplifying and interpreting spatial trends in GWCC concen-
trations by allowing rapid determination of which wells have the overall
-178-
OSWER-9950.1
highest and lowest GWCC concentrations. Table 6-5 presents an example of
a data set analyzed by a ranking procedure.
The ranking can be performed using the mean, median, maximum, or
minimum concentration values in the summary statistics table describing
the values from each GWCC-well combination. For example, the mean
concentration from each well is ranked from lowest to highest for each
GWCC. The well with the lowest mean concentration of a GWCC will receive
a value of 1: the well with the next highest concentration of the same
GWCC will receive a value of 2, and so on. If two or more wells have the
identical mean concentration, then the ranks for these wells will be
averaged and applied to all wells with the same mean concentration. This
procedure should be repeated for each GWCC that was detected at ieast
once at every well in the monitoring system. The pH values may be ranked
from highest to lowest rather than from lowest to highest, depending on
whether the ground-water contamination is likely to result in an increase
or decrease in pH. It is also useful to calculate an overall average
rank for each well by averaging the ranks across all GWCCs associated
with the well. These.ranks should be presented in a table using GWCCs as
column headings, and well codes as row headings. It may be helpful to
group GWCCs with similar chemistry (e.g., volatile organics, metals,
salts, etc.) and order the rows based on the wells with spatial proximity
(e.g., upgradient, downgradient in plume, downgradient out of plume,
shallow screen depth). This will facilitate identification of specific
groups of wells where high concentrations of GWCC were detected.
6.10.4 Graphic Displays of Data
Ground-water data should be plotted to allow evaluation of temporal
changes in GWCC concentrations over time. Each plot should consist of a
X or horizontal axis, which represents time with year and month
identified at intervals. The Y or vertical axis should represent the
concentrations of GWCCs. The plots may be constructed using the mean
values from the GWCC-well-date summary statistics table, and one plot
-17%
WELL
17A
2A
4A
11A
3*
9h
1A
90
ISA
13A
1UA
14A
7A
12A
16A
100
TABLE 6-5
AN EXAWlE OF I-KM R/‘&KS OF THE HEAN CCMENTRATIONS FCR EACH
WCCMLL COMBINATION CAN BE USED TO SWLIFY AND PRESENT CWCENTR.ATIO)(
WTA COLLECTED FOR A VARIETY OF GUCCs IN A NU?BER OF MDNITORING WELLS
RA.W OF HEAN RANK OF MEAN
CHROHIM LEAD
CONCENTRATION CONCENTRATION
3 3
3 3
3 3
5 3
1 6
6 3
2 a
4 7
8 9
7 12
12 10
9 16
11 I1
14 15
IO 14
13 13
RANK OF MAN RAW OF BEAN
TCE P!c
CONCENTRATION CONCENTRATION
1
4 3
2 6
5 3
3 7
12 8
b 11
10 9
11 10
7 12
9 15
8 14
14 13
13 lb
3
AVERAGE HELL
RANK ACROSS
GUCC
2.50
3.00
3.00
3.75
3.75
4.25
5.00
7.75
8.50
9.50
10.75
11.00
11.50
12.75
12.75
13.75
-180-
OSWER-9950.1
could be presented for each GWCC/well combination as in Figure 6-5.
Alternatively, it may be more insightful to plot the data from several
wells or GWCCs on one graph, as in Figure 6-6, provided the lines do not
overlap excessively.
It may also be useful to plot data on facility maps, so that trends
in GWCCs both vertically and horizontally can be evaluated. The summary
statistics from the GWCC-well table can be used to provide data for
plotting. A map of the facility, which identifies well locations, should
be used to depict horizontal trends in concentrations. Geological cross
sections and/or a facility map may be useful for plotting vertical trends
in GWCC concentrations. The mean concentrations can be placed near each
well location, similar to the construction of potentiometric maps
described earlier. It may also be helpful to plot isopleth contours of
concentration on the maps.
6.11 Rate of Migration
An assessment plan should specify the procedures the owner/operator
will use to determine the rate of constituent migration in ground water.
A rapid approach will generally be required for determining the rate of
migration during interim status assessments. Migration rates can be
determined by monitoring the concentration of GWCCs over a period of time
in monitoring wells aligned in the direction of flow. If these wells are
located both at the edge and the interior of the plume, subsequent
analysis of the monitoring data can then provide an estimate of the rate
of migration, both of the contaminant front as a whole and of individual
constituents within the plume. This approach does not necessarily provide
a reliable determination of the migration rates that will occur as the
contaminant plume continues to move away from the facility in light of
potential changes in geohydrologic conditions. More importantly, this
approach requires the collection of a time series of data of sufficient
duration and frequency to gauge the movement of contaminants. Such a
-181-
FIGURE 6-5. PLOT OF CHROMIUM CONCENTRATIONS OVER TIME (WELL 9A)
-182-
OSWER-3950.1
250
200
150
100
50
O- I I I I I I I I I I
APR
1984
MAY JUN JUL AUG SEP OCT
TIME INTERVAL
(MONTHS)
NOV DEC JAN FEl3
1985
FIGURE 6-6. CHROMIUM AND LEAD CONCENTRATIONS OVER TIME (WELL 9A)
483-
delay is normally inappropriate during initial assessment of ground-water
contamination, since a relatively quick determination of at least an
estimate of migration rates is required to deduce the impact of
ground-water contamination and to formulate an appropriate reaction.
Estimates of migration rates can be based on aquifer properties obtained
during the site investigation and knowledge of the physico-chemical
properties of contaminants known to be present. By recognizing the
various factors that can affect transport processes of the GWCCs, the
owner/operator can obtain approximate potential rates of migration during
an initial assessment phase. Continued monitoring of the plume to verify
rates of migration during assessment monitoring should serve as a basis
for identifying additional monitoring well locations
Initial approximations of contaminant migration rates based on
ground-water flow rates are not reliable without verification because of
potential differential transport rates among various classes of chemical
constituents. Differential transport rates are caused by several factors
including:
0 Dispersion due to diffusion and mechanical mixing;
l Retardation due to adsorption and electrostatic interactions; and
l Transformation due to physical, chemical, and/or biological
processes.
Dispersion results in the overall dilution of the contaminant and
blurring at plume boundaries. Dispersion can result in a contaminant's
arriving at a particular location before the arrival time computed solely
on average rates of ground-water flow. Alternatively, retardation
processes can delay the arrival of contaminants beyond that calculated by
the average rates of ground-water flow. Local geology will also affect
constituent migration rates. Relating rates of constituent migration to
rates of ground-water flow is appropriate for a quick approximation
during the initial assessment phase, but this should be followed by a
more comprehensive study of migration rates.
-184-
OSWER-9950.1
Simple slug tests are not the preferred method for determining the
aquifer characteristics. The slug test is limited to the immediate
vicinity where it is performed, and its results often cannot be projected
across an entire site.
At those facilities where sufficient immiscible contaminants have
leaked to form and migrate as a separate immiscible phase (see
Figure 6-71, additional analysis will be necessary to evaluate the
migration of these contaminants away from the facility. Chapter Five
contains a discussion of the ground-water monitoring techniques that can
be used to sample multi-phased contamination. The formation of separate
phases of immiscible contaminants in the subsurface is largely controlled
by the rate of infiltration of the immiscible contaminant and the
solubility of that contaminant in ground water. Immiscible contaminants
generally have some limited solubility in water. Thus, some amount of
immiscible contaminant leaking from the facility will enter into solution
in ground water and migrate away from the facility as dissolved
constituents. If the amount of immiscible fluid reaching ground water
exceeds the solubility constant, however, the ground water in the upper
portion of the water table aquifer will become saturated, and the
contaminant will form a separate immiscible phase.
At this point, the behavior and migration of the contaminants
present in the immiscible phase will be strongly influenced by their
density relative to ground water. If the immiscibles are less dense than
ground water, the immiscibles will tend to coalesce on the surface of the
potentiometric surface and form and migrate as a separate immiscible
layer floating on the ground water. If the density of the immiscible
contaminants is similar to that of ground water, the immiscible will tend
to mix and flow as a separate phase with the ground water, creating a
condition of multiphase flow.
If the density of the immiscibles is greater than ground water, the
immiscibles will tend to sink in the aquifer (see Figure 6-7). As the
-1a5-
immiscibles sink and reach unaffected ground water in a deeper portion of
the aquifer, more of the immiscible contaminant will tend to enter into
solution in ground water and begin to migrate as dissolved constituents.
If enough of the dense immiscible contaminants are present, however, some
portion of these contaminants will continue to sink as a separate
immiscible phase, until a formation of reduced permeability is reached.
At this point, these contaminants will tend to coalesce and migrate as a
layer of dense immiscibles resting on the geologic barrier.
In each of these cases, the contaminants present in the separate
immiscible phase may migrate away from the facility at rates different
from that of ground water. In many cases, they will migrate at rates
slower than or equivalent to ground water, but in some instances migra-
tion rates can be greater. In addition, migration of the immiscibles may
not be in the direction of ground-water flow. However, it is important
to reemphasize that some amount of these contaminants will invariably
dissolve in ground water and migrate away from the facility as dissolved
constituents.
Light immiscible contaminants will migrate downgradient to form a
floating layer above the saturated zone (see Figure 6-7). The direction
of ground-water flow will dictate the movement of this light immiscible
layer. Important factors involved in its migration rate include the
intrinsic permeability of the medium and the density and viscosity of the
contaminants. With time, an ellipsoidal plume develops, overlying the
saturated zone as depicted in Figure 6-7. While it is possible to
analyze the behavior of the light immiscible layer using analytical or
numerical models, the most practical approach for determining the rate
and direction of migration of such a light immiscible layer during an
assessment may be to observe its behavior over time with appropriately
located monitoring wells.
-186-
OSWER-9950.1
CONTAMlNATlON
SOURCE AREA I
F
SAND
GROUND-WATER FLOW * .
LEGEND
.-1.1.. LIGHT IMMISCIBLE PLUME
- . - MAIN PLUME
- - HEAVY IMMISCIELE PLUME
.,o- PIEZOMETRIC SURFACE
FIGURE 6-7 GENERAL SCHEMATIC OF MULTIPHASE CONTAMINATION IN SAND
-187-
The migration of a layer of dense immiscibles settled on a confining
layer may be strongly influenced by gravity. Depending on the slope of
the confining layer in the gradients used to calculate flow rates. A
program of continued monitoring of the dense immiscible layer should
always be included in the assessment plan to verify direction and rate of
movement.
6.12 Reviewing Schedule of Implementation
The assessment plan should specify a schedule of Implementation.
Each assessment program will have to include the amount of work revolved
in the assessment and other local factors such as weather and
availability of equipment and personnel. The schedule should include a
sufficient number of milestones, so that the Agency can judge whether
sufficient progress is being made toward the completion of the
assessment. Any continued monitoring undertaken during the maintenance
phase of assessment should be scheduled at least on a quarterly basis.
Activities planned to initially determine whether contamination has
actually occurred should not unnecessarily delay the implementation of a
comprehensive assessment. When an extensive program to collect additional
data to remedy inadequacies in currently available data is to be under-
taken, these activities should require only a short period for completion.
Additional analysis of water quality data should require no more than
15 days to 30 days. Sampling to determine actual concentrations of
hazardous waste constituents should require only time enough for sample
collection and analysis, followed by a brief period for subsequent
analysis of the data.
A thorough discussion of monitoring well placement, and monitoring
well design and construction, can be found In Chapters Two and Three,
respectively. A discussion of the ground-water monitoring techniques
necessary to effectively characterize a multiphase containment migration
is also given in Chapter Four of this document.
-188-
OSWER-9950.1
REFERENCES
Freeze, R.A. and J.A. Cherry. 1979. Groundwater. Prentice-Hall Inc.
Hoaglin, D.C., F. Mosteller, and Hal Takey. 1985. Exploring Data
Tables, Trends, and Shapes. John Wiley and Sons, Inc. 475 pp.
MacKay, D.M., P.V. Roberts and J.A. Cherry. 1985. Transport of Organic
Contaminants in Ground-Water, Engineering Science & Technology,
Vol. 19, No. 5, pp. 284-392.
U.S. Environmental Protection Agency. 1979. Water-Related Environmental
Fate of 129 Priority Pollutants, Volume 1, Introduction, Technical
Background, Metals and Inorganics, Pesticides, and PCBs.
EPA-440/4-79/029a.
U.S. Environmental Protection Agency. 1979. Water-Related Environmental
Fate of 129 Priority Pollutants, Volume 2, Halogenated Aliphatic
Hydrocarbons, Halogenated Ethers, Monocyclic Aromatics, Pthalate
Esters, Polycyclic Aromatic Hydrocarbons, Nitrosamines, Miscellaneous
Compounds. EPA-440/4-79/029b.
U.S. Environmental Protection Agency. 1983. Ground-Water Monitoring
Guidance for Owners and Operators of Interim Status Facilities.
National Technical Information Service. PB83-209445.
U.S. Environmental Protection Agency. September 1985. Protection of
Public Water Supplies from Ground-Water Contamination. EPA-625/4-85/
016.
-189-
OSWER-9950.1
GLOSSARY
AR t-Test - Averaged replicate t-test.
Adsorb - Adherence of atoms, ions, or molecules to the surface of another
substance.
Aliphatic Hydrocarbons - Class of organic compounds characterized by
straight or branched chain arrangement of the constituent carbon atoms.
Analyte - A specific compound or element of interest undergoing analysis.
Annular Sealant - Material used to seal the space between the borehole
and the casing of the well. Annular sealants prevent surface
contaminants from entering the well.
Annular Space - The open space formed between the borehole and the well
casing.
Anticline - A fold, usually from 100 meters to 300 kilometers in width,
that is convex upward with the oldest strata at the center.
Appendix VII Monitoring Requirements - A compilation of constituents
arranged by EPA hazardous waste numbers which caused the Administrator
to list the waste as an EP Toxic Waste (E) or Toxic Waste (T) in 40 CFR
§ 261.31 and § 261.32.
Appendix VIII Constituents - A list of 297 toxic constituents (Part 261)
which, if present in a waste, may make the waste hazardous. The waste
containing these constituents poses a substantial hazard to human health
or the environment when improperly treated, stored, transported or
disposed.
Aquiclude - A geologic formation which may contain ground water but is
incapable of transmitting significant quantities of ground water under
normal hydraulic gradients.
Aquifer Adsorptive Characteristics - Ability of an aquifer to retain
atoms, ions, or molecules.
Aquifer Degradation Characteristics - Aquifer contamination can be
characterized by parameters such as pH, total organic halogens, total
organic carbon, temperature, and specific conductance.
Aromatic Hydrocarbons - Class of unsaturated cyclic organic compounds
containing one or more ring structures. The name aromatic is derived by
the distinctive and often fragrant odors of these compounds.
-191-
Assessment Monitoring - A program of monitoring ground water under
interim status requirements. After a release of contaminants to ground
water has been determined, the rate of migration, extent of
contamination, and hazardous constituent concentration gradients of the
contamination must be identified.
Assessment Plan - The written detailed plan drawn up by the owner/operator
which describes and explains the procedures the owner/operator intends to
take to perform assessment monitoring.
Attenuation - To reduce, weaken, dilute, or lessen in severity. value, or
amount such as the attenuation of contaminants as they migrate from a
particular source.
Background Concentrations - A schedule of sampling and analysis that
is completed during the first year of monitoring. All wells in the
monitoring system must be sampled on a quarterly basis to determine
drinking water characteristics, ground-water quality, and contamination
indicator parameters. For each upgradient well, at least four replicate
measurements must be made for the contamination indicator parameters.
Background Mean - The arithmetic average of a set of data, used as a
control value in subsequent statistical tests.
Background Variance - The variance is the measure of how far an
observation value departs from the mean. Background refers to the
observations used for control in subsequent statistical tests.
Basement - The oldest rocks recognized in a given area, a complex of
metamorphic and igneous rocks that underlies all the sedimentary
formations.
Bentonite - A sedimentary rock largely comprised of clay minerals that
has a great ability to absorb water and swell in volume.
Bluooey Line - Air supply line during drilling operations.
Borehole - A circular hole drilled or bored into the earth, usually for
exploratory or economic purposes, such as a water well or oil well.
Borehole Geophysics (Geophysical Borehole Logging) - A general term that
encompasses all techniques in which a sensing device is lowered into a
borehole for the purpose of characterizing the associated geologic
formations and their fluids. The results can be interpreted to determine
lithology, geometry resistivity, bulk density, porosity, permeability,
and moisture content and to define the source, movement, and physical/
chemical characteristics of ground water.
-192-
OSWER-9950.1
CABF t-Test - Co&ran's Approximation to the Behrens-Fisher t-Test. ~I_-
'Carbonate Environments -~-~ - Refers to sedimentary rock environments composed
of calcium or magnesium carbonate.
Casing - The pipe between the intake (screen) section and the surface,
serving as a housing for pumping equipment and conduit for the piunped
water.
Chain of Cust& - Method for documenting the history and possession of a
sample fromThe time of its collection through its analysis and data
reporting to its final disposition.
Chemical Standards - Materials made from ultra-pure compounds used to
calibrate laboratory analytical equipment.
Chemical Spike (Spike) - A sample that contains a measured amount of a I_-
known analytc used for determining matrix interferences.
Cluster - (see Well Cluster). -- _-
Coefficient of Variation -- - The standard deviation divided by the mean of
a set of data. (Note: the coefficient of variation can be expressed as
a percentage by multiplying the number obtained by iO0).
Color - A diagnostic property of a rock, mineral, or sediment.
Components- of Variability - The characteristics that vary from one
statistical population to another, such as well locations, and analytical
lab errors.
Concentration Profiles - Graphic representations of the horizontal and
vertical locations of contaminant concentration levels on maps and
cross-sections.
Confined Aquifer - An aquifer under greater than atmospheric pressure
bounded above and below by impermeable layers with distinctly lower
permeabilities (aquitards) than the aquifer itself.
Confining Layer - A geologic stratum exhibiting low permeability and
having little or no intrinsic permeability.
Core - A continuous columnar sample of the lithologic units extracted
from a borehole. Such a sample preserves stratigraphic contacts and
structural features.
f!ccrosive Environments - Subsurface zones containing ground water or soil
corrosive to monitoring well constructzon materials.
-193-
Dedicated (Sampling Equipment) - Sampling equipment (e.g., bladder pump,
bailer) which is reserved for use in only one monitoring well.
Denqsiticn. Environment - A geographically restricted complex where a
sediment accumulates, described in geomorphic terms and characterized by
physical, chemical, and biological conditions (e.g., flood plain, lake,
beach).
Dielectric - Substance having a very low electrical conductivity.
Direct Methods for Hydrogeological Investigations - Methods (e.g.
borehole logging, pump tests) which involve the drilling, collection,
observation, and analysis of geologic materials, water samples, and
drawdown/recovery data.
Dispersivity - Ability of a contaminant to disperse within the ground
water by molecular diffusion and mechanical mixing.
Disposal Facility - A facility as defined in 40 CFR 260.10 where hazardous
waste is intentionally placed into or on land or water, and at which waste
will remain after closure of the facility.
Dolomite - A carbonate sedimentary rock composed predominantly of
CaMg(CO312.
Downgradient - In the direction of decreasing static head.
Downgradient Well - A well which has been installed hydraulically
downgradient of the site, and is capable of detecting the migration of
contaminants from a regulated unit. Regulations require the installation
of three or more downgradient wells depending upon the site- specific
hydrogeological conditions and potential zones of contaminant migration.
Drawdown - The lowering of the water level in a well as a result of
withdrawal.
Drilling Mud - Fluids which are used during the drilling of a borehole or
well to wash soil cuttings away from the drill bit and adjust the
specific gravity of the liquid in the borehole so that the sides of the
hole do not cave in prior to installation of a casing.
Drive Pipe - Casing consisting of the drive shoe and riser. This casing
follows the auger bit as it advances.
Drive Shoe - Steel coupling or band at the bottom edge of the casing
reinforced to withstand drive pressures during cable tool and drill-
through casing driver methods.
-194-
OSWER-9950.1
Dunnett's Modification - Dunnett's version of the t-Test. Uses Dunnett's
calculated t-statistics rather than the Student's t-statistics.
Electrical Resistivity (ER) - A surficial geophysical method whereby
known current is applied to spaced electrodes in the ground and the
resulting electrical resistance is used to detect changes in earth
materials between and below the electrodes. ER is particularly useful
for facilities receiving electrically conductive wastes (e.g., inorganic)
at sites characterized by settings having minimal quantities of high
resistance materials.
Electromagnetic Conductivity (EM) - A surficial geophysical method
whereby induced currents are produced and measured in conductive
formations from electromagnetic waves generated at the surface. EM is
used to define shallow ground water zones characterized by high dissolved
solids content.
Equipment Blank - Chemically pure solvent (typically reagent grade water)
that is passed through an item of field sampling equipment and returned
to the laboratory for analysis, to determine the effectiveness of
equipment decontamination procedures.
Equipotential - Equal pressure. Equipotential lines are lines drawn
between points of equal pressure.
Esters - Class of organic compounds derived by the reaction of an organic --
acid with an alcohol.
False Negative - Contamination has occurred but the results of the t-Test
fail to indicate contamination.
False Positive - No contamination has occurred, but the results of the
t-test indicate contamination.
Field Blank - A laboratory-prepared sample of Type II-Reagent grade water
or pure solvent which is transported to the sampling site for use in
QA/QC evaluation of field sampling procedures. See equipment blank and
trip blank.
Filter Pack - Sand or glass beads that are placed in the annulus of the
wall between the borehole wall and the well screen to prevent formation
material from entering through the well screen. Glass beads are smooth,
uniform, clean, well rounded, and siliceous. The filter pack typically
extends 2 feet above the screen.
Floaters - Light phase organic liquids in ground water capable of forming
an immiscible layer which can float on t'ne water table.
-195-
Flow Net - A set of intersecting equipotential lines and flow lines
representing a two-dimensional steady flow through porous media.
Fluvio-Glacial Depositional Environment - A complex melange of glacially
borne and riverine sediments deposited at the head of a melting glacier.
The sediments range in grain size from clays to boulders, and in places
are typically unsorted.
Fracture Zone - A thickness of strata that has undergone mechanical
failure due to stress (e.g., cracks, joints, and faults).
Geophysical Borehole Logging - See Borehole Geophysics.
Glacial Till - Unsorted and unstratified sediment originating directly ___---
from glacial ice (i.e., not reworked by glacial meltwater).
Goodness of Fit - A statisticai test to determine the likelihood that
sample data have been generated from a population that conforms to a
specified type of probability distribution.
Grain Size - The general dimensions of the particles in a sediment or
rock, -- or of the grains of a particular mineral that make up a sediment or
rock. It is common for these dimensions to be referred to with broad
terms, such as fine, medium, and coarse. A widely used grain size
classification is the Udder-Wentworth grade scale.
Ground Penetrating Radar (GPR) - A geophysical method used to identify --.
surface formations which will reflect electromagnetic radiation. GPR
is useful for defining the boundarles of buried trenches and other
subsurface installations on the basis of time-domain ref;ectrometry.
Ground-Water Detection Monitoring Program - A monitoring well system ~..._. -_ ~.
capable of yielding ground-water samples for analysis. Upgradient wells
must be installed to obtain representative background ground-water
quality In the uppermost aquifer and be unaffected by the facility.
Downgradient we:?-, must be placed immediately adjacent to the hazardous
waste management area(s) to detect hazardous waste or hazardous waste
constituents migrating from the facility.
Halogenated Hydrocarbons - An organic compound containing one or more
halogens (e.g., fluorine, chlorine, bromine, and iodine).
Hazardous Waste - A solid waste which exhibits any of the hazardous
characteristics defined in 40 CFR 5261
excluded ds a hazardous waste.
.2 and has not been specifically
Categorical list of hazardous waste are
provided In 40 CFR 5261.3.
-136-
OSWER-9950.1
Hazardous Waste Constituent - A constituent which causes a waste to be
classified hazardous based upon the criteria cited in 40 CFR 55261.2 and
261.3.
Hazardous Waste Management - The collection, source separation,
transportation, processing, treatment, recovery, and disposal of
storage,
hazardous waste.
Hazardous Waste Management Area - The area within a facility's property
boundary which encompasses one or more hazardous waste management unit or
cell.
Headspace - The empty volume in a sample container between the water
level and the cap.
HeavingSand - Unconsolidated ---_ sand that cannot maintain the integrity of
the borehole wall.
High Corrosion Potential - Material with a high propensity for
electrochemical degradation.
High-Yield Well- - A relative term referring to a well capable of quick
recovery after it has been purged of at least three casing volumes (i.e.,
samples can be collected immediately after purging).
Hydraulic Conductivity - A coefficient of proportionality which describes _. -. -
the rate at which a fluid can move through a permeable medium. It is a
function of the media and of the fluid flowing through it.
Hvdraulic Connection - The hydraulic relationship between two different -----. ..._~~
lithologic layers.
Hydraulic Head - Water-level elevation in d well or piezometer. The -.. .__--
elevation typically referenced to mean sea level to which water rises as
a result of hydrostatic pressure.
Illite (Illitic) - A general name for a group of three layer, mica-like
clay minerals. These clay minerals are intermediate in composition and
structure (between muscovite and montmorillonite).
Indicator Parameters - pH, specific conductance, total organic carbon
(TCC), total organic halogens (TOX).
Indirect Nethods for Hydroqeoloqiqations - Methods which ___-. -.
include the measurement or remote sensing of various physical and/or
chemical properties of the earth (e.g., electromagnetic conductivity,
elp_ctricai resistivity, specific conductance, geophysical logging, aerial
photography).
-1g7-
Interim Status Detection Monitoring - Ground-water monitoring conducted
under 40 CFR 265, Subpart F.
Intrinsic - The characteristic of a porous medium to
transmit liquid under a hydraulic gradient, it is independent of the
liquid itself.
Ion Exchange Capacity - Measured ability of a formation to adsorb charged
atoms or molecules.
Karst Topography (Karst) - A topographic area which has been created by
the dissolution of a carbonate rock terrain. This type of topography is
characterized by sinkholes, caverns, and lack of surface streams.
Ketones - Class of organic compounds where the carbonyl group is bonded
to two alkyl groups.
Landfill - A disposal facility or part of a facility where hazardous
waste is placed in or on the land, and which is not a land treatment
facility, a surface impoundment, or an injection well.
Leach - To wash or drain by percolation.
Leachate - A solution produced by the movement or percolation of liquid
through soil or solid waste and the subsequent dissolution of certain
constituents in the water.
Leachate Management System - A method of collecting leachate and
directing it to a treatment or disposal area.
Less Than Detection Limits - A phrase which indicates that a chemical
constituent was either not identified or not quantified at the lowest
level of sensitivity of the analytical method being employed by the
laboratory. Therefore, the chemical constituent either is not present in
the sample, or it is present in such a small concentration that it cannot
be measured by the analytical procedure.
Limestone - Sedimentary rock primarily made up of calcium carbonate.
Liner - A continuous layer of natural or man-made materials lining the
bottom and/or sides of a surface impoundment, landfill, or landfill cell
that restricts the downward or lateral escape of hazardous waste,
hazardous waste constituents, or leachate.
Lithology - The systematic description of rocks, in terms of mineral
composition and texture.
-198-
OSWER-9950.1
Low-Yield Well - A relative term referring to a well that cannot recover
in sufficient time after well evacuation to permit the immediate
collection of water samples.
Mature Karst - Karst environment where the physical features (e.g.,
sinkholes, caves) are well defined (see Karst).
Maximum Value - In a set of data, the measurement having the highest
numerical value.
Mean - The sum of all measurements collected over a statistically
significant period of time (e.g., one year) divided by the number of
measurements.
Median - The middle point in a set of measurements ranked by numerical
value. If there are an even number of measurements, the medium is the
mean of the two central measurements.
Mineralogy - The study of minerals, including their formation, occurrence,
properties, composition, and classification.
Minimum Value - In a set of data, the measurement having the lowest
numerical value.
Mounding - A phenomenon usually created by the recharge of ground water
from a manmade structure into a permeable geologic material. Associated
ground-water flow will be away from the manmade structure in all
directions.
Mud - See Drilling Mud.
Non-Dedicated Sampling Equipment - Equipment used to sample more than a
single sampling point.
Normal Distribution - The character of data that follows the Gaussian
distribution (bell) curve.
Number of LT Detection Limit Values - The number of times a chemical
parameter was not detected by a given analytical procedure over a
statistically significant period of time (e.g., one year).
Octanol-Water Partition Coefficient - A coefficient representing the ratio
of solubility of a compound in octanol to its solubility in water. As
the octanol-water partition coefficient increases, water solubility
decreases.
-199-
Organic Polymers - Drilling fluid additives comprised of long-chained,
heavy organic molecules. Drilling fluid additives are used to increase
drilling rates and drilling fluid yields, thereby decreasing operational
costs.
Organic Vapor Analyzer - A field monitoring device used to determine the
concentrations of organic compounds in air using flame ionization or
photoionization detection systems.
Outwash Sand - Stratified sediment (usually sand and gravel) removed from
a glacier by meltwater streams and deposited beyond the active margin of
a glacier.
Oxidizing Acids - An acid (e.g., HN03) which tends to lose electrons in
a reaction.
PVC - Abbreviation for polyvinyl chloride.
Permeability - The capacity of a porous rock, sediment, or soil to
transmit a fluid.
Petrographic Analysis - Systematic description and classification of
rocks.
Photoionization Analyzer - See Organic Vapor Analyzer.
Phreatic Zone - See Saturated Zone.
Piezometers - Generally a small diameter, non-pumping well used to
measure the elevation of the water table or potentiometric surface.
Plume Characterization - Provides information on concentration profiles
and rates of migration.
Polyethylene - A plastic composed of synthetic crystalline polymer of
ethylene (HzC:CHz}. Polymer may be low density (branched) or high
density (linear).
Polypropylene - A plastic composed of synthetic crystalline polymer of
propylene (C&In.
Potentiometric Data - Ground-water surface elevation values obtained at
wells and piezometers. The data is primarily used to construct potentio-
metric maps indicating the ground-water flow direction and elevation.
Potentiometric Surface (Piezometric Surface) - The surface that represents
the level to which water from a given aquifer will rise by hydrostatic
pressure. Xhen the water-bearing zone is the uppermost unconfined
aquifer, the potentiomecric surface is identical to the water table.
-2OO-
OSWER-9950.1
Pump Test - A test made by pumping a well for a period of time and
observing the change in hydraulic head in adjacent wells. A pump test
may be used to determine degree of hydraulic interconnection between
different water-bearing units, as well as the recharge rate of a well.
Purged Water - Wastewater from wells undergoing evacuation or being used __-
for aquifer testing.
Qualified Professional in Geology - A professional, by degree, experience,
or certification, specializing in the study of the earth material science.
Rate of Migration - The time a contaminant takes to travel from one
stationary point to another. Generally expressed in units of time/
distance.
Regional Administrator - The Regional Administrator of the appropriate
Regional Office of the Environmental Protection Agency, or the authorized
representative.
Regulated Unit - Hazardous waste management unit. The number of regulated
units will define the extent of the hazardous waste management area.
Retardation - Preferential retention of contaminant movement in the
subsurface zone. Retention may be a result of adsorbtion processes or
solubility differences.
Sampling and Analysis Plan - A detailed document describing the proce-
dures used to collect, handle, and analyze ground-water samples for
detection or assessment monitoring parameters. The plan should detail
all quality control measures which will be implemented to ensure that
sample collection, analysis, and data presentation activities meet the
prescribed requirements.
Saturated Zone (Phreatic Zone) - A subsurface zone below which all rock
pore space is filled with water.
Seismic Prospecting - Any of the various geophysical methods for
characterizing subsurface properties based on the analysis of elastic
waves artificially generated at the surface (e.g., seismic reflection,
seismic refraction).
Shelby Tube or Split Spoon Sampler - Devices used in conjunction with a
drilling rig to obtain an undisturbed core sample of the strata.
Significant Digits - The number of digits reported as the result of a
calculation or measurement (exclusive of following zeroes).
Sinkers - Dense phase organic liquids which coalesce in an immiscible
layer at the bottom of the saturated zone.
-201-
Slug Test - A single well test to determine the in-situ hydraulic --
conductivity of an aquifer by the instantaneous addition or removal of
a known quantity (slug) of water into or from a well. and the subsequent
measurement of the resulting well recovery time.
Smectite - A commonly used name for the montmorillonite group of clay
minerals. These clay minerals have swelling Droperties and a high cation
exchange capacity.
Solution Channel - A tubular or planar channel formed by solution in
carbonate-rock (Karst) terrains.
Standard Deviation - The positive square root of the variance. The
variance is the average of the squares of the differences between the
actual measurements and the mean.
Stratigraphy - The science (study) of original succession and age of rock
strata, also dealing with their form, distribution, lithologic composi-
tion, fossil content, and geophysical and geochemical properties.
Stratigraphy also encompasses unconsolidated materials (i.e., soils).
Structural Anomaly - A geologic feature, especially in the subsurface,
distinguished by geophysical, geological, or geochemical means, which is
different from the general surroundings,
Surface Impoundment - A facility or part of a facility which is a natural
topographic depression, man-made excavation, or diked area formed
primarily of earthen materials (although it may be lined with man-made
materials), which is designed to hold an accumulation of liquid wastes or
wastes containing free liquids, and which is not an injection well.
Examples of surface impoundments are holding, storage, settling, and
aeration pits, ponds, and lagoons.
T-Test - The t-test is a statistical method used to determine the
significance of difference or change between sets of initial background
and subsequent parameter values.
TOC - Total organic carbon.
TOX - Total organic halogens.
Teflon --- - Trade name for polyperfluorethylene.
Texture - The interrelationship between the size, shape, and arrangement
of minerals or particles In a rock.
Total ?lumber of Values - The n:lmber of measurements (including less than
detection values) made for a chemical parameter over a statistically
significant period of time (e.g., one year).
-202-
OWER-9950.1
Transformatio>! - Process of establishing correspondence between elements
in one set of data to elements in another set of data, such that each
element in the first set corresponds to a unique element in the second
set.
Tremie Method - Method whereby bentonitelcement slurries are pumped .- -
uniformly within the annular space of a well.
TripBlank -__ - A field blank that is transported to the sampling site,
handled the same as other samples, then returned to the laboratory for
analysis in determining QA/Qc of sampie handling procedures.
Type II Water - Water prepared by using a still (deionized supply water
may be necessary) designed to produce a distillate having a conductivity
of less than 1.0 umho/cm at 2 5*C and a maximum total matter content of
0.1 mg/l.
Undulatinq - A periodic rise and fall of a surface; having a wavy outline
or appearance.
Unsaturated Zone - A subsurface zone above the water table in which the
interstices of a porous medium are only partially filled with water.
Also referred to as Vadose Zone.
UJq_radient ~-__ - In the direction of increasing static head.
Upgradient Well - One or more wells which are placed hydraulically
upgradient of the site and are capable of yielding ground-water samples
that are representative of regional conditions and are not affected by
the regulated facility.
Uppermost Aquifer - The geologic formation, group of formations, or part
of a formation that contains the uppermost potentiometric surface capable
of yielding a significant amount of ground water to wells or springs and
may include fill material that is saturated. There should be very
limited interconnection, based upon pumping tests. between the uppermost
aquifer and lower aquifers. Consequently, the uppermost aquifer includes
all interconnected water-bearing zones capable of significant yield that
overlie the confining layer.
Vadose Zone - See Unsaturated Zone.
Volatile Constituents - Solid or liquid compounds which are relatively
unstable at standard temperature and pressure and undergo spontaneous
phase change to a gaseous state.
Volatile Organics - Liquid or solid organic compounds with a tendency to
pass into the vapor state.
-203-
Wastewater Treatment System - A collection of treatment processes
designed and built to reduce the amount of suspended solids, bacteria,
oxygen-demanding materials, and chemical constituents in wastewater.
xater Table - The water level surface below the ground at which the ..- .--__--
vadose zone ends and the phreatic zone begins. It is the level to which
a well screened in t‘ne unconfined aquifer would fill with water.
3ell - A shaft or pit dug or bored into the earth, generally of a
:iyiindrical form, and often waLled with tubing or pipe to prevent the
c=i:th froT. zay:ir,g in.
.Y.d 11 Cluster - A weli cluster consists of two or more wells completed
(screened) to ‘different depths in a single borehole or a series of
ho~~hoies in close proximity to each other. From these weils, water
Sainpi2S that are representative of the different horizons within one or
:nors aquifers can be collected.
T.’ .ie11 Euacuation - Process of removing stagnant water from a well prior to
sampling.
:GRay Diffraction - An analytical technique used for mineralogical
iharacteri-iation. A sample is exposed to a filtered and monochromatic
beam of X-rays and the reflected energy is measured and used to identify
soil collold types, degree of interleafing, or interstratification, and
,*rariatlons in interplatelet spacings.
Zone of Potential Contaminant Miqration - Any subsurface formation or - ---___
layer which is permeable and would preferentially channei the fiow of
contaminants away from a regulated facility.
-204-
OSWER-9950.1
REFERENCES
Anderson, T.W., Sclove, L. Stanley L. 1986. The Statistical Analysis of
Data, Second Edition. The Scientific Press, Palo Alto, California.
Bates, Robert L. and J.A. Jackson. 1980. Glossary of Geology, Second
Edition. American Geological institute,
Zohn, Hinrich L., Brian L. McNeal, George A. O'Conner; 1979. Soil
Chemistry. John Wiley & Sons, New York.
Century Systems Corporation. Date unknown. Operating and Service Manual
for Century Systems' Portable Organic Vapor Analyzer (OVA) Model
OVA-128, Revision C.
Driscoll, F.G. 1986. Groundwater and Wells, Second Edition. Johnson
Division, St Paul, Minnesota.
Environmental Protection Agency Interim Status Standards for Owners and
Operators of Hazardous Waste Facilities: 40 CFR 265. Environmental
Reporter. March 29, 1985.
Environmental Protection Agency Interim Status Standards for Owners and
Operators of Hazardous Waste Facilities: 40 CFR 265. Environmental
Reporter. April 4, 1986.
Environmental Protection Agency Interim Status Standards for Owners and
Operators of Hazardous Waste Facilities: 40 CFR 265. Environmental
Reporter. November 15, 1985.
Hays, W.L. 1981. Statistics, Third Edition. Holt, Rinehart and
Winston, New York, New York.
HNu Systems Inc. 1975. Instruction Manual for Model P1 101
Photoionization Analyzer.
Keller, Edward A. 1976. Environmental Geology. Charles E. Merrill
Publishing Company, Columbus, Ohio.
Kohler, Heinz. 1985. Statistics for Business and Economics. Scott,
Foreman and Co., Illinois.
The Condensed Chemical Dictionary, Tenth Edition. 1981. Revised by
Gessner G. Hawley, Van Nostrand Reinhold Company.
-205-
USEPA/EMSL. March 1979. Handbook for Analytical Quality Control in
Water and Wastewater Laboratories, EPA-600/4-79-019.
-206-
OSWER-9950.1
INDEX
AR t-Test, 130, 131, 133, 136
Adsorb, 78, 114
Aliphatic Hydrocarbons, 78
Analyte, 108, 109
Annular Sealant, 82, 83
Annular Space, 84, 85
Anticline. 39, 41
Aquiclude, 90
Aromatic Hydrocarbons, 78
Assessment Monitoring, 120, 124,
137, 140, 143, 144, 145
Assessment Plan, 145, 146, 147
Attenuation, 163
Background Concentrations, 136,
138
Background Mean, 123, 136
Background Variance, 123
Basement, 39
Bentonite. 77, 83, 88
Borehole, 6, 8, 9, 73, 74, 76,
77
Borehole Geophysics, 154
CABF t-Test, 130
Carbonate Environments, 64
Casing, 78-86, 99
Chain of Custody, 97, 98, 114,
119
Chemical Standards, 38
Chemically Spiked, 98
Cluster, 26, 55, 164
Coefficient of Variation, 174
Color, 58
Components of Variability, 132
Concentration Profiles, 143
Confining Layer, 5, 8, 12, 35,
36, 100, 161, 188
Core, 162
Corrosive Environments, 78
Dielectric, 80
Dispersivity, 49, 50, 156, 157
Downgradient, 132
Downgradient Monitoring Well, 45,
46, 47, 49, 51, 107, 123, 137,
139, 148
Drawdown, 33, 165
Drive Pipe, 75
Drive Shoe, 74
Dunnett's Modification, 131
Equipment Blank, 119
Equipotential, 58
Esters, 78
False Negative, 131, 135
False Positive, 131, 134, 135,
137, 139, 148, 150
Field Blank, 119
Filter Pack, 78, 82
Floaters, 56, 100, 101
Flow Net, 28, 29
Glacial Till, 47, 58
Goodness of Fit, 132, 133, 134
Grain Size, 58
Halogenated Hydrocarbons, 78
Hazardous Waste, 46, 52, 143,
164, 167, 168
Hazardous Waste Constituent, 46,
52, 151, 157, 162, 164, 167
Hazardous Waste Management, 125
Hazardous Waste Management Area, 51
Headspace, 114
Heaving Sand, 73
High Corrosion Potential, 78
Hydraulic Conductivity, 5, 3, 11,
15, 17, 30, 31, 50, 35, 156, 161
Hydraulic Communication, 62
Hydraulic Head, 26, 30, 31, 62,
157
Indicator Parameters, 54, 136,
139, 145, 150
Intrinsic Permeability, 186
Ion Exchange Capacity, 161
Karst, 47, 64, 69
Ketones, 78
Landfill, 64
Leach, 78
Leachate, 53, 150, 157
Limestone, 36, 39, 66
Liner, 50
Lithology, 6, 50. 56, 170
-207-
Mature Karst, 64
Maximum Value, 174
Mean, 174, 179
Median, 174, 179
Mineralogy, 8, 15, 17, 161
Minimum Value, 174
Mud, 77
Non-Normality, 133
Normal Distribution, 134
Organic Polymers, 77
Organic Vapor Analyzer, 100, 152
Outwash Sand, 58
Oxidizing Acids, 78
PVC, 78, 80, 106
Permeability, 18, 19, 34, 35, 36,
53, 54, 152, 156, 163, 186
Petrographic Analysis, 15, 17
Photoionization Analyzer, 100
Piezometer, 24, 26, 28, 71, 162
Plume Characterization, 144, 145,
167
Polyethylene, 78, 106, 109, 112
Polypropylene, 78, 109, 112
Potentiometric Data, 66
Potentiometric Surface, 6, 24,
26, 30, 35, 36, 39, 49, 52, 53,
55, 64, 90, 100, 154, 162
Pump Test, 33
Purged Water, 104
RCRA Monitoring Well, 71
Rate of Migration, 168, 170, 181
Retardation, 156, 157
Sampling and Analysis Plan, 97,
98, 108, 165
Saturated Zone, 54, 78, 161, 186
Shelby Tube, 12
Side-by-Side, 94
Sinkers, 56, 100
Slug Test, 32, 185
Split Spoon Sampler, 12
Standard Deviation, 174
Stratigraphy, 9, 11
T-Test, 28, 123, 124, 130
TOC, 105, 111, 114
TOX, 105, 111, 114
Teflon, 78
Texture, 58
Total Number of Values, 174
Transformation, 134, 163, 164,
167, 184
Tremie Method, 84
Trip Blank, 118, 119
Type II Water, 107, 109
Undulating, 58
Unsaturated Zone, 15, 80
Upgradient, 132
Upgradient Monitoring Well, 45,
46, 51, 66, 67, 69, 123, 133,
138, 136, 137
Uppermost Aquifer, 1, 5, 8, 34,
35, 58
Vadose Zone, 49, 80
Volatile Constituents, 107
Volatile Organics, 78, 105, 114, 152
Well (Monitoring Well), 24, 47,
51, 71, 99, 100, 101, 102, 116
Well Cluster, 55, 56, 165
Well Evacuation, 97, 102, 107,
108, 116
X-Ray Diffraction, 8, 15, 17
-208-
OSWER-9950.1
APPENDIX A
EVALUATION WORKSHEETS
OSWER-9950.1
APPENDIX A.1
CHARACTERIZATION OF SITE HYDROGEOLOGY WORKSHEET
The following worksheets have been designed to assist the enforcement
official in evaluating the program the owner/operator used in characterizing
hydrogeologic conditions at his site. This series of worksheets has been
compiled to parallel the information presented in Chapter 1 of the TEGD.
I. Review of Site Hydrogeologic Investigatory Techniques
A. Was the site investigation and/or data collection
performed by a qualified professional in geology?
B. Did the owner/operator survey the following existing
regional data:
1. U.S.G.S. Maps?
2. Water supply well logs?
3. Other (specify)
C. Did the owner/operator use the following direct
techniques in the hydrogeologic assessment:
1. Soil borings/rock corings?
2. Materials tests (e.g., grain size analyses,
standard penetration tests, etc.)?
3. Piezometer installation for water level
measurements at different depths?
4. Slug tests?
5. Pump tests?
6. Geochemical analyses of soil samples?
7. Other (specify)
D. Did the owner/operator use the following indirect
techniques to supplement direct techniques data:
1. Geophysical well logs?
2. Tracer studies?
3. Resistivity and/or electromagnetic conductance?
4. Seismic survey?
5. Hydraulic conductivity measurements of cores?
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N)
(Y/N) __
A-1
E.
F.
G.
H.
I.
J.
K.
L.
M.
6. Aerial photography?
7. Ground penetrating radar?
8. Other (specify) ~-
(Y/NIP-
(Y/N) __-
Did the owner/operator document and present the
raw data from the site hydrogeologic assessment?
Did the owner/operator document methods (criteria)
used to correlate and analyze the information?
Did the owner/operator prepare the following:
1. Narrative description of geology?
2. Geologic cross sections?
3. Geologic and soil maps?
4. Boring/coring logs?
5. Structure contour maps of aquifer and aquitard?
6. Narrative description of ground-water flows?
7. Water table/potentiometric map?
8. Hydrologic cross sections?
Did the owner/operator obtain a regional map of the
area and delineate the facility?
If yes, does this map illustrate:
1. Surficial geology features?
2. Streams, rivers, lakes, or wetlands near the facility?
3. Discharging or recharging wells near the facility?
Did the owner/operator obtain a regional
hydrogeologic map?
If yes, does this hydrogeologic map indicate:
1. Major areas of recharge/discharge?
2. Regional ground-water flow direction?
3. Potentiometric contours which are consistent with
observed water level elevations?
Did the owner/operator prepare a facility site map?
If yes, does the site map show:
1. Regulated units of the facility (e.g., landfill
areas, ipoundments)?
2. Any seeps, springs, streams, ponds, or wetlands?
(Y/NIP
(Y/NIP
(Y/NIP
(Y/N)___
(Y/N)
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) .-
(Y/NIP
(Y/NIP
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N) __-
(Y/NIP
(Y/N)-
(Y/NIP
A-2
OSWER-9950.1
3. Location of monitoring wells, soil borings,
or test pits? (Y/N) --
4. How many regulated units does the facility have? -__
If more than one regulated unit then,
l Does the waste management area encompass all
regulated units? (Y/N) --
Gr
0 Is a waste management area delineated for each
regulated unit? (Y/N)-_m-_
II. Characterization of Subsurface Geology of Site
A. Soil boring/test pit program:
1. Were the soil borings/test pits performed under
the supervision of a qualified professional? (Y/N)-
2. Were the borings placed close enough to accurately
portray stratigraphy with minimal reliance on
inference? (Y/N) --
3. If not, did the owner/operator provide documentation
for selecting the spacing for borings? (Y/NIP
4. Were the borings drilled to the depth of the first
confining unit below the uppermost zone of
saturation? (Y/N)-
5. Indicate the method(s) of drilling:
l Auger (hollow or solid stem)
l Mud rotary
l Air rotary
l Reverse rotary
l Cable tool
l Jetting ___-
l Other (specify)
6. Were continuous sample corings taken? (Y/N)-
7. How were the samples obtained (check method[s])
0 Split spoon
l Shelby tube, or similar
l Rock coring
l Ditch sampling
l Other (explain)
8. Were the continuous sample corings logged by a
qualified professional in geology?
9. Does the field boring log include the following
information:
l Hole name/number?
l Date stared and finished?
l Geologist's name?
(Y/N)-
(Y/N)-
(Y/N) __-
(Y/N)__
A-3
l Driller's name?
l Hole location (i.e., map and elevation)?
l Drill rig type and bit/auger size?
l Gross petrography (e.g., rock type) of
each geologic unit?
l Gross mineralogy of each geologic unit?
l Gross structural interpretation of each
geologic unit and structural features
(e.g., fractures, gouge material, solution
channels, buried streams or valleys,
identification of depositional material)?
l Development of soil zones and vertical extent
and description of soil type?
l Depth of water-bearing unit(s) and vertical
extent of each?
l Depth and reason for termination of borehole?
l Depth and location of any contaminant encountered
in borehole?
l Sample location/number?
0 Percent sample recovery?
l Narrative descriptions of:
-- Geologic observations?
-- Drilling observations?
10. Were the following analytical tests performed on the
core samples:
l Mineralogy (e.g., microscopic tests and x-ray
diffraction)?
l Petrographic analysis:
- degree of crystallinity and cementation of
matrix?
- degree of sorting, size fraction (i.e,
sieving), textural variations?
- rock type(s)?
- soil type?
- approximate bulk geochemistry?
- existence of microstructures that may effect
or indicate fluid flow?
0 Falling head tests?
l Static head tests?
l Settling measurements?
l Centrifuge tests?
l Column drawings?
B. Verification of subsurface geological data
1. Has the owner/operator used indirect geophysical methods
to supplement geological conditions between borehole
locations?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/Nly
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)___
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(-f/N)-
A-4
OSWER-9950.1
2. Does the number of borings and analytical data indicate
that the confining layer displays a low enough
permeability to impede the migration of contaminants
to any stratigraphically lower water-bearing units?
3. Is the confining layer laterally continuous across
the entire site?
4. Did the owner/operator consider the chemical
compatibility of the site-specific waste types
and the geologic materials of the confining layer?
5. Did the geologic assessment address or provide
means for resolution of any information gaps of
geologic data?
6. Does the laboratory data corroborate the field
data for petrography?
7. Does the laboratory data corroborate the field
data for mineralogy and subsurface geochemistry?
C. Presentation of geologic data
1. Did the owner/operator present an adequate number
of geologic cross sections of the site?
2. Do each of these cross sections:
0 identify the types and characteristics of
the geologic materials present?
l define the contact zones between different
geologic materials?
l note the zones of high permeability or
fracture?
l give detailed borehole information including:
-- location of borehole?
-- depth of termination?
-- location of screen (if applicable)?
-- depth of zone of saturation?
-- depiction of any geophysical logs?
3. Did the owner/operator provide a topographic map which
was constructed by a licensed surveyor?
4. Does the topographic map provide:
l contours at a maximum interval of two-feet?
l locations and illustrations of man-made
features (e.g., parking lots, factory
buildings, drainage ditches, storm drains,
pipelines, etc.)?
l descriptions of nearby water bodies?
l descriptions of off-site wells?
0 site boundaries?
0 individual RCRA units?
l delineation of the waste management area(s)?
0 solid waste management areas?
l well and boring locations?
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/NIP
(Y/NIP
(Y/NIP
(Y/NIP
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/NIP
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)
(Y/N)-
(Y/N)
A-5
5. Did the owner/operator provide an aerial photo-
graph depicting the site and adjacent off-site
features?
6. Does the photograph clearly show surface water
bodies, adjacent municipalities, and residences
and are these clearly labelled?
III. Identification of Ground-Water Flow Paths
A. Ground-water flow direction
1.
2.
3.
4.
5.
6.
7.
8.
9.
Was the well casing height measured by a
licensed surveyor to the nearest 0.01 feet?
Were the well water level measurements taken
within a 24 hour period?
Were the well water level measurements taken
to the nearest 0.01 feet?
Were the well water levels allowed to stabilize
after construction and development for a
minimum of 24 hours prior to measurements?
Was the water ievel information obtained
from (check appropriate one):
l multiple piezometers placement in single
boreholes?
l vertically nested piezometers in closely spaced
separate boreholes?
Did the owner/operator provide construction
details for the piezometers?
How were the static water levels measured (check
method(s).
- Electric water sounder
- Wetted tape
- Air line
- Other (explain)
Was the weli water level measured in wells
drilled to an equivalent depth below the
saturated zone, or screened at an equivalent
depth below the saturated zone?
Has the owner/operator provided a site water table
(potentiometric) contour map? If yes,
l Do the potentiometric contours appear logical
based on topography and presented data?
(Consult water level data)
l Are ground-water flowlines indicated?
l Are static water levels shown?
l Can hydraulic gradients be estimated?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) -.
(Y/N) --
A-6
OSWER-9950.1
10. Did the owner/operator develop two, or more,
hydrologic cross sections of the vertical flow
component across the site?
11. Do the owner/operator's flow nets include:
0 piezometer locations?
l depth of screening?
0 width of screening?
B. Seasonal and temporal fluctuations in ground-water level
1. Do fluctuations in static water levels occur?
l If yes, are the fluctuations caused by any of
the following:
-- Off-site well pumping
-- Tidal processes or other intermittent natural
variations (e.g., river stage, etc.)
-- On-site well pumping
-- Off-site, on-site construction or changing
land use patterns
-- Deep well injection
-- Waste disposal practices
-- Seasonal variations
-- Other (specify)
2. Has the owner/operator documented the source and
patterns that contribute to or affect the ground-water
flow patterns below the waste management area?
3. Do the water level fluctuations alter the general
ground-water gradients and flow directions?
4. Based on water level data, do any head differ-
entials occur that may indicate a vertical flow
component in the saturated zone?
5. Did the owner/operator implement means for gauging
long term effects on water movement that may result
from on-site or off-site construction or changes
in land-use patterns?
C. Hydraulic conductivity
1. How were hydraulic conductivities of the subsurface
materials determined?
0 Single-well tests (slug tests)?
l Multiple-well tests (pump tests)?
2. If single-well tests were conducted, was it done
by:
- Adding or removing a known volume of water?
or
- Pressurizing well casing
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) ~
A-7
3. If single well tests were conducted in a highly
permeable formation, were pressure transducers
and high-speed recording equipment used to
record the rapidly changing water levels?
4. Since single well tests only measure hydraulic
conductivity in a limited area, were enough
tests run to ensure a representative measure
of conductivity in each hydrogeologic unit?
5. Is the owner/operator's slug or pump test data
consistent with existing geologic information
(e.g., boring logs)?
6. Were other hydraulic conductivity properties
determined?
7. If yes, provide any of the following data, if
available:
l Transmissivity
0 Storage coefficient
0 Leakage
0 Permeability
0 Porosity
l Specific capacity
l Other (specify)
D. Identification of the uppermost aquifer
1. Has the extent of the uppermost aquifer in the
facility area been defined? If yes,
l Are soil boring/test pit logs included?
l Are geologic cross-sections included?
2. Is there evidence of confining (competent,
unfractured, continuous, and low permeability)
layers beneath the site?
l If yes, was continuity demonstrated through the
evidence of lack of drawdown in the upper well
when separate, closely-spaced wells (one screened
at the uppermost part of the water table, and
the other screened on the lower side of the
confining layer) are pumped simultaneously?
3. Was hydraulic conductivity of the confinin! unit
determined by direct field measurements to be
of sufficient low permeability to prevent passage
of contaminants to saturated, stratigraphically
lower units?
4. Does potential for other hydraulic interconnect-
tion exist (e.g., lateral incontinuity between
geologic units, facies changes, fracture zones,
cross cutting structures, or chemical corrosion/
alteration of geologic units by leachate)?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) ~
(Y/NIP
(Y/N)p
(Y/NIP
(Y/N) -
(Y/N)-
(Y/N) __-
A-3
OSWER-9950.1
IV. Conclusions
A. Subsurface geology
1. Has sufficient data been collected to adequately
define petrography and petrographic variation?
2. Has the subsurface geochemistry been adequately
defined?
3. Was the boring/coring program adequate to define
subsurface geologic variation?
4. Was the owner/operator's narrative description
complete and accurate in its interpretation
of the data?
5. Does the geologic assessment address or provide
means to resolve any information gaps?
B. Ground-water flow paths
1. Did the owner/operator adequately establish the
horizontal and vertical components of ground-
water flow?
2. Were appropriate methods used to establish
ground-water flow paths?
3. Did the owner/operator provide accurate
documentation?
4. Are the potentiometric surface measurements
valid?
5. Did the owner/operator adequately consider the
seasonal and temporal effects on the ground-
water?
6. Were sufficient hydraulic conductivity tests
performed to document lateral and vertical
variation in hydraulic conductivity in the
the entire hydrogeologic subsurface
site?
below
C. Uppermost aquifer
1. Did the owner/operator adequate
uppermost aquifer?
ly def ine the
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/NIP
(Y/N)-
A-9
APPENDIX A.2
PLACEMENT OF DETECTION MONITORING WELLS WORKSHEET
The following worksheets are designed to assist the enforcement officer's
evaluation of an owner/operator's approach for selecting the number, location,
and depth of all detection phase monitoring wells. This series of worksheets
has been compiled to closely track the information presented in Chapter 2 of
the TEGD. The guide for the evaluation of an owner/operator's placement of
monitoring wells is highly dependent upon a thorough characterization of the
site hydrogeology as described in Chapter 1 of the TEGD and Appendix A.1
worksheets.
I. Placement of Downgradient Detection Monitoring Wells
A.
B.
C.
D.
E.
F.
G.
Are the ground-water monitoring wells or clusters located
immediately adjacent to the waste management area? (Y/N)-
Does the owner/operator provide a rationale for the
location of each monitoring well or cluster? (Y/N)-
Does the owner/operator provide an explanation for the
density of the ground-water monitoring wells? (Y/N)-
Has the owner/operator identified the screen length(s)
of each monitoring well or cluster? (Y/N)-
What length screens has the owner/operator employed in
the ground-water monitoring wells on site?
Does the owner/operator provide an explanation for the
screen lengths of each monitoring well or cluster?
Do the actual locations of monitoring wells or clusters
correspond to those identified by the owner/operator?
(Y/N)-
(Y/NIP
II. Placement of Upgradient Monitoring Wells ___~
A. Has the owner/operator documented the location of each
upgradient monitoring well or cluster? (Y/NIP
B. Does the owner/operator provide an explanation for the
location(s) of the upgradient monitoring wells? (Y/N)-
A-10
OSWER-9950.1
C. rWhat length screens has the owner/operator employed in
the background monitoring well(s)?
D. Does the owner/operator provide an explanation for the
screen length(s) chosen? (Y/N) ___-
E. Are the upgraident monitoring wells installed in the
same portion of the uppermost aquifer as the downgradient
monitoring wells? (Y/N)-
F. Does the actual location of each background monitoring
well or cluster correspond to that identified by the
owner/operator? (Y/N)--
III. Conclusions
A. Downgradient Wells
Do the location, number, and screen lengths of the ground-
water monitoring wells or clusters in the detection
monitoring system allow for the immediate detection
of a release of hazardous waste or constituents from the
hazardous waste management area? (Y/Nlpp
B. Upgradient Wells
Do the location and screen lengths of the upgradient
(background) ground-water monitoring wells ensure
the capability of collecting ground-water samples
representatiave of upgradient (background) ground-water
quality including any ambient heterogeneous chemical
characteristics? (Y/NJemm-
A-11
APPENDIX A.3
MONITORING WELL DESIGN AND CONSTRUCTION WORKSHEET
The following worksheets have been designed to assist the enforcement
officer in evaluating the techniques used by an owner/operator for designing
and constructing monitoring wells. This series of worksheets has been
compiled to parallel the information presented in Chapter 3 of the TEGD.
I. Monitoring Well Design -~. --. __--
A. Complete the attached well construction summary sheet for the
monitoring well unless similar documentation is already available
from the owner/operator. Include the locations where the well
intercepts changes in geological formation.
II. Drilling Methods
A. What drilling method was used for the well?
l Hollow-stem auger
l Solid-stem auger
l Cable tool
l Air rotary
l Water rotary
l Mud rotary --
l Reverse rotary
e Jetting
l Air drill with casing hammer
l Other (specify)
B. Were any drilling fluids (including water) or additives
used during drilling? (Y/N)-
If yes, specify
Type of drilling fluid --- ..-..-. ..-_.. - .-
Source of water used
Foam -.-__-- ..-- ~-.
Polrymers .~ ---- .
Other
C. Was the drilling fluid, or additive, analyzed? (Y/N)-
D. Was the drilling equipment steam-cleaned prior to drilling
the well? (Y/N)-
A-12
CSWER-9950.1
MOJCCT
¶I?f
co0RelnrTff
OATC COYPL f TfO
SuPf WlSf D Or
WELL w.
AQuInt a
filler belw plu99ed
ClcvatioP of 3OttOr Of bOre%‘r
Ucll Construction Suzarl:.
A-13
E. Was
1.
F. Did
the
1.
compressed air used during drilling?
If yes, was the air treated to remove oil (e.g.,
filtered)?
the owner/operator document procedure for establishing
potentiometric surface?
If yes, how was the location established?
(Y/N)-
(Y/NJ-
(Y/N)-
G. Formation samples
1. Were continuous formation sample cores collected
initially during drilling?
2. How were the samples obtained?
l Split spoon
l Shelby tube
l Core drill
l Other (specify)
3. Indicate the intervals at which formation samples were
collected
(Y/N)-
4. Identify if any physical and/or chemical tests were per-
formed on the formation samples (specify)
III. Monitoring Well Construction Materials
List of Potential Construction Materials for the Saturated Zone
1. Stainless steel (316, 304, 2205)
2. Fluorocarbon resins (specify)
3. Other (specify)
Teflon
A. Identify construction materials (by number) and diameters
(ID/OD)
Diameter
Material (ID/OD)
1. Primary Casing
2. Secondary or outside casing
(double construction)
3. Screen
A-14
OSWSR-9950.1
B. How are the sections of casing and screen connected?
l Pipe sections threaded
l Couplings (friction) with adhesive or solvent
l Couplings (friction) with retainer screws
l Other (specify)
C. Were the materials steam-cleaned prior to installation? (Y/N)-
Other cleaning methods (specify)
IV. Well Intake Design and Well Development
A. Was a well intake screen installed?
1. What is the length of the screen for the well?
(
2. Is the screen manufactured?
B. Was a filter pack installed?
1. Wase the material used to construct the filter pack
chemically inert? Specify the material
Y/N)
Y/N)
Y/N)
(Y/N)-
2. Has a turbidity measurement of the well water ever
been made? (Y/N)-
C. Well development
1. What technique was used for well development?
l Surge block
l Bailer
l Air surging
l Water pumping
l Other (specify)
V. Annular Space Seals
A. Is the annular space in the saturated zone directly above
the filter pack filled with?
l Sodium bentonite (specify type and grit)
l Cement (specify neat or concrete)
l Other (specify)
1. Was the seal installed by?
l Dropping material down the hole and tamping
l Dropping material down the inside of
hollow-stem auger
l Tremie pipe method
l Other (specify)
A-15
VI.
B. Was a different seal used in the unsaturated zone?
If yes,
1. Was this seal made with?
l Sodium bentonite (specify type and grit)
l Cement (specify neat or concrete)
l Other (specify)
2. Was this seal installed by?
l Dropping material down the hole and tamping
l Dropping material down the inside of
hollow-stem auger
l Tremie pipe method
l Other (specify)
C. Is the upper portion of the borehole sealed with a concrete
cap to prevent infiltration from the surface?
D. Is the well fitted with an above-ground protective device?
z. Has the protective cover been installed with locks to
prevent tampering?
Field Tests/Field Demonstration
A. Do field measurements of the following agree with
reported data:
1. Casing diameter?
2. Well depth?
3. Water level elevation?
B. If the existing well is being field demonstrated, complete
Questions 1 through 7.
1.
2.
3.
4.
5.
Is the location of the demonstration well hydraulically
equivalent to the existing well?
Was the demonstration well installed using EPA-approved
methods and materials?
How were the wells evacuated (e.g., bailer or bladder
pump)?
existing well:
demonstration well:
Were the wells sampled concurrently?
Were the wells each sampled using the appropriate EPA
methodology?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
A-16
OSWER-9950.1
6. What parameters were the ground water samples analyzed
for?
7. Are the values for these parameters equivalent for each
well (i.e., within the acceptable standard deviations)? (Y/N)-
VII. Conclusions
A. Do the design and construction of the owner/operator's
ground-water monitoring wells permit depth discrete ground-
water samples to be taken? (Y/N)-
B. Are the samples representative of ground-water quality? (Y/N)-
C. Are the ground-water monitoring wells structurally stable? (Y/N)-
D. Does the ground-water monitoring well's design and con-
struction permit an accurate assessment of aquifer
characteristics? (Y/N)
A-17
APPENDIX A.4
SAMPLING AND ANALYSIS WORKSHEET
The following worksheets have been designed to assist the enforcement
officer in evaluating the techniques an owner/operator uses to collect and
analyze ground-water samples. This series of worksheets has been compiled
based on the information provided in Chapter 4 of the TEGD.
I. Review of Sample Collection Procedures
A. Measurement of well depths elevation:
B.
C.
D.
E.
1. Are measurements of both depth to standing water
and depth to the bottom of the well made?
2. Are measurements taken to the nearest centimeter
or 0.01 foot?
3. What device is used?
---.
4. Is there a reference point(s) established by a
licensed surveyor?
Detection of immiscible layers:
1. Are procedures used which will detect light phase
immiscible layers?
2. Are procedures used which will detect dense phase
immiscible layers?
Sampling of immiscible layers:
1. Are the immiscible layers sampled separately prior to
well evacuation?
2. Do the procedures used minimize mixing
with water soluble phase?
Well evacuation:
1. Are low yielding wells evacuated to dryness?
2. Are high yielding wells evacuated so that at least
three casing volumes are removed?
3. What device is used to evacuate the wells?
.~
4. If any problems are encountered (e.g., equipment
malfunction) are they noted in a field logbook?
Sample withdrawal:
1. For low-yielding wells, are first samples tested for
pH, temperature, and specific conductance after the
well recovers?
(Y/N)
(Y/N) __~
(Y/N)
(-f/N)
(Y/N)
(Y/N)
(Y/N). .--
(Y/N) _
(Y/N)--.
OSWER-9950.1
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Are samples collected and containerized in order of
the parameters volatilization sensitivity?
For higher-yielding weils, are samples retested for
pH, temperature, and specific conductance to determine
purging efficiency?
Are samples withdrawn with either fluorocarbon resins
or stainless steel (304, 316, 2205) sampling devices?
Are sampling devices either bottom valve bailers
or positive gas displacement bladder pumps?
If bailers are used, is fluorocarbon resin-coated wire,
single strand stainless steel wire, or monofilament
used to raise and lower the bailer?
If biadder pumps are used, are they operated in a
continuous manner to prevent aeration of the sample?
If bailers are used, are they lowered slowly to
prevent degassing of the water?
If bailers are used, are the contents transferred
to the sample container in a way that will minimize
agitation and aeration?
Is care taken to avoid placing clean sampling equipment
on the ground or other contaminated surfaces prior to
insertion into the well?
If dedicated sampling equipment is not used, is
equipment disassembled and thoroughly cleaned between
sampies?
If samples are for inorganic analysis, does the clean-
ing procedure include the following sequential steps:
a. Nonphosphate detergent wash?
b. Dilute acid rinse (HNO3 or HCl)?
c. Tap water rinse?
d. Type II reagent grade water?
If samples are for organic analysis, does the cleaning
procedure include the following sequential steps:
a. Nonphosphate detergent wash?
b. Tap water rinse?
C. Distilled/deionized water rinse?
d. Acetone rinse?
e. Pesticide-grade hexane rinse?
Is sampling equipment thoroughly dry before use?
Are equipment blanks taken to ensure that sample
cross-contamination has not occurred?
If volatile samples are taken with a positive gas
displacement bladder pump, are pumping rates below
100 ml/min?
(Y/N) ---
(Y/N) __~
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) ___-
(Y/N) --
(Y/N) --
(Y/NIP
(Y/N) --
(Y/N)-
(Y/N) --
(Y/N)-
(Y/N)-.
(Y/N)-
(Y/N)
(Y/N) ___
(Y/N)-
A-19
F. In-situ or field analyses:
1. Are the following labile (chemically unstable) parameters
determined in the field:
a. pH?
b. Temperature?
C. Specific conductivity?
d. Redox potential?
e. Chlorine?
f. Dissolved oxygen?
g- Turbidity?
h. Other (specify)
2. For in-situ determinations, are they made after well
evacuation and sample removal?
3. If sample is withdrawn from the well, is parameter
measured from a split portion?
4. Is monitoring equipment calibrated according to
manufacturers' specifications and consistent with
SW-846?
5. Is the date, procedure, and maintenance for equipment
calibration documented in the field logbook?
II. Review of Sample Preservation and Handling Procedures
A. Sample containers:
1. Are samples transferred from the sampling device
directly to their compatible containers?
2. Are sample containers for metals (inorganics) analyses
polyethylene with polypropylene caps?
3. Are sample containers for organics analysis glass
bottles with fluorocarbon resin-lined caps?
4. If glass bottles are used for metals samples are
the caps fluorocarbon resin-lined?
5. Are the sample containers for metal analyses cleaned
using these sequential steps?
it
Nonphosphate detergent wash?
1:l nitric acid rinse?
C. Tap water rinse?
d. 1:l hydrochloric acid rinse?
e. Tap water rinse?
f. Type II reagent grade water rinse?
6. Are the sample containers for organic analyses cleaned
using these sequential steps?
a. Nonphosphate detergent/hot water wash?
b. Tap water rinse?
C. Distilled/deionized water rinse?
d. Acetone rinse?
e. Pesticide-grade hexane rinse?
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)__
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)___
(Y/N)-
(Y/NIF
(Y/N)-
(Y/N)-
(Y/N) ---
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
A-20
OSWER-9950.1
7. Are trip blanks used for each sample container type
to verify cleanliness? (Y/N)-
B. Sample preservation procedures:
1. Are samples for the following analyses cooled to 4°C:
;:
TOC? (Y/N)-
TOX? (Y/N)-
c. Chloride? (Y/N)-
d. Phenols? (Y/N)-
e. Sulfate? (Y/N)-
f. Nitrate? (Y/N)-
g- Pesticides/Herbicides? (Y/N)
h. Coliform bacteria? (Y/N)-
i. Cyanide? (Y/N)-
.
ii.
Oil and grease? (Y/N)
Volatile, semi-volatile, and nonvolatile organics? (Y/N)-
2. Are samples for the following analyses field acidified to
pH <2 with HNO3:
a. Iron?
b. Manganese?
C. Sodium?
d. Total metals?
e. Dissolved metals?
f. Radium?
9. Gross alpha?
h. Gross beta?
3. Are samples for the following analyses field acidified
to pH <2 with H2SO4:
a. Phenols?
b. Oil and grease?
4. Is the sample for TCC analyses field acidified to
pH <2 with H2SO4 or HCl?
5. Is the sample for TOX analysis preserved with
1 ml of 1.1 M sodium sulfite?
6. Is the sample for cyanide analysis preserved with
NaOH to pH >12?
7. Are pesticides pH adjusted to between 6 and 8 with
NaOH or H2SO4?
C. Special handling considerations:
1. Are organic samples handled without filtering?
2. Are samples for volatile organics transferred to
the appropriate vials to eliminate headspace over
the sample?
3. Are samples for metal analysis split into two
portions?
4. Is the sample for dissolved metals filtered
through a 0.45 micron filter?
(Y/N) ___
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
A-21
5. 1s the second portion not filtered and analyzed
for total metals?
6. Is one equipment blank prepared each day of
ground-water sampling?
III. Review of Analytical Procedures
A. Laboratory analysis procedures:
1. Are all samples analyzed using an EPA-approved
method (SW-846)?
2. Are appropriate QA/Qc measures used in laboratory
analysis (e.g., blanks, spikes, standards)?
3. Are detection limits and percent recovery (if
applicable) provided for each parameter?
4. If a new analytical method or laboratory is used,
are split samples run for comparison purposes?
5. Are samples analyzed within specified holding
times?
B. Laboratory logbook:
1. Is a laboratory logbook maintained?
2. Are experimental conditions (e.g., temperature,
humidity, etc.) noted?
3. If a sample for volatile analysis is received
with headspace, is this noted?
4. Are the results for all QC samples identified?
5. Is the time, date, and name of person noted
for each processing step?
IV. Review of Chain-of-Custody Procedures
A. Sample labels:
1. Are sample labels used?
2. Do they provide the following information:
a. Sample identification number?
b. Name of collector?
c. Date and time of collection?
d. Place of collection?
e. Parameter(s) requested:
3. Do they remain legible even if wet?
B. Sample seals:
1. Are sample seals placed on those containers to
ensure the samples are not altered?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
A-22
OSWER-9950.1
C. Field logbook:
1. Is a field logbook maintained?
2. Does it document the following:
a. Purpose of sampling (e.g., detection or
assessment)?
b. Identification of well?
c. Total depth of each well?
d. Static water level depth and measurement
technique?
e. Presence of immiscible layers and
detection method?
f. Collection method for immiscible layers
and sample identification numbers?
g. Well yield - high or low?
h. Purge volume and pumping rate?
i. Time well purged?
j. Well evacuation procedures?
k. Sample withdrawal procedure?
1. Date and time of collection?
m. Well sampling sequence?
n. Types of sample containers and sample
identification numbers?
o. Preservative(s) used?
p. Parameters requested?
q. Field analysis data and method(s)?
r. Sample distribution and transporter?
s. Field observations?
l Unusual well recharge rates?
l Equipment malfunction(s)?
l Possible sample contamination?
l Sampling rate?
t. Field team members?
U. Climatic conditions and air temperature?
D. Chain-of-custody record:
1. Is a chain-of-custody record included with
each sample?
2. Does it document the following:
a. Sample number?
b. Signature of collector?
c. Date and time of collection?
d. Sample type?
e. Identification of well?
f. Number of containers?
g. Parameters requested?
h. Signatures of persons involved in the
chain-of-possession?
i. Inclusive dates of possession?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
A-23
E. Sample analysis request sheet:
1. Does a sample analysis request sheet accompany
each sample? (Y/N)-
2. Does the request sheet document the following:
a. Name of person receiving the sample? (Y/N)-
b. Date of sample receipt? (Y/N)-
c. Laboratory sample number (if different than
field number)? (Y/N)-
d. Analyses to be performed? (Y/N) -
T. Laboratory logbook:
1. Is a laboratory logbook maintained? (Y/N)-
2. If so, does it document the following:
a. Sample preparation techniques (e.g., extraction)? (Y/N)-
b. Instrumental methods? (Y/N)-
c. Experimental conditions? (Y/N)-
V. Review of Quality Assurance/Quality Control
A. Is the validity and reliability of the laboratory and
field generated data ensured by a QA/QC program?
a. Dwr tha @/QC program include:
1. Docwentation of any deviations from approved
procedures?
2. Collection and analysis of trip blanks and
equipment blanks?
3. Documentation of analytical results for:
a. Laboratory blanks?
b. Standards?
c. Duplicates?
d. Spiked samples?
C. Are approved statistical methods used?
D. &a Qc rumples used to correct data?
E. Are all data critically examined to ensure it
has been properly calculated and reported?
VI. Review of Indicators of Data Quality
A. Reporting of low and zero concentration values:
1. Do specific concentration values accompanying
measurements reported as less than a limit of
detection?
2. Is the magnitude of detection limits consistent
throughout the data set for each parameter?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) __
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
A-24
ZSWER-9950.1
3. Have techniques described in Appendix B of
40 CFR §136 been used to determine the detection
limits?
4. Has the method for using less than detection
limit data in presentations and statistical
analysis been documented?
B. Significant digits:
1. Are constituent concentrations reported with
a consistent number of significant digits?
2. Are all indicator parameters reported with
at least three significant digits?
C. Missing data values:
1. Is the monitoring data set complete?
2. Are t-test comparisons between upgradient and
downgradient wells attempted despite missing
data provided that:
a. At least one upgradient and one downgradient
well were sampled?
b. In the case of a missing quarterly
sampling set, values are assigned by
averaging corresponding values for
the other three quarters?
c. In the case of missing replicate values
from a sampling event, values are assigned
by averaging the replicate(s) which are
available for that sampiing event?
D. Outliers:
1. Have extreme values (outliers) of constituent
concentrations deleted or otherwise modified
because of:
a. Incorrect transcription?
b. Methodological problems or an unnatural
catastrophic event?
c. Are these above occurrences fully
documented?
2. Are true but extreme values unaltered and
incorporated in the analysis?
E. Units of measure:
1. Are all units of measure reported accurately?
2. Are the units of measure for a given chemical
parameter used consistently throughout the
report?
(Y/N)
(Y/N)
(Y/N) --
(Y/N)
(Y/N) --
(Y/N)
(Y/N)
(Y/N) .-.
(Y/N)
(Y/NIP.,
(Y/N)-
(Y/N)-....
(Y/N)
(Y/N)-
3. Do the reporting formats clearly indicate
consistent units of measure throughout so that
no ambiguity exists (i.e., do the units
accompany each parameter instead of a
statement, "all values are ppm unless
otherwise stated")?
VII. Conclusions
A. Does the sampling and anaiysis plan permit the owner/
operator to detect and, where applicable, assess the
nature and extent of a release of hazardous constituents
to ground water from the monitored hazardous waste
management facility?
(Y/N) ~_
(Y/NIP
A-26
OSWER-9950.1
APPENDIX A.5
PRESENTING DETECTION MONITORING DATA WORKSHEET
The following worksheets have been designed to assist the enforcement
official in evaluating the method an owner/operator uses in presenting and
statistically analyzing detection monitoring data. This series of worksheets
has been compiled to parallel the information provided in Chapter 5 of the
TEGD.
I. Presenting Detection Monitoring Data
A. Is the owner/operator using the data reporting sheets
as described in the TEGD (Chapter 5)?
B. Have all the detection monitoring data collected by the
facility been obtained and reviewed?
(Y/N)-
(Y/N)-
II. T-test and Number of Wells
A. Which t-test is in use:
1. Cochran's Approximation to the Behrens-Fisher
(CABF t-test)?
2. Averaged replicate t-test (AR t-test)?
3. Other, describe:
B. Does the facility have more than one upgradient monitoring
well? (Y/N)-
III. First Year's Data
A. Have upgradient wells been monitored to establish background
concentrations of the following data on a quarterly basis for
one year:
1. Appendix III parameters (5265.92(b)(l))? (Y/N)
Ground-water quality parameters (§265.92(b)(2))?
--
2. (Y/N)-
3. Ground-water contamination indicator parameters
(5265.92(b)(3))? (Y/N)-
B. Were four replicate measurements obtained from each
upgradient well during the first year of quarterly detec-
tion monitoring for indicator parameters [§265.92(b)(3)]? (Y/N) _-- -
r -. Have the background mean and variance been determined for
the 5265.92(b)(3) parameters using ail the data obtained
from the upgradient wells during the first year of sampling? (Y/N)
A-27
D. Are background statistics determined from missing data
using the criteria discussed in Chapter Four?
IV. Subsequent Year's Data
A. Is monitoring data collected after the first year being
compared with background data to determine possible
groundwater contamination?
B. Is the identified approved t-test being used properly to
determine possible ground-water contamination?
C. Are the ground-water quality parameters in §265,92(b)(2)
being measured at least annually?
D. Are the indicator parameters in §265.92(b)(3) being
measured in at least four replicate samples from each
well in the detection monitoring network at least
semi-annually?
z. Are the indicator parameters coliected on a semi-annual
basis being used to estimate the mean and variance?
F. Is the elevation of the water table at each monitoring
well determined each time a sample is collected?
V. Conclusions
A. Is the owner/operator adequately reporting and statis-
tically analyzing the facility's monitoring well data?
B. If the t-test indicated a significant increase in IP's for
downgradient wells, were they resampled and reanalyzed?
C. If the resamplifig still indicated a significant increase,
was assessment monitoring begun?
(Y/N) __
U/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N
(Y/N
(Y/N )
A-23
APPENDIX A.6
ASSESSMENT MONITORING
The following worksheets have been designed to assist the enforcement
officer in evaluating an owner/operator's assessment phase ground-water
monitoring program. This series of worksheets has been compiled to mrallel
the information presented in Chapter 6 of the TEGD.
I. Review of Hydrogeologic Descriptions
A. Has the site's hydrogeologic setting been well characterized
(refer to Appendix A.1 of TEGD)?
1.
2.
3.
4.
5.
6.
7.
Has the regional and local hydrogeologic setting
been thoroughly described?
Is there sufficient direct field information?
Is the information accurate and reliable?
Was the evaluation performed by a hydrogeologist?
Did indirect investigatory methods correlate with
direct methods?
Have all possible migration pathways been identified?
Will the description of the hydrogeologic setting aid
in characterizing the rate and e:ctent of the plume
migration?
II. Review of Detection Monitoring System Description
A. Is the detection monitoring system capable of detecting
all contaminant leakage that may be escaping from the
facility (refer to Appendix A.2 of TEGD)?
1. Are the well designs and construction parameters
fully documented?
2. Have the downgradient wells been strategically
located so as to intercept migrating contaminants?
3. Are upgradient wells positioned so that they are
not effected by the facility?
4. What are the screened intervals?
5. Are the well construction materials (e.g., casing,
screen, seals, packing) comprised of material that
will not affect the ground-water quality?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)
(Y/N)
A-29
III. Review of Description of Approach for Making First Determination -.-.- - --.
A. Did the detection monitoring system consistently yield
statistically equivalent concentrations for ali indicator
parameters?
B. If no:
1. Were the results based on the Student's t-test at the
0.01 level of significance? (Single-tailed t-test for
testing significant increases and two-tailed t-test
for testing significant differences in pH values.)
2. Were the calculations performed correctly?
3. If the results are deemed as a false positive, did
the owner/operator fully document the reasoning?
4. Is there any reasonable cause to believe that faulty
data are responsible for the false positive claim?
5. Can or will deficiencies in well design, sample
collection, sample preservation, or analysis be
corrected?
6. If the owner/operator intends to collect additional
data to remedy any inadequacies, will this collection
result in an acceptable delay in assessing the extent
of contamination at the site?
7. Will positive results of these determinations initiate
a driiling program for assessment monitoring?
IV. Review of Approach for Conducting Assessment
A. Have the assessment monitoring objectives been clearly
defined in the assessment plan?
1. Does the plan include analysis and/or re-evaluation
to determine if significant contamination has occurred
in any of the detection monitoring wells?
2. Does the plan provide for a comprehensive program of
investigation to fully characterize the rate and
extent of contaminant migration from the facility?
3. Does the plan call for determining the concentrations
of hazardous wastes and hazardous waste constituents
in the ground water?
4. Does the plan employ a quarterly monitoring program?
B. Does the assessment plan identify the investigatory
methods that will be used in the assessment phase?
1. Is the role of each method in the evaluation fully
described?
(Y/N) --
(Y/N)
(Y/N) --_
(Y/N) --
(Y/N)
(‘f/N) __
(Y/N)-
(Y/N)
(Y/N)-
(Y/N) _-
(Y/N) __-
(Y/N) __
(Y/N)-
(Y/N)-
(Y/N)-
A-30
OSWER-995'0.1
2. Does the plan provide sufficient descriptions of the
direct methods to be used?
3. Does the plan provide sufficient descriptions of the
indirect methods to be used?
4. Will the method contribute to the further characteri-
zation of the contaminant movement?
C. Are the investigatory techniques utilized in the assess-
ment program based on direct methods?
1. Does the assessment approach incorporate indirect
methods to further support direct methods?
2. Will the planned methods called for in the assessment
approach ultimately meet performance standards for
assessment monitoring?
3. Are the procedures well defined?
4. Does the approach provide for monitoring wells similar
in design and construction as the detection monitoring
wells?
5. Does the approach employ taking samples during drill-
ing or collecting core sampies for further analysis?
D. Are the indirect methods to be used based on reliable
and accepted geophysical techniques?
1.
2.
3.
4.
5.
Are they capable of detecting subsurface changes
resulting from contaminant migration at the site?
Is the measurement at an appropriate level of
sensitivity to detect ground-water quality changes
at the site?
Is the method appropriate considering the nature
of the subsurface materials?
Does the approach consider the limitations of
these methods?
Will the extent of contamination and constituent
concentration be based on direct methods and sound
engineering judgment? (Using indirect methods to
further substantiate the findings)
E. Does the assessment approach incorporate any mathematical
modeling to predict contaminant movement?
1. Will site specific measurements be utilized to
accurately portray the subsurface?
2. Will the derived data be reliable?
3. Will the model be adequately calibrated with
observed physical conditions?
4. Have the assumptions been identified?
5. Have the physical and chemical properties of the
site-specific wastes and hazardous waste constituents
been identified?
(Y/N)
(Y/N)
(Y/N) -.
(Y/N)-
(Y/N)-.
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-.
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
f-f/N)-
(Y/N)
(Y/N)-
(Y/Nlpmm
A-3:
v. Review of Assessment Monitoring Wells
A. Does the assessment plan specify:
1. The number, location, and depth of wells?
2. The rationale for their placement and identify the
basis that will be used to select subsequent sampling
locations and depths in later assessment phases?
B. Does the assessment period consist of a phased investiga-
tion so that data gained in initial rounds may help guide
subsequent rounds?
1. Do initial rounds incorporate geophysical techniques
to approximate the limits of the contaminant plume?
2. Has information from the triggering well (well show-
ing elevated contaminant concentrations) been incor-
porated in the initial design and specifications?
3. Is the sampling program designed adequately to portray
a three dimensional plume configuration?
4. Are evaluation procedures in place that will provide
further guidance for subsequent monitoring?
C. Does sufficient hydrogeologic data exist in the direction
of the contaminant plume?
1. Does the subsurface setting provide any information
on possible transport mechanisms and attenuation
processes?
2. Are provisions made to secure additional data as
needed?
3. Are hydrogeologic descriptions updated as additional
data become available?
D. Sampling density:
1.
2.
3.
4.
5.
-1s the number of monitoring well clusters sufficient
to define the horizontal boundaries of the plume?
Are the well clusters placed both perpendicular and
parallel to plume migration from the triggering well?
Are the well clusters placed both inside and outside
the contaminant plume to identify its horizontal
boundaries?
Are sampling locations situated so as to identify
areas of maximum contaminant concentration within
the plume?
Does the sampling density correlate with the size
of the plume and the geologic variability?
(Y/N)--
(Y/N) ___
(Y/N) ___
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) -_
(Y/N)-
(Y/N)
(Y/N)
A-32
OSWER-9950.1
E. Sampling depths:
1.
2.
3.
4.
5.
6.
7.
8.
Are the intervals over which the samples are collected
clearly identified?
Are the well screens within each cluster positioned
to sample the full extent of the predicted vertical
distribution of hazardous waste constituents?
Are the well screens depth discrete to the extent
possible to minimize dilution effects?
Are there sufficient wells in each cluster to
verbally define plume margins?
Are there wells within each cluster that are
screened within the plume?
Are the wells placed alternating lower and higher
screened wells to reduce the effect of drawdown on
the sampling horizons?
Are there high fluctuations in ground-water levels,
or is the subsurface characterized by fractured
consolidated formations that may otherwise require
longer screen lengths?
Are the wells screened to identify vertical concen-
tration gradients and maximum concentrations of the
contaminants?
VI. Review of Monitoring Well Desiqn and Construction
A. Are the well design and construction specification require-
ments equivalent to the detection requirements detailed in
Chapter 3? (Y/N)
1s provided for: B. Are well design and construction detai
1. Drilling methods?
2. Well construction materials?
3. Well diameter?
4. Well intake structures and procedu
development?
5. Placement of annular seals?
res for well
C. Are all these details approved and recommended considering
the characteristics of the site?
VII. Review of Sampling and Analysis Procedures
A. Does the list of monitoring parameters include all
hazardous waste constituents from the facility?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)
(Y/N)-
(Y/N)
(Y/N)
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
A-33
1. Does the water quality parameter list include other
important indicators not classified as hazardous
waste constituents?
2. Does the owner/operator provide documentation for
the listed wastes which are not included?
B. Have the procedures been detailed for sample collection?
1. Do the procedures include evacuation of the borehole
prior to sample collection?
2. Are special procedures delineated for collection of
separate phase immiscible contaminants?
3. Has the equipment been identified?
4. Do the procedures include decontamination of equipment
5. :!-ive pumping rates, duration. and position in the well
from which water will be evacuated been specified?
C. Do the procedures include Provisions for sample preser-
vation and shipment?
D. Do the procedures specify:
1. Type of sample containers?
2. Filtering procedures?
3. Preservation techniques?
4. Storage and time elements involved?
5. Proper documentation?
E. Do these procedures correspond to recommended procedures
(SW-846 or EPA-approved procedures) for sampling and
preservation?
F. Do the sampling and analysis procedures identify analyti-
cal Procedures for each of the identified monitoring
parameters?
G. Do the analytical procedures include:
1. Detailed description and reference of approved
analyticai methods?
2. QA/QC procedures?
3. Location of laboratory performing analysis?
4. Proper documentation?
H. Does the sampling and analysis plan establish procedures
for chain of custody control?
I. Do these procedures include:
1. Sample labels?
2. Sample seals?
3. Field logbook?
4. Chain of custody record?
5. Sample analysis request sheet?
6. Laboratory logbook?
?
(Y/N)
(Y/NJ --
(Y/N)
(Y/N) _ --.-
(Y/N)
(Y/N)
(Y/N) --
(Y/N)
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
(Y/N
1
)
1
A-34
OSWER-9950.1
J. Do the procedures specify how assessment monitoring data
will be evaluated to determine if contamination has
actually occurred?
1. Will the evaluation delineate the full extent of
contaminant migration?
2. Will significant changes in containment concentration
or movement be identified?
3. Are the evaluation procedures suitable and objective?
K. Does the assessment plan clearly describe the procedures
that will be used for evaluating monitoring data during
the assessment?
L. Does the plan provide for evaluation of its methodologies
to ensure each method is properly executed during the
assessment period?
M. Is a list of all detection monitoring and assessment monitor-
ing (if applicable) data available from the owner/operator?
1. Do these lists include:
l Field quality control samples (e.g., sample container
and equipment blanks)?
l Laboratory quality control samples (e.g., replicates,
spiked samples, etc.)?
l Method detection limits?
2. Are the lists prepared using a format which presents:
l Codes that identify GWCCs?
l Well number?
l Date?
l Units of measure?
l Less than (LT) detection limit values?
l Concentrations of GWCCs?
N. Has the owner/operator prepared summary statistics tables
of the GWCC data?
1. Do the summary statistics tables include:
l Number of LT detection limit values?
l Total number of values?
l Mean?
l Median?
l Standard deviation?
l Coefficient of variation?
l Minimum value?
l Maximum value?
2. Are there summary statistics tables that present:
l GWCC?
l GWCC by well number?
l GWCC by well number and date?
l Quality control data?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)--.-
(Y/N)-
(Y/NIP
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(YINIF
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) --
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/NIP
(Y/N)-
A-35
0. Has the owner/operator simplified the statistical data?
1. Was the data simplified using a ranking procedure for
each GWCC-well combination?
2. Has the ranking procedure been applied to each GWCC
which was detected at least once at every well in the
monitoring system?
P. Did the owner/operator display the data graphically?
1. Were the data plotted graphically to evaluate
temporal changes?
2. Were the data plotted on facility maps to evaluate
spatial trends?
VIII. Review of Migration Rates
A. Did the owner/operator's assessment plan specify the pro-
cedures to be used to determine the rate of constituent
migration in the ground-water?
B. Do the procedures incorporate a periodic re-evaluation of
sampling data to continually monitor the rate and extent
of
1.
2.
3.
4.
5.
6.
contaminant migration? -
Do the procedures clearly establish ground-water flow
rates and direction downgradient from the detection
wells?
Are the methods employed suitable for these determina-
tions?
Are the limitations of these methods known and
documented?
Do the evaluations incorporate chemical and physical
characteristics of the contaminants and the media?
Are adsorptive and degradative processes considered
in determining any retardation of contaminant movement?
Have the assumptions been identified and documented?
C. Does the assessment plan evaluate the presence of
immiscible phase layers?
1. Do the procedures specify detection and collection
of light and dense phase immiscibles prior to well
evacuation?
2. Has the owner/operator used the slope of the water
table and the velocity of ground-water flow to estimate
light phase immiscible migration?
3. Has the owner/operator defined the configuration of
the confining layer to predict dense phase immiscible
migration?
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)
(Y/N)-
(Y/N)-
(Y/N)-
(Y/NIP
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N)-
(Y/N) -
A-36
OSWER-9950.1
IX. Reviewing Schedule of Implementation
A. Has the owner/operator specified a schedule of implementa-
tion in the assessment plan?
B. Does the schedule for implementing assessment monitoring
data include a timetable for a comprehensive site evalua-
tion for contamination?
C. Does the timetable include:
1. A number of milestones used to judge if sufficient
progress is being made toward the completion of the
assessment during implementation?
2. The determination if contamination has occurred?
3. Completing an initial comprehensive assessment of
contamination at the site?
4. Implementing a program for continued monitoring after
fully characterizing contamination at the site?
D. Does this represent an acceptable time frame?
X. Conclusions
A.
B.
C.
D.
E.
F.
G.
H.
Has the owner/operator adequately characterized site
hydrogeology to determine contaminant migration?
Is the detection monitoring system adequately designed
and constructed to immediately detect any contaminant
release?
Are the procedures used to make a first determination of
contamination adequate?
Is the assessment plan adequate to detect, characterize,
and track contaminant migration?
Will the assessment monitoring wells, given site hydro-
geologic conditions, define the extent and concentration
of contamination in the horizontal and vertical planes?
Are the assessment monitoring wells adequately designed
and constructed?
Are the sampling and analysis procedures adequate to
provide true measures of contamination?
Do the procedures used for evaluation of assessment
monitoring data result in determinations of the rate of
migration, extent of migration, and hazardous constituent
composition of the contaminant plume?
(Y/N)-
(Y/N)-
(Y/N
(Y/N
(Y/N
(Y/N
1
1
(Y/N)-
(Y/N
(Y/N
I-.-
)-
(Y/N)-
(Y/N)-
(Y/NJ-
(Y/N)-
(Y/N)-
(Y/N)-
A-37
I. Are the data collected at sufficient duration and frequency
to adequately determine the rate of migration? (Y/N)-
J. Is the schedule of implementation adequate? (Y/N)-
K. Is the owner/operator’s assessment monitoring plan adequate? (Y/N)-
1. If the owner/operator had to implement his assessment
monitoring plan, was it implemented satisfactorily? (Y/N)-
A-38
OSWER-9950.1
APPENDIX B
A STATISTICAL PROCEDURE FOR ANALYZING
INTERIM STATUS DETECTION MONITORING DATA:
METHODOLOGY AND APPLICATION
OSWER-9950.1
APPENDIX B
A STATISTICAL PROCEDURE FOR ANALYZING INTERIM STATUS
DETECTION MONITORING DATA: METHODOLOGY AND APPLICATION
1.0 INTRODUCTION
This appendix describes a statistical methodology for evaluating
ground-water data collected under Subpart F of 40 CFR § 265. The
methodology is presented in the context of an example data set from an
idealized RCRA facility subject to the interim status ground-water
monitoring requirements. The data structures were designed to illustrate
several characteristics of RCRA interim status ground-water concentration
data. The data presented in this appendix are more extensive over time
and space than the data available from most RCRA facilities. It is used
here to illustrate the importance of an extensive and rigorous data
collection program and because it is easier to simplify a detailed
example than to design details based on a simple example.
Enforcement officials should understand that a proper statistical
analysis and evaluation protocol involves more than a simple calculation
procedure and that decisions must be made during the course of conducting
preliminary data analyses, exploration, and summary. To help with the
preparatory analyses, Appendix B offers a series of preliminary procedures
which provide guidance on data characterization and summary, evaluation
of the background data distribution, and methods for confronting a variety
of data structure features including values less than (LT) a limit of
detection, seasonal fluctuations in concentration, and violation of the
assumptions required for the t-test.
2.0 DATA DESCRIPTION, PREPARATION, AND SUMMARY
2.1 Data Description
The data analyzed in this example include measurements of total
organic carbon (TOC) in parts per million (ppm) and total halogenated
B-1
organics (TGX) in parts per billion (ppb) from four upgradient wells and
six downgradient wells. Background ground-water quality was characterized
by sampling the four upgradient wells bimonthly for a year. The down-
gradient and upgradient wells were sampled quarterly after the first
year. This example includes data from the background characterization
period and one quarterly sampling episode that was conducted after the
background characterization. Four replicate measurements were obtained
for every chemical parameter each time a well was visited for sampling.
Table 1 is a listing of the TOX and TCC data used to characterize the
background ground-water quality, and Table 2 is a listing of the data
obtained during a subsequent quarterly sampling.
2.2 Data Preparation
2.2.1 Averaging the Replicate Measurements
Prior to further evaluation, the data should be prepared for
analysis by taking the average of the replicate measurements from each
well. The averaging of the replicate measurements is the first step
required for the averaged replicate t-test.
The methodology for averaging the replicates depends on how many of
the four replicate measurements are LT detection limit values. If all of
the values measured are LT a limit of detection, then the replicate
average value assigned to the well for that sampling period is LT the
limit of detection. However, if none of the replicate concentration
measurements from a well are LT a limit of detection, then the simple
averaging method described in Table 3 can be applied. The most difficult
situation is when the replicate measurements consist of a mixture of
values that are greater than or equal to a limit of detection and values
that are LT a limit of detection. In this instance, Cohen's Method,
which is referenced in Chapter Four, may be appropriate. Cohen's Method
assumes that the data are selected from a normally distributed population
and only requires calculation of the mean and variance of the values
B-2
OSWER-9950.1
TABLE 1
A LISTING OF THE TOTAL ORGANIC CARBON (TCC) AND TOTAL
HALOGENATED ORGANIC (TOX) BACKGROUND DATA FROM FOUR
UPGRADIENT WELLS SAMPLED BIMONTHLY FOR A YEAR
Month Well Replicate TOC
@Pm)
TOX
(ppb)
1 1 A 60.3 (10.0
B 60.9 <lO.O
r
; 61.2 60.7 (10.0 (10.0
2 A 58.3 15.2
B 58.2 13.4
C 58.0 18.0
D 58.4 <lO.O
3 A 61.4 22.0
B 61.5 16.2
C 61.4 16.3
D 61.0 15.9
A 64.2 13.0
B 64.0 13.9
C 63.2 13.7
D 63.3 13.8
A 63.2 11.0
B 63.2 12.2
C 63.4 <lO.O
D 64.0 <lO.O
A 59.9 12.4
B 60.1 13.3
C 59.7 16.6
D 59.7 11.9
A 61.4 18.4
B 61.8 17.0
C 61.3 19.2
D 62.0 19.9
A 65.7 13.8
B 66.1 13.9
C 65.8 13.0
D 65.9 13.2
(Continued)
B-3
TABLE 1 (Continued)
A LISTING 3F THE TOTAL ORGANIC CARBON (TOC) AND TOTAL
HALCGENATED ORGANIC (TOX) BACKGROUND DATA FROM FOUR
UPGRADIENT WELLS SAMPLED BIMONTHLY FOR A YEAR
Month Well Replicate
5 1 A
B
C
D
7
4
1
A
B
C
D
A
B
8"
;;
A
B
C
D
TCC
(mm) ^_ ..-- -.
70.2
71.8
69.9
69.8
TOX
(wb)
11.8
12.0
(10.9
<lo.3
62.0 14.3
62.7 20.0
62.0 13.6
62.2 14.2
63.8 21.2
62.0 20.8
63.2 21.8
63.4 20.8
65.5 (10.0
65.5 (10.0
65.4 14.0
65.0 14.1
69.2 (10.3
68.4 (10.0
68.8 (10.0
69.0 12.0
59.7 16.0
59.2 17.0
59.1 17.0
60.0 21.0
61.2 la.9
61.1 17.7
61.5 18.2
61.7 17.0
64.0 (10.0
64.1 <lO.O
64.3 13.7
64.6 13.3
(Continued)
3-4
OSWER-9950.1
TABLE 1 (Continued)
A LISTING OF' THE TOTAL ORGANIC CARBON (TOC) AND TOTAL
HALCGENATED ORGANIC (TOX) BACKGROUND DATA FROM FOUR
UPGRADIENT WELLS SAMPLED BIMONTHLY FOR A YEAR
Month Well Replicate TCC
(PPm)
TOX
(ppb)
3 A
B
c
D
4 A 63.3 (10.0
B 63.7 12.3
c 63.4 13.8
D 63.5 12.4
11 1
4
66.7 12.2
65.9 <lO.O
66.2 12.0
66.2 12.7
57.7 15.7
57.9 14.9
57.8 15.2
57.7 13.7
61.0 19.9
60.5 15.4
60.2 14.8
60.5 16.3
62.9 (10.0
62.8 (10.0
62.4 13.3
62.0 13.8
58.2 14.7
58.3 14.6
58.1 14.3
58.3 14.6
60.7 21.7
60.0 21.4
60.4 21.5
60.4 21.5
61.6 13.8
61.6 12.0
61.9 12.3
62.0 12.2
B-5
TABLE 2
AN EXAMPLE OF TOX AND TOC DATA COLLECTED DURING A SEMIANNUAL
MONITORING EPISODE AFTER THE FIRST YEAR OF BACKGROUND MONITORING
Well Location Replicate TCC TOX
(PPm1 (wb)
1 Upgradient
2 Upgradient
3 Upgradient
A 71.7 11.4
B 72.3 15.3
C 70.9 11.2
D 72.4 12.8
A 62.9 24.7
B 64.7 23.8
C 63.0 21.4
D 63.2 27.8
A 62.9 19.4
B 64.2 18.6
C 63.5 19.2
D 63.4 19.0
A
B
C
D
64.8 (10.0
64.3 (10.0
64.8 <lO.O
64.8 11.2
A 69.3 18.2
B 68.4 18.3
C 67.9 18.1
D 68.5 18.1
A 76.4 12.4
B 75.9 12.7
C 75.8 12.3
D 75.8 12.1
A 70.1 17.3
B 70.1 12.4
C 70.2 19.8
D 64.2 15.4
A 89.4 29.4
B 88.6 29.2
C 33.7 29.2
D 38.4 24.5
Upgradient 4
5 Downgradient
6 Downgradient
7 Downgradient
8 Downgradient
(Continued)
B-6
OSWER-9950.1
TABLE 2 (Continued)
AN EXAMPLE OF TOX AND TOC DATA COLLECTED DURING A SEMIANNUAL
MONITORING EPISODE AFTER THE FIRST YEAR OF BACKGROUND MONITORING
Well Location Replicate TOC TOX
@pm) (ppb)
9 Downgradient A 59.7 16.2
B 60.1 16.4
C 60.1 16.2
D 58.3 16.1
10 Downgradient A 62.1 23.4
B 62.3 27.2
C 62.0 18.1
D 62.2 22.7
B-7
TABLE 3
METHODS FOR CALCULATING SUMMARY STATISTICS
DESCRIBING THE REPLICATE MEASUREMENTS
The background and monitoring well averages resulting from the
methodology described below become the data values that are used
in the averaged replicate t-test.
BACKGROUND WELLS
Average of the Replicates
'b
%,ij = kil % ijk/p , b
'Where: Xb,ijk = Concentration
well, the jth
measurement.
k = 1 to pb
measurement from the ith background
sampling period, and the kth replicate
Where i = 1 to nb, j = 1 to ob, and
Variance Among the Replicates -__ .
7 'b - 7
Coefficient of Variation Among the Replicates
cvb,ij = (sb,ij/xb,ij) l 100
MONITORING WELLS
Aver* of the Replicates .~
” i = z Xm ik/Pm , k=l ’
Where: G,ik = A quarterly concentration measurement from the ith
monitoring well and the kth replicate measurement.
Where i = 1 to r+.,, and k = 1 to pm.
(Continued)
B-8
OSWER-9950.1
TABLE 3 (Continued)
Methods for Calculating Summary Statistics
Describing the Replicate Measurements.
Variance Among the Replicates
S2 m.i = ki: lXm,ik - Xm i)2/(Pm-1) ,
Coefficient of Variation Among the Replicates
cvm i = (Sm pm i) l 100 I I ,
B-9
greater than or equal to the detection limit and the proportion of values
LT the detection limit. Cohen's methodology in the context of the
averaged replicate t-test as applied to RCRA interim status facilities is
described in Table 4, and the parameter estimates required to complete
the calculations are included in Table 5.
Examples of averaging the replicate measurements under the three
scenarios described above are presented in Table 6. These methods apply
regardless of how many replicate measurements are available. If no
replicate measurements were taken, there is no need for preparatory
averaging, and the single measured value from the well is used in the
analysis.
2.2.2 Additional Summary Statistics Describing the Replicate
Measurements
It is also advisable to evaluate the variance and standard deviation
among the replicate measurements. Although this component of variability
is not considered in the averaged replicate test, it does provide an
indication of the consistency of the replicate measurements and therefore
a notion of how the owner/operator's sampling and laboratory protocols
(depending on when and how the samples are split and collected) are
performing. Another, more interpretable, measure of variability is the
coefficient of variation. The coefficient expresses the standard
deviation in terms of a percent of the mean. Large coefficients of
variation are generally unacceptable and suggest poor laboratory quality
control. Table 3 describes the methodology for calculating the variance
and coefficient of variation among the replicate measurements. Tables 7
and 3 display the summary statistics which describe the replicate
measurements taken during the background characterization period for TOC
and TOX, respectively. Table 9 includes the summary statistics
describing the replicate measurements taken during the first monitoring
period.
B-10
OSWER-9950.1
TABLE 4
A METHODOLOGY FOR CALCULATING THE MEAN AND VARIANCE
OF THE REPLICATE MEASUREMENTS WHEN SOME OF THE REPLICATE
MEASUREMENTS ARE LESS THAN A LIMIT OF DETECTION
The mean and variance of the values greater than or equal to the
limit of detection must be calculated using the methodology described
in Table 3. An example application of this methodology is presented
in Table 6 as Case 3.
BACKGROUND
Estimate T hmii as follows:
-r -
T s2 ' b,ij = , b ij"%,ij - DL J2 b,ij
-t
Where: xb,ij = Mean of the measurements above or equal to the
limit of detection from the ith background well
sampled on the jth sampling period. This mean is
computed as follows:
%
%,ij = kc1 %,ijk'p& =
Where: %,ijk =
Pb =
Measurements above or equal to the
limit of detection
Number of measurements above or
equal to the limit of detection
'i,ij = Variance of the measurements above the limit of
detection from the ith background well sampled on
the jth sampling period. This variance is computed
as follows:
%
S2’ b,ij = & (T,ijk - X;,ij’2/‘P; - 1)
k=l
DL b,ij = Detection limit for measurements from the ith
background well sampled on the jth sampling period.
(Continued)
B-11
TABLE 4 (Continued)
A METHODOLOGY FOR CALCULATING THE MEAN AND VARIANCE
OF THE REPLICATE MEASUREMENTS WHEN SOME OF THE REPLICATE
MEASUREMENTS ARE LESS THAN A LIMIT OF DETECTION
Obtain values for hb,ij and Xb,i. as follows:
hb,ij = Proportion of the replicate measurements below the limit
of detection at well i on sampling period j.
xb,ij = A parameter estimate obtained from entering Table 5 with
Tb,ij and hb,ij.
Replicate mean and variance estimates considering the LT detection
limit values:
,ij = <,ij - , 'b ij'~,ij - DLb,ij)
S2 s2' + x -' b,ij = b,ij b,ij"b,ij - DLb,ij)2
MONITORING WELL
Estimate Tm i as follows: I
T m,i = s - DLm i)2 ,
Where: <,i = M ean of the measurements above or equal to the
limit of detection from the ith monitoring well.
This mean is computed as follows:
Where: Xm ik = Measurements above or equal to the r limit of detection
PI;I = Number of measurements above or
equal to the limit of detection
(Continued)
B-12
OSWER-9950.1
TABLE 4 (Continued)
A METHODOLOGY FOR CALCULATING THE MEAN AND VARIANCE
OF THE REPLICATE MEASUREMENTS WHEN SOME OF THE REPLICATE
MEASUREMENTS ARE LESS THAN A LIMIT OF DETECTION
S2 m,i = Variance of the measurements above the limit of
detection from the ith monitoring well. This variance
is computed as follows:
I
s2’ m,i = kir lXh,ik - 1)
=
DLm i = Detection limit for measurements from the ith , monitoring well.
Obtain values for hm i and Xm i as follows: I I
hm,i = Proportion of the replicate measurements below the
the limit of detection at well i.
x m,i = A parameter estimate obtained from Table 5 using
T m,i and hm,i-
Replicate mean and variance estimates, considerinq the LT detection
limit values:
- Am $i; i - DLm $ , , I
si i = s;li + xm pi; i - DLm $2 , , , , I
B-13
T
.oo
.05
.10
.15
.20
.25
.30
.35
.40
.45
.50
.55
.60
.65
.70
.75
.80
.85
.90
.95
1.00
I
TABLE 5
VALUES OF X FOR ESTIMATING THE MEAN AND VARIANCE
OF A NORMAL DISTRIBUTION WHEN LESS THAN DETECTION
LIMIT VALUES ARE PRESENT
.Ol .10 .20 .25 .30 .40
.010100
.010551
.010950
011310
:011642
011952
:012243
012520
:012784
.013036
.013279
.013513
.013739
.013958
.014171
.014378
.014579
.014775
.014967
.015154
.015338
.11020
. 11431
.11804
. 12148
.12469
. 12772
.13059
.?3333
. 13595
.13847
. 14090
.14325
14552
:14773
.14987
. 15196
.15400
* 15599
, 15793
.15983
.16170
.24268 .31862 .4021
.25033 .32793 .4130
.25741 .33662 .4233
.26405 .34480 .4330
.27031 .35255 .4422
.27626 .35993 .4510
.28193 .36700 .4595
.28737 .37379 .4676
.29260 .28033 .4755
.29765 .38665 .4831
.30253 .39276 .4904
.30725 .39870 .4978
.31184 .40447 .5045
.31630 .41008 .5114
.32065 -41555 .5180
.32489 I42090 .5245
.32903 .42612 .5308
.33307 .43122 .5370
.33703 043622 .5430
.34091 .44112 .5490
.34471 .44592 .5548
.5961
.6101
.6234
.6361
.6483
.6600
-6713
.6921
.6927
.7029
* 7129
7225
:7320
.7412
a7502
.7590
. 7676
.7761
.7844
.7925
.8005
(Continued)
B-14
OSWER-9950.1
T ' .50 .60 .70 . 80 .90
.oo .8368 1.145 1.561 2.176 3.283
.05 .8540 1.166 1.585 2.203 3.314
.lO .8703 1.185 1.608 2.229 3.345
.15 .8860 1.204 1.630 2.255 3.376
.20 *go12 1.222 1.651 2.280 3.405
.25 .9158 1.240 1.672 2.305 3.435
.30 (9300 1.257 1.693 2,329 3.464
.35 I9437 1.274 1.713 2.353 3.492
.40 I9570 1.290 1.732 2.376 3.520
.45 .9700 1.306 1.751 2.399 3.547
.50 .9826 1.321 1.770 2.421 3.575
.55 .9950 1.337 1.788 2.443 3.601
.60 1.007 1.351 1.806 2.475 3.628
.65 1.019 1.366 1.825 2.486 3.654
.70 1.030 1.380 1.841 2.507 3.679
.75 1.042 1.394 1.858 2.528 3.705
.80 1.053 1.408 1.875 2.548 3.730
.85 1.064 1.422 1.892 2.568 3.754
.90 1.074 1.435 1.908 2.588 3.779
1.00 1.095 1.461 1.940 2.626 3.827
-
TABLE 5 (Continued)
VALUES OF X FOR ESTIMATING THE MEAN AND VARIANCE
OF A NORMAL DISTRIBUTION WHEN LESS THAN DETECTION
LIMIT VALUES ARE PRESENT
h
From: A Clifford Cohen (19611, Technometrics 3:538
B-15
TABLE 6
EXAMPLE CALCULATIONS WHICH ILLUSTRATE HOW TO ESTIMATE
THE REPLICATE AVERAGE WHEN: (1) ALL THE VALUES ARE LESS THAN
A LIMIT OF DETECTION, (2) ALL VALUES ARE GREATER THAN A LIMIT
OF DETECTION, AND (3) THE VALUES CONSIST OF A MIXTURE
OF VALUES ABOVE, EQUAL, AND BELOW A LIMIT OF DETECTION
CASE 1: All values are less than a limit of detection
January, Well No. 1
Replicate gx (ppb)
A (10.0
B <lO.O
8" (10.0
D <lO.O
The replicate average is <lO.O
CASE 2: All values are greater than the limit of detection -.- ~-_
March, Well No. 4
Replicate TOX (ppm)
A 65.7
B 66.1
C 65.8
D 65.9
'b
%,ij = 21 xb ijk"b I
= (65.7 + 66.1 + 65.8 t 65.9)/4
= 65.88
CASE 3: The values consist of-a mixture of values above, equal ant
below a limit of detection -. .-_
January, Yell No. 2
Replicate TOX (ppb)
A 15.2
B 13.4
C 18.0
D <lO.O
(Continued)
B-16
OSWER-9950.1
TABLE 6 (Continued)
EXAMPLE CALCULATIONS WHICH ILLUSTRATE HOW TO ESTIMATE
THE REPLICATE AVERAGE WHEN: (1) ALL THE VALUES ARE LESS THAN
A LIMIT OF DETECTION, (2) ALL VALUES ARE GREATER THAN A LIMIT
OF DETECTION, AND (3) THE VALUES CONSIST OF A MIXTURE
OF VALUES ABOVE AND BELOW A LIMIT OF DETECTION
Mean of the values greater than or equal to a limit of detection
\,ij = kil yb,ijk'%
= (15.2 t 13.4 t 18.0)/3
= 15.53
Variance of the values greater than or equal to a limit of detection
%
s2 ’ b,ij = 1 'X1; ijk- 2; ij,"/(p,-l)
k=l ' ,
= ((15.2 - 15.53)2 t --- t
(18.0 - 15.53)2/(3-1)
= 5.373
Proportion of values LT the limit of detection
hb,ij = l/4 = 0.25
Detection limit
DLb,ij = 10
Estimate of Tb,ij
Tb ij = s2 I b'ij'(<,ij , , - DLb ij)2
= 5.373/(15.53 - 1012
= 0.178
(Continued)
B-17
TABLE 6 (Continued)
EXAMPLE CALCULATIONS WHICH ILLUSTRATE HOW TO ESTIMATE
THE REPLICATE AVERAGE WHEN: (1) ALL THE VALUES ARE LESS THAN
A LIMIT OF DETECTION, (2) ALL VALUES ARE GREATER THAN A LIMIT
OF DETECTION, AND (3) THE VALUES CONSIST OF A MIXTURE
CF VALUES ABOVE A!JD BELOW A LIMIT OF DETECTION
The value of lb,ij interpoiated using Table 5 is 0.3495.
'The :nean, considering the less-than-detection limit values, is:
<,ij = <,ij - gb ij (< ij - DLb .j) I I .1
= 15.53 - .3495(15.33 - 10)
= 13.60
B-18
TABLE 7
SUMMARY STATISTICS DESCRIBING THE REPLICATE MEASUREMENTS
OF TOC (ppm) THAT WERE TAKEN DURING THE ESTABLISHMENT
OF BACKGROUND CONCENTRATIONS
Well Month N Prop
<DL Mean Variance
1 1 4 0 60.78 0.14 0.38 0.62
3 4 0 63.45 0.14 0.38 0.60
5 4 0 70.43 0.87 0.93 1.32
7 4 0 68.85 0.12 0.34 0.50
9 4 0 66.25 0.11 0.33 '0.56
11 4 0 62.53 0.17 0.41 0.66
2 1 4 0 58.23 0.03 0.17 0.29
3 4 0 59.85 0.04 0.19 0.32
5 4 0 62.23 0.11 0.33 0.53
7 4 0 59.50 0.18 0.42 0.71
9 4 0 57.78 0.01 0.10 0.17
11 4 0 58.23 0.01 0.10 0.16
3
4
1 4 0 61.33 0.05 0.22 0.36
3 4 0 61.63 0.11 0.33 0.54
5 4 0 63.10 0.60 0.78 1.23
7 4 0 61.38 0.08 0.28 0.45
9 4 0 60.55 0.11 0.33 0.55
11 4 0 60.38 0.38 0.29 0.48
1 4 0 63.68 0.25 0.50 0.78
3 4 0 65.88 0.03 0.17 0.26
5 4 0 65.35 0.06 0.24 0.36
7 4 0 64.25 0.07 0.27 0.41
9 4 0 63.48 0.03 0.17 0.27
11 4 0 61.78 0.04 0.21 0 . a 3
Std. Dev. C.V.
B-19
TABLE 8
SUMMARY STATISTICS DESCRIBING THE REPLICATE MEASUREMENTS
OF TOX (ppb) THAT WERE TAKEN DURING THE ESTABLISHMENT
OF BACKGROUND CONCENTRATIONS
Well Month N Prop
<DL Mean Variance Std. Dev. C.V.
1 1*
3**
5**
7**
3**
11**
2 ;**
3
5
7
9
II
3 1
3
5
7
9
11
4 1
3
5**
7**
9**
11
0
2
2
1
3
2
3
4
4
4
4
4
4
4
4
4
4
4
4
4
2
2
3
4
1.00 <lO.O -- --
0.50 10.12 3.09 1.76
0.50 10.30 3.05 1.75
0.75 8.26 4.00 2.00
0.25 11.55 1.85 1.36
0.50 10.58 10.39 3.30
--
17.37
16.99
24.21
11.78
31.19
0.25 13.58 15.95 3.99 29.40
0 13.55 4.47 2.11 15.57
0 15.53 8.00 3.00 19.32
0 17.75 4.92 2.22 12.49
0 14.88 0.72 0.85 5.71
0 14.55 0.03 0.17 1.19
0 17.60 8.63 2.94 16.70
0 18.63 1.55 1.25 6.68
0 21.15 9.22 0.47 2.23
0 17.95 3.64 0.80 4.47
0 16.60 5.22 2.29 13.76
0 21.53 0.02 0.13 0.59
0 13.60 0.17 0.41 3.00
0 13.48 0.20 0.44 3.28
0.50 10.66 13.73 3.71 34.76
0.50 10.56 10.37 3.22 30.49
0.25 11.88 3.38 1.84 15.48
0 12.58 0.68 0.83 6.57
.--.
*All values were LT the limit of detection.
**Cohen's method was used to calculate the summary statistics.
3-23
OSWER-9950.1
TABLE 9
SUMMARY STATISTICS DESCRIBING THE REPLICATE MEASUREMENTS
TAKEN DURING THE FIRST MONITORING PERIOD FOLLOWING
THE ESTABLISHMENT OF BACKGROUND
Well Chemical
Location Parameter k Prop
<DL Mean Variance Std. Dev. C.V.
l/Up
2 /up
3/up
4/up
7/Dawn
8/Dawn
g/Down
lo/Down
TOX (ppb)
TOC (ppm)
TOX
TOC
TOX
TOC
TOX
TOC
TOX
TOC
TOX
TOC
TOX
TOC
TOX
TOC
TOX
TCC
TOX
TOC
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
12.68 3.75 1.89 14.91
71.83 0.48 0.69 0.96
24.43 7.00 2.65 10.83
63.45 0.71 0.84 1.33
19.05 0.12 0.34 1.79
63.50 0.29 0.54 0.84
8.96 2.69 1.64 0.18
64.68 0.06 0.25 0.39
18.18 0.10 0.01 0.53
68.53 0.58 0.34 0.85
12.38 0.06 0.25 2.02
75.98 0.08 0.29 0.38
16.23 9.75 3.12 19.24
68.65 8.80 2.07 4.32
28.08 5.69 2.39 8.50
38.78 0.19 0.44 0.49
16.23 0.02 0.13 0.78
59.55 0.73 0.85 0.35
22.85 13.94 3.73 16.34
62.15 0.02 0.13 0.21
B-21
2.2.3 Transformation of pH Measurements to Hydrogen Ion
Concentration
It may also be valuable in the case of interim status detection
monitoring parameters to consider transformation of the pH scale to
hydrogen ion concentration. This methodology is explained in Table 10.
The hydrogen ion concentration scale can be used for statistical
comparisons rather than pH scale measurements.
2.3 Data Summary
One of the most important initial steps is to review and evaluate
the ground-water data using summary statistics, tables, data plots, and
maps. The background data should be considered collectively and on a
well-by-well basis. Also, it is informative to consider whether there
are seasonal influences on the concentration measurements from particular
wells.
Most statistical software packages offer procedures that provide
univariate summary statistics of data and subsets of data. Table 11 is
an example of output that describes the background TCC and TOX averaged
repiicate data. These are quite informative with respect to the mean
background concentration, the variability of the background concentration,
percentile estimates, the presence of outliers, and the distributional
shape of the concentration measurements. Chapter Six also discusses the
use of summary statistics.
Another informative display of data involves plotting replicate
average concentrations over time. This permits a visual comparison among
the upgradient wells and indicates whether there appear to be seasonal or
unusual, extreme events. Figures 1A and 2B are plots of the averaged
replicate TOC and TOX data measured in the upgradient wells during the
year of background characterization.
B-22
OSWER-9950.1
TABLE 10
METHODOLOGY FOR TRANSFORMING THE pH MEASUREMENTS TO
HYDROGEN ION CONCENTRATIONS
The pH is equal to the negative base ten logarithm of the hydroqen
ion concentration:
pH = -loglo[H30+[
Where : k30+1 = moles/liter of H30t
The hydrogen ion concentration is therefore equal to:
IH30+1 = 10'pH
B-23
TABLE 11
A SUMMARY DESCRIPTION OF TEi TOC (ppm) AND TOX (ppb) AVERAGED
REPLICATE DATA COLLECTED FROM THE UPGRADIENT WELLS
CURING THE BACKGROUND CHARACTERIZATION PERIOD
N
“E*N
slu DLV
SXEWESS
Los
CV
l:rlE4n.O
scNPuI(
MJl -= 0
Y:KI(IML
STEM LLAC
70 6
bb 8
bb 2
IS wl bclS
b?..lWI M
J.10651 VUIKE
0.7413.Sl KIRTMIS
94on6.7 css
5.01561 SlO MEAN
9b.1*11 Wrn>lYl
150 Wrn~l~l
ES
o.*SllD3 AIc4CY
I
1a 219
b1 251457
b0 *5b%bb
se 225e
WI
- - - -. -- - - l --
VILW? Gown
I?. 775 I
5b.225 2
59.5 1
39.85 1
60.375 I
LO.55 1
I
1
1
s
b
7
4
I
- -. - - - - +
PCPCEN-TT) PCRCEMS PERCENTS
CELL cm VALUE C0.h-T CELL CUJ VALUE CamT CELL CL?!
e.2 4.t bO.775 1 4.2 3J.3 (2.525 1 6.t se.3
1.1 12.5’ bl .325 1 6.2 37.5 63.1 1 *.t 12.5
c.2 lb.7 bI.IIS 1 4.2 41.7 (3.95 1 1.L Lb.7
4.2 20.e bl .btS 1 4.2 45.b (1.475 1 4.2 70.6
4.2 25.0 Il.775 1 4.2 so.0 b1.b75 I e.2 75 b
b.2 2e.t Lt.225 1 4.2 54.2 be.25 I 6.2 79.2
E33.53?
O.b50‘46
O.OOOI
0.0001
O.Jb
WxPLOl
0
I
I .-----,
I + I .-----*
I I
WAM-fILESlDC~:4 I
70.4cs qn!
b4.1017. 95%
b2 90%
bO.bI(T 107
70.525
70.0312
b7.549q
SS.22S
Ckl?ERES
WEST NIOcLIT
S7.775 bS.35 Id.225 bS.b75
SS.225 bb.25
57.775 I/.
1%
s7.aa75
57.775
IV.5
sq.es
be.bS
70.42s
It.bS
l.be75
M.225
WDPRAL wcaMILITv PLOT
71t . 6.
I . *+..*.
**.....
,...Y
+*-
-
-.“I
57. ..*.a.
*----*----*----*----.----,----*----.----*----*----,
-2 -1 0 11 *a
wnctNTs
VlLlJt mwT CELL M
LS. 15 1 4.t 01.1
LB.175 1 4.t S7.S
bb.25 1 a.2 qt.7
bb.85 I a.2 95.e
70.625 1 4.2 100.0
TOX -.
MEKTS aJANlILL5lDEF’r.~ lXrPEH2S
N 24 SLA IGTS 24 100% rux 21.525 99% 21.525 LCUEST HIMEST
MEAN 14.035 SLm 1x.64 75% P3 17.15 95% 11.4312 I.tb 17.75
STC CEV J. b5700 VMIAXE 11.3742 50% RED 11.sb5 90% 19.M75 10 17.95
SKEY(CSS 0.517442 KLIITOSIS -0.u9072 25% et 10.b 10% 10.0b 1D.l.Z lS.btS
uss 5035.1b CfS
CV 2b.0569 ST.0 MlN
T:t!EAH=O ta.eott Woe'ITI
s&N nw( 150 R)W’lSl
NM -- 0 CI
Y:mRRAuL o.wds31 wmen
307.boe
0.766**9
O.ODOl
0% fun
RANGE
1.26
13.Zb5
b .74999
b.Cb
5%
1%
b.b9S
e.2b
lQ.3 21.15
lD.W 21.525
STEfl LEAF
20 IS
IO b
lb bbbe
14 595 It b554b
10 OlYb759
s
b
83 1
----,----.----.----*
PEPCENTS PEI)CEHTS
VALUE cm CLLL CLH VALW OQM CELL M
b.2b 1 4.2 4.2 I@.bb 1 5.1 29.2
IO 1 4.2 0.3 11.55 1 6.I 13.3
10.12 1 6.2 It.5 11.0 1 4.2 17.5
10 5 1 4.2 tb.7 11.575 1 4.2 Cl.7
10 5b I 4.2 t0.b 13.*75 1 5.2 *S.b to.se I 6.E tS.0 13.55 I 4.t 50.0
@.@OOl P3-at
mE
1.3
BOXPLOT
I
I
+-----,
I + I
.---mm*
l -----*
I
FREWtKT
~NL wmmL2Ty PLOT
21t . ..*..
I
.L....
HI”
151 +..-
I
.rm ”
. .,“.em,.
9t . l .+t*
+----*----,----,----* ----,----,----.----*----*----*
-I -1 0 r1 l 2
TABLE
PERCEMTS
VkLUt toun CELL M
PCRCCtrrS
VALU2 WINT CELL CUl
13.W 1 4.2 W.t 17.b 1 6.1 79.t
15.b 1 *.t sb.1 17.75 1 6.t a3 3
14.55 I 6.t bZ.5 17.95 1 4.2 87.5
lC.175 I 4.t Lb.7 le. b25 1 4.2 91.7
15.525 I 4.1 70.0 Il.15 1 4.t 95.6
1b.b 1 4.2 75.0 t1.s25 1 4.2 100.0
B-24
oSxE&?-9g5~.:
FIGLJRE i
PLOTS OF TOX (ppb) AND TCC (ppm) CONCENTRATIONS VERSUS TIME IN THE FOUR
U~GRADI~ WELLS THAT WERE USED T3 CHARACTERIZE BACKGROUND CONCENTRATIONS
TCX I 2%.
21
*o
I*
10
I7
16
15
I4
II
I2
11
LO
9
I .
7 .
‘ .
___,_------.-------,------- .-------,-----.-*-------I--------.-------*-------,-------.-~-----.-------~-------.-------.-------.--
,5,w,2 Oc,~2~82 2'rrEBll ,b".PbL 05AAIbZ 25AFI82 15?'A(*182 OrJW&t 26~HI2 ICJULbZ 05AL'ZlZ 2lAUGbZ ‘LILPlt DtOCTb2 22XTbtllMJW32
.
t
.
.
..-------.-------*-------.-------*-------.-------.-------.-------.------ *.-------.------.
TbZ CSAP”bt 15MrbZ 04-2 ,sA,M I~JULbZ 05AUGbZ I,,&32 ,2,tW2 Ot'XTd2 2ZQCT8211'~
3-25
FIGURE 2
x;oRMAi PR:EABI;,ITY PLC:S :;F TOX (ppb) AND TCC (pm) CONCENTRATION VERSUS THE
ycm SC3RES Em? T:-iE "ATA WHICH ARE PL3TTED AS STARS (*) AND FROM DISTRIBUTION
',<:;I{ THE Si\ll'E "EAN .UI! VARIANCE: AS THE DATA jJHICH ARE REPRESENTED BY THE LINE.
. -*-------.----.-.-------.-.~.--.--------------* ..---------. .--------------.---- __*___e.mm.-
-2 c -L.5 0 -0 0 0 0.b 0 1.s t.0
3-25
OSWER-9950.1
3.0 SEASONAL TRENDS
3.1 Characterization of Seasonality
During the analysis of interim status detection monitoring data, it
is important to consider seasonal trends in concentration. The presence
of time or seasonal effects introduces a factor that may obscure the
presence of, or falsely indicate, leakage from the hazardous waste unit.
This is because there are times of the year when concentrations are
normally higher or lower than the average. In such a situation, if a
downgradient well is sampled during a period when concentrations are
high, the statistical test may suggest the presence of contamination when
actually the high values are the result of normal seasonal concentration
increase,
In order to
the ground-water
evaluate whether seasonal influences are reflected in
concentration measurements, one should plot the data
plotted over time. Figures 1A and 1B indicate that the TCC data for all
wells in the system appear to increase during mid-year and decrease
during the winter. In contrast, the TOX data reveal no clear seasonal
trends.
3.2 Methods for Reducinq the Adverse Effects of Seasonally
Influenced Data
Two methods are available for considering seasonal fluctuations in
interim status ground-water monitoring data. The first method can be
applied when one year of background data are used in the analysis and
simply calls for the seasonal effect to be included in the variance
estimate used for the averaged replicate t-test. Essentially, this
method includes the additional variability caused by seasonality in the
t-test error term. As a result, comparisons of monitoring well data with
the background data will not lead to inaccurate contamination assessments
because the seasonal variability will have been accounted for in the
error term. Under this method, the difference between the upgradient and
downgradient mean must exceed the differences expected by seasonal change
in order to indicate contamination.
B-27
The other method uses a seasonal correction methodology. Under this
approach, the background and monitoring data are corrected to reduce the
tendency for the data values to become seasonally large or small, but
retain their original error structure. This method requires that the
upgradient wells have been monitored for more than one year (Chapter Five
discusses the situations and considerations that may lead to a
modification of the background data set).
The seasonal correction is performed separately for each well and
chemical parameter. Table 12 presents an example application of the
seasonal correction methodology. First, monthly averages of the average
replicate values are calculated by averaging across years for each
month. Then, an overall average is calculated for all the averaged
replicate values across all years and months. Finally, the adjusted
means are calculated by taking an averaged replicate value then
subtracting the monthly mean and adding the overall background mean.
The data from subsequent monitoring events must also be corrected if
seasonally adjusted data have been used to establish the background
statistics. The monitoring data are corrected in a similar fashion by
subtracting the monthly averages from the background data and then adding
the overall average from the background data to the averaged replicate
monitoring data values.
Several problems may arise in the use of seasonal correction. If
monitoring data were collected on an even month, say April (4), then,
because the background data are only available for odd numbered months,
the monthly averages from the two adjacent months (March and May) could
be averaged to estimate a monthly average for correcting the April
monitoring event.
Finally, after the background data have been corrected, it is useful
to replot the data for summary and review purposes.
B-28
XWER-9950.1
TABLE 12
AN ILLUSTRATION OF HOW TO PERFORM A SIMPLE SEASONAL CORRECTION
USING TOC (ppm) DATA FROM MONITORING WELL NO. 1
The seasonal correction can only be performed if more than one year of
background data are available. Consult Chapter Five for when and how to
update background data.
Month
Averaged Replicate Values Adjusted Means***
1982 1983 1984* Monthly Means** 1982 1983 1984
1 60.78 58.23 61.33 60.11 66.59 64.04 67.14
3 63.45 69.85 61.47 64.92 64.45 71.30 62.47
5 70.43 82.23 79.10 77.25 59.10 70.90 67.77
7 68.85 79.41 69.27 72.51 62.26 72.82 62.68
9 66.25 54.78 60.41 60.48 71.69 60.22 65.85
11 62.53 58.13 60.00 60.22 68.23 63.83 65.70
Overall Background Mean %b,i = 65.92
*The data from 1983 and 1984 have not been discussed elsewhere in Appendix B.
These are included because the seasonal correction methodology requires more
than one year of data.
**Monthly means are calculated by averaging for a particular month all of the
measurements taken during the month over the prior monitoring.
***The adjusted means are calculated by taking an averaged replicate value then
subtracting the monthly mean and adding the overall background mean. For
example. the adjusted monthly mean for May 1983 was calculated as follows:
82.23 - 77.25 t 65.92 = 70.90
4.0 GOOCNESS-OF-FIT
Before applying the t-test to the data, it is also important for
owner/operators to evaluate whether their replicate average data have
been sampled from a normally distributed population of concentration
measurements. Many background data sets will be too small to reasonably
evaluate with respect to distributional shape; for example, a single-well
upgradient system sampled quarterly only yields four replicate average
values.
4.1 Graphical Methods -
One simple method for evaluating data distributions is to plot the
data on a normal probability plot and overlay a plot of the data expected
from a normal distribution that has the same mean and variance as the
data. If the sampling data deviate substantially from the data expected
from a normal distribution, then the data may not have been sampled from
a normal distribution. The methodology for developing normal probability
plots is well documented (e.g., Neter and Wasserman, 1974; and Shapiro,
1980) and will not be described.
Figures 2A and 2B are normal probability plots of the replicate
averages of the TOC and TOX data, respectively. In these instances, the
data approximate a reasonably normal distribution. The replicate
averages, because of a fundamental statistical principle referred to as
the central limit theorem, will tend to approach a normal distribution.
However, in some instances, the normal distribution will not be
appropriate and lognormal estimates of the mean and variance may be
useful. Aithchison and Brown (1957) present methodologies for estimating
lognormal distribution parameters. Enforcement officers should not,
however, allow owner/operators to simply take the natural logarithms of
their data prior to analysis because this will reduce the ability of the
statistical procedure to detect contamination.
B-30
OSWER-9950.1
4.2 Hypothesis Testing Methods
Another set of methods that can be used to evaluate the
distributional shape of replicate averages uses statistical tests, One
problem with statistical goodness-of-fit hypothesis testing is that few
tests are useful with small sample sizes. The benefit is that unlike the
visual comparison of a line with data points, there is no subjectivity
associated with a statistical goodness-of-fit hypothesis test. The null
hypothesis that the data follow a normal distribution is either accepted
or rejected. If the hypothesis is rejected, then the lognormal theory
referenced above may be useful.
One statistical goodness-of-fit test, which performs well on small
sample sizes and tests the null hypothesis that the data values are
random samples from a normal distribution against an unspecified
alternative distribution, is the Shapiro-Wilk, W statistic (Shapiro and
Wilk, 1965).
The enforcement officer should respond to complaints regarding the
non-normality of data by insisting that owner/operators evaluate, either
graphically or via a statistical test, the goodness-of-fit of their data
distributions. Enforcement officers should also understand that
parametric methods such as the t-test are robust to departures from
normality and that the outcome of the statistical evaluation is not
altered by small deviations from normality, particularly when larger
sample sizes are available (Harris, 1975). Finally, interim status
facilities are required by 40 CFR §265 to use a Student's t-test and
therefore cannot use a nonparametric statistical procedure to circumvent
the requirement for normally distributed data.
5.0 ANALYSIS OF MONITORING WELL DATA COLLECTED AFTER CHARACTERIZATION OF
THE BACKGROUND GROUND-WATER QUALITY
After development of the background ground-water concentrations
interim status, owner/operators must sample their entire well systems
B-31
semiannually. The purpose is to determine whether any weil in the
monitoring system has concentrations that are larger than (or in the case
of pH, different from) those established during the characterization of
the background water quality.
Data collected during May 1983 from the four upgradient and six
downgradient wells are presented in Table 2. The data consist of four
replicate measurements of TOC and TOX from each of the ten wells. The
replicate measurements are averaged prior to analysis using the
methodology described earlier in Appendix B. Table 9 presents the
averaged replicate monitoring data.
6.0 THE AVERAGED REPLICATE T-TEST
6.1 Calculation Methodology
Once the replicates are averaged and summary statistics, which
describe the background data, are developed, the calculation of the test
statistic is straightforward. Table 13 describes the methodology for
calculating the required input statistics and test statistics. Table 14
presents example calculations that compare the background TOX data with
data from downgradient Well 6.
Observe that Cohen's method is also used in these calculations.
This is because during background characterization, all four replicates
from Well 1 measured during the first month of monitoring were less than
the limit of detection. Therefore, as described earlier, the replicate
average was also <lO.O ppb of TOX. Cohen's method was needed to estimate
the background summary statistics from the replicate average data.
6.2 Control of the False Positive Rate
The test statistics from the calculations described in Table 13 are
compared with critical values from the t-distribution that have been
adjusted to control the overall false positive probability for the waste
B-32
GSXER-9950.1
TABLE 13
A DESCRIPTION OF THE METHODOLOGY USED TO CALCULATE THE
TEST STATISTIC FOR THE AVERAGED REPLICATED T-TEST
The notation assumes that data were obtained from every upgradient
well every time they were sampled during the background characteri-
zation period. Alternative and more complicated methods which
require estimating the contribution from several components of
variance, fractional degree of freedom estimates, and linear
combinations of mean square estimates can also be used to provide
unbiased estimates of the background variance.
WITHOUT LESS THAN DETECTION LIMIT VALUES
Background Mean
s = ,I 1 'b,ij'%"b
i=l j=l
Background Variance --
2
'b =
WITH LESS THAN DETECTION LIMIT VALUES
Background Mean of All Nondetection Limit Values
Where: I$ = Number of averaged repiicate values greater than or
equal to the limit of detection in the background
data set.
. <,ij =
Average replicate values greater than or equal to
the limit of detection in the background data set.
(Continued)
B-33
TABLE 13 (Continued)
A DESCRIPTION OF THE METHODOLOGY USED TO CALCULATE THE
TEST STATISTIC FOR THE AVERAGED REPLICATED T-TEST
Background Variance of All Nondetection Limit Values
Cohen's Adjustment
Tb = +I/(< - DLb)2
hb = proportion of values less than a limit of detection
Xb = from Table 5 based on values of hb and Tb.
Adjusted Background Mean
- -I -Q
)6 = 'b - hb'y, - DLb)
Adjusted Background Variance
s; = sg+ x b< ( - DLb)2
AVERAGED REPLICATE TEST STATISTIC
* Xm i - yo ,
tm,i =
I
'b 1 + WI-lb l Ob)
B-34
OSKER-9950.1
TABLE 14
MPLE CALCULATIONS OF THE METHODOLOGY DESCRIBED IN TABLE 13,
WHICH COMPARE THE TOX AVERAGED REPLICATE BACKGROUND DATA
WITH THE TOX DATA FROM DOWGRADIENT WELL 6
Backqround Mean, Variance, and Standard Deviation of All Averaqed
Replicates Above a Limit of Detection
-!
Xb = (10.12 + 10.30 + **- + 11.88 + 12.58)/23
= 14.21
2'
Sb = ((14.21-10.21J2 + -** + (14.2l*12.58)2)/(23-:)
= 13.22
Cohen's Adjustment C* 1 c
Tb = 14.21/( - lo.@)2
= 0.746
hb = l/24 = 0.042
xb = 0.061 (From Table 5)
Adjusted Background Mean, Variance and Standard Deviation of the Averaqed
Replicates
'b = 14.21 - 0.61(14.21 - 10.0)
= 13.95
2
'b = 13.22 + 0.61(14.21 - 10.0j2
= 14.30
'b =
(Continued)
B-35
TABLE 14
EXAMPLE CALCULATIONS OF THE METHODOLOGY DESCRIBED IN TABLE 13,
-?JHICH COMPARE THE TOX AVERAGED REPLICATE BACKGROUND DATA
WITH THE TOX DATA FROM DOWGRADIENT WELL 6
The Averaged Replicate Value from Monitoring Well No. 6
x m,6 = (12.4 + 12.7 t 12.3 t 12.1)/4
= 12.38
The Averaged Replicate Test t-Statistic
*
tm,6 = (12.38 - 13.95)/(3.78d=)
= - 0.407
B-36
OSWER-9950.1
management unit. The probability depends on the monitoring event under
evaluation and considers that multiple downgradient wells are being
tested and that the concentrations of four indicator parameters are being
measured. Critical values based on Bonferroni t-statistics are used for
each individual comparison to control the false positive rate at one
percent for the entire facility. Miller (1981) discusses Bonferroni
t-statistics and methods for estimating critical values. Tables 15
and 16 include tabulations of critical values (one and two tailed,
respectively) to use for individual comparisons that control the overall
facility false positive rate at one percent.
6.3 Evaluation of Whether There Is a Suggestion of Contamination
The test statistics (t*) calculated for each well using the
methodology described in Table 13 are presented in Table 17. The test
statistics are compared with the Bonferroni critical test statistics
(t,) using the following decision rules:
l If specific conductivity, TOC, or TOX are being evaluated and
if t* is less than t,, then there is no statistical indication
that the concentrations are higher in the well under comparison
than in the background data. If t* is larger than t, then
there is a statistical indication that the concentrations are
higher in the well under investigation.
0 If pH is being evaluated and if It*1 (absolute value of t*) is
less than t,, then there is no statistical indication that
pH has changed. If It*1 is larger than t,, then there is a
statistical indication that pH has changed. If t* is negative,
then pH increased: if t* is positive, then pH decreased relative
to background.
6.4 Evaluation of the Power and False Negative Rate
The false negative rate and power for each chemical parameter can
be evaluated after characterization of the background ground-water
quality. As described in Chapter Five, this is an important evaluation
procedure because it allows evaluation of the false negative rate, that
is, the probability that a difference in mean concentration of a specified
B-37
Total No
of Wells
4
5
6
7
8
9
10
11
12
13
14
15
TABLE 15
ONE TAILED CRITICAL (t,) VALUES WHICH CONYROL THE
OVERALL SIGNIFICANCE LEVEL AT ONE PERCENT
Deg-ress of Freedom Associated with the
Averaged Replicate Test Statistic
3 7 11 15 19 23 27 31 3s
___. -..---.-..- -. - -. .-- - ..- -.-
6.297 4.543 4.365 3.841 3.712 3.628 3.568 3.524 3,490
6.534 4.609 4.175 3.939 3.803 3.714 3.651 3.604 3.569
6.729 4.793 4.265 4.019 3.876 3.703 3.718 3.669 3.569
6.896 4.689 4.342 4.086 3.939 3.842 3.714 3.724 3.388
7.041 4.972 4.408 4.145 3.992 3.893 3.823 3.771 3.490
7.169 5.045 4.466 4.136 3.039 3.937 3.865 3.812 3.569
7.285 5.111 4.518 4.242 4.082 3.977 3.904 3.849 3.632
7.390 5.171 4.566 4.283 4.120 4.013 3.938 3.882 3.685
7.487 5.225 4.609 4.321 4.154 4.046 3.969 3.912 3.731
I ! 7.576 5.276 4.548 4.356 4.186 4.076 3.398 3.940 3.1-11
7.657 5.322 4.685 4.388 4.216 4.103 4.024 3.366 3.307
7.136 5.366 4.719 4.418 4.243 4.129 4.049 3.989 3.839
1.----..- --- -- -. .- -.- - .._. --
B-38
OSWER-9950.1
TABLE 16
TWO TAILED CRITICAL (t,) VALUES WHICH CONTROL THE
OVERALL SIGNIFICANCE LEVEL AT ONE PERCENT
Degress of Freedom Associated with the
Total No. Averaged
of Wells / 3 7 11 15 19 23 27 31 35
4 7.041 4.972 4.408 4.145 3.992 3.893 3.823 3.771 3.731
5 7.285 5.111 4.518 4.242 4.154 4.046 3.969 3.912 3.869
6 7.487 5.225 4.609 4.321 4.154 4.046 3.969 3.912 3.869
7 7.659 5.322 4.685 4.388 4.216 4.103 4.024 3.966 3.920
8 7.808 5.406 4.751 4.446 4.269 4.153 4.072 4.012 3.965
9 7.941 5.481 4.810 4.496 4.315 4.197 4.114 4.052 4.004
10 8.061 5.547 4.862 4.542 4.357 4.236 4.151 4.088 4.039
11 8.169 5.608 4.909 4.583 4.394 4.271 4.185 4.120 4.071
12 8.269 5.663 4.952 4.621 4.429 4.304 4.215 4.150 4.100
13 8.361 5.714 4.992 4.655 4.460 4.333 4.244 4.177 4.126
14 8,446 5.761 5.029 4.687 4.489 4.360 4.270 4.202 4.150
15 8.525 5.805 5.063 4.717 4.516 4.386 4.294 4.226 4.173
B-39
TABLE 17
THE RESULTS OF THE AVERAGED REPLICATE T-TEST WHICH
COMPARE BACKGROUND TOC AND TOX DATA WITH THE DATA
COLLECTED DURING THE SUBSEQUENT MONITORING PERIOD
This analysis assumes that pH and specific conductance were
also monitored.
Monitoring
Well
WC (ppm) TOX (ppb)
'm x,-q) t* %rn Xm-i$ t*
1 71.83 9.29 2.857 12.68 -1.27 -0.329
2 63.45 0.91 0.280 24.43 10.48 2.716
3 63.50 0.96 0.295 19.05 5.10 1.322
4 64.68 2.14 0.658 8.96 -4.99 -1.293
5 68.53 5.99 1.842 18.18 4.23 1.096
6 75.98 13.44 4.133* 12.38 -1.57 -0.407
7 68.85 6.11 1.879 16.23 2.28 0.597
8 88.78 26.24 8.070* 28.08 14.13 3.663
9 59.55 -2.99 -0.920 16.23 2.28 0.591
10 62.15 -0.39 -0.120 22.85 8.90 2.307
tc(overal1 alpha=0.01, k=40, df=23) = 3.98
TOX Xb = 13.95 ppb, TOC %= 62.54 ppm
TOX sb{= = 3.858, TCC sb,/= = 3.252
*The concentrations measured in the well are statistically larger than
the concentrations measured during the background characterization
period.
B-40
OSWER-9950.1
magnitude will not be detected by the statistical procedure. The
complement of the false negative rate is the power of the statistical
test, which is the probability that the procedure will detect a
difference.
A power and false positive evaluation should be performed at a
concentration threshold which causes the test to indicate a statistically
significant difference and at several concentrations that are less than
the difference detected by the statistical test. The reason for perform-
ing this analysis is that smaller differences between the background and
downgradient data concentrations than were detected by the statistical
test may suggest contamination of the ground water by the unit being
monitored. If the statistical procedure is only able to detect large
differences as being statistically significant, then more samples or
alternative approaches may be necessary.
Table 18 presents the results of such an analysis using the TOX and
TOC data. Table 19 is a power table taken from Cohen (1969) that is
required for the analysis. Table 18 indicates that the AR t-test as
applied to these data performs well. Contamination would only be missed
a large percentage of the time if the contamination resulted in only a
1 ppm for TOC or 1 ppb for TOX difference between upgradient and
downgradient.
B-41
TABLE 18
A POWER ANALYSIS 3F THE AVERAGED REPLICATE T-TEST CONDUCTED ON THE
TOC AND TOX DATA USING '=HE IYET:-:GDOLOGY DESCRIBED IN COHEN (1969)
Constants Required for ths_Analysis
Difference Detected Standard Background
as Significant Deviation Sample Size
i -
-_t' Sb l+i/25 l tc = Xm - Xb 'b n
xc 3.252 9 3.977 = 12.93 3.186 24
TOX 3.85% l 3.977 = 15.34 3.780 24
Power and False NegaJi,?e Rate Analysis as a Function of the Mean - - -.
Different= Between the Bac'kground Data and Data from a Monitorinq Well
x
b
- ”
False Negative
Difference 'b Power Rate __--.-
TCC (ppm) 12.93 5.74 >.995 <.005
TC?X (ppbl 15.34 5.74 >. 995 c.005
TCC 10.0 5.56 >. 995 <.005
TOX 10.0 4.3c >. 995 <.005
TOC 3.0 1.33 0.96 0.04
TOX 3 . 13 i . 12 0.86 0.14
Tot 1.0 0.44 0.14 0.86
TOX 1.0 0.37 0.09 0.91
B-12
OSWER-9950.1
TABLE 19
A powl3~ TABLE (Cohen, 1969)
n dc
It
9
I.49
1.3b
‘0
II
$2
13
I4
1.20
1.21
1.15
1.10
I.05
15
tb
I?
III
I9
1.0)
.97
.a
.92
.I
20
21
22
23
n
1:
JJI
.79
.n
25 .76
26 .IL
17 .n
20 .7I
29 .70
.H
1:
45
.a
.o
.b2
.62
.bl
A0
.59
.5e
.56
.55
.Sb
.I0
iI
01
01
01
01
01
01
ol
01
01
01
01
01
01
01
01
ol
01
:
02
02
02
02
02
02
02
0
02
02
02
02
02
02
02
02
.20
02 01
02
02
02
02 02
02 02
02 02 02
w
OY a
:
03 OY
: 03
:
2 OS
OS bs
2 OS
OL
2 05 05
.10 .50 .bO .70 .eo I.00 I.20 I.lrO
0)
04
05
05
09 12 21
II I5 25
09 12 I7
IO 14 20
II IC 22
I2 18 25
lb 20 27
29
2
u
56
::
Ill
19
21
22
24
26
::
iz
10
10
II
t4 22 32
I5 24 )rc
I6 25 36
I7 27 38
10 28 10
19
20
21
::
I2
ii
LLl
50
56
58
60
63
65
14 2b
,5 25
(5 26
I6 27
17 28
::
b0
::
52
14
56
::
17
11
IO
I9
20
15
::
59
50
8:
bh
66
67
20
22
23
w
26
%2
55
::
62
60
71
*
76
76
;:
57
:
67
70
::
70
80
2
x;
:
91
92
n
z
z
96
96
97
2
99
n
50
:t
56
58
loo
120
IS0
lb0
I&l
200
250
:E
400
450
500
600
E
900
loo0
dc .I0
.51
.5l
.50
.50
.49
02
02
02
02
02
.18 02
.46 02
A5 02
.u 02
.42 03
.bl
A0
::89
.3e
03
03
03
01
03
::I
.I1
.29
.27
03
2
05
05
.26
.23
.2l
.20
.I@
.I?
06
2
IO
I2
lb
.I6
.I5
.I4
.I3
.I2
.I2
lb
20
24
::
37
.I0
I4
lr(
I5
16
16
I7
t8
20
21
21
2h
26
27
:‘o
32
::
zs
60
66
ii
::
9?
CM
l
.10
27
20
:;
32
34
36
::
14
47
50
52
P
57
59
69
77
ii
91
97
99
.
.50
I5
b7
49
51
53
::
::
68
02
35”
97 90
99
l
-70
--
07
07
z
09
09
IO
IO
II
II
I2
15
Ill
21
2%
*70
El1
02
Is
86
87
88
91
z
95
%
g
99
99
l
:69
b5
::
66
x:
m
92
95
97
l C- valuer belar thlr pdnt are 9rrtcr thm .995.
.80 1.00 1.20 l.bO
9199 + l
92 99
93 99
* l
95
96
97
5
99
2
l
B-43
7.0 REFERENCES
Aithchison, J., and J.A.C. Brown. 1957. The Lognormal Distribution.
Cambridge University Press, New York.
Cohen, C. 1961. Tables for Maximum Likelihood Estimates from Single
Truncated and Singly Censored Samples. Technometrics 3:535-541.
Cohen, J. 1969. Statistical Power Analysis for the Behavioral Sciences.
Academic Press, New York.
Miller, R.G. 1981. Simultaneous Statistical Inference. Springer-Verlag,
New York.
Neter, J., and W. Wasserman. 1974. Applied Linear Statistical Models.
Richard D. Irwin, Inc., Illinois.
Peiser, A.M. 1943. Asymptotic Formulas for Significance Levels of
Certain Distributions. Annals of Mathematical Statistics 14:56-62.
Shapiro, S.S., and M.B. Wilk. 1965. An Analysis of Variance Test for
Normality (complete samples). Biometrica 52:591-611.
Shapiro, S.S. 1980. How to Test Normality and Other Distributional
Assumptions. In: The ASQC Basic References in Quality Control:
Statistical Techniques. Vol. 3, Ed. E.J. Dudewicz, American Society
of Quality Control, Milwaukee, Wisconsin.
B-44
OSWER-9950.1
APPENDIX C
DESCRIPTION OF SELECTED GEOPHYSICAL METHODS
AND ORGANIC VAPOR ANALYSIS
OSWER-9950.1
APPENDIX C
SELECTED GEOPHYSICAL METHODS AND ORGANIC VAPOR ANALYSIS
This Appendix is a presentation of several investigative techniques
capable of augmenting data gathered from boreholes and ground-water
monitoring wells. The five methods are:
1. Ground Penetrating Radar (GPR)
2. Electromagnetic Conductivity (EM)
3. Resistivity
4. Seismic Refraction/Reflection
5. Organic Vapor/Soil Gas Analysis
The summaries of EM and resistivity focus on surficial and not
borehole methods. Although surficial and borehole techniques operate
under the same physical principles, the reader should be aware that
surficial and borehole techniques have different characteristics.
Surficial methods can be undertaken without regard to the number of
location or boreholes therefore providing a great deal of flexibility
to the investigation without disturbing the subsurface. Borehole EN and
resistivity, however, offer a much higher degree of resolution at depth
in the vicinity of a single borehole or between two or more.
The effectiveness of geophysical methods and organic vapor/soil
gas analysis increases if several techniques are used conjunctively.
For instance, EM, resistivity and organic vapor analysis are highly
correlative in the field where organic contamination exists.
C-i
GROUND PENETRATING RADAR..._(S-,P_R)*
Ground penetrating radar (GPR) uses high frequency radio waves to
acquire subsurface informlation. From a small antenna which is moved
slowly scross the surface of the ground, energy is radiated downward into
zhe subsurface, then reflected back to the receiving antenna, where
variations in the return signal are continuously recorded: this produces
a continuous cross-sectional "picture" or profile of shallow subsurface
conditions. These responses are caused by radar wave reflections from
interfaces of materials having different electrical properties. Such
reflections are often associated with natural geohydrologic conditions
such as bedding, cementation, moisture and clay content, voids, fractures,
and intrusions, as well as man-made objects. The radar method has been
used at numerous HWS to evaluate natural soil and rock conditions, as
well as to detect buried wastes.
Radar responds to changes in soil and rock conditions. An interface
between two soil or rock layers having sufficiently different electrical
properties will show up in the radar profile. Buried pipes ar.d other
discrete objects will also be detected.
Depth of penetration is highly site-specific, being dependent upon
the properties of the site's soil and rock. The method is limited in
depth by attenuation, primarily due to the higher electrical conductivity
of subsurface materials. Generally, better overall penetration is
achieved in dry, sandy or rocky areas: poorer results are obtained in
moist, clayey or conductive soils. However, many times data can be
obtained from a considerable depth in saturated materials, if the
specific conductance of the pore fluid is sufficiently low. Radar
penetration from one to ten meters is common.
*GPR has been called by various names: ground piercing radar, ground
probing radar and subsurface impulse radar. It is also known as an
electromagnetic method (which in fact it is); however, since there are
many other methods :Ihich are also Slcctromagnetic, the term G?R has come
into common use today, <and ~111 be used herein.
. . : - -
CSWER-9950.1
The continuous nature of the radar method offers a number of
advantages over some of the other geophysical methods. The continuous
vertical profile produced by radar permits much more data to be gathered
along a traverse, thereby providing a substantial increase in detail.
The high speed of data acquisition permits many lines to be run across a
site, and in some cases, total site coverage is economically feasible.
Reconnaissance work or coverage of large areas can be accomplished using
a vehicle to tow the radar antenna at speeds up to 8 KPH. Very high
resolution work or work in areas where vehicles cannot travel can be
accomplished by towing the antenna by hand at much slower speeds.
Resolution ranges from centimeters to several meters depending upon the
antenna (frequency) used.
Initial in-field analysis of the data is permitted by the picture-
like quality of the radar results. Despite its simple graphic format,
there are many pitfalls in the use of radar, and experienced personnel
are required for its operation and for the interpretation of radar data.
Radar has effectively mapped soil layers, depth of bedrock, buried
stream channels, rock fractures, and cavities in natural settings.
Radar applications to HWS assessments include:
l Evaluation of the natural soil and geologic conditions.
l Location and delineation of buried waste materials, including
both bulk and drummed wastes.
l Location and delineation of contaminant plume areas.
l Location and mapping of buried utilities (both metallic and
non-metallic).
The radar system discussed in this document is a readily available
impulse radar system. Continuous wave (CW) or other impulse systems
exist, but they are generally one of a kind, being experimental instru-
ments, and are not discussed here.
-- "3
Figure C-l shows a simplified block diagram of a radar system.
The system consists of a control unit, antenna, graphic recorder, and
an optional magnetic tape recorder. In operation, the electronics are
typically mounted in a vehicle. The antenna is connected by a cable by
hand. System power is usually supplied by a small gasoline generator.
Various antennas may be used with the system to optimize the survey
results for individual site conditions and specific requirements.
c-4
OSWER-9955.1
GRAPHIC RECORDER
ANTENNA CONTROLLER TAPE RECORDER
GROUND SURFACE -
FIGURE C-l
BLOCK DIAGRAM OF GROUND PENETRATING RADAR SYSTEM.
RADAR WAVES ARE REFLECTED FROM SOIL/ROCK INTERFACE.
C-5
ELECTROMAGNETICS (EM)! -__
The electromagnetic (EM) method provides a means of measuring the
electrical conductivity of subsurface soil, rock and ground water.
Electrical conductivity is a function of the type of soil and rock, its
porosity, its permeability, and the fluids which fill the pore space. In
most cases, the conductivity (specific conductance) of the pore fluids
will dominate the measurement. Accordingly, the EM method is applicable
both to assessment of natural geohydrologic conditions and to mapping of
many types of contaminant plumes. Additionally, trench boundaries,
buried wastes and drums, as well as metallic utility lines can be located
with EM techniques.
Natural variations in subsurface conductivity may be caused by
changes in soil moisture content, ground water specific conductance,
depth of soil cover over rock, and thickness of soil and rock layers.
Changes in basic soil or rock types, and structural features such as
fractures or voids may also produce changes in conductivity. Localized
deposits of natural organics, clay, sand, gravel, or salt rich zones will
also affect subsurface conductivity.
Many contaminants will produce an increase in free ion concentration
when introduced into the soil or ground water systems. This increase
over background conductivity enables detection and mapping of contaminaed
soil and ground water at HWS, landfills, and impoundments. Large amounts
*The term electromagnetic has been used in contemporary literature as a
descriptive term for other geophysical methods, including GPR and metal
detectors which are based on electromagnetic principles. However, this
document will use electromagnetic (EM) to specifically imply the measure-
ment of subsurface conductivites by low-frequency electromagnetic induc-
tion. This is in keeping with the traditional use of the term in the
geophysical industry from which the EM methods originated. While the
authors recognize that there are many electromagnetic systems and manu-
facturers, the discussion in this section is based solely on instruments
which are calibrated to read in electrical conductivity units and which
have been effectively and extensively used at hazardous waste sites.
OSWER-9950.1
of organic fluids such as diesel fuel can displace the normal soil
moisture, causing a decrease in conductivity which may also be mapped,
although this is not commonly done. The mapping of a plume will usually
define the local flow direction of contaminants. Contaminant migration
rates can be established by comparing measurements taken at different
times.
The absolute values of conductivity for geologic materials (and
contaminants) are not necessarily diagnostic in themselves, but the
variations in conductivity, laterally and with depth, are significant.
It is this variation which enables the investigator to rapidly find
anomalous conditions.
Since the EM method does not require ground contact, measurements
may be made quite rapidly. Lateral variations in conductivity can be
detected and mapped by a field technique called profiling. Profiling
measurements may be made to depths ranging from 0.75 to 60 meters.
Instrumentation and field procedures have been developed recently which
make it possible to obtain continuous EM profiling data to a depth of
15 meters. The data is recorded using strip chart and magnetic tape
recorders. This continuous measurement allows increased rates of data
acquisition and improved resolution for mapping small geohydrologic
features. Further, recorded data enhanced by computer processing has
proved invaluable in the evaluation of complex hazardous waste sites.
The excellent lateral resolution obtained from EM profiling data has been
used to advantage in efforts to outline closely-spaced burial pits, to
reveal the migration of contaminants into the surrounding soil, or to
delineate fracture patterns.
Vertical variations in conductivity can also be detected by the EM
method. A station measurement technique called sounding is employed for
this purpose. Data can be acquired from depths ranging from 0.75 to
60 meters. This range of depth is achieved by combining results from
c-7
a variety of EM instruments, each requiring different field application
techniques. Other EM systems are capable of sounding to depths of
1,000 feet or more, but have not yet been used at HWS and are not
adaptable to continuous measurements.
Profiling is the most effective use of the EN method. Continuous
profiling can be used in many applications to increase resolution, data
density, and permit total site coverage at critical sites.
At HWS, applications of EM can provide:
Assessment of natural geohydrologic conditions;
Locating and mapping of burial trenches and pits containing drums
and/or bulk wastes;
Locating and mapping of plume boundaries;
Determination of flow direction in both unsaturated and saturated
zones;
Rate of plume movement by comparing measurements taken at
different times; and
Locating and mapping of utility pipes and cables which may affect
other geophysical measurements, or whose trench may provide a
permeable pathway for contaminant flow.
This document discusses only those instruments which are designed
and calibrated to read directly in units of conductivity.
The basic principle of operation of the electromagnetic method is
shown in Figure C-2. The transmitter coil radiates an electromagnetic
field which induces eddy currents in the earth below the instrument.
Each of these eddy current loops, in turn, generates a secondary electro-
magnetic field which is proportional to the magnitude of the current
flowing within that loop. A part of the secondary magnetic field from
each loop is intercepted by the receiver coil and produces an output
voltage which (within limits) is linearly related to subsurface
c-a
OSWER-9950.1
Phase -
Sensing
Circuits
Ond
Amplifiers r 1 Chorl ond
Meg Topr
Recorders
FIELD
GROUND SURFACE -
> INDUCED
SECONDARY FlELDS
\
CURRENT
LOOPS FROM CURRENT LOOPS
\- /
SENSED BY
RECEIVER COiL
FIGURE C-2
BLOCK DIAGRAM SHOVING E?4 PRINCIPLE OF OPERATIONS
c-9
conductivity. This reading is a bulk measurement of conductivity; the
cumulative response to subsurface conditions ranging all the way from the
surface to the effective depth of the instrument.
The sampling depth of EM equipment is related to the instrument's
coil spacing. Instruments with coil spacings of 1. 4, 10, 20, and
40 meters are commercially available. The nominal sampling depth of an
EM system is taken to be approximately 1.5 times the coil spacing.
Accordingly, the nominal depth of response for the coil spacings given
above is 1.5, 6, 15, 30, and 60 meters.
The conductivity value resulting from an EM insrument is a
composite, and represents the combined effects of the thickness of soil
or rock layers, their depths, and the specific conductivities of the
materials. The instrument reading represents the combination of these
effects. extending from the surface to the arbitrary depth range of the
instrument. The resulting values are influenced more strongly by shallow
materials than by deeper layers, and this must be taken into
consideration when interpreting the data. Conductivity conditions from
the surface to the instrument's nominal depth range contribute about
75 percent of the instrument's response. However, contributions from
highly conductive materials lying at greater depths may have a
significant effect on the reading.
ELM instruments are calibrated to read subsurface conductivity in
millimhos per meter (mm/m). These units are related to resistivity units
in the following manner:
1000/~mi11imhos/meter) = 1 ohm-meter
1000/(millimhos/meter) = 3.28 ohm-feet
1 millimho/meter = 1 siemen
The advantage of using millimhos/meter is that the common range of
resistivities from 1 to 1000 ohm-meters is covered by the range of
conductivities from 1000 to 1 millimhos/meter. This makes conversion of
units relatively easy.
c-10
CSWER-9950.1
Most soil and rock minerals, when dry, have very low conductivities
(Figure C-3). Cn rare occasions, conductive minerals like magnetite,
graphite and pyrite occur in sufficient concentrations to greatly
increase natural subsurface conductivity. Xost often, conductivity is
overwhelmingly influenced by water content and the following soil/rock
parameters:
l The porosity and permeability of the material:
l the extent to which the pore space is saturated;
l the concentration of dissolved electrolytes and colloids in the
pore fluids; and
l the temperature and phase state (i.e., liquid or ice) of the pore
water.
A unique conductivity value cannot be assigned to a particular material,
because the interrelationships of soil composition, structure and pore
fluids are highly variable in nature.
In areas surrounding HWS. contaminants may escape into the soil and
the ground-water System. In many cases, these fluids contribute large
amouns of electrolytes and colloids to both the unsaturated and saturated
zones. In either case, the ground conductivity may be greatly affected,
sometimes increasing by one to three orders of magnitude above background
values. However, if the natural variations in subsurface conductivity
are very low, contaminant plumes of only 10 to 20 percent above
background may be mapped.
In the case of spills involving heavy nonpolar, organic fluids such
as diesel oil, the normal soil moisture may be displaced, or a sizeable
pool of oil may develop at the water table. In these cases, subsurface
conductivites may decrease causing a negative EM anomaly. (A negative
anomaly will occur only if substantial quantities of nonconductive
contaminants are present.)
c-11
I(
Cloy and Marl
Loam
Top Soil
Clayey Soils
Sandy Soils
Loose Sands
River Sand and Gravel
Glacial Till
Chalk
Limestones
Sandstones
Basalt
Crystalline Rocks
Conductivity (millimhos /meter)
IO2 IO' I IO” IO-*
FIGURE C-3
RANGE OF ELECTRICAL CONDUCTIVTIES IN NATURAL SOIL AND ROCK.
(Modified After Culley et al.)
c-12
3SWER-9950.1
RESISTIVITY
The resistivity method is used to measure the electrical resistivity
of the geohydrologic section which includes the soil, rock, and ground
water. Accordingly, the method may be used to assess lateral changes and
vertical cross sections of the natural geohydrologic settings. In
addition, it can be used to evaluate contaminant plumes and locate buried
wastes at hazardous waste sites.
Application of the method requires that an electrical current be
injected into the ground by a pair of surface electrodes. The resulting
potential field (voltage) is measured at the surface between a second
pair of electrodes. The subsurface resistivity can be calculated by
knowing the electrode separation and geometry of the electrode positions,
applied current, and measured voltage. (Resistivity is the reciprocal of
conductivity, the parameter directly measured by the EM technique.)
In general, most soil and rock minerals are electrical insulators
(highly resistive); hence the flow of current is conducted primarily
through the moisture-filled pore spaces within the soil and rock.
Therefore, the resistivity of soils and rocks is predominantly controlled
by the porosity and permeability of the system, the amount of pore water,
and the concentration of dissolved solids in the pore water.
The resistivity technique may be used for "profiling" or "sounding."
Profiling provides a means of mapping lateral changes in subsurface
electrical properties. This field technique is well suited to the
delineation of contaminant plumes and the detection and location of
changes in natural geohydrologic conditions. Sounding provides a means
of determining the vertical changes in subsurface electrical properties.
Interpretation of sounding data provides the depth and thickness of
subsurface layers having different resistivities. Commonly up to four
layers may be resolved with this technique.
c-13
Applications of the resistivity method at hazardous waste sites
include:
l Locating and mapping contaminant plumes;
l Establishing direction and rate of flow of contaminant plumes;
l Defining burial sites by
- locating trenches,
- defining trench boundaries,
- determining the depths of trenches; and
l Defining natural geohydrologic conditions such as
- depth to water table or to water-bearing horizons,
- depth to bedrock, thickness of soil, etc.
Most dry mineral components of soil and rock are highly resistive
except for a few metallic ore minerals. Under most circumstances, the
amount of soil/rock moisture dominates the mesurement greatly reducing
the resistivity value. Current flow is essentially electrolytic, being
conducted by water contained within pores and cracks. A few minerals
like clays actually contribute to conduction. In general, soils and
rocks become less resistive as:
l Moisture or water content increases;
l Porosity and permeability of the formation increases;
l Dissolved solid and colloid (electrolyte) content increases; and
l Temperature increases (a minor factor, except in areas of
permafrost).
Figure C-4 illustrates the range of resistivity found in commonly-
occurring soils and rocks. Very dry sand, gravel, or rock as encountered
in arid or semi-arid areas will have very high resistivity. As the empty
pore spaces fill with water, resistivity will drop. Conversely, the
resistivity of earth materials which occur below the water table but lack
pore space (such as massive granite and limestone) will be relatively
high and will be primarily controlled by current conduction along cracks
c-14
OSWER-9950.1
Cloy and Marl
Loom
Top Soil
C I0 yey Soils
Sandy Soils
loose Sands
River Sond ond Grovet
Gtociat Till
Chalk
Limestones
Sandstones
eosott
Crystalline Rocks
I
Resistivity (ohm - meters )
10’ JO2 IO5 IO’ IO3 IO”
FIGURE C-4
RANGE OF RESISTIVITIES IN COMMONLY-OCCURRING SOILS AND ROCKS
(Modified after Culley et al.>
c-15
and fissures in the formation. Clayey soils and shale layers generally
have low resistivity values, due to their inherent moisture and clay
mineral content. In all cases, an increase in the electrolyte, total
dissolved solids (TDS) or specific conductance of the system will cause a
marked increase in current conduction and a corresponding drop in
resistivity. This fact makes resistivity an excellent technique for the
detection and mapping of conductive contaminant plumes.
It is important to note that no geologic unit or plume has a unique
or characteristic resistivity value. Its measured resistivity is
dependent on the natural soil and rock present, the relative amount of
moisture, and its specific conductance. However, the natural resistivity
value of a particular formation or unit may remain within a small range
for a given area.
Figure C-5 is a schematic diagram showing the basic principles of
operation. The resistivity method is inherently limited to station
measurements, since electrodes must be in physical and electrical contact
with the ground. This requirement makes the resistivity method slower
than a noncontract method such as EM.
Many different types of electrode spacing arrays may be used to
make resistivity measurements; the more commonly used include Wenner,
Schlumberger, and dipole-dipole. Due to its simple electrical geometry,
the Wenner array will be used as an example in the remainder of this
section; however, its use is not necessarily recommended for all site
conditions. The choice of array will depend upon project objectives and
site conditions and should be made by an experienced geophysicist.
Using the Wenner array, potential electrodes are centered on a line
between the current electrodes; and equal spacing between electrodes is
maintained. These "A" spacings used during HWS evaluation commonly range
from 0.3 meter to more than 100 meters. The depth of measurement is
related to the "A" spacing and may vary depending upon the geohydroiogy.
C-16
OSWER-9950.1
Current
face
Current Flow
Through Earth
Apparent resistivity values using the Uenner array are calculated
from the measured voltage and current and the spacing between electrodes
as shown in the following equation:
a=2 AV/I
where a = apparent resistivity (ohm-meters or ohm-feet)
A = “A” Spacing (meters or feet)
v = potential (volts)
I = current tampers)
FIGURE C-5
DIAGRAM SHOVING BASIC CONCEPT OF RESISTIVITY MEASUREMENT
c-17
Current is injected into the ground by the two outer electrtodes
which are connected by cables to a DC or low-frequency AC current source.
(If true DC is used, special nonpolarizing electrodes must be used.) The
distribution of current within the earth is influenced by the relative
resistivity of subsurface features. For example, homogenous subsurface
conditions will have the uniform current flow distribution and will yield
a resistivity value characteristic of the sampled section. On the other
hand, current distribution may be pulled downward by a low-resistivity
(lower than that of the surface layer, due to the influence of the lower
resistivity material at depth.
The current flow within the subsurface produces an electric field
with lines of equal potential, perpendicular to the lines of current
(Figure C-5). The potential field is measured by a voltmeter at the two
inner electrodes.
C-18
OSWER-9950.1
SEISMIC REFRACTION --
Seismic refraction techniques are used to determine the thickness
and depth of geologic layers and the travel time or velocity of seismic
waves within the layers. Seismic refraction methods are often used to
map depths to specific horizons such as bedrock, clay layers, and water
table. In addition to mapping natural features, other secondary
applications of the seismic method include the location and definition of
burial pits and trenches at HWS.
Seismic waves transmitted into the subsurface travel at different
velocites in various types of soil and rock and are refracted (or bent)
at the interfaces between layers. This refraction affects their path of
travel. AII array of geophones on the surface measures the travel time of
the seismic waves from the source to the geophones at a number of
spacings. The time required for the wave to complete this path is
measured, permitting a determination to be made of the number of layers,
the thicknesses of the layers and their depths, as well as the seismic
velocity of each layer. The wave velocity in each layer is directly
related to its material properties such as density and hardness.
A seismic source, geophones. and a seismograph are required to make
the measurments. The seismic source may be a simple sledge hammer with
which to strike the ground. Explosives and any other seismic sources may
be utilized for deeper or special applications. Geophones implanted in
the surface of the ground translate the received vibrations of seismic
energy into an electrical signal. This signal is displayed on the
seismograph, permitting measurement of the arrival time of the seismic
wave. Since the seismic method measures small ground vibrations, it is
inherently susceptible to vibration noise from a variety of natural and
cultural sources.
At HWS, seismic refraction can be used to define natural geohydro-
logic conditions, including thickness and depth of soil and rock layers,
C-19
their composition and physical properties, and depth to bedrock or water
table. It can also be used for the detection and location of anomalous
features. such as pits and trenches, and for evaluation of the depth of
burial sites or landfills. (In contrast to seismic refraction, the
reflection technique, which is common in petroleum exploration, has not
been applied to HWS. This is primarily because the method cannot be
effectively utiiized at depths of less than 20 meters.)
Although a number of elastic waves are inherently associated with
the method, conventional seismic refraction methods that have been
employed at XWS are concerned only with the compressional wave (primary
or P-wave). The compressional wave is also the first to arrive which
makes its identification relatively easy.
These waves move through subsurface layers. The density of a layer
and its elastic properties determine the speed or velocity at which the
seismic wave will travel through the layer. The porosity, mineral
\:omposition, and :Jater content of the layer affect both its density and
elasticity. Table Z-1 lists a range of compresslonal wave velocities in
~01~01~ geologic materials. It can be seen from these tables that the
s;;~sr;~ic velocities for different types of soil and rock overlap, so
,rr~ow~rlg the velocities of these layers alone does not permit a unique
determination of their composition. However, Lf this knowledge is
.:;jmb~r:id with geologic Information, it can be used intelligently to
;.:~i:t~fy geologic strata.
in general, velocity values are greater for:
l dense rocks than light rocks.
. older rocks than younger rocks.
4 igneous rocks than sedimentary rocks.
4 solid rocks than rocks with cracks or fractures.
5-20
OSWER-9950.1
TABLE C-l
RANGE OF VELOCITIES FOR COMPRESSIONAL WAVES IN SOIL AND ROCK
(After Sakosky, 1950)
Material Velocity (Meters/set)
Weathered surface material
Gravel or dry sand
Sand (wet)
Sandstone
Shale
Chalk
Limestone
Salt
Granite
Metamorphic rocks
305 - 610
465 - 915
610 - 1,830
1,830 - 3,970
2,750 - 4,270
1,830 - 3,970
2,140 - 6,100
4,270 - 5,190
4,380 - 5,800
3,050 - 7,020
C-21
l unweathered rocks than weathered rocks.
0 consolidated sediments than unconsolidated sediments.
l water-saturated unconsolidated sediments than dry unconsolidated
sediments.
0 wet soils than dry soils.
Figure C-6 shows a schematic view of a 12-channel seismic system in
use and the compessional waves traveling through a two-layered system of
so11 over bedrock. A seismic source produces seismic waves which travel
in all directions into the ground. The seismic refraction method,
however, is concerned only with the waves shown in Figure C-6. One of
these waves, the direct wave, travels parallel to the surface of the
ground. A seismic sensor (geophone) detects the direct wave as it moves
along the surface layer. The time of travel along this ?ath is related
to the distance between the sensor and the source and the material
composing the layer.
If a denser layer with a higher velocity, such as bedrock, exists
below the surface soils, some of the seismic waves will be bent or
refracted as they enter the bedrock. This phenomenon is similar to the
refraction of light rays when light passes from air into water and is
described by Snell's law. One of these refracted waves, crossing the
interface at a critical angle. will move parallel to the top of the
bedrock at the higher velocity of the bedrock. The seismic wave
travelling along this interface will continually release energy back into
the upper layer by refraction. These waves may then be detected in the
surface at various distances from the source (Figure C-6).
Beyond a certain distance (called the critical distance), the
refracted wave will arrive at a geophone before the direct wave. This
happens even though the refraction path is longer, because a sufficient
portion of the wave's path occurs in the higher velocity bedrock.
c-22
OSWER-9950.1
Geophone A way
FIGURE C-6
FIELD LAYOUT OF A 12-CXANNEL SBISHOGRAF'H SHWING THE PATH
OF DIRECT AND REFRACTED SEISMIC WAVES IN A TW-LAYER SOIL/ROCK SYSTEM
C-23
Measurement of these first arrival times and their distances from the
source permits calculation of layer velocities, thicknesses and bedrock
depth. Application of the seismic method is generally limited to
resolving three to four layers.
The preceding concepts are based upon the fundamental assumptions
that:
1. Seismic velocities of geologic layers must increase with depth.
This requirement is generally met at most sites.
2. Layers must be of sufficient thickness to permit detection.
3. Sesimic velocities of layers must be sufficiently different to
permit resolution of individual layers.
There is no way to establish from the seismic data alone whether a hidden
layer (due to 1 and 2 above) is present; therefore, correlation to a
boring log or geologic knowledge of the site must be used to provide a
cross check. If such data is not available, the interpreter must take
this into consideration in evaluating the data.
Variations in the thickness of the shallow soil zone, inhomo-
geneities within a layer, or irregularities between layers will often
produce geologic scatter or anomalies in the data. This data scatter
is useful information, revealing some of the natural variability of the
site. For example, a zone containing a number of large boulders in a
glacial till deposit will yield inconsistent arrival times, due to
variable seismic velocities between the boulders and the clay matrix.
An extremely irregular bedrock surface as is often encountered in karst
limestone terrain, likewise, will produce scatter in the seismic data.
The seismic refraction technique uses the equipment shown in
Figure C-6. The seismic source is often a simple ten-pound sledge hammer
or drop weight which strikes the ground, generating a seismic impulse.
Explosives and a variety of other excitation sources are also used for
the greater energy levels rquired for information at deeper layers.
C-24
OSWER-9950.1
Seismic waves are detected by geophones implanted in the surface of
the ground at various distances from the source. The geophone converts
the seismic wave's mechanical vibration into an electrical signal in a
manner similar to that of a microphone. This signal is carried by cable
to the seismograph.
The seismograph is an instrument which electronically amplifies and
then displays the received seismic signal from the geophone. The display
may be a cathode ray tube, a single-channel strip chart, or a thermal
printer, commonly used on multi-channel systems. The identification and
measurement of the arrival time of the first wave from the seismic source
is obtained from this presentation. The time is measured in milliseconds,
with zero time or start of trace intitiated by the source, which provides
a trigger signal to the seismograph.
Travel time is plotted against source-to-geophone distance producing
a time/distance (T/D) plot.
l The number of line segments indicates the number of layers.
l The slope of each line segment is inversely proportional to the
seismic velocity in the corresponding layer.
l Break points in the plot (critical distance, X) are used with the
velocities to calculate layer depth.
The seismic line must be centered over the required information area
and overall line length must be three to five times the maximum depth of
interest. Resolution is determined by the geophone spacing. Spacings of
3 to 15 meters are commonly used; however, closer spacings may be
necessary for very high resolution of shallow geologic sections.
c-25
ORGANIC VAPOR/SOIL GAS ANALYSIS
Organic contaminant vapors present in the vadose zone may be
assessed using a variety of techniques. One method is the use of organic
vapor detectors such as OVAs, explosimeters and Draeger tubes to detect
volatile organics. Two major strategies may be adopted, jointly or
separately, depending on whether wells are in place at the time of
investigation:
:. Monitoring the well head space.
2. Monitoring the vadose zone directly by lowering a probe into
shallow, hand-augurred holes.
Gaseous sample constituents can be identified in detail using a
portable gas chromatograph. An alternative methodology is an analysis of
soil gas. Under this methodology, a ten-liter sample of soil gas is
drawn through a probe which is mechanically driven into the ground to a
depth of about ten feet. Two cubic centimeters of gas are injected into
a portable gas chromatograph to ascertain its organic constituents. It
is useful to know what class of organics is present in order to choose
the gas chromatography method which provides the highest resolution,
i.e., photoionization/aromatics, electron-capture/halogenated hydro-
carbons. The 2 cc sample is injected by syringe to the gas chromatograph
through a dewatering napthalon tubing. This method is limited in two
major ways:
1. Coarse, pe;bly/cobbly strata prevent penetration of the probe,
in which case holes may be hand-augured.
2. The presence of shallow, saturated zones, especially low
permeability formations severely restricts the development of a
gas envelope and thus limits the applicability of the method.
Soil gas analysis is a vadose zone monitoring technique and
cannot be used effectiveiy where the water table or saturation
is shallow.
Organic vapor/soil gas analysis is most effective when used in
conjunction with other investigat:ve methods. Although it provides an
OSWEIR-9950.1
analysis of the volatile organics and thus provides a preliminary
characterization of the subsurface contamination, it is limited to a
fraction of the total hazardous constituents and needs augmentation.
C-27
k U.S. COVtHNMiS’T PXITTISC @I;FICi: 1986: 62!-735/60541