United States Robert S. Kerr EPA/600/R-94/205
Environmental Protection Environmental Research Laboratory
Agency Ada, OK 74820
Research and Development
v>EPA Ground Water Sampling -
A Workshop Summary
Dallas, Texas
November 30 - December 2, 1993
-------
EPA/600/R-94/205
Ground Water Sampling —
A Workshop Summary
Dallas, Texas
November 30 - December 2,1993
Robert S. Kerr Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Ada, OK 74820
-------
Notice
The information in this document has been funded wholly or in part by the United States Environmental Protection Agency.
It has been subjected to the Agency's peer review and administrative review, and it has been approved for publication as an
EPA document.
All research projects making conclusions or recommendations based on environmentally related measurements and funded
by the Environmental Protection Agency are required to participate in the Agency Quality Assurance Program. This project
did not involve environmentally related measurements and did not involve a Quality Assurance Project Plan. Mention of
trade names or commercial products does not constitute endorsement or recommendation for use.
-------
Foreword
EPA is charged by Congress to protect the Nation's land, air and water systems. Under a mandate of national environmental
laws focused on air and water quality, solid waste management and the control of toxic substances, pesticides, noise and
radiation, the Agency strives to formulate and implement actions which lead to a compatible balance between human
activities and the ability of natural systems to support and nurture life.
The Robert S. Kerr Environmental Research Laboratory is the Agency's center of expertise for investigation of the soil and
subsurface environment. Personnel at the laboratory are responsible for management of research programs to: (a) determine
the fate, transport and transformation rates of pollutants in the soil, the unsaturated and the saturated zones of the subsurface
environment; (b) define the processes to be used in characterizing the soil and subsurface environment as a receptor of
pollutants; (c) develop techniques for predicting the effect of pollutants on ground water, soil, and indigenous organisms; and
(d) define and demonstrate the applicability and limitations of using natural processes, indigenous to the soil and subsurface
environment, for the protection of this resource; and (e) provide technical assistance to characterize and remediate
contaminated soils and ground water.
The dissemination, review and implementation of new environmental research findings is essential in providing the
background information required for practitioners and policy makers working in the areas of environmental protection and
restoration. This information is critically needed in the area of ground-water sampling, where recent improvements in
technology have far outpaced our ability to evaluate and incorporate new methods into sampling protocols and technical
guidance documents.
The primary objective of this workshop was to provide a forum for the presentation and discussion of recent research
findings on ground-water sampling for researchers, practitioners, regulators and policy makers. Secondary objectives were:
to improve communication and the transfer of information between these diverse groups, encourage consistency in ground-
water sampling programs where appropriate, and identify research and technology transfer needs. Participants included
representatives of universities, private industry, environmental consultants, the U.S. Environmental Protection Agency, the
U.S. Geological Survey, the Department of Energy, and several state environmental agencies.
Clinton W. Hall
Director
Robert S. Kerr Environmental
Research Laboratory
111
-------
IV
-------
Contents
Notice ii
Foreword iii
1. Executive Summary 1
2. Agenda 5
3. Extended Abstracts 7
4. Poster Abstracts 59
5. Small Group Discussions 67
Appendices
Appendix A: Steering Committee and Other Attendees
Appendix B: Questions to Consider for Small Group Discussions
Appendix C: Bibliography
Appendix D: Glossary
Appendix E: Sampling Forms
-------
VI
-------
I. Executive Summary
Monitoring purposes have evolved in the past decade
reflecting an increasing emphasis on assessment and
remedial action performance over detection monitoring. The
recognition of many potential contamination sources as well
as the regulatory inclusion of a large number of sites in need
of investigation have spurred this evolution in practice. New
more efficient methods and tools for hydrogeochemical
characterization have been developed which can make site
investigations more reliable and cost-effective. Improved
field screening techniques, geophysical, portable analytical,
and computerized hydrologic data acquisition systems,
lighter drilling and boring tools, multi-level samplers, and
hydraulic "push" technologies are some of the tools which
may be applied to subsurface studies. Professional practices
and standards for all monitoring have lagged behind
methodological research and development. Regulatory
acceptance of new tools and methods has been slow in
general. However change is inevitable and, if properly
implemented, could lead to more uniformly reliable and less
expensive investigations. Improved training and
certification of hydrogeologic professionals could also aid in
this regard. Decreased reliance on wells as primary data
collection points follows on the recognition that many
inorganic and most organic contaminants have an
appreciable fraction of their mass associated with subsurface
solids. Also, wells designed for detection and assessment
monitoring purposes frequently become obsolete when
vadose or saturated zone remedial action has begun. It may
be anticipated that as the familiarity with and supporting
documentation for more effective methods grow, more
stringent standards of professional practice will evolve as
well. In summary, the state of current monitoring practice is
steadily changing to more reliable, cost-effective techniques
which should improve the spatial coverage, accuracy and
precision of data collection efforts.
The primary objective of this workshop was to provide a
forum for the presentation and discussion of recent research
findings on ground water sampling in the context of
continuing advances in environmental monitoring
technologies and changing environmental regulatory
requirements. Participants included researchers,
practitioners, regulators and policy makers from
governmental and non-governmental sectors. Secondary
objectives of the workshop were: to improve
communication and the transfer of information between
these diverse groups, encourage consistencey in ground
water sampling programs where appropriate, and identify
research and technology transfer needs.
Ground-Water Monitoring Goals and Objectives
One of the most important steps in the design and
implementation of a ground-water monitoring program
involves defining the data, analyses, and information
required to meet the monitoring program objectives.
Common examples of monitoring objectives may include
leak detection at hazardous and solid waste land disposal
facilities, hazardous waste site assessments, corrective
action evaluations, and ground-water resource evaluations.
In a broad sense, monitoring objectives define the data or
information that the investigator needs to support decisions
or conclusions. Monitoring objectives are developed to
satisfy regulatory requirements, support resource
assessments and research goals, and support the
development of site conceptual models.
In defining monitoring objectives, data quality objectives
(DQOS) are used to specify the type and quality of data
required to support decisions. DQOS describe the overall
level of uncertainty that a decision-maker is willing to
accept in results derived from environmental data. This
uncertainty is used to specify the quality of the measurement
data required, usually in terms of objectives for accuracy,
precision, bias, representativeness, comparability, and
completeness. DQOS are defined prior to the initiation of
the field and laboratory work. Also, DQOS are
communicated to field and laboratory personnel performing
the work to make informed decisions during the course of
the project to attain those DQOS. The procedures used to
characterize the hydrogeology of a site, to design and
construct a monitoring network, to collect and analyze
environmental samples, and to evaluate analytical results
should ensure that the data are of the type and quality
necessary to meet the objectives of the monitoring program.
The development and refinement of a site conceptual model
underlies all ground-water monitoring programs. A
conceptual model is an understanding of the hydrogeologic
characteristics of a site, and of how the hydrogeologic
characteristics (e.g., geology and geochemistry of the site
and the distribution of contaminant migration pathways) are
integrated into a hydrogeologic system that contains
interacting and dynamic components. Development of a site
conceptual model is an iterative process. After a ground-
water monitoring system has been installed and numerous
ground-water samples have been collected, the conceptual
model for a site may be further refined, depending on the
DQOS.
-------
Design and Installation of Subsurface Sampling Points
The design and installation of sampling points for
subsurface investigations and monitoring have undergone
substantial change since the days when wells of one type or
another were the dominant data collection platforms. A
variety of gas, solid and liquid sampling techniques are in
use for field screening and to improve spatial coverage of
subsurface conditions, particularly for volatile organic
contaminants. These techniques may effectively substitute
for permanent sampling installations. They can be employed
during drilling or in conjunction with various "push"
technologies which rely on hydraulic or percussion
advancement principles.
The utilization of soil gas probes, piston-coring devices, and
"drill-tool" water sampling devices (e.g.," Hydropunch",
screened augers/pumps, etc.) during drilling has led to an
improved ability to identify hydrostratigraphic units and
zones of contamination. In these instances, the location,
design, construction and performance of short-screened (i.e.,
<2m), narrow diameter monitoring wells or multi-level
samplers can enable more focused and reliable
hydrogeochemical data collection. Fewer drill holes and less
disturbance of subsurface conditions result from the use of
these emerging techniques which also reduce the time and
cost involved in monitoring efforts.
Wells are traditionally installed to address one or more of
the following objectives: regulatory monitoring; water level
measurement; hydraulic conductivity estimation; or to
measure or evaluate some other hydrogeochemical site
characteristic. Generally speaking, wells are designed to
yield enough water for sample collection for regulatory
purposes; however, this is not always practical in low water-
yielding formations. Conventional monitoring well designs
are generally inadequate in these types of formation and
alternative designs are needed together with sampling
methodologies which do not introduce unwanted artifacts in
the collected samples.
Well Purging and Sampling
Traditional approaches to purging and sampling ground
water are undergoing significant reevaluation. The most
common method involves purging a well at a high pumping
rate or bailing, until a fixed number of casing volumes
(usually 3-5) is evacuated, followed by sample collection.
This approach has raised concerns about the
representativeness of samples collected using these
methods, especially if the sampling objectives include
monitoring of contaminants. Concerns include entrainment
of immobile particles and the possible need to filter samples
to remove those artifacts, costs of pumping and disposing of
large volumes of contaminated water, and uncertainties in
interpreting the source of the sampled water. Many of the
Workshop participants recommended a method which is
referred to as low-flow (with minimal-drawdown) purging
and sampling. The principal differences between this and
more traditional approaches centers on the rate of pumping
and the criteria for deciding that purging is complete. The
newer method calls for slow flow rates for purging and
sampling in order to minimize chemical and hydrological
disturbance in and around the well. Furthermore, the
completion of purging is gauged on site-specific criteria
(stabilization of water quality parameters) rather than on a
fixed number of well volumes pumped. Conceptually,
formation water flowing through the screened section of the
well is purged and then sampled no faster than it enters the
well bore under natural hydrological flow conditions. The
criteria for the appropriate rate for purging and sampling is
hydrological: pumping rates should produce no net (or at
least minimal) drawdown of the water table. Under these
conditions, low-flow sampling removes water from only the
screened zone and stabilization of water quality indicator
parameters can be used as a criteria for deciding when
formation water has been accessed and sampling can begin.
The ideal endpoint would be stabilization of the
concentration of a contaminant (or other species of interest);
however, some indicator parameters may be correlated with
different classes of contaminants (e.g., dissolved oxygen
with volatile organics, and turbidity with metals). A
conservative approach was recommended which included
the use of dissolved oxygen and specific conductance for
volatile contaminants, turbidity and specific conductance for
metals (and metalloids) and hydrophobic organics, and
perhaps also the use of oxidation-reduction (redox)
potentials for both cases.
There are several advantages to the low-flow, minimal
drawdown method, including the following:
• More Representative Samples — Minimal
disturbance of the sampling point and reduced stress
on the formation results in low turbidity samples
which are representative of the "mobile" load of
contaminants (dissolved and colloidassociated)
present in the formation. By minimizing "artifactual"
turbidity, this method also reduces the need for
filtration of samples and the costs of analyses of
filtered and unfiltered samples.
• Waste minimization— The volume of purge water
required to access formation water is much less than
for more traditional purging methods because low-
flow sampling conditions remove water from only
the screened zone. This minimizes the volume of
water that will require waste disposal at
contaminated sites.
• Spatial Resolution in Sampling —A fundamental
benefit of the decreased purge volume is that a
-------
smaller volume of the aquifer is sampled. This
represents a significant improvement in our ability to
detect and resolve contaminant distributions, which
may vary greatly over small distances in three-
dimensional space. In fractured clay or rock, for
example, most of the water comes from the fractures.
Because the fracture porosity is so small, the
sampling process may draw in water from a very
large volume of the deposit. This could greatly dilute
the concentration of contaminants in the sample.
Analogous flow problems can occur in granular
aquifers due to vertical heterogeneities in hydraulic
conductivity in typically layered sediments.
Switching to the low-flow sampling method met with some
resistance on the part of sampling practitioners because of
potential problems with data comparison and interpretation
of temporal trends due to differences in sample collection
methods. There was also concern that increased time may be
required for purging and sampling at slow flow rates, and
additional equipment may be needed. It was also questioned
whether the method was practical in tight or low water-
yielding formations where monitoring wells are often urged
to dryness. However, recognizing the disadvantages, the
advantages of the low-flow sampling method in providing a
higher quality sample that more closely represents the
mobile dissolved and colloidal components in the formation
suggest that this is the direction in which the state of
professional practice must proceed. In the case of low
water-yielding formations, extension of the low-flow
method to more "passive" sampling is proposed for further
study as an alternative to current practices in such geologic
settings.
Colloidal Transport and Ground-Water Sampling
Recognition of the potential role of colloids in facilitating
contaminant transport has heightened awareness of the need
to obtain ground-water samples that are representative of
naturally mobile colloids. Carefully collected field evidence
shows that commonly-used sampling protocols (bailing,
rapid pumping) produce ground-water samples in which
colloids have been artificially entrained. That is, bailed and
rapidly pumped samples often contain substantial turbidity
that is not representative of conditions within the
subsurface. In practice, the suspended particles causing this
turbidity have been removed from the samples by filtering
in the field. Usually, membrane filters with 0.45 m pores
have been used to remove turbidity despite biases
introduced by their use. The consensus of the workshop
endorsed low-flow sampling techniques (described above)
which make it possible to obtain ground-water samples that
are relatively free of turbidity, without resorting to filtering,
by withdrawing water at relatively slow rates which induce
little or no drawdown of the water level in a monitoring
well. The use of the term "low-stress" sampling in place of
low-flow was proposed to emphasize the importance of
minimizing disturbance or stress to the subsurface system.
Concern about the potential effect of colloid-facilitated
transport is limited to certain classes of low-volubility,
surface-reactive contaminants, including radionuclides (e.g.,
Pu, Am, U, Co, Sr, Cs), heavy metal cations (Cu, Pb),
inorganic contaminants typically in anionic form (Cr(VI),
As), as well as high molecular weight organic compounds
(polychlorinated biphenyls, polycyclic aromatic
hydrocarbons). For other contaminants, such as low
molecular weight, non-surface-reactive contaminants (e.g.,
volatile organic compounds), the significance of colloids
can be deemed insignificant, based on published partitioning
values. Further research is needed to clearly identify what
types of subsurface environments pose the greatest risk for
this mode of contaminant transport.
Filtration and Sample Handling
The decision to collect filtered or unfiltered ground-water
samples should be based on ground-water monitoring and
data quality objectives. Field filtration should not
necessarily be the default choice. Consideration must be
given to what the application of field filtration is attempting
to accomplish, and filtration should not be used as a
corrective measure for poor sampling practices. To estimate
truly dissolved concentrations of elemental species, the
smaller the filter pore size the better, with 0.1 pm generally
the most practical for field usage. Dissolved concentrations
of elemental species are required for geochemical modeling
purposes and other determinations (e.g., alkalinity). In-line
field filtration is considered the most desirable approach to
minimize sample handling, sample exposure to the
atmosphere, and facilitate the expeditious transfer of the
water sample to the sample container. For assessment of
colloidal mobility, samples must not be filtered. This places
more importance on the manner in which samples are
collected and the efforts taken to exclude non-mobile or
artifactual suspended particles from the sample.
Compatibility of unfiltered samples with the eventual
analytical method imposed on the sample must be
considered. Unfiltered samples collected to evaluate the
"mobile" contaminant loading (i.e., dissolved + colloidal)
and having high turbidity must undergo digestion prior to
analysis. This issue requires special attention in the site-
specific sampling plan.
Sample transfer from the sampling device to the sample
container should be accomplished with as little disturbance
as possible. This operation is relatively straightforward with
low-flow sampling techniques, however is problematic
using bailers. The excessive volume of sample required for
different regulatory programs was also seen as a major
impediment to improving our ability to collect
-------
"representative" samples, particularly in low water-yielding
formations. Wells in these geologic settings may recharge at
extremely slow rates, requiring hours of sampling time to
fulfill regulatory sample volume requirements. In many
cases, current analytical requirements for sample volumes
are considerably less than "regulatory" requirements which
are based on outdated analytical methods and
instrumentation. Advantages of reduced sample volume
requirements are: lower sample container filling time at
lower flow rates, lower bottle costs, lower shipping costs,
and reduction of liquid wastes back at the laboratory.
Documentation and Technology Transfer
Despite the fact that more focused, real-time
hydrogeochemical data collection has been enabled by
recent technological advances, the growth of the knowledge
base and improvements in field sampling practice are
seriously lagging. Additional and better documentation is
required at virtually every step of the sampling process.
This will permit better interpretation of the collected data
and more assurance that data quality objectives are being
met. Improvements in technology transfer from the research
community to the user communities is also needed. Training
and perhaps certification of field sampling personnel should
be encouraged and efforts should be expanded in this area
by state and federal regulatory agencies.
Summary edited by Robert W. Puls, Robert S. Kerr
Environmental Research Laboratory, U.S. Environmental
Protection Agency and written by Michael J. Barcelona,
University a/Michigan, John E McCarthy, Oak Ridge
National Laboratory, and James R. Brown and Robert W.
Puls, U.S. Environmental Protection Agency.
-------
II. Agenda
TUESDAY. NOVEMBER 30. 1993
7:30 - 8:00 a.m. General Introduction
8:00 - 8:30 a.m. A Federal Perspective on the Acquisition of Representative
Ground Water Samples: A Historical Review
8:30 - 9:00 a.m. Sampling Program Purpose and Design Considerations
9:00 - 9:30 a.m. The Relationship of Monitoring Well Design, Construction, and
Development to Turbidity in Wells, and Related Implications for
Ground Water Sampling
9:30 -10:00 a.m. Use of Low-Flow or Passive Techniques for Sampling
Ground Water
Robert W. Puls
James R. Brown and
Andrew L. Teplitzky
Michael J. Barcelona
David M. Nielsen
Robert W. Puls
10:15 - 10:45 a.m. Sampling Colloids and Colloid-Associated Contaminants
in Ground Water
Debera A. Backhus,
J. N. Ryan, D.M. Groher,
J.K. MacFarlane, and
P.M. Gschwend
10:45 - 11:15 a.m. Need for Practical Approaches to Conduct Multilevel Sampling
11:15 - 12:45 p.m. Effects of Well Design and Time of Pumping on Concentrations of
Volatile Organic Compounds in Ground Water Samples
11:45 a.m. Lunch
1:00-1:30 p.m. Common Sampling and Analytical Procedures Viewed in the
Context of Data Quality Needs
1:30 - 2:00 p.m. Considerations in Selecting Filtered or Unfiltered Samples for
Analyses of Metals in Ground Water Samples
2:00 - 2:30 p.m. A Study of the Impact of Monitoring Well Purging and Filtering
Techniques on Metals Concentrations in Ground Water Samples
from the Auburn Road Landfill Site in Londondery, N.H.
Gary A. Robbins
Jacob Gibs
and Thomas E. Reilly
William R. Mabey
and Nancy Barnes
Dennis E. Reece
Carol A. White
2:30 - 3:00 p.m. Ground Water Sampling of Fractured Clay and Rock
3:30 - 4:00 p.m. Evaluation of Field-Filtration Variables for Representative Samples
of Trace Metals in Ground Water
4:00 - 4:30 p.m. Monitoring Well Sampling—You Can't Always Get What You
Want, But Can You Get What You Need?
Larry McKay,
Kent Novakowski, and
John McCarthy
Karl F. Pohlmann,
Gary A. Icopini, and
Charlita G. Rosal
Jack Connelly
-------
4:30 - 5:00 p.m. Abstract of Statistical Comparison of Metal Concentrations in
Filtered and Unfiltered Ground Water Samples
5:30 - 7:30 p.m. Poster Session
Robert Gibbons
and Martin Sara
WEDNESDAY. DECEMBER 1. 1993
8:00-11:30 a.m.
ll:30a.m.
1:30-5:00 p.m.
Small Group Discussions
Monitoring Goals and Objectives
Well Design, Construction, and Development
Well Purging and Sampling
Turbidity and Colloid Transport
Sample Handling and Analysis
Lunch
Small Group Discussions
Monitoring Goals and Objectives
Well Design, Construction, and Development
Well Purging and Sampling
Turbidity and Colloid Transport
Sample Handling and Analysis
THURSDAY. DECEMBER 2. 1993
8:00 - 11:00 a.m. Reports from Small Group Discussions
11:00-Noon
Wrap-Up
Michael J. Barcelona,
Linda Aller, Robert W.
Puls, Joseph N. Ryan, and
Karl Pohlmann
Robert W. Puls
-------
III. Extended Abstracts
The following extended abstracts are provided for the oral presentations made by invited speakers on the first day of the
Workshop. Speakers were selected by the Steering Committee and the topics were intended to provide some foundation for
subsequent discussions which occurred both formally and informally during the balance of the Workshop. The first day's
presentations covered a wide range of sampling-related topics from a diverse group of individuals active in ground-water
sampling. It should be noted that the abstracts were not part of the formal peer-review process which the remainder of this
document underwent, and represent opinions or personal points of view in many cases. As a result, some presentations are in
conflict with others. An attempt was purposely made to provide a forum for the exchange of conflicting views and data on
the first day, and thus stimulate discussions for the remainder of the Workshop.
-------
A REVIEW OF FEDERAL
REGULATIONS AND GUIDANCE
FOR GROUND-WATER SAMPLING
James R. Brown and Andrew L. Teplitzky
Several Federal regulations mandate the collection and
analysis of ground-water samples to protect ground-water
resources and to determine the anthropogenic impacts on
ground-water quality: The Resource Conservation and
Recovery Act (RCRA); the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA); the
Toxic Substances Control Act (TSCA); and the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA).
Annually, these regulations result in the collection of
hundreds of thousands of ground-water samples. Some of
the regulations use general performance standards for
ground-water sample collection requirements. Other
regulations require specific procedures for the collection of
ground-water samples.
The RCRA program regulates, in part, owners and
operators of municipal solid waste landfills (MSWLFs),
hazardous waste management facilities, and underground
storage tanks. Ground-water monitoring regulations at
RCRA MSWLFs prohibit the filtration of ground-water
samples. RCRA hazardous waste ground-water
monitoring regulations and underground storage tank
standards do not specify whether the samples should be
filtered or not. Implementation of specific sampling
protocols for hazardous waste disposal facilities and
underground storage tanks is done at the State or EPA
Region level, mainly through technical guidance
documents. EPA's ground-water sampling guidance for
hazardous waste disposal facilities is found in the RCRA
Ground-Water Monitoring Technical Enforcement
Guidance Document (TEGD) (USEPA, 1986), and RCRA
Ground-Water Monitoring: Draft Technical Guidance
(GWM) (USEPA, 1992).
The TEGD recommends:
1) the use of sampling equipment and procedures that
minimize sample agitation and reduce or eliminate
contact with the atmosphere;
2) that low-yielding wells be purged at a rate that does
not cause recharge water to be excessively agitated;
and
The TEGD also recommends that samples collected for the
analysis of organic compounds not be filtered. However,
samples collected for the analysis of metals should be split
into two portions: one portion should be filtered and
analyzed for dissolved metals, and one portion should not be
filtered and analyzed for total metals.
GWM recommends the use of low rate purging and
sampling procedures, discourages the use of bailers, and
generally does not recommend the filtration of ground-water
samples used for leak detection purposes.
The CERCLA program (commonly referred to as
"Superfund") regulates the remediation of uncontrolled and
abandoned hazardous waste sites. The regulations specify
only general performance standards for collecting ground-
water samples. Several guidance documents, however,
prescribe ground-water sampling procedures for various
sampling objectives.
A Compendium of Superfund Field Operations Methods
(USEPA, 1987), recommends that the sampling protocol be
dictated by the study objectives. The Compendium states
that if the study objective is to assess the migration
mechanisms in conjunction with migration pathways, then it
is necessary to know the concentration of dissolved and total
constituents.
The Risk Assessment Guidance for Superfund (USEPA,
1989), notes that data from filtered and unfiltered
ground-water samples are useful for evaluating chemical
migration in ground water, and that a comparison
between these samples can provide important
information on the form in which a chemical exists in
ground water. The Risk Assessment Guidance also
specifies the use of data from unfiltered samples for
estimating exposure concentrations.
A Superfund Ground-Water Issue Paper entitled, "Ground-
Water Sampling for Metals Analysis," (Puls and Barcelona,
1989), also makes several ground-water sampling
recommendations for the Superfund program. The paper
recommends:
3) that sampling rates for volatile organic
compounds not exceed 100 milliliters per
minute.
1) the collection of unfiltered samples to determine a
conservative estimate of contaminant loading in an
aquifer;
-------
2) the use of small pore-size filters (<0.45 micron) to
determine the concentration of dissolved metals;
3) the use of low-rate purging and sampling
4)
me use 01 low-rare purging ana sampling
procedures (e.g., 100 milliliters per minute); and
that bailers not be used to collect ground-water
samples.
Ground-water samples are also collected under provisions of
the Toxic Substances Control Act (TSCA). TSCA requires
that materials containing concentrations of polychlorinated
biphenyls in excess of 50 parts per million must be disposed
of in a landfill with a ground-water monitoring network.
The TSCA regulations (40 CFR § 761.75(b)(6)(ii) and (iii))
outline general performance standards for ground-water
sampling methods, but do not address specifically whether
ground-water samples should be filtered or unfiltered.
Under § 3(c)(2)(B) of the Federal Insecticide, Fungicide,
and Rodenticide Act (FIFRA), EPA may require pesticide
registrants to submit monitoring data from ground-water
studies to support the registration of pesticide products.
These studies are performed when residues of the pesticide
have been detected in ground water or when EPA suspects
that the pesticide will leach to ground water based on a
review of the environmental fate data. FIFRA's Draft
Guidance for Ground-Water Monitoring Studies (USEPA,
1988) recommends the filtration of ground-water samples;
however, this practice is being re-evaluated and revised
FIFRA guidance may change the sampling recommendation.
State implementation of the Federal RCRA ground-water
sampling requirements exhibits varying practices across the
nation. Information submitted to EPA by State
environmental agencies on ground-water sampling
procedures at MSWLFs and hazardous waste management
facilities reveals a non-uniform approach to the acquisition
of representative ground-water samples. During the
summer of 1993, all 50 States submitted publically available
information about ground-water sampling policies and
regulations to EPA. The data show that in both the
municipal solid waste and hazardous waste programs:
1) some States allow the filtration of ground-water
samples;
2) some States do not allow the filtration of ground-
water samples;
3) some States require or recommend the collection of
both filtered and non-filtered ground-water
samples; and
Figures 1 and 2 exhibit the distribution of these ground-
water sampling practices throughout the United States for
the municipal solid waste and hazardous waste programs
respectively. The figures do not suggest any apparent
regional trends in sampling practices used by the various
State agencies. A comparison of these sampling practices to
a generalized hydrogeologic map of the United States also
fails to depict any deliberate consistency in sampling
practices within a distinct ground-water region (Figure 3).
Overall, that data show that ground-water sampling
practices lack consistency within EPA Regions, within
ground-water regions of the U.S., among States, and
between the municipal solid waste and hazardous waste
programs within States.
As made evident by this brief review of various Federal
environmental regulations requiring ground-water
monitoring, sampling requirements vary: some regulations
allow the filtration of ground-water samples and others do
not. Some Federal regulations are not specific with regard
to how the samples should be collected, relying instead
upon guidance from State environmental agencies to
develop the sampling protocol. A general survey of these
State sampling policies and regulations shows no apparent
consistency with respect to regulatory objectives or
hydrogeologic setting. Therefore, a need exits to develop
uniform ground-water sampling guidance that considers the
objectives of the ground-water monitoring regulatory
programs and incorporates site-specific hydrogeologic
information.
4) some States allow either filtered or non-filtered
ground-water samples.
-------
Stales With Policy «r Regulations Thai Generally Suppon Filtering Ground-water Samples
Slates With Policy or Regulations That Generally Suppon Not Filtering Ground-water Samples
' {Including Sums That Require or Recommend Both Filtered and Non-filtered Samples i.e., W and !L)
t'* ','1 Slates With Policy ur KcgulaiHMs Thai Generally Allow Eith
-------
Unsoned Glacial Till
Unconsolidated Grave), Sand, Sitt, and Clay (Western
Alluvial Region Includes Basin and Range \folcamics)
: Consolidated Sediments (Predominantly Sandstone and
: Shale with Minor Umestone and Dolomite)
\ CartKmaic Karsi Regions
Igneous and Mctamorphic Crystalline Rocks
Figure 3. Generalized Hydrogeologic Map of the United States.
References
Federal Insecticide, Fungicide, and Rodenticide Act, 7
U.S.C. § 3(c)(2)(B).
Ground-water Monitoring Requirements for Landfill
Disposal of PCBs, 40 CFR § 761.75(b)(6)(ii) and (iii),
(1992).
Puls, Robert W. and Michael J. Barcelona, (1989). Ground-
Water Sampling for Metals Analysis. Superfund Ground-
Water Issue Paper (EPA/540/4-89/001).
USEPA, (1986). RCRA Ground-Water Monitoring
Technical Enforcement Guidance Document, Office of
Waste Programs Enforcement and Office of Solid Waste and
Emergency Response, (OSWER-9950.1).
USEPA, (1987). A Compendium of Superfund Field
Operations Methods. Office of Emergency and Remedial
Response, (EPA/540/P-87/001; OSWER Directive: 9355.0-
14).
USEPA, (1988). Draft Guidance for Ground-Water
Monitoring Studies. Office of Pesticide Programs.
USEPA, (1989). Risk Assessment Guidance for Superfund:
Interim Final Guidance. Office of Emergency and Remedial
Response (EPA/540/1-89/002).
USEPA, (1992). RCRA Ground-Water Monitoring: Draft
Technical Guidance. Office of Solid Waste (EPA/530-R-93-
001).
11
-------
SAMPLING PROGRAM PURPOSE
AND DESIGN CONSIDERATIONS
Michael J. Barcelona
Ground-water and subsurface monitoring programs have
diverse purposes and objectives which inevitably change as
detective efforts evolve into assessments and from there to
remedial-action performance evaluations. Common
objectives to many subsurface site characterization and
monitoring programs allow us to approach network designs
from a basic set of criteria for representativeness, accuracy,
precision, specifity, and sensitivity.
Detective monitoring programs have as their main
objectives: the characterization of the site and placement of
sampling points so as to permit detection of indicator
parameters (e.g. TOC, pH, conductance, metals or volatile
organic compounds (VOC's) etc. Assessment monitoring
efforts would follow detection of releases and seek to
determine the specific extent and magnitude of contaminant
distributions. In turn, programs planned to support the
design and implementation of remedial actions increase in
detail to the point where real differences in contaminant
mass removal may be determined over time. Fundamental
to all of these purposes are the following objectives1:
1. Avoidance of gross errors2 (i.e. false
negative data collection due to poor
sampling techniques, choice of location or
focusing on symptoms rather than media
where the bulk of the contaminant mass
resides),
2. Minimization of costs in dollars, time and
human resources by avoidance of redun-
dancy in data collection, while providing a
basis for detecting real trends3, and,
3. Complete data collection of both
hydrogeologic parameters and chemical
concentrations sufficient to permit decision-
making at a known level of confidence.
In all of these efforts, representativeness, accuracy,
precision, specificity and sensitivity are of value to the
monitoring effort. Good professional standards of practice
apply and it is time that these standards are applied to both
field and laboratory techniques.
Representativeness
Though representativeness in sampling is often parroted
back in plans and reports, there exists a substantial gap
between what is commonly practiced and what we can
identify as standards for professional practice from the
scientific literature. In simplest terms, representativeness
stems from the purposes of the investigation recognizing
that purposes change (as does the dollar value of the data)
and all sites merit some level of specific detective and site
characterization effort. This comes from a recognition that
the basic geologic descriptions and a consistent hydrogeo-
chemical framework which constitute a conceptual model of
a site have lasting value for the future.
Natural variability in background and "impacted"
hydrogeochemical conditions provides limits on the
distributions of chemical concentrations and aquifer
properties. These can be described in statistical terms.
Representativeness, therefore, implies that the sampled
population reflects at least some properties of the total
population. Values which lie beyond these limits will
require some explanation. The site's unique characteristics
will determine the extent to which the statistical
distributions of critical parameters (e.g. major hydro-
stratigraphic units, hydraulic conductivity, transmissivity,
background and contaminant concentrations) must be
known. Clearly, presuming that the site has been
characterized to a degree that it would pass peer review
could establish if in fact good professional practice had been
performed. It is at this point that the basis for a thorough
monitoring effort may address major deficiencies within
program constraints.
The literature has shown numerous examples of deficiencies
which can and should be remedied in routine monitoring
efforts3'4'5. Progress has been in a number of operational
areas which deserve more in-depth consideration in ongoing
efforts.
Focus on the Media
A focus on the correct medium (i.e. solid associated
contaminants, rather than symptoms in soil gas or water) for
the major mass of contaminants is essential to accurate
monitoring efforts. Table 1 shows data which reflects the
dominant nature of solid-associated volatile organic
contaminant mass. This has also been observed by many
groups. Here, the guidance from the Agency with regard to
monitoring practice (e.g. bulk solid sampling without
preservation) is at odds with good professional performance.
If in fact, the correct medium is left under-represented with
respect to chemical data collection, it is often the case that
12
-------
redundant well completions rather than in-depth
hydrogeologic data collection and interpretation may have
been made. The major liability here is that though some
information exists for the major contaminants, both
background conditions and critical design parameters for
remedial action are lacking. The major pitfalls here are:
1. Overly rigid work plans which do not permit
needed exploratory field screening or
borehole/solid sampling for
hydrogeochemical parameters in
background and contaminated zones,
2. The use of known biased procedures in
water (e.g. bailer sampling for VOC's) or
solids (e.g. bulk jar sampling) without
sufficient allowances for mineralogic, site
specific contaminant sorption/desorption
parameters, or hydraulic conductivity
determinations.
In addition, whether by dint of tradition or the "parrot-
syndrome" (i.e. parroting back the format and detail of
earlier studies) monitoring efforts may fail to provide
sufficient hydrogeologic data and its integration with the
chemical data to suit the reasonable long-term interests of
the monitoring work6-7. From this perspective, it may be
irrelevant that precise, specific or sensitive analytical
determinations have been made since the focus of the
investigation may have been off the mark. It is far more
important that high quality decisions have been made in the
field. It true quality is the key (and why has this become
less of a value?), the quality of decision making can only be
assured if active peer-review at the planning stage becomes
a part of the overall field program.
Critical Considerations For Improving Professional
Practice
What is vitally important here in the judgement of good
professional practice? From the literature we can glean the
following:
1. Implement consistent field screening
techniques for both chemical and
hydrogeologic parameters with careful lab
confirmation of a percentage of the total
samples 8>9;
2. Minimize planned well completions where
they will afford the most long-term
information term and collect sufficient
background data for meaningful comparison;
3. Recognize that field sampling can often be
the most significant source of error. Avoid
the use of "different" well design, purging
and sampling techniques across the site so
that chemical concentration comparisons are
in fact statistically comparable, (this also
means rejection of biased purging and
sampling techniques (i.e. the bailer); and,
4. Permit sufficient time and flexibility in the
program's directions to avoid "dead-end"
data collection which wastes time, human
resources and dollars to no useful end10'11.
The foregoing discussion stresses the value of a balanced
approach to site characterization and monitoring, which
Table 1. Relative Masses (|0g) of TCA in ground water and aquifer solids for a representative 1 liter aquifer
element/25)
Well
15
31
32
37
Ground
Water
13.5
85.7
11.1
83.7
MeOH Preserved
Solid
21.0
516.0
30.9
239
Bulk Jar, 4°C
Solid
-
52
-
-
% Total
TCA In
Solid
60
86
74
74
denotes no detectable levels of TCA in bulk jar sample by static headspace gas chromatography-Hall
electrolytic conductivity detection (7 day holding time) (Barcelona, M. J. and D. M. Shaw, 1992 unpublished
data).
13
-------
values completeness in the data collection effort. 2.
Refinement of the hydrogeochemical dataset via an
evolutionary, DQO-driven investigative effort will serve
both the regulated and regulating communities' goals most
closely5. We must reject the use of poor practices and
educate the entire community that these sites will not simply
go away. They may yield eventually to a well-founded,
documented monitoring approach based on geologic and
hydrogeologic fact.
3.
Table 2 contains an example of results of a controlled data
collection effort by using dedicated bladder pumps and
consistent purging and sampling procedures from a large
network for VOC's in ground water. Here the magnitude of
natural variability (what we are trying to evaluate) far
outweighs the errors from analytical or sampling
procedures. In addition, the use of dedicated devices has
been shown to minimize costs associated with network
operation12.
Many other examples of controlled data collection can be
pointed out from the literature to aid future monitoring
network designs13. These approaches have the merit of
lower cost as well as a focus on the correct media for
decision-making purposes.
References
1. Barcelona, M.J., Overview of the Sampling Process, in
Principles of Environmental Sampling, Keith, L. H.
Ed., ACS Professional Reference Book, American 8.
Chemical Society, Washinton, DC, 1988, Chapter 1.
5.
6.
Sanders, T.G., R.C. Ward, J.C. Loftis, T.D. Steele, D.D.
Adrian, and V. Yevjevich, Design of Networks for
Monitoring Water Quality, Water Resources
Publications, Littleton, CO, 1983, 328 pp.
Gammage, R.B., and B.A. Berven, Hazardous Waste
Site Investigations - Towards Better Decisions, Lewis
Publishers, 1992, 288 pp.
4. Barcelona, M.J., and Helfrich, J.A., Realistic
Expectations for Ground Water Investigation in the
1990's, in Current Practices in Ground- Water and
Vadose Zone Investigations, Nielson, D. M. and Sara,
M. N., Eds., American Society for Testing and
Materials. Philadelphia, PA, 1992, 431 pp.
U.S.EPA. Data Quality Objectives for Remedial
Response Activities Development Process, U.S. EPA.
540/G-87/003, March 1987.
Calabrese, E. J. and P.T. Kostecki, Risk Assessment and
Environmental Fate Methodologies, Lewis Publishers,
Chelsea, MI, (1992), 150 pp.
National Research Council, Ground Water and Soil
Contamination Remediation Toward Compatible
Science, Policy and Public Perception, Water Science
and Technology Board, National Academy Press,
Washington, DC, (1990), 261 pp.
First International Symposium on Field Screening
Methods for Hazardous Waste Site Investigations,
Table 2. Overall Mean, Relative Standard Deviation and Percentage of Total Variance Attributable to
Laboratory or Field (Sampling) Error, and Natural Variability (November 1990-September 1992)
Overall
VOC
Relative
Mean
(Std. Dev.)
o/.
Percent of Total Variability
Lab Field
Natural
111 TCA
TCE
C12DCE
11DCA
11DCE
119.5
29.8
45.2
44.3
16.3
(36%)
(43%)
(32%)
(28%)
(31%)
1.29
1.95
1.69
1.02
3.61
3.26
12.75
4.72
5.22
4.15
95.45
85.30
93.59
93.76
92.24
14
-------
U.S.EPA., U.S. Army Toxic and Hazardous Materials
Agency, Instrument Society of America, October, 1988.
9. Robbins, G. A., R. D. Bristol and V. D. Roe, AField
Screening Method for Gasoline Contamination Using a
Polyethylene Bag Sampling System, Ground Water
Monit. Rev. 9(3):87-97, (1989).
10. National Symposium on Measuring and Interpreting
VOC's in Soils: State of the Art and Research Needs.
Univ. of Wisconsin Extension, U.S. Environmental
Protection Agency, Oak Ridge National Laboratory,
U.S. Department of Energy, U.S. Toxic and Hazardous
Materials Agency, America Petroleum Institute.
January 12-14, 1993, Las Vegas, NY
11. Black, S.C., Defining Control Sites and Blank Sample
Needs, Chapter?, pp. 1077-117, in Principles of
Environmental Sampling, L. H. Keith (ed.). ACS
Professional Reference Book, American Chemical
Society, Washington, DC, (1988) 458 pp.
12. Lovejoy, S. and D.M. Eisenberg, Cost Effectiveness of
Dedicated Sampling Systems for Ground Water
Monitoring, pp. 89-96 in Proceedings of HAZMACON
'89 Hazardous Materials Management Conference and
Exhibition, April 18-20, 1989, Santa Clara, CA, T.
Bursztynsky andM. Loss, editors. Assoc. of Bay Area
Governments, P.O. Box 2050, Oakland, CA 94604-
2050.
13. Nielsen, D.M. and M.N. Sara, Current Practices in
Ground Water and Vadose Zone Investigations ASTM
1118, American Society for Testing and Materials,
Philadelphia, PA, (1992), 431 pp.
15
-------
THE RELATIONSHIP OF
MONITORING WELL DESIGN,
CONSTRUCTION
AND DEVELOPMENT TO
TURBIDITY IN WELLS, AND
RELATED
IMPLICATIONS FOR GROUND-
WATER SAMPLING
David M. Nielsen
A strong correlation exists between improper ground-water
monitoring well design, construction and development and
high levels of turbidity in ground-water samples. This
correlation is independent of the type of geologic material in
which the wells are installed. Research conducted by the
author, prior to the compilation of two recently approved
ASTM standards (D-5092, Standard Practice for Design and
Installation of Ground-Water Monitoring Wells in Aquifers;
D-5342, Standard Guide for Development of Ground-Water
Monitoring Wells in Granular Aquifers) indicates that
proper well design, construction and development can
alleviate turbidity problems in most unconsolidated
formation materials (sands and gravels). However, the same
research demonstrated that current well design, construction
and development practices are not sufficiently advanced to
limit turbidity in monitoring wells installed in
predominantly fine-grained unconsolidated formation
materials (silts and clays).
Recently proposed regulatory restrictions on ground-water
sample filtration appear to be founded primarily on the
premise that proper monitoring well design construction and
development practices can alleviate sample turbidity
problems in all wells. Current research demonstrates other
wise. Though proper well design, construction and
development are important steps toward eliminating
sedimentladen ground-water samples in some hydrogeologic
settings, these measures will not suffice in all situations.
Field filtration of ground-water samples intended for metals
analysis, while not a suitable substitute for proper well
design and construction in coarse-grained formation
materials, is necessary to ensure comparability of analytical
data from wells installed in predominantly fine-graine
formation materials. To disallow field filtration of samples
from these wells is to fail to recognize the limitations of
current well design technology.
Work conducted on this subject by others (Paul et al., 1988)
has demonstrated th vigorous development of wells installed
in fine-grained glacial till actually aggravates turbidity
problems and that standard well design practices, even when
supplemented by additional controls on sediment
production, failed to ameliorate turbidity problems.
Additional research is required to develop suitable well
design, construction and development guidelines applicable
to predominantly finegrained formation materials. This is
particularly critical in waste disposal facilities, at which
monitoring wells are or well be required by regulation, are
purposely sited in areas where these geologic materials
prevail.
16
-------
USE OF LOW-FLOW OR PASSIVE
SAMPLING TECHNIQUES FOR
SAMPLING GROUND WATER
Robert W. Puls
Introduction
It is generally accepted that monitoring wells must be
purged to access formation water to obtain "representative"
ground water quality samples. Historically anywhere from
3 to 5 well casing volumes have been removed prior to
sample collection to evacuate the standing well water and
access the adjacent formation water. However, a common
result of such purging practice is highly turbid samples from
excessive downhole disturbance to the sampling zone. This
disturbance includes the following: mixing of stagnant
casing water with water which resides in the screened
interval and the formation; aeration; degassing; and
excessive turbidity due to high pump rates or the continual
plunger action of bailers. The excessive turbidity created
can impact estimations for both "dissolved" and "total"
metals determinations, and cause other artifacts which may
adversely affect sample quality. An alternative purging
strategy has been proposed using pumps which permit much
lower flow rates (< 1 liter/min) and placement within the
screened interval of the monitoring well. The advantages of
this approach include less disturbance to the sampling zone,
increased spatial resolution of sampling points (i.e., sample
smaller portion of the aquifer), less variability, less purge
time (and volume), and low-turbidity samples. The overall
objective is a more passive approach to sample extraction
with the ideal being to match the intake velocity with the
natural ground water flow velocity (i.e. negligible
drawdown during purging). The volume of water extracted
to access formation water is generally independent of well
size and capacity and dependent upon well construction,
development, hydrogeologic variability and pump flow rate.
This is particularly evident where dedicated sampling
systems are employed. A detailed understanding of the
hydrogeochemistry of the sampling zone, particularly with
respect to aquifer heterogeneities, is important in both the
design of the sampling system and the interpretation of the
resultant sampling data for site assessment and remedial
evaluation purposes.
Required purge volume or purge duration is evaluated
through continuous monitoring of water quality parameters
(WQPs) such as dissolved oxygen, specific conductance,
oxidation-reduction (redox) potential and turbidity to
determine the presence of formation water. Research has
shown (Puls et al. 1992; Backhus et al. 1993; Barcelona et
al. 1994) that purging at these lower rates with various types
of pumps (peristaltics, low-speed submersibles, and bladder
pumps) does indeed produce low turbidity and generally
high quality samples. A perceived disadvantage of such
strategies however is the additional time required to purge
the wells. In general however, this same research has shown
that the volumes required to access formation water are less
than 2 casing volumes and in deeper wells are actually only
fractions of a casing volume, suggesting that the purge
volume is independent of well size or casing volume. This
previous research was conducted exclusively with
portable sampling systems. Studies which have utilized a
downhole camera during the purging process (Puls and
Powell, 1992; Kearl et al. 1992) have suggested that the
installation of the sampling devices themselves causes
the most disturbance to the sampling point. If
installation of the sampling system can be avoided
through the use of dedicated sampling systems, then this
could greatly reduce the time required to obtain
representative formation samples and also greatly reduce
the volume of water brought to the surface.
This study was conducted to evaluate the use of low-flow
purging methods in dedicated sampling installations in
routine monitoring wells. Stabilization of water quality
indicator parameters, together with time series sampling of
contaminants and water chemistry was employed to evaluate
the purge volume required to access formation water.
Study Site
The field site is located at the U.S. Coast Guard (USCG)
Support Center near Elizabeth City, North Carolina, about
100 km south of Norfolk, Virginia and 60 km inland from
the Outer Banks of North Carolina. The base is located on
the southern bank of the Pasquotank River, about 5 km
southeast of Elizabeth City. The topography of the site is
essentially flat and between two and three meters above sea
level. The river has a width of approximately 3.2 km along
the USCG base's northern boundary and a depth about 3
meters. A chrome plating shop, located within an aircraft
hangar on the base, had been in use for more than 30 years,
and had discharged acidic chromium wastes through a hole
in the concrete floor into the soils immediately below the
shop's foundation and the underlying aquifer (Puls et al.
1994).
17
-------
The site geology consists of typical Atlantic coastal plain
sediments characterized by complex and variable sequences
of surficial sands, silts and clays. In the vicinity of the
plating shop, the surface soils are silty clays. These overlie
a thin sandy clay layer at about 1.5 m, which overlie a
sequence of sands and silty fine sands. In some locations, a
dense gray clay layer substitutes for the sandy clay layer at
1.5 m. Fine to medium sands dominate from 4 to 20 m. A
dense gray clay unit (Yorktown Confining Unit) persists at a
depth of 20 m. This depth is slightly variable and dips
gently from north to south.
Materials and Methods
Monitoring Wells and Sampling Pumps
Eight different monitoring wells at the site were used in the
study. With the exception of two of the wells, these were all
5 cm (2 in.) diameter schedule 40 polyvinylchloride (PVC)
wells with 0.25 mm (0.01 in) slotted screens. Other
characteristics of the sampled wells used in the study are
listed in Table 1. The wells ranged in depth from 4.6 to 15.2
meters below ground surface. System volume (Table 1)
refers to the volume of water in the tubing and the water
quality parameter measurement device (QED PurgeSaver).
Well volume (Table 1) refers to the water in the monitoring
well itself which varied somewhat over time due to
fluctuations in the water table level. Two of the wells (MW
25, MW 31) were installed without casing. For these two
wells, a permanently dedicated PVC bladder pump (QED
Inc.) and a permanently dedicated variable-speed
submersible pump (Redi-Flo 2, MP1; Grundfos Inc.) were
encased in 0.25 mm (0.01 in.) slotted screens, sealed and
connected with 0.63 cm (0.25 in.) teflon-lined polyethylene
tubing. These units were lowered inside 7 cm (2.75 in.)
hollow-stem augers to the desired depth and the formation
was allowed to collapse in around the units. For the caseless
Grundfos, a sandpack was used inside the screen, whereas
for the caseless QED pump glass beads were used as the
packing or filter material. These systems are referred to as
permanently dedicated, because they cannot be removed for
servicing. The same type pumps (Grundfos Redi-Flo2 and
QED bladder) and tubing were used in the traditional
monitoring wells as well.
Purging and Sampling Procedures
The pumps (except for the permanently installed pumps)
were set with the pump intake at approximately mid-screen.
Following installation (April, 1992), most of the pumps
remained in place throughout the study (April, 1992 - June,
1993). Data was first collected in August, 1992, and then
again in February, March, and June of 1993. The sampling
procedures described in previous publications (Puls et al.
1992; Puls and Powell, 1992) were generally followed,
except all purge water was collected in sequential 500 ml
increments for analysis of volatile organics
(trichloroethylene [TCE], dichloroethylene [DCE], and
vinyl chloride), major cations and metals, anions, and
chromium. Water levels were measured and recorded
prior to purging and monitored continuously while
purging to minimize drawdown (<0.1 m). Flow rates
ranged from 0.22-0.55 L/min. A flow-through cell with
data logger (QED PurgeSaver) was used to continuously
monitor pH, temperature, dissolved oxygen (DO), and
specific conductance. Purging and continuous sampling
continued beyond equilibration of water quality indicator
parameters to a maximum of 21 liters. Equilibration was
defined as three successive readings within +10% for
DO and turbidity, +3% for specific conductance, and
+0.05 for pH. At the flow rates utilized, these readings
were taken every three minutes. Equilibration criteria
were based on evaluation of preliminary plotted WQP
data and equipment accuracy. Temperature was recorded
but not used for stabilization.
Results and Discussion
Purging results in terms of water quality parameter
(WQP) equilibration and contaminant concentration
(CC) equilibration are shown in Table 1. WQP and CC
equilibration volumes were independent of well depth or
well volumes. WQP equilibration volumes ranged from
4 to 10 L, while CC equilibration volumes ranged from
2.5 to 7.5 L. The two dedicated permanent wells (MW
23, 31) had some of the lowest CC equilibration volumes
as might be expected due to the minimal stagnant well
water above the pump (0.64-0.81 L in buried pump
tubing). MW 22, the deepest well, had the smallest CC
equilibration volume for all wells used in the study,
while the largest CC equilibration volume was for the
shallowest well (MW 2).
The equilibration trends for the WQP's were similar for
all wells in the study. Specific conductance increased
slightly and equilibrated prior to dissolved oxygen and
turbidity. Dissolved oxygen decreased and equilibrated
after specific conductance and prior to turbidity.
Turbidity followed a generally exponential decline.
Contaminant concentration equilibration volumes were
less than or equal to WQP equilibration volumes.
Differences in initial and final equilibration contaminant
concentrations were generally less than 20% for TCE, c-
DCE, vinyl chloride and chromium.
Using a spherical conceptual model for the aquifer volume
sampled, CC equilibration was achieved within a 13-16.5
cm radius (assuming a porosity of 0.38). Alternatively, if
casing volume is defined by only the water within the
screened interval of the well where the pumps are located,
CC equilibration was attained between 1.1-2.4 screened
interval casing volumes (SICV) for the dedicated pumps in
the 5 cm monitoring wells.
18
-------
Summary and Conclusions
An alternative conceptual model is proposed which
considers the evacuation of only a portion of the screened
interval volume rather than the entire casing volume.
Depending upon well depth and the analytes of interest,
various types of pumps may be used with the pump intake
located within the screened interval and at the desired
sampling depth. Significant reductions in purge volume
have been attained using dedicated systems, and thorough
economic analyses may show this to be a cost-effective
alternative for routinely-sampled monitoring wells.
While this strategy works most effectively where well
screens are short (<1.5 m) and located within a relatively
homogeneous geologic zone, they may also be effective in
longer screened intervals and where geologic
heterogeneities may exist. This remains to be tested. The
utility of these techniques has not been explored in open
boreholes in fractured rock. Proper well construction and
well development and complete documentation of all
sampling activities becomes increasingly important where
such strategies are employed. A detailed understanding of
the hydrologic and geologic variability of the system is
essential in establishing sampling points and designing the
overall sampling program.
Disclaimer
Although the research described in this article has been
funded wholly or in part by the United States Environmental
Protection Agency, it has not been subjected to the Agency's
peer and administrative review and therefore may not
necessarily reflect the views of the Agency and no official
endorsement may be inferred.
References
Backhus, D.A., J.A. Ryan, D.M. Groher, J.K. MacFarlane.
1993. Sampling Colloids and Colloid-Associated
Contaminants in Ground Water. Ground Water, 31(3), 466-
479.
Barcelona, M.J., H.A. Wehrmann and M.D. Varljen. 1994.
Reproducible Well Purging Procedures and VOC
Stabilization Criteria for Ground-Water Sampling. Ground
Water, 32(1).
Table 1. Characteristics of wells used in the study and water quality parameter and contaminant concentration
equilibration data, U.S. Coast Guard Support Center, Elizabeth City, NC.
Well
Depth1 Screen2 Device3 WQPEV4 CCEV5 SystemVol6 Well Vol.'
2 4.6 1.5 CS 7.5 7.5
13 4.6 1.5 CS 7.0 4.5
13 4.6 1.5 CS 7.0 7.0
14 6.1 1.5 BL 10.0 4.5
15 4.6 1.5 BL 6.0 5.5
16 4.6 1.5 CS 4.0 3.5
22 15.2 3.1 CS 7.5 2.5
23 7.1 0.9 BLp 7.5 3.5
31 4.9 0.5 CSp 6.5 3.5
1 Depth = well depth in m,
2 Screen = screen length in m,
3 CS = low-speed centrifugal submersible pump, BL = bladder pump, CSp
centrifugal submersible pump, BLp = permanently buried bladder pump.
4 WQPEV = water quality parameter equilibration volume in liters,
5 CCEV = contaminant concentration equilibration volume in liters,
6 System volume = volume (in liters) of tubing and flow-through cell,
7 Well volume = volume (in liters) of water in well casing and well screen.
0.6
0.6
0.6
0.9
0.8
0.8
1.3
0.8
0.6
= permanently
5.6
5.6
5.6
8.7
5.6
5.6
27.2
0.5
0.3
buried low-speed
19
-------
Kearl, P.M., N.E. Korte, andT.A. Cronk. 1992. Suggested
Modifications to Ground Water Sampling Procedures Based
on Observations from the Colloidal Borescope. Ground
Water Monitoring Review, Spring, 1992, 155-160.
Puls, R.W. andR.M. Powell. 1992. Acquisition of
Representative Ground Water Quality Samples for Metals.
Ground Water Monitoring Review, 12(3), 167-176.
Puls, R.W., D.A. Clark, B. Bledsoe, R.M. Powell, and CJ.
Paul. 1992. Metals in Ground Water: Sampling Artifacts and
Reproducibility. Hazardous Waste and Hazardous
Materials, 9(2).
Puls, R.W., CJ. Paul, D.A. Clark, and J. Vardy. 1994.
Transport and Transformation of Hexavalent Chromium
Through Soils and into Ground Water. J. Soil
Contamination. To be published June, 1994.
20
-------
SAMPLING COLLOIDS AND
COLLOID-ASSOCIATED
CONTAMINANTS
IN GROUND WATER
D.A. Backhus, J.N. Ryan, D.M. Groher, J.K.
MacFarlane and P.M. Gschwend
Groundwater monitoring wells are used to obtain general
information about groundwater quality and specific
information about concentrations and speciation of mobile
contaminants in the vicinity of a well. This information is
used to determine whether a given facility is currently in
compliance with regulations. In addition, data from
monitoring wells can be used to provide clues to as to the
processes affecting contaminant fate in the subsurface,
allowing development of models for accurate predictions of
spacial and temporal contaminant distributions. These
predictions allow for assessment of risks under various
scenarios and evaluation of alternative remediation
strategies. It is crucial that samples obtained from
monitoring wells accurately reflect in situ mobile
contaminant concentrations, as the cost of a bad decision
based on inaccurate data can be great.
In the past, it was generally assumed that contaminants
existed in the saturated zone either as associated with the
aquifer material and therefore immobile or dissolved in and
moving with the groundwater. Hence, when a turbid
groundwater samples was obtained from a monitoring well
it was assumed that the turbidity was an artifact of well
construction, preparation, or sampling procedures.
Filtration was and continues to be used to remove the
particles causing this presumed artifact turbidity. In recent
years, we have come to realize that in some subsurface
systems, turbidity causing materials may actually be present
in an aquifer as mobile colloidal species (see review by
McCarthy and Zachara, 1989). Filtration of groundwater
samples from these aquifers may remove both mobile
colloids and artifact particles. If the mobile colloids are
inherently hazardous (e.g., pathogenic bacteria, asbestos
particles, precipitates of radioactive materials or toxic
metals) or if significant quantities of contaminants sorb onto
these mobile colloids, then it is important that samples
analyzed to determine compliance, assess risks, and develop
models include these, as well as, dissolved contaminant
species. Removal of these mobile colloids from
groundwater samples could lead to underestimation of
contaminant concentrations and mobility, affecting
determinations of both compliance and current risks.
Failure to account for colloid- associated contaminants in
predictive models used for decision making could lead to
serious errors in setting clean-up priorities and in designing
cost effective remediation strategies.
Examination of the role of colloid-associated contaminants
in subsurface systems requires an ability to distinguish
between artifact particles and mobile colloids in
groundwater samples. We have designed and tested a
groundwater sampling system which avoids inclusion of
artifact colloids/particles in samples and minimizes losses of
mobile colloids and changes in colloid character during
sample collection and storage (Backhus et al., 1993). This
sampling system (Figure 1) incorporates slow prolonged
pumping using a positive-displacement pump with a low
sample contact surface area constructed of stainless steel
and teflon (model SP-202, Fultz, Inc., Lewistown, PA). The
pumping rate is controlled by a variable voltage AC-DC
converter powered by a generator, allowing minimum
steady pumping rates of about 100 mL/min to be maintained
at most wells. An inflatable packer or packers are used to
isolate the sampling zone. Groundwater from this sampling
zone is pumped to the surface through a continuous piece of
polypropylene (0.64 cm id) or aluminum (0.48 cm id)
tubing. At the surface, groundwater flows directly into a
flow-through monitoring cell to allow measurement of water
chemistry parameters (pH, Eh, dissolved oxygen (DO),
temperature, and specific conductance) during well purging.
In addition, samples are obtained periodically during well
purging to monitor turbidity. These samples are collected
by overfilling and tightly capping a cuvette or vial
and examined by using a submicron particle analyzer or
turbidity meter. Once the well is deemed sufficiently
purged, unfiltered groundwater samples for geochemistry
and colloid analyses are collected in glass bottles designed
to allow closure without inclusion of head space; therefore,
they minimize the exchange of gases between the sample
and the atmosphere during storage. Prior to sampling,
these bottles are filled with argon. Samples are collected by
inserting the sampling tube to the bottom of the bottle and
overfilling the bottle (by at least half of its volume).
Samples for specific contaminant analyses are collected in a
similar manner in suitable sampling vessels (e.g., 40 mL
glass vials or 4 L bottles). Additional groundwater colloid
21
-------
Figure 1. Sampling setup for collection of ground water from monitoring wells when colloids or colloid associated
chemicals may be of importance.
samples are collected by joining the end of the sampling
tube to a 5 mL plastic syringe tip and allowing the plunger
to be displaced backwards to rinse and then fill the syringe.
The filled syringe is connected to a filter holder containing a
15 or 30 nm pore size Nuclepore polycarbonate membrane
filter. The sample is forced through the filter by pressure
from weights (0.8 kg/cm2), then rinsed with 5 ml of
distilled, deionized water, and placed in a covered
petri plate in a desiccator to dry.
Samples collected as indicted above should provide accurate
information about in situ groundwater geochemistry, mobile
colloid concentrations, and total mobile contaminant loads,
as well as, the characteristics of mobile colloids (e.g., size,
morphology, and elemental composition from scanning
electron microscopy /energy dispersive x-ray analysis of
colloids caught on field filters). This sampling scheme (1)
minimizes collection of artifact particles sheared from
aquifer materials by using slow pumping rates; (2) avoids
precipitation of artifact colloids and changes in the character
of mobile colloid by using flow-through sampling methods
and sample storage in DO bottles which minimize changes
in groundwater chemistry as the sample is brought to the
surface and sealed; (3) assures that prior to sampling, the
well is sufficiently purged of artifact particles which may
have been introduced during construction or formed due to
atmospheric exposure via the well casing; and (4) minimizes
losses of mobile colloids by avoiding filtration of ground-
water samples.
Groundwater samples have been collected from numerous
sites across the United States using this careful sampling
scheme (Gschwend and Reynolds, 1987; Backhus and
Gschwend, 1990; Gschwend et al., 1990; Ryan and
Gschwend, 1990; Groher, 1990). Collectively, the results of
these studies provide information both about groundwater
sampling and about the existence of mobile colloids in the
subsurface. Regarding sampling, similar results were found
at most wells when groundwater turbidity was examined as
a function of time or volume of water removed from the
well. Initially high turbidity level, decreased to a stable
level after several hours of slow pumping (e.g., Figure 2).
At "background"wells, turbidity levels approached those
observed for distilled water blanks (e.g., Figure 2). At
contaminated wells (within a contaminant plume or where
geochemistry was altered due to natural processes), turbidity
22
-------
'Background" well
"Contaminated" well
>s
*->
'w
C
•g
cm
c
'C
t>
1C
o
to
t
VI
W
C
3
O
106 Oct
105
104
103
102
10
1987
• Field
O Laboratory
I
•
I
•
0
JO
1 1
o
^
fin
IFI
<$
creased
o-w Rate
5 10 15 20
Volume Pumped (L)
25
105
104
103
102
10
Oct 1987
• Field
OLaboratory
p If •
0 10 20 30 40 50 60
Volume Pumped (L)
Figure 2. Turbidity timecourses from wells at the Delaware Site.
levels stabilized at significantly higher levels (e.g., Figure
2) indicating that mobile colloids may be present. When the
timecourses of other monitored parameters were compared
with turbidity, some parameters (O2, pH, Eh) were generally
found to stabilize before turbidity (O2 vs turbidity in Figure
3). At some sites, a few parameters did not reach a stable
level during well purging and sampling (e.g., total
nonvolatile organic carbon, TOC and specific conductance,
Figure 3). At the one site where we monitored contaminant
concentrations as well as water chemistry parameters and
turbidity, the timecourse for lower solubility contaminants
(Pyrene, Chrysene, and Benzo (a) anthracene) were found to
closely mimic the turbidity timecourse (Figure 3). Little
variation was seen from the more soluble and more
abundant contaminant naphthalene. These data suggest that
turbidity may provide a better indication of required well
purging times than other typically monitored parameters for
the less soluble, possibly colloid-associated contaminants.
At a few sites, the effect of pumping rate on turbidity was
examined. Turbidity initially increased with pumping rate
(Figure 2 and Figure 4), but then decreased to a stable
level again after pumping at the higher rate for a while
(Figure 4). The stable level eventually achieved at the
higher rate was not always the same as that observed at the
slow rate ("Background" vs Contaminated well Figure 4).
At the "Background" well, the stable turbidity level was
higher at the increased flow rate, while at the contaminated
well, the same stable turbidity level was reached at both
pumping rates. Finally, comparison of bailed and slowly
pumped groundwater samples indicate that the method
chosen effects the results obtained (Figure 4). At the New
York site, bailed samples contained 10 - 100 times greater
colloid concentrations, and up to 750 times greater
polycyclic aromatic hydrocarbon concentrations than were
detected in slowly pumped samples. The results of our
sampling experiences indicate that both the parameter used
to judge the adequacy of purging efforts prior to sampling
and the sampling method (pumping rate, and pumping vs
bailing) may effect geochemistry parameters and
contaminant and colloid concentrations observed in
groundwater samples.
Regarding the existence of mobile colloids in aquifers, at
each of the sites where stable turbidity levels in
Contaminated wells were significantly greater than levels
found in "Background" wells, colloid size characteristics
(size, composition, surface charge, stability) and
groundwater geochemistry were examined to assess whether
colloids collected in the samples were likely to be natural in
situ colloids. Based on alteration of the regional
geochemical conditions by the "contaminated" plume and
expected colloid solubilities and surface charges,
explanations have been surmised for the presence of mobile
colloids at several of the sites sampled. At the
Massachusetts, it is hypothesized that the changes in
groundwater chemistry due to the influence of the plume of
secondarily treated sewage lead to precipitation of colloids
(Gschwend and Reynolds, 1987). At the New Jersey,
Delaware (Ryan and Gschwend, 1990), and Nevada sites
(Gschwend et al., 1990) it is hypothesized that changes in
groundwater chemistry lead to dissolution of secondary
mineral phases (iron oxyhydroxides or calcium carbonates)
which had cemented colloidal material to the aquifer solids
and consequently mobilization of colloidal material.
Results obtained at numerous sites using the careful
groundwater sampling protocol described above indicate
that mobile colloids do exist in the subsurface. However,
there is insufficient information available, at this point, to
23
-------
ID*
s
« f o4
1 f '
" i
« I 103
1
10
40
"L 30
j
| 20
1-
0
{
30000
j 20000
1 10000
J
0
TURBIDITY eo
• 4.0
_ - 9.0
• • «? * f 2.0
1.0
a&f
DISSOLVED OXYGEN
•
1
\
V
10 20 30 40 0 10 20 30 40
• — • mow* 300°
D — D CHRYSENE +
•XD 7 ""
\\ ,." 1
^\"\ / * 1000
SPECIFIC CONDUCTANCE
/' ^
V ^
•
•
10 20 30 40 0 10 80 30 40
NAPHTHALENE *°°
f 260
A A _•_ t1
•-• »-«^ » 200
g 160
£
1 100
J 60
. . 1 . 0
TOC (non-volatile)
\
~~ *~^»
^~^«
10 20 30 40
Volume Pumped (L)
10 SO 30
Volume Pumped (L)
Figure 3. Monitored parameter timecourses from well D at the Connecticut Site.
oacKgrouna wen
"0 20 40 60 80 100 120 140
Volume pumped (L)
uom.ammaLea wen
>, 106
Icr io5
flj |
C | 104
00 m
B q lO3
ij. 102
(0
o
OT 10
July 1988
D D Bailed
O Pumped
D
o ^ o
O
O t Increased O Q O Q
<^ Q 1 Flow rate dHsO
106
106
10*
103
102
10
D D
o
%fcO 0 ° 0
t Increased
1 Flow Rate
July 1988
D Bailed
O Pumped
O CP
0 20 40 60 80 100
Volume Pumped (L)
Figure 4. The effect of pumping rate and sampling method on observed turbidity, NY site.
24
-------
judge the wide spread significance of mobile colloids.
There is sufficient information to warrant continued
examination of the role of mobile colloidal materials in the
transport of subsurface contaminants. Examination of the
role of colloids requires careful sampling methods which
incorporate: slow prolonged pumping, monitoring turbidity
during purging, maintaining in situ groundwater chemistry
conditions throughout sampling and storage, avoiding
filtration of samples which are used to assess colloid
concentration and total mobile loads of contaminants, and
collection of auxiliary data to confirm findings regarding the
existence of mobile colloids. If no evidence supporting the
existence of colloids and colloid-associated contaminants is
found at a given site, then more practical sampling schemes
may be instituted with comparable results.
References
Backhus, D.A., J.N. Ryan, D.M. Groher, J.K. MacFarlane,
and P.M. Gschwend. 1993. Sampling Colloids and Colloid-
Associated Contaminants in Ground Water. Ground Water,
31(3): 466-479.
Backhus, D.A. and P.M. Gschwend. 1990. Use of
fluorescent polycyclic aromatic hydrocarbon probes to study
the impact of colloids on pollutant transport in groundwater.
Environ. Sci. Technol. v. 24, pp. 1214-1223.
Groher, D.M. 1990. An investigation of the factors affecting
the concentrations of polycyclic aromatic hydrocarbons in
groundwater at coal tar waste sites. M.S. thesis. Dept of
Civil Eng., Massachusetts Institute of Technology. 145 pp.
Gschwend, P.M., and M.D. Reynolds. 1987. Monodisperse
ferrous phosphate colloids in an anoxic groundwater plume.
J. Contam. Hydrol. v.l, pp. 309-327.
Gschwend, P.M., D.A. Backhus, J.K. MacFarlane, and A.L.
Page. 1990. Mobilization of colloids in groundwater due to
infiltration of water at a coal ash disposal site. J. Contam.
Hydrol. v. 6, pp. 307-320.
McCarthy, J. F. and J. M. Zachara. 1989. Subsurface
transport of contaminants. Environ. Sci. Technol. v. 23, pp.
496-502.
Ryan, J. N. and P. M. Gschwend. 1990. Colloid
mobilization in two Atlantic Coastal Plain aquifers: Field
studies. Water Resour. Res. v. 26, pp. 307-322.
25
-------
NEED FOR PRACTICAL
APPROACHES TO CONDUCT
MULTILEVEL SAMPLING
Gary A. Robbins
Robbins (1989), Robbins and Martin-Hayden (1991) and
Martin-Hayden et al. (1991) have shown that vertical mass
averaging in typical monitoring well results in water quality
samples that ate influenced by well screen length, vertical
concentration gradients, vertical variations in formation and
backfill hydraulic conductivity, the amount of water
removed for purging, and water levels achieved in a well
during purging and when sampling. Because of vertical
mass averaging, samples taken from wells can be highly
misleading in terms of the absolute and relative abundances
of water quality contituents. This in turn can lead to gross
misinterpretations of contaminant distributions, transport
properties, and physical, chemical and biological conditions
and processes that influence overall water quality. It
follows that efforts directed at obtaining representative
samples from typical monitoring wells are futile. Mass
averaging during well purging and sampling will result in a
well-biased and non-representative sample irrespective of
sampling method.
It should be kept in mind that monitoring wells were not
designed to provide representative samples. They are
simply an outgrowth of combining geotechnical engineering
practice (i.e., the availability of hollow stem augers),
everyday water well practice (i.e., constructing wells with
casing and screens), and public health practice (i.e.,
collection of samples for standard laboratory analysis). That
is, on a historical basis, monitoring well were not developed
in consideration of the three-dimensional nature of
groundwater flow and quality.
It is suggested here that research efforts be refocused on
developing practical approaches for conducting multilevel
sampling. Multilevel sampling issues requiring resolution
include: methods for constructing multilevel sampling nests
in hollow stem auger holes, methods for ground water
sampling using direct push technologies, methods for
collecting water quality samples during drilling, testing
seals between samplers, defining the vertical spacing
between samplers and their intake lengths, preventing intake
clogging in fine grain formations, and sampling contituents
exhibiting high vertical gradients. When dealing with non-
aqueous phase liquids (NAPL), methods are needed for
sampling mobile and residual NAPLs, and avoidance of
cross contamination when samplers are constructed through
zones of NAPL.
References
Robbins, G.A., 1989. Influence of Using Purged and
Partially Penetrating Monitoring Wells on Contaminant
Detection, Mapping and Modeling. Ground Water Journal,
v. 27, No. 2, p. 155-162.
Robbins, G.A. and Martin-Hayden, J.M., 1991. Mass
Balance Evaluation of Monitoring Well Purging, Part I.
Theoretical Models and Implications for Representative
Sampling. Journal of Contaminant Hydrology, v. 8, p. 203-
224.
Martin-Hayden, J.M., Robbins, G.A., and Bristol, R.D.,
1991. Mass Balance Evaluation of Monitoring Well
Purgeing, Part II. Field Tests at a Gasoline Contamination
Site. Journal of Contaminant Hydrology, v. 8, p. 225-241.
26
-------
Effects of Well Design and Time of
Pumping on Concentrations
of Volatile Organic Compounds in
Ground-Water Samples
Jacob Gibs and Thomas E. Reilly
Because a water sample collected from a well is an
integration of water from different depths along the well
screen, concentrations of analytes measured in the sample
can be biased if analyte concentrations are not uniform
along the length of the well screen. The concentration in the
sample is a function of variations in well-screen inflow rate
and analyte concentration with depth. These relations were
investigated at a site with gasoline-contaminated
groundwater in Galloway Township, Atlantic County, New
Jersey. Numerical simulation of the integration of water
along the well screen at this site and in a hypothetical
system was used to determine the total mass of selected
volatile organic compounds entering the screen by assuming
a layered porous medium in which each layer is
characterized by uniform hydraulic conductivity and
chemical concentration.
A well screen with seven short screened intervals was
designed and installed at two locations at the Galloway
Township site. Independent samples were collected from the
seven screened zones at each location. The sample
concentrations from each screened zone were flow-rate
weighted and integrated to simulate a sample concentration
from a 5-foot-long, 2.375-inch-outside-diameter
conventional wire-wound screen. The integrated volatile
organic compound concentration was as little as 28 percent
of the maximum concentration observed in samples from
the multiscreened well.
Numerical simulation of the integration of water along the
well screen at this site and in hypothetical heterogeneous
ground-water systems also was used to investigate the
temporal variation in water quality. Concentrations of
constituents that are not uniformly distributed along the
screened interval of the well can vary during purging and
continued pumping after purging as a result of vertical
variations in flow rate in the vicinity of the well screen. This
variation in water quality with time also can be affected by
well design characteristics, such as the use of a filter pack
and screen length. Results of numerical simulations of flow
associated with a hypothetical well design show that at a
constant pumping rate the percentage of total flow into the
screened zone at the ends of the screen that is derived from
increases in vertical flow as screen length decreases or when
a filter pack is used. Thus, a water sample collected from a
well in which the screen length is short or a filter pack is
used represents water from a vertical interval in the aquifer
that is larger than the actual length of the well screen.
27
-------
COMMON SAMPLING AND
ANALYTICAL PROCEDURES
VIEWED IN THE CONTEXT OF
DATA QUALITY NEEDS
William R. Mabey and Nancy Barnes
The practice of sampling in the monitoring and investigation
of groundwater quality has experienced significant advances
in knowledge in the last decade. While some of the
methods now in use were adapted from the water supply
industry, many other methods reflect modifications that
resulted from the recognition of shortcomings of some
methods in obtaining water samples that provide reliable
chemical data. Research has been vital in recognizing these
shortcomings and developing better methods as well as
increasing our knowledge of groundwater chemistry,
hydrogeology, and remedial technologies. However,
incorporating these changes as acceptable professional
methods has been sometimes slow because of the
precedence of existing data with the older methods, costs of
equipment and training field personnel, and the lack of
familiarity and the uncertainties with the newer methods.
Recognizing that groundwater data are collected for specific
applications, it is appropriate to depart from historical
precedence, policy, or preoccupations regarding
uncertainties, and rather focus on the methods as they
provide quality data. Quality data can be defined as those
data that are fully described so that they may be intelligently
used (Campbell and Mabey, 1985). This definition is
consistent with the Data Quality Objectives (DQO) as set
forth in USEPA guidance. In this DQO context, data are
evaluated according to the criteria of accuracy,
representativeness, comparability, completeness and
precision. Clearly, data must have known qualities so that
the data user can decide whether the data can be applied to
the intended application.
The DQO objectives listed above are well-established in the
evaluation of analytical chemical data. The accuracy and
precision attainable for these data are measureable using
stated procedures. The data validation process according to
the Contract Laboratory Program (CLP) provide the basis
for expression of the quality of analytical data, and where
any qualifications on the data are valuably discussed in a
summary report format. A review of the analytical
procedures and CLP criteria shows that negative bias (that
is, lower concentrations than actually present) is reasonably
inherent in many methods because of losses during sample
handling and analyses. While a complete discussion of
analytical data quality is not the purpose of this paper, it is
important to recognize that data acceptable in the CLP
program may have losses of over 50% for some semi-
volatile organic chemicals (Guide to Environmental
Analytical Methods, 1992).
Appropriately, there is increasing concern for the quality of
data from groundwater monitoring and investigation
programs because of the way in which wells are
constructed, developed, purged, and sampled (Puls and
Powell, 1992). An evaluation of some procedures now in
common use shows that the accuracy of resulting data
should be of concern. In particular, some negative bias is
often likely because of losses as a result of some
groundwater purging and sampling procedures. As with
other measurements, good precision (that is, repeatability)
can be a misleading if accuracy is not also attained. It
should also be recognized that databases can have the effect
that the very qualities that contribute to a measured value
can be lost, and data from the database then are
inappropriate for comparison purposes.
It is ironic that the implementation of new sampling
methods can be viewed as a "conflict" between the DQO
criteria of comparability and representativeness. While the
established methods have merits of familiarity and
established use, research and other experience is showing
that some methods may not give representative samples
based on what can be shown or is reasonably of concern
based on the newly gained scientific knowledge. However,
attempts to develop methods that produce data more
representative of our understanding of groundwater
conditions are hampered by reasons presented in the
opening paragraph. Often overlooked is the realization that
some known bias in data may be acceptable for particular
applications, and that having the most accurate or
representative value may not be necessary.
One area where research has revealed a negative bias in
metals data because of sampling practices is in the filtration
of groundwater samples. Previously, turbid water samples
were often filtered, usually with a 0.45 micron filter, to
remove paniculate materials from the water sample. The
rationale was that the turbidity was an artifact of well
28
-------
construction and sampling, and filtering provided a true
dissolved concentration of analytes in the groundwater
sample. However, when research and some field
observations observed that colloidal material was mobile in
some saturated zone systems, filtration was discouraged to
avoid removing the collodial material (Puls and Barcelona,
1989). Thus, to remove a negative bias in metals
concentrations, the sampling procedure has changed to
possibly provide a positive bias in the data. As an
alternative procedure, the low flow rate purging and
sampling of a monitoring well has been employed to
achieve low turbidity and more representative groundwater
samples (Puls and Powell, 1992).
Several field sampling efforts by Montgomery Watson have
attempted to evaluate the significance of sample turbidity on
the concentrations of metals in groundwater samples. In
one project in California, wells previously installed in
another investigation study were found to have significant
turbidity and high metals concentrations. These wells were
purged at rates as low as 0.5 gallons/minute using a bladder
pump until the temperature, electrical conductivity (EC) and
turbidity parameters stabilized. Typically, turbidity was the
last parameter to stabilize according to DQO criteria, and
the times for stabilization ranged from one to seven hours,
with purge volumes ranging from 15 to 365 gallons (15 to
40 casing volumes). The wells were then sampled first
using the bladder pump, and then using a bailer. For each
sampling procedure a water sample was filtered using a 0.45
micron filter so as to provide a comparison of the effect of
filtration. Sampling with the bailer provided turbid samples
(>200 NTU) and the bailed, unfiltered water samples
typically showed two-fold higher concentrations of metals
present (including barium, chromium, copper, zinc, etc) than
the unfiltered water collected with a bladder pump (NTU
values of 4 and 5). However, the metal concentrations of
the bailed, filtered samples were generally comparable to
those of the filtered and unfiltered samples collected with
the bladder pump (generally less than 10% difference)
These data then provide empirical evidence that filtration
can produce representative groundwater samples for some
monitoring wells at this site. It is also of note that for
several monitoring wells the iron concentrations in the
filtered water samples collected with the pump were higher
than the concentrations in the filtered, bailed water samples,
a result that is rationalized by the aeration of the bailed
sample, the subsequent oxidation of ferrous (iron) to the
ferric state, and removal of the solid ferric oxide/hydroxide
by filtration.
At another site, a Montgomery Watson sampling team
collected groundwater samples over a several month period
with varying pumping rates to develop an understanding of
the metals concentrations as a function of turbidty. The total
suspended solids (TSS) was also determined by the
laboratory as a quantitative measure of the paniculate
content of the samples. A log/log plot of TSS against the
field-measured turbidity showed a roughly linear
relationship. Plots of TSS against the concentrations of
several metals also showed the expected relationship of
higher concentrations at higher TSS values. This
relationship was most notable for lead; however, when the
lead concentrations were divided by the iron or aluminum
concentrations the resulting ratios were roughly independent
of the TSS values. If it is accepted that the iron and
aluminum values are associated with their natural
abundances in soils, it is also reasonable that the lead
concentrations observed are also associated with the natural
abundance of soil constituents. Such a normalization
approach also has shown value in explaining high metal
concentrations in water samples at other sites.
Measures to improve the quality of sampling data clearly are
needed. The improvements may include improved sampling
procedures or obtaining additional information to support
the measured numbers. For example, measurement of redox
potentials or other groundwater chemistry parameters are
useful support the presence of reduction products such as
vinyl chloride. Because of the time and equipment costs of
the low flow rate purge/sample collection procedure, it also
seems reasonable that empirical evidence of the
comparability of the sample with that of a filtered water
sample may be the basis for planning a monitoring program,
where water samples may be filtered during three events in
a quarterly monitoring program while using the low flow
rate purge/sample collection procedure once a year as a
quality assurance measure.
While procedures to document measurement performance
have been developed for analytical data quality purposes,
such general procedures are difficult to implement for some
field sampling efforts because of the various sampling
methods and the differing hydrogeologic conditions
encountered. However, the quality of data would be
substantially improved if a more formal documentation
process for reporting well installation, development, purging
and sampling were applied, with any limitations or crucial
observations being reported and discussed in the text of the
report. This information is critical for future sampling and
evaluation of groundwater data where the qualities of the
water samples should be compared before the analytical
data are compared.
Finally, research observations or experience at unique sites
needs to be put in the appropriate context rather than being
put in general procedures that are incorporated in all
applications. For example, calculations indicate that the
contribution of a chemical sorbed on colloidal material to
the total measured concentration of a chemical in a
groundwater sample is likely to be negligible unless a large
amount of collodial material is present, the chemicals of
concern are detectable at very low concentrations, or they
29
-------
are highly sorbed. Such an understanding is important for
collecting quality data in a cost effective sampling program.
In conclusion, it is important to recognize the qualities of
the groundwater samples for the correct use of the data. The
selection of sampling procedures should be made with
regard to the uses of the data and limitations of the data, and
this information needs to be documented for data users adn
for planning future sampling efforts.
The authors gratefully acknowledge the assistance of
Montgomery Watson colleagues Dr. Eric Wendlandt in
Walnut Creek, CA, Dr. Peter LeVon in Salt Lake City, UT,
and Mr. Marshall Pauly and Mr. Rodney Vlieger in Des
Moines, IA.
References
J.A. Campbell and W.R. Mabey. 1985. A Systematic
Approach for Evaluating the Quality of Groundwater
Monitoring Data, Groundwater Monitoring Review,
V(4):58-62.
R.W. Puls and MJ. Barcelona. 1989. Ground Water
Sampling for Metals Analyses. EPA/540/4-89/001
R.W. Puls and R.M. Powell. 1992. Acquistionof
Representative Ground Water Quality Samples.
Groundwater Monitoring Review, Summer.
R.E. Wagner, Editor. 1992. Guide to Environmental
Analytical Methods. Genium Publishing Co., New York.
30
-------
CONSIDERATIONS IN SELECTING FILTERED
OR UNFILTERED SAMPLES FOR ANALYSES OF
METALS IN GROUNDWATER SAMPLES
Dennis E. Reece
A basic problem in establishing uniform groundwater
sampling procedures is that procedures that efficiently
provide representative samples for an intended use at one
site may not for others. The decision to use filtered or
unfiltered samples for metals can depend on a number of
site specific factors including:
• intended use of the data
• monitor well construction and conditions
• aquifer properties
• data quality objectives process (the data to be
collected and the associated procedures should be
developed to specifically make decisions or answer
questions for making decisions)
Analysis of unfiltered samples containing suspended solids
associated with well installation or sampling can result in
high metal concentrations which are not representative of
metal concentrations in the groundwater. Alternately, if
metal contaminants are present in the form of mobile
colloids, analysis of filtered samples may not identify the
colloidal fraction of the contamination.
The intended use for the data to be collected is important.
For detection monitoring the main concern is with
consistently identifying the concentrations of mobile
constituents present in the groundwater with adequate
reproduceability to allow determination of whether the data
indicate increases relative to site background conditions.
Determining the mobile concentrations of constituents is
often of primary concern during assessments and corrective
actions. Determining total (dissolved and solid phase)
concentrations present in a well is usually of prime concern
for use of the data in human health risk assessment or for
evaluating suitability of a well for use as a drinking water
supply.
Well construction and development practices and aquifer
properties influence the potential for significant well
siltation and for potential presence of contaminants in
mobile colloidal form.
The Data Quality Objectives process can be used on a site
specific basis to select sampling procedures which are
appropriate for the specific data uses and site conditions.
There is considerable variability in the experience and
technical knowledge among those involved in groundwater
sampling. Complex sampling requirements are often not
implemented appropriately. Cost of the procedures can also
be an important factor in how well the sampling procedures
are implemented. Selection of procedures for collecting
suitable groundwater samples should consider the
complexity and cost of implementation, and should be
practical for a wide range of site conditions rather than ideal
conditions. Required procedures should be implementable
and enforceable.
The selection of filtered or unfiltered samples should be
considered relative to two contradictory issues which may
be present at a site: siltation of monitor wells and possible
presence of contaminants in mobile colloidal form.
by:
The severity of these problems is usually controlled
• well construction
• well development
• aquifer properties
• sampling procedures
Siltation is a widespread problem which can have drastic
impacts on results for metals when unfiltered samples are
analysed. It is an easily identifiable problem and is
sometimes more easily addressed than the issue of whether
mobile colloids are present. Siltation is often greatest in
fine-grained low yield zones such as thin silty or clayey
intervals which are often required to be monitored. If the
sampling conditions result in high solids content or
nonreproduceable solids content in samples, analysis of
unfiltered samples for metals may result in useless data.
Siltation problems can often be mitigated at the well by
proper selection of procedures for well design and
construction, well development and sample collection.
However, experience indicates that these are not always
effective or practical for mitigating siltation.
Siltation problems (for metals analysis) may also be
mitigated by sample pretreatment such as decanting,
centrifuging, macrofiltration, microfiltration or analysis for
both filtered and unfiltered samples. These methods,
however, have limitations and pose a risk of introducing
31
-------
contamination during pretreatment or of removing mobile
colloidal material actually present in the groundwater, thus
reducing the representativeness of the sample.
The issue of possible presence of mobile contaminants in
colloidal form is a less easily identifiable problem. Relative
to siltation, significant colloidal transport of contaminants is
probably a much less widespread problem. The impacts on
monitoring data are also probably less severe for most sites.
Transport of contaminants in colloidal form may be of more
concern in high yield aquifers than in low yield aquifers.
Geochemical conditions in the aquifer matrix may also be a
controlling factor in the potential importance of colloidal
transport.
Colloidal material present in water samples may result from
either artificial means related to well construction and
sampling or from transport of colloidal material in the
groundwater. Proper well design, construction,
development and sampling procedures can reduce the
artificial sources. Microfiltration of the samples may
eliminate from the sample both undesired artificially
introduced colloids and mobile colloids actually present in
the groundwater. Analysis of unfiltered samples will
include colloidal material, if any is present, but will provide
representative data only if siltation does not result in
artificially high suspended solids content.
Groundwater data from three sites were examined to
illustrate the range of impact of siltation on metal results
when unfiltered samples are analyzed for metals.
Site 1 is a landfill site at which waste had not yet been
placed and at which there was no previous development or
use of the site. Four quarters of background monitoring had
been completed on unfiltered samples. The monitoring
system included five monitor well pairs. Each pair included
a monitor well completed in a shallow clayey unit and a
second monitor well completed in a deeper sand unit. The
wells were reportedly sampled using nondedicated
peristaltic or bladder pumps. Turbidity values for the
samples indicated relatively high and variable suspended
solids content in the shallow monitor wells in the clay unit.
Mean turbidity values for these wells varied from < 100 to
7,000 ntu. Values for four of the sand unit wells were much
lower and less variable. Mean turbidity values for these
wells ranged from 15 to 25 ntu. One of the sand unit wells
exhibited higher turbidity values which were suspected to
have resulted from well construction or development
procedures. The high turbidity of the shallow clay unit
wells resulted in high and variable concentrations of some
metals for the unfiltered samples (in particular arsenic,
beryllium, cobalt, copper, nickel, chromium and zinc).
Relatively high correlations were observed between
turbidity values and metal concentrations for the shallow
clayey unit wells. Metal concentrations were highly
variable from quarter to quarter for each shallow clayey unit
well and among different wells due to variability of
suspended solids content. Analysis of the unfiltered samples
limited usefulness of intrawell or interwell statistical
comparison tests. Total dissolved solids values, by contrast
with turbidity values, were relatively constant from quarter
to quarter indicating that dissolved concentrations were
relatively constant. Further evaluation of the data did not
indicate that contamination was present at the site.
Site 2 was a site at which monitor wells were completed in a
relatively thick high yield sand and gravel aquifer. Twenty-
one wells were sampled using high yield dedicated pumps.
Both filtered (0.45 micron) and unfiltered samples were
collected and analyzed for thirteen metals. Turbidity was
low for all samples. Comparison of the results did not
indicate that filtering the samples lowered the
concentrations of metals.
Groundwater samples were collected at Site 3 from soil
borings using downhole tools without installation of
monitor wells. Sampling conditions at this site
approximated a worst case condition with respect to
siltation. All samples were very turbid and contained very
high suspended solids content. Comparison of results of
analyses for 23 metals for filtered (0.45 micron) and
unfiltered samples indicated markedly higher concentrations
in the unfiltered samples for all 23 metals except selenium
and thallium.
32
-------
A STUDY OF THE IMPACT OF MONITORING
WELL PURGING AND FILTERING
TECHNIQUES ON METALS
CONCENTRATIONS IN GROUNDWATER
SAMPLES FROM THE AUBURN ROAD
LANDFILL SITE IN LONDONDERRY, N.H.
Carol White
Introduction
A study of the impact of various sampling techniques on
metals concentrations, in particular iron and arsenic, was
conducted as part of a larger remedial investigation at the
Auburn Road Landfill Site in Londonderry, N.H. A major
objective of the remedial investigation was the development
of a detailed understanding of the site geochemistry.
Collection of samples representative of the in-situ
groundwater chemistry was essential to the success of the
study. At the suggestion of EPA, a low flow-rate purging
and sampling method was employed to collect groundwater
samples from monitoring wells at the site.
Site Description
The Auburn Road Landfill Site is a Superfund site located in
southeastern New Hampshire in the Town of Londonderry.
Municipal and industrial wastes were disposed in three
separate landfills at the site. Contaminants present in the
groundwater at the site as a result of these disposal activities
include low levels several organic compounds: 2-butanone,
trichloroethane, tetrachloroethane, benzene and toluene, and
arsenic. The site consists of glacial outwash sands
overlying moderately fractured metamorphic bedrock. Over
150 monitoring wells have been installed at the site vicinity.
The majority of the wells are constructed of 2-inch diameter
polyvinylchloride screen and riser; screen lengths generally
range from 1 to 15 feet in length.
Sampling Methods
Prior to sample collection, all monitoring wells were purged
using either a peristaltic pump with dedicated tubing, or a
bladder pump. Where possible, dedicated tubing was
installed in each monitoring to mid-screen level several days
or weeks prior to sampling. This enabled the purging to be
conducted with minimal disturbance of the water column.
In general, purging rates ranged from 0.2 to 0.3 1/min.
During purging a flow-through cell equipped with pH, Eh,
specific conductance, dissolved oxygen and temperature
sensors was used to obtain in-line measurements. An
turbidimeter with a separate in-line flow-through cell was
utilized for turbidity measurements. Purging was
considered complete when turbidity measurements had
stabilized (+ 10%). Initially, during the 1991 sampling
program, a minimum 2.5 well volumes were purged prior to
sampling. After a few sampling rounds it became clear that
generaly the in-line measurements stabilized within one well
volume or less, and the minimal well volume criteria was
dropped. All field measurements, purging volumes and
other pertinent sample data were recorded for each well
during the sampling episode.
Results and Discussion
Low-flow Purge Rate: Filtered vs. Unfiltered Samples
In 1991, the low-flow rate sampling methodology was used
to collect groundwater samples from 52 of the existing
monitoring wells at the site. Both filtered and unfiltered
samples were collected for iron and arsenic analysis. Pre-
washed, disposable 0.45 micron filters were used to collect
the filtered groundwater samples from the pump discharge.
As shown on Figure 1, samples collected from wells
included in this study showed an excellent correlation
between filtered and unfiltered arsenic concentrations.
Filtered and unfiltered iron concentrations were also well
correlated, except near the method detection limits.
Low-flow Rate Unfiltered vs. High-flow Rate Filtered
Samples
In 1992, a comparison of the unfiltered samples collected at
a low-flow purge rate (0.2 to 0.3 1/min) versus filtered
samples collected at a high-flow purge rate (> 0.5 1/min)
was conducted. A comparison of the sample results for iron
and arsenic are presented on Figure 2. In all cases, the
filtered samples obtained with high-flow rate method
yielded higher iron and arsenic concentrations than the
unfiltered, low-flow rate samples. These data suggest that
the high-flow rate sampling method may in fact mobilize
particles that can pass through a 0.45 micron filter resulting
in elevated metals concentrations. These results also suggest
33
-------
1.000.000-
e
O
LU
o:
1,000—
11HUM i i nun i i mini i iiiiiiil i 11mill i i nn
(Moeti torn 5 to 26 we*
IIUII i i IIIHI i i MIIN i 11 iiuii i IIIIHII i i Mini
10 100 1,000 10,000 100 on- 1.000.000
FILTERED TOTAL IRON (jig/I)
UJ
V)
Cf.
o
UJ
cr
i i i i i nil i i i i 11 ill i ; i i 11
niiny -r-riiii
100 1.000
FILTERED TOTAL ARSENIC (M8/0
IRON
Slow Ping* Rate
Filtered w. UnRltend
AubiTK Ron Unffll Sto
Londooctany, NH
SAMPLKG EVENTS
April-May 1991
July-August 1981
ARSENIC
Slow Purge Rate
Filtered vi. Unfittered
Auburn R«od Landfill Site
Londonderry, NH
SAMPLING EVENTS
< May-June 1991
• July-August 1991
FIGURE 1
48
Figure 1. Correlations between filtered and unfiltered iron (top) and arsenic (bottom).
34
-------
100,000 —|
80,000 —
60,000 —
40,000 —
20,000 —
IRON
Auburn Roma Landfill Site
Londonderry, NH
C-2
A-41
| | Filtered, Stow Purge, May '81
H|^B Unfltered, Slow Purge. May •SI
Filtered. Fact Purge. Dec. 1892
Unfittered, Slow Purge, Dec 1992
600
400 —
200 —
ARSENIC
Auburn Road Landfll Site
Londonderry, NH
C-2
Filtered. Slow Purge. May '91
UnBtered, Stow Purge, May '91
Filtered, Fast Purge. Dec. 1892
Unratered, Stow Purge, Dec 1992
FIGURE 2
49
Figure 2. Comparison of filtered vs. unfiltered, slow purge and filtered, fast purge unfiltered, slow purge in different
wells for iron (top) and arsenic (bottom).
35
-------
that the data obtained with the low-flow rate sampling
method are more representative of the in-situ ground water
chemistry.
References
Baedecker, N.J. and W. Back. 1979. Hydrogeochemical
processes and chemical reactions at a landfill. Groundwater.
Vol. 17, No. 5 pp. 429-437
Barcelona, M.J., T.R. Holm, M.R. Schock and G.K George.
1989. Spatial and temporal gradients in aquifer oxidation-
reduction conditions. Water Resources Research Vol.
25,No. 5 pp. 991-1003.
Chapelle, F.H. 1993. Groundwater Microbiology and
Geochemistry. John Wiley and Sons, New York, NY. 424 p.
Puls, R.W. and M.J. Barcelona. 1989. Groundwater
sampling for metals analysis. Superfund Ground Water
Issue Paper USEPA-ORD, OSWER EPA/540/4-89/001. 6 p.
Sevee and Maher Engineers. 1992. Pre-Design Investigation
for the Remediation of Groundwater, Auburn Road Landfill
Site, Londonderry, N.H.
Sevee and Maher Engineers. 1993. Supplement II
Investigation Report, Supplemental Pre-Design
Investigation for the Remediation of Groundwater, Auburn
Road Landfill Site, Londonderry, N.H..
Walton-Day, K., D. Macalady, M. Brooks, V Tate. 1990.
Field methods for measurement of groundwater redox
chemical parameters. Groundwater Monitoring Review.
Vol. 10, No. 4 pp. 81-89
36
-------
GROUNDWATER SAMPLING IN
FRACTURED CLAY AND ROCK
Larry D. McKay, Kent Novakowski, and John F. McCarthy
Introduction
Deposits of fractured clay and rock are widespread and
many contaminated sites are located in such materials.
Groundwater flow and contaminant migration in these
materials are controlled by different factors than in granular
media and sampling/monitoring programs must reflect these
differences if they are to be effective.
Hydraulic Properties of Fractured Media
In a fractured porous medium, flow is largely controlled by
the distribution, orientation, length and aperture of fractures
or joints. In many materials flow through the blocks of
"matrix" between the fractures is much less than through the
fractures. Measurements of hydraulic conductivity are
largely dependent on how many hydraulically-conductive
fractures are intersected by the measurement zone (typically
in a borehole). As a result, measured values can vary greatly
and often depend on the length and orientation of the
measurement zone. Not all fractures are hydraulically-
conductive and at some sites investigators have found that a
few large aperture fractures may dominate an entire flow
system. One of the most critical parameters governing flow,
aperture, cannot be directly measured and is almost always
inferred from hydraulic or solute transport data, usually
using the "cubic law" (Snow, 1970). Hence there is always a
large degree of uncertainty associated with fracture aperture
values. The fracture porosity of a deposit generally
represents only a small fraction (10-2 to 1O5) of the total
volume of a deposit and is often much less than the
intergranular porosity of the deposit (<0.01 to 0.7).
Case Study - Borehole Flow Meter Survey at Oak Ridge
Reservation
Recent borehole flow meter surveys at the Oak Ridge
Reservation in Tennessee (Will, et al., 1992) illustrate some
of the problems encountered in fractured media. Based on
drilling records, core samples and geophysical/downhole-
camera surveys of a 405 foot deep borehole, CH-9, it
appeared that the shales at this site were highly fractured
with typical fracture spacings of a few inches to a few feet.
However, an electromagnetic flow meter survey under
ambient conditions (no pumping) indicated that flow was
restricted to two narrow zones at 135 and 330 foot depth.
Flow was found to enter the deeper zone, then flow up the
well bore and exit into the shallow fracture zone with a flow
rate of up to 0.2 gpm or about 700 gal per day. This presents
several potential problems: possible mixing of contaminated
and uncontaminated waters; and if you don't have a flow
meter and the well is to be completed with a multilevel
sampler to prevent uphole flow, how do you decide where to
put the sampling zones?
Contaminant Transport in Fractured Porous Media
The transport of contaminants through fractured p.m. is
highly dependent on the physical properties of the
contaminants with solutes, colloids and immiscible phase
liquids behaving in radically different manners. Solute
transport is strongly influenced by matrix diffusion, which is
the transfer of solute mass from the zone of rapid flow
within the fractures into the relatively immobile pore water
in the blocks of matrix between fractures. Recent field
experiments (McKay et al., 1993) and model simulations
(ex: Sudicky & McLaren, 1992) have shown that matrix
diffusion is sufficient to retard migration of a non-reactive
solute by several orders of magnitude or more relative to
fracture flow velocities in fractured high porosity sediments.
Although we expect to see the greatest retardation in high
porosity clays and shales recent experiments (Birgersson &
Neretieks, 1990) indicate that this can be significant even in
granitic rock.
Colloidal contaminants, because of their larger diameters,
are not as strongly influenced by matrix diffusion and can
actually migrate faster than non-reactive solutes (McKay et
al., 1993). Immiscible phase contaminants, particularly
DNAPL's (dense non-aqueous phase liquids) are also not
strongly influenced by matrix diffusion and can move
rapidly downwards through fracture systems with
movement controlled by the density and viscosity/interfacial
tension properties of the fluid, and the size of fracture
apertures (or openings). The residual DNAPL, which coats
the fracture walls, may then be slowly dissolved and
transported away by the flowing groundwater or may
diffuse directly into the matrix pore water (Keuper &
McWhorter, 1991).
Implications for Groundwater Monitoring
Some common problems which are expected in fractured
media include:
37
-------
• contaminant plumes which are very irregular in shape
and relative concentration (due to variations in the
distribution of hydraulically-conductive fractures),
• difficulty in distinguishing which fractures are contri-
buting to flow in a borehole and hence difficult to
determine where the contaminants are coming from.
• potential for ambient flow along the borehole,
resulting in mixing or (even worse) spreading of
contaminants into previously uncontaminated zones
• problems with sample dilution due to large volume of
sampling interval in conventional boreholes/wells
(Novakowski, 1992).
• purging prior to sampling may draw in water from a
large volume of the deposit (because of the very low
fracture porosity) and this water may not be
representative of what was initially in the fractures
and is likely not in equilibrium with pore water in the
matrix adjacent to the fractures. For example a 10
litre water sample taken from a well in a rock with a
fracture porosity of 1O4 could draw water from up to
100m3 of the rock.
Improvement of Monitoring and Sampling Methods
Monitoring and sampling methods should be evaluated on a
site-specific basis with respect to both the properties of the
fractured media and to the type of contaminant. Possible
improvements for existing wells could include:
• use of downhole methods including flow meter logs,
temp, logs and conductivity logs to identify location
of hydraulically-conductive fracture zones.
• installation of packers or seals to minimize borehole
storage volumes and to prevent flow along the
borehole
• sampling using minimal purging
New monitoring programs could include
• angled or horizontal boreholes to increase the
probability of intersecting vertical fractures (where
needed)
• use of monitoring wells designed to minimize storage
volume and prevent flow along the borehole
(examples: Westbay system, Waterloo multi-level)
• sampling of matrix pore water directly from core
samples. In clays a variety of methods have been
developed including: leaching with de-mineralized
water, squeezing out pore water in high pressure
cells, displacement with toluene, and finally direct
measurement of vapour phase contaminants in small
holes drilled into the core sample.
• measurement of contaminant concentrations along
streams to identify specific fracture discharge zones
(Clappetal., 1992)
References
Birgersson, L., and I. Neretnieks, Diffusion in the matrix
of granitic rock: Field test in the Stripa mine, Water Resour.
Res., 26(11), 2833-2842, 1990.
Clapp, R.B., D.S. Hicks, O.K. Solomon, D.M. Borders,
D.D. Huff, and H.L. Boston, Groundwater, surface water,
and movement of contaminants at the Oak Ridge
Reservation, Abstract presented at American Geophysical
Union meeting, San Francisco, CA, Dec., 1992.
Kueper, B.H., and D.B. McWhorter, The behaviour of
dense, non-aqueous phase liquids in fractured clay and rock,
Ground Water, 29(5), 716-728, 1991.
McKay, L.D., J.A. Cherry, R.C. Bales, M.T. Yahya, and C.P.
Gerba, Afield example of bacteriophage as tracers of
fracture flow, Environ. Sci. and Technol., Vol. 27, No. 6, p.
1075-1079, 1993.
McKay, L.D., R. W Gillham, and J.A. Cherry, Field experi-
ments in a fractured clay till: 2. Solute and colloid transport,
Water Resour. Res., 29(12), 3879-3890, 1993.
McKay, L.D., and J.A. Cherry, Groundwater research in
clay-rich glacial tills in southwestern Ontario, Paper
presented at Intl. Assoc. of Hydrol. Conference, Hamilton,
Ontario, May, 1992.
Novakowski, K.S., The analysis of tracer experiments
conducted in divergent radial flow fields, Water Resour.
Res., 28(12), 3215-3225, 1992.
Snow, D.T., The frequency and apertures of fractures in
rock, J. RockMech. Min. Sci., Vol. 7, 23-40, 1970.
Sudicky, E. A. and R.G. McLaren, The LTG technique for
large-scale simulation of mass transport in discretely
fractured porous formations, Water Resour. Res., 28(2), 499-
514, 1992.
Will, A.S., J.R. Kannard, A.R. Day, and L.B. Shannon,
Additional borehole geophysical logging at waste area
grouping 1 at Oak Ridge National Laboratory, Oak Ridge,
TN, ORNL, Environ. Restoration Div., Technical
Memorandum 01-04, 1992.
38
-------
EVALUATION OF FIELD-
FILTRATION VARIABLES FOR
REPRESENTATIVE SAMPLES OF
TRACE METALS IN GROUND
WATER
Karl F. Pohlmann , Gary A. Icopini and
Charlita G. Rosal
Abstract
Selected ground-water sampling and field-filtration methods
were evaluated to determine their effects on trace metal
concentrations in ground-water samples. The study focused
on conditions where traditional approaches may produce
samples with high particle concentrations and concomitant
elevated metals concentrations. These samples are often
filtered to remove suspended particles before laboratory
chemical analysis; however, filtration may also remove
colloidal particles that may be important to the transport of
trace metals. Two filtration variables were evaluated in this
study: (1) collecting samples without filtration, and (2)
filtering samples with 0.45-|j,m pore size filters. Samples
were collected with a bailer, a submersible pump at a "low"
rate of 0.3 L/min, and a submersible pump at a "moderate"
rate of 1.0 L/min. Pump discharge rates were controlled by
pump speed rather than by flow restrictors or valves.
The results showed that some sampling methods entrained
large quantities of particles greater than 0.45 |jm in size.
The quantities and sizes of these particles suggested that
they were not mobile in ground water under natural flow
conditions but were primarily artifacts of well construction,
development, and purging. Analysis of unfiltered samples
containing high concentrations of these artifactual particles
and associated metals resulted in metal concentrations that
were often orders-of-magnitude higher than in
corresponding 0.45-(j,m-filtered samples. This effect was
most consistent and pronounced in bailed samples, because
operation of the bailer caused the greatest agitation in the
sampling zone. The use of pumps at low to moderate rates
resulted in minimal concentration differences between
unfiltered and 0.45-jjm-filtered samples in all but the most
turbid wells, and reflected the entrainment of only minor
amounts of artifactual particles larger than 0.45 |jm in size.
The three sample collection methods produced similar
results when samples from less turbid wells were filtered,
however, the pumping methods produced the most
consistent overall results. Little variation was evident
between filtered and unfiltered pumped samples, reflecting
minimal agitation in the sampling zone and sample during
purging and sample collection. Use of submersible pumps at
low speeds may reduce the uncertainty in results when
collecting samples of inorganic ground-water constituents
that have the potential to associate with particles in ground
water.
Introduction
Ground-water samples are commonly field-filtered to
remove sediments mobilized during well construction
and sampling because inclusion of these particles may
bias analytical determinations, leading to elevated and
erroneous concentrations of mobile contaminants (Puls
et al., 1991; Backhus et al., 1993). However,
indiscriminant field filtration using 0.45-jjm filters
ignores the presence of colloidal particles in ground
water that may exist between the extremes of solutes and
sediments. The small size of colloids facilitates their
mobility in certain ground-water systems and provides
them with high ratios of surface area to mass, which
increases their relative sorptive capabilities (McDowell-
Boyer et al., 1986). The association of metals with
colloids has been shown to provide a potentially
important mechanism for transport of these metals in
ground water (McCarthy and Zachara, 1989).
Routine filtration of ground-water samples may have
particularly important implications for metal determinations
when metals are associated with particles larger than the
filter pore size and/or when turbid samples are collected.
Turbid samples may result when bailers or submersible
pumps operated at moderate to high discharge rates (greater
than 1 L/min) are used in inadequately designed,
constructed, or developed wells, or wells completed in
formations containing fine-grained sediments. Collecting
samples at rates that approach natural ground-water
advective flow velocities may minimize disturbance in the
sampling zone, reduce entrainment of normally immobile
species, and thereby alleviate the need to filter samples. This
approach to sampling has been advocated by several
researchers, with maximum suggested pumping rates of 100
to 300 mL/min (Ryan and Gschwend, 1990; Puls et al.,
1990; Backhus etal., 1993).
The U.S. Environmental Protection Agency (EPA) is
interested in the implications of field-filtration on metal
39
-------
concentrations in ground-water samples. This paper
summarizes some of the results of a field study addressing
this issue (the work presented here is fully described in
Pohlmann et al., 1994). The objectives of the study were to
(1) evaluate the impacts on trace-metal concentrations of
filtration with 0.45-^m pore size filters versus not filtering;
and (2) investigate the effects of interactions between field
filtration and other sampling variables including sampling
device, pumping rate, and well turbidity. The study focused
on sampling in conventional standpipe monitoring wells
under conditions where traditional approaches to sampling
may produce turbid samples. Three field sites were visited:
an active municipal solid waste landfill in Wisconsin, a
closed solid waste landfill in Washington, and a site
contaminated by industrial waste in Nevada.
Methods and Materials
The monitoring wells sampled in this study were
constructed of poly vinyl chloride, and were 5.1 cm in
diameter, with the exception of one 10.2 cm diameter well.
The top of the well screens ranged from 2 to 19m below
ground surface, with well screen lengths of 0.6 to 6.0 m.
The static water level ranged from 1 to 14m below ground
surface. Volumes of water within the well screens ranged
from 1.2 to SOL.
The results of three sample collection methods will be
described in this paper. The first method used a dual check
valve bailer with a volume of approximately 0.4 L. Samples
were transferred from the bailer directly to sample bottles
for unfiltered samples or to a filtration vessel for filtered
samples. Compressed nitrogen gas was used to drive the
samples through either membrane filters or disposable
cartridge filters. The second sampling method was a
submersible centrifugal pump (CP) operated at an
appropriate speed to produce a flow rate at the surface of
approximately 300 mL/min. Filtration was conducted on-
line with disposable cartridge filters. The third method was
a bladder pump (BP) operated at an appropriate speed and
pressure to produce a flow rate at the surface of
approximately 1 L/min. Filtration was conducted in the
same manner as for the centrifugal pump. The pumps and
bailer were positioned to collect samples from about 0.6 m
below the top of the well screen.
Measurements of turbidity, dissolved oxygen (DO),
temperature, electrical conductivity (EC), and pH of the
pump discharge were made on-line, while measurements of
these parameters for the bailer discharge were made off-line.
Stabilization of these parameters provided an indication of
equilibrium between incoming ground water, the action of
the sampler, and stagnant water in the well; thereby
suggesting that purging was complete. The relative values of
these parameters also provided a means for comparing the
sampling methods with respect to their ability to minimize
disturbance in the sampling zone. Estimates of particle size
distribution were determined gravimetrically by serial
ultrafiltration using microfilters of 5.0 |jm, 0.4 |jm, 0.1 |jm,
and 0.03 |jm pore size.
Results and Discussion
The relative disturbance in the sampling zone caused by a
sampling method was most evident in the field
measurements of turbidity and DO, particularly under low
well-yield conditions. When the discharge rate exceeded the
well yield, the increasing hydraulic gradient between the
formation and the well mobilized large quantities of
particles, thereby elevating turbidity values. Continued
removal of water from the well dewatered the filter pack,
leading to gravity drainage of pore water and sediments and
continually increasing turbidity values. Bailer turbidity
values were further elevated by the surging action of the
bailer which mobilized large quantities of particles in the
well. Elevated DO values of the bailer and BP at 1 L/min in
low-yield wells reflect the formation of a large air-water
interface which increased the potential for oxygenation of
incoming ground water as the filter pack was dewatered.
The bailer caused additional aeration of the samples as a
result of the increased exposure to the atmosphere during
sample collection and transfer. The lower discharge rate of
0.3 L/min, which was generally closer to the well yield,
resulted in less variability and more representative values of
turbidity and DO, as well as lower purge volumes.
Somewhat less variable results were observed between
sampling methods in wells where the purging and sampling
rate did not exceed the well yield. Under these conditions,
hydraulic gradients into the well were minimal, the filter
pack was not dewatered, and turbidity was generally lower.
The two pumping methods produced similar values of most
field measurements, while the surging action of the bailer
produced turbidity values that were approximately two
orders of magnitude higher than those produced by the
pumps. Likewise, DO values in bailed samples were
elevated with respect to the pumped values, an artifact of the
bailing process. As a result, the pumps produced equilibrium
DO and turbidity conditions with relatively low purge
volumes, while the bailer produced high values of these
parameters and did not reach equilibrium after greater purge
volumes.
In almost every case, samples collected by bailer contained
higher particle concentrations than those collected by the
pumps, with the greatest differences occurring at the most
turbid wells. Furthermore, the size distribution of particles
in most bailed samples was highly skewed toward larger
particles, with over 96 percent larger than 0.45 |jm, and
generally over 93 percent larger than 5.0 |jm. The quantities
and sizes of these particles suggest that they were not
mobile in ground water under natural flow conditions but
40
-------
were primarily the artifacts of well construction,
development, and purging and were mobilized by agitation
in the sampling zone caused by bailing. The particle size
distribution in samples pumped from the most turbid wells
(the low-yield wells) were also skewed toward larger
particles, but total particle concentrations were much lower
than in the bailed samples. In the less turbid (high-yield)
wells, total particle concentrations in pumped samples were
orders-of-magnitude lower than in bailed samples, reflecting
the lower degree of agitation caused by the pumping
methods. Also, particle sizes in the pumped samples were
generally more uniformly distributed; approximately 50
percent of the particles were larger than 0.45 |jm.
Differences in metal concentrations between filtered and
unfiltered samples were most evident in low-yield and
highly turbid wells, particularly when the samples were
collected by bailer. In fact, several metals present in
unfiltered bailed samples were below detection levels in the
corresponding filtered samples. The large differences in
concentration between filtered and unfiltered bailed samples
reflect the association of metals with the high concentrations
of artifactual particles entrained during bailing. For
example, iron in the sampling zone likely existed as iron
hydroxide particles, particles containing elemental iron, and
ferrous iron sorbed to particle surfaces. Removal of the
majority of particles during filtration therefore greatly
reduced iron concentrations in the filtered samples. Other
metals likely existed as aqueous species sorbed to particle
surfaces, or as elemental components of particles originating
as aquifer solids, and their concentrations were similarly
reduced by filtration. Additionally, ferrous iron may have
oxidized and precipitated during bailing, transfer, and
filtering of the samples, and then removed during filtration.
Finally, the formation of a thick filter cake during filtration
of bailed samples likely reduced the effective pore size of
the filter membrane, thereby blocking passage of some
particles smaller than 0.45 |jm; this would further reduce the
concentrations of associated metals in the sample.
Trace metal concentrations in unfiltered samples pumped
from low-yield and highly turbid wells were generally lower
than in unfiltered samples bailed from the same wells. This
reflects the lower degree of agitation associated with
pumping and, as a result, the lower artifactual particle
concentrations. Removal of the larger particles in the
pumped samples did, however, cause filtered samples to
contain lower metal concentrations than unfiltered samples,
though the differences in concentration were much lower
than in bailed samples. Unfiltered metal concentrations in
samples pumped at 1 L/min were often slightly higher than
in samples pumped at 0.3 L/min, but the concentrations in
the filtered samples from both pumps were essentially the
same. Furthermore, metal concentrations in filtered pumped
samples did not differ significantly from those in filtered
bailed samples.
In less turbid and high-yield wells, unfiltered bailed samples
usually contained the highest metal concentrations of all
samples, but the differences between these concentrations
and concentrations in filtered samples were much smaller
than for low-yield and turbid wells. Several metals showed
only slight differences between filtered and unfiltered
results in bailed samples. These results reflect the lower
proportion of artifactual particles removed during filtration
as compared to the low-yield and turbid wells, but also are
related to metal speciation at each well. Differences between
filtered and unfiltered pumped samples were minimal, and
the concentrations were essentially the same as those in the
filtered bailed samples, despite the variability in proportion
of particles smaller than 0.45 |j,m. This suggests that many
metals existed primarily as dissolved species and/or were
associated with particles smaller than 0.45 |jm in the less
turbid and high-yield wells included in this study.
Conclusions
The effects of field filtration on trace metal concentrations
were most evident when a bailer was used to sample low-
yield and/or turbid wells. Concentrations in unfiltered bailed
samples were up to several orders-of-magnitude higher than
in filtered bailed, filtered pumped, and unfiltered pumped
samples. Elevated metal concentrations in unfiltered bailed
samples reflected the entrainment of large quantities of
normally immobile artifactual particles and their associated
matrix metals, and unknown quantities of contaminant
metals. Pumping at low to moderate rates in low-yield and/
or turbid wells resulted in less agitation in the sampling
zone, lower particle concentrations, and reduced effects of
field filtration on metal concentrations.
The effects of field filtration were the least evident in high-
yield wells and/or wells having lower turbidity. Samples
bailed from these wells exhibited much smaller differences
between unfiltered and 0.45-(j,m-filtered samples. However,
bailing clearly mobilized artifactual particles that caused
elevated metal concentrations in most unfiltered bailed
samples. In contrast, samples pumped from these wells
exhibited virtually no differences between unfiltered and
filtered samples, reflecting the minimal entrainment of
artifactual particles larger than 0.45 |jm during sampling at
low to moderate pumping rates. Concentrations in filtered
samples bailed from high-yield wells and/or from wells
having lower turbidity were generally equivalent to
concentrations in pumped samples. This reflects the removal
of larger, normally immobile artifactual particles and
associated metals from the bailed samples.
Although the three sample-collection methods generally
produced similar results when samples from less turbid
wells were filtered, the pumping methods produced the most
consistent overall results. Most metals showed little
variation between filtered and unfiltered pumped samples,
41
-------
reflecting the minimal agitation in the sampling zone and
sample during purging and sample collection. Use of
submersible pumps at low speeds may reduce the
uncertainty in results when collecting samples of inorganic
ground-water constituents that have the potential to
associate with particles in ground water.
Acknowledgements
The authors wish to acknowledge the technical assistance of
James Brown, Jane Denne, Robert Puls, and Andrew
Teplitzky; and the logistical assistance of Jack Connelly,
Brian Haelsig, Nadine Romero, and Bernie Zavala. This
work was supported by the EPA Environmental Monitoring
Systems Laboratory, Advanced Monitoring Systems
Division, under Cooperative Agreement #CR815774 with
the University and Community College System of Nevada,
Desert Research Institute, Water Resources Center.
Notice
The U.S. EPA, through its Office of Research and
Development (ORD), prepared this extended abstract for a
workshop presentation and proceedings. It does not
necessarily reflect the views of the EPA or ORD.
References
1. Backhus, D.A., J.N. Ryan, D.M. Groher, J.K.
MacFarlane, and P.M. Gschwend. Sampling Colloids
and Colloid-Associated Contaminants in Ground Water.
Ground Water, 31 (3): 466-479, 1993.
2. McCarthy, J.F., and J.M. Zachara, Subsurface Transport
of Contaminants. Environmental Science Technology,
23(5): 496-502, 1989.
3. McDowell-Boyer, L.M., J.R. Hunt, and N. Sitar.
Particle Transport Through Porous Media. Water
Resources Res., 22(13): 1901-1921, 1986.
4. Pohlmann, K.F., G.A. Icopini, R.D. McArthur, and C.G.
Rosal. Evaluation of Sampling and Field-Filtration
Methods for the Analysis of Trace Metals in Ground
Water. U.S. Environmental Protection Agency, Las
Vegas, Nevada, in preparation, 1994.
5. Puls, R.W. and M.J. Barcelona. Groundwater Sampling
for Metals Analysis. EPA/540/4-89/001, U.S.
Environmental Protection Agency, Ada, Oklahoma,
1989, 6 pp.
6. Puls, R.W., J.H. Eychauer, and R.M. Powell. Colloidal-
Facilitated Transport of Inorganic Contaminants in
Ground Water: Part I. Sampling Considerations. EPA/
600/M-90/023, U.S. Environmental Protection Agency,
Ada, Oklahoma, 1990. 12 pp.
Puls, R.W, R.M. Powell, D.A. Clark, and C.J. Paul.
Facilitated Transport of Inorganic Contaminants in
Ground Water: Part II. Colloidal Transport. EPA/600/
M-91/040, U.S. Environmental Protection Agency, Ada,
Oklahoma, 1991. 12pp.
Ryan, J.N. and P.M. Gschwend. Colloid Mobilization in
Two Atlantic Coastal Plain Aquifers: Field Studies.
Water Resources Res., 26(2): 307-322, 1990.
42
-------
MONITORING WELL SAMPLING -
YOU CAN'T ALWAYS GET WHAT
YOU WANT BUT CAN YOU GET
WHAT YOU NEED?
Jack Connelly
Introduction
In 1985, Wisconsin was one of the first states to develop
guidelines for installing monitoring wells, and in 1990 we
became the only state to develop comprehensive rules for
monitoring well installation and development. Wisconsin's
monitoring well installation rule includes many
requirements to ensure that the water produced from a
monitoring well is representative of groundwater.
Wisconsin also recognized the need for consistent sampling
procedures, so in 1987 we developed Groundwater
Sampling Procedures Guidelines that we require landfill
owners and consultants to follow. We are proud of both our
sampling procedures guidelines and our monitoring well
installation rule, and these documents are in demand by
many states' researchers and regulators. The guidelines
require field filtering of groundwater samples collected for
inorganic analysis. This requirement is based upon our 15+
years of infield sampling experience and the geology of our
state. Therefore, when the final Subtitle D criteria were
published, we were shocked that EPA had banned the use of
field filtering. We were especially surprised that such a
rigid policy would be included in an otherwise flexible rule.
We surveyed those sampling Wisconsin landfills to
investigate the impact of changing our sampling
requirements. The results from 305 respondents (see
Appendix I for complete results) show that:
1. 80 percent of the samplers use a bailer to purge
their wells,
2. 89 percent use a bailer to retrieve samples from the
wells,
3. over 55 percent of the sampled wells are turbid.
Many of the landfill monitoring wells in Wisconsin are
turbid because we encourage locating landfills in fine-
grained soils to limit contaminant migration. Wells
screened in tight soils often do not clear up during
development, and if bailed samples are not filtered, metals
results are erroneously high, variable and misleading. Table
1 lists results from an upgradient unimpacted well at a
foundry landfill. We typically see very high metals values
in unfiltered samples ("Bailer Unfiltered" results) that also
vary from one sampling period to the next. The variability
and elevated metals results are caused by well turbidity, not
by groundwater contamination. Columns C and D in Table
1 more accurately represent groundwater quality. Many
such sites with turbid wells would be required to perform
Assessment Monitoring under Subtitle D because of falsely
elevated metals values.
The impact of banning field filtering in Wisconsin would be
significant. Wisconsin has 15 to 20 years of data based on
filtered samples. We have detected contamination and taken
action at many sites relying primarily on this method. To
determine which sites to investigate we do not rely on
unfiltered turbid samples that produce falsely elevated
metals values but instead rely on VOCs and trends in non-
metallic inorganic parameters.
Table 1. Metals Concentrations (jJ^g/L) from
Unimpacted Well at Foundry Site
Arsenic
Cadmium
Chromium
Lead
A
MCL
50
5
100
15
B
Bailer
Unfiltered
120
9.2
450
320
C
Bailer
Filtered
<2
<0.5
<2
<5
D
Low Flow
Unfiltered
<1
0.29
1.2
<1
Recent advances in monitoring technology have produced
the relatively new low-flow pumping technique (LFPT),
improving the quality of unfiltered samples, and recent
experience shows that the LFPT for purging and sampling
can produce reliable samples. After review of the articles
referenced in the Subtitle D criteria as background to this
issue, we agree that colloidal material may indeed move
through some aquifers. Since contaminants may be
adsorbed on colloids, the filtering of groundwater samples
43
-------
could remove one source of contamination. However, as
illustrated in the data from columns C and D of Table 1, the
significance of the colloidal source is unclear. Few studies
have actually compared results from contaminated samples
that have been bailed and filtered with results from samples
collected using the LFPT. The evidence from these studies,
discussed below, has been mixed and inconclusive. In
addition, there has not been a study of the practicality of
using the LFPT.
Is Low-Flow Pumping Practical?
EPA contracted with the Desert Research Institute (DRI) to
study the filtering issue for a year. Wisconsin's Outagamie
County Landfill was one of three sites DRI chose. DRI
collected both filtered and unfiltered samples using the
LFPT one week after the County collected samples using a
bailer and field filtering. All three types of sampling results
were similar for metals. We were impressed with the
LFPT's ability to minimize turbulence and aeration. The
LFPT produced "clear" samples from wells that had
produced highly turbid samples when bailed the previous
week.
We made the following additional observations about the
LFPT:
1. There is a significant amount of bulky equipment
required: a heavy reel, 100 feet of hose and
electrical cable, a regulator, a 60-90 pound
generator (or battery or nitrogen tank depending on
pump type), meters for measuring conductivity,
DO, pH, temperature and turbidity and a flow
through cell. (Many Wisconsin wells cannot be
reached by vehicle.)
2. It is time-consuming to transport and set-up the
equipment, purge the well and collect the sample.
(An average of two wells were sampled per day.)
3. The pump sometimes had difficulties pumping at
very low rates.
4. The equipment would be very difficult to operate in
Wisconsin type winter conditions (e.g. if the pump
turned off, water remaining in the tubing would
freeze in a matter of minutes).
5. Low-flow pumping could not be used in wells that
could be purged dry. (Many monitoring wells in
Wisconsin can be purged dry.)
One of our biggest practical concerns with the LFPT is the
amount of time required to sample wells. A typical
Wisconsin landfill remaining open under Subtitle D would
have 50 monitoring wells. It would take samplers 1 week to
sample such a site using a bailer and field filtering, while it
would take 3-5 weeks to sample the site using the LFPT. An
ideal set-up for sampling would be a dedicated low-flow
pump because there would be less equipment to transport
and freezing would be less likely since the tubing is in the
well, not exposed to the air and less time would be needed
for decontamination. However, it is not realistic to expect
that most landfill owners will install a dedicated pump
system in each of its monitoring wells immediately to
comply with the Subtitle D criteria. As part of our survey of
groundwater samplers we asked them what type of
dedicated sampling equipment they were using. Seventy-six
percent of the samplers did not use any type of dedicated
system, 16% used dedicated bailers and only 8% used
dedicated pumps. Although dedicated pumps have been
available for a number of years, most landfill owners in
Wisconsin are not choosing to use them.
We evaluated the LFPT at facilities with high levels of
metals, such as plating companies. At the Riverside Plating
Company a consultant sampled several monitoring wells
using a bailer and collected both a filtered and unfiltered
sample from each well. We sampled the same wells using
the LFPT during the month following the consultant's
sampling. We experienced a number of problems during
this sampling. We had hoped to sample 3 to 4 wells in a day
but found our rate of sampling to be similar to DRI's - a
maximum of 2-3 wells per day. The pump would shut off
unexpectedly while purging some of the wells. It had to be
restarted at a high rate which created added turbulence and
increased the purging time. It took approximately 45
minutes to set up and calibrate the equipment prior to
sampling each well and slightly less time to take down the
equipment following sampling. Purging and sampling,
excluding set up and take down time, took anywhere from
1-4 hours for each well.
How Do Results Compare?
To date, the metals results comparing bailing and field
filtering to low-flow pumping have been inconclusive.
Table 1 represents data collected at an unimpacted well in
Wisconsin where metals are high only in bailed and
unfiltered samples due to turbidity. The bailing and field
filtering and low- flow pumping produced very similar
results indicating no metals contamination. Researchers
sampling for arsenic using the two methods found
somewhat lower values using bailing and field filtering than
low-flow pumping (Puls et al. 1992 and Puls and Powell
1992), but the values were close enough that under our
enforcement standards and procedures, we would have
taken the same enforcement action using the results from
either method. Chromium results have varied depending on
the level of contamination. Table 1 illustrates similar results
at low levels of contamination. Table 2, representing
preliminary results from Wisconsin's study, illustrates
44
-------
Table 2. Chromium Concentrations (|Jg/L) at Two Impacted Wells - Riverside Plating Co.
WellP-3
Bailer
Unfiltered
Bailer
Filtered
Low Flow
Unfiltered
WellP-4
Bailer
Unfiltered
Bailer
Filtered
Low Flow
Unfiltered
Chromium
77
20
14
340
470
292
similar results for the two methods at medium levels of
chromium contamination (see Well P-3) and exaggerated
levels of contamination using bailing and field filtering at
high levels of chromium contamination (see Well P-4). Puls
et al. (1992) also found exaggerated levels of chromium
when using the bailing and field filtering method. However,
in a separate study Paul and Puls (1993) found "very little
difference" in chromium concentrations between the bailer
and low-flow pumps at high levels of chromium
contamination.
Conclusion
Wisconsin has over 9,000 landfill monitoring wells, of
which over 85 percent are bailed and over 50 percent
produce turbid samples. If turbid samples collected from
these wells are not filtered, the results will be erroneously
high for metals and these erroneously high values will force
many landfill owners into Assessment Monitoring under
Subtitle D.
A logical alternative to bailing is the low-flow pumping
technique, which can produce clear samples without
filtering in many circumstances. However, before requiring
landfills and others to use this technique two critical
questions need resolution:
1. Does bailing and field filtering produce
significantly different results from low-flow
pumping? (The results of the comparison to date
have been preliminary and inconclusive.)
2. Can the low-flow pumping technique be
implemented practically? (The evidence indicates
it would be difficult to implement in the field but
more evidence should be gathered.)
We propose postponing the ban on field filtering while EPA
and the states gather additional evidence to answer these
two questions. What risk is involved in postponing the ban?
Very little, if any, at municipal solid waste landfills. Our
experience has been that metals rarely migrate beyond 100
feet of our landfills. We have found that more mobile
constituents such as VOCs and chloride are far better
indicators of contaminant releases from landfills than are
metals.
In Wisconsin we have used results from our current
techniques of bailing and field-filtering for inorganics and
bailing without filtering for VOCs to require groundwater
investigations at 100 landfills and implement remedial
action at over 50 of these landfills. We believe that our
monitoring and remediation programs would not be
significantly improved by adding requirements for low-flow
pumping.
We suggest that EPA use some of the flexibility used
throughout Subtitle D, at least while we gather additional
data to answer the two questions posed above. We
recommend that EPA:
1. Lift the ban on field filtering until more research is
completed.
2. Collect additional metals data from individual
wells using both bailing and field filtering and low-
flow pumping.
3. Determine whether there is a significant difference
between the metals values using the two
techniques.
4. Evaluate ways to improve bailing and field filtering
if there are significant differences between the two
techniques.
5. Evaluate the practicality of the low-flow pumping
45
-------
technique, especially for sites in northern climates,
in fine grained soils where wells can be purged dry
and sites with a large number of monitoring wells.
No matter which sampling technique seems best
scientifically, if it can't be implemented, it serves no
purpose. Returning to the title of my presentation, you can't
always get what you want. What all of us would love to
have is the ideal, perfect sampling technique. But if you try
sometime, and balance the ideal with what works in the
field, you just might find, you get what you need.
References
Paul, CJ. andR.W. Puls. 1993. Comparison of Ground-
Water Sampling Devices Based On Equilibration of Water
Quality Indicator Parameters. EPA/600/A-93/005, 13 pp.
Puls, R.W. andR.M. Powell. 1992. FOCUS PAPER:
Acquisition of Representative Ground Water Quality
Samples for Metals. Ground Water Monitoring Review, v.
12, no. 3, pp. 167-176.
Puls, R.W., D.A. Clark, B. Bledsoe, R.M. Powell and CJ.
Paul. 1992. Metals in Ground Water: Sampling Artifacts
and Reproducibility. Hazardous Waste and Hazardous
Materials, v. 9, no. 2, pp. 149-162.
46
-------
RESULTS OF WISCONSIN'S GROUNDWATER SAMPLING PROCEDURES SURVEY
This survey was sent on October 13, 1993 to those sampling groundwater at 470 Wisconsin landfills, of which 305
had replied at the time the results were compiled.
1. At about how many of the following types of Wisconsin landfills do you collect groundwater samples?
(e.g., if you sample at 2 municipal and 3 demolition landfills, fill in 2 in front of "Municipal Solid Waste"
and 3 in front of "Demolition")
58% Municipal Solid Waste 8% Demolition
29% Industrial 5% Other
2. Of the above landfills, about how many are
31% Active (taking waste) 54% Closed
3. At about how many landfills do you use the following equipment to purge the wells? (e.g., if you use a
bailer at 3 landfills and a bladder pump at 21andfills, fill in 3 in front of "Bailer' and 2 in front of "Bladder
Pump")
80% Bailer 0.2% Gas Displacement Pump
12% Bladder Pump 1.6% Air Lift Pump
0.2% Centrifugal Pump 0.9% Peristaltic Pump
9% Submersible Pump 1.6% Suction Lift Pump
0.9% Others (list brand name of pump if unsure of type)
4. About how many well volumes do you remove from the well when purging wells which you cannot purge
dry?
3.7 number of well volumes purged
5. About how much time elapses between the time you finish purging a well which recharges rapidly and the
time that you sample it?
(Circle One)
None, we have no such well 8.7%
less than 30 minutes 60%
30 to 60 minutes 6%
1 to 2 hours 3.3%
2 to 4 hour 1.1%
4 to 6 hours 0.0%
47
-------
6 to 12 hours 1.1%
12 to 24 hours 11%
25 to 48 hours 4.3%
49 to 72 hours 0.0%
6. At about how many landfills do you use the following equipment to retrieve the samples from the well?
40% Bailer 6% Submersible Pump
49% Bailer with Bottom .0% Gas Displacement Pump
Emptying Device
11% Bladder Pump 0.7% Peristaltic Pump
1.6% Centrifugal Pump 0.9% Suction Lift Pump
0.9% Others (List brand name of pump if unsure of type)
7. About what percent of all monitoring wells that you sample produce turbid water (i.e., water is not clear)?
18% <25% 21% 25-50% 27% 51-75% 34% >75%
8. Do you filter samples for inorganics? 84% Yes 14% No
(e.g., alkalinity, hardness)
If Yes, is filtering done in 66% Field
18% Lab
9. Do you filter samples for metals? 88% Yes 12% No
If Yes, is filtering done in 77% Field
77% Lab
10. If field filtering, do you use: 68% A transfer container
25% An in line filter system
11. About how long is it, on the average, between the time you take a sample and the time that it is filtered?
(Circle One)
0 minutes (in line filtering) 18%
Less than 15 minutes 45%
15 to 60 minutes 16%
1 to 2 hours 6.5%
48
-------
2 to 3 hours 3.3%
More than 3 hours 11%
12. About how many landfills that you sample have dedicated sampling equipment?
16% Landfills with dedicated bailers (separate bailers for each well)
7.6% Landfills with dedicated pump system (separate pump for each well)
13. Do you use distilled water (also includes deionized and reagent grade water) to rinse equipment between
wells?
95% Yes 5%. No
If Yes, where is it usually obtained?
22% Grocery store
59% Laboratory
14% Other (please specify)_/wo5/ often: Culligan, or office purification equipment
If No, what do you use?
"River water " or "Just dry it off" were notable responses
Used to rinse equipment
The figures that follow graphically illustrate the answers to several of the sampling questions asked in the survey.
49
-------
How long is it between the time you
take a sample, and when it is filtered?
Omln
<1imm
1MO mtn
1-thr
RESPONSE
»-Jhf
Do you filter for inorganics
and/or metals? If so, where?
100-
•0-
*o-
70-
Ul
D. *>-
ao-
otlll»rtnsforlr>o
-------
About what percent of monitoring wells
that you sample produce turbid water?
UJ
o
DC
-------
How long is it between the time you
take a sample, and when it is filtered?
LU
o
DC
UJ
O.
emhi
1*-*0m*
1-thr
RESPONSE
»-Jhr
Do you filter for inorganics
and/or metals? If so, where?
10SV
•o-l
Ul
O
UJ
D.
ao-'
10-'
FIcMtlltor tor Inorganic* Nofln*rlng«orlr»rB«nlo» UbflltMfornwti'l*
Ub filter tor Inoreinloi FMdlMtrfor nwtiK NofUtoringtoriMtolt
RESPONSE
52
-------
How long is it between the time you
purge a well, and when it is sampled?
T0~
5 «.
LU
o
tr
111 10-
o.
1C-
n
11 n - - 1 n
no rapid rtdwrg* *»40mln *-4hr» '*-12hra
<»0mtn 1-tlm 4-»hf» 1»44hra M-TShn
RESPONSE
-------
STATISTICAL COMPARISON OF
METAL CONCENTRATIONS IN
FILTERED AND UNFILTERED
GROUND-WATER SAMPLES
the sample would be roughly divided in thirds. The results
presented are not dependent on this classification system,
and can be replicated with the continuous turbidity
measurements as well.
Results
Robert D. Gibbons & Martin N. Sara
Introduction
The promulgation of Subtitle D on October 9, 1991 with it's
restriction on filtering ground-water samples has generated
much controversy. From USEPA's perspective, there is
concern that by filtering the sample potential contribution of
colloidal transport to off-site migration may be overlooked
(i.e., molecules of metals bound to paniculate matter in the
ground water). From the states and regulated community
perspective, there is concern over what is being measured;
ground water or the formation adjacent to the monitoring
well. Increases in turbidity may lead to higher unfiltered
metal concentrations, and since waste disposal facilities
invariably have more downgradient than upgradient
monitoring wells, the probability of a turbid sample is more
likely in a downgradient well than an upgradient well by
chance alone. This type of "false positive" result will lead to
expensive and unneeded site assessment. Alternatively, if
higher turbidity or differences in formation are associated
with upgradient well measurements, effects of
contamination may go undetected (i.e., false negative
results).
Methodology
Paired measurements (i.e., filtered and unfiltered) were
obtained for 16 metals [calcium (Ca), magnesium (Mg),
sodium (Na), arsenic (As), cadmium (Cd), mercury (Hg),
selenium (Se), silver (Ag), chromium (Cr), copper (Cu),
lead (Pb), barium (Ba), manganese (Mn), potassium (K),
zinc (Zn), and iron (Fe)] in mg/1, and sample turbidity in
NTU, from each of 4 waste disposal companies. Each
company had paired data for upgradient and downgradient
monitoring wells from several different facilities. This
report is based solely on complete samples in which filtered
and unfiltered metal concentrations, and turbidity were
measured in the ground-water sample.
There were a total of 9689 complete records from 12 sites
and 155 monitoring wells. Sample sizes ranged from 506
complete measurements for Ba to 803 complete
measurements for Ca. For the purpose of illustration,
turbidity was divided into low (< 10 NTU), medium (11-50
NTU), and high (> 50 NTU). These cutpoints were based on
the empirical frequency distribution of turbidity such that
This section presents results from the comparison of filtered
and unfiltered samples using the entire database of complete
sample (i. e., those in which turbidity and both filtered and
unfiltered measurements were simultaneously available).
Summary statistics are presented in Tables 1 and 2. Table 1
displays detection frequency for filtered and unfiltered
samples by compound and sample turbidity level (~.e., low,
medium, and high). Table 2 displays the mean concentration
and standard deviation for those cases in which a metal was
detected in both filtered and unfiltered samples. Also, Table
2 lists results for each metal and turbidity level.
Table 3, displays summary statistics for Fe overall, and for
each turbidity level in 45 upgradient measurements. Table 3
reveals that the average upgradient unfiltered concentration
(see Total in Table 3) is five times higher than filtered
samples and the standard deviation i.e., variability) is six
times higher. Although variability is always higher in
unfiltered samples, the magnitude of the difference is
proportional to turbidity of the sample.
In addition to the effect of the increased variability of the
unfiltered samples on the false negative rate of the statistical
test, the relationship with turbidity can produce large
numbers of false positive results as well. Table 4 displays
the average and maximum iron concentrations for each of
the 11 downgradient wells at the facility for filtered and
unfiltered samples. Table 4 reveals widespread variability in
iron concentrations for the unfiltered samples across
downgradient wells. Some wells show consistently high (I
and 2) or consistently low (3, 4, 5, and 11) iron
concentrations whereas other wells show widespread
variation for unfiltered samples i.e., ratio of maximum to
mean values of 10 to 1). Results for the filtered samples
were far more consistent (ratio of maximum to mean of less
than 2 to 1).
The variance component estimates i.e., inter-site, inter-well,
and intra-well) for filtered, unfiltered and their difference
i.e., unfiltered-filtered) are displayed in Table 5. Table 5
reveals that variability from site to site is reasonably
consistent for filtered and unfiltered samples. Variability
from well to well is approximately twice as large for
unfiltered versus filtered samples. Variability within wells is
over five times as large for unfiltered versus filtered
samples.
These results suggest that if unfiltered samples are the basis
for these analyses, false negative rates i.e., failure to detect
54
-------
contamination when it is present) will, in fact, be
considerably larger for than if filtered samples are used.
Conversely, the doubling of the well to well component of
variation for unfiltered samples relative to filtered samples,
will produce large numbers of false positive results as well,
particularly in upgradient versus downgradient detection
monitoring comparisons. Turbid samples will continuously
be mistaken for contaminated samples and needless site
assessments will be routinely performed. On the other hand,
an unscrupulous owner/operator might simply churn the
upgradient well samples i.e., inaease turbidity), and
meticulously collect downgradient samples, which would
greatly increase statistical limits and rarely result in
statistically significant differences whether contamination
was present or not. Such practice would have little or no
effect on filtered samples.
Summary
The data presented show how unfiltered measurements can
lead to increases in both false positive and false negative
rates due to increased variability of the unfiltered results
(i.e., false negative) and the relationship between
concentration and sample turbidity that can be mistakenly
interpreted as contamination (ie., false positive). The data
presented indicate that turbid samples will continuously be
mistaken for contaminated samples and needless technical
efforts will be expended by both the states' and facility
owners' staff. Conversely, by not filtering, upgradient
samples collected in turbid wells could mask releases in
downgradient wells where unturbid samples were carefully
collected.
The purpose of this report is to clearly demonstrate the
enormous price that is paid in both potential false positive
and false negative detection monitoring decisions when
unfiltered samples are used as the basis of testing statistical
hypotheses regarding site impact. It is clearly shown that
unfiltered sample concentrations exhibit extreme variability,
which is in large part due to sample turbidity and colloidal
transport. Furthermore, even if a statistical adjust for the
effects of turbidity was performed (i.e., in effect hold
turbidity constant), differences between unfiltered and
filtered metal concentrations still exist. Are these differences
due to colloidal transport or the effects of the formation? No
definitive conclusion can be readily drawn since these
effects are confounded in empirical data. It should be noted,
however, that there are special geological conditions that are
required for colloidal transport, but the increased variability
in unfiltered samples is seen consistently in all sites
examined regardless of whether conditions are conducive
for colloidal transport or not. In addition, the effects are
seen both in upgradient and downgradient wells indicating
that in these data, the increased concentrations associated
with unfiltered samples are not due to colloidal transport of
metals released from a facility.
From a statistical perspective, the question is whether
filtered or unfiltered samples lead to a preferential balance
between false positive and false negative results when used
in ground-water detection monitoring programs. To select
unfiltered samples to cover the possibility of colloidal
transport, at the expense of both greatly inaeased false
positive and false negative results is foolish at best. Analysis
of extensive monitoring data, from numerous waste disposal
facilities using upgradient and downgradient monitoring
wells in which both filtered and unfiltered samples were
simultaneously collected, clearly showed that filtered
samples dramatically minimize both false positive and false
negative rates relative to unfiltered samples. The principal
factor in this difference was due to the turbidity of the
sample.
The results of this study have profound implications for
public policy in relation to the monitoring of municipal solid
waste landfills. Use of unfiltered samples will dramatically
increase false negative results in those sites in which
upgradient wells exhibit turbidity or are drilled in different
formations that vary in terms of their concentration of tht
metal in question. Both conditions will lead to large
variability in upgradient samples and produce statistical
limit estimates that are quite large relative to what could be
obtained had filtered samples been used. In contrast, when
the small number of upgradient wells are not representative
of either the formations or turbidity in the far greater
number of downgradient wells, the variability in the
upgradient wells will underestimate the true background
variability for the site as a whole and large numbers of false
positive results will occur. The reason, of course, is that the
unfiltered samples are influenced heavily by sample
turbidity and geological formation, neither of which is
explicitly controlled in the detection monitoring process. To
the extent that there are far greater numbers of downgradient
wells, there is a correspondingly greater chance of having
turbidity or differences in formation adversely impacting the
concentration in a, downgradient well and falsely
concluding that it is the site that has impacted ground water.
Conversely, if the upgradient samples are highly turbid or
vary in turbidity, variability in the measured concentrations
will be large and statistical tests of contamination will be
powerless to detect real contamination when it occurs. All
analyses performed in this study yield exactly the same
conclusion.
The reader should note that the effects observed here were
consistent for both upgradient and downgradient wells. If
colloidal transport is the method by which contaminants are
transported off-site, why should the same types of
variability be observed in upgradient wells? These results
clearly show that whether colloidal transport is real or not,
the price paid for using unfiltered samples is enormous, both
in terms of missing real contamination when it exists and in
detecting contamination when it is not present.
55
-------
Table 1 Detection Frequencies for Filtered and Unfiltered Samples (Expressed as a Proportion of Total Measurements)
Metal
Ag
As
Ba
Ca
Cd
Cr
Cu
Fe
Hg
K
Mg
Mn
Na
Pb
Se
Zn
Low Turbidity
Filtered Unfiltered
.04
.08
.23
.99
.10
.04
.22
.58
.01
.85
.99
.75
.92
.15
.05
.59
.07
.14
.31
1.0
.14
.09
.29
.87
.01
.87
1.0
.81
.96
.-23
.04
.68
Med Turbidity
Filtered Unfiltered
.01
.13
.36
1.0
.07
.07
.12
.64
.01
.94
.99
.84
.98
.07
.04
.61
.06
.22
.48
1.0
.18
.15
.29
.98
.02
.95
1.0
.94
.98
.22
.03
.78
High
Filtered
.02
.32
.60
.99
.06
.11
.28
.82
.02
.82
.98
.88
1.0
.24
.14
.62
Turbidity
Unfiltered
.12
.48
.70
.99
.16
.55
.57
.99
.05
.83
1.0
.99
1.0
.52
.08
.76
56
-------
Table 2 Mean and Standard Deviation (SD) of Metal Concentrations for Filtered and Unfiltered Samples When Both
were Detected
Metal Low Turbidity
Filtered Unfiltered
Mean SD Mean SD
Ag
As
Ba
Ca
Cd
Cr
Cu
Fe
K
Mg
Mn
Na
Pb
Se
Zn
.005
.010
.146
93.4
.005
.021
.091
6.68
11.9
121.
6.96
286.
.091
.016
.350
.001
.018
.174
139.
.003
.016
.134
14.6
27.0
432.
13.5
1001
.034
.030
.607
.004
.018
.253
96.1
.007
.021
.111
20.0
13.2
114.
7.26
297.
.073
.016
.385
.001
.024
.649
120.
.007
.013
.148
61.4*
31.7*
393.
13.5*
1055
.278
.031
.638*
Med Turbidity
Filtered Unfiltered
Mean SD Mean SD
.004
.009
.139
102.
.009
.023
.143
10.5
9.87
184.
4.60
360.
.022
.314
.194
.000
.008
.112
126.
.006
.031
.250
23.5
19.5
654.
8.11
1178
.030
.483
.385
.023
.014
.180
108.
.008
.032
.152
20.1
9.71
171.
4.76
369.
.033
.347
.217
.000
.016
.153*
138.
.006
.032
.242
48.8*
19.4
580.
8.00*
1216
.033*
.549
.381 *
High Turbidity
Filtered Unfiltered
Mean SD Mean SD
.019
.028
.152
356.
.011
.012
.070
21.3
20.5
163.
5.99
314.
.158
.030
.350
.043
.055
.144
1152
.007
.013
.173
34.1
36.7
609.
11.8
1163
.124
.036
1.18
.042 0.47
.032 0.59
.461 .995
465. 1341
.011 .006
.051 .053
.146 .239
50.9 104*
24.2 38.6*
179. 640.*
6.83 12.4*
324. 1174
.182 .134
.041 .059
.602 1.26*
* indicates statistically significant difference between filtered and unfiltered samples based on a paired t-statistic
Table 3 Summary Statistics for Iron (mg/1) in Filtered and Unfiltered Samples in
Upgradient Wells (1/2 MDL substituted for Nondetects)
Turbidity
Total
Low
Medium
High
Mean
.439
.152
.472
1.569
Filtered
SD
.760
.305
.348
1.145
N
45
32
5
8
Mean
2.044
.369
1.970
8.790
Unfiltered
SD
4.530
1.014
1.066
7.682
N
45
32
5
8
57
-------
Table 4 Average and Maximum Iron Concentrations (mgA) in Downgradient Wells (1/2 MDL substituted for
Nondetects)
Well
Filtered
Mean Max
Unfiltered
Mean Max
Turbidity
Mean Max
I
2
3
4
5
6
7
S
9
10
11
60.05
62.25
.SO
.06
.06
.11
.39
.06
.06
5.90
.43
87.30
106.00
4.90
.08
.08
.94
4.80
.08
.08
7.89
.92
60.02
56.78
.90
.41
.85
2.65
5.27
8.62
13.55
12.58
.49
89.70
104.00
5.60
1.16
1.98
29.50
53.50
53.60
153.00
24.40
1.00
123
392
9
5
21
16
198
235
215
741
2
220
705
49
12
85
55
21 90
1650
2550
3470
8
Table 5 GLS Variance Component Estimates for Unfiltered, Filtered, and the Difference Between Filtered and
Unfiltered Results For Fe in mg/1
Effect
Unfiltered
Filtered
Difference
Inter-Site SD
Inter-Well SD
Intra-Well SD
17.65
27.43
66.72
14.30
13.30
12.30
11.09
20.59
64.65
Variance components listed as standard deviations (SD)
58
-------
IV. Poster Abstracts
PRESENTERS
John D. Gray, New York State Electric and Gas Company
Robert M. Powell, ManTech Environmental Technology, Inc.
Cynthia Paul, U.S. EPA/Robert S. Kerr Environmental Research Laboratory
Nic. E. Korte, Oak Ridge National Laboratory
Natalie Park, Westinghouse Savannah River Company
Daniel Ronen, Weizmann Institute
Gary Robbins, University of Connecticut (no abstract)
59
-------
IN-LINE FILTERING OF GROUND WATER SAMPLES FROM MONITORING WELLS IN
GLACIAL TILL
John D. Gray
Richard E. Wardwell
ABSTRACT
New York State Electric & Gas (NYSEG) conducted an extensive research project at a solid waste (fly ash) landfill to
compare and evaluate the analytical results from filtered and unfiltered ground water samples. Two years of site-specific data
for dissolved (filtered) and total (unfiltered) concentrations of 22 metals were collected. Elemental analyses of the soil
constituents of the glacial till from the site were performed. In addition, field testing of alternative sampling techniques to
reduce turbidity in ground water samples was conducted.
The following results were determined from the research project:
• It is theoretically impossible to design a monitoring well filter pack which will prevent the movement of fines (silt
and clay) into a well bore during sampling.
• The high percentage of fines in the glacial till and fractured bedrock at the site makes it is impossible to assure a
turbid free sample from the site monitoring wells regardless of the well installation or sampling technique.
• The chemical composition of the native soil contains significant amounts of aluminum, chromium, copper, iron,
manganese and zinc.
• Unfiltered samples show increases in the fore mentioned metals in direct relation to the amount of turbidity in the
ground water sample.
• Filtering removes suspended soil particles but does not change the chemistry of the ground water samples collected
at the site.
The results from this study demonstrate that the use of unfiltered samples will produce elevated concentrations for those
metals which comprise the soil matrix. These elevated concentrations will trigger invalid exceedances of New York State
ground water standards and will mask any water quality trends associated with landfill activities. Therefore, filtered
(dissolved metals) concentrations are necessary for assessing the impacts of the landfill because they are more representative
of constituents actually moving in the ground water.
60
-------
Passive Sampling of Ground Water Monitoring Wells without Purging: Multilevel Well
Chemistry and Tracer Disappearance
Robert M. Powell and Robert W. Puls
ABSTRACT
It is essential that the sampling techniques utilized in ground-water monitoring provide data that accurately depicts the water
quality of the sampled aquifer in the vicinity of the well. Due to the large amount of monitoring activity currently underway
in the U.S. it is also important that the techniques be efficient. It would be desirable to minimize the requirements of
sampling time, equipment, and quantity of contaminated waters pumped to the surface, without loss of data integrity. If
representative samples could be acquired without purging the wells, increased sampling efficiency could potentially be
achieved.
Purging of multiple borehole volumes is largely routine, based on studies that show changes in the water chemistry as it
stands in the casing and is subjected to atmospheric exposure at the top of the column. However, little data is available
depicting water chemistry in the screened intervals of wells at equilibrium flow conditions, i.e. with little or no disturbance to
the natural flow regime or disruption of the overlying casing waters.
This study examines the differences in water chemistry between the casing and screened interval volumes of four wells at a
field site, then compares the results to purged values for the same wells. Tracer experiments, utilizing both colloidal particles
and dissolved species as tracers, are presented to illustrate differences in natural flushing between the screened and cased
intervals. The data from the tracer removal was then utilized to estimate ground water flow velocities in the vicinities of the
boreholes.
61
-------
Turbidity Effects on Volatile Organic Contaminants in Ground Water Samples
Cynthia J. Paul, Robert W. Puls and Lisa R. Secrest
ABSTRACT
According to the RCRA Ground Water Monitoring Technical Enforcement Document (TEGD), ground water samples
collected for contaminant analysis should have a turbidity of less than five nephelometric turbidity units (NTU). Some feel
turbidity levels greater than five NTUs would not be representative of the aquifer; however, many monitoring wells produce
samples with turbidity levels greater than five NTUs due to site-specific hydrogeochemistry or improper well design or
construction. However, our previous work has shown that proper sampling techniques can produce samples with low
turbidity values even in wells traditionally considered turbid. Little research has been conducted to determine the impact of
turbidity on volatile organic compounds (VOCs) in recovered ground water samples. This study was designed to
differentiate turbidity effects from unknown field effects that could influence the apparent concentrations of VOCs in
ground water samples. Laboratory batch studies and field investigations were performed to evaluate the effects of solids
(turbidity) on VOC concentrations. Three different solids (recovered aquifer material, kaolinite, and Na-montmorillonite)
were used in the turbidity "spikes" to assess turbidity effects on sample quality. During the laboratory portion of this study,
these solids were used to determine sorption and volume displacement effects on VOC concentrations under controlled
laboratory conditions. The procedure included adding known amounts of solid to a simulated ground water solution and
then spiking each sample with varying amounts of TCE. The same solids were used in the field portion of the study to
differentiate turbidity effects on sampling methodology. Sample VOA vials were prespiked with known amounts of solids.
VOC samples were collected into these vials as well as vials containing no solids for comparison purposes. The field study
also included alteration of the sampling procedure to intentionally increase turbidity levels by entrainment of natural aquifer
materials in the collected samples. Water quality indicators (pH, dissolved oxygen, turbidity, specific conductance, redox,
and temperature) were monitored during well purging and samples were collected after all parameters reached equilibration.
Results of this study indicate that increased turbidity levels in ground water samples have no impact on VOC concentrations
(i.e., TCE and its degradation products). Sampling methodology appears to be the most important consideration when
collecting ground water samples for VOCs.
62
-------
A Field Comparison of Micropurging and Traditional Groundwater Sampling Using Analytical
Data and Observations with the Colloidal Borescope
Nic E. Korte and Peter M. Kearl
ABSTRACT
The recent literature discusses micropurge sampling techniques as a possible replacement for traditional purge and sample
methods. Micropurge sampling involves a dedicated sampling pump, a sampling rate of approximately 100 mL/min, and
purging of only the sample pump and tubing. If micropurging can yield reliable groundwater samples, then significant
quantities of purge water can be eliminated, thereby reducing costs and minimizing waste. Unfortunately, evaluating
sampling methods for yielding representative water samples of natural groundwater systems is difficult. The monitoring well
represents an unnatural intrusion into the subsurface, and natural variations in the chemistry of the groundwater further
complicate the system. One approach to overcoming these problems is to duplicate groundwater sampling using both
conventional and repetitive micropurge sampling followed by a statistical comparison of the results. Using this approach, a
series of experiments is being conducted at two sites where samples are analyzed for selected organic and inorganic
constituents. Analysis of data collated to date using a paired t-test indicates that within a 95% confidence interval, there was
no significant difference between the sampling methods for both site contaminants and the majority of naturally occurring
analytes. Analytes that showed a significant difference were redox sensitive (e.g., iron and manganese) or were present in
such low concentrations that analytical reliability was a factor. The analytical results were supported by observations
performed with the colloidal borescope. Borescope observations demonstrated significant impacts on the hydrodynamic flow
system when using traditional sampling methods. Results of the study suggest replacing traditional purging and sampling
with micropurging, due to the reliability and cost-effectiveness of the method.
63
-------
Random Variability of Metals Measurements in Ground Water Reduced through Revised
Sampling Techniques
Natalie M. Park
ABSTRACT
The potential to reduce the random variability or imprecision in metals measurements in groundwater samples by as much
as 90% was demonstrated through the application of revised groundwater sampling techniques. Revised ground water
sampling procedures incorporated recent research recommendations to cease filtering samples collected for metals analyses
and to move away from the conventional approach of "purge a minimum of 3 to 5 well volumes until pH, specific
conductance and temperature are stable." Since historical data indicted that pH and conductivity usually reached stability
within one to two well volumes, the new approach requires purging a minimum of two-well volumes (as a conservative
measure) and uses turbidity instead of temperature as a stabilization parameter. To reduce time, the first two well volumes
are purged at a rate that can be sustained without creating surging, generally 2-5 gal/min. Then the purging rate is reduced
to 0.25 gal/min and purging continued until pH, conductivity, and turbidity measurements stabilize. Samples collected for
metals analyses are not filtered, whereas previously 0.45 micron filters were used.
Dedicated, variable speed pumps replaced faster pumping, single speed submersible pumps in 40 wells. These wells
monitor a contaminant plume characterized by low pH (3-5), high nitrate (as N) and sodium concentrations, elevated
specific conductance (51 to 2200 |j,S/cm), and metals including aluminum, iron, manganese, barium zinc, copper, cobalt,
cadmium and zinc. Pumps are routinely set toward the bottom of the PVC screens to accommodate fluctuations in water
levels. The wells were installed using mud rotary drilling in Coastal Plain interbedded sands, silts and clays at depths of 20
to 70 feet (saturated screened intervals varied from 5 to 15 feet).
After a year of quarterly sampling using the revised method, comparisons with historical data revealed that for some wells
there was an average 75% to 90% reduction in random variability for measurements of aluminum, zinc, manganese,
barium, iron, cobalt, cadmium and copper. This average reduction in variability was observed for manganese data from
32% of all wells; for iron, zinc and barium data from 25% of all wells; and for 35% of the wells in which cobalt, copper,
and cadmium were present. On the average, the reduction in random variability was accompanied by a 50% decrease in
metal concentration. Turbidity measurements stabilized below 5 NTU in 38 of the wells sampled; stable values ranged
from 0.2 to 50 NTUs and were generally achieved within 15 minutes to 1.5 hours. Since the revised method did not use
filtering, the decrease in metals concentrations and variability can be attributed to a combination of the slower pumping rate
and the addition of turbidity as a stabilization parameters.
64
-------
A Multi-Layer Sampler for the Study and Monitoring of Chemical Processes and Transport
Phenomena in Aquifers
Daniel Ronen
ABSTRACT
A multi-layer sampler (MLS) was developed and utilized for: a) sampling detailed undisturbed groundwater chemical
profiles; b) sampling gases in both the saturated and the unsaturated zone; c) deriving detailed vertical profiles of the
horizontal component of the specific discharge; d) characterizing suspended particles under natural gradient flow
conditions. Sampling is based on the dialysis-cell method; the sampling volume is defined by the desired sampling interval.
The MLS is portable, cheap and easy to operate.
The results of the field study conducted in Israel revealed: 1) intensive biochemical activity as reflected by the consumption
of dissolved O2 with the concomitant oxidation of organic matter and the development of an anoxic layer, and the
production of N2O (up to 400 |j,g/l) and CO2 (log PCO2 from -1.7 to -1.3); 2) the presence of an almost stagnant water layer
(q = 0.5 m/y) down to a depth of 60 cm below the water table; 3) the presence of microscale isothermal water parcels
(characteristic vertical and horizontal length dimensions on the order of less than 1 m) which differ from each other in their
chemical composition and density and are characterized by very sharp boundaries between them; 4) microscale Eulerian
variations in the flux, mineralogical composition and size of suspended particles under natural gradient flow conditions (q
= 11 to 16 m/y). It is postulated that the gases produced during the biodegradation of the organic matter accumulate as a
distinct gas phase (bubbles) down to a depth of 1 m below the water table, reducing groundwater flow. The replenishment
of the aquifer by water of different chemical composition and the almost stagnant conditions prevailing at the water table
region (where mechanical dispersion by advection is negligible) lead to the development of microscale parcels of water of
different chemical composition. It is suggested that haline convection is a major transport and mixing mechanism at the
water table region. It is also postulated that due to the dramatic increase in pCO2 part of the carbonate cement of the rocks
dissolve and detritial CaC03, quartz and clay are released as colloidal particles. In the prevailing anoxic conditions of
groundwater at the study site (DO <1 mg/1) colloidal stability is enhanced by organic matter coating of particles.
65
-------
66
-------
V. Small Group Discussions Summaries
Prior to arriving at the workshop, participants were sent a package which included reference materials, brief abstracts for
presentations on the first day, small group discussion topics and focus questions for the small group discussion topics. In
addition, the participants were queried as to their preference for small group discussion assignment. These assignments were
made based on participants' responses, and each participant was assigned to two topical groups. The focus questions are
provided as Appendix B of this document. These questions were formulated by the Steering Committee in advance of the
workshop and as a result of formal meetings, phone conversations, and other communications. The small groups were
directed to attempt to answer as many questions as possible during their respective discussions, but were also granted the
latitude to let the group dynamics determine discussion direction as appropriate.
67
-------
V. Small Group Discussions
Participants
(by topic area)
1. MONITORING GOALS AND OBJECTIVES
Group Leaders: Barcelona, Brown
Participants: Bourbon, Franks, Gronwald, Lee, Mabey, McKay, Park,
Reece, Rightmire, Romero, Stelz, Teplitzky, Willey, Zavala
2. WELL DESIGN, CONSTRUCTION, AND DEVELOPMENT
Group Leaders: Aller, Gardner
Participants: Connelly, Franks, Gray, Gronwald, McKay, Nielsen, Parker,
Sara, Stelz, Taylor
3. WELL PURGING AND SAMPLING
Group Leaders: Puls, McCarthy
Participants: Backhus, Barcelona, Connelly, Gibs, Hall, Korte, Mangion,
Martin, Park, Parker, Paul, Pohlmann, Powell, Robbins, Ronen,
White, Zavala
4. TURBIDITY AND COLLOID TRANSPORT
Group Leaders: Ryan, Mangion, Willey
Participants: Backhus, Lee, Martin, McCarthy, Puls, Reece, Rightmire,
Romero, Ronen, Sridharan, Taylor, Teplitzky, White
5. SAMPLE HANDLING AND ANALYSIS
Group Leaders: Pohlmann, Rosal, Beldsoe
Participants: Bourbon, Brown, Clark, Gibs, Gray, Hall, Korte, Mabey,
Nielsen, Paul, Powell, Robbins, Sara, Sridharan
68
-------
Monitoring Goals and
Objectives
Michael J.Barcelona and James R. Brown
Introduction
This group considered a number of questions dealing with
the goals and objectives of monitoring efforts. From the
outset it was recognized that there are a wide variety of
monitoring purposes and programmatic goals-each with
their own objectives. The group acknowledged that main
objectives and the uses of data (i.e., both chemical and
hydrogeologic) collected in monitoring programs change
with time as the complexity and detail of subsurface
conditions become apparent. The participants also noted
that much has been learned about techniques which lead to
more accurate and reproducible data collection. It was
agreed that adoption of improved standards for professional
practice in monitoring network design, construction, and
operation should be encouraged; and that the use of error-
prone monitoring techniques should be discouraged
accordingly. In this regard the newest revision of Agency
Guidance ("RCRA Ground-Water Monitoring: Draft
Technical Guidance," EPA/530-R-93-001) goes a long way
towards highlighting the differences between currently
recommended practices and traditionally used practices.
Objectives of Monitoring Programs
Monitoring objectives include four main types (i.e.
detection, assessment, corrective-action evaluation and
resource evaluation), along with "hybrid" variations such as
site-assessments for property transfers and water availability
investigations. Monitoring purposes and objectives may
change as contamination or water quality problems are
discovered. However, there are a number of common
objectives for monitoring which should be recognized as
important regardless of initial purpose. These objectives
include: 1) Development of a conceptual model that
incorporates elements of the regional geology to the local
geologic framework. The conceptual model development
also includes initial site characterization efforts to identify
hydrostratigraphic units and likely flow-paths using a
minimum number of borings and well completions; 2) Cost-
effective and documented collection of high quality data
utilizing simple, accurate, and reproducible techniques; and,
3) Refinement of the conceptual model based on
supplementary data collection and analysis efforts which
evolve in complexity and the level of spatial detail. These
fundamental objectives would serve many types of
monitoring programs and provide a basis for future efforts
as purposes and objectives expand.
The Extent that Program Objectives, Site
Characteristics, or Constituents of Concern Provide
Criteria for Representativeness
The group noted from the outset that this topic applies to the
representativeness of chemical and hydrogeologic data
collected via wells, borings, piezometers, geophysical and
soil gas measurements, lysimeters, and temporary sampling
points. There is a need to get more quantitative about our
definition of what representativeness entails in the context
of controlled evolutionary site characterization and
monitoring efforts. Representativeness arises from a
recognition of the statistical variability of individual
subsurface physical properties, and contaminant or major
ion concentration levels with a need to explain extreme
values. Subsurface variability is a fact and good
professional practice should seek to maximize
representativeness by using proven accurate and
reproducible techniques to define limits on the distribution
of measurements collected at a site. An investigative site
characterization model was proposed to more systematically
approach the goal of consistent data collection.
The model emphasizes a recognition of the causes of the
variability (e.g., use of inappropriate technology such as
using bailers to purge wells; imprecise or operator
dependent methods) and the need to control avoidable
errors.
Flexible Sampling Designs for Meeting Multiple
Monitoring Objectives
Detailed site characterization is central to all purposes and
the basis for site characterization resides in the geologic
framework and identification of major hydro-stratigraphic
units. Fundamental data on subsurface lithology, head-
differences and background geochemical conditions for
example, should not change much (except to be refined
appropriately over time). Each sampling point has a proper
use or uses which should be documented at a level which is
appropriate for the program's data quality objectives. While
these sampling points cannot always fulfill multiple
monitoring objectives (e.g., detection, assessment,
69
-------
Program Objectives
*
Data Quality Objectives
Evolutionary Site
Characterization
Sampling and
Analytical Protocols
pplication
I ^- Refined Protocols -^- — — — -^- Decision-Making
corrective action), the data that they yield will always
contribute to the fundamental information needed for any
monitoring program.
Elements of a Sampling Protocol: Accuracy, Precision
and Sensitivity Needed to Meet Monitoring Objectives
A sampling protocol is a documented set of procedures and
steps to accomplish a defined task. Since data quality
objectives (DQO's) entail the identification of necessary
levels of accuracy, precision, sensitivity and completeness
for specific analytes (or subsurface properties), the DQO's
should drive the complexity of the sampling protocols. It
was recognized that data validation methods which focus
solely on the analytical process do not assure that data will
be of sufficient quality to meet DQO's. The major sources
of controllable error occur in the field (e.g., wrong locations,
poor technology choices, operator errors and incomplete
documentation), and it is vitally important to discourage the
continued use of methods proven to be inadequate to meet
DQO's. Suggestions were made to expand educational
efforts on technology choices, and the potential usefulness
of QA summary reports which apply to field as well as
laboratory data collection efforts.
Application of Field-Screening Techniques and Related
Performance Criteria
The value of field-screening techniques to aid in
evolutionary site characterization and assessment efforts has
been documented in the literature. Since detection efforts
involve the evaluation of potential pathways for
contaminant movement, and assessment efforts focus on the
extent of movement through such pathways, their
monitoring objectives and DQO's are comparable. The
levels of accuracy, precision and sensitivity for many
constituents of concern approachable by field techniques are
sometimes equal to laboratory-based analytical techniques.
In some cases (e.g., volatile organics) the use of field
techniques may be favored over lab analyses because
sample handling, storage, preservation steps are minimized
or eliminated prior to sample analysis. Laboratory methods
have the advantages of higher sensitivity and specificity
(e.g., mass-spectrometry). Their use should be encouraged
for confirmation purposes.
In all cases, the performance criteria for all screening
methods (including chemical analyses and hydrogeologic
analyses), should be determined by the data quality
objectives of the monitoring program. After the DQO's are
established, field screening procedures/tools can be selected
on the basis of their performance capabilities.
Criteria for the use of field screening methods should
include: 1) documentation of daily calibration and method
control with external standards/audits; 2) laboratory
confirmation of a percentage of samples; and, 3)
development of a field methods manual analogous to
laboratory CLP manuals under Agency auspices.
It is essential that continued training and overall quality
improvement should be undertaken to maintain/elevate field
practice to the state-of-the-art level. This should include
training courses, technical information transfer of important
scientific publications, and some form of certification and
auditing. The use of "ASTM-like" standards and quality
control summary reports would result in quality
improvement of hydrogeologic field work.
70
-------
Factors (e.g., Sampling Techniques, Extraction Rates,
and Location) that Influence Monitoring Objectives
The group consensus was that program objectives influence
the approach and complexity of the sampling and analytical
protocols, and not vice-versa. Program objectives are
usually motivated by regulation, although considerations of
experimental design establish monitoring objectives for
research projects. There will always be difficult situations
where the availability of water (e.g., low yielding
formations), accessibility of specific hydrostratigraphic
units, or unpredictable gradients may need to be factored
into the objectives; but the specifics of sampling protocols
should always be traceable to the DQO's.
Chemical or Physical Speciation of Contaminants
Influence on the Design of a Monitoring Program
Speciation includes the individual physical (i.e., sorbed,
colloidal, etc.) and chemical (i.e., mineral, complexed,
varying oxidation states) forms of an element or compound
which make up the total amount present in a subsurface
sample. The potential for multiple species with differences
in mobility, toxicity, and stability should necessarily expand
the scope of program objectives and influence the detail and
rigor of DQO's, sampling, and analytical protocols. In these
instances, renewed focus on the characterization of
background conditions and the total mass distribution of
contaminants becomes very important regardless of program
purpose. The proper selection of sample location and
sampling for the media in which the bulk of contaminant
mass may reside rather than symptoms (e.g. soil, gas or H2O
for sorbed contaminants) are of major concern in this
regard.
In addition, the hydrogeochemical conceptual model of the
site is as important as the site's geological conceptual
model. Ion balance and geochemical assessments should be
performed.
Alternative Methods for Designing Monitoring Networks
Given the fundamental value of a site geologic framework
consistent with regional hydrogeology and likely
contaminant behavior, alternative design methods include:
1) the diagnostic use of a flow and transport models to
select sampling locations where they can provide the most
information return; 2) Coupled field-screening/conceptual
model refinement and modeling work confirmed by further
site characterization and assessment efforts; 3) Integrated
use of geostatistical, stochastic fate and transport, and
optimization models applied in iterations to identify likely
limits on mass distributions and benchmarks for
performance as site-characterization proceeds. More
sophisticated methods such as decision-analysis/
optimization techniques should be made more accessible to
the practitioners via technology improvement and transfer
efforts.
Documentation of Well Construction and Well
Development (Extends to demand criteria, three-
dimensional site characterization, and sampling
protocols)
The group was unanimous in acknowledging the need for
improved standards for professional well construction and
development practices. This need demands improved detail
in documentation of monitoring efforts. Useful inclusions
are: geologic descriptions of core material, grain-size
determinations prior to well screen design and construction,
and the development of meaningful criteria for well
development, purging and sampling. Extending the rigor of
traditional laboratory-based QA/QC procedures to all
aspects of field activities would result in substantial
improvements in practice and the value of monitoring data
for decision-making.
71
-------
Well Design, Construction and
Development
Linda Aller and Steven Gardner
The purpose of a monitoring well is to create a point within
the desired monitored zone such that an "unaffected"
representative sample can be collected. While the ease or
difficulty with which this can be accomplished is dependent
in part upon the hydrogeologic setting, additional questions
arise in almost any situation. These questions extend to
include whether or not even an effective water-level
sampling point can be designed in fine-grained materials
and fracture-flow dominated formations. More widespread,
pervasive questions arise as to what effects the process of
drilling and installing the well have on the analytical
samples that are desired to be collected from the well. The
purpose of this section is to focus on the "artifacts" that are
introduced in the well installation process and on the ways
and means that are currently available to minimize these
effects.
Artifacts can be defined as unwanted residual effects. These
effects are introduced through both the drilling process and
the well installation process (often through the materials
added) and may be either enhanced or minimized by the
development process. Other sources of artifacts, such as
interconnection of formations, are also of concern. A
discussion of the potential artifacts and possible
minimization of those artifacts is contained in the following
paragraphs. Areas where additional research is needed are
also noted.
Drilling Artifacts
Drilling Techniques
Drilling techniques are the crudest link in the drilling,
installation and sampling scenario. The drilling methods
employed to install monitoring wells were adapted from the
water well, soil dynamics and oil industry. None of the
technologies provided by this historical base have been
adequate to provide the rigorous quality control needed for
monitoring wells. This is clearly evident when the
complexities of the chemical environment being monitored
are carefully considered. Typical drilling techniques used
for monitoring well installation have been described by
Aller et al. (1989), Sara (1994) and USEPA (1992). As
such, it is only recently that these techniques are beginning
to be modified to reflect monitoring needs. Smearing of
borehole walls, compaction of borehole walls, transport of
formation material and drilling fluid into different zones are
all areas where artifacts are of concern.
The smearing of borehole walls is a concern where
conventional augering techniques are employed or where
casing is driven and then pulled back. The smearing effect
seals off fractures in fine-grained formations, thereby
potentially reducing the amount of flow to the well or even
sealing off the formation of concern entirely. Because
fractures are typically the most prominent pathways for
contaminant migration in the fine-grained formations, the
question arises as to whether or not the contaminant
pathway has been blocked such that the contaminant will be
present in the formation but not be found in the well. One
way to minimize smearing of the borehole wall is to
minimize auger rotation in the screened zone. The concept
is that if the auger can be simply "screwed" in and then
retracted without turning, the smearing will be less.
Although this can be performed relatively easily in most
formations, the typical drilling procedure used is to rotate
the auger to move cuttings up the annulus, thereby
minimizing the amount of shear that the rig must overcome
when withdrawing the augers. There is much practical
experience in this area, but little published research that
better defines the drilling process and the formation
constraints. The working group identified a research need
to develop more effective ways to minimize this smearing
artifact.
Compaction of borehole walls and transport of formation
material is of concern in both augering techniques, casing
driving, and in rotary drilling. Compaction of the borehole
walls is similar to smearing of the walls as discussed above
with the concern being that pathways and therefore
contaminants are possibly "shut off." Transport of
formation material, along the borehole, downward into the
screened zone, is a concern where upper contaminant zones
are present. The amount of contaminated material necessary
to cause a detection with today's detection limits of parts per
quadrillion is minimal. Several techniques for installing
monitoring wells in areas where this is a concern have been
successfully used. Keely and Boetang (1987) have written
about driving casing while augering. Other multiple casing
72
-------
methods for installing surface casing through the
contaminated zone and then drilling deeper have also been
successful.
Drilling Fluids
There is substantial question created by the use of drilling
fluids during drilling with regard to artifact effects on the
monitoring well. Drilling fluids can be air, water or drilling
mud. Using air as a drilling fluid can cause chemical
changes in the formation due to oxidation. Lubricants in air
necessary for compressor operation can result in the
introduction of hydrocarbons through air entrainment into
the formation. Use of an in-line filter that has not reached
its useful life capacity can minimize this risk.
Adding water as a drilling fluid can: flush and dilute
indigenous water; change formation chemistry; precipitate
certain minerals, possibly seal off preferential pathways of
flow to the well, and substantially alter the soil and water
chemistry in the vicinity of the bedrock. Where water is
used, the amount of water added should always be measured
and the chemistry of the water should be known. Where
water is added, the question of how much water must be
removed in order to "get all the water back out" assumes
that no chemical changes have taken place and it is a matter
of just recovering the injected water. Although the amount
of water varies by formation and flow characteristics, the
rule of thumb for removal of three times the volume added
is frequently quoted. Using no water or minimizing water
use is the preferred alternative. One alternative is to pump
out and reuse formation water where the use of water is
needed. Some of the problems with the uses of drilling
fluids are discussed in Schalla (1986).
Other concerns related to drilling fluid include the use of
drilling mud and additives and the use of lubricants on
piping and joints. Drilling with mud involves the formation
of a filter cake on the borehole wall as an integral part of the
drilling process. Mud intrusion into the formation occurs as
part of the drilling process. The depth of intrusion is a
function of the geologic formations as imparted by other
factors such as wall cake thickness and character. There is
always concern that the mud filter cake (wall cake) will not
be completely removed during the development process.
The result will be that, once again, significant pathways of
contaminant migration are permanently sealed off
(particularly in fracture-controlled flow). Of concern also is
that the clay-sized particles in the mud will attenuate any
electrically charged contaminants that do migrate toward the
well. Because mud is difficult to remove during the
development process, the use of mud should be restricted.
The use of additives in mud to overcome drilling difficulties
is also inappropriate. This practice simply furthers the
potential to cause physical or chemical changes that will be
introduced and remain as a result of the particular additive.
Brobst and Brobka (1987) discuss the effects of drilling
fluid on sample chemistry.
The use of downhole hammer lubricants and petroleum-based
lubricants on rotary drill pipe connections are of concern in
monitoring well drilling. However, some type of lubricant is
necessary for the drilling machinery to operate without
damaging the equipment. Where lubricants are essential, there
are non-petroleum based synthetics that are better choices for
lubricants.
Installation Artifacts
Casing/Screen Artifacts
Installation artifacts are a result of casing and screen
materials and/or annular fill materials. Artifacts due to
casing and/or screen materials include: long-term
incompatibility of casing/screen materials with formation
water quality; short-term casing/screen impacts by sorption/
desorption and leaching; impacts by solvent cements; and
leakage through casing joints. The ability of the casing and
screen material to resist degradation by the water with
which it will come into contact is paramount from two
different standpoints. First, what are the immediate water
quality effects during sampling and second, what are the
water quality effects when the casing/screen materials
deteriorate and allow water to migrate along the borehole?
The corrosion of metal casing has long been a concern in
acidic environments. The swelling of polyvinyl chloride
(PVC) by solvents is also documented. These are real
concerns that should be addressed by choosing casing/
screen materials that are long-term compatible. When wells
are installed to meet short-term goals (ie. less expensive
materials that might not be longterm compatible), the well
may or may not stay a short-term well. Frequently, wells
are required to remain in place for a regulatory purpose. In
other situations, cost of abandonment may become a factor
as funding or priorities change and the well will remain.
Finally, personnel familiar with the concept of short-term
versus long-term well installation and any anticipated
incompatibility problems may be disengaged from the job.
With this loss, others may not recognize that a potential
problem exists and the well is left in place. Therefore,
prudent monitoring well installation does not employ
incompatible casing/screen materials.
Short-term effects on sample integrity due to: leaching of
analytes of interest or analytes that interfere with analyses;
sorption of analytes of interest; possible desorption of
analytes of interest should water quality improve; and
diffusion of organics through polymeric materials, are also
of concern in some environments. Significant discussions
73
-------
of the "real effects" of different materials including PVC
and PTFE are included in the following papers: Bianchi-
Mosquera and Mackay (1992), Cowgill (1988), Gillham and
O'Hannesin (1990), Hewitt (1989), Hewitt (1992), Jones
and Miller (1988), Parker (1991), Parker (1992), Parker et
al. (1990), Reynolds et al. (1990), Miller (1982), Reynolds
and Gillham (1985), Barcelona and Helfrich (1986),
Barcelona and Helfrich (1988), Barcelona et al. (1983),
Barcelona et al. (1985), Barcelona et al. (1988), Marsh and
Lloyd (1980), Barcelona (1984), Junk et al. (1974), Boettner
et al. (1981), Curran and Tomson (1983), Parker and Jenkins
(1986), Tomson et al. (1979), and Rivett et al. (1991).
The use of solvent cements in joining polymer casings is
inappropriate due to dissolution of the cements and
subsequent detection in samples from the wells. Sosebee et
al (1983) have documented these problems. The industry,
therefore, uses flush-joint threaded casing. However, these
joints are of concern when, or if, they are not sealed
properly and allow water to enter the well. The use of o-
rings and/or wrapping national pipe thread threads (NPT)
with polyfluoromer tape minimizes leakage potential. If
tape is used on threads that are not NPT, the tape may
actually increase the potential for the joint to leak. The
American Society for Testing and Materials (ASTM)
Standard F-480 specifies pressure ratings for the joints in
polymer casing. Usage of materials following this standard
with a pressure rating of 25 pounds per square inch (psi) or
greater will minimize the potential for joint leakage.
Annular Fill Artifacts
The materials that may be used in the annular space during
monitoring well installation include: the filter pack; fine
sand on top of the filter pack to minimize grout
contamination; bentonite; bentonite/cement mixtures and
cement. All materials may cause sampling artifacts if not
installed properly or used in inappropriate situations. A
filter pack should consist of inert, non-reactive materials.
Not all sands are silica and therefore the material used as a
filter pack should be verified during monitoring well
installation.
Bentonite is typically added to a monitoring well above the
filter pack to minimize grout contamination. The use of
bentonite as a "seal" against contaminants is a misnomer.
Bentonite is available in a variety of commercial forms and
is added either in solid form (by pellets) or in the form of a
slurry. If bentonite is introduced in such a manner that
formation water moving toward the well screen contacts the
bentonite, sorption of electrically-charged contaminants,
both organic and inorganic, sorption by the clay particles
will occur. In this setting, contaminants within the
formation may either be detected in the well in
concentrations less than the formation, or not at all. One
solution to this situation is to add a physical barrier between
the bentonite and the filter pack, such as a packer. Another
approach to this problem is to add a layer of fine sand
between the bentonite and the filter pack.
Many questions about the proper use of, and effective
emplacement of, bentonite have arisen. If granular
bentonite is added to a well and adequate time for hydration
is not allowed, the bentonite may not provide the intended
purpose. The time of hydration varies by pellet size, but it
is questionable whether or not a hydration time less than
two hours is appropriate in any situation. When granular
bentonite is added through a standing water column, there
are questions about whether or not the pellets will "bridge"
selectively and thus leave pathways for unwanted water to
enter the annular space. Although many pellets are "coated"
to delay instant hydration when falling in a water column, it
is generally agreed that the maximum column of standing
water through which pellets should be attempted to be
dropped is 30 feet. Documented studies by manufacturers
or researchers are necessary before this recommendation can
be changed.
Bentonite is a clay with a high shrink-swell potential.
This property makes it effective as long as it remains
hydrated. When the bentonite dries out, it shrinks and
cracks. The use of bentonite in the unsaturated zone is
questionable because of the ability of the bentonite to
stay hydrated in the unsaturated zone. Whether or not
fully hydrated bentonite will stay saturated in the
unsaturated zone in the presence of capillary action on
the clay is questionable. The conditions under which the
bentonite will remain hydrated are related to the
formation material with which it is in contact. If the
bentonite does not stay saturated, the clay will crack and
an unspecified amount of time and moisture will be
necessary to re-swell the clay. The concern is that these
cracks will allow rapid movement of contaminants along
the annular space in this area until sufficient moisture re-
hydrates the bentonite. Restricted use of bentonite in the
unsaturated zone is recommended until research has
been conducted to determine whether or not fully
hydrated bentonite placed the unsaturated zone will
remain fully hydrated. Field experience indicates that in
some cases, the bentonite dehydrates. Research is
needed.
Use of bentonite as an additive to cement is a popular
practice. Originally, the bentonite was added to control
shrinkage of the cement when it cured and to improve its
pumpability. Halliburton research on bentonite/neat cement
mixtures indicates that bentonite does not control
shrinkage. Research is needed to determine if and where
the use of bentonite/cement mixtures are appropriate and/or
desirable.
Cement is typically used to fill the annular space above a
74
-------
bentonite "plug" to the surface. Cement, if it enters the
filter pack, can significantly raise the pH of the water.
Concerns about the precipitation of cations, and elevation of
specific conductance are valid if cement contamination of
the filter pack occurs. Steps should be taken to make sure
that either fine sand or properly emplaced and hydrated
bentonite is used in the well and properly separated from the
well screen. Mixing neat cement at a ratio of 5.5 to 6
gallons of water per 94 pound bag of cement will also
minimize grout infiltration by decreasing the mobility of the
grout. When adding cement to the well below the water
table, or in wells with an unsaturated zone greater than 20
feet, the grout should be emplaced using a tremie discharge
pipe and pumping grout from the bottom of the formation
until returns are seen at the surface. Emplacing cement
with a submerged discharge tremie pipe reduces the risk of
contamination, through weak zones developing in the grout.
Development
Development is the process by which all artifacts from
drilling are supposed to be removed from the well.
However, development becomes more difficult when it must
penetrate into the filter pack and/or formation. For example,
the cutting head of a 4.25-inch ID Central Mine Equipment
auger creates a borehole of 8.25 inches. However, Mobile
augers create an 11 inch borehole for a 4 inch ID auger
flight. If two inch monitoring wells are installed in this
borehole, the filter pack thickness will range between
approximately 3 inches and 4.5 inches if the screen is
perfectly centered in the borehole. Removing smearing
effects from the borehole wall through the filter pack is
difficult, if not impossible. Other methods have similar
limitations, particularly those such as mud rotary that
produce a "skin" effect on the borehole wall. It is very
difficult to remove this modified surface from the entire
screen length.
Other concerns about development relate to air entrainment,
addition of water, and problems caused by excessive
sediment in the well during development. Entrainment of
air in the formation has traditionally been associated with
development by air lifting. Acceptable methods of
development in monitoring wells do not include air lifting
because of the potential for the air to cause chemical
alteration of the formation water, to oxidize and/or
precipitate constituents within the formation and to
physically "block" the formation with air if the well
becomes "air-locked".
The addition of water during the development process is a
questionable practice. The addition of water to the
formation dilutes the indigenous water and can cause
chemical changes resulting in undesirable chemical
reactions within the formation. Precipitation of constituents
can physically block pathways of migration for
contaminants. Water may not be able to be retrieved from
the formation and thus may alter the water chemistry
permanently. Most formations should yield enough water to
be developed using the formation water if proper methods
and techniques are utilized. The addition of water to
compensate for choosing an inappropriate technique is not
acceptable. If enough time is allotted, most wells can be
developed with formation water. This may involve the
process being repeated over a number of days in low yield
wells. If water is added, the amount of water that will need
to be removed varies based on formation, flow paths and
flow regimes. In the absence of better data, the rule of
thumb is to remove no less than three times the volume of
water added. The water should be removed as soon as
possible.
Excessive sediment in a monitoring well during
development is a concern because of the potential to force
the fine sediments into the formation and not be able to
retrieve the sediments. Clay-size particles and some clay
mineral particles have the potential to sorb contaminants
from ground water and also to reduce pathways of migration
for contaminants. The latter is especially true in formations
where flow is fracture-controlled. The problem of excessive
sediment in a monitoring well during development can be
minimized by using appropriate development methods and
proper techniques. The use of a well "sump" is not a
substitute for inadequate development or improper well
design.
Monitoring well development should be conducted using
only a combination of the following methods: surging/
pumping or overpumping/backwashing. Surging is
conducted by using a surging tool that is manufactured or
made in the field. The critical design elements of a surging
tool are a relatively close fit with the casing and means in
the tool by which the water may pass through the tool so
that the casing will not collapse during development. During
surging, the development takes place by surging an interval
(usually approximately two feet), beginning at the top of the
screen and continuing until the screen is completely
developed. The sediment-laden water must be removed
between every interval or two (depending on the amount of
fines in the water) to minimize forcing fine material too far
into the formation.
Overpumping and backwashing rely on the same principle
of moving water in and out of the formation to dislodge and
move the finer particles into and out of the well. Methods
that only pump in one direction are usually ineffective in
proper well development. Historically, a major problem has
been that most drilling rigs that are used to install
monitoring wells are not properly equipped to develop
monitoring wells. This is particularly true with regard to
surging and pumping, alternatively.
75
-------
The most frequently made mistake in well development is
"giving up." It is important to recognize that adequate
development takes time. Development time may range from
two hours to more than three days in order to be effective.
Remember that the thicker the filter pack, the longer the
development time required, and the greater the energy
requirement for surging.
Properly designed wells with appropriately sized well screen
and filter packs also minimize well development time.
Traditionally monitoring wells were installed with either a
10 or 20-slot screen and a filter pack of material that would
bridge (80% or more) on the selected slot size. This
"design" is often unsuitable in extremely fine-grained silts
and clays, with the resultant wells being high in turbidity
when sampled. Very little research has been done on very
fine grained, pre-packed filters, ceramic filters, etc. and their
impact on sample quality. References on monitoring well
design and found in the American Society for Testing and
Materials Standard D5092-90 (1990), Aller et al. (1989) and
Nielsen (1991).
Questions as to how much of a filter pack volume will be
removed during development and the effect that volume
reduction will have on the integrity of the well are
significant. Similarly, loss of the filter pack to the formation
and concomitant volume reduction and associated effects
were also questioned. It was agreed that appropriately
designing the well with a screen and filter pack picked for
the formation at hand should minimize this concern.
However, research into these related problems is warranted.
Other Artifacts
Other artifacts that may affect the chemistry of water in the
monitoring well include: microbial activity in the well;
interzonal leakage across a screened zone; and migration
along a casing due to ineffective grout adherence to casing
materials. Microbes in the well, particularly what are
known collectively as "iron bacteria", are not uncommon.
These microbes have the capability to alter the natural
chemical environment and change reducing conditions to
oxidizing conditions and vice versa. The magnitude of the
effect of these organisms are not known because the
organisms are not well understood. In the water well
industry, these organisms are considered as a "nuisance"
because they are not toxic. The traditional methods of
rehabilitation that have been used to temporarily destroy
their protective sheaths are inappropriate in a monitoring
well due to addition of chemicals to the well. The chemical
methods used in water wells have been relatively ineffective
over time. The full effect of these organisms in monitoring
wells is not known. This is an area where research is
warranted. Clearly, field procedures are lacking.
Where screens cross more than one zone of flow with
different hydraulic gradients, two concerns are raised. First,
the potential for mixing in the borehole exists; and second
the opportunity for interzonal mixing through the well and
into the zone with the lower hydraulic head exists. The
concern can be minimized by understanding the flow
characteristics of the site, developing a site hydrogeologic
model and using short, specific screen lengths were the
situation is anticipated. It is commonly, and incorrectly
assumed, that there is generally "mixing" in an aquifer that
is considered to be hydraulically "isotopic". This fails to
recognize that most samples are zone specific, and
dependent upon specific flow lines.
Questions as to the adherence of casing to certain types of
grout have been raised as a possible source of contamination
along the casing. Kurt and Johnson (1982) and Molz and
Kurt (1979) have looked at some of the issues of grout
adherence to casing and effects of temperature on casing
materials.
Well Design
Traditional Thinking
Many of the artifacts mentioned above are the result of the
acceptance and desire to install traditional monitoring wells
with traditional methods. Indeed the most frequently
installed size monitoring well is two inch ID, although some
states require four-inch wells in certain programs such as
hazardous waste monitoring or underground storage tanks.
Further, company policies may also dictate the installation
of four-inch monitoring wells. Some deeper wells are also
larger diameter and some older wells were built in the days
when sampling equipment required larger diameters.
Wells are traditionally installed to address one or more of
the following objectives: regulatory monitoring; water level
measurement; hydraulic conductivity determination; or "
other investigative reasons". Inherent in the "typical
thinking" is the necessity to design a monitoring well that
will yield enough water to collect the volumes of water
necessary to test for regulatory purposes. While this may be
desirable, it may not always be practical in low-yield
formations.
Low Yield Formations
Conventional monitoring well designs are inadequate to
meet all the objectives of representative monitoring in low-
yield formations. In order to obtain a reliable water quality
sample, it is desirable not to dewater the top of the screen
when collecting a water sample. It is desirable, in fact, not
76
-------
to lower the water level at all. This either requires a change
in thought pattern or an inventive sampling methodology
(which was summarily decided to be outside the scope of
this group and to be determined by others).
If it is possible to change our frame of reference, then other
options for designing low yield wells exist. Perhaps special
purpose wells should be considered. The use of pore water
samples from cores may be appropriate in some situations.
McKay (1993) has used one-half inch well completions in
low yield formations in order to try to minimize many of the
artifacts previously discussed. Installation of a small
diameter well in a pushed shelby tube, split spoon hole or
with a drive point may minimize the smearing effect
associated with augering. These types of completions will
necessarily limit the sample volume available at any one
time. Typically the sample collection rate should be limited
to the recovery rate of the well. If the water level is held
constant in larger diameter wells, the effect is the essentially
the same.
Determination of what is a low-yield formation is
particularly important, but difficult with current methods
and thinking. It is inappropriate to use only grain size for
determination as to when a conventional well design is
appropriate. The depositional model, stratigraphic model
and diagenetic model are all essential. In fine-grained
formations, the fracture flow pathways may override the
significance of grain size. McKay etal. (1993a; 1993c) has
discussed flow and flow rates in fine grained deposits. One
strongly identified research need is for development of
better and more accurate ways to measure hydraulic
conductivity on a meaningful scale in low-yield formations.
Other traditional problems with turbidity in fine-grained
formations were raised. A research need was identified to
determine whether or not low flow sampling techniques
always yielded turbidity-free samples, (e.g. how is colloidal
transport being sampled?).
Fractured Rock Completions
Open hole completions have been common in the water well
industry in fractured rock in order to maximize the amount
of water that enters the borehole. However, open hole
completions in monitoring wells are not acceptable.
Fractured rock monitoring wells should be designed to
prevent cross contamination between zones of different
hydraulic head as discussed above. The identification of
specific zones to be monitored is often very difficult in a
rock hole. This technology is to be dealt with elsewhere.
Overall Monitoring Issues
The discussion about the "true" definition of aquifer in the
monitoring well context has been discussed since it has been
desirable to monitor low yield formations. However, the
important concept to remember is that monitoring of an
aquifer may be desirable, but monitoring of pathways for
migration is essential. This necessitates monitoring of zones
that have not traditionally been considered as aquifers.
An adequate monitoring well network cannot be designed
until a site investigation is completed. For site investigations
and monitoring well installation to be effective and to
reduce unwanted artifacts, qualified, trained geologists as
well as drillers and helpers are needed in the field.
Unfortunately, less experienced geologists are traditionally
relegated to the field while the "more experienced"
personnel move on to more prestigious office and
management assignments. This is due, in part, to cost-
saving measures by the customer as well as the provider.
Further, many of the "field personnel" are not adequately
trained, if at all, to make the important field decisions.
Indeed, many do not even know that they are making
important decisions. Either a complete attitude change is
necessary to assign trained individuals to the field, or a more
formalized training process is necessary for field personnel.
Certification is not necessarily the answer; training is.
Part of a successful site investigation is in depth description
of the geology and hydrogeology at the site. In order for the
appropriate level of detail to be developed complete and
accurate description of samples is necessary. Part of this
process relates to training and experience; part relates to the
type of information that needs to be collected. There are
many classification systems that provide guidance as to the
types of information to record when visually classifying
samples. The Unified Soil Classification System is most
frequently used because it was developed to identify
important engineering properties. While these
characteristics are important, additional descriptions such as
environment of deposition, diagenetic changes,
hydrostratigraphic units and lithofacies descriptions are
equally important for determining pathways of migration.
Characteristics, such as color, must be defined in a
reproducible manner. An accepted classification system or
combination of systems is necessary to adequately describe
lithology for monitoring well purposes. The use of standard
references, such as color charts for both soils and rock, add
reproducibility to the description. Descriptions of many of
the parameters that should be recorded are described in Sara
(1994).
Monitoring well design, construction and development are
the most important part of a monitoring system. When
compared with analytical costs, the installation and
development is the least expensive portion of the sampling
process. However, costs are typically cut during the
installation phase. Later, considerable sums of money are
spent on repeat analytical results to justify or rectify poor or
inadequate installation. Awareness needs to be raised as to
77
-------
the importance of spending adequate sums of money up-
front for long-term benefit.
Questions were raised as to whether or not it is appropriate
to make generalizations about design considerations for
hydrogeologic settings. The consensus was that for
overview purposes, generalizations were appropriate, but for
detailed design this approach was inappropriate. Site specific
information and monitoring objectives should always be
considered in lieu of using generalizations.
One last statement about monitoring well installation that
has been widely publicized, but that is still important is the
presence of adequate annular working space during
monitoring well installation. A minimum two inch working
annular space is necessary to minimize bridging of annular
fill materials and to maximize the central placement of the
screen within the well. However, it must be remembered
that where this annular space is filled with gravel or sand-
pack around the screen, the larger the annular space the
more difficult the development.
Summary of Important Items
General
Although many items were discussed the major points of
discussion are summarized as follows:
1) A site investigation must be completed before an
adequate monitoring well network can be chosen.
2) We are interested in monitoring "pathways of
migration", not necessarily "aquifers" by
definition.
3) Do not skimp on monitoring well costs because
design, construction and development are the most
important parts of the system, and the least
expensive over the long haul.
4) Conventional monitoring well designs are
inadequate to meet all objectives in low hydraulic
conductivity formations. (Detection monitoring
water sample volumes may not be possible.)
Grain size alone should not be used to determine
when conventional wells are appropriate. We need
to change our frame of reference. We are not
monitoring system "averages" or "typical" zones.
We must monitor potential pathways of
contamination.
5) Methods are needed to monitor in low hydraulic
conductivity formations. Better ways to measure
and evaluate hydraulic conductivity in low
hydraulic conductivity formations are needed.
6) There is a need to expand (and possibly
standardize) existing visual classification systems
for samples (i.e. USCS or other) to include
geologic interpretation of environment of
depositions and hydrostratigraphic units.
7) The need for qualified, trained field personnel is
emphasized.
Special Considerations
The following were considered important special
considerations that should be considered in the future:
1) The use of special purpose wells in low hydraulic
conductivity formations (i.e. one half-inch well,
lysimeters) should be considered;
2) The use of pore water samples should be
considered in low hydraulic conductivity
formations;
3) Long, open-hole completions in fractured rock are
inappropriate due to the potential for inter-
formation migration; and
4) Guidance on well design can be developed for
hydrogeologic settings where broad-scale overview
and knowledge transfer is the objective. However,
this approach is not appropriate for detailed well
design.
Research Needs
1) Smearing along borehole walls during monitoring
well installation, particularly in auger drilling, was
identified as a problem particularly in low
hydraulic conductivity formations and/or where the
clay content of the formation is relatively high and
flow is primarily in fractures. A research need was
identified to better define the components that
cause smearing and to suggest and/or design
alternatives to overcome and/or minimize this
problem.
2) Bentonite is sometimes used in the unsaturated
zone as an annular fill material. A research need
was identified to answer questions as to whether or
not fully-hydrated bentonite placed in the
unsaturated zone remains fully hydrated over time.
If the bentonite desiccates in only certain
situations, these circumstances need to be
identified.
78
-------
3) Low hydraulic conductivity formations are difficult
to monitor using traditional monitoring well
designs. However, effectively defining a low
hydraulic conductivity formation is even more
difficult. Traditional methods for measuring
hydraulic conductivity have been proven to yield
hydraulic conductivity values orders of magnitude
too low. A research need was identified to develop
testing methods (both in the laboratory and in-situ)
that will be more representative of the flow
characteristics of the formation that intersects the
fractures as well as through the bulk of the
formation.
Guidance Needs
There was strong agreement that guidance on
wells in low conductivity formations needs to be
developed.
Reference Documents
There are many documents that will help to understand the
issues that were discussed by this group. The following
documents provide basic information and contain references
on additional subjects. The reference section contains
additional references that deal with more specific subject
matter.
1) American Society for Testing and Materials
Standard D5092-90, Standard Practice for Design
and Installation of Ground Water Monitoring Wells
in Aquifers, vol. 4.08, Philadelphia, Pennsylvania,
12pp..
2) American Society for Testing and Materials,
(1994). Ground Water and Vadose Zone
Investigations, Philadelphia, Pennsylvania, 396 pp.
3) Aller, L., T. W. Bennett, G. Hackett, R. J. Petty, J.
H. Lehr, H. Sedoris, D. M. Nielsen and J. Denne,
1989. Handbook for the Suggested Practices for
the Design and Installation of Ground-Water
Monitoring Wells, National Water Well
Association, Dublin, Ohio, 398 pp.
4) USEPA, 1992. RCRA Ground-Water Monitoring:
Draft Technical Guidance, Office of Solid Waste,
EPA/530-R-93-001, PB93-139-350, Washington,
DC.
5) Sara, M. N., 1994. Standard Handbook for Solid
and Hazardous Waste Facility Assessments, Lewis
Publishers, Ann Arbor, Michigan.
6) Nielsen, D. M., ed., 1991. Practical Handbook of
Ground-Water Monitoring, Lewis Publishers,
Chelsea, Michigan, 717 pp.
7) Driscoll, F. G., 1986. Groundwater and Wells,
Johnson Division, St. Paul, Minnesota, 1089 pp.
References
American Society for Testing and Materials, (1994).
Specification for Thermoplastic Well Casing, Pipe and
Couplings, Made in Standard Dimension Ratio (SDR), SCH
40, SCH 80, Standard F-480-94, vol. 8.04 Philadelphia,
Pennsylvania, 25 pp.
Barcelona, M. J., 1984. TOC Determinations in Ground
Water, Ground Water, vol. 22, no. 1, pp. 18-24.
Barcelona, M. J., G. K. George andM. R. Schock, 1988.
Comparison of Water Samples form PTFE, PVC and SS
Monitoring Wells, United States Environmental Protection
Agency, Office of Research and Development,
Environmental Systems Monitoring Laboratory, Las Vegas,
EPA 600/X-88/091,37pp.
Barcelona, M. J., J. P. Gibb and R. Miller, 1983. A guide to
the Selection of Materials for Monitoring Well Construction
and Ground-Water Sampling, Illinois State Water Survey,
SWS Contract Report 327, Champaign, Illinois, 78 pp.
Barcelona, M. J., J. A. Helfrich and E. E. Garske, 1985.
Sampling Tubing Effects on Ground-Water Samples,
Analytical Chemistry, vol. 57, no. 2, pp. 460-464.
Barcelona, M. J. and J. A. Helfrich, 1986. Well
Construction and Purging Effects on Ground-Water
Samples, Environmental Science & Technology, vol. 20, no.
11, pp. 1179-1184.
Barcelona, M. J. and J. A. Helfrich, 1988. Laboratory and
Field Studies of Well-Casing Material Effects, Proceedings
of the Ground Water Geochemistry Conference, National
Water Well Association, Dublin, Ohio, pp. 363-375.
Bianchi-Mosquera, G. C. and D. M. Mackay, 1992.
Comparison of Stainless Steel vs. PTFE Miniwells for
Monitoring Halogenated Organic Solute Transport, Ground
Water Monitoring Review vol. 12, no. 4, pp. 126-131.
Boettner, E. A., G. L. Ball, Z. Hollingsworth and R. Aquino,
1981. Organic and Organotin Compounds Leached from
PVC and CPVC Pipe, United States Environmental
Protection Agency Report, EPA/600/1-81-062, 102 pp.
79
-------
Brobst, R. B. and Buszka, P. M, 1986. The Effect of Three
Drilling Fluids on Ground Water Sample Chemistry, Ground
Water Monitoring Review, vol. VI, no. 1, pp. 62-70.
Cowgill, U. M., 1988. The Chemical Composition of
Leachate from a Two-Week Dwell-Time Study of PVC Well
Casing and Three-Week Dwell-Time Study of Fiberglass
Reinforced Epoxy Well Casing, in A. G. Collins and A. I.
Johnson, eds., Ground-Water Contamination: Field
Methods, American Society for Testing and Materials,
Philadelphia, Pennsylvania, STP 963, pp. 172-184.
Curran, C. M. and M. B. Tomson, 1983. Leaching of Trace
Organics into Water from Five Common Plastics, Ground
Water Monitoring Review, vol. 3, no. 3, pp. 68-71.
Gillham, R. W. and S. F O'Hannesin, 1990. Sorption of
Aromatic Hydrocarbons by Materials Used in Construction
of Ground-Water Sampling Wells, in D. M. Nielsen and A. I.
Johnson, eds., Ground-Water and Vadose Zone Monitoring,
American Society for Testing and Materials, Philadelphia,
Pennsylvania, STP 1053, pp. 108-122.
Hewitt, A. D., 1989. Leaching of Metal Pollutants from
Four Well Casings Used for Ground-Water Monitoring,
CRREL Special Report 89-32, U. S. Army Cold Regions
Research and Engineering Laboratory, Hanover, New
Hampshire.
Hewitt, A. D., 1992. Potential of Common Well Casing
Materials to Influence Aqueous Metal Concentration,
Ground Water Monitoring Review, vol. 12, no. 2, pp. 131-
136.
Jones, J. N. and G. D. Miller, 1988. Adsorption of Selected
Organic Contaminants onto Possible Well Casing Materials,
in A. G. Collins and A. I. Johnson, eds., Ground-Water
Contamination: Field Methods, American Society for
Testing and Materials, Philadelphia, Pennsylvania, STP 963,
pp. 185-198.
Junk, G. A., H. J. Svec, R. D. Vick and M. J. Avery, 1974.
Contamination of Water by Synthetic Polymer Tubes,
Environmental Science and Technology, vol. 8, no. 13, pp.
1100-1106.
Keely, J. F. and K. Boetang, 1987. Monitoring Well
Installation, Purging, and Sampling Techniques - Part I:
Conceptualizations, Ground Water, v. 25, no. 3, pp. 300-
313.
Kurt, C. E. andR. C. Johnson, 1982. Permeability of Grout
Seals Surrounding Thermoplastic Well Casing, Ground
Water, vol. 20, no. 4, pp. 415-419.
Marsh, J. M. and J. w. Lloyd, 1980. Details of
Hydrochemical Variations in Flowing Wells, Ground Water,
vol. 18, no. 4, pp. 366-373.
McKay, L. D., J. A. Cherry and R. W. Gillham, 1993a.
Field Experiments in a Fractured Clay Till: 1. Hydraulic
Conductivity and Fracture Aperature, Water Resources
Research, vol. 29, no. 4, pp. 1149-1162.
McKay, L. D., R. W. Gillham and J. A. Cherry, 1993b.
Field Experiments in a Fractured Clay Till: 2. Solute and
Colloidal Transport, Water Resources Research, vol. 29, no.
12, pp. 3879-3890.
McKay, L. D., J. A. Cherry, R.C. Bales, M.T Yahya and
C.P Gerba, 1993c. AField Example of Bacteriophage as
Tracers of Fracture Flow, Environmental Science and
Technology, vol. 27, no. 6, pp. 1075-1079.
Miller, G. D., 1982. Uptake and Release of Lead,
Chromium and Trace Level Volatile Organics Exposed to
Synthetic Well Casings, Proceedings of the Second National
Symposium on Aquifer Restoration and Ground-Water
Monitoring, National Water Well Association, Dublin, Ohio,
pp. 236-245.
Molz, F. J. and C. E. Kurt, 1979. Grout-Induced
Temperature Rises Surrounding Wells, Ground Water, vol.
17, no. 3, pp. 264-269.
Parker, L. V and T F. Jenkins, 1986. Suitability of
Polyvinyl Chloride Well Casings for Monitoring Munitions
in Ground Water, Ground Water Monitoring Review, vol. 6,
no. 3, pp. 92-98.
Parker, L. V, A. D. Hewitt and T. F. Jenkins, 1990.
Influence of Casing Material on Trace-Level Chemicals in
Well Water, Ground Water Monitoring Review, vol. 10, no.
2, pp. 146-156.
Parker, L. V, 1991. Discussion of The Effects of Latex
Gloves and Nylon Cord on Ground Water Sample Quality,
by J. L. Canova and M. G. Muthig, Ground Water
Monitoring Review, vol. 11, no. 4, pp. 167-168.
80
-------
Parker, L. V, 1992. Suggested Guidelines for the Use of
PTFE, PVC, and Stainless Steel in Samplers and Well
Casings, Current Practices in Ground Water and Vadose
Zone Investigations, in D. Nielsen and M. Sara, eds.,
American Society for Testing and Materials, Philadelphia,
Pennsylvania, STP 1118, pp. 217-229.
Reynolds, G. W. andR. W. Gillham, 1985. Absorption of
Halogenated Organic Compounds by Polymer Materials
Commonly Used in Ground-Water Monitors, Proceedings of
the Second Canadian/American Conference on
Hydrogeology, National Water Well Association, Dublin,
Ohio pp. 125-132.
Reynolds, G. W., J. T. Hoff and R. W. Gillham, 1990.
Sampling Bias Caused by Materials Used to Monitor
Halocarbons in Groundwater, Environmental Science
Technology, vol. 24, no. 1, pp. 135-142.
Rivett, M., S. Feenstra and J. Cherry, 1991. Field
Experimental Studies of a Residual Solvent Source
Emplaced in the Ground Water Zone, Proceedings of the
Conference on Petroleum Hydrocarbons and Organic
Chemicals in Ground Water, National Water Well
Association, Dublin, Ohio, pp. 283-299.
Schalla, R., 1986. A Comparison of the Effects of Rotary
Wash and Air Rotary Drilling Techniques on Pumping Test
Results, Proceedings of the Sixth National Symposium and
Exposition on Aquifer Restoration and Ground Water
Monitoring, National Water Well Association, Dublin, Ohio,
pp. 7-26.
Sosebee, J. B., P. C. Geiszler, D. L. Winegardner and C.
Fisher, 1983. Contamination of Ground Water Samples
With Poly (Vinyl Chloride) Adhesives and Poly (Vinyl
Chloride) Primer From Monitor Wells, in Proceedings:
American Society of Testing and Materials Second
Symposium on Hazardous and Industrial Solid Waste
Testing, ASTM Special Publication #805, pp. 38-49.
Tomson, M. B., S. R. Hutchins, J. M. King and C. H. Ward,
1979. Trace Organic Contamination of Ground Water:
Methods for Study and Preliminary Results, III World
Congress on Water Resources, Mexico City, Mexico, vol. 8,
pp. 3701-3709.
81
-------
Well Purging and Sampling
Robert W. Puls and John F. McCarthy
Traditional approaches to purging and sampling ground
water involve purging a well at a high pumping rate until a
fixed number of casing volumes (usually 3-5) is evacuated,
followed by sample collection at a lower pumping rate.
This approach has raised concerns about the
representativeness of samples collected using these
methods, especially if the sampling objectives include
monitoring of contaminants. Briefly, the concerns include
entrainment of immobile particles (Puls et al. 1992) and the
possible need to filter samples to remove those artifacts
(Puls and Powell, 1992; Backhus et al. 1993), costs of
pumping and disposing of large volumes of contaminated
water (Barcelona et al., 1994; Korte and Kearl, this report),
and uncertainties in interpreting the source of the sampled
water (Martin-Hayden and Robbins, in revision). The goal
of this discussion group was to evaluate approaches to
purging and sampling and determine if new techniques
should be recommended. Group discussions addressed
questions related to the objectives of purging, methods of
well purging and sampling, the advantages and
disadvantages of low-flow rate purging, site-specific
considerations in purging and sampling methods, and trade-
offs among sampling devices.
The general consensus of the discussion group was that, in
many cases, new methods should be adopted for purging
and sampling wells; the recommended method is generally
referred to as low-flow, minimal drawdown purging and
sampling. The principal differences between this and more
traditional approaches centers on the rate of pumping and
the criteria for deciding that purging is complete. The
newer method calls for slow flow rates for purging and
sampling in order to minimize chemical and hydrological
disturbance in and around the well. Furthermore, the
completion of purging is gauged on site-specific chemical
criteria (stabilization of water quality parameters) rather
than on a fixed number of well volumes pumped.
Objectives of Well Purging
The objective of well purging is to obtain formation water
from the targeted sampling point with no alteration of water
chemistry. The location within the subsurface from which
that formation water is actually drawn depends on a number
of factors including how we purge. There was general
agreement among the members of the discussion group that
water in the well casing, and perhaps even water in the
screened interval of the well, is different than the formation
water. This is due to a variety of factors including the
following: gas diffusion into and out of the standing water
column, potential alteration of water chemistry from contact
with the well casing, filter pack and annular sealing
materials, and surface infiltration. The similarity of the
water in the screened interval to that of the formation will
depend on the following: ground water flow velocity and
direction, screen length, well diameter, well depth, distance
from the screen to the water table, geologic and hydrologic
heterogeneities in the screened interval and the degree of
connectedness of the well to the aquifer. It was also
generally agreed that these same factors and considerations
would apply whether the sampling approach involved
'portable', or 'semi-permanent' dedicated sampling systems,
multilevel samplers or whether the sampling was carried out
in monitoring wells or using screening tools (e.g.
Geoprobe). The volume of water that needs to be purged to
obtain formation water will, however, depend on the
particular sampling approach, and monitoring system
selected, and the hydrogeologic characteristics of the site.
Methods of Well Purging and Sampling
It was generally agreed by the discussion group that
disturbance of the sampling zone or point should be
minimized and that purging methods which tend to alter
water chemistry or the integrity of the sample should be
avoided. The objectives of the sampling program should be
considered in selecting a purging method. There was
general agreement that consistency and adequate
documentation are essential and currently lacking in many
sampling programs.
For purposes of site assessment and remedial performance,
low-flow purging which produces minimal drawdown of the
water table was generally recommended, with no change of
flow rates during sampling. Conceptually, formation water
flowing across the screened section of the well is sampled
no faster than it enters the well bore under natural
hydrological flow conditions. The criteria for the
appropriate rate for purging and sampling is hydrological:
pumping rates should produce no net (or at least minimal)
drawdown of the water table. There are several advantages
to this method, a principal one being that the volume of
purge water is much less than for more traditional purging
methods. Low-flow sampling conditions have been
demonstrated to remove water from only the screened zone,
without the need for packers to isolate the stagnant
82
-------
overlying casing water (Powell and Puls, 1993; Gillham et
al. 1985; Robin and Gillham, 1987). Although flow rates
are slower, much less water needs to be purged, with the
benefit that, in contaminated sites, a much smaller volume
of water will require subsequent waste disposal. More
fundamentally, however, a key benefit of the decreased
volume of purge water is that a smaller section or volume of
the aquifer is sampled. This represents a significant
improvement in our ability to detect and resolve
contaminant distributions, which may vary greatly over
small distances in three-dimensional space.
The traditional method of evacuating a standard 3-5 well
volumes samples a much larger portion of the aquifer, and
provides a larger volume-averaged concentration. That
averaged value is directly related to the volume of water
purged, the geologic setting, and the placement of the
sampling point. For some sampling objectives, such as
determining a large volume-averaged number for water
resource analysis purposes, the traditional approach may be
valid. However, much of the discussion centered around
contaminant detection and long-term monitoring at landfills
and hazardous waste sites. In most cases, the large volume
of aquifer sampled complicates interpretation of data
concerning the concentration and spatial distribution of
contaminants. If the well screen is long and intersects only
a small portion of a contaminant plume, a biased low
concentration value would be produced due to mixing of
uncontaminated portions of the aquifer. Similarly, in a
fractured clay or rock, most of the water comes from the
fractures. Because the fracture porosity is so small, the
sampling process may draw in water from a very large
volume of the deposit. This could greatly dilute the
concentration of contaminants in the sample.
Recent research has highlighted the significant chemical and
physical heterogeneities that exist even within porous media
(Davis etal. 1993; Nikolaidis et al. 1994). Flow problems
analogous to those associated with fractured rock can occur
in granular aquifers due to vertical heterogeneities in
hydraulic conductivity in typically layered sediments (Hess
et al. 1991). For this reason, it was generally agreed that for
many sampling purposes the use of smaller screen lengths
(1-5 ft) was best. Multi-level samplers, which are available
in combination with pumps or which use dialysis cells for
the 'passive' collection of water (Ronen et al. 1987), were
also recommended for further resolution of the spatial
distribution of contaminants in a formation. These sampling
devices may the best choice for sampling water in low-
permeability formations and in fractured rock (in
combination with packers). It was suggested that low-flow
purging and sampling techniques or the use of multilayered
sampling devices might also be capable of obtaining useful
information in long-screened wells. However, this would
depend on the establishment of good hydraulic connection
between the well and the adjoining formation. This topic
requires additional research.
The group consensus was that common water quality
indicator parameters, such as specific conductance,
dissolved oxygen and turbidity should be used to determine
the endpoint of purging, i.e., when formation water has been
accessed. Ideally, the concentration of a contaminant (or
other species of interest) would be measured over time
during purging to determine when its concentration
stabilized. In most cases, this approach would not be
practical. However, some studies (Puls et al. in review;
Barcelona et al. 1994; Backhus et al. 1993) have indicated
correlation of some indicator parameters with different
classes of contaminants (e.g. dissolved oxygen with volatile
organic compounds, and turbidity with metals and
hydrophobic organic compounds). While the measurement
of dissolved oxygen can be problematic at low
concentrations (< 1 ppm), many reliable field-portable
chemical measurement techniques (colorimetric) and
improved dissolved oxygen probes have recently been
developed which should help in obtaining more accurate
and stable measurements. A conservative approach was
recommended which included the use of dissolved oxygen
and specific conductance for volatile contaminants,
turbidity, dissolved oxygen and specific conductance for
metals (and metalloids) and semi-volatiles, and perhaps also
the use of oxidation-reduction (redox) potentials in both
cases. It must be recognized that erroneous or highly
variable parameter measurements can result from improper
methods of sample collection or analysis. Dissolved oxygen
or electrode potential measurements will change rapidly if a
sample is open to the atmosphere, and turbidity can increase
within a few minutes if air is introduced into a sample
containing ferrous iron. In many cases, sampling
practitioners are using flow-through cells which can be
connected in-line with the sampling system. These devices
generally include most of the above-mentioned indicator
parameters as well as pH and temperature. The latter two
measurements are useful data, but are generally insensitive
as purging indicators. In addition to measurements to
determine stabilization of the water quality parameters,
time-series measurements of the change in the depth to the
water table should also be recorded to assure hydrologic
stabilization with minimal drawdown of the water table.
The frequency for measurement of the indicator parameters,
and the criteria to decide that the parameters are stabilized
needs to consider pump flow rate and the precision of the
monitoring instruments. Some suggested criteria were
discussed: stabilization to within 10% (for turbidity and
dissolved oxygen) for three consecutive readings taken three
minutes apart under conditions where flow rates ranged
from 100-500 ml/min; three successive readings within five
times the reproducibility of the instrument; and, individual
well evaluation of indicator parameter stabilization plots.
83
-------
The same use of indicator parameters was recommended in
fractured rock settings.
Finally, it was concluded that education and training of
ground-water samplers should be expanded, and
certification programs possibly initiated. It was strongly
recommended that EPA be involved in these technology-
transfer activities. Field sample collection currently
represents the greatest potential source of error in site
assessment and remedial evaluations. Increased education
and certification and the establishment of quality-assured
and consistent sampling protocols is critical to the future
success of multi-million dollar environmental restoration
programs.
Low-Flow Purging
Low-flow purging refers to the intake velocity of the
sampling device downhole and the resulting induced
formation water velocity, not the average flow rate at the
surface. The latter can be manipulated using flow valves or
other obstructions to produce low surface flow rates while
the subsurface induced flow may be extremely rapid and
impart significant disturbance within the sampled zone. The
overall objective is a more passive approach to sample
extraction with the ideal being to match the intake velocity
with the natural ground water flow velocity thus inducing
minimal drawdown of the water table.
Required purge volume or duration is evaluated through
continuous monitoring of water quality parameters such as
dissolved oxygen, specific conductance, oxidation-reduction
(redox) potential and turbidity for evaluation of the presence
of formation water. Research has shown that purging at
these lower rates with various types of pumps (peristaltics,
low-speed submersibles, and bladder pumps) does indeed
produce low turbidity and generally high quality samples
(Pulsetal. 1992;Kearletal. 1992). A perceived
disadvantage of such strategies is the additional time
required to remove the traditional 3 to 5 casing volumes. In
general, however, the volumes required to access formation
water are much less than 3 to 5 casing volumes, and in
deeper wells are actually only fractions of a casing volume.
Indeed the volume of water extracted to access formation
water is generally independent of well size and capacity
(Puls et al. in review; Puls, this report). The criteria for the
initiation of sampling are stabilization of the water quality
indicator parameters listed above. It should be noted that
excessively stringent stabilization guidelines may result in
longer purging. If chemically distinct zones exist within the
formation adjacent to the screened interval, they may be
accessed with prolonged pumping and result in changes in
water quality indicator values. The design of such sampling
points should be discouraged; however where they already
exist, caution should be exercised in acquiring and
interpreting the data from such sampling points.
The most important parameters affecting purge volume
appear to be hydraulic and geologic heterogeneity, water
chemistry, pumping rate, device size and whether the
sampling devices are used in a portable or dedicated
fashion. Significant reductions in purge volume have been
noted when the problems associated with disturbances
during installation of pumps immediately prior to sampling
are avoided (Puls and Powell, 1992). This can be achieved
by installing portable pumps at least a day before sampling,
or through the use of dedicated pumps. A thorough
economic analysis may show dedicated systems to be a cost-
effective alternative for routinely-sampled monitoring wells.
In general, the advantages of low-flow purging include the
following:
• low turbidity samples which are representative of
the 'mobile' load of contaminants present
(dissolved and colloid-associated),
• minimal disturbance of the sampling point,
• reduced stress on the formation (minimal
drawdown),
• less mixing of stagnant casing water with
formation water,
• reduced need for filtration and therefore less time
required for sampling,
• smaller purging and sampling volume which
decreases waste disposal costs and sampling time.
Whereas the disadvantages of low-flow purging are:
• resistance to change on the part of sampling
practitioners due to lack of long-term data base,
• problems with data comparison and interpretation
of temporal trends due to differences in sample
collection methods,
• increased time that may be required for purging
and sampling at slow flow rates,
• the need for additional (& costly) equipment,
• difficulty of implementation, and
• potential for increased gas exchange and sorption
of contaminants onto tubing surfaces due to
increased residence time of fluid in the tubing.
Recognizing the disadvantages, the advantages of the low-
flow-rate sampling in providing a higher quality sample that
more closely represents the mobile dissolved and colloidal
components in the formation suggest that this is the
direction that the state of professional practice must
proceed.
Regardless of the purging and sampling protocols selected
for a particular sampling objective, proper well construction
84
-------
and well development is important and intimately
interrelated to the issue of well purging. A detailed
understanding of the hydrologic and geologic variability of
the system is also essential in establishing sampling points
and in designing the overall sampling program. There was
general agreement among the group from a technical
perspective, that minimal drawdown, low-flow purging was
a better way to sample for most ground water sampling
programs.
Site-Specific Considerations
The overall goals of the sampling program or the sampling
objectives will drive how the sampling points are located,
installed, and the choice of sampling methods. Likewise,
site-specific hydrogeologic factors will affect these
decisions. Unlike water supply wells, wells installed for
ground-water quality assessment and restoration programs
are often installed in low water yielding settings. In fact, it
was the consensus of the group that the use of "typical"
ground-water monitoring wells in such locations should
undergo a serious review. Alternative types of sampling
points and sampling methods are needed in these types of
environments. The discussions focused on five general
classes of hydrogeologic settings:
• high permeability formations, screened below the
water table;
• high permeability, water table wells (i.e. screened
across the water table);
• low permeability formations, deep, screened below
the water table;
• low permeability formations, screened across the
water table; and
• fractured rock.
For the high permeability wells, either screened deep and
below the water table or screened across the water table, it
was generally agreed that use of low-flow purging and
sampling techniques was the preferred method for sample
collection in most instances. The use of low-flow techniques
were also advocated for the low-permeability, deep wells,
with the caveat being to avoid dewatering of the screened
interval. This may require extremely low-flow purging (<
100 ml/min) and may be technology-limited. Where devices
are not readily available to pump at such low flow rates, the
primary consideration was still to avoid dewatering of the
well screen. This may require repeated recovery of the water
during purging while leaving the pump in place within the
well screen. It was suggested that comparisons be made
between samples recovered using low-flow purging
techniques and using passive sampling techniques. The
latter would essentially entail acquisition of the sample with
no or very little purging using a dedicated sampling system
installed within the screened interval.
The most problematic of the above settings were those in
low permeability water table wells and fractured rock. In
the former case, there was a serious concern that an
adequate sample could not be obtained using traditional
monitoring wells and standard sampling devices. The group
consensus was that, once again, the primary consideration in
purging such wells was to avoid dewatering the well
screens. Use of low-flow techniques may be impractical in
these settings, depending upon the water recharge rates.
The sampler and the end-user of data collected from such
wells needs to understand the limitations of the data
collected, i.e. a strong potential for underestimation of
actual contaminant concentrations for volatile organics,
potential false negatives for filtered metals and potential
false positives for unfiltered metals. Once again it was
recommended that more passive sampling techniques be
investigated and compared with current and low-flow
purging techniques.
In fractured rock formations, a low- to no-flow purge was
recommended in conjunction with the use of packers to
isolate the sampling zone in the borehole. It is imperative in
such settings to identify flow paths or water-producing
fractures prior to sampling, using tools such as borehole
flowmeters. The spatial resolution issue referred to above is
particularly of issue in fractured rock. That is, the volume
purged can make a dramatic difference on the constituent
concentrations obtained.
For low-yielding wells or fractured rock formations, the
unnecessarily large sample volumes often required by some
regulatory programs is a significant problem. For example,
in a rock with a fracture porosity of 10-4, the 10 L sample
volume required by many protocols could conceivably
include the water from a volume of 100 m3 of rock. This
could greatly dilute the concentration of contaminants in the
sample and the concentrations may not be in equilibrium
with the adjoining matrix (McKay, this report). It was
strongly recommended that the different EPA programs
review their sample volume requirements. In many cases
they are unnecessary, and in fact have the result of
encouraging poor sampling practices.
Sampling Systems and Devices
The use of standard monitoring wells and devices to sample
ground water in low permeability formations should be
reevaluated and research should be directed toward the
development of alternative approaches. In most other
settings, many currently available devices (e.g. bladder
pumps, low-speed centrifugal submersible pumps etc.) are
entirely appropriate and adequate for the collection of water
samples. There was extensive discussion on the merits of
85
-------
using bailers of any type. The majority of participants
agreed with the statement that "the standards of professional
practice have passed bailers by" for most situations. Bailers
were deemed useful in water table wells and for sampling
light non-aqueous phase liquids (LNAPL's). Variability in
sampling technique was considered a major limitation to
their use in providing representative and reproducible data.
Production of excessive turbidity and aeration were also
seen as significant drawbacks or limitations.
It was recommended that studies be undertaken to evaluate
the cost-effectiveness of using dedicated sampling systems
as opposed to portable sampling systems for some sampling
programs. Research has indicated (Puls, this report;
Barcelona, 1994) that the benefits of low-flow purging and
sampling techniques are more fully realized in such
sampling systems. These include: less purge volume, time
savings in the field and in field preparation, less
decontamination and production of additional wastes from
decontamination procedures, and improvements in sampling
consistency and sampling results (e.g. better
reproducibility). The potential disadvantages from dedicated
systems were unknown effects from system deterioration
over the long term and up-front costs for installation and
implementation.
Documentation
Throughout the group discussions, the topic of
documentation of all sampling activities, including sampling
preparation and subsequent sample handling, transportation
and storage was brought up. It was unanimously agreed that
current documentation efforts and documentation
requirements are inadequate, regardless of environmental
program. This was targeted as a major need or area for
improvement in the ground-water sampling field. Included
in the appendix of this document are examples of a field-
sampling log forms which would be useful for providing
some of this information. Regardless of the sampling
objective or sampling approach employed, insufficient
documentation will make the interpretation of the resulting
data inadequate.
Research Needs
Identified research needs by the group included the
following:
• more thorough comparison of low-flow sampling
data (unfiltered) with filtered conventional
sampling data,
• comparison of low-flow sampling and "passive"
sampling techniques,
1 evaluations of cost-effectiveness of dedicated
sampling systems,
1 development of new methods and instruments for
acquiring water samples in low water-yielding
formations.
References
Backhus, D.A., J.N. Ryan, D.M. Groher, J.K. MacFarlane,
and P.M. Gschwend. 1993. Sampling Colloids and Colloid-
Associated Contaminants in Ground Water. Ground Water,
31(3), 466-479.
Barcelona, M. J., H. Allen Wehrmann, and Mark D. Varljen.
1994. Reproducible Well Purging Procedures and VOC
Stabilization Criteria for Ground-Water Sampling. Ground
Water, 32(1),.
Davis, J.A., CC. Fuller, J.A. Coston, K.M. Hess, andE.
Dixon. 1993. Spatial Heterogeneity of Geochemical and
Hydrologic Parameters Affecting Metal Transport in Ground
Water. USEPA Environmental Research Brief, EPA/600/S-
93/006, 22 pp.
Gillham, R.W., M.J.L. Robin, J.F. Barker, and J.A. Cherry.
1985 Field Evaluation of Well Flushing Procedures. API
Publ. 4405, 110pp.
Hess, K.M., S.H. Wolf, M.A. Celia, and S.P Garabedian.
1991. Macrodispersion and Spatial Variability of Hydraulic
Conductivity in a Sand and Gravel Aquifer, Cape Cod,
Massachusetts. USEPA Environmental Research Brief, EPA/
M-91/005, 9 pp.
Kearl, P.M., N.E. Korte, andT.A. Cronk. 1992. Suggested
Modifications to Ground Water Sampling Procedures Based
on Observations from the Colloidal Borescope. Ground
Water Monitoring Review, Spring, 155-160.
Martin-Hayden, J.M. and G.A. Robbins. 1994. Plume
Distortion and Apparent Attenuation Due to Concentration
Averaging in Monitoring Wells. In revision for publication
in Ground Water.
McKay, L., K. Novakowski, and J. McCarthy. 1994.
Ground-water Sampling of Fractured Clay and Rock. In
Ground Water Sampling Workshop, USEPA, Nov. 30-Dec.
2, 1993, Dallas, TX.
Nikolaidis, N.P, G.A. Robbins, M. Sherer, B. McAninch, G.
Binkhorst, and S.L. Suib. 1994. Vertical Distribution and
Partitioning of Chromium in a Glacio-Fluvial Aquifer.
Ground Water Monitoring and Remediation,.
86
-------
Powell, R.M. and R.W. Puls. 1993. Passive Sampling of
Ground-Water Monitoring Wells Without Purging:
Multilevel Well Chemistry and Tracer Disappearance. J.
Contam. Hydrol. 12,51-77.
Puls, R.W. and R.M. Powell. 1992. Acquisition of
Representative Ground Water Quality Samples for Metals.
Ground Water Monitoring Review, Summer, 167-176.
Puls, R.W., D.A. Clark, B. Bledsoe, R.M. Powell, and C.J.
Paul. 1992. Metals in Ground Water: Sampling Artifacts and
Reproducibility. Hazardous Waste & Hazardous Materials,
9(2), 149-162.
Puls, R.W. 1994. Use of Low-Flow or Passive Techniques
for Sampling Ground Water. In Ground Water Sampling
Workshop, USEPA, Nov. 30-Dec. 2, 1993, Dallas, TX.
Puls, R.W. and C.J. Paul. 1995. Low Flow Purging and
Sampling of Ground-Water Monitoring Wells With
Dedicated Systems. Accepted, Ground Water Monitoring
and Remediation, Winter issue, 1994.
Robin, M.J.L. and R.W. Gillham. 1987. Field Evaluation of
Well Purging Procedures. Ground Water Monitoring
Review,.7(4). 85-93.
Ronen, D., M. Magaritz and I. Levy. 1987. An In Situ
Multilevel Sampler for Preventive Monitoring and Study of
Hydrochemical Profiles in Aquifers. Ground Water
Monitoring Review, 7(4), 69-74.
87
-------
Turbidity and Colloid
Transport
Joseph N. Ryan, Steve Mangion and Dick
Willey
Background
Realization of the potential role of colloids in facilitating
contaminant transport (McCarthy and Zachara, 1989) has
heightened our awareness of the need to obtain ground
water samples that are representative of the naturally mobile
colloids. Carefully collected field evidence shows that the
commonly-used sampling protocols (bailing, rapid
pumping) produce ground water samples in which colloids
have been artificially entrained (Puls et al., 1992; Backhus
et al., 1993). In such samples, the naturally mobile colloid
fraction is overestimated. That is, bailed and rapidly
pumped samples often contain substantial turbidity that is
not representative of conditions within the subsurface. In
practice, the suspended particles causing this turbidity have
been removed from the samples by filtering in the field.
Usually, membrane filters with 0.45 |j,m pores have been
used to remove turbidity despite the biases introduced by
their use (Kennedy et al., 1974; Danielsson, 1982; Johnson
and Wangersky, 1985).
Recently, the Solid Waste Program of the U.S.E.P.A. issued
regulations that banned the field filtration of ground water
samples (40 CFR, 1993). The ban requires samples from
ground water monitoring systems be analyzed for the total
amounts of contaminants in unfiltered samples. If the
samples have been obtained using techniques that overly
stress the subsurface system, the resulting samples may be
highly turbid. Such samples, when analyzed, typically have
high contaminant concentrations.
The field research cited above has also demonstrated that
ground water samples relatively free of turbidity can be
obtained without resorting to filtering by withdrawing
ground water at relatively slow rates. Collecting ground
water samples following these low-stress protocols requires
an investment in equipment and time that some site
investigators contend that they cannot afford. It should also
be noted that low-stress protocols minimize the amount of
purge water that may need special handling and eliminate
time needed to conduct the field filtration step.
The goal of these subgroup discussions was to reach a
consensus on the best protocol (based on current scientific
knowledge) for obtaining ground water samples that are
representative of the actual mobile colloid load. As a
premise to these discussions, the attendees agreed that, for
certain low-solubility contaminants like radionuclides,
metals, and high-molecular weight hydrophobic organic
compounds, research has shown that colloid-facilitated
transport is significant. The discussions leading to
consensus on following four important topics will be
presented in this report:
1. Criterion for Adequacy of Well Purging
2. Distinguishing Dissolved and Colloid-Bound
Contaminants
3. Distinguishing Mobile Colloids and Artifactual
Colloids
4. Identifying Sites with Colloid-Facilitated Transport
Potential
Owing to the recent promulgation of the U.S.E.P.A. ban on
field filtering, the discussions often included debates over
the adequacy of collecting turbid ground water samples and
"fixing" the improperly collected samples by field filtering.
After summaries of the consensuses on the four topics listed
above, this report will address field filtering.
Criterion for Adequacy of Well Purging
A suspended sediment concentration criterion should be
established to obtain ground water samples that are
representative of the mobile colloid load. This criterion
should specify that the suspended sediment concentration
should reach a stable level during purging and prior to
sampling. This criterion should be applied in a manner
analogous with the monitoring of other parameters during
purging; i.e., the sample turbidity should be monitored
during purging and purging should continue until a stable
turbidity is achieved. The best means to achieve the stable
suspended sediment concentration is through pumping the
ground water from the well. Based on the scientific
evidence currently available, low-stress purging is
recommended. The best way to specify low-stress pumping
is to require minimal or no drawdown of the water level.
The scientific evidence presented at the workshop clearly
demonstrated that the suspended sediment concentration of
the sample decreases during purging to some stable level. It
is this stable suspended sediment concentration that is
-------
representative of the actual mobile colloid load. Careful
field sampling has shown that this stable colloid
concentration range from <1 to as high as 100 mg L"1 with
higher concentrations typically occurring in shallow,
unconfined aquifers subjected to changes in pore water
chemistry (McCarthy and Degueldre, 1993).
In the field, the decrease in suspended sediment
concentration toward a stable level can be effectively
monitored by some type of light scattering technique. A
turbidity meter is the most practical instrument to use for
this purpose. The extracted ground water can be routed
through a flow-through cell in the turbidity meter for
constant monitoring of the suspended sediment
concentration (measured in nephelometric turbidity units,
NTU) without exposing the sample to the atmosphere. The
turbidity meter can be added to the suite of instruments used
to monitor the stabilization of other chemical parameters
during purging (pH, dissolved oxygen, specific
conductance, etc.). Careful field tests have shown that these
chemical parameters typically stabilize more rapidly than
the suspended sediment concentration of the extracted
ground water (Puls et al., 1992; Backhus et al., 1993).
A stable colloid concentration is easily achieved by
withdrawing ground water using a pump positioned at a
fixed depth during purging and sampling. Careful field
studies have demonstrated that low-stress pumping produces
samples with stable suspended sediment concentrations that
are two to three orders of magnitude lower than those
produced by bailing in the same wells (Puls et al., 1992;
Backhus et al., 1993). The tests by Backhus et al. (1992)
showed that bailing continued to produce samples with high
turbidity even after 60 pore volumes of purging and that
bailed samples also contained up to 750 times greater
concentrations of high molecular weight poly cyclic
aromatic hydrocarbons than pumped samples from a coal
tar-contaminated site.
The careful field studies currently recommend that pumping
be performed at relatively slow withdrawal rates to
minimize suspension of "artifactual" colloids, or sediment
that is attached to sediments at natural ground water flow
velocities. Calculations have shown that a pumping rate of
about 100 mL min"1 will keep shear rates below those
believed to mobilize attached colloids in a typical
unconfined aquifer (Ryan, 1988). Rather than specify a
particular withdrawal rate, which is strongly dependent on
aquifer properties, a "minimal-" or "no-drawdown"
guideline appears to be the best way to assure that the
pumping-induced flow rates near the well were not
"stressing" the aquifer and causing artifactual colloid
mobilization. However, some field data shows that
pumping at a range of withdrawal rates has no effect on the
stable suspended sediment concentration (Puls et al., 1992).
In fact, at higher pumping rates, the stable level was reached
in a shorter purging time, although the higher pumping rates
produced larger colloids. Further research is needed to
elucidate the effect of pumping rate on the stable suspended
sediment concentration.
A particular concern of many of the attendees was sampling
from "borderline" aquifers ~ formations with low
permeability. Sampling and purging requirements are
difficult, if not impossible to meet, in such formations.
However, it is clear that bailing samples from these
formations will exacerbate turbidity problems, while low-
stress pumping ("minimal drawdown") produce samples of
lower turbidity. Also, there is a possibility that passive
sampling techniques (Magaritz et al., 1990) may prove
especially valuable in such formations.
Distinguishing Dissolved and Colloid-Bound
Contaminants
A representative sample used to determine the transport of
the contaminant should include both the mobile dissolved
and mobile colloidal fractions of the contaminant. For some
contaminants (e.g., volatile organic compounds), the mobile
colloidal fraction may be deemed insignificant a priori
based on published partitioning values. If the transport
behavior of the contaminant capable of significant
partitioning to colloidal phases is to be determined and
predicted, then it is always necessary to measure and
distinguish between the dissolved and colloidal fractions.
However, in a monitoring mode, where only the presence of
the contaminant is to be determined, it is not necessary to
distinguish between the dissolved and colloidal fractions;
only the total mobile contaminant concentration is needed.
Scientific evidence shows that certain low-solubility
contaminants are susceptible to colloid-facilitated transport
(McCarthy and Zachara, 1989). These low-solubility
contaminants are generally those that are surface-reactive;
i.e., they adsorb strongly to mineral and organic colloids. In
the field, colloid-facilitated transport has been observed for
radioactive (e.g., Pu, Am, U, Co, Sr, Cs) and non-
radioactive metal (Cu, Pb) in the cationic form (Means et
al., 1978; Buddemeier and Hunt, 1988; Magaritz et al.,
1990; Penrose et al., 1990). Laboratory experiments have
shown that the transport of metals and other inorganic
elements typically present in the anionic form (e.g., Cr(VI),
As) is facilitated by positively charged metal oxides.
Organic colloidal phases (NOM) also enhance the transport
of high molecular weight organic compounds (PAH,
pesticides, polychlorinated biphenyls, etc.) in laboratory
columns, but direct evidence of this phenomenon has not
been observed in the field.
For contaminants that do not fit into the categories listed
above, colloid-facilitated transport is unlikely. Once the
nature and abundance of the colloidal fraction is known,
89
-------
specific estimates of the potentially colloid-bound fraction
can be estimated based on known partition coefficients.
Given some reasonable safety margin, it is reasonable to
state that colloid-facilitated transport can be ruled out for
many low molecular weight, non-surface-reactive
contaminants. In these cases, it is not necessary to obtain
representative samples of the mobile colloidal fraction;
hence, many of the added precautions (turbidity monitoring,
low-stress pumping) recommended for sampling for colloids
may not be needed. However, the site investigator should
keep in mind an important caveat before dispensing with the
colloid sampling precautions: samples collected without
taking precautions to avoid nonrepresentative suspended
sediment can never be used to make judgements about the
transport of contaminants that may be colloid-associated.
The need for quantifying separate dissolved and colloidal
fractions of contaminants depends on the reason for
collecting the data. If the ground water sample has been
collected solely to monitor the containment of the
contaminant (e.g., a landfill), then the site investigator need
only be concerned with total mobile contaminant
concentrations without regard for the dissolved or colloidal
fraction. Note, however, that sampling techniques designed
for colloids still must be used for this measurement because
part of the total mobile contaminant may be colloid-bound.
If the groundwater sample has been collected to assess the
fate and transport of the contaminant, then it is essential to
quantify both the dissolved and colloidal fractions because
the transport of each fraction will be different.
Uncertainty still exists concerning the best method for
distinguishing between the dissolved and colloidal fractions
of a contaminant. The most common technique to isolate
the dissolved fraction is filtration. Currently, the 0.45 |j,m
pore size filters are needed to accurately isolate the
dissolved fraction. Improvements in the separation have
been made by using membrane filters with pore sizes
ranging from 1.0 down to 0.01 |j,m; however, these filters
are subject to rapid clogging. A more efficient means of
isolating the dissolved fraction is high molecular weight
ultrafiltration (10 to 100K nominal molecular weight
cutoff). Ultrafiltration must be performed in the laboratory;
thus, it requires more careful sample handling and storage
techniques. Centrirugation, another laboratory technique,
may also be useful in separating the dissolved and colloidal
fractions (Salbu et al., 1985).
Distinguishing Mobile Colloids and Artifactual Colloids
Collecting samples following the low-stress protocol is our
best assurance that we have obtained samples representative
of the truly mobile colloid population. However, it is still
necessary to ascertain that the colloids retrieved truly
mobile or artifactual. At many sites, this may not be
"readily" achievable. Detailed knowledge of the site
hydrogeochemistry and proper characterization of the
colloids are necessary to make this decision. Colloid
"veracity" can be determined by consideration of colloid
size, composition, possible origin, and the geochemistry of
"background" vs. "contaminated" samples.
Even after collecting ground water samples using techniques
designed to obtain representative dissolved and colloidal
fractions of the contaminant, the site investigator must judge
whether the colloids in the sample are truly mobile or not.
To do this, the site investigator must characterize the
colloids and gain thorough knowledge of the site
hydrogeochemistry. Without this thorough analysis of the
"veracity" of the colloids, the site investigator runs the risk
of accounting for colloid transport when, in fact, the colloids
are not truly mobile.
Colloid characterization techniques are considered esoteric
by all but the research community; however, many of the
analyses are more "routine" and less expensive than many
site investigators may realize. Such analyses include
scanning electron microscopy (colloid size, morphology,
and concentration), energy- or wave-dispersive x-ray
spectroscopy (elemental composition), photon correlation
spectroscopy and other light scattering techniques (size and
concentration), microelectrophoresis (surface charge), and
x-ray diffraction (mineralogy). Detailed methods for these
characterization techniques have been summarized in the
literature (McCarthy and Degueldre, 1993).
Knowledge of the colloid character is not sufficient; the site
investigator must seek to gain a thorough understanding the
hydrologic, geologic, and chemical character of the site.
This necessarily includes knowledge of (1) the direction and
velocity of the ground water flow, (2) the mineralogic
composition of the aquifer, (3) chemical composition of the
groundwater. Also, the site investigator must know how
these properties have been changed or been influenced by
the contaminant plume. This knowledge can be used to
assess the likelihood of mobile colloids in the samples. The
presence of colloids in the ground water must "make sense"
in the hydrogeochemical setting at the site. To make this
judgement, site investigators must become familiar with the
basics of colloid transport.
Key to this assessment is the presence of wells sampling
both "background" and "contaminated" portions of the
aquifer. The background and contaminated samples can be
used to explain the possible effect of the contaminant plume
on colloid mobilization. The geochemical changes wrought
by the advancing plume may mobilize colloids through
some change in the pore water chemistry; in these cases,
colloid transport may advance no further than the
contaminant plume which caused its mobilization. In such a
case, detailed understanding of the colloid transport may not
be necessary. Currently, field data is needed to substantiate
this phenomenon.
90
-------
Identifying Sites with Colloid-Facilitated Transport
Potential
Given a contaminant for which colloid-facilitated transport
is possible, the potential for colloid-facilitated transport
cannot be ruled out based on the hydrogeochemical
character of a site because colloids have been found
everywhere they have been sought. With our current
understanding of colloid mobilization and transport, it is
possible to identify geochemical conditions under which
colloid-facilitated transport would be likely.
Using careful sampling methods, colloids have been found
everywhere that researchers have looked for them
(McCarthy and Degueldre, 1993). The nature and
abundance of colloids varies widely from site to site;
however, colloids are apparently ubiquitous. Whether or not
the colloids present are capable of facilitating contaminant
transport is a question that must be addressed on a site-by-
site basis. At each site, we must consider whether the
mobile colloids are present at a sufficiently high
concentration and are capable of sufficiently strong binding
of contaminants to significantly enhance contaminant
transport. Based on the hydrogeochemical characteristics of
sites, we cannot rule out the possibility of colloid-facilitated
transport a priori.
We can identify combinations of aquifer mineralogy and
pore water chemistry where we expect colloid mobilization
to be significant. Two good examples of aquifers
susceptible to colloid mobilization have been presented in
the literature. The first involves the infiltration of anoxic,
organic matter-rich water (an analog for a landfill leachate)
into a sediment composed of quartz sand coated by ferric
oxyhydroxides and kaolinite. The infiltrating water
promoted the mobilization of the kaolinite colloids by
dissolution and reversal of the surface charge of the ferric
oxyhydroxide cement (Ryan and Gschwend, 1990; 1992).
The second involves the infiltration of acidic, carbon
dioxide-rich water into a carbonate-cemented aquifer. The
carbonate cement dissolved and released aluminosilicate
mineral colloids (Gschwend et al., 1990; Ronen et al.,
1992). Based on field observations such as these and
general knowledge of colloidal interactions, we can predict
that colloid mobilization may be a particular problem for
certain sites.
Bailing and Field Filtering
The general consensus of the Turbidity and Colloidal
Transport subgroup was that, based on scientific evidence,
bailing and field filtering do not provide samples that are
representative of the mobile colloid fraction. Nevertheless,
a number of the attendees, particularly those representing
state regulatory agencies, contended that E.RA. ban on field
filtering should be rescinded for detection monitoring at
solid waste facilities (subtitle D). These regulators were
solely concerned with ground water sampling for
monitoring purposes. They foresaw that the ban would
result in an unusually large number of "false positive"
analyses in samples collected by bailing, the most common
sampling method. Although the intent of monitoring is to
measure only the mobile form of the compound of interest,
analyses of unfiltered turbid samples would measure both
mobile and immobile forms of the compound of interest as
well as natural forms of the compound.
Metals leaching from landfills was the major concern of the
state regulators. Currently, many state regulations require
the routine monitoring of metals in samples from landfill
monitoring wells. Because metals interact with colloidal
mineral surfaces and organic matter, the importance of
accurately determining the truly mobile metal fraction and
the potential for colloid-facilitated transport is heightened.
Representatives from the states of Wisconsin and Michigan
were particularly concerned with highly turbid samples
taken from glacial till aquifers, which are common in those
states. Glacial till contain abundant clay that can be
mobilized by the high groundwater velocities induced by
bailing. It is thought that the metals adsorbed to the
mobilized clays are actually immobile in the aquifer, hence
the suspended sediment is currently filtered out using 0.45
[im filters to attempt to isolate the mobile fraction of the
metals. This filtrate is typically called the "dissolved"
fraction, although it is well known that (1) some dissolved
metals may become adsorbed to the suspended sediment
trapped on the filter and (2) some colloidal metal (associated
with colloids smaller than the pore size) may pass through
the filter.
The state of Washington representative was concerned with
the inability to distinguish between natural levels of certain
compounds and levels caused by contamination. Many
Washington ground waters and sediments contain high
levels of arsenic, making the distinction between natural
levels and contamination difficult. Bailed samples
containing abundant suspended sediments may contain
natural arsenic originally present in mineral lattices that is
released by sample acidification and analyzed as a
contaminant.
The state regulators agreed that current scientific evidence
shows that the low-stress pumping protocol for obtaining
metals samples is the best method. In defense of the bailing
and field filtering protocol, the state regulators (and others)
raised two issues: (1) cost and (2) consistency. The state
regulators asserted that most site investigators required to
monitor ground water cannot afford the equipment and time
required to sample ground water properly. They also point
91
-------
out that a change in sampling techniques will render their
extensive data bases inconsistent with newly-acquired
results.
Some members of the subgroup, particularly those from
academic and federal research laboratories, countered these
arguments by reiterating that (1) the perceived necessity of
field-filtering highlighted the fact that bailing was
inappropriate for collecting samples for metals and (2)
filtering introduces an unknown bias into metals analysis
(i.e., filtering could result in either under- or overestimation
of the mobile metal concentration). The researchers pointed
out that the cost of making decisions on poorly-collected
samples is potentially much greater than the cost of
sampling correctly. Following the same reasoning,
continuation of inappropriate sampling methods for the sake
of consistency is not a valid reason ~ if bad data were
collected before, that makes collection of good data even
more important.
Some of the researchers did not share the state regulators'
concern for "false positive" measurements of high metal
concentrations. High metal concentrations in a turbid
sample may simply indicate that metals adsorbed to the
sediments were pulled into the well by the vigorous bailing
action. These metals must have been mobile enough to
reach that point. The contention that high natural levels of
some metals obscures our ability to distinguish contributions
from leaking landfills was also not deemed a valid excuse to
continue poor sampling techniques (e.g., high As
concentrations in Washington). Increases in levels of metals
above natural levels should be evident from properly located
background wells. If such wells are not available, then it is
scientifically advisable to discontinue monitoring the
troublesome metal and substitute a more easily measured
parameter indicative of leachate (e.g., organic carbon,
chloride).
References
40 CFR Parts 257 and 258, Solid Waste Disposal Facility
Criteria; Final Rule. Federal Register 56 (196), 50978-51119
Backhus D.A., Ryan J.N., GroherD.M., MacFarlane J.K.,
and Gschwend RM. (1993) Sampling colloids and colloid-
associated contaminants in ground water. Ground Water 31,
466-479.
Buddemeier R.W and Hunt J.R. (1988) Transport of
colloidal contaminants in groundwater: radionuclide
migration at the Nevada Test Site. Appl. Geochem. 3, 535-
548.
Danielsson L.G. (1982) On the use of filters for
distinguishing between dissolved and paniculate fractions in
natural waters. Water Res. 16, 179-182.
Gschwend RM., Backhus D.A., MacFarlane J.K., Page A.L.
(1990) Mobilization of colloids in groundwater due to
infiltration of water at a coal ash disposal site. J. Contain.
Hydrol. 6, 307-320.
Johnson B.D. and Wangersky PJ. (1985) Seawater filtration:
Particle flow and impaction considerations. Limnol.
Oceanogr. 30, 966-971.
Kennedy V.C., Zellweger G.W., and Jones B.F (1974) Filter
pore-size effects on the analysis of Al, Fe, Mn, and Ti in
water. Water Resour. Res. 10, 785-790.
Magaritz M., Amiel A. J., Ronen D., and Wells M.C. (1990)
Distribution of metals in a polluted aquifer: A comparison of
aquifer suspended material to fine sediments of the adjacent
environment. J. Contam. Hydrol. 5, 333-347.
McCarthy J.F and Degueldre C. (1993) Sampling and
characterization of colloids and particles in groundwater for
studying their role in contaminate transport. In
Environmental Particles (eds. Buffle J. and van Leeuwen
H.P), Lewis Publishers, 247-315.
McCarthy J.R. and Zachara J.M. (1989) Subsurface
transport of contaminants. Environ. Sci. Technol. 23, 496-
502.
Penrose W.R., Polzer W.L., EssingtonE.H., Nelson D.M.,
and Orlandini K.A. (1990). Mobility of plutonium and
americium through a shallow aquifer in a semiarid region.
Environ. Sci. Technol. 24, 228-234.
Puls R.W., Clark D.A., Bledsoe B., Powell R.M., and Paul
C.J. (1992) Metals in groundwater: Sampling artifacts and
reproducibility. Haz. Waste Haz. Mater. 9, 149-162.
Ronen D., Magaritz M., Weber U., Amiel A. J., and Klein E.
(1992) Characterization of suspended particles collected in
groundwater under natural gradient flow conditions. Water
Resour. Res. 28, 1279-1291.
Ryan J.N. (1998) Groundwater Colloids in Two Atlantic
Coastal Plain Aquifers: Colloid Formation and Stability.
M.S. Thesis, Dept. Civil Eng., Massachusetts Institute of
Technology, Cambridge, MA.
Ryan J.N. and Gschwend P.M. (1990) Groundwater colloids
in two Atlantic Coastal Plain aquifers: Field studies. Water
Resour. Res. 26, 307-322.
Ryan J.N. and Gschwend P.M. (1992) Effect of iron
diagenesis on clay colloid transport in an unconfmed sand
aquifer. Geochim. Cosmochim. Acta 56, 1507-1521.
Salbu B., Bjornstad H.E., Lindstrom N.S., Lydersen E.,
92
-------
Brevik E.M., Rambaek J.P., and Paus P.E. (1985) Size
fractionation techniques in the determination of elements
associated with paniculate or colloidal material in natural
fresh waters. Talanta 32, 907-913.
93
-------
Sample Handling and Analysis
Karl F. Pohlmann, Charlita Rosal and Bert
Bledsoe
Objectives of Sampling Program
Program objectives must be sufficiently defined so that data
collection efforts are designed and carried out in a way that
will produce data appropriate for the needs of the program.
"Data quality objectives" are developed to define the types,
quality, and quantity of data required by the various aspects
of the program. In this sense, "data quality" addresses the
purpose(s) for which the data are to be used. Once these
data objectives are defined, appropriate sampling and
analytical methodology (protocol) are evaluated and chosen.
Examples of several types of programs that might have
different data quality objectives are detection and
assessment monitoring related to regulatory activities,
resource evaluation, and geochemical modeling.
Most decisions regarding sampling methodology are made
before field operations begin, however, flexibility must be
built into the sampling protocol to deal with unexpected or
changing conditions. The group leaned toward basing
decisions on well or site-specific conditions as opposed to
applying general rules governing all sites, regardless of site-
specific conditions.
Field Measurements
Three categories of field measurements were defined: in
situ, purging, and field analyses. "In situ measurements"
are absolute values of constituents of properties that,
because of their unstable nature, must be made under
conditions as close to in situ conditions as possible, usually
conducted in a well or in situ device. Examples include
temperature and dissolved gases. Note that project
objectives may not require the additional efforts required to
collect data of this resolution.
"Purging measurements" are conducted to evaluate the
progress or efficiency of monitoring well purging and are
usually conducted at the well head. Because these
measurements are made on samples that have been removed
from their native physicochemical environment, their
accuracy may be lower than for in situ measurements. Use
of a flow-through cell may make measurements easier but
doesn't necessarily improve data accuracy or precision.
Examples of purging parameters include temperature, EC,
pH, DO, turbidity, other field analytes, and lab analytes.
Measurements of the constituents of interest to the sampling
program are preferred but are not often practical. Note that
the sampling device may influence the quality of certain
parameters; such as DO and turbidity from bailed samples,
and temperature from many devices. Therefore, these
parameters may not be equally weighted as purging criteria.
The last value measured is often taken as representing
sample conditions, however, the accuracy of these
measurements must be evaluated to determine whether they
meet data quality objectives. Field evidence suggests that if
measurements are made in-line, turbidity should be
measured first to avoid potential particle settling in flow-
through cells for other measurements. However, some
turbidimeters may heat the ground-water leading to error in
temperature measurements. Also, the effect of flow-through
cells on pump discharge rate should be considered; they can
act as flow restrictors. When they are disconnected for
bottle filling (as recommended), pump discharge rates may
increase dramatically.
"Field analyses" are made for the purpose of on-site
characterization or are measurements of constituents and
properties considered too unstable for laboratory analysis.
Field-deployable analytical techniques (GC, GC/MS, XRF,
ion chromatography, immunochemistry, fiber optics) may
be useful for both detection and assessment but will require
the development of strong, standardized protocols for
quality control. Current use is primarily for site
characterization, but application of these techniques to
routine monitoring may reduce the costs and uncertainty
associated with sample collection, handling, and analysis,
particularly for aromatic hydrocarbons. Examples of
unstable parameters that are usually best measured at the
well head include pH, turbidity, DO, metals speciation (Fe
II, Cr IV), alkalinity, sulfide, nitrite, dissolved gases, and
electrode potential (for Eh). Note that holding times for
some of these parameters is 24 hours. Constituents that
require preservation for later analysis are those traditionally
analyzed in the laboratory, including major ion chemistry,
trace metals, organics, colloids, etc. Preservation
techniques are described below.
Field Filtration
The decision to collect filtered or unfiltered samples must be
based on data quality objectives and, therefore, the
94
-------
objectives of the overall program. Field-filtering of certain
constituents should not be the default, however.
Consideration should be given to what the application of
field-filtration is trying to accomplish. The group agreed
that specific reasons exist for collection of both filtered and
unfiltered samples, as outlined below.
Under conditions of excessive turbidity caused by certain
geologic or sampling conditions, field filtration is often
suggested to reduce artifacts caused by the presence of
excessive loads of suspended sediments that threaten to bias
the analyses. However, potential implications on sample
accuracy and overall data quality must be recognized and
evaluated. For example, filtration with 0.45 |jm filters may
remove particles important to transport of colloidally-
associated contaminants (both larger than the initial pore
size and smaller, as the filter cake builds up). It was
generally agreed that field-filtration should not be used to
correct for improperly designed or constructed monitoring
wells, inappropriate sampling methods, or poor sampling
technique. The objective should be to carry out all aspects
of sampling so as to produce a sample that is minimally
disturbed. Turbidity may be a good criterion to evaluate
sample disturbance and the question was raised as to
whether turbidity guidelines could be developed.
Truly dissolved concentrations of major ions and trace
metals are used for flow system analysis and equilibrium
geochemical modeling. Therefore, samples collected for
these purposes must be filtered because inclusion of
paniculate material in the analyses may lead to erroneous
dissolved concentrations that will impact geochemical
equilibrium calculations. Alkalinity samples must also be
filtered if significant paniculate calcium carbonate is
suspected. The presence of this material is likely to impact
the alkalinity titration although filtration itself may alter the
CO2 composition of the sample and therefore affect the
results. Field filtration may also be conducted to maintain
consistency with historic data collected with routine field-
filtration of samples. This approach may be appropriate in
certain situations (e.g. flow system analysis) but the
determination must be made as to whether filtration is
appropriate to program objectives. Samples for analysis of
colloid composition should be collected on membrane
filters or by field ultrafiltration (Backhus et al., 1993).
Although filtration may be necessary in the cases suggested
above, filtration of a sample may cause a number of
unintended changes to occur, possibly leading to filtration-
induced artifacts during sample analysis and uncertainty in
the results (Horowitz et al., 1992). Some of these
unintended changes may be unavoidable but the factors
leading to them must be recognized. Also, their effects can
be minimized by the consistent application of certain
filtration guidelines. These issues, which include filter type,
media, pore size, and others, must be addressed to identify
and minimize potential sources of uncertainty when filtering
samples.
In-line filtration was generally considered the most desirable
approach because it allows better consistency through less
sample handling, and minimizes sample exposure to the
atmosphere. In-line filtration can be accomplished through
the use of disposable filter cartridges or membrane filters in
an in-line filter apparatus. Disposable filter cartridges have
the advantage of greater sediment handling capacity when
compared to traditional membrane filters. The filter media,
including materials and pore structure may impact break-
through of certain metals. To obtain truly dissolved
constituents, the smaller pore size the better. The smallest
pore size practical in the field seems to be 0.1 [im, but the
choice depends on particle loads, required sample volume,
and project objectives. Also, the commercial availability of
filter pore sizes less than 0.45 |jm is limited.
Conditioning is required to remove preservatives, wetting
agents, residues of filter manufacturing, and leachable
compounds from filter media. Volumes of sample water of
500 mL to 1 L are generally accepted, but the volume
necessary depends on sample turbidity. If turbidity is high,
the filters may have to be preconditioned with DI water. For
analysis of very low metal concentrations, the filters may
require acid-washing. Once filtration has begun, a filter
cake may develop as particles larger than the pore size
accumulate on the filter membrane. The result is that the
effective pore diameter of the membrane is reduced and
particles smaller than the stated pore size are excluded from
the filtrate. Possible solutions include prefiltering,
minimizing particle loads to begin with, and reducing
sample volume. Finally, an inert environment may be
necessary during off-line filtration to minimize oxidation of
highly reduced species and may be accomplished through
the use of oxygen-free sample bottles in a glove box. For
in-line filtration, oxygen-free bottles alone will suffice.
If the objective of the sampling program is total recoverable
constituents, whether they be major ions or trace metals,
then the samples must not be filtered. In particular, samples
for the analysis of colloid-associated constituents must be
collected unfiltered. Likewise, samples for certain classes
of analytes are generally collected unfiltered, such as many
organics and nutrients. Particle load may be immaterial to
analysis of certain types of organic compounds (Paul et al.,
in preparation).
Collecting unfiltered samples eliminates many of the
uncertainties associated with field-filtration but should not
be attempted without addressing the issue of sample
collection. For example, sampling methods that disturb the
sampling zone and sample may suspend normally immobile
95
-------
particles and entrain them in the sample (e.g. bailers and
pumps at high flow rates) leading to analysis of naturally-
occurring matrix metals and "false positive" results. This is
a particular problem in turbid wells, where excessive
turbidity may result from improper design, construction, and
development. The following approaches may be useful in
minimizing potential sources of uncertainty when collecting
unfiltered samples.
Well design, construction, and development must be
appropriate to hydrogeochemical conditions present at the
site. Disturbance of the geologic matrix during drilling
must be minimized to reduce the effects of well construction
artifacts. When purging and sampling, low pumping rates
(300 mL/min and below) have been demonstrated to
minimize disturbance of the sample zone and sample when
compared to high pumping rates and bailing. The key is to
minimize particle concentrations in the sample thereby
minimizing artifacts from well design, construction, and
installation; purging; and sampling. However, the pump
intake position must be carefully chosen when using low
pumping rates because water drawn into the pump
originates in a relatively small vertical segment of the well
intake. If the pump intake is positioned away from the zone
of contamination, dilution within the wellbore may prevent
detection of the contaminant of interest. A borehole velocity
survey may provide indication of most conductive zones;
other techniques may also be applied to detect contaminated
zones. The measurement of selected field parameters during
purging, particularly in-line measurements of DO and
turbidity, provides an indication of the progress of purging
and when parameter equilibrium conditions have been
attained. Other parameters may lead to underpurging while
turbidity may be too conservative. Disturbance of the
sampling zone may be minimized, in turn reducing purge
volumes, through the use of dedicated sampling systems
rather than portable systems.
Finally, compatibility of unfiltered samples with the
analytical method must be considered, especially if the
samples contain significant quantities of suspended
particles. Also, make sure the lab doesn't filter the sample
that was painstakingly collected unfiltered in the field!
Preservation of Samples
It should be recognized that potential uncertainty associated
with sample preservation may be reduced or eliminated for
certain analytes by conducting the analyses in the field.
Although routine field analysis is not currently widespread,
it appears that present trends are leading in that direction.
For those analyses conducted in the laboratory, there are
several important issues that must be addressed to ensure
sample integrity is preserved from the field to the lab.
The transfer of the sample from the sampling device (bailer,
pump tubing) to bottle (or filtration device if filtration is
conducted off-line) must be accomplished with as little
disturbance to the sample as possible. This operation is
relatively straightforward when using low-speed pumps but
can be more problematic when using bailers , although
several techniques are currently available for simplifying the
transfer and reducing agitation of samples from bailers. If
field filtration is conducted, the operation must be carried
out in such a way that sample integrity is preserved, as
discussed above. To minimize oxidation of highly reduced
species, an inert environment may be necessary during off-
line filtration.
Chemical preservation is used to minimize reactions such as
precipitation of metal oxides and hydroxides and
biodegradation of organic compounds. Chemicals may also
be used to control a specific reaction; such as addition of
sulfuric acid to samples collected for ammonia
determination to lower the pH and form the stable
ammonium ion. These preservation techniques have
evolved over many years of laboratory analyses and appear
adequate in most cases (see SW-846). However, questions
remain about preservation of samples collected for analysis
of volatile organic compounds. Recommended chemical
preservatives include HC1, sodium bisulfate, sulfuric acid,
and mercuric chloride. These substances have proven
effective for preservation, particularly for aromatic
hydrocarbons, which are highly susceptible to
biodegradation, greatly increasing holding times past the
standard 14 days (Maskarinec et al., 1990; Roe et al., 1989).
Increasing the holding times of chemically-preserved
organic samples might reduce project costs associated with
resampling when holding times were not met. On the other
hand, some studies suggest that if not chemically preserved,
samples for volatile aromatic hydrocarbons may biodegrade
in very short time periods (Roe et al., 1989).
Physical preservation includes temperature of the samples
during transport to the laboratory and type of sample
containers. Maintenance of sample temperature at 4°C may
be critical to control biodegradation in organic samples that
are not chemically preserved. However, the question was
raised whether ice chests arrive at the laboratory at the
required temperature, and what actions are taken if the
samples are warmer. In contrast, SW-846 does not specify
cooling of samples collected for most metal analyses, but
this is a common practice in the field. Under certain
conditions, cooling of metal samples may lead to
precipitation of certain constituents. For very sensitive
analyses, maintaining the samples near in situ ground-water
temperature may be a better approach. Questions were
raised as to the effect of collecting samples in bottles that
may have been stored on ice (for convenience) and therefore
96
-------
are much cooler than ground water. Likewise, concerns
were expressed about the effects on VOC concentrations of
filling VOA vials that had been warmed by exposure to the
sun.
The types of sample containers appropriate for a wide range
of analytes are given in SW-846. Because organic
compounds may diffuse through plastic bottles and caps,
glass bottles and PTFE-lined caps are specified for samples
collected for organic analyses. Though plastic bottles and
caps are considered acceptable for many inorganic analyses,
the use of glass bottles minimizes exchange of gases with
the atmosphere that might lead to sample degradation of
very sensitive constituents (e.g., oxidation of highly reduced
metals species). Likewise, stoppers designed to eliminate
headspace may also be necessary to minimize this sample
degradation. Other physical preservation requirements are
given in SW-846.
Reducing the volume of samples sent to the laboratory can
minimize some of the problems discussed above under
filtration. Other advantages of reducing sample volumes
include increased ability to collect samples in low-yield
wells, lower bottle filling times at low pumping rates, lower
bottle costs, lower shipping costs, and reduced sample
disposal problems at the laboratory. Sample volume can
easily be reduced over many present practices. For
example, a full suite of metals can easily be analyzed from a
50 to 100 mL sample, and only 5 mL is required for purge
and trap GC analysis of organics.
Shipment of samples to the laboratory remains
controversial. Sample preservation techniques and holding
times would suggest that certain samples (e.g., metal) could
be shipped via conventional ground methods. However, this
method causes samples from the same sampling event to be
"split up" since VOC samples are commonly shipped by
overnight carrier, and might possibly lead to chain-of-
custody problems.
Documentation
There was clear consensus that more and better
documentation is required at virtually every step of the
sampling process in order to better interpret data collected
(field and lab) and to aid identification of areas where data
quality objectives are not being met. Discussion of general
topics follow.
Detailed descriptions of all methods, activities, and data are
critical. Examples include field conditions (well, weather)
that might be significant to the results, personnel, purging
methods (device, depth, rate, indicator parameters),
sampling methods (device, depth, rate), field measurement
instrumentation (type, methods, calibration data), specific
preservation information (filtration, chemical, physical), and
shipping information. Standardization of documentation
(data forms) at some level was discussed. Generic
"templates" of various forms could be developed and
adapted for widespread or project-specific use. More
widespread and uniform documentation is facilitated by
standardizing data records. Likewise, detailed,
comprehensive forms prompt for information that might
otherwise go unrecorded.
Automation of documentation is a future direction and has
some very appealing advantages. For example, permanent
bar codes applied directly to sample bottles in the field
might streamline sample tracking from the field to the lab
and into the final report. Personal Data Assistants may
provide an effective means for field data entry and storage
(for example all purging data could be recorded directly). A
copy of the data diskette sent to the laboratory with the
samples would provide laboratory personnel with a better
understanding of conditions under which the samples were
collected, and how they have been preserved.
Communication between field and lab personnel must be
open and proceed in both directions. As examples, lab
personnel should be fully aware of field conditions and
measurements, while field personnel should be cognizant of
laboratory requirements and the results of QC sampling.
Sending field notes with samples may facilitate this
communication.
Field Operations
Virtually all hydrogeologic and hydrochemical analyses are
based on information that is either measured in the field or
analyzed from samples collected in the field. As a result, it
is imperative that field personnel be qualified to carry out
the tasks required to ensure that data are of the quality
required to meet program objectives. At a minimum,
structured training of field personnel should provide an
understanding of project objectives, background in the
operating principles of field instrumentation and other
equipment, overview of the principles of sample
preservation, and guidance in documenting the sampling
event. A certification program might be initiated if certain
minimum qualifications for field personnel could be
developed. Since certification programs exist for analytical
laboratories, it seems reasonable that some type of
certification be instituted at the point where the data
originates. Regular audits of field operations might also be
useful for ensuring that appropriate procedures are followed
to meet established data quality objectives (QA/QC
program).
Research Needs
1. Although there was general agreement that
minimizing disturbance of ground-water samples
97
-------
provides higher quality data, questions remained
about the steps that must be taken to achieve
significant improvement in results. In particular,
there was some concern expressed as to whether
low-flow pumping/unfiltered samples actually
provided results that were of significantly higher
quality than more traditional approaches and
whether the improvements in results justified the
costs associated with major changes in sampling
protocol.
2. There was also a general interest in improving our
understanding of colloid-facilitated transport in
different hydrogeologic environments. Questions
such as How mobile are colloids? How far do they
travel? What are they composed of? and What
types of contaminants are likely to be associated
with colloids? were raised. In reality, many of
these questions are currently being addressed in the
literature. The overriding question was Are
colloids actually important to contaminant
transport at solid waste landfills, or is their role in
facilitating contaminant transport rather
insignificant? What might be needed to answer
this question is a comprehensive survey of the
colloid-transport literature to complement the work
that comes out of Group 4 (Turbidity and Colloid
Transport).
Guidance Needs
Several important issues were raised that may warrant the
development of technical guidance or recommendations and
are summarized below.
1. The important potential and increased use of field-
deployable analytical techniques requires the
development of strong, standardized protocols for
quality control.
2. Standardized turbidity guidelines should be
developed to assist evaluation of sample
disturbance.
3. The issues of sample holding times, chemical
preservation of certain classes of organic samples,
and sample shipment methods could be revisited to
update current practices.
4. Documentation could be improved and made more
uniform through development of standard data
forms, templates, etc.
5. There was concern about the inappropriate
application of low-flow techniques, particularly in
monitoring wells with long screens. If the pump
intake is positioned away from the zone of
contamination, dilution within the wellbore may
prevent detection of the contaminant. A borehole
velocity survey may provide indication of the most
conductive zones but may be impractical and
doesn't detect the contaminated zones. Therefore,
alternatives should be investigated and reported.
References
1. Backhus, D.A., J.N. Ryan, D.M. Groher, J.K.
MacFarlane, and P.M. Gschwend. Sampling
Colloids and Colloid-Associated Contaminants in
Ground Water. Ground Water, 31 (3): 466-479,
1993.
2. Horowitz, A.J., K.A. Elrick, and M.R. Colberg.
The Effect of Membrane Filtration Artifacts on
Dissolved Trace Element Concentrations. Wat.
Res., 26 (6): 753-763,1992.
3. Maskarinec, M.R, L.H. Johnson, S.K. Holladay,
R.L. Moody, C.K. Bayne, and R.A. Jenkins.
Stability of Volatile Organic Compounds in
Environmental Water Samples during Transport
and Storage. Environ. Sci. Technol., 24 (11): 1665-
1670, 1990.
4. Roe, V.D., M.J. Lacy, J.D. Stuart and G.A.
Robbins. Manual Headspace Method to Analyze
for the Volatile Aromatics of Gasoline in
Groundwater and Soil Samples. Anal. Chem, 61:
2584, 1989.
98
-------
Appendix A - Steering Committee
and Other Attendees
A-l
-------
Appendix A - Steering Committee
Puls, Robert W.
U.S. EPA/RSKERL
919 Kerr Research Drive
P.O. Box 1198
Ada, OK 74820
405/436-8543
405/436-8529 (fax)
Barcelona, M. J.
Western Michigan University
1007 Everett Tower
Kalamazoo, MI 49008-5150
616/387-5501
616/387-4988 (fax)
Bledsoe, Bert
U.S. EPA/RSKERL
919 Kerr Research Drive
P.O. Box 1198
Ada, OK 74820
405/436-8605
405/436-8529 (fax)
Brown, James R.
U.S. EPA/OSW (5303W)
401M Street, SW
Washington, DC 20460
703/308-8656
703/308-8617 (fax)
Denne, Jane
AMD/AMW
U.S. EPA/EMSL-LV
P.O. Box 93478
Las Vegas, NV 89193-3478
702/798-2655
Mangion, Steve
U.S. EPA/Region V
HSRLT-5J
77 W. Jackson Blvd.
Chicago, IL 60604
312/353-7499
312/353-9176 (fax)
McCarthy, John F.
Environmental Sciences Div.
Oak Ridge National Lab
P.O. Box 2008
Oak Ridge, TN 37831-6036
615/576-6606
Rich Steimle
U.S.EPA/TIO(5102W)
401M Street, SW
Washington, DC 20460
703/308-8846
703/308-8528 (fax)
Willey, Richard
U.S. EPA/Region I
HSS-CAN7
JFK Federal Bldg.
One Congress Street
Boston, MA 02203
617/573-9639
617/573-9662 (fax)
A-2
-------
Other Attendees
Aller, Linda
Geodyssey
2700 E. Dublin-Granville Road
Suite 530
Columbus, OH 43231
614/882-4991
614/882-4260 (fax)
Backhus, Debera
Indiana University
School of Public and Environmental Affairs
10th St. and Fee Lane,Room 425
Bloomington, IN 47405
812/855-4944
812/855-7802 (fax)
Bourbon, John
U.S. EPA/Region II
2890 Woodbridge Ave.
Edison, NJ 08837-3679
908/321-6729
908/321-6788 (fax)
Clark, Don A.
U.S. EPA/RSKERL
919 Kerr Research Drive
P.O. Box 1198
Ada, OK 74820
405/436-8562
405/436-8529 (fax)
Connelly, Jack
Bureau of Solid & Hazardous Waste
Dept. of Natural Resources
101 S. Webster Street
Box 7921
Madison, WI 53707
608/267-7574
608/267-2768 (fax)
Franks, Charles
U.S. EPA/Region I (HER-CAN6)
JFK Federal Bldg.
One Congress Street
Boston, MA 02203
617/573-9678
617/573-9662 (fax)
Gardner, Steven P.
U.S. EPA/ORD/EMSL-LV
944 E. Harmon Ave.
Las Vegas, NV 89119-6748
702/798-2580
702/798-2692 (fax)
Gibs, Jacob
U.S. Geological Survey
810 Bear Tavern Road,Suite 206
West Trenton, NJ 08628
609/771-3977
Gray, John D.
New York State Electric and Gas Company
Corporate Drive, Kirkwood Industrial Park
P.O. Box 5227
Binghampton, NY 13902-5227
607/762-8879
607/762-8457 (fax)
Gronwald, Keith
New York State DEC
50 Wolf Road
Albany, NY 12233-4013
518/457-1860
518/485-7733 (fax)
Hall, Steve
U.S. EPA/Region IV
960 College Station Road
Athens, GA 30605
706/546-3173
Korte, Nic E.
Oak Ridge National Lab
P.O. Box 2567
Grand Junction, CO 81502
303/248-6210
303/248-6147 (fax)
Lee, Roger W.
U.S. Geological Survey-WRD
8011 Cameron Road
Austin, TX 78754
512/873-3023
512/873-3090
A-3
-------
Mabey, William R.
Montgomery Watson
365 Lennon Lane
Walnut Creek, CA 94598
510/975-3521
510/975-3412 (fax)
Martin, Jr., William A.
Florida Dept. of Environmental Protection
2600 Blair Stone Road
Twin Towers Office Bldg.
Tallahassee, FL 32399-2400
904/921-0190
904/922-4939 (fax)
McKay, Larry D.
Dept. of Geological Sciences
University of Tennessee
306 G&G Bldg.
Knoxville, TN 37996-1410
615/974-0821
615/974-2368 (fax)
Nielsen, David
Nielsen Ground-Water Science
4686 State Road 605 South
Galena, OH 43021
614/965-5026
Park, Natalie
Westinghouse Savannah
River Company
Bldg. 773-57A
Savannah River Site
Aiken, SC 29808
803/725-5923
803/725-8041 (fax)
Parker, Louise
USACRREL
72 Lyme Road
Hanover, NH 03755
603/646-4393
603/646-4640 (fax)
Paul, Cynthia
U.S. EPA/RSKERL
919 Kerr Research Dr.
P.O. Box 1198
Ada, OK 74820
405/436-8600
405/436-8529 (fax)
Pohlmann, Karl
Desert Research Institute
P.O. Box 19040
Las Vegas, NV 89132-0040
702/895-0485
702/895-0427 (fax)
Powell, Robert M.
ManTech Environmental Technology
919 Kerr Research Drive
P.O. Box 1198
Ada, OK 74820
405/436-8676
405/436-8501 (fax)
Reece, Dennis
Woodward-Clyde Consultants
P.O. Box 66317
Baton Rouge, LA 70896
504/751-1873
504/753-3616 (fax)
Rightmire, Craig
Oak Ridge National Lab
P.O. Box 2008
Oak Ridge, TN 37831-6036
615/574-7285
Robbins, Gary
Dept. of Geology & Geophysics
University of Connecticut
U-45, Room 207
345 Mansfield Street
Storrs, CT 06269-2045
203/486-1392
203/486-1382 (fax)
Romero, Nadine
Washington Dept. of Ecology
P.O. Box 47600
Olympia, WA 98504-7600
206/407-6116
206/407-6102 (fax)
Ronen, Daniel
Weizmann Institute of Science
76100Rehovot
Israel
972-8-342546
972-8-344124 (fax)
A-4
-------
Rosal, Charlita
U.S. EPA/EMSL-LV
P.O. Box 93478
Las Vegas, NV 89193-3478
702/798-2179
702/798-2692 (fax)
Ryan, Joseph N.
University of Colorado
Campus Box 428
Boulder, CO 80309
303/492-0772
303/492-7317 (fax)
Sara, Martin
Rust Environment & Infrastructure
1240 E. Diehl Road
Neighborville, IL 60563
708/955-6686
Sridharan, Lakshmi
Bureau of Solid & Hazardous Waste
Dept. of Natural Resources
101 S. Webster Street Box 7921
Madison, WI 53707
608/266-0520
608/267-2768 (fax)
Stelz, William G.
U.S. EPA/OMMSQA (8205)
401M Street, SW
Washington, DC 20460
202/260-5798
202/260-4346 (fax)
Taylor, Allan B.
Michigan Dept. of Natural Resources
Work Management Division
P.O. Box 30241
Lansing, MI 48909
517/373-4799
517/373-4797 (fax)
Teplitzky, Andrew L.
U.S. EPA/OSW (5306)
401 M Street, SW
Washington, DC 20460
202/260-4536
White, Carol
Sevee & Maher Engineers, Inc.
4 Blanchard Road
P.O. Box 85A
Cumberland Center, ME 04021
207/829-5016
207/829-5692 (fax)
Zavala, Bernie
U.S. EPA/Region X
1200 Sixth Ave.
Seattle, WA 98101
206/553-1562
206/553-0119 (fax)
A-5
-------
Appendix B -
Questions to Consider
for
Small Group Discussions
B-l
-------
Monitoring Goals and Objectives
1. What are the objectives of the monitoring programs?
2. To what extent do/should the program objectives, site characteristics, or constituents of concern
provide criteria for representativeness?
3. Can we identify purpose and objective-driven sampling designs if monitoring networks are to be
used for different purposes over the life of the networks (e.g., detection networks phased into
assessment networks which may later be phased into corrective action networks)?
4. Given a sampling protocol, will the data collected meet monitoring objectives? What accuracy and
sensitivity will the sampling and analysis protocol provide?
5. Can field screening techniques be applied equally to detection and assessment monitoring pro-
grams? How can we provide performance criteria for field analyses that are credible and reliable?
6. Knowing that contaminant concentration values are dependent upon how the sample is collected, its
rate of extraction, and the point where it's drawn - how do these factors influence monitoring objec-
tives?
7. What effects do/should the physical/chemical speciation of contaminants play in the design of a
monitoring program?
8. What alternative methods are available for designing monitoring well networks?
9. Do we need better documentation of well construction and well development?
B-2
-------
Well Design, Construction, Development
1. What artifacts are introduced by well design, construction, and development? How do these arti-
facts compare in magnitude to sampling procedure errors?
2. What procedures/criteria should be followed to reduce the artifacts associated with well design,
construction, and development? Can general guidelines be recommended? What existing guidance
documents are avilable for reference? Are there needs for new guidance?
3. What are reliable/effective monitoring well development techniques and criteria?
4. Can guidance on well design be developed for specific hydrogeologic settings (i.e., grain size/
sorting/uniformity coefficient/hydraulic conductivity ranges)?
5. Can we provide criteria for properly designed wells without identifying some grain size/sorting/
uniformity coefficient/hydraulic conductivity ranges for specific types of installations?
B-3
-------
Well Purging and Sampling
1. Why do we purge wells? Are the reasons/justification the same for dedicated and non-dedicated
sampling systems?
2. How should wells be purged? What should be the endpoints of sufficient purging? Do we need
better documentation of how wells are purged? Is there a difference between different purging
methods?
3. What are the advantages and disadvantages of low-flow purging and sampling techniques?
4. Is one set of criteria of endpoints sufficient to define "acceptable" purging and sampling for all
cases, or will endpoints of purging be site-specific? (i.e., do hydrogeological conditions affect how
we should purge and sample, e.g., unconsolidated formations, fractured rock, low-yielding wells,
Karst solution cavities). How does purging and sampling affect the spatial resolution of informa-
tion on contaminant distribution in different hydrogeological environments?
5. Can/should we formulate a short list of recommended types of sampling devices? What are the
trade-offs of dedicated vs. non-dedicated pumps? Are bailers useful and under what conditions?
B-4
-------
Turbidity & Colloidal Transport
1. Is it advisable to have a turbidity criteria for well purging, and if so what would it be?
2. Does a representative sample always include the mobile solute and mobile colloids? Is it always
necessary to distinguish between dissolved and colloid-bound contaminants?
3. Can we readily distinguish between truly mobile colloids and immobile solids or artifacts solids
from sampling?
4. Can we identify hydrogeochemical characteristics for sites where colloidal mobility could be a
significant transport mechanism?
B-5
-------
Sample Handling and Analysis
1. How does filtering/not filtering produce "representative" samples and/or address the objectives of
the sampling program?
What is "dissolved"? What is "mobile"? What is "total"?
2. If filtration is deemed necessary, how can sampling/filtering-induced error (artifacts) be reduced?
3. If a decision is made to not filter, what approach is required to reduce error, particularly with
regards to sample collection?
4. What ground-water constituents must be (or are easily) analyzed in the field and what constituents
require preservation for later analysis?
5. What must be done at the time of sample collection to best preserve samples for later analysis in the
laboratory?
6. Do we need better documentation of sample handling and preservation procedures used?
B-6
-------
Appendix C -Bibliography
c-i
-------
Appendix C - Bibliography
Aller, L., T.W. Bennett, G. Hackett, RJ. Petty, J.H. Lehr, H.
Sedoris, D.M. Nielsen and J. Denne. 1989. Handbook for
the Suggested Practices for the Design and Installation of
Ground-Water Monitoring Wells, National Water Well
Association, Dublin, Ohio, 398 pp. EPA/600/4-89-034,
NTIS PB 90-159807.
American Society of Testing and Materials (ASTM). 1981.
Standard Specification for Thermoplastic Water Well Casing
Pipe and Couplings Made in Standard Dimension Ratios
(SDR). F-480,1987 Annual Book of ASTM Standards,
Philadelphia, pp. 1028-1033.
American Society of Testing and Materials (ASTM). 1986.
Standard Specification for Poly(Vinyl Chloride) (PVC)
Plastic Pipe, Schedules 40, 80, and 120. D1785, 1987
Annual Book of ASTM Standards, Philadelphia, pp. 89-101.
American Society for Testing and Materials (ASTM). 1990.
Standard Practice for Decontamination of Field Equipment
Used at Non-Radioactive Waste Sites #D5088-90.
Philadelphia, Pennsylvania. American Society for Testing
and Materials, pp. 1078-1080.
American Society for Testing and Materials(ASTM). 1994.
Ground Water and Vadose Zone Investigations,
Philadelphia, Pennsylvania, 396 pp.
American Society for Testing and Materials (ASTM). 1994.
Specification for Thermoplastic Well Casing, Pipe and
Couplings, Made in Standard Dimension Ratio (SDR), SCH
40, SCH 80, Standard F-480-94, Vol. 8.04, Philadelphia,
Pennsylvania, 25 pp.
American Society for Testing and Materials Standard
D5092-90, Standard Practice for Design and Installation of
Ground Water Monitoring Wells in Aquifers, Pennsylvania,
12pp.
American Society of Testing and Materials (ASTM)
Subcommittee D18.2105 on Design and Installation of
Ground-Water Monitoring Wells. 1989. Draft Standard,
Proposed Recommended Practice for Design and
Installation of Ground Water Monitoring Wells in Aquifers.
ASTM, 47 pp.
Backhus, D.A., J.N. Ryan, D.M. Groher, J.K. MacFarlane,
and P.M. Gschwend. 1993. Sampling Colloids and Colloid-
Associated Contaminants in Ground Water. Ground Water.
31(3): 466-479.
Backhus, D.A. and P.M. Gschwend. 1990. Use of
fluorescent polycyclic aromatic hydrocarbon probes to study
the impact of colloids on pollutant transport in groundwater.
Environ. Sci. Technol. 24:1214-1223.
Baedecker, N.J. and W. Back. 1979. Hydrogeochemical
processes and chemical reactions at a landfill. Ground
Water. 17(5):429-437
Barcelona, M.J., 1984. TOC determinations in ground
water. Ground Water, 22(1): 18-24.
Barcelona, MJ. 1988. Overview of the Sampling Process, in
Principles of Environmental Sampling, Keith, L.H., Ed.,
ACS Professional Reference Book, American Chemical
Society, Washington, D.C., Chapter 1.
Barcelona, M.J., O.K. George and M.R. Schock, 1988.
Comparison of Water Samples form PTFE, PVC and SS
Monitoring Wells, United States Environmental Protection
Agency, Office of Research and Development,
Environmental Systems Monitoring Laboratory, Las Vegas,
NV, EPA 600/X-88/091, 37 pp.
Barcelona, M.J., and J.P. Gibb. 1988. Development of
Effective Ground-Water Sampling Protocols. In: Collins,
A.G. and Johnson, A.I., 1988. pp. 17-26.
Barcelona, M.J., J.P. Gibb, J.A. Helfrich and E.E. Garske.
1985. Practical Guide for Ground-Water Sampling. EPA/
600/2-85/104,169 pp.
Barcelona, M.J., J.P. Gibb and R. Miller. 1983. A Guide to
the Selection of Materials for Monitoring Well Construction
and Ground-Water Sampling, Illinois State Water Survey,
SWS Contract Report 327, Champaign, Illinois, 78 pp.
Barcelona, M.J. and J.A. Helfrich. 1986. Well construction
and purging effects on ground-water samples. Environ. Sci.
& Technol. 20(11): 1179-1184.
C-2
-------
Barcelona, MJ. and J.A. Helfrich. 1988. Laboratory and
Field Studies of Well-Casing Material Effects, Proceedings
of the Ground Water Geochemistry Conference, National
Water Well Association, Dublin, Ohio, pp. 363-375.
Barcelona, MJ. and J. A. Helfrich, 1992. Realistic
Expectations for Ground Water Investigation in the 1990's,
in Current Practices in Ground-Water and Vadose Zone
Investigations. Nielson, D. M. and Sara, M. N., Eds.,
American Society for Testing and Materials. Philadelphia,
PA, 431pp.
Barcelona, M.J., J.A. Helfrich and E. E. Garske. 1985.
Sampling tubing effects on ground-water samples.
Analytical Chemistry, 57(2):460-464.
Barcelona, M.J., J.A. Helfrich, andE.E. Garske. 1988.
Verification of Sampling Methods and Selection of
Materials for Ground-Water Contamination Studies. In:
Collins, A.G., and Johnson A.I. pp. 221-231.
Barcelona, M.J., J.A. Helfrich, E.E. Garske, and J.P. Gibb.
1984. A laboratory evaluation of groundwater sampling
mechanisms. Ground Water Monitoring Review. 4(2): 32-41.
Barcelona, M.J., T.R. Holm, M.R. Schock and O.K. George.
1989. Spatial and temporal gradients in aquifer oxidation-
reduction conditions. Water Resources Research, 25(5):
991-1003.
Barcelona, M.J., J.F. Keely, W.A. Pettyjohn, and A.
Wehrmann. 1987. Handbook: Ground Water. EPA/625/
687/016, 212 pp.
Barcelona, M.J., H. A. Wehrmann, andM.D. Varljen. 1994.
Reproducible well purging procedures and VOC stabilization
criteria for ground-water sampling. Ground Water. 32(1)12-
22.
Barnes, I. 1964. Field Measurement of Alkalinity and pH.
U.S. Geological Survey Water-Supply Paper 1535-H. 11 pp.
Benson, R.C., M. Turner, P. Turner, and W. Vogelsong.
1988. In Situ, Time Series Measurements for Long-Term
Ground-Water Monitoring. In: Collins, A.G., and Johnson,
A.I. 1988. pp. 58-72.
Bianchi-Mosquera, G.C. andD.M. Mackay. 1992.
Comparison of stainless steel vs. PTFE miniwells for
monitoring halogenated organic solute transport. Ground
Water Monitoring Review. 12(4):126-13 1.
Birgersson, L., and I. Neretnieks, Diffusion in the matrix of
granitic rock: Field test in the Stripa mine, Water Resour.
Res., 26(11): 2833-2842, 1990.
Black, S.C. 1988. Defining Control Sites and Blank Sample
Needs, Chapter 7, pp. 1077-117. In: Principles of
Environmental Sampling, L. H. Keith (ed.). ACS
Professional Reference Book, American Chemical Society,
Washington, D.C., 458 pp.
Boettner, E.A., G.L. Ball, Z. Hollingsworth and R. Aquino.
1981. Organic and Organotin Compounds Leached from
PVC and CPVC Pipe, U.S. EPA Report, EPA/600/1-81-062,
102 pp.
Brobst, R.B. and P.M. Buszka. 1986. The effect of three
drilling fluids on ground water sample chemistry. Ground
Water Monitoring Review, VI(1):62-70.
Buddemeier R.W. and J.R. Hunt. 1988. Transport of
colloidal contaminants in groundwater: radionuclide
migration at the Nevada Test Site. Appl. Geochem. 3:535-
548.
Buddemeier, R.W. and J.H. Rego. 1986. Colloidal
Radionuclides in Groundwater. FY85 Annual Report.
Lawrence Livermore National Laboratory, Livermore, CA,
UCAR 10062/85-1.
Calabrese, E.J. and P.T. Kostecki. 1992. Risk Assessment
and Environmental Fate Methodologies, Lewis Publishers,
Chelsea, MI. 150 pp.
Campbell, J.A. and W.R. Mabey. 1985. A systematic
approach for evaluating the quality of groundwater
monitoring data. Ground Water Monitoring Review,
V(4):58-62.
Champ, D.R., W.F. Merritt, and J.L. Young. 1982. Potential
for Rapid Transport of Pu in Groundwater as Demonstrated
by Core Column Studies. In: Scientific Basis for Radio-
active Waste Management. Vol. 5, Elsevier Sci.Publ., NY.
Champlin, J.B.F. and G.G. Eichholz. 1976. Fixation and
remobilization of trace contaminants in simulated subsurface
aquifers. Health Physics. 30: 215-219.
C-3
-------
Chapelle, F.H. 1993. Groundwater Microbiology and
Geochemistry. John Wiley & Sons, New York, NY. 424 pp.
Clapp, R.B., D.S. Hicks, O.K. Solomon, D.M. Borders, D.D.
Huff, andH.L. Boston. 1992. Groundwater, surface water,
and movement of contaminants at the Oak Ridge
Reservation, Abstract presented at American Geophysical
Union meeting, San Francisco, CA.
Clark, S.B., N.M. Park, and R.C.Tuckfield. 1992. Effects of
Sample Collection Device and Filter Pore Size on
Concentrations of Metals in Groundwater Samples. In:
Grundfos Pumps Corporation, pp. 13-20.
Cowgill, U.M. 1988. The Chemical Composition of
Leachate from a Two-Week Dwell-Time Study of PVC Well
Casing and Three-Week Dwell-Time Study of Fiberglass
Reinforced Epoxy Well Casing. In: Ground-Water
Contamination: Field Methods, A. G. Collins and A. I.
Johnson, eds. American Society for Testing and Materials,
Philadelphia, Pennsylvania, STP 963, pp. 172-184.
Curran, C.M. andM.B. Tomson. 1983. Leaching of trace
organics into water from five common plastics. Ground
Water Monitoring Review, 3(3):68-71.
Dablow, John S. Ill, G. Walker, andD. Persico. 1988.
Design Consideration and Installation Techniques for
Monitoring Wells Cased with TEFLON PTFE, in A.G.
Collins and A.I. Johnson, eds., Ground-Water
Contamination: Field Methods. ASTM STP 963, ASTM,
Philadelphia, pp. 199-205.
Dalton, M., B. Huntsman, andK. Bradbury. 1991.
Acquisition and Interpretation of Water-Level Data. In:
Nielsen, David M. 199 1. pp. 367-396.
Danielsson L.G. 1982. On the use of filters for
distinguishing between dissolved and paniculate fractions in
natural waters. Water Res. 16:179-182.
Davis, J.A., C.C. Fuller, J.A. Coston, K.M. Hess, and E.
Dixon. 1993. Spatial Heterogeneity of Geochemical and
Hydrologic Parameters Affecting Metal Transport in Ground
Water. U.S. EPA Environmental Research Brief, EPA/600/
S93/006, 22 pp.
Driscoll, F.G. 1986. Groundwater and Wells, Johnson
Division, St. Paul, Minnesota, 1089 pp.
Dunnivant, P.M., P.M. Jardine, D.L. Taylor, and i.F.
McCarthy. 1992. Cotransport of cadmium and
hexachlorobiphenyl by dissolved organic carbon through
columns containing aquifer material. Environ. Sci. Technol.
26(2):360-368.
Eichholz, G.G., E.G. Wahlig, G.F. Powell and T.F. Craft.
1982. Subsurface migration of radioactive waste materials
by paniculate transport. Nuclear Tech. 5 8: 511 -519.
Enfield, C.G. and G. Bengtsson. 1988. Macromolecular
transport of hydrophobic contaminants in aqueous
environments. Ground Water. 26(1): 64-70.
Electric Power Research Institute (EPRI). 1986. Changes in
the Chemical Integrity of Groundwater Samples due to
Sampling Devices and Procedures. EPRI Environment
Division Technical Brief RP 2485-7, 2 pp.
Electric Power Research Institute (EPRI). 1985a. Field
Measurement Methods for Hydrogeologic Investigations: A
Critical Review of the Literature. EPRI Report EA-4301,
Research Project 2485-7, Palo Alto, CA. 260 pp.
Electric Power Research Institute (EPRI). 1985b.
Preliminary Results on Chemical Changes in Groundwater
Samples due to Sampling Devices. EPRI Report EA-4118,
Research Project 2485-7, Palo Alto, CA. 54 pp.
Federal Insecticide, Fungicide, and Rodenticide Act, 7
U.S.C. § 3(c)(2)(B).
Federal Register, 40 CFR Parts 257 and 258, Solid Waste
Disposal Facility Criteria; Final Rule. 56 (196), 50978-
51119.
Ficken, J.R. 1988. Recent Development of Downhole
Water Samplers for Trace Organics. In: Collins, A.G., and
Johnson, A.I. 1988. pp. 253-257.
Freeze, R.A. and J.A. Cherry. 1979. Ground Water.
Prentice-Hall, Inc., Englewood Cliffs, New Jersey.
Gammage, R.B, andB.A. Berven. 1992. Hazardous Waste
Site Investigations - Towards Better Decisions. Lewis
Publishers, 288 pp.
Garske, E.E., andM.R. Schock. 1986. An inexpensive flow-
through cell and measurement system for monitoring
selected chemical parameters in ground water. Ground
Water Monitoring Review. 6(3):79-84.
C-4
-------
Gibb, J.P., R.M. Schuller, and R.A. Griffin. 1981.
Procedures for the Collection of Representative Water
Quality Data from Monitoring Wells. Illinois State Water
Survey/Geological Survey Cooperative Ground-Water
Report No. 7. Champaign, IL.
Gibs, J. andT.E. Imbrigiotta. 1990. Well-purging criteria for
sampling purgeable organic compounds. Ground Water.
28(l):68-78.
Gillham, R.W. and S.F. O'Hannesin. 1990. Sorption of
Aromatic Hydrocarbons by Materials Used in Construction
of Ground-Water Sampling Wells, in D. M. Nielsen and A. I.
Johnson, eds., Ground-Water and Vadose Zone Monitoring,
American Society for Testing and Materials, Philadelphia,
PA. STP 1053, pp. 108-122.
Gillham, R.W., MJ. L. Robin, J. F. Barker and J.A. Cherry.
1983. Ametican Petroleum Institute. API Publication 4367,
206 pp.
Gillham, R.W., M.J.L. Robin, J.F. Barker, and J.A. Cherry.
1985. Field Evaluation of Well Flushing Procedures. API
Publ. 4405, 110pp.
Gillham, R.W., MJ.L. Robin, J.F. Barker, and J.A. Cbeny.
1983. Groundwater Monitoring and Sample Bias.
Environmental Affairs Department, American Petroleum
Institute. 206 pp.
Gounaris, V., P.R. Anderson, and T.M. Holsen. 1993.
Characteristics and environmental significance of colloids in
landfill leachate. Environ. Sci. Technol. 27(7): 1381-1387.
Grisak, G.E., R.E. Jackson, and J.F. Pickens. 1978.
Monitoring Ground Water Quality: The Technical
Difficulties. In: Establishment of Water Quality
Monitoring Programs. Everett, L.G., and Schmidt, K.D.
(Editors). American Water Resources Association.
Groher, D.M. 1990. An investigation of the factors affecting
the concentrations of poly cyclic aromatic hydrocarbons in
groundwater at coal tar waste sites. M.S. thesis. Dept of
Civil Eng., Massachusetts Institute of Technology. 145 pp.
Ground-Water Monitoring Requirements for Landfill
Disposal of PCBs. 1992. 40 CFR § 761.75(b)(6)(ii) and (iii),
(1992).
Gschwend P.M., D.A. Backhus, J.K.MacFarlane, and A.L.
Page. 1990. Mobilization of colloids in groundwater due to
infiltration of water at a coal ash disposal site. J. Contam.
Hydrol. 6:307-320.
Gschwend, P.M. andM.D. Reynolds. 1987. Monodisperse
ferrous phosphate colloids in an anoxic groundwater plume.
J. Contam. Hydrol., 1:309-327.
Harvey, R.W., L.H. George, R.L. Smith, and D.R. LeBlanc.
1989. Transport of microspheres and indigenous bacteria
through a sandy aquifer: Results of natural- and forced-
gradient tracer experiments. Environ. Sci. Technol.,
23(l):51-56.
Hem, D. 1985. Study and Interpretation of the Chemical
Characteristics of Natural Water. U.S.G.S. Water Supply
Paper 2254.
Hem, J.D. and C.E. Roberson. 1967. Form and Stability of
aluminum hydroxide complexes in dilute solution. U.S.
Geol. Water Supply Pap. 1 827-A,A24.
Hendry, M.J., RW. Gillham, and J.A. Cherry. 1983. An
integrated approach to hydrogeologic investigations - a case
history. Journal of Hydrology. Elsevier Science Publishers
B.V., Amsterdam - Printed in the Netherlands.
pp. 211-232.
Herzog, B.L., Sheng-Fu J. Chous, J.R. Valkenburg, and R.A.
Griffin. 1988. Changes in volatile organic chemical
concentrations after purging slowly recovering wells.
Ground Water Monitoring Review, pp. 93-99.
Herzog, B., J. Pennine, and G. Nielsen. 1991. Ground-Water
Sampling, in D.M. Nielsen, ed., Practical Handbook of
Ground-Water Monitoring. Lewis Publishers, Chelsea, MI,
pp. 449-499.
Hess, K.M., S.H. Wolf, M.A. Celia, and S.P. Garabedian.
1991. Macrodispersion and Spatial Variability of Hydraulic
Conductivity in a Sand and Gravel Aquifer, Cape Cod,
Massachusetts. USEPA Environmental Research Brief,
EPA/M-91/005, 9pp.
Hewitt, A.D. 1989. Leaching of Metal Pollutants from Four
Well Casings Used for Ground-Water Monitoring, CRREL
Special Report 89-32, U. S. Army Cold Regions Research
and Engineering Laboratory, Hanover, NH.
Hewitt, A.D. 1992. Potential of common well casing
materials to influence aqueous metal concentration. Ground
Water Monitoring Review. 12(2): 131 -136.
C-5
-------
Hingston, F.J., A.M. Posner, and J.P. Quirk. 1971. Discuss.
Faraday Sci. 52, 334-342.
Holm, T.R., O.K. George, andM.J. Barcelona. 1988.
Oxygen transfer through flexible tubing and its effects on
ground water sampling results. Ground Water Monitoring
Review. 8(3):83.
Horowitz, A.J., K.A. Elrick, andM.R. Colberg. 1992. The
effect of membrane filtration artifacts on dissolved trace
element concentrations. Wat. Res. 26 (6):753-763.
Houghton, R.L., and M.E. Berger. 1984. Effects of Well
Casing Composition and Sampling Method on Apparent
Quality of Ground Water. Proceedings of the Fourth
National Symposium and Exposition on Aquifer Restoration
and Ground Water Monitoring, pp. 203-213.
Hsieh, P.A., and S.P. Neuman. 1985. Field Determination of
the Three-Dimensional Hydraulic Conductivity Tensor of
Anisotropic Media, 1. Theory. Water Resources Research,
1(11):1655-1665.
Imbrigiotta, T.E., J. Gibs, T.V. Fusillo, G.R. Kish, and J.J.
Hochreiter. 1988. Field Evaluation of Screen Sampling
Devices for Purgeable Organic Compounds in A.G. Collins
and AJ. Johnson, eds., Ground Water Contamination: Field
Methods. ASTM STP 963, ASTM, Philadelphia, pp. 258-
273.
Johnson B.D. and PJ. Wangersky. 1985. Seawater filtration:
Particle flow and impaction considerations. Limnol.
Oceanogr. 30: 966-971.
Jones, J.N. and G.D. Miller. 1988. Adsorption of Selected
Organic Contaminants onto Possible Well Casing Materials,
in A. G. Collins and A. I. Johnson, eds., Ground-Water
Contamination: Field Methods, American Society for
Testing and Materials, Philadelphia, Pennsylvania, STP 963,
pp. 185-198.
Junk, G.A., H.J. Svec, R.D. Vick and M.J. Avery. 1974.
Contamination of water by synthetic polymer tubes.
Environmental Science and Technology, 8(13): 1100-1106.
Kaplan, D.I., Bertsch, P.M., andD.C. Adriano. 1993.
Soilborne mobile colloids as influenced by water flow and
organic carbon. Environ. Sci. Technol., 27(6): 1193-1200.
Kearl, P.M., N.E. Korte, and T.A. Cronk. 1992. Suggested
modifications to ground water sampling procedures based on
observations from the colloidal borescope. Ground Water
Monitoring Review, Spring 1992:155-160.
Keely, J.F. 1982. Chemical time-series sampling. Ground
Water Monitoring Review. 2(4):29-38.
Keely, J.F. and K. Boeteng. 1987. Monitoring well
installation, purging, and sampling techniques - Part 1:
conceptualizations, Ground Water, 25(3):300-313.
Keely, J.F., and F. Wolf. 1983. Field applications of
chemical time-series sampling. Ground Water Monitoring
Review, pp. 26-33.1.
Kim, J.I., G. Buckau, F. Baumgartner, H.C. Moon and D.
Lux. 1984. Colloid Generation and the Actinide Migration
in Gorleben Groundwaters. In Scientific Basis for Nuclear
Waste Management, Vol. 7, Gary L. McVay, Ed., Elsevier,
NY. pp. 31-40.
Kueper, B.H., and D.B. McWhorter. 1991. The behaviour of
dense, non-aqueous phase liquids in fractured clay and rock.
Ground Water. 29(5):716-728.
Kennedy V.C., G.W. Zellweger, andB.F. Jones. 1974. Filter
pore-size effects on the analysis of Al, Fe, Mn, and Ti in
water. Water Resour. Res. 10:785-790.
Knobel, L.L. and L.J. Mann. 1993. Sampling for purgeable
organic compounds using positive-displacement piston and
centrifugal submersible pumps: a comparative study.
Ground Water Monitoring and Remediation. 13(2).
Kurt, C.E. andR.C. Johnson. 1982. Permeability of grout
seals surrounding thermoplastic well casing, Ground Water,
20(4):415-419.
Lang. K.T., M.H. Stutz, L.V. Parker, A.D. Hewitt, and T.F.
Jenkins. 1989. Influence of Well Casing Materials on
Chemical Species in Ground Water. Proceedings of the Fifth
Annual Waste Testing and Quality Assurance Symposium,
Washington, p. 29.
Lovejoy, S. andD.M. Eisenberg. 1989. Cost Effectiveness
of Dedicated Sampling Systems for Ground Water
Monitoring, pp. 89-96 in Proceedings of HAZMACON '89
Hazardous Materials Management Conference and
Exhibition, April 18-20, 1989, Santa Clara, CA. T.
Bursztynsky andM. Loss, editors. Assoc. of Bay Area
Governments, P.O. Box 2050, Oakland, CA 94604-2050.
C-6
-------
Magaritz M, A.J. Amiel, D. Ronen, and M.C. Wells. 1990.
Distribution of metals in a polluted aquifer: A comparison of
aquifer suspended material to fine sediments of the adjacent
environment. J. Contam. Hydrol. 5:333-347.
Marsh, J.M. and J.W. Lloyd. 1980. Details of hydro-
chemical variations in flowing wells. Ground Water,
18(4):366-373.
Martin-Hayden, J.M., Robbins, G.A., and Bristol, R.D.,
1991. Mass Balance Evaluation of Monitoring Well
Purging, Part 111. Field Tests at a Gasoline Contamination
Site. Journal of Contaminant Hydrology, v. 8, p. 225-241.
Martin-Hayden, J.M. and G.A. Robbins. 1994. Plume
Distortion and Apparent Attenuation Due to Concentration
Averaging in Monitoring Wells. U.S. EPA, Internal Report,
in review.
Maskarinec, M.P., L.H. Johnson, S.K. Holladay, R.L.
Moody, C.K. Bayne, andR.A. Jenkins. 1990. Stability of
volatile organic compounds in environmental water samples
during transport and storage. Environ. Sci. Technol.,
24(11):1665-1670.
McCarthy J.F. and C. Degueldre. 1993. Sampling and
characterization of colloids and particles in groundwater for
studying their role in contaminate transport. In:
Environmental Particles (eds. Buffle J. and van Leeuwen
H.P.), Lewis Publishers, 247-315.
McCarthy, J.F., and J.M. Zachara, Subsurface transport of
contaminants. Environmental Science Technology,
23(5):496-502, 1989.
McDowell-Boyer, L.M., J.R. Hunt, andN. Sitar. 1986.
Particle transport through porous media. Water Resources
Res. 22(13):1901-1921.
McKay, L.D., and J.A. Cherry. 1992. Groundwater research
in clay-rich glacial tills in southwestern Ontario, Paper
presented at Intl. Assoc. of Hydrol. Conference, Hamilton,
Ontario, May, 1992.
McKay, L.D. and J.A. Cherry. 1993. A field example of
bacteriophage as tracers of fracture flow. Environmental
Science and Technology. 27(6): 1075-1079.
McKay, L.D., J.A. Cherry, R.C. Bales, M.T. Yahya, and C.P.
Gerba. 1993. A field example of bacteriophage as tracers of
fracture flow. Environ. Sci. and Technol. 27(6).
McKay, L.D., J.A. Cherry andR.W. Gillham. 1993a. Field
experiments in a fractured clay till: 1. Hydraulic conductivity
and fracture aperature. Water Resources Research.
29(4): 1149-1162.
McKay, L.D., R.W. Gillham and J.A. Cherry. 1993b. Field
experiments in a fractured clay till: 2. Solute and colloidal
transport, Water Resources Research, 29(12):3879-3890.
Means, J.C. and R. Wijayaratne. 1982. Role of natural
colloids in the transport of hydrophobicpollutants. Science
215(19): 968-970.
McKay, L., K. Novakowski, and J. McCarthy. 1994.
Ground-water Sampling of Fractured Clay and Rock. In
Ground Water Sampling Workshop, U.S. EPA, Nov. 30-Dec.
2,1993, Dallas, TX.
Michalski, M. 1989. Application of temperature and
electrical conductivity logging in ground water monitoring.
Ground Water Monitoring Review, 9(3): 112-118.
Mickam, J.T., R. Bellandi,and B.C. Tifft Jr. 1989.
Equipment decontamination procedures for ground water
and vadose zone monitoring programs: Status and prospects.
Ground Water Monitoring Review. 9(2).
Miller, G.D. 1982. Uptake and Release of Lead, Chromium
and Trace Level Volatile Organics Exposed to Synthetic
Well Casings, Proceedings of the Second National
Symposium on Aquifer Restoration and Ground-Water
Monitoring, National Water Well Association, Dublin, Ohio,
pp. 236-245.
Mioduszewski, D., and J.L. Irwin. 1985. Dedication-An
Answer to Reliable and Cost-Effective Ground-Water
Sampling. Proceedings, National Conference on Hazardous
Wastes and Environmental Emergencies, HMCRI,
Cincinnati, Ohio. pp. 106-110.
Molz, F.J., 0. Guven, and J.G. Melville. 1990. A New
Approach and Methodologies for Characterizing the
Hydrogeologic Properties of Aquifers. EPA Project
Summary. EPA 600/S2-90/002.
Molz, F.J. and C.E. Kurt. 1979. Grout-induced temperature
rises surrounding wells. Ground Water. 17(3):264-269.
National Research Council. 1990. Ground Water and Soil
Contamination Remediation Toward Compatible Science,
Policy and Public Perception, Water Science and Technology
Board, National Academy Press, Washington, D.C., 261 pp.
C-7
-------
National Symposium on Measuring and Interpreting VOC's
in Soils: State of the Art and Research Needs. 1993. Univ.
of Wisconsin Extension, U.S. Environmental Protection
Agency, Oak Ridge National Laboratory, U.S. Department
of Energy, U.S. Toxic and Hazardous Materials Agency,
America Petroleum Institute. January 12-14, 1993, Las
Vegas, NV.
Nielsen, D.M., ed. 1991. Practical Handbook of Ground-
Water Monitoring, Lewis Publishers, Chelsea, MI. 717 pp.
Nielsen, D.M., and A.I. Johnson (editors). 1990. Ground
Water and Vadose Zone Monitoring. American Society for
Testing and Materials, Ann Arbor, MI.
Nielsen, D.M. andM.N. Sara. 1992 Current Practices in
Ground Water and Vadose Zone Investigations ASTM 1118,
American Society for Testing and Materials, Philadelphia,
PA. 431pp.
Nielsen, D.M. and G.L.Yeates. 1985. A comparison of
sampling mechanisms available for small-diameter ground
water monitoring wells. Ground Water Monitoring Review.
5(2):83-99.
Nielsen, D.M. and G.L. Yeates. 1985. A Comparison of
Sampling Mechanisms Available for Small-Diameter
Ground Water Monitoring Wells. Proceedings of the 5th
National Symposium and Exposition on Aquifer Restoration
and Ground Water Monitoring, pp. 237-270.
Nelson, D.M., W.R. Penrose, J.O. Karttunen and P.
Mehlhaff. 1985. Effects of dissolved organic carbon on the
adsorption properties of plutonium in natural waters.
Environ. Sci. Technol. 19(1): 127-13 1.
Nightingale, H.I. and W.C. Bianchi. 1977. Ground water
turbidity resulting from artificial recharge. Ground Water.
15(2): 146-152.
Nikolaidis, N.P., G.A. Robbins, M. Sherer, B. McAninch, G.
Binkhorst, and S.L. Suib. 1994. Vertical Distribution and
Partitioning of Chromium in a Glacio-Fluvial Aquifer.
Ground Water Monitoring and Remediation,.
Novakowski, K.S. 1992. The analysis of tracer experiments
conducted in divergent radial flow fields. Water Resour.
Res. 28(12), 3215-3225.
O'Melia, C.R. 1980. Aquasols: The behavior of small
particles in aquatic systems. Environ. Sci. Technol.
14(9): 1052-1060.
Palmer, C.D., J.F. Keely, and W. Fish. 1987. Potential for
solute retardation on monitoring well sand packs and its
effect on purging requirements for ground-water sampling.
Ground-Water Monitoring Review, 7: 40-47.
Panko, A.W., and P. Earth. 1988. Chemical Stability Prior
to Ground-Water Sampling: A Review of Current Well
Purging Methods. In: Collins, A.G., and Johnson A.I. pp.
232-239.
Parker, L.V. and T.F. Jenkins. 1986. Suitability of polyvinyl
chloride well casings for monitoring munitions in ground
water. Ground Water Monitoring Review. 6(3):92-98.
Parker, L.V., T.F. Jenkins, andP.B. Black. 1989. Evaluation
of Four Well Casing Materials for Monitoring Selected
Trace Level Organics in Ground Water. CRREL Report 89-
18, U.S. Army Cold Regions Research and Engineering
Laboratory, Hanover, NH 03775.
Parker, L.V., A.D. Hewitt and T.F. Jenkins. 1990. Influence
of casing material on trace-level chemicals in well water.
Ground Water Monitoring Review. 10(2): 146-156.
Parker, L.V. 1991. Discussion of the effects of latex gloves
and nylon cord on ground water sample quality, by J. L.
Canova and M. G. Muthig. Ground Water Monitoring
Review. 11(4): 167-168.
Parker, L.V. 1992. Suggested Guidelines for the Use of
PTFE, PVC, and Stainless Steel in Samplers and Well
Casings, Current Practices in Ground Water and Vadose
Zone Investigations, in D. Nielsen and M. Sara, eds.
American Society for Testing and Materials, Philadelphia,
Pennsylvania, STP 1118, pp. 217-229.
Paul, C.J. andR.W. Puls. 1993. Comparison of Ground-
Water Sampling Devices Based On Equilibration of Water
Quality Indicator Parameters. EPA/600/A-93/005, 13pp.
Pearsall, K.A. and D.A.V. Eckhardt. 1987. Effects of
selected sampling equipment and procedures on the
concentrations of trichloroethylene and related compounds in
ground water samples. Ground-Water Monitoring Review,
Spring, pp. 64-73.
C-8
-------
Penrose W.R., W.L. Polzer, E.H. Essington, D.M. Nelson,
and K.A. Orlandini. 1990. Mobility of plutonium and
americium through a shallow aquifer in a semiarid region.
Environ. Sci. Technol. 24:228-234.
Pohlmann, K.F., R.P. Blegen, and J.W. Hess. 1991. Field
Comparison of Ground Water Sampling Devices for
Hazardous Waste Sites: An Evaluation Using Volatile
Organic Compounds. U.S. Environmental Protection
Agency, Environmental Monitoring Systems Laboratory,
Las Vegas, Nevada. National Technical Information
Service, Springfield, Virginia.
Pohlmann, K.F. and J.W. Hess. 1988. Generalized ground
water sampling device Matrix. Ground Water Monitoring
Review. Fall:82-84.
Pohlmann, K.F., G.A. Icopini, R.D. McArthur, and C.G.
Rosal. 1994. Evaluation of Sampling and Field-Filtration
Methods for the Analysis of Trace Metals in Ground Water.
U.S. Environmental Protection Agency, Las Vegas, NV, in
preparation.
Powell, R.M. andR.W. Puls. 1993. Passive Sampling of
Ground-Water Monitoring Wells Without Purging:
Multilevel Well Chemistry and Tracer Disappearance. J.
Contam. Hydrol. 12,51-77.
Puls, R.W. 1994. Use of Low-Flow or Passive Techniques
for Sampling Ground Water. In: Ground Water Sampling
Workshop, USEPA, Nov. 30-Dec. 2,1993, Dallas, TX.
Puls, R.W., D.A. Clark, B. Bledsoe, R.M. Powell, and C.J.
Paul. 1992. Metals in ground water: sampling artifacts and
reproducibility. Hazardous Waste and Hazardous Materials,
9(2): 149-162.
Puls, R.W., and M.J.Barcelona. 1989. Filtration of ground
water samples for metals analysis. Hazardous Waste and
Hazardous Materials, 6(4):385-393.
Puls, R.W. and M.J. Barcelona. 1989. Ground Water
Sampling for Metals Analyses. EPA/540/4-89/001.
Puls, R. W. and J.H. Eychaner. Sampling Ground Water for
Inorganics - Pumping Rate, Filtration, and Oxidation
Effects, Fourth National Outdoor Action Conference on
Aquifer Restoration, Ground Water Monitoring and
Geophysical Methods, National Water Well Association,
Dublin, OH, 1990.
Puls, R.W., J.H. Eychauer, and R.M. Powell. 1990.
Colloidal-Facilitated Transport of Inorganic Contaminants in
Ground Water: Part 1. Sampling Considerations. EPA/ 600/
M-90/023, U.S. EPA, Ada, OK, 1990.12 pp.
Puls, R.W. and C.J. Paul. 1995. Low flow purging and
sampling of ground-water monitoring wells with dedicated
systems. Accepted, Ground Water Monitoring and
Remediation, Winter issue, 1994.
Puls, R.W., C.J. Paul, D.A. Clark, and J. Vardy. 1994.
Transport and transformation of hexavalent chromium
through soils and into ground water. J. Soil Contamination.
To be published June, 1994.
Puls, R.W. and R.M. Powell. 1992. Acquisition of
representative ground water quality samples for metals.
Ground Water Monitoring Review, 12(3): 167-176.
Puls, R.W. and R.M. Powell. 1992. Transport of inorganic
colloids through natural aquifer material: Implications for
contaminant transport. Environ. Sci. Technol. 26(3): 614-
621.
Puls, R.W., R.M. Powell, D.A. Clark, and C.J. Paul. 1991.
Facilitated Transport of Inorganic Contaminants in Ground
Water Part 11. Colloidal Transport. EPA/600/M-91/040,
U.S. EPA, Ada, OK. 12pp.
Reynolds, M.D. 1985. Colloids in Groundwater. Masters
Thesis. Mass. Inst. of Tech. Cambridge, MA.
Reynolds, G.W. andR.W. Gillham. 1985. Absorption of
Halogenated Organic Compounds by Polymer Materials
Commonly Used in Ground-Water Monitors, Proceedings of
the Second Canadian/American Conference on
Hydrogeology, National Water Well Association, Dublin,
OH. pp. 125-132.
Reynolds, G.W., J.T. Hoff andR.W. Gillham. 1990.
Sampling bias caused by materials used to monitor
halocarbons in groundwater. Environmental Science
Technology. 24(1): 135-142.
Rivett, M., S. Feenstra and J. Cherry. 1991. Field
Experimental Studies of a Residual Solvent Source
Emplaced in the Ground Water Zone, Proceedings of the
Conference on National Water Well Association, Petroleum
Hydrocarbons and Organic Chemicals in Ground Water,
Dublin, OH, pp. 283-299.
C-9
-------
Robbins, G.A., 1989. Influence of Using Purged and
Partially Penetrating Monitoring Wells on Contaminant
Detection, Mapping and Modeling. Ground Water Journal,
v. 27, No. 2, p. 155-162.
Robbins, G.A., R.D. Bristol and V.D. Roe. 1989. A field
screening method for gasoline contamination using a
polyethylene bag sampling system. Ground Water Monit.
Rev. 9(3):87-97.
Robbins, G.A. and Martin-Hayden, J.M., 1991. Mass
Balance Evaluation of Monitoring Well Purging, Part 1.
Theoretical Models and Implications for Representative
Sampling. Journal of Contaminant Hydrology, v. 8, p. 203-
224.
Robin, M.J.L. and R.W. Gillham. 1987. Field evaluation of
well purging procedures. Ground Water Monitoring Review.
7(4). 85-93.
Roe, V.D., M.J. Lacy, J.D. Stuart and G.A. Robbins. 1989.
Manual headspace method to analyze for the volatile
aromatics of gasoline in groundwater and soil samples.
Anal. Chem. 61:2584.
Ronen, D., M. Magaritz and I. Levy. 1987. An in situ
multilevel sampler for preventive monitoring and study of
hydrochemical profiles in aquifers. Ground Water
Monitoring Review. 7(4), 69-74.
Ronen D., M. Magaritz, U.Weber, A.J. Amiel, andE. Klein.
1992. Characterization of suspended particles collected in
groundwater under natural gradient flow conditions. Water
Resour. Res. 28:1279-1291.
Ross, S. and I. Morrison. Colloidal Systems and Interfaces,
New York, John Wiley & Sons, 1988.
Ryan, J.N. 1998. Groundwater Colloids in Two Atlantic
Coastal Plain Aquifers: Colloid Formation and Stability.
M.S. Thesis, Dept. Civil Eng., Massachusetts Institute of
Technology, Cambridge, MA.
Ryan, J.N. and P.M. Gschwend. 1990. Colloid mobilization
in two atlantic coastal plain aquifers: field studies. Water
Resources Res. 26(2): 307-322.
Ryan J.N. and Gschwend P.M. 1990. Groundwater colloids
in two atlantic coastal plain aquifers: Field studies. Water
Resour. Res. 26:307-322.
Ryan J.N. and Gschwend P.M. 1992. Effect of iron
diagenesis on clay colloid transport in an unconfined sand
aquifer. Geochim. Cosmochim. Acta. 56:1507-1521.
Salbu B., H.E. Bjomstad, N.S. Lindstrom, E. Lydersen, E.M.
Brevik, J.P. Rambaek, and P.E. Paus. 1985. Size
fractionation techniques in the determination of elements
associated with paniculate or colloidal material in natural
fresh waters. Talanta 32:907-913.
Sanders, T.G., R.C. Ward, J.C. Loftis, T.D. Steele, D.D.
Adrian, and V. Yevjevich. 1983. Design of Networks for
Monitoring Water Quality, Water Resources Publications,
Littleton, CO. 328 pp.
Sandhu, S.S. and G.L. Mills. 1987. Kinetics and
Mechanisms of the Release of Trace Inorganic
Contaminants to Ground Water from Coal Ash Basins on the
Savannah River Plant. Savannah River Ecology Lab.
Aiken, SC. DOE/SR/15170--1
Sara, M.N. 1994. Standard Handbook for Solid and
Hazardous Waste Facility Assessments, Lewis Publishers,
Ann Arbor, MI.
Scalf, M.R., J.F. McNabb, W.J. Dunlap, R.L. Cosby, and J.
L. Fryberger. 1981. Manual of Ground-Water Quality
Sampling Procedures, National Water Well Assoc. 93 pp.
Schalla, R. 1986. A Comparison of the Effects of Rotary
Wash and Air Rotary Drilling Techniques on Pumping Test
Results, Proceedings of the Sixth National Symposium and
Exposition on Aquifer Restoration and Ground Water
Monitoring, National Water Well Association, Dublin, OH,
pp. 7-26.
Sevee and Maher Engineers. 1992. Pre-Design Investigation
for the Remediation of Groundwater, Auburn Road Landfill
Site, Londonderry, NH.
Sevee and Maher Engineers. 1993. Supplement II
Investigation Report, Supplemental Pre-Design
Investigation for the Remediation of Groundwater, Auburn
Road Landfill Site, Londonderry, NH..
Sheldon, R.W. andW.H. Sutcliffe, Jr. 1969. Retention of
marine particles by screens and filters. Limnol. and
Oceanogr. 14(3):441-444.
Sheppard, J.C., M.J. Campbell and J.A. Kittrick. 1979.
C-10
-------
Retention of neptunium, americium and curium by diffusible
soil particles. Environ. Sci. Technol. 13(6):680-684.
Snow, D.T. 1970. The frequency and apertures of fractures
in rock. J. Rock Mech. Min. Sci., 7:23-40.
Sosebee, J.B., P.C. Geiszler, D.L. Winegardner and C.
Fisher. 1983. Contamination of Ground Water Samples
With Poly (Vinyl Chloride) Adhesives and Poly (Vinyl
Chloride) Primer From Monitor Wells, in Proceedings:
American Society of Testing and Materials Second
Symposium on Hazardous and Industrial Solid Waste
Testing, ASTM Special Publication #805, pp. 38-49.
Standard Methods for the Examination of Water and
Wastewater. 1989. Clesceri, L.S., A.E. Greenberg, andR.
Rhodes Trussell, eds. 17th Edition, American Public
Health Association, Washington D.C.
Sudicky, E.A. and R.G. McLaren. 1992. The LTG technique
for large-scale simulation of mass transport in discretely
fractured porous formations. Water Resour. Res.,
28(2):499-514.
Summers, K.V, and S.A. Gherini. 1987. Sampling
Guidelines for Groundwater Quality. Palo Alto, CA:
Electric Power Research Institute.
Tai, D.Y., K.S. Turner and L.A. Garcia. 1991. The use of a
standpipe to evaluate ground water samplers. Ground Water
Monitoring Review. Winter: 125-132.
Takayanagi, K. and G.T.F. Wong. 1984. Organic and
colloidal selenium in south Chesapeake Bay and adjacent
waters. Marine Chem. 14:141-148.
Tomson, M.B., S.R. Hutchins, J.M. King and C.H. Ward.
1979. Trace Organic Contamination of Ground Water:
Methods for Study and Preliminary Results, III World
Congress on Water Resources, Mexico City, Mexico, Vol. 8,
pp. 3701-3709.
USEPA. 1979. Methods for Chemical Analysis of Water
and Wastes. EPA/600/4-49/020.
USEPA. 1986. RCRA Ground-Water Monitoring Technical
Enforcement Guidance Document, Office of Waste
Programs Enforcement and Office of Solid Waste and
Emergency Response, (OSWER-9950.1).
USEPA. 1987. A Compendium of Superfund Field
Operations Methods. Office of Emergency and Remedial
Response, (EPA/540/P-87/001; OSWER Directive: 9355.0-
14).
USEPA. 1987. Data Quality Objectives for Remedial
Response Activities Development Process, USEPA. 540/G-
87/003, March.
USEPA. 1988. Draft Guidance for Ground-Water
Monitoring Studies. Office of Pesticide Programs.
USEPA. 1988. First International Symposium on Field
Screening Methods for Hazardous Waste Site Investigations,
U.S. Army Toxic and Hazardous Materials Agency,
Instrument Society of America.
USEPA. 1989. RCRA Sampling Procedures Handbook.
USEPA. 1989. Risk Assessment Guidance for Superfund:
Interim Final Guidance. Office of Emergency and Remedial
Response (EPA/540/1-89/002).
USEPA. September 1990. Handbook - Ground Water,
Volume I: Ground Water and Contamination. EPA/625/6-
90/016a, 141 pp.
USEPA. July 1991. Handbook - Ground Water, Volume II:
Methodology. EP A/625/6-90/016b, 144pp.
USEPA. November 1991. Seminar Publication - Site
Characterization for Subsurface Remediation. EPA/625/4-
91/026, 259 pp.
USEPA. 1992. RCRA Ground-Water Monitoring: Draft
Technical Guidance, Office of Solid Waste, EPA/530-R-93-
001, PB93-139-350, Washington, DC.
U.S. EPA. 1992b. Site Assessment Program.. Draft short
sheet "Key Issues in Site Assessment Sampling" dated June
30, 1992. Transmitted to MPCA by Memorandum from U.S.
EPA Region V Site Assessment Section, Chicago, IL, on
July 23, 1992.
U.S. EPA. 1993a. Office of Research and Development.
Subsurface Characterization and Monitoring Techniques: A
Desk Reference Guide, Volume I: Solids and Ground Water,
Appendices A and B; U.S. Environmental Protection
Agency, Washington, DC. EPA/625/R-93/003a
C-ll
-------
U.S. EPA. 1993b. Office of Research and Development.
Subsurface Characterization and Monitoring Techniques: A
Desk Reference Guide, Volume II: The Vadose Zone, Field
Screening and Analytical Methods; Appendices C and D;
U.S. Environmental Protection Agency, Washington, DC.
EPA/625/R-93/003b
Unwin, I, and V. Maltby. 1988. Investigations of
Techniques for Purging Ground-Water Monitoring Wells
and Sampling Ground Water for Volatile Organic
Compounds. In: Collins, A.G., and Johnson, A.I. pp. 240-
252.
Wagemann, R. and G.J. Brunskill. 1975. The effect of filter
pore-size on analytical concentrations of some trace elements
in filtrates of natural water. Intern. J. Environ. Anal. Chem.
4: 75-84.
Wagner, R.E. Editor. 1992. Guide to Environmental
Analytical Methods. Genium Publishing Co., New York.
Walton-Day, K., D. Macalady, M. Brooks, V. Tate. 1990.
Field methods for measurement of groundwater redox
chemical parameters. Ground Water Monitoring Review,
10(4):81-89
Will, A.S., J.R. Kannard, A.R. Day, andL.B. Shannon. 1992.
Additional borehole geophysical logging at waste area
Grouping 1 at Oak Ridge National Laboratory, Oak Ridge,
TN, ORNL, Environ. Restoration Div., Technical
Memorandum 01-04.
Wilson, L.C., and J.VRouse. 1983. Variations in water
quality during initial pumping of monitoring wells. Ground
Water Monitoring Review. Ground Water Publishing
Company, Dublin, OH.
Yao, K., M.T. Habibian and C.R. O'Melia. 1971. Water and
waste water filtration: Concepts and applications. Environ.
Sci. Technol. 5(11): 1105-1112.
Yeskis, D., K. Chiu, S. Myers, J. Weiss, and T. Bloom.
1988. A Field Study of Various Sampling Devices and
Their Effects on Volatile Organic Contaminants. Second
National Outdoor Action Conference on Aquifer
Restoration, Ground-Water Monitoring and Geophysical
Methods, NWWA. May 23-26. pp. 471-479.
C-12
-------
Appendix D - Glossary
D-l
-------
Appendix D - Glossary
Absorption
Acid-extractable metals
Adsorption
Advection
Aggregation
Aliquot
Anisotropy
Annular space, annulus
Aquifer
Aquifer, confined
Aquifer, unconfined
Aquitard
Artifact
Assessment (investigation)
Assessment monitoring
ASTM cement types
The process by which one substance is taken into and included within another substance,
as the absorption of water by soil or nutrients by plants.
The concentration of metals in solution after treatment of an unfiltered sample with hot
dilute mineral acid.
The increased concentration of molecules or ions at a surface, including exchangeable
cations and anions on soil particles.
The process by which solutes are transported by the motion of flowing ground water.
The act of soil particles cohering so as to behave mechanically as a unit.
One of a number of equal-sized portions of a water sample that is being analyzed.
The condition under which one or more of the hydraulic properties of an aquifer vary
according to the direction of flow.
The space between two concentric tubes or casings, or between the casing and the
borehole wall. This would include the space(s) between multiple strings of tubing/
casings in a borehole installed either concentrically or multi-cased adjacent to each other.
A geologic formation, group of formations, or part of a formation that is saturated, and is
capable of providing a significant quantity of water.
An aquifer that is overlain by a confining bed. The confining bed has a significantly
lower hydraulic conductivity than the aquifer.
An aquifer in which there are no confining beds between the zone of saturation and the
surface. There will be a water table in an unconfined aquifer. Watertable aquifer is a
synonym.
A lithologic unit that impedes ground water movement and does not yield water freely to
wells or springs but that may transmit appreciable water to or from adjacent aquifers.
Where sufficiently thick, may act as a ground water storage zone. Synonymous with
confining unit. (9)
A product of artificial character due to extraneous agency.
The study of a particular area or region for defining the appropriateness of the area for
waste disposal.
An investigative monitoring program that is initiated after the presence of a contaminant
in ground water has been detected. The objective of this program is to determine the
concentration of constituents that have contaminated the ground water and to quantify the
rate and extent of migration of these constituents.
Portland cements meeting the requirements of ASTM C 150 (Standard Specifications for
Portland Cement). Cement types have slightly different formulations that result in
D-2
-------
various characteristics which address different construction conditions and different
physical and chemical environments. They are as follows:
ASTM Type I (Portland) A general-purpose construction cement with no special
properties.
Bailer
Ballast
Bar
Baseline
Bedrock
Bentonite clay
Blow-in
ASTM Type II (Portland) A construction cement that is moderately resistant to sulfates
and generates a lower heat of hydration at a slower rate than ASTM Type I.
ASTM Type III (Portland; high early strength) A construction cement that produces a
high early strength. This cement reduces the curing time required when used in cold
environments, and produces a higher heat of hydration than ASTM Type I.
ASTM Type IV (Portland) A construction cement that produces a low heat of hydration
(lower than Borehole Log: The record of geologic units ASTM Types I and II) and
develops strength at a slower rate.
ASTM Type V (Portland) A construction cement that is a high sulfate resistant
formulation. Used when there is severe sulfate action from soils and ground water.
A hollow tubular receptacle used to facilitate withdrawal of fluid from a well or borehole.
Materials used to provide stability to a buoyant object (such as casing within a borehole
filled with water).
A unit of pressure equal to one million dynes per square centimeter.
A surveyed condition which serves as a reference point to which later surveys are
coordinated or correlated.
The more or less continuous body of rock which underlies the overburden soils.
An altered deposit of volcanic ash usually consisting of sodium montmorillonite clay.
The inflow of ground water and unconsolidated material into a borehole or casing caused
by differential hydraulic heads; that is, caused by the presence of a greater hydraulic head
outside of a borehole/casing than inside.
Borehole
Borehole geophysics
A circular open or uncased subsurface hole created by drilling.
The general field of geophysics developed around the lowering of various probes into a
well.
Borehole log
Bulk density, soil
The record of geologic units penetrated, drilling progress, depth, water level, sample
recovery, volumes and types of materials used, and other significant facts regarding the
drilling of an exploratory borehole or well.
The mass of dry soil per unit bulk volume. The bulk volume is determined before drying
to constant weight at 105 degrees Centigrade.
D-3
-------
Bulk specific gravity
Bulk volume
Caliper log
Capillary forces
Capillary fringe
Casing
Casing, protective
Casing, surface
Cation-exchange
Caving; sloughing
Cement; Portland cement
Channels
Chemical activity
Clay films
Clay
The ratio of the bulk density of a soil to the mass of unit volume of water.
The volume, including the solids and the pores, of an arbitrary soil mass.
A borehole log of the diameter of an uncased well.
The forces acting on soil moisture in the unsaturated zone, attributable to molecular
attraction between soil particles and water.
The zone immediately above the water table, where water is drawn upward by capillary
attraction.
Pipe, finished in sections with either threaded connections or bevelled edges to be field
welded, which is installed temporarily or permanently to counteract caving, to advance
the borehole, and/or to isolate the zone being monitored.
A section of larger diameter pipe that is emplaced over the upper end of a smaller
diameter monitoring well riser or casing to provide structural protection to the well and
restrict unauthorized access into the well.
Pipe used to stabilize a borehole near the surface during the drilling of a borehole that
may be left in place or removed once drilling is completed.
The interchange between a cation and solution and another cation on the surface of any
surface-active material such as clay colloid or organic colloid. (1) Cation-exchange
capacity (CEC): The sum total of exchangeable cations that a soil can absorb. Expressed
in milli-equivalents per 100 grams or per gram of soil (or of other exchangers such as
clay).
The inflow of unconsolidated material into a borehole which occurs when the borehole
walls lose their cohesive strength.
Commonly known as Portland cement. A mixture that consists of calcareous,
argillaceous, or other silica, alumina-, and iron-oxide-bearing materials that is
manufactured and formulated to produce various types which are defined in ASTM C
150. Portland cement is also considered a hydraulic cement because it must be mixed
with water to form a cement-water paste that has the ability to harden and develop
strength even if cured under water (see ASTM Cement Types).
Voids that are significantly larger than packing voids. They are generally cylindrical
shaped and smooth walled, have regular conformation, and have relatively uniform cross-
sectional size and shape.
The molal concentration of an ion multiplied by a factor known as the activity
coefficient.
Coating of clay on the surfaces of soil peds and mineral grains and in soil pores. (Also
called clay skins, clay flows, illuviation cutans, argillans or tonhautchen.)
(a) A soil separate consisting of particles > 0.002 mm in equivalent diameter, (b) A
textural class.
D-4
-------
Clay mineral
Coarse texture
Colloid
Colloidal particles
Conceptual model
Conductance (specific)
Conductivity, hydraulic
Confining bed
Confining unit
Contaminant
d-10
d-60
Darcy's law
Deflocculate
Degradation
Degree of consolidation
(percent consolidation)
Naturally occurring inorganic crystalline material found in soils and other earthy
deposits, the particles being clay sized; that is, > 0.002 mm in diameter.
The texture exhibited by sand, loamy sands, and sandy loams except very fine sandy
loams.
The phase of a colloidal system made up of particles have dimensions of 10 -10,000
angstroms (1 -1000 manometers) and which is dispersed in a different phase.
Particles that are so small that the surface activity has an appreciable influence on the
properties of the particle.
A written or illustrated visualization of geologic/hydrogeologic/environmental conditions
of a particular area.
A measure of the ability of the water to conduct an electric current at 770 F (250C). It is
related to the total concentration of ionizable solids in the water. It is inversely
proportional to electrical resistance.
See soil water.
A body of material of low hydraulic conductivity that is stratigraphically adjacent to one
or more aquifers. It may lie above or below the aquifer.
A term that is synonymous with "aquiclude," "aquitard," and "aquifuge;" defined as a
body of relatively low permeable material stratigraphically adjacent to one or more
aquifers.
An undesirable substance not normally present or an unusually high concentration of a
naturally occurring substance in water or soil.
The diameter of a soil particle (usually in millimeters) at which 10% by weight of the
particles of a particular sample are finer. Synonymous with the effective size or effective
grain size.
The diameter of a soil particle (usually in millimeters) at which 60% by weight of the
particles of a particular sample are finer.
A law describing the rate of flow of water through porous media. (Named for Henry
Darcy of Paris who formulated it in 1856 from extensive work on the flow of water
through sand filter beds.)
(a) To separate the individual components of compound particles by chemical and/or
physical means, (b) To cause the particles of the disperse phase of a colloidal system to
become suspended in the dispersion medium. (1)
The breakdown of substances by biological action.
The ratio, expressed as a percentage of: (1) The amount of consolidation at a given time
within a soil mass, to (2) The total amount of consolidation obtainable under a given
stress condition.
D-5
-------
Degree of saturation
Density
Deposit
Depression curve
Detection monitoring
Diagenesis
Differential water capacity
Direct methods
Direct precipitation
Discharge area
Discharge velocity
Direct methods
Discontinuity
Disperse
Disposal
The extent or degree to which the voids in rock contain fluid (water, gas, or oil). Usually
expressed in percent related to total void or pore space.
The mass or quantity of a substance per unit volume. Units are kilograms per cubic
meter or grams per cubic centimeter.
Material left in a new position by a natural transporting agent such as water, wind, ice. or
gravity, or by the activity of man.
Record of profile of water table as a result of pumping.
A program of monitoring for the express purpose of determining whether or not there has
been a contaminant release to ground water.
The chemical and physical changes occurring in sediments before consolidation or while
in the environment of deposition.
The absolute value of the rate of change of water content with soil water pressure. The
water capacity at a given water content will depend on the particular desorption or
adsorption curve employed. Distinction should be made between volumetric and specific
water capacity.
Methods (e.g., boreholes and monitoring wells) which entail the excavation or drilling,
collection, observation, and analysis of geologic materials and water samples.
Water that falls directly into a lake or stream without passing through any land phase of
the runoff cycle.
An area in which there are upward components of hydraulic head in the aquifer. Ground
water is flowing toward the surface in a discharge area and may escape as a spring, seep,
or baseflow, or by evaporation and transpiration.
An apparent velocity, calculated from Darcy's law, which represents the flow rate at
which water would move through an aquifer if the aquifer were an open conduit. Also
called specific discharge.
Methods (e.g., boreholes and monitoring wells) which entail the excavation or drilling,
collection, observation, and analysis of geologic materials and water samples.
(a) Boundary between major layers of the Earth which have different seismic velocities.
(b) Interruption of the homogeneity of a rock mass (e.g. joints, faults, etc.).
(a) To break up compound particles, such as aggregates, into the individual component
particles, (b) To distribute or suspend fine particles, such as clay, in or throughout a
dispersion medium, such as water.
The discharge, deposit, injection, dumping, spilling, leaking, or placing of any solid
waste or hazardous waste into or on any land or water so that such solid waste or
hazardous waste or any constituent thereof may enter the environment or be emitted into
the air or discharged into any waters, including ground waters.
D-6
-------
Dissolved metals
Dissolution
DNAPL
DNAPL entry location
DNAPL site
DNAPL zone
Downhole geophysics
Drawdown
Drill cuttings
Drilling fluid
Dynamic equilibrium
Effective porosity
Effective solubility
Those constituents (metals) of an unacidified sample that pass through a 0.45pm
membrane filter.
The process where soluble organic components from DNAPL dissolves in groundwater
or infiltration and forms a groundwater contaminant plume. The duration of remediation
measures (either clean-up or containment) is determined by the 1) the rate of the
dissolution process that can be achieved in the field, and 2) the mass of soluble
components in the residual DNAPL trapped in the aquifer.
A Dense Non-aqueous Phase Liquid. Also known as free product or a sinking plume
(sinker).
The area where DNAPL has entered the subsurface.
A site where DNAPL has been released and is now present in the subsurface as an
immiscible phase.
The portion of a site affected by free-phase or residual DNAPL in the subsurface either
the vadose zone or saturated zone). The DNAPL zone has organics in the vapor phase
(unsaturated zone), dissolved phase (both unsaturated and saturated zone), and DNAPL
phase (both unsaturated and saturated zone).
Techniques that use a sensing device that is lowered into a borehole for the purpose of
characterizing geologic formations and their associated fluids. The results can be
interpreted to determine lithology, resistivity, bulk density, porosity, permeability, and
moisture content and to define the source, movement, and physical/chemical
characteristics of ground water.
A lowering of the water table of an unconfmed aquifer or the potentiometric surface of a
confined aquifer caused by pumping of ground water from wells.
Fragments or particles of soil or rock, with or without free water, created by the drilling
process.
A fluid (liquid or gas) that may be used in drilling operations to remove cuttings from the
borehole, to clean and cool the drill bit, and to maintain the integrity of the borehole
during drilling.
A condition in which the amount of recharge to an aquifer equals the amount of natural
discharge.
The amount of interconnected pore space through which fluids can pass, expressed as a
percent of bulk volume. Part of the total porosity will be occupied by static fluid being
held to the mineral surface by surface tension, so effective porosity will be less than total
porosity.
The actual aqueous solubility of an organic constituent in groundwater that is in chemical
equilibrium with a mixed DNAPL (a DNAPL containing several organic constituents).
The effective solubility of a particular organic chemical can be estimated by multiplying
its mole fraction in the DNAPL mixture by its pure phase solubility.
D-7
-------
Equilibrium constant
Equivalent diameter
Equipotential line
Equipotential surface
Equivalent weight
Field capacity
Fill
Film water
Fine texture
Fissure flow
Flow net
Flow, steady
Flow, unsteady
Fluid potential
Flush joint or flush coupled
Fracture
The number defining the conditions of equilibrium for a particular reversible chemical
reaction.
In sedimentation analysis, the diameter assigned to a non-spherical particle, it being
numerically equal to the diameter of a spherical particle of the same density and velocity
of fall.
A line in a two-dimensional ground-water flow field such that the total hydraulic head is
the same for all points along the line.
A surface in a three-dimensional ground-water flow field such that the total hydraulic
head is the same everywhere on the surface.
The concentration in parts per million of a solute multiplied by the valence charge and
then divided by its formula weight in grams.
The maximum amount of water that the unsaturated zone of a soil can hold against the
pull of gravity. The field capacity is dependent on the length of time the soil has been
undergoing gravity drainage.
Man-made deposits of natural soils or rock products and waste materials.
A layer of water surrounding soil particles and varying in thickness from 1 or 2 to
perhaps 100 or more molecular layers. Usually considered as that water remaining after
drainage has occurred because it is not distinguishable in saturated soils.
Consisting of or containing large quantities of the fine fractions, particularly of silt and
clay. (Includes all clay loams and clays; that is, clay loams, sandy clay loam, silty clay
loam, sandy clay, silty clay, and clay textural classes. Sometimes subdivided into clayey
texture and moderately fine texture.) See soil texture.
Flow of water through joints and larger voids.
The set of intersecting equipotential lines and flowlines representing two dimensional
steady flow through porous media.
The flow that occurs when, at any point in the flow field, the magnitude and direction of
the specific discharge are constant in time.
The flow that occurs when, at any point in the flow field, the magnitude or direction of
the specific discharge changes with time. Also called transient flow or nonsteady flow.
The mechanical energy per unit mass of fluid at any given point in space and time.
Casing or riser with ends threaded such that a consistent inside and outside diameter is
maintained across the threaded joints or couplings.
A break in a rock formation due to structural stresses. Faults, shears, joints, and planes of
fracture cleavage are all types of fractures.
D-8
-------
Fracture trace
Free energy
Free-phase DNAPL
Free water
(gravitational water)
(ground water)
(phreatic water)
Gamma-gamma
radiation log
Geohydrology
The surface representation of a fracture zone. It may be a characteristic line of vegetation
or linear soil-moisture pattern or a topographic sag.
A measure of the thermodynamic driving energy of a chemical reaction. Also known as
Gibbs free energy or Gibbs function.
Immiscible liquid exiting in the subsurface with a positive pressure such that it can flow
into a well. If not trapped in a pool, free phase DNAPL will flow vertically through an
aquifer or laterally down sloping fine-grained stratigraphic units. Also called mobile
DNAPL or continuous phase DNAPL.
Water that is free to move through a soil or rock mass under the influence of gravity.
A borehole log in which a source of gamma radiation as well as a detector are lowered
into the borehole. This log measures bulk density of the formation and fluids.
Science of the occuffence, distribution, and movement of water below the surface of the
Earth.
Geomorphology
The description of the present exposed surfaces of the crust of the Earth, and seeks to
interpret these surfaces in terms of natural processes (chiefly erosion) which lead or have
led to their formation.
Geophysical borehole logging
Geophysics
Geotechnical
Gradation (grain-size
distribution) (texture)
Grading
Grain-size analysis
(mechanical analysis)
(particle/size analysis)
Granule
Gravelpack
See Downhole Geophysics.
The study of all the gross physical properties of the Earth and its parts, particularly
associated with the detection of the nature and shape of unseen subsurface rock bodies by
measurement of such properties and property contrasts. Small scale applied geophysics
is now a major aid in geological reconnaissance.
Pertaining to Geotechnics, which is the application of scientific methods to problems in
engineering geology.
The proportions by mass of a soil or fragmented rock distributed in specified particle-size
ranges.
A'well-graded' sediment containing some particles of all sizes in the range concerned.
Distinguish from'well sorted', which describes a sediment with grains of one size.
The process of determining grainsize distribution.
A natural soil aggregate or ped which is relatively nonporous. See soil structure and soil
structure types.
Common nomenclature for the preferred terminology, primary filter of a well (see
primary filter pack).
D-9
-------
Gravel
Ground water
Ground-water basin
Ground water, confined
Ground-water flow
Ground water level
Ground-water mining
Ground water, perched
Ground water regime
(ground water)
Ground water, unconfined
Round or semirounded particles of rock that will pass a 3-in. (76.2 mm) sieve and be
retained on a No. 4 (4.75 mm) U.S. standard sieve.
(a) The water contained in interconnected pores located below the water table in an
unconfined aquifer or located in a confined aquifer, (b) The portion of the total
precipitation which at any particular time is either passing through or standing in the soil
and the underlying strata and is free to move under the influence of gravity.
A rather vague designation pertaining to a ground-water reservoir which is more or less
separate from neighboring ground-water reservoirs. A ground-water basin could be
separated from adjacent basins by geologic boundaries or by hydrologic boundaries.
The water contained in a confined aquifer. Pore-water pressure is greater than
atmospheric at the top of the confined aquifer.
The movement of water through openings in sediment and rock which occurs in the zone
of saturation.
The level below which the rock and subsoil, to unknown depths, are saturated.
The practice of withdrawing ground water at rates in excess of the natural recharge.
The water in an isolated, saturated zone located in the zone of aeration. It is the result of
the presence of a layer of material of low hydraulic conductivity, called a perching bed.
Perched ground water will have a perched water table.
Water below the land surface in a zone of saturation.
The water in an aquifer where there is a water table.
Grout
Head, total
Heterogeneous
Homogeneous
Horizon
Hydration
Hydraulic conductivity
A low permeability material placed in the annulus between the well casing or riser pipe
and the borehole wall (i.e., in a single cased monitoring well), or between the riser and
casing (i.e., in a multi-cased monitoring well), to maintain the alignment of the casing
and riser and to prevent movement of ground water or surface water within the annular
space.
The sum of the elevation head, the pressure head, and the velocity head at a given point
in an aquifer.
Pertaining to a substance having different characteristics in different locations. A
synonym is nonuniform.
Pertaining to a substance having identical characteristics everywhere. A synonym is
uniform.
See soil horizon.
The physical binding of water molecules to ions, molecules, particles, or other matter.
A coefficient of proportionality describing the rate at which water can move through a
D-10
-------
Hydraulic diffusivity
Hydraulic gradient
Hydraulic head
Hydrochemical facies
Hydrodynamic dispersion
Hydrograph
Hydrogeology
Hydrologic equation
Hydrologic unit
Hydrostatic pressure
Hydrostratigraphic unit
Hygroscopic water
Ideal gas
permeable medium. The density and kinematic viscosity of the water must be considered
in determining hydraulic conductivity.
A property of an aquifer or confining bed defined as the ratio of the transmissivity to the
storativity.
The change in total head with a change in distance in a given direction. The direction is
that which yields a maximum rate of decrease in head.
The elevation with respect to a specified reference level at which water stands in a
piezometer connected to the point in question in the soil. Its definition can be extended
to soil above the water table if the piezometer is replaced by a tensiometer. The
hydraulic head in systems under atmospheric pressure may be identified with a potential
expressed in terms of the height of a water column. More specifically it can be identified
with the sum of gravitational and capillary potentials, and may be termed the hydraulic
potential.
Bodies of water with separate but distinct chemical compositions contained in an aquifer.
The process by which ground water containing a solute is diluted with uncontaminated
ground water as it moves through an aquifer.
A graph that shows some property of ground water or surface water as a function of time.
The study of the natural (and artificial) distribution of water in rocks, and its relationship
to those rocks. Inasmuch as the atmosphere is a continuation of the hydrosphere, and is
in physical and chemical balance with it, there is a close connection with meteorology.
An expression of the law of mass conservation for purposes of water budgets. It may be
stated as inflow equals outflow plus or minus changes in storage.
Geologic strata that can be distinguished on the basis of capacity to yield and transmit
fluids. Aquifers and confining units are types of hydrologic units. Boundaries of a
hydrologic unit may not necessarily correspond either laterally or vertically to
lithostratigraphic formations.
A state of stress in which all the principal stresses are equal (and there is no shear stress).
A formation, part of a formation, or a group of formations in which there are similar
hydrologic characteristics allowing for grouping into aquifers or confining layers.
Water adsorbed by a dry soil from an atmosphere of high relative humidity, water
remaining in the soil after "air-drying" or water held by the soil when it is in equilibrium
with an atmosphere of a specified relative humidity at a specified temperature, usually
98% of relative humidity at 25 degrees Centigrade.
A gas having a volume that varies inversely with pressure at a constant temperature and
that also expands by 1/273 of its volume at 0°C for each degree rise in temperature at
constant pressure.
D-ll
-------
Immobilization
Indirect methods
Infiltration
Infiltration capacity
Infiltration rate
Injection well
Interflow
Intermediate zone
Intrinsic permeability
Ion exchange
Isotropy
Jetting
Kinematic viscosity
Laminar flow
Law of mass action
Leach
The conversion of an element from the inorganic to the organic form in microbial tissues
or in plant tissues.
Methods which include the measurement or remote sensing of various physical and/or
chemical properties of the earth (e.g., electromagnetic conductivity, electrical resistivity,
specific conductance, geophysical logging, aerial photography).
The flow of water downward from the land surface into and through the upper soil layers.
The maximum rate at which infiltration can occur under specific conditions of soil
moisture. For a given soil, the infiltration capacity is a function of the water content.
(a) A soil characteristic determining or describing the maximum rate at which water can
enter the soil under specified conditions, including the presence of an excess of water, (b)
The rate at which a soil under specified conditions can absorb falling rain or melting
snow; expressed in depth of water per unit time (cm/sec; in/hr).
A well drilled and constructed in such a manner that water can be pumped into an aquifer
in order to recharge it.
The lateral movement of water in the unsaturated zone during and immediately after a
precipitation event. The water moving as interflow discharges directly into a stream or
lake.
That part of the unsaturated zone below the root zone and above the capillary fringe.
Pertaining to the relative ease with which a porous medium can transmit a liquid under a
hydraulic or potential gradient. It is a property of the porous medium and is independent
of the nature of the liquid or the potential field.
A process by which an ion in a mineral lattice is replaced by another ion which was
present in an aqueous solution.
The condition in which hydraulic properties of the aquifer are equal in all directions.
When applied as a drilling method, water is forced down through the drill rods or casing
and out through the end aperture. The jetting water then transports the generated cuttings
to the ground surface in the annulus of the drill rods or casing and the borehole. The term
jetting may also refer to a development technique (see well screen jetting).
The ratio of dynamic viscosity to mass density. It is obtained by dividing dynamic
viscosity by the fluid density. Units of kinematic viscosity are square meters per second.
That type of flow in which the fluid particles follow paths that are smooth, straight, and
parallel to the channel walls. In laminar flow, the viscosity of the fluid damps out
turbulent motion. Compare with Turbulent flow.
The law stating that for a reversible chemical reaction the rate of reaction is proportional
to the concentrations of the reactants.
To cause water or other liquid to percolate through soil.
D-12
-------
Leaky confining layer
Lineament
Lysimeter
Lysimeter
Maximum contaminant level
Molality
Morphology
Multi-cased well
Multiport systems
N-value
Natural gamma radiation log
Neat cement
Negative pressure
Neutron log
Observation well
Overburden
Overland flow
Oxidation-reduction potential
A low-penneability layer that can transmit water at sufficient rates to furnish some
recharge to a well pumping from an underlying aquifer. Also called aquitard.
A natural linear surface feature longer than 1500 meters.
A field device containing a soil column and vegetation which is used for measuring
actual evapotranspiration.
(a) A device for measuring percolation and leaching losses from a column of soil under
controlled conditions, (b) A device for measuring gains (precipitation and condensation)
and losses (evapotranspiration) by a column of soil.
The highest concentration of a solute permissible in a public water supply as specified in
the National Interim Primary Drinking Water Standards for the United States.
A measure of chemical concentration. A one-molal solution has one mole of solute
dissolved in 1000 grams of water. One mole of a compound is its formula weight in
grams.
See soil morphology.
A well constructed by using successively smaller diameter casings with depth.
A single hole device in which points are installed that are capable of sampling or
measuring at multiple levels within a formation or series of formations.
The number of blows required to drive the sampler of the Standard Penetration test its
last 12 inches (300 mm).
A borehole log that measures the natural gamma radiation emitted by the formation
rocks. It can be used to delineate subsurface rock types.
A mixture of Portland cement (ASTM 150) and water.
A pressure less than the local atmospheric pressure at a given point.
A borehole log obtained by lowering a radioactive element, which is a source of neutrons,
and a neutron detector into the well. The neutron log measures the amount of water
present; hence, the porosity of the formation.
A nonpumping well used to observe the elevation of the water table or the potentiometric
surface. An observation well is generally of larger diameter than a piezometer and
typically is screened or slotted throughout the thickness of the aquifer.
The loose soil, sand, silt, or clay that overlies bedrock.
The flow of water over a land surface due to direct precipitation. Overland now generally
occurs when the precipitation rate exceeds the infiltration capacity of the soil and
depression storage is full. Also called Horton overland flow.
The potential required to transfer electrons from the oxidant to the reductant and used as
2 qualitative measure of the state of oxidation in wastewater treatment systems.
D-13
-------
Packer
Parent material
Particle density
Particle size
A transient or dedicated device placed in a well that isolates or seals a portion of the well,
well annulus, or borehole at a specific level.
The unconsolidated and more or less chemically weathered mineral or organic matter
from which the solum of soil is developed by pedogenic processes.
The mass per unit volume of the soil particles. In technical work, usually expressed as
grams per cubic centimeter. See bulkeny, soil.
The effective diameter of a particle measured by sedimentation, sieving, or micrometric
methods.
Particle-size analysis
Particle-size distribution
Penetrability
Percent saturation
(degree of saturation)
Perched water table
Percolation
Permeability, soil
Determination of the various amounts of the different separates in a soil sample, will
usually be sedimentation, sieving, micrometry, or combinations of these methods.
The amounts of the various soil separates in a soil sample, usually expressed as weigbt
percentages.
The ease with which a probe can be pushed into the soil. (May be expressed in units of
distance, speed, force, or work depending on the type of penetrometer used).
The ratio, expressed as a percentage, of. (a) The volume of water in a given soil or rock
mass, to (b) The total volume of intergranular space (voids).
A water table usually of limited area maintained above the normal free water elevation by
the pressure of an intervening relatively impervious confining stratum.
The flow or trickling of a liquid downward through a contact or filtering medium. The
liquid may or may not fill the pores of the medium.
(a) The ease with which gases, liquids, or plant roots penetrate or pass through a bulk
mass of soil or a layer of soil. Since different soil horizons vary in permeability, the
particular horizon under question should be designated, (b) The property of a porous
medium itself that relates to the ease with which gases, liquids, or other substances can
pass through it. Previously, frequently considered the "k" in Darcy's law. See Darcy's
law and soil water.
Permeameter
pH, soil
Physical properties (of soil)
A laboratory device used to measure the intrinsic permeability and hydraulic conductivity
of a soil or rock sample.
The negative logarithm of the hydrogen-ion activity of a soil. The degree of acidity (or
alkalinity) of a soil as determined by means of a glass, quinhydrone, or other suitable
electrode or indicator at a specified moisture content or soil-water ratio, and expressed in
terms of the pH scale.
Those characteristics, processes, or reactions of a soil which are caused by physical
forces and which can be described by, or expressed in, physical terms or equations.
Sometimes confused with and difficult to separate from chemical properties; hence, the
terms "physical-chemical" or "physiochemical". Examples of physical properties are
bulk density, water-holding capacity, hydraulic conductivity, porosity, pore-size
distribution, etc.
D-14
-------
Phreatic water
Piezometer
Piezometer nest
Piezometric surface
Piping
Piston sampler
Plume
Pool and lens
Pollutant
Pore-size distribution
Pore space
Porosity
Potentiometric surface
Primary filter pack
Profile, soil
Water in the zone of saturation.
A nonpumping well, generally of small diameter, which is used to measure the elevation
of the water table or potentiometric surface A piezometer generally has a short well
screen through which water can enter.
A set of two or more piezometers set close to each other but screened to different depths.
(a) The surface at which water will stand in a series of piezometers, (b) An imaginary
surface that everywhere coincides with the static level of the water in the aquifer.
An underground flow of water with a sufficient pressure gradient to cause scour along a
preferred path.
A tube with an internal piston used for obtaining relatively undisturbed samples from
cohesive soils.
The zone of contamination containing organics in the dissolved phase. The plume
usually will originate from the DNAPL zone and extend downgradient for some distance
depending on site hydrogeologic and chemical conditions. To avoid confusion, the term
"DNAPL plume" should not be used to describe a DNAPL pool; "plume" should be used
only to refer to dissolved-phase organics.
A zone of free-phase DNAPL at the bottom of an aquifer. A lens is a pool that rests on a
fine-grained stratigraphic unit of limited areal extent. DNAPL can be recovered from a
pool or a lens if a well is placed in the right location.
Any solute or cause of change in physical properties which renders water unfit for a
given use.
The volume of the various sizes of pores in a soil. Expressed as percentages of the bulk
volume (soil plus pore space).
The volume between mineral grains in a porous medium.
The ratio of the volume of void spaces in a rock or sediment to the total volume of the
rock or sediment.
A surface that represents the level to which water will rise in tightly cased wells. If the
head varies significantly with depth in the aquifer, then there may be more than one
potentiometric surface. The water table is a particular potentiometric surface for an
unconfined aquifer.
A clean silica sand or sand and gravel mixture of selected grain size and gradation that is
installed in the annular space between the borehole wall and the well screen, extending an
appropriate distance above the screen, for the purpose of retaining and stabilizing the
particles from the adjacent strata. The term is used in place of "gravel pack."
A vertical section of the soil through all its horizons and extending into the parent
material.
D-15
-------
PTFE tape
Pumping cone
Pumping test
Radial flow
Rating curve
Reaction, soil
Recharge area
Recharge basin
Recharge boundary
Recovery
Regolith
Representativeness
Representative point
Riser
Residual
Joint sealing tape composed of polytetrafluoroethylene.
The area around a discharging well where the hydraulic head in the aquifer has been
lowered by pumping. Also called cone of depression.
A test made by pumping a well for a period of time and observing the change in
hydraulic head in the aquifer. A pumping test may be used to determine the capacity of
the well and the hydraulic characteristics of the aquifer. Also called aquifer test.
The flow of water in an aquifer toward a vertically oriented well.
A graph of the discharge of a river at a particular point as a function of the elevation of
the water surface.
The degree of acidity or alkalinity of a soil, usually expressed as a pH value. Descriptive
terms commonly associated with certain ranges in pH are: extremely acid, less than 4.5;
very strongly acid, 4.5 to 6.0; slightly acid, 6.1 to 6.5; neutral, 6.6 to 7.3; slightly alkaline,
7.4 to 7.8; moderately alkaline, 7.9 to 8.4; strongly alkaline, 8.5 to 9.0; and very strongly
alkaline, greater than 9. 1.
An area in which there are downward components of hydraulic head in the aquifer.
Infiltration moves downward into the deeper parts of an aquifer in a recharge area.
A basin or pit excavated to provide a means of allowing water to soak into the ground at
rates exceeding those that would occur naturally.
An aquifer system boundary that adds water to the aquifer. Streams and lakes are typical
recharge boundaries.
The rise in water level in a pumping well and nearby observation wells after ground-
water pumpage has ceased.
The upper part of the earth's surface that has been altered by weathering processes. It
includes both soil and weathered bedrock.
The characteristic of a specific scientific experiment that makes it an adequate sample of
the general case.
a) A location in surface waters or ground waters at which specific conditions or
parameters may be measured in such a manner as to characterize or approximate the
quality or condition of the water body; or b) A location in process or waste waters at
which specific conditions or parameters are measured and will adequately reflect the
actual condition of those waters or waste waters for which analysis was made.
The pipe extending from the well screen to or above the ground surface.
Immiscible phase liquid held in the pore spaces or fractures by capillary forces (negative
pressure on DNAPL), Residual will remain trapped within the pore of the porous media
unless the viscous forces (caused by the dynamic force of water against the DNAPL ) are
greater than the capillary forces holding the DNAPL in the pore. At most sites the
hydraulic gradient required to mobilize all of the residual trapped in an aquifer is usually
much greater than can be produced by wells or trenchers.
D-16
-------
Residual saturation
Resistivity log
Rock
Runoff
Sand
Sand model
Sanitary landfill
Saturated zone
Saturation
Secondary filter pack
Secondary porosity
Sediment sump
Sedimentation
Seepage velocity
Seismic refraction
The fraction of available pore space containing residual DNAPLS, or the saturation level
where free-phase DNAPL becomes residual DNAPL. In the vadose zone, residual
saturation range up to 20% of total pore volume while in the saturated zone residual
saturations range up to 50% of total pore volume.
A borehole log made by lowering two current electrodes into the borehole and measuring
the resistivity between two additional electrodes. It measures the electrical resistivity of
the formation and contained fluids near the probe.
Natural solid mineral matter occurring in large masses or fragments.
The total amount of water flowing in a stream. It includes overland flow, return flow,
interflow, andbaseflow.
(a) A soil particle between 0.05 and 2.0 mm in diameter, (b) Any one of five soil
separates, namely: very coarse sand, coarse sand, medium sand, fine sand, and very fine
sand. See soil separates, (c) A soil textural class. See soil texture.
A scale model of an aquifer, which is built using a porous medium to demonstrate
ground-water flow.
The disposal of solids and, in some instances, semisolid and liquid wastes by burying the
material to shallow depths, usually in unconsolidated materials.
The zone in which the voids in the rock or soil are filled with water at a pressure greater
than atmospheric. The water table is the top of the saturated zone in an unconfined
aquifer.
A condition reached by a material, whether it be in solid, gaseous, or liquid state, that
holds another material within itself in a given state in an amount such that no more of
such material can be held within it in the same state. The material is then said to be
saturated on in a condition of saturation.
A clean, uniformly graded sand that is placed in the annulus between the primary filter
pack and the over-lying seal, or between the seal and overlying grout backfill or both, to
prevent movement of seal, or grout, or both into the primary filter pack.
The porosity developed in a rock after its deposition or emplacement, through such
processes as solution or fracturing.
A blank extension beneath the well screen used to collect fine-grained material from the
filter pack and adjacent strata. The term is synonymous with rat trap or tail pipe.
The process of subsidence and deposition of suspended matter carried by water,
wastewater, or other liquids, by gravity. It is usually accomplished by reducing the
velocity of the liquid below the point at which it can transport the suspended material.
The actual rate of movement of fluid particles through porous media.
A method of determining subsurface geophysical properties by measuring the length of
time it takes for artificially generated seismic waves to pass through the ground.
D-17
-------
Silt
Single-cased well
Single-point resistance log
Site assessment
Slug test
Soil
Soil auger
Soil mineral
Soil moisture
Soil morphology
Soil physics
Soil science
Soil separates
(a) A soil separate consisting of particles between 0.005 and 0.002 mm in equivalent
diameter. See soil separates, (b) A soil texture class. See soil texture.
A monitoring well constructed with a riser but without an exterior casing. (9)
A borehole log made by lowering a single electrode into the well with the other electrode
at the ground surface. It measures the overall electrical resistivity of the formation and
drilling fluid between the surface and the probe.
A formal means of exploring and characterizing a proposed waste management facility or
location so that all physical factors are identified and so quantified as to serve as the basis
of an environmentally sound design and operational plan.
An aquifer test made by either pouring a small instantaneous charge of water into a well
or by withdrawing a slug of water from the well. A synonym for this test, when a slug of
water is removed from the well, is a bail-down test.
(a) The unconsolidated mineral material on the immediate surface of the Earth that serves
as a natural medium for the growth of land plants, (b) The unconsolidated mineral matter
on the surface of the Earth that has been subjected to and influenced by genetic and
environmental factors of: parent material, climate (including moisture and temperature
effects), macro- and microorganisms, and topography, all acting over a period of time and
producing a product, soil, that differs from the, material from which it is derived in many
physical, chemical, biological, and morphological properties and characteristics.
A tool for boring into the soil and withdrawing a small sample for field or laboratory
observation. Soil augers may be classified into several types as follows: (a) Those with
worm-type bits, unenclosed; (b) Those with worm-type bits enclosed in a hollow
cylinder; and those with a hollow cylinder with a cutting edge at the lower end.
(a) Any mineral that occurs as a part of or in the soil, (b) A natural inorganic compound
with definite physical, chemical, and crystalline properties (within the limits of
isomorphism), that occurs in the soil.
The water contained in the unsaturated zone.
(a) The physical constitution, particularly the structural properties of the soil profile as
exhibited by the kinds, thickness, and arrangement of the horizons in the profile, and by
the texture, structure, consistency, and the porosity of each horizon, (b) The structural
characteristics of the soil or any of its parts.
The organized body of knowledge concerned with the physical characteristics of soil and
with the methods employed in their determinations.
That science dealing with soils as a natural resource on the surface of the Earth including
soil formation, classification, and mapping, and physical, chemical, biological, and
fertility properties of soil per se; and these properties in relation to their management.
Mineral particles, < 2.0 mm in equivalent diameter, ranging between specified size limits.
The names and size limits of separates recognized in the U.S.D.A. system are: very
coarse sand, 2.0 to 1.0 mm; coarse sand, 1.0 to 0.5 mm; medium sand, 0.5 to 0.25 mm;
D-18
-------
Soil solution
Soil structure
Soil suction
Soil texture
fine sand, 0.25 to 0.10 mm; very fine sand, 0.10 to 0.05 mm; silt, 0.05 to 0.002 mm; and
clay, < 0.002 mm. The U.S.C.S. particle and size range are as follows: coarse sand, 2.0
to 4.76 mm; medium sand, 0.42 to 2.0 mm; fine sand, 0.074 to 0.42 mm; fines (silt and
clay), < 0.074 mm. (Note: U.S.C.S. silt and clay designations are determined by response
of the soil to manipulation at various water contents rather than by measurement of size.)
The aqueous liquid phase of the soil and its solutes.
The combination or arrangement of primary soil particles into secondary particles, units,
or peds. These secondary units may be, but usually are not, arranged in the profile in
such a manner as to give a distinctive, characteristic pattern. The secondary units are
characterized and classified on the basis of size, shape, and degree of distinctness into
classes, types, and grades, respectively.
A measure of the force of water retention in unsaturated soil. Soil suction is equal to a
force per unit area that must be exceeded by an externally applied suction to initiate water
flow from the soil. Soil suction is expressed in standard pressure terms.
The relative proportion of the various soil separates in a soil as described by the classes
of soil texture.
Soil water diffusivity
The hydraulic conductivity divided by the differential water capacity (care being taken to
be consistent with units), or the flux of water per unit gradient of moisture content in the
absence of other force fields.
Soil water pressure
Soil water
The pressure (positive or negative), relative to the external gas pressure on the soil water,
to which a solution identical in composition to the soil water must be subjected in order
to be in equilibrium through a porous permeable wall with the soil water. May be
identified with the capillary potential defined above.
A general term emphasizing the physical rather than the chemical properties and bebavior
of the soil solution.
Soil-moisture tension
Solid waste disposal facilities
Solubility product
Specific capacity
See moisture tension (or pressure).
A facility or part of a facility at which solid waste is intentionally placed into or on any
land or water, and at which waste will remain after closure.
The equilibrium constant that describes a solution of a slightly soluble salt in water.
An expression of the productivity of a well, obtained by dividing the rate of discharge of
water from the well by the drawdown of the water level in the well. Specific capacity
should be described on the basis of the number of hours of pumping prior to the time the
drawdown measurement is made. It will generally decrease with time as the drawdown
increases.
Specific electrical conductance
Specific retention
The ability of water to transmit an electrical current. It is related to the concentration and
charge of ions present in the water.
The ratio of the volume of water the rock or sediment will retain against the pull of
gravity to the total volume of the rock or sediment.
D-19
-------
Specific weight
Specific yield
Spontaneous potential log
Stability
Stagnation point
Subsoil
Static water level
Storage, specific
Storativity
Stratified
Structure
Subsoil
Surface sealing
Suspended metals
Target monitoring zone
The weight of a substance per unit volume. The units are newtons per cubic meter.
The ratio of the volume of water a rock or soil will yield by gravity drainage to the
volume of the rock or soil. Gravity drainage may take many months to occur.
A borehole log made by measuring the natural electrical potential which develops
between the formation and the borehole fluids.
The condition of a structure or a mass of material when it is able to support the applied
stress for a long time without suffering any significant deformation or movement that is
not released by the release of stress.
A place in a ground-water flow field at which the ground water is not moving. The
magnitude of vectors of hydraulic head at the point are equal but opposite in direction.
In general concept, that part of the soil below the depth of plowing. (2)
The elevation of the top of a column of water in a monitoring well or piezometer that is
not influenced by pumping or conditions related to well installation, hydrologic testing,
or nearby pumpage.
The amount of water released from or taken into storage per unit volume of a porous
medium per unit change in head.
The volume of water an aquifer releases from or takes into storage per unit surface area
of the aquifer per unit change in head It is equal to the product of specific storage and
aquifer thickness. In an unconfined aquifer, the Storativity is equivalent to the specific
yield. Also called storage coefficient.
Arranged in strata, or layers. The term refers to geologic material. Layers in soils that
result from the processes of soil formation are called horizons; those inherited from the
parent material are called strata.
One of the larger features of a rock mass, like bedding, foliation, jointing, cleavage, or
brecciation; also the sum total of such features as contrasted with texture. Also, in a
broader sense, it refers to the structural features of an area such as anticlines or synclines.
See also soil structure.
In general concept, that part of the soil below the depth of plowing. (2)
The orientation and packing of dispersed soil particles in the immediate surface layer of
the soil, rendering it relatively impermeable to water.
Those constituents (metals) of an unacidified sample that are retained by a 0.45 mm
membrane filter.
The ground water flow path from a particular area or facility into which monitoring wells
will be screened. The target monitoring zone should be a stratum (strata) in which there
is a reasonable expectation that a vertically placed well will intercept migrating
contaminants.
D-20
-------
Target
Tensiometer
Tension
Tension, soil water
Test pit
Total metals
Throughflow
Time-series
Total potential (of soil water)
Transects
Transmissivity
Tremie pipe
Turbidity
Turbulent flow
Undisturbed sample
Unsaturated flow
In detection monitoring programs, the ground water flow path from a particular area or
facility into which monitoring wells will be screened. The target monitoring zone should
be a stratum (strata) in which there is a reasonable expectation that a vertically placed
well will intercept migrating contaminants.
(a) A device used to measure the soil-moisture tension in the unsaturated zone, (b) A
device for measuring the negative pressure (or tension) of water in soil in situ; a porous,
permeable ceramic cup connected through a tube to a manometer or vacuum gauge.
The condition under which pore water exists at a pressure less than atmospheric.
The expression, in positive terms, of the negative hydraulic pressure of soil water.
A shallow excavation made to characterize Transpiration: water loss from leaves and
other plant the subsurface.
The concentration of metals determined on an unfiltered sample after vigorous digestion,
or the sum of the concentrations of metals in both dissolved and suspended fractions.
The lateral movement of water in an unsaturated zone during and immediately after a
precipitation event. The water from throughflow seeps out at the base of slopes and then
flows across the ground surface as return flow, ultimately reaching a stream or lake.
A series of statistical data collected at regular intervals of time; a frequency distribution
in which the independent variable is time.
The amount of work that must be done per unit quantity of pure water in order to
transport reversibly and isothermally an infinitesimal quantity of water from a pool of
pure water, at a specified elevation and at atmospheric pressure, to the soil water (at the
point under consideration). The total potential (of soil water) consists of the following:
In ecology, a sample area (usually elongate or linear) chosen as the basis for studying a
particular assemblage of organisms.
The rate at which water of a prevailing density and viscosity is transmitted through a unit
width of an aquifer or confining bed under a unit hydraulic gradient. It is a function of
properties of the liquid, the porous media, and the thickness of the porous media.
A pipe or tube that is used to transport filter pack materials and annular sealant materials
from the ground surface into the borehole annulus or between casings and casings or riser
pipe of a monitoring well.
Heavily suspended and colloidal organic and inorganic material in water.
That type of flow in which the fluid particles move along very irregular paths.
Momentum can be exchanged between one portion of the fluid and another. Compare
with Laminar flow.
A soil sample that has been obtained by methods in which every precaution has been
taken to minimize disturbance to the sample.
The movement of water in a soil which is not filled to capacity with water.
D-21
-------
Unsaturated zone
Uppermost aquifer
Vadose water
Viscosity
Void ratio
Voids
Water content
Water table
Weathering
Well completion diagram
Well, fully penetrating
Well interference
Well, partially penetrating
Well screen
Well screen jetting
(Hydraulic Jetting)
The zone between the land surface and the water table. It includes the root zone,
intermediate zone, and capillary fringe. The pore spaces contain water at less than
atmospheric pressure, as well as air and other gases. Saturated bodies, such as perched
ground water, may exist in the unsaturated zone.
The geologic formation nearest the natural ground surface that is an aquifer, as well as
lower aquifers that are hydraulically interconnected with this aquifer within the facility's
property boundary.
Water in the zone of aeration.
The property of a fluid describing its resistance to flow. Units of viscosity are newton-
seconds per meter squared or pascal-seconds. Viscosity is also known as dynamic
viscosity.
The ratio of. (a) The volume of void space, to (b) The volume of solid particles in a given
soil mass.
Entities which are interconnected with each other either through voids of dissimilar size
and shape, through narrow necks, or through intersection with voids of similar size and
shape.
The ratio of the volume of soil moisture to the total volume of the soil. This is the
volumetric water content, also called volume wetness.
The surface in an unconfined aquifer or confining bed at which the pore water pressure is
atmospheric. It can be measured by installing shallow wells extending a few feet into the
zone of saturation and then measuring the water level in those wells.
All physical and cbemical changes produced in rocks, at or near the earth's surface, by
atmospheric agents.
A record that illustrates the details of a well installation..
A well drilled to the bottom of an aquifer, constructed in such a way that it withdraws
water from the entire thickness of the aquifer.
The result of two or more pumping wells, the drawdown cones of which intercept. At a
given location, the total well interference is the sum of the drawdowns due to each
individual well.
A well constructed in such a way that it draws water directly from a fractional part of the
total thickness of the aquifer. The fractional part may be located at the top or the bottom
or anywhere in between the aquifer.
A filtering device used to retain the primary or natural filter pack; usually a cylindrical
pipe with openings of a uniform width, orientation, and spacing.
When jetting is used for development, a jetting tool with nozzles and a high pressure
pump is used to force water outwardly through the screen, the filter pack, and sometimes
into the adjacent geologic unit.
D-22
-------
Appendix E- Sample Forms
E-l
-------
Ground Water Sampling Log
Project Site Well No. Date
Well Dep
Sampling
Water Le
Measurin
Sampling
Pump Ra
Time
th Scr
Device
vel (before sampl
g Point
een Lenc
th Well Diameter
Tu
ng)
Ding type
Water Le
Oth«
vel (after samplinc
sr Info
Casing T'
/pe
3)
Personn
te
el
PH
Temp
Cond
Dis.O 2
Turb
QConc
Notes
Type of Samples Collected
Information: 2 in = 617 ml/ft, 4 in = 2470 ml/ft; Volcy| = 7tr2h, Volsphere= 4/37tr3
E-2
-------
Ground Water Sampling Log
(with automatic data logging for most water quality parameters
Proiect
Well Depth
Sampling Device
Measuring Point
Site
Screen Length
Well No.
Well Diameter
Tubing type
Other Info
Date
Casing Type
Sampling Personnel
Time
Pump Rate
Turbidity
Alkalinity
[ ] Cone
Notes
Type of Samples Collected
Information: 2 in => 617 ml/ft, 4 in => 2470 ml/ft; Volcy| = 7tr2h, Volsphere= 4/37tr3
E-3
-------
WILLIAMS AIR FORCE BASE
SAMPLING INFORMATION FORM
Job Number:
Sample I.D._
Sampling Method:
n Pump Q Submersible Q Bla
Ql Bailer Q PVC Q Tefl
Field Instruments:
D pH
n Temperature
n Conductivity
l~l Turbidity
CH Dissolved Oxygen
Well Number:
dder Q Piston Q Peristaltic
on Q Stainless
Brand
Serial /I.D. #
Measured Water Level Depth (before sampling) =
Measured Water Level Depth (after sampling) =
Description of Sample Water:
(include color, odor, etc.)
Date
Time
Initials
Field Filtration Performed
SAMPLE MEASUREMENTS
Temp. (°C)
Print Name:
Signature:
pH (S.U.)
Cond.
(units)
Turbidity
(NTU)
Dissolved
02(mg/l)
Date
Time
Initials
Print Name:
Signature:
E-4
-------
WELL PURGING - FIELD WATER QUALITY MEASUREMENTS FORM* 8U.1a,r.aiwL_0(_
\
i
!
I
i
1
\
CI
2
I
'i
U
1
C/
c
1
]
a
I
|
*e
1
'
.
i
i
^s
g
i
u,
c;
Q
1
1
I
£
1
i
'5
1
•5
2
a
C
1
I
j
1
a
| Submersible pump with direct line to Flow Cell used for all measurements? (Y, N)* Pump Intake or Bailer set at '(ft. below MP)
j
j
o!
*t
fil
*
1
ifl
Jig
Ujl
Cumulative Volume
Purged (specify units
below) ( )
lifiF
if'
i
1
i
j
a.
0
Date
CALIBRATION Time: military
Comment*
:
|
i
CD
to
Q
£
•q
1
o
>
CO
c
c
a
o
c
"1
:
3
i
5
I Measurements protocol fo
1
LL.
E-5
-------
°
•s
.c
fvl"
"B
CM
0)
•JC
oi
O
Cr5
1-
z
UJ
^^
UJ
rr
^
V)
<
UJ
^
fc
_J
^f
3
a
UJ
^~
<
5
Q
UJ
U_
0
z
0
a.
^
LLJ
c
E
0
c
!
E
c
I
c
"c
0
(J
I
c
^
'c
0
it
_c
5
c
"c
c
s
^
n
.
s
a
c
'z
a
i
«
c
HI
E
E
0
U
«
*^
ra
E
Qi
d)
u
c
ra
T3
'3
O
ID
5 ^
3 ~ '
l-
^
§~3>
E_
_
££
X
Q.
"c5 5 o
^ •o i
Hi 0 3
O ~
E "
|~
Calibration
"ro
Q
0)
E
i-
1
•a
c
c
o
1
8
en
"ol
0
c
-c
Q
"a
c
C
£
E-6
-------
GROUND WATER SAMPLING INFORMATION FORM*
^^^^^^^^^^^^" General Information ^^^^^^^^^"
Location (Site/Facility Name)
Project Name/#
Field Personnel
Sampling Organization
Weather C£ ?
Sampling Point (common name)
Type (mon. well, spring, etc.)
Field Sample (Event) ID#*
Facility ID (for IGWIS data entry)
Station ID (for IGWIS data entry)
Sheet of_
Side 1 of 2*
Read from left to right |]^p
Well Depth (ft below MP)
Static Depth to Water (below MP)
Water Column Length (L) (ft)
Condition: Securely Locked?
L.
Casinq Diameter (inches'!
Static DTW (ft below GS1
10 °1 "' One WC Volume (cu ft) '° ' "'
YorN Station (Well) Damaged? YorN
top - bottom
np=n Interval (d=pth h=hw fss) f
Pat0 Tim0
One WC Volume (nalsl
Surface Contamination (visible) vorN
^
^
r
PID/FID Reading @ Wellhead*
Free Product (circle UIAPLOTDNAPLT
Well Purging Equipment
Purging Date/Time
Pump/Bailer Intake Set at
Ami. Purged before Sampling
L.
Concentration ppm
Detected/Sampled? YorN / YorN
PumD. bailer?
Start £> /
Feet below MP
Gals/WC Volumes /
Rankgrnnnri Cnnn
Appearance
Type*
Finish $
Avg Purge Rate
Purge Protocol of WCV's met?
1
com
dom
YorN
^
Field Water-Quality Measurements and Observations
Date/Time Measurements Began .
1 Purge Rate for Measurements (gpm) .
Submersible Pump with direct line to Flow Cell used for all Field Water Quality Measurements?
All Field Measurement Instruments Calibrated according to Protocol?
All Field Water Quality Parameters Stabilized according to Protocol Criteria just before filling sample containers? .
The Measurements below Represent: (\) stabiliization, (2) sample water collected, (3) both a and 2, (4) other*: .
Sample Appearance: ® Odor:
Field Measurement
Temperature
Electrical Conductivity
Specific Conductance
pH
Dissolved Oxygen
Eh
Turbidity
Value
°C
nMhos/cm
nMhos/cm
Standard Units
mg/l
mV
NTU
Military Time
Comments*
= meter reading x magnitude x k
EC corrected to 25 °C
Sample Collection
Sampling Device (type of pump/bailer)*
Permanently Installed Pump? YorN
Dedicated Equipment?
Sample Medium (well water, LNAPL, etc.)*
YorN Used Same Equip, for Purge? _
Pump Intake/Bailer Set at (ft below MP) _
Date/Time Sampling Began
Depth to Water (ft below MP)
QC Samples Collected? YorN
( see reverse*)
All Field Protocols were followed with no exceptions (Y,N)
Remarks (1)* (include protocol exceptions)
Form Completed by
Interval Samples Represent (ft below GS) Top =
Date /Time Sampling Finished _
Depth to Water (ft below MP) _
Sample Withdrawal Rate _
Enter Protocol Codes*
^ .
2.
Date.
•n GWS #2
sed 9-2-93
E-7
-------
Side 2 of 2
GROUND WATER SAMPLING INFORMATION FORM*
(Reverse Side)
ABBREVIATIONS
ft. feet MP Measuring Point GS Ground Surface
DTW Depth to Water WC Water Column cu.ft. cubic feet
Y Yes (circle if appropriate) N No (circle if appropriate) gals gallons
PID Photo lonization detector FID Flame lonization detector ppm parts per million
gpm gallons per minute Amt. amount k cell constant
EC Electrical Conductivity LNAPL light non-aqueous phase liquid (floater) DNAPL dense non-aqueous phase liquid
(sinker)
GENERAL INFORMATION
The "Field Sample (Event) ID#" should be constructed from the date and time that the first sample container of a purposefully associated set of sample
containers is filled. This set of samples would normally be collected vary closely together in time and include containers for a number of analytical
parameters and QC samples. QC samples are normally assigned temporary aliases (see below). For example, if the first of a set of containers is filled at
1:30 PM on December 19, 1992, the Field Sample Event ID#forall containers in the set should be 9212191330.
WELL INFORMATION
The water column length (L) is calculated by subtracting the depth to water (DTW) from the well depth. L = well depth - DTW. However, both of these
distances must be referenced to the same datum: either from the measuring point (MP) or from ground surface (GS). This form was designed with the
assumption that both the well depth and static water level values are referenced to the MP.
For convenience, a blank was included to also enter depth to water below GS in case the well depth referenced to the MP is unknown or cannot be
measured directly. In addition, this value will indicate where the static water level is relative to the open (screened) interval which is referenced to GS. For
the calculation of L in this case, the "stick up", the distance from the MP to GS, needs to be looked up or measured in the field. If the MP is above GS, then
the stick up is a positive number for this calculation. Enter the stick up distance here ft. (to the nearest 0.1 ft.). DTW (from GS) = DTW (from MP) - stick
up; L = well depth (from GS) - DTW (from GS).
One water column volume = OT2L. The units conversion from cubic feet to gallons is as follows: OT2 [ft.2] L [ft.[ [7.48 gallons/ft3]. r = well radius in feet (since
well specifications are normally given as diameter in inches, the diameter must be converted from inches to feet and then divided by one-half to yield r, in
feet). Examples of well diameter/gallons perft. ofWC: 170.041 gals; 270.163; 470.653; 671.47; 872.61.
PURGING
Measure the concentration of organic vapors inside the well immediately after removing the wellhead cap. On the front side of this form, circle whether a
PID or a FID was used, then enter wellhead and ambient background readings. Here specify the calibration gas , lamp voltage make & model
# of the instrument here .
If free product was detected, describe appearance, thickness, etc. (free product samples collected {Y, N }):
Supplemental description of purging equipment:
FIELD WATER QUALITY MEASUREMENTS AND OBSERVATIONS
If a flow cell was not used, describe how measurements were taken (note whether or not measurements were taken down hole):
Other Comments and Observations
SAMPLE COLLECTION
Sampling equipment details (Mfgr., Model*, tubing, etc.):
Quality Control Samples
Fictional sampling point name(s) end field sample event ID#(s) (aliases) can be used for QC samples on sample labels and chain of custody sheets to
distinguish them from primary samples without tipping off laboratories. List aliases here to document their association with primary sample identifiers on
front side of sheet. Name(s)/lD#(s) Indicate total # of QC samples collected: Replicates Splits Trip blanks Field ambient air blanks
Field methods blanks
Protocol codes: 1. Indicate the type of sampling protocol followed by selecting from codes (A-F) below and entering it on the front of this form. Specify the
name of the agency and the name of the agency program that approved the protocol. If none, write "none.") A slightly modified agency
program standard sampling protocol, approved as a non-site-specific protocol B) An unmodified or slightly modified agency program standard sampling
protocol, approved as a non site-specific protocol C) A non site-specific protocol approved by an agency D) A detailed but non agency-approved, site-
specific sampling protocol with adequate QA/QC procedures was followed; E) A detailed but non agency-approved sampling protocol without adequate QA/
QC procedures was followed; B) None of the above protocol conditions were known to be met (comment):
Protocol code: 2
A) Sampling observed by (agency) to meet all field protocols except as noted below: (agency signature)
B) Sampling observed by "neutral" observer (signature) all field protocols except as noted below;
C) Neither A or B applies (comment):
PROTOCOL EXCEPTIONS
List/discuss protocol exceptions for sampling-related field work (attach additional sheet if necessary):
Other Remarks(2)
E-8
------- |