Handbook: Ground Water, Volume 2: Methodology


          United States      Office of
          Environmental Protection   Research and Development
          Agency         Washington, DC 20460
EP A/625/6-90/016tl
July 1991
SEPA    Handbook
          Ground Wat^r
          Volume II: Methodology

-------

-------
         Handbook
       Ground Water
    Volume Me  Methodology
                                    EP A/625/6-90/016b
                                          July 1991
   U.S. Environmental Protection Agency
   Office of Research and Development
Center for Environmental Research Information
         Cincinnati, OH 45268
                                 Printed on Recycled Paper

-------
                                           NOTICE
This document has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for publication. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.

This document is not intended to be a guidance or support document for a specific regulatory program.
Guidance documents are available from EPA and must be consulted to address specific regulatory issues.

-------
                                          Contents
Chapter 1.  Monitoring Well Design and Construction
Chapter 2.  Ground-Water Sampling	
Chapter 3.  Transport and Fate of Contaminants in
Chapter 4.  Ground-Water Tracers	
Chapter 5.  Introduction to Aquifer Test Analysis.
Chapter 6.  Models and Computers in Ground-Water Investigations	  117
                                    Page
     	    1
	 .'..'.'."..'.'.'.'.'..'.'"  22
the Subsurface	   41
	   67
                                      96
                                            Jii

-------
                                        Acknowledgment
Many individuals contributed to the preparation and review of this handbook. The document was prepared by
Eastern Research Group, Inc.,  for EPA's Center for Environmental Research Information,  Cincinnati, OH.
Contract administration was provided by the Center for Environmental Research Information.

Volume I, Ground Water and Contamination, was published in September 1990 (EPA/625/6-90/016a). Volume
II, Methodology describes various investigative approaches and techniques. Although extensively revised, part
of Volume II was obtained from previous publications, "Handbook: Ground Water"  (EPA/625/6-87/016) and
"Protection of Public Water Supplies from Ground-Water Contamination" (EPA/625/4-85/016).

Authors and Reviewers

Michael J. Barcelona - Western Michigan University, Kalamazoo, Ml
Russell Boulding - Eastern Research Group, Inc., Arlington, MA
Ralph C. Heath - Private Consultant, Raleigh, NC
Wayne A. Pettyjohn - Oklahoma State University, Stillwater, OK
Ron Sims - Utah State University, Logan, UT
Judy Sims - Utah State University, Logan, UT
Paul van der Heijde - IGWMC, Holcomb Research Institute
H. Allen Wehrman - Illinois State Water Survey, Champaign, IL

Project Officer

Carol Grove - EPA-CERI, Cincinnati, OH
                                             iv

-------
Preface
The subsurface environment of ground water is ch aracterized by a complex interplay of physical, geochemical
and biological forces that govern the release, transport and fate of a variety of chemical substances. There are
literally as many varied hydrogeologic settings as there are types and numbers of contaminant sources. In
situations where ground-water investigations are most necessary, there are frequently many variables of land
and ground-water use and contaminant source characteristics which cannot be fully characterized.
The impact of natural ground-water recharge and discharge processes on distributions of chemical constituents
is understood for only a few types of chemical species. Also, these processes may be modified by both natural
phenomena and man's activities so as to further complicate apparent spatial or temporal trends in water quality.
Since so many climatic, demographic and hydrogeologic factors may vary from place to place, or even small areas
within specific sites, there can be no single "standardj" approach for assessing and protecting-the quality of ground
water that will be applicable in all cases.

Despite these uncertainties, investigations are under! way and they are used as a basis for making decisions about
the need for, and usefulness of, alternative corrective and preventive actions. Decision makers, therefore, need
some assurance that elements of uncertainty are minimized and that hydrogeologic investigations provide reliable
results.

A purpose of this document is to discuss measures th at can be taken to ensure that uncertainties do not undermine
our ability to make reliable predictions about the response of contamination to various corrective or preventive
measures.

EPA conducts considerable research in ground water to support its regulatory needs. In recent years, scientific
knowledge about ground-water systems has been increasing rapidly. Researchers in the Office of Research and
Development have made improvements in technology for assessing the subsurface, in adapting techniques from
other disciplines to successfully identify specific contaminants in ground water, in assessing the behaviorof certain
chemicals in some geologic materials and in advancing the state-of-the-art of remedial technologies.

An important part of EPA's ground-water research program is to transmit research information to decision makers,
field managers and the scientific community. This! publication has been developed to assist that effort and,
additionally, to help satisfy an immediate Agency need to promote the transfer of technology that is applicable to
ground-water contamination control and prevention]

The need exists for a resource document that brings together available technical information in a form convenient
and
for ground-water personnel within EPA and state
proper ground-water management. The information contained in this
is applicable to many programs that deal with the ground
or support document for a specific regulatory program
local governments on whom EPA ultimately depends for
handbook is intended to meet that need. It
-water resource. However, it is not intended as a guidance
GUIDANCE DOCUMENTS ARE AVAILABLE Ffi6M EPA AND MUST BE CONSULTED TO ADDRESS
SPECIFIC REGULATORY ISSUES.

-------

-------
                                             Chapter 1
                     MONITORING WELL DESIGN AND CONSTRUCTION
The principal objective of constructing monitoring /veils
is  to  provide access to  an otherwise inaccessible
environment.  Monitoring wells  are used to evaluate
topics within  various disciplines, including geology,
hydrology,  chemistry, and biology. In ground-water
quality monitoring, wells are used for collecting grcjund-
water samples, which upon  analysis may allow
description of a contaminant plume, or the movement of
a particular chemical (orbiological) constituent, or ensure
that potential contaminants are not moving  past a
particular point.                               '
Ground-Water Monitoring Program Goals
                                           nbie
Each purpose for ground-water monitoring-ambient
monitoring,  source  monitoring, case preparation
monitoring,  and  research  monitoring-must satisfy
somewhat different requirements, and may necessitate
different strategies for well design  and construction
(Barcelona and others, 1984). At the outset.the goals
of the intended monitoring program must be clearly
understood and thought should be given to the potential
future use of the wells in other, possibly different,
monitoring programs.

Regional investigations of ground-water quality involve
ambient monitoring.  Such investigations seek to
establish an overall picture of the quality of water vlithin
all or parts of an aquifer. Generally, sample collection is
conducted routinely over a period of many years to
determine changes in quality overtime. Often, changes
in quality are related to long-term changes in land use
(e.g., the effects of urbanization). Monitoring conducted
for Safe Drinking Water Act compliance generally! falls
in this category.

Samples commonly  are collected from a  variety of
public and private water supply wells for ambient quality
investigations. Because of the diversity of sources, the
data obtained  through some  ambient monitoring
programs  may  not  meet the strict well design and
construction requirements imposed by the three other
types of monitoring. However,  such programs are
important for detecting significant changes in aquifer
water quality overtime and space and protecting public
health.

Regulatory monitoring at potential contaminant sources
is considered source monitoring. Under this  type of
program, monitoring wells are located and designed to
detect the movement of specific pollutants outside of
the boundaries of a particular facility (e.g., treatment,
storage, or disposal). Ground-water sampling to define
contaminant plume extent and geometry would fall into
this monitoring classification.  Monitoring well design
and construction are  tailored to the site geology and
contaminant chemistry.   With  source monitoring,
quantitative aspects of analytical results become most
important because the  level  of contaminant
concentration may require specific regulatory action.

Monitoring  for case preparation, such as  legal
proceedings in environmental enforcement, requires a
level of detail  similar to source monitoring.  Source
monitoring,  in  fact, often  becomes a part  of  legal
proceedings to establish whether or not environmental
damage ha,s occurred and to identify the responsible
party. This is a prime example of one type of monitoring
program evolving into another. The appropriateness
and integrity of monitoring well design and construction
methods will come  under close scrutiny  in  legal
proceedings. In such cases, the course of action taken
during the monitoring investigation, the decisions made
concerning well design and construction, and the reasons
for those decisions must be clearly established and
documented.

Monitoring for research generally requires a  level of
sophistication beyond that required of any other type of
monitoring (this, of course, depends upon the types and
concentrations of constituents being sought and the
overall objectives of the research). Detailed information
is often needed  to support the basic concepts and
expand understanding of the complex mechanisms of

-------
ground-water movement and solute/contaminant
transport.

The goals of any proposed ground-water monitoring
program should be clearly stated and understood before
making any decisions on the types  and numbers of
wells  needed, their locations, depth, constituents of
interest,  and  methods  of collection,  storage,
transportation, and analysis.

As each of these decisions is made, consideration must
be  given to the  costs involved in each step of the
monitoring program and how compromises in one step
may affect the integrity and outcome of the other steps.
Forexample, costsavings in well construction materials
may so severely limit the  usefulness of a well that
another well may need to be constructed at the same
location for the reliable addition of a single chemical
parameter.

Monitoring Well Design Components

Monitoring well design and construction methods follow
production well design and construction techniques; a
monitoring well .however, is  built specifically to give
access to the ground water so that a "representative"
sample of water can be withdrawn and analyzed. While
well efficiency and yield is important, the ability to
produce large amounts of water for supply purposes is
not the primary objective.

Emphasis is placed instead on constructing a well that
will provide  easily obtainable ground-water samples
thatwill give reliable, meaningful information. Therefore,
materials  and techniques used for constructing  a
monitoring well must not materially alter the quality of
the water being sampled. An understanding of the
chemistry  of suspected  pollutants and  the geologic
setting in which the monitoring well is to be constructed
play a major role  in the drilling technique and well
construction materials used.

Several components need to be considered in monitoring
well design:  location and number of wells, diameter,
casing and screen material, screen length and depth of
placement, sealing material, well development, and
well security. Often, discussion of one component will
impinge upon other components.

Location and Number
Locating monitoring wells  spatially  and  vertically to
ensure that the ground-water flow regime of concern is
being monitored is obviously one of the most important
components in ground-water quality monitoring design.
Monitoring well locations (sites) andthe numberofwells
in the monitoring program are closely linked. The number
of wells and their location are principally determined by
the purpose of  the monitoring program.  In most
monitoring situations, the goal is to determine the effect
that some surface or near-surface activity has had on
nearbyground-water quality. Mostdissolved constituents
will descend vertically through the unsaturated zone
beneath the area of activity and then, upon reaching the
saturated zone, move horizontally in the direction of
ground-water flow. Therefore, monitoring wells  are
normally completed downgradient in the first permeable
water-bearing unit encountered. Consideration should
be given to natural (seasonal) fluctuations, which can
amount to several feet throughout the year and from
one year to the next, and artificial fluctuations brought
about largely by pumping, which can amount to several
tens of feet in only a few hours.  Artificial fluctuations
also are caused by lagoon operation, which can cause
a rise or "mound" in the water table.

Preliminary boreholes and monitoring wells can be
constructed to collect  and analyze geologic material
samples, to measure ground-water levels, and to collect
water-quality samples, all  of which provide a guide to
thefuture placement of additional wells. Accurate water-
level information must be obtained to determine if local
ground-waterf low paths and gradients differ significantly
from the regional appraisal.

The  analysis  of  water-quality samples from  the
preliminary wells can direct the placement of additional
wells. Such data are helpful in the vertical arrangement
of sampling points, especially for a contaminant that is
denser than water. Without some preliminary chemical
data, it is usually very difficult to determine the location
of the most contaminated zone.

Site geology, site hydrology,  source characteristics,
contaminant characteristics, and the size of  the area
under investigation all help determine where  and how
many wells should be constructed. Certainly, the more
complicated the geology and hydrology, the more
complex the contaminant and source, and the largerthe
area being investigated, the greater the number of
monitoring wells that will be required.

Diameter
In the past, the diameter of a monitoring well was based
primarily on the size of the device (bailer, pump, etc.)
used to withdraw the water samples. This practice was
similarto that followed for water supply well design. For
example, a domestic water well is commonly 4 to 6
inches in  diameter,  which is of sufficient size to
accommodate a submersible pump capable of delivering
from 5 to 20 gallons per minute.  Municipal, industrial,

-------
and irrigation wells have greater diameters to handle
larger pumps, and to increase the available screen
open area so the well can produce water efficiently.

This practice worked well in very permeable formations,
where an aquifer capable of furnishing large volumes of
water was present. However, unlike most water-su Dply
wells, monitoring wells are quite often completed in very  2l
marginal water-producing zones. Pumping one or nWe
well volumes of water (the amount of water stored in the
well casing under nonpumping conditions) from a |well
built in low-yielding materials (Gibb and others, 1981)
may present a serious problem if the well has a large
diameter.

Figure 1-1 illustrates the amount of water in storage, per
foot of casing for different well casing diameters. Well
casings with diameters of 2 and 6 inches will contain
0.16 and 1.47 gallons of water  per foot of casing,
respectively.  Purging four well volumes  from a well
containing 10 feet of water would require removal of 6.4
gallons of water from a 2-inch well and 58.8 gallons of
waterfrom a 6-inch well. Under low-yielding conditions,
it can take considerable time to recover enough v\ ater
from the well to collect a sample (see Figure  1-2)
                     3456

                    Wall Diameter (Inches)
Figure 1-1. Volume of Water Stored Per Foot of
Well Casing for Different Diameter Casings (from
Rmaldo-Lee, 1983)
                                                  3!   30
                                                      20
                                                      10
                                                                     Wall Diameter (Inches)

                                                       Assumptions: K = 1 x 10"' cm/sec, well screen = 10', 10' of water
                                                                above screen, 6' of water Instantaneously
                                                                removed

                                                   Figure 1-2. Time Required for Recovery After
                                                   Slug of Water Removed (from Rinaldo-Lee, 1983)
                                                   In addition, when hazardous constituents are present in
                                                   the ground water, the purged water must be properly
                                                   disposed.  Therefore, the quantity of water pumped
                                                   from the well should be minimized for reasons of safety,
                                                   as well as disposal cost. Cost of well construction also
                                                   is a consideration. Wells less than 4 inches in diameter
                                                   are much less expensive  than large diameter wells in
                                                   terms of both cost of materials and cost of drilling.

                                                   For these reasons and with the advent of a variety of
                                                   commercially available small-diameterpumps (less than
                                                   2 inches OD) capable of lifting water over 100 feet, 2-
                                                   inch ID wells have become the  standard in monitoring
                                                   well technology.

                                                   Large diameter wells can be useful in situations where
                                                   monitoring may be followed by remedial actions involving
                                                   reclamation and treatment of the contaminated ground
                                                   water. In some instances, the  "monitoring" well may
                                                   become a "supply" well to  remove contaminated water
                                                   for treatment. Larger diameter wells also merit
                                                   consideration when monitoring is required at depths of
                                                   hundreds of feet and in situations where the additional
                                                   strength of large diametercasing is needed. For sampling
                                                   at several depths beneath one location,  several
                                                   monitoring wells have been nested in a single borehole
                                                   (Johnson, 1983). This type of technique will require
                                                   drilling a larger diameter hole to  accommodate the
                                                   multiple well casings. Again, the use of smallerdiameter

-------
casing provides advantages by allowing more wells to
be nested in the borehole, thus easing construction and
reducing drilling expenses.

Casing and Screen Material
The type of material used for a monitoring well can have
a distinct effect on the quality of the water sample to be
collected (Barcelona and others, 1985; Gillham and
others, 1983; and Miller, 1982). The materials of choice
should retain their structural integrity for the duration of
the monitoring  program  under actual  subsurface
conditions. They should neither adsorb nor leach
chemical  constituents  that  would  bias  the
representativeness of the samples collected.

Galvanized steel casing  can impart iron, manganese,
zinc, and cadmium to many waters, and steel casing
may contribute iron and manganese to a sample. PVC
pipe has been shown to  release and adsorb  trace
amounts of various organic constituents to water after
prolonged exposure (Miller, 1982). PVC solvent cements
used to attach sections of PVC pipe also have  been
shown to release significant  quantities  of organic
compounds.

TeflonR and glass are among the most inert materials
considered for  monitoring well construction. Glass,
however, is difficult and expensive to use under most
field conditions. TeflonR also is very expensive; with
technological advances,  TeflonR-coated casings and
screens may become available. Stainless steel also
offers desirable properties for monitoring, but it too is
expensive.

A reasoned strategy for ground-water monitoring must
consider the effects of  contaminated  water on well
construction materials as well. Unfortunately, there is
limited published information on  the performance of
specific materials  in varied hydrogeologic settings
(Pettyjohn  and  others,  1981). The  following is a
preliminary ranking of commonly used materials exposed
to different solutions representing the principal soluble
species present in hazardous waste site investigations
(Barcelona and others, 1984). They are listed in  order
of best to worst in terms  of chemical resistance:

       TefIonR
       Stainless Steel 316
       Stainless Steel 304
       PVC Type 1
       Lo-Carbon Steel
       Galvanized Steel
       Carbon Steel

Polyvinyl chloride (PVC Type I) is very chemically
resistant except to low molecular weight ketones,
aldehydes, and chlorinated solvents. Generally, as the
organic content of a solution increases, direct attack on
the polymer matrix or solvent absorption, adsorption, or
leaching may occur. This reaction, however, has not
been  observed with  TeflonR.  Provided that sound
construction practices are followed, TeflonR can be
expected to outperform all other casing and sampling
materials (Barcelona and others, 1984).

Stainless steels are the most chemically resistant of the
ferrous materials. Stainless steel, however, may be
sensitive to the chloride ion, which can  cause pitting
corrosion,  especially over long-term exposures under
acidic conditions. Given the similarity in price, workability,
and performance, the remaining ferrous  materials (lo-
carbon, galvanized steel, and  carbon)  provide little
advantage over one  another  for casing/screen
construction.

Significant levels of organic components found in PVC
primers and  adhesives  (such as tetrahydrofuran,
methylethylketone,     cyclohexanone,     and
methylisobutylketone) were detected in well water
several months after well installation (Sosebee and
others, 1982). The presence of compounds such as
these can mask the presence of other similar volatile
compounds (Miller, 1982). Therefore, when using PVC
and othersimilar materials, such as ABS, polypropylene,
or polyethylene, for well construction, threaded joints
are the preferred means forconnectingsectionstogether.

In many situations, it may be possible to compromise
accuracy or precision for initial cost, depending on the
objectives of the  monitoring program. For example, if
the contaminants of interest are already defined and
they do not include substances that might bleed or sorb,
it may be  reasonable to use wells cased with a less
expensive material.

Wells constructed of less than optimum materials might
be used for sampling  if identically fabricated wells are
constructed in uncontaminated parts of the monitored
aquifer to  provide ground-water samples for use as
"blanks" (Pettyjohn and others, 1981). Such blanks,
however, may not adequately  address problems of
adsorption on, or leaching from,  the casing material
induced by contaminants in the ground water. It may be
feasible to use two or more kinds of casing materials in
the saturated zone and above the seasonal high water
table, such as TeflonR or stainless steel, and use  a
rnore appropriate material, such as PVC or galvanized
steel casing, above static water level.

Trying to save money by compromising on material
quality or suitability, however, may eventually increase
program cost by creating the need for reanalysis, or

-------
worse, monitoring well reconstruction. Each case
requires  careful  consideration  and the  analytical
laboratory should be fully aware of the construction
materials used.

Care also must be taken in preparing the casing and
well screen materials priorto installation. At a minimum,
materials should be washed with detergent and rinsed
thoroughly with clean water. Steam-cleaning and tiigh
pressure, hot water cleaners provide excellent cleaning
of cutting oils and lubricants left on casings and screens
after their manufacture (particularly for metal casing
and screen materials). To ensure that these and other
sampling materials are protected from contamination
prior to placement down-hole,  materials should be
covered (with plastic sheeting oTotheFmaterial), and
kept off the ground.                          T

All wells should allow free entry of water. The water
produced  should be as clear and silt-free as possible.
For drinking water supplies, sediment in the raw water
can create additional pumping and treatment costs and
lead to the  general unpalatability of the water. With
monitoring wells, sediment-laden water can greatly
lengthen filtering time and create chemical interfere ice
in sample analyses.

Commercially manufactured well screens are preferred
for monitoring wells so long as the screen slot siz
appropriate. Sawed or torch-cut  casing may
                                             3 is
                                              be
satisfactory in deposits where medium to coarse shnd
or gravel predominate. In formations where fine se
                                              nd,
silt, and clay predominate, sawed or torch-cut slots will
be too large to retain the aquifer materials, and the well
may clog or fill with sediment. The practice of sawing
slots  in PVC pipe should be avoided  in monitoring
situations where organic chemicals are of concern,
because this procedure exposes fresh surfaces of PVC,
increasing the possibility of releasing  compound
ingredients or reaction products.

It may be helpful to have several slot-sized well screens
on site  so that the proper manufactured screen and
slots can be placed in the hole afterthe aquifer materials
have been inspected. Gravel pack of a size compatible
with the selected screen slot size will further help rejtain
the finer fractions of material and allow free entry of
water into the well by creating a zone of higher
permeability around the well screen.

For  natural-packed  wells,  where  relatively
homogeneous, coarse materials predominate,  a slot
size should be selected based on the effective size [and
uniformity coefficient of the formation materials.  The
effective size is equivalent to the sieve size that will
retain 90 percent (or passes 10 percent) of the formation
material; the uniformity coefficient is the ratio of the
sieve size that will retain 40 percent (orpass 60 percent)
of the formation material to the effective size (Aller and
others  1989).   If an artificial pack is used, a uniform
gravel-pack size that is from three to five times the 50
percent retained size of the formation and a screen size
that will retain at least 90 percent of the pack material
should be selected (Walker, 1974). The gravel-pack
should be composed of clean, uniform quartz sand.

The  gravel-pack should be placed carefully to avoid
bridging in the hole and to allow uniform settling around
the screen.  A tremie pipe can be used to guide the sand
to the bottom of the hole and around the screen. The
pipe should be lifted slowly as the annulus between the
screen and borehole as the borehole fills. If the depth of
water standing in the annulus is not great, the sand can
be simply poured from the surface. The volume of sand
required to fill the annulus to the desired depth (usually
about 1 foot above the top of the screen) should be
calculated.  Field measurements  should be taken  to
confirm that the pack has reached this level before
backfilling or sealing procedures start.

Screen Length and Depth of Placement
The  length of screen and the depth at which it is placed
depend,  to a  large degree, on the behavior of the
contaminant as it moves through the unsaturated and
saturated zones,  and  on the goal of the monitoring
program. When monitoring an aquifer used as a water
supply, the entire thickness  of  the  water-bearing
formation could be screened (just as a production well
might be). In regional aquifer studies, production wells
commonly are used for sampling. Such samples would
provide water integrated over the entire thickness of the
water-bearing zone(s), and would be similar in quality to
what would be found in a drinking-water supply.

When specific depth intervals must be sampled at one
location, vertical  nesting of wells is common.  This
technique is often necessary when the saturated zone
is too thick to adequately monitor with one long-screened
section (which would dilute the collected sample). Since
contaminants tend to stratify within the saturated zone,
collecting a sample integrated over a thick zone will
provide  little or no information on the depth and
concentration that a contaminant may have reached.
Furthermore, nested wells provide information on the
water level or potential that exists at each well screen.
These data are essential to an understanding of the
vertical component of flow.

Screen lengths of 1 to 2 feet are common in detailed
plume  geometry investigations. Thick  aquifers would
require that several wells be completed at different
depth intervals. In such situations (and depending on

-------
the magnitude of the aquifersaturated thickness), screen
lengths of no more than 5 to 10 feet should be used.
Monitoring wells can be constructed in separate holes
placed closely together or in one larger diameter hole,
as  shown in Figure  1-3. Vertical movement of
contaminants in  the well bore before and after well
completion may be difficult to prevent since it is difficult
to seal several wells in one hole.Thus, multiple holes
may needto be drilled to ensure well integrity. Specially
constructed installations have been developed to sample
a large number of points vertically over short intervals
(Morrison and Brewer,  1981;  Pickens, 1981; and
Torstensson, 1984;  Figures 1-4 and 1-5).
              sna
                       -Bukm ITyp.1
                    -F«I«fS»ndlTvp,l
                       Screened kilotval
Figure 1-3. Typical Multlwell Installations (from
Johnson,1983)


In other situations, only the first water-bearing zone
encountered will require monitoring (for example, when
monitoring near a potential contaminant source  in a
relatively impermeable glacial till). The "aquifer" or zone
of interest in such an instance may be only several
inches to  a few feet thick. Screen length should be
limited to 1 to 2 feet in these cases to minimize siltation
problems from surrounding fine-grained materials and
possible dilution  effects from water contributed by
uncontaminated zones.

Because of the chemical reactions that occur when
ground water contacts  the  atmosphere,  particularly
when dealing with volatile compounds, the screened
section should  not be aerated. Generally, well depth
should assure that the screened section is always fully
submerged. The design should consider fluctuations in
the elevation of the top of the saturated zone caused by
seasonal variations or human-induced changes.
FMd IniulUlkw
    ,— End Op
                                                                                 Crott Sectkm of
                                                                                 S«m(*ng Point
                                                    Gfouod Surfaca
                                                              -/
                                                                 -PVCPip«
                                                                 - Coupling
                                                               >
                                                                                              *-Se«tn
                                                                 -EndCip
                                                   Figure 1-4. Schematic Diagram of a Multilevel
                                                   Sampling Device (from Pickens and others, 1981)
                                                     BontoniM SurfKI  f~"
                                                                '  '"'* Ci«nd-W«W»L«vel '•_*>•;
                                                        A. Snub As Lilt S*H)!*t with PVC Citing
                                                   Figure 1-5. Single (a) and Multiple (b) Installation
                                                   Configurations for an Air-Lift Sampler (from
                                                   Morrison and Brewer, 1981)
                                                   Monitoring for contaminants  with densities different
                                                   than waterdemands special attention. In particular, low
                                                   density organic compounds, such as gasoline, will float
                                                   ontheground-watersurface(Gillham and others, 1983).
                                                   Monitoring wells constructed to detect floating
                                                   contaminants should contain screens that extend above
                                                   the zone of saturation so that these lighter substances
                                                   can enter the well. The screen length and position must

-------
accommodate the magnitude and depth of variations in
water-table elevation. However, the thickness of floating
products in the well does not necessarily indicate the
thickness of the product in the aquifer.

Sealing Materials and Procedures
It is critical that the screened part of each  monitoring
well access ground water from a specific depth interval.
Vertical movement of ground water in the vicinity of the
well can greatly  influence sample quality  (Keith -and
others, 1982). Rainwater can infiltrate backfill, potentially
diluting or contaminating samples; vertical seepagje of
leachate along  the well casing will also produce
unrepresentative samples (particularly  important  in
multilevel installations such as in Figures 1-3,1-4, land
1 -5). Even more importantly, the creation of a conduit in
the annulus of the monitoring well that could contribute
to or hasten the spread of contamination is to be strictly
avoided. Several  methods have  been  employed
successfully to isolate contaminated zones duringlthe
drilling process (Burkland and Raber, 1983; Perry and
Hart, 1985).

Monitoring wells  are usually sealed with neat cerr ent
grout, dry bentonite  (powdered,  granulated,  or
pelletized), or bentonite slurry. Well seals usually are
installed at two places within the annulus:  one just
above the screened interval and the other at the ground
surface  to inhibit downward leakage of  surface
contaminants.

Bentonite traditionally has been considered to provide
a much  better seal than cement.  However,  recent
investigations on the use of clay liners for hazardous
waste disposal have shown that some  organic
compounds migrate through bentonite with little or no
attenuation (Brown, and others, 1983).  Therefore,
cement may offer some benefits over bentonite.

Bentonite most often is used as a down-hole sea  to
prevent vertical migration within the well annulus. When
bentonite must be placed below the water table  (or
where water has risen in the borehole), it is recommended
that a bentonite slurry be tremied down the annulus to
fill the hole from the  bottom upward. In  collapsible
materiai conditions, where-the"borehole has collapsed
to a point just above  the water table, dry bentonite
(granulated or pelletized works best) can be pou
down the hole.
red
Bentonite clay has appreciable ion-exchange capacity,
which may interfere with the chemistry of collected
samples when the seal is adjacent to the screen or v/ell
intake. When improperly placed,  cement grout has
been known to seriously affect the pH of sampled wa er.
Therefore, special attention and care should be exercised
during placement of a down-hole seal. Approximately 1
foot (at a minimum) of gravel-pack or naturally collapsed
material should extend above the top of the well screen
to ensure that the sealing materials do  not migrate
downward into the well screen. If the sealing material is
too watery before being placed down the hole, sealing
matejiajsjnay settle or migrate into the gravel-pack or
screened area, and the fine materials in the seal may
penetrate the natural or artificial pack.

While a neat cement (sand and cement, no gravel) grout
is often recommended,  especially for surface sealing,
shrinkage and cracking of the cement upon curing and
weathering can create an improper seal. Shrink-resistant
cement  (such as Type K Expansive  Cement) and
mixtures of small amounts of bentonite with neat cement
have been used successfully to help prevent cracking.

Development
Development is a facet of monitoring well installation
that often is overlooked. During the drilling process,
fine-grained materials smearon the sidesoftheborehole,
forming a  mud  "cake" that reduces  the  hydraulic
conductivity of the materials opposite the screened part
of the well. To facilitate entry of water into the monitoring
well  (a  particularly important factor for  low-yielding
geologic materials), this mud cake must be broken
down and the fine-grained materials removed from the
well or well bore. Development also removes  fluids,
primarily water, which  are introduced to the water-
bearing formations during the drilling process.

Additionally, monitoring wells must be developed to
provide water free of suspended solids for sampling.
When sampling for metal  ions and other  inorganic
constituents, water samples must  be filtered and
preserved at the well site at the time of sample collection.
Improperly developed monitoring wells will  produce
samples containing suspended sediments that will both
bias the chemical analysis of the collected samples and
frequently cause clogging of  the field  filtering
mechanisms.

The time and money spent for this important procedure
will expedite  sample filtration and result in  samples
more representative of water contained in the formation
being monitored. The time saved in field filtration alone
will more than offset the cost of development.

Successful development  methods  include bailing,
surging, and flushing with airorwater. The basic principle
behind each method is to create reversals of flow in and
out of the well (and/or borehole), which tend to break
down the mud cake and draw the finer materials into the
hole for removal. This process also aids in removing the
finer fraction of materials in proximity to the borehole,

-------
leaving behind  a "natural" pack  of coarser-grained
materials.

Years ago, small-diameter well  development most
commonly was achieved through use of a bailer. The
bailer was about the only "instrument" that had been
developed for use in such wells. Rapidly dropping and
retrieving the bailer in and out of the water caused a
back-and-forth action of water in the well, moving some
of the more loosely bound fine-grained materials into
the well where they could be removed.

Depending on the depth of water in the well, the length
of the well screen, and the volume of water the bailer
could displace, this method was not always very efficient.
"Surge blocks," which could fit inside 2-inch diameter
wells,  provided some improvement over bailing
techniques. Such devices are simply plungers that,
when moved  vigorously up and down, transfer that
energy to an in-and-out action on the water near the well
screen. Surge blocks have the potential to move larger
quantities of water with higher velocities, but they pose
some risk to the well casing and  screen if the surge
block fits too tightly orifthe up-and-down action becomes
too vigorous.  Improved  surge block design has been
the subject of some recent investigation (Schalla and
Landick, 1985).

In more productive aquifers, "overpumping" has been a
popular method for well development. With this method,
a pump is alternately turned on (usually at  a slightly
higher rate than the well can sustain) and off to simulate
a surging action in the well. A problem with this method
is that overpumping does not create as pronounced an
outward movement of water  as  does  surging.
Overpumping may tend to bridge  the fine and coarse
materials, limiting the movement of the fine  materials
into the well and thereby limiting the effectiveness of the
method.

Pumping with air also has been used effectively (Figure
1-6). Better development has been accomplished by
attaching differently shaped devices to the end of an
airline to force the air out into the formation. Figure 1-7
shows an example. Such a device causes a much more
vigorous action on the movement of material in proximity
to the  well screen while also pushing water to the
ground surface.

Air development techniques rhay expose field crews to
hazardous constituents when highly contaminated
groundwateris present. The technique also may cause
chemical reactions with species present in the ground
water,  especially volatile  organic compounds. Care
also must be  taken to filter the injected air to prevent
contamination of the well environmentwith oil and other
lubricants present in the compressor and airlines.
Comprised Air
Jl
••-••[
• o . - I
o .
o . •
0 . 0
o '
. o
o
o
• o '
d
. * o
ft ' '
Air-Watw ,
Mixture
• o .
0 0 • .
0 '. 0 . '
• ' Nozzla -^
• . * a
0,o'
• o . ""
' o .
d . .
o

—
=
1
.
:
;f
[
;
I
;
I



1







>•
'I
l-V.o.

—

|
^
~
^
1
';
r
^:
: ;
N
- , ^
i ~
- ;
: -
J .
. " o
-tf— Well Screen
o
" " 0 •
o •
O
Stnd Formation
'. * '
'. • •
.«•; .'. o
•i.;t^i :.
.' e' "a • 'o
• O '
• . o
• o ' o . .
-* " * l>
. ' O -
0' • 0 ' .

                  Air Development
Figure 1-6. Well Developments with Compressed
Air
                    Continuous Slot Well Scresn
                  -Jet Nozzle
Figure 1-7. The Effects of High-Velocity Jetting
Used for Well Development through Openings in
a Continuous-Slot Well Screen
                                               8

-------
Development procedures  for  monitoring  wells  in
relatively unproductive geologic materials are somewhat
limited.  Due to the low hydraulic conductivity of tljie
materials, surging of water in and out of the well casing
is extremely difficult. Also, when the well is pumped, the
entry rate of the water is  inadequate  to effectively
remove fines from the well  bore and the gravel-pack
material outside the well screen.

Where an open borehole can be sustained in this type
of geologic setting, clean water can be circulated down
the well casing, out through the screen, and back up the
borehole (Figure 1-8). Relatively high water velocities
can be maintained and the mud cake from the borehole
wall can be broken  down  effectively and removed.
Because of the low hydrau lie conductivity of the geologic
materials outside the well, only a small amount of water
will penetrate the formation being  monitored. Trjis
development procedure can be done before and after
placement of a gravel-pack but must be conducted
before a well has been sealed. Afterthe gravel-pack has
been installed, water should not be circulated too quickly
or the gravel-pack will be lifted out of the boreho e.
Immediately following development, the well should be
sealed,  backfilled, and pumped for a short period jto
stabilize the formation around the outside of the screen
and to ensure that the well will produce fairly cle|ar
water.
                     Water
Figure 1-8. Well Development by Back-Flushing
with Water
Security
For most monitoring well installations, some precautions
must be exercised to protect the surface portions of the
well from damage.  In many instances, inadvertent
vehicular accidents  do occur; also, monitoring  well
installations seem particularly vulnerable to grass
mowers. Vandalism is often a major concern, from
spontaneous "hunters" looking  for a likely target to
premeditated destruction of property associated with
an unpopular operation. Several simple solutions can
be employed to help minimize the damage due to
accidental collisions. However, outwitting the determined
vandal may be an impossible undertaking and certainly
an expensive one.

The basic problem in maintaining the physical condition
of any monitoring  well is anticipating the hazards that
might befall that particular installation. Some situations
may call for making the well highly visible whereas
others may require keeping the well inconspicuous.

Where the most  likely problem is one  of vehicular
contact, be it mowers, construction traffic, orothertypes
of two-, three-, orfour-wheeled traffic, the first thing that
can be done is to make the top of the well easy to see.
It should extend far enough above ground to be visible
above grass, weeds, or small shrubs. If that is not
practical, use a "flag" that extends above the  well
casing. A flag is also helpful for periods when leaves or
snow have buried  low-lying objects.

The well casing should also be painted a bright color
(orange and yellow are the most visible). This not only
makes the well more visible but also protects metal
casing material  from rusting. Care  should be taken to
prevent paint from getting  inside the well casing or in
threaded fittings that may contact sampling equipment.

The owners/operators of the site being monitored should
also know the location of each installation. They should
receive maps clearly and exactly indicating the position
of the wells, and their employees should be informed of
the importance of those installations, the cost associated
with them, and the difficulty involved in replacement.

The segment of the well that extends above the ground
also can  be reinforced,  particularly  if the well  is
constructed of  PVC or TeflonR. The  well could be
constructed such that only the portion of the well above
the water table is metal. In this manner, the integrity of
the sample is maintained as ground watercontacts only
inert material, and the physical condition of the well is
maintained as the upper metal portion is better able to
withstand impact.

There are two argu ments to consider when constructing

-------
a well in this manner. The arguments focus on the weak
point in the well construction: at or near the juncture of
the metal and nonmetal casings. One argument suggests
that a longer section of metal casing is superior because
its additional length in the ground provides more strength.
Thus, a break is less likely to occur (although the casing
is likely to be bent). The other argument suggests that
should a break in the casing occur, a shorter length of
metal casing is superior because  a break nearer to
ground surface is easier to repair. Each argument has
its merits; only experience with site conditions is likely
to produce the best solution.

The use of "well protectors" is another popular solution
that involves the use of a larger diameter steel casing
placed around the monitoring well at the ground surface
and extending several feet below ground (Figure 1-9).
The protectors are usually seated in the cement surface
seal to a depth below the frost line.

Commonly, well protectors are equipped with a locking
cap, which ensures against tampering with the inside of
the well. Dropping  objects down the well may clog the
well screen or prohibit the sampling device from reaching
water, and the quality of the ground  water may  be
altered, particularly where small quantities (perhaps
drops) of an organic liquid may be suff icientto completely
contaminate the well.
       Monitoring Well
                                _Well Protector with
                                Lockable Cap
Figure 1-9. Typical Well Protector Installation
Problems associated with vandalism run from simple
curiosity to outright wanton destruction. Obviously, sites
within secured, fenced  areas are less likely to be
vandalized. However, there is probably no sure way to
deter the determined vandal, short of posting a 24-hour
guard. In such situations, well protectors are a must.
The wells should be kept as inconspicuous as possible.
However, the benefits of "hiding" monitoring wells must
be weighed against the costs of delays  in finding them
for sampling and the potential costs  for  repairs or
maintenance on untried security designs.

In some situations, it might be a good policy to notify the
public of the need for the monitoring wells. Properly
asserting that  each  well serves an  environmental
monitoring purpose and that the  wells have been
constructed to ensure public well-being may create a
civic conscience that would help to minimize vandalism.

As with all the  previously mentioned monitoring well
components, no single solution will best meet every
monitoring situation. Knowledge of the social, political,
and economic conditions of the geographic area  and
circumstances surrounding the need for ground-water
monitoring will dictate, to a large degree, the type of well
protection needed.

Monitoring Well Drilling Methods

As might be expected, different drilling techniques  can
influence the quality of a ground-water sample. This
applies to the drilling method employed  (e.g., augered,
driven, or rotary), as well as the driller. There is no
substitute for a conscientious driller willing to take the
extra time and care  necessary to  complete a good
monitoring well installation.

Among the criteria used to select an appropriate drilling
method are the following factors,  listed in order of
importance:

1. Hydrologic information
   a.  type of formation
   b.  depth of drilling
   c.  depth of desired screen setting below the top of
        the  zone of saturation
2. Types of pollutants expected
3. Location of drilling site, i.e., accessibility
4. Design of monitoring well
5. Availability of drilling  equipment

Table 1 -1 summarizes several different drilling methods,
and their advantages and disadvantages when used for
monitoring  well  construction.  Several  excellent
publications are referenced  for detailed discussions
(Campbell and Lehr, 1973;  Fenn and others, 1977;
                                                 10

-------
         Method
                                     Drilling Principle
                                            Advantages
                                               Disadvantages
       Drive Point
     Auger, Hollow-
     and Solid-stem
         Jetting
       Cable-tool
      (Percussion)
1.25 to 2 inch ID casing with
pointed screen mechanically
driven to depth.
Successive 5-foot flights of spiral-
shaped drill stem are rotated into
the ground to create a hole.
Cuttings are brought to the surface
by the turning action of the auger.
Washing action of water forced out
of the bottom of the drill rod clears
hole to allow penetration. Cuttings
brought to surface by water flowing
up the outside of the drill rod.
Hole created by dropping a heavy
"string" of drill tools into well bore.
crushing materials at bottom.
Cuttings are removed occasionally
by bailer. Generally, casing is
driven just ahead of the the bottom
of the hole; a hole greater than 6
inches in diamerter is usually
made.
Inexpensive.

Easy to install, by hand if
necessary.

Water samples can be collected as
driving proceeds.

Depending on overburden, a good
seal between casing and formation
can be achieved.
Inexpensive.

Fairly simple operation.  Small rigs
can get to difficult-to-reach areas.
Quick set-up time.

Can quickly construct shallow wells
in firm, noncavey materials.

No drilling fluid required.

Use of hollow-stem augers greatly
facilitates collection of split-spoon
samples.

Small-diameter wells can be built
inside hollow-stem flights when
geologic materisl are cavey.
Inexpensive. Driller often not
needed for shallow holes.

In firm, noncavey deposits where
hole will stand open, weU
construction fairly simple.
Can be used in rock formations as
well as unconsolidated formations.

Fairly accurate logs can be
prepared from cuttings if collected
often enough.

Driving a casing ahead of hole
minimizes cross-contamination by
vertical leakage of formation
waters.

Core samples can be obtained
easily.
Difficult to sample from smaller diameter
drive points if water level is below suction
lift Bailing possible.

No formation samples can be collected.

Limited to fairly soft materials. Hard to
penetrate compact, gravelly materials.

Hard to develop. Screen may become
clogged if thick clays are penetrated.

PVC and Teflon casing and screen are not
strong enough to be driven. Must use
metal construction materials which may
influence some water quality determina-
tions.

Depth of penetration limited, especially in
cavedy materials. Maximum depths 150
feet.

Cannot be used in rock or well-cemented
formations. Difficult to drill in cobbles/
boulders.

Log of well is difficult to interpret without
collection of split spoons due to the lag
time for cuttrings to reach ground surface.

Vertical leakage of water through borehole
during driling is likely to occur.

Solid-stem limited to fine grained,
unconsolidated materials that win not
collapse when  unsupported.

With hollow-stem flights, heaving materials
can present a problem. May need to add
water down auger to control heaving or
wash materials from auger  before
completing well.

Somewhat slow, especially with increasing
depth.

Extremely difficult to use in  very coarse
materials, i.e., cobbles/boulders.

A water supply is needed that is under
enough pressure to penetrate the geologic
materials present

Difficult to interpret sequence  of geologic
materisl from cuttings.

Maximum depth 150 feet, depending on
geology and water pressure capabilities.

Requires an experienced driller.

Heavy steel drive pipe used to keep hole
open and drilling "tools" can limit
accessibility.

Cannot run some geophysical logs due to
presence of drive pipe.

Relatively slow drilling method.
Table 1-1. Advantages and Disadvantages of Selected Drilling Methods for Monitoring Well
Construction

                                                                  11

-------
         Method
                                      Drilling Principle
                                             Advantages
                                             Disadvantages
     Hydraulic Rotary
     Reverse Rotary
        Air Rotary
     Air-Percussion
  Rotary or Downhote-
        Hammer
Rotating bit breaks formation;
cuttings are brought to the
surface by a circulating fluid
(mud). Mud is forced down the bit.
and up the annulus between the drill
stem and hole waU. Cuttings are
removed by settling in a "mud pir at
the ground surface and the mud is
circulated back down the drill stem.
Similar to Hydraulic Rotary method
except the drilling fluid is circulated
down the borehole outside the drill
stem and is pumped up the Inside, just
the reverse of the normal rotary
method.  Water Is used as the drilling
fluid, rather than a mud, and the hole
is kept open by the hydrostatic
pressure of the water standing in the
borehole.
Very similar to Hydraulic Roatary, the
main difference being that air is used
as the primary drilling fluid as opposed
to mud or water.
Air Rotary with a reciprocating hammer
connected to the bit to fracture rock.
Drilling is fairly quick in all types of
geologic materials.

Borehole will stay open from
formation of a mud wall on sides of
borehole by the circulating drilling
mud. Eases geophysical logging
and well construction.

Geologic cores can be collected.

Virtually unlimited depths possible.
Creates a very "clean" hole, not
dinted with drilling mud.

Can be used in all geolotc
formations.

Very deep penetrations possible.

Split-spoon sampling possible.
Can be used in all geologic
formations; most successful in
highly fractured environments.

Useful at any depth.

Fairly quick.

Drilling mud or water not required.
Very fast penetrations.

Useful in all geoloogic
formations.

Only small amounts of water
needed for dust and bit
temperature control.

Cross contamination potential
can be reduced by driving
casing.
Expensive, requires experienced driller
and fair amount of peripheral equipment

Completed well may be difficult to
develop, especially small-diameter wells,
because of mud wall on borehole.

Geologic logging by visual inspection of
cuttings is fair due to persence of drilling
mud.  Thin beds of sand, gravel, or day
may be missed.

Presence of drilling mud can contaminate
water samples, especially the organic,
biodegradable muds.

Circulation of drilling fluid through a
contaminated zone can create a hazard at
the ground surface with the mud pit and
cross-contaminate clean zones during
circulation.

A large water supply is needed to
maintain hydrostatic pressure in deep
holes and when highly conductive
formations are encountered.

Expensive-experienced driller and much
peripheral equipment required.

Hole diameters are usually large,
commonly 18 inches or greater.

Cross-contamination from circulating
water likely.

Geologic samples brought to surface are
generally poor, circulating water will
"wash" finer materials from sample.

Relatively expensive.

Cross-contamination from vertical
communication possible.

Air will be mixed with water in the hole
and that which is blown from the hole,
potentially creating unwanted reactions
with contaminants; may affect "represen-
tative" samples.

Cuttings and water blown from the hole
can pose a hazard to crew and
surrounding environment if toxic
compounds encountered.

Organic foam additives to aid cuttings
removal may contaminate samples.

 Relatively expensive.

 As with most hydraulic rotary methods,
 the rig is fairly heavy, limiting accessibil-
 ity.

 Vertical mixing of water and air creates
 cross-contamination potential.

 Hazard posed to surface environment if
 toxic compounds encountered.

 Organic foam additives for cuttings'
 removal may contaminate samples.
Table 1-1. Continued
                                                                   12

-------
Johnson, Inc., 1972; and Scalf and others, 1981). The
table  also gives a concept of  the  advantages and
disadvantages  that  need  to be considered  wrlen
choosing a  drilling technique for different site and
monitoring situations (see, also, Lewis, 1982; Luhdbrff
and Scalmanini, 1982; Minning, 1982; and Voyt|ek,
1983).

Hollow-stem  augering is one of the most  desirable
drilling methods for constructing monitoring wells. No
drilling fluids are used and disturbance of the geologic
materials penetrated is minimal. Depths are usually
limited to no more than 150 feet.  Typically, auger rigs
are not used when consolidated rock must be penetrated.
In formations where the borehole will not stand open,
the monitoring well can be constructed inside the hollow-
stem auger priorto its removal from the hole. Genera lly,
this limits the diameter of the well that can be built 13 4
inches. The hollow-stem has an added advantage in
offering the ability to collect continuous in situ geologic
samples without removal of the auger sections.
The  solid-stem auger is most useful in fine-grain
pd,
unconsolidated materials that will not collapse when
unsupported. The method is similar to the hollow-st 3m
except that the auger flights must be removed from the
hole to allowthe insertion of the well casing and screen.
Cores cannot be collected when using a solid-stem.
Therefore, geologic sampling must rely on cuttings that
come to the  surface, which is an unreliable method
because the depth from which the cuttings are derived
is not precisely known.                          |

Cable-tool drilling is one of the oldest methods used in
the water  well industry.  Even though the rate
penetration is rather slow, this method offers many
advantages for monitoring well construction. With
cable-tool, excellent formation samples can be collec
and  the presence of thin permeable  zones can
 of
tie
:ed
be
detected. As drilling progresses, a casing is normally
driven and this provides an ideal temporary cas ng
within which to construct the monitoring well.

In air-rotary drilling, air is forced down the drill stem and
back  up the borehole  to remove the cuttings. This
technique has been found to be particularly well sui:ed
to drilling in fractured rock. If the monitoring is intended
for  organic  compounds, the  air must be filtered to
ensure that oil from the air compressor is not introduced
into the formation to be monitored. Air-rotary should jiot
be used in highly contaminated environments because
the water and cuttings blown out of the hole are difficult
to control and can pose a hazard to the drill crew and
observers. Where volatile compounds are of  interest,
air-rotary can volatilize them and cause water samples
to be unrepresentative of in situ conditions. The use of
foam additives to aid cuttings' removal presents the
opportunity for organic contamination of the monitoring
well.

Air-rotary with  percussion hammer increases  the
effectiveness of air-rotary for  materials likely to cave
and highly  creviced formations. Addition of  the
percussion hammer gives air-rotary the ability to drive
casing,  which reduces the loss of air circulation in
fractured rock and aids in maintaining an open hole in
soft formations. The capability to construct monitoring
wells inside the driven casing, prior to its being pulled,
adds to the appeal of air-percussion. However, the
problems with contamination and crew safety must be
considered.

Reverse circulation rotarydrilling has limited application
for  monitoring well construction. Reverse circulation
rotary requires that large quantities of water be circulated
down the borehole  and up the drill stem to remove
cuttings. If permeable formations are  encountered,
significant quantitiesof watercan move into the formation
to be monitored, thus altering the quality of the water to
be sampled.

Hydraulic or "mud" rotary is probably the most popular
method used in the waterwell industry. Hydraulic rotary,
however, presents some disadvantages for monitoring
well construction. In hydraulic rotary technique, adrilling
mud (usually bentonite) is circulated down the drill stem
and up the  borehole to remove  cuttings. The  mud
creates a wall on the side of the borehole that must be
removed from the  screened area by development
procedures. With small diameterwells, the drilling mud
is not always completely removed. The ion-exchange
potential of most drilling muds is high and may effectively
reduce the concentration of trace metals in water entering
the well. In addition,  the use of biodegradable, organic
drilling muds, rather than  bentonite, can introduce
organic components to water sampled from the well.

Most ground-water monitoring wells will be completed
in glacial or unconsolidated materials, and generally will
be  less than 75  feet in depth. In these applications,
hollow-stem augering usually will be the method of
choice. Solid-stem auger, cable-tool, and air-percussion
also offer advantages depending on the geology and
contaminant of interest.

Geologic Samples
Permit applications fordisposal of waste materials often
require that geologic samples be collected at the disposal
site. Investigations  of ground-water  movement  and
contaminant transport also should include the collection
of geologic samples for physical inspection and testing.
                                                13

-------
Stratigraphic samples  are  best collected  during
monitoring well drilling.

Samples can be collected continuously, at each change
in Stratigraphic unit, or, in homogeneous materials, at
regular intervals. These samples may later be classified,
tested, and analyzed for physical properties, such as
particle-size distribution, textural classification, and
hydraulic conductivity, and for chemical analyses, such
as ion-exchange capacity, chemical composition, and
specific parameter teachability.

Probably the most common method of material sampling
is a "split-spoon" sampler.  This device consists of a
hollow cylinder, 2  inches in diameter, that is 12 or 18
inches long, and split in half lengthwise. The halves are
held together with threaded couplings at each end; the
top end attaches to the drill rod, and the bottom end is
a drive shoe (Figure 1 -1 Oa). The sampler is lowered to
the bottom of the hole and driven ahead of the hole with
a weighted "hammer" striking an anvil  at the upper end
of the drill rod. The sample is forced up the inside of the
hollow tube and is held in place with a basket trap or flap
valve. The trap or valve allows the sample to enter the
tube but not exit,  although retention  of noncohesive,
sandy material in  the tube is often difficult. After the
sampler is withdrawn from the  hole, the sample is
removed by unscrewing the couplings and separating
the collection tube.

Anothercommon sampler is the thin wall tube or "Shelby"
tube. These tubes are usually 2 to  5-1/2  inches in
diameter and about 24 inches long. The cutting edge of
the tube is sharpened and the upper end is attached to
a coupling head by means of cap screws or a retaining
pin (Figure 1-1 Ob). A Shelby tube has a minimum ratio
of wall area to sample  area and creates the least
disturbance to the sample of any drive-type sampler in
current use (for hydraulic conductivity tests, minimal
disturbance is critical). After retraction, the tube is
disconnected from the head and the sample is forced
fromthetubewith a jackorpress. If sample preservation
is a majorconcern, the tube can be sealed and shipped
to the laboratory.

Apart from permit requirements, material samples are
very helpful for deciding  at what depth to complete a
monitoring  well.  Unexpected changes encountered
during drilling can alter preconceived ideas concerning
the local ground-waterflow regime. In  many instances,
the driller will be able to  detect a  variation in the
formation by a change in penetration rate,  sound, or
"feel" of the drilling rig. However, due to the lag time for
cuttings to come to the surface  and the  amount of
mixing the cuttings may undergo as they come up the
           Kl
          44-

             :...

          til
      a.
b.
Figure 1-10. Cross-Sectional Views of (a) Split
Spoon and (b) Shelby Tube Samplers (from
Mobile Drilling Co., 1972)

borehole, the only way to exactly determine the character
of the subsurface is to stop drilling and collect a sample.

Case History

Several different types of  monitoring wells were
constructed during the investigation of a volatile organic
contaminant plume  in northern Illinois (Wehrmann,
1984). A brief summary of the types of wells employed
and the reasons for their use helps illustrate how an
actual ground-water quality monitoring problem was
approached.

During the final  weeks of a 1 -year study of nitrate in
ground water in north-central Illinois, the presence of
several organic compounds was detected in the drinking
water of all five homes sampled within a large rural
residential  subdivision. The principal compound,
trichloroethylene (TCE), was present in concentrations
ranging from 50 to 1,000 micrograms per liter (u.g/L). All
the homes in the subdivision used private wells, 65 to 75
feet  deep, that  tapped a  surficial sand and gravel
deposit. Figure 1-11 shows a geologic cross section of
the study area.

Two immediate concerns needed to be addressed.
First, how many other water wells were affected and,
second, what was the contaminant source? Early
thoughts  connected the TCE to the contamination
potential of the large number of septic systems in the
subdivision. Earlier work (Wehrmann, 1983)  had
established a south-southwest direction of ground-
                                               14

-------
High Terrace T
WW
Rock River Floodplain Low Terrace I
Glenwood-St. Peter Sandstone
WW - Water Well
T - Tollway boring
;':':[•'.:.-] Outwash sand snd gravel ^-Vj-':.'Vji-:'--.';.'-.:}i':'--.'V::' v.5'•;
i |V',vM Lacustrine sands, silt and

Organic materials (or) buried soil XiivvV-'.'^Vvi^'
BOO-
L-BOO
Figure 1-11. East-West Cross Section Across R^ock River Valley at Roscoe (from Berg and others,
1981)
water flow beneath the subdivision. Because the area
upgradient of the subdivision was primarily farmland,
several monitoring wells were placed in that area to help
confirm or deny the possibility that the septic systems
were the source of the TCE.

Five "temporary" monitoring wells were construct 3d
upgradient of the affected subdivision. Original plans
called for driving a 2-inch diameter sandpoint to deptjhs
from 40 to 70 feet. Water samples would be collected at
10-foot intervals as the point was driven. Once 70 feet
was reached, the sandpoint would be pulled, the hole
properly abandoned, and the point driven at a new
sampling location. The first hole was to be placed north
(upgradient) of a domestic well found to be highly
contaminated, and additional holes were to be placed
successively in an upgradient direction across the fie d.
In this manner, ground-watersamples could be collected
quickly at many depths and locations, the well materials
recovered, and the field left relatively undisturbed.

Once drilling commenced, however, it became clear
that driving sandpoints into the coarse sand and grav el
was not possible. Consequently, an air-percussion rig
was brought on site and a new approach was established.
A 4-inch diameter screen, 2 feet long and with a drive
shoe, was welded to a 4-inch diameter steel casing.
This assembly was driven by air hammer to the desired
sampling depth. The bottom of the drive shoe, being
open, forced the penetrated geologic materials into the
casing and screen. These materials were then removed
by air rotary once the desired depth was reached. All
well construction materials were steam-cleaned priorto
use to avoid cross-contamination. Figure 1-12 shows
the locations of the temporary well sitesand the analytical
results forTCE from samples bailed at depths of 40 and
50 feet.

The temporary sampling program revealed that the
contaminant source was outside of the subdivision. Due
to the construction and sampling methods employed for
these wells, emphasis was not placed on the quantitative
aspects of the sampling results; however, important
qualitative conclusions were made. The temporary wells
confirmed the presence of VOCs directly upgradient of
the subdivision and provided information for the
15

-------
I
                                        HOUSE WELL
                                         SAMPLED
                                        SAME WEEK.
                                      2178 ppb TCE 6> 65"
           Figure 1*12. Locations and TCE Concentrations for Temporary Monitoring Wells at Roscoe, Illinois
           (from Wehrmann, 1984)
           subsequent location and  depth of  nine  permanent
           monitoring wells.

           Due to the problems associated withorganic compound
           leachability and adsorptionf rom PVC casing and screen,
           flush-threaded stainless steel casing and screen, 2
           inches  in diameter, were used for the  permanent
           monitoring wells.  The screens were 2 feet long with
           0.01-inch wire-wound slot openings. All materials
associated with the monitoring well  construction,
including the drill rig, were steam-cleaned prior to the
commencement of drilling to avoid organic contamination
from cutting oils and grease. Priorto use, the casing and
screen materials were kept off site in a covered, protected
area.  To ensure that the sandy materials would not
collapse after drilling, casing lengths and the screen
were  joined  aboveground and placed  inside of the
augers before the auger flights were pulled out of the
                                                          16

-------
hole.  The sand and gravel below the water tabl
collapsed around the screen and casing as the auger
were removed. To prevent vertical movement of watejr
down along the casing, about 3 feet of a wet bentonite/
cement mixture was placed in the annulus just above
the water table. Cuttings  (principally clean, fine to
medium sand) were backfilled above the bentonite/
cement seal to within 4 feet of  land surface. Another
bentonite/cement mixture was placed to form a seal at
ground surface, further preventing movement of water
down along the well casing. A 4-inch diameter steel
protective cover with locking cap was placed over the
casing and into the surface  seal to protect against
vandalism.

The nine wells were drilled at four locations with paired
wells at three sites and a nest of three wells at one site
(Figure 1-13). The locations were based on the analytical
results of the samples taken from the temporary wells
and basic knowledge of the ground-water flow direction.
                                 JH'E HAVEN] 11 j-M
                                 3CIVISION HILH     I

                            \l=f^^  "^TvER

Figure 1-13. Location of Monitoring Well Nests ar|d Cross Section A-A* at Roscoe, Illinois (from
Wehrmann, 1984)
                                               17

-------
Locations were numbered as nests 1 through 4 in order
of their construction. Nest 1, located immediately north
of the affected subdivision, consists of three wells
completed at depths of approximately 60, 70, and 80
feet below ground surface. Nest 2 consists of two wells,
one 50 feet deep, and the other 60 feet deep. Nest 3
consists of two wells 40 and 55 feet deep, while nest 4
consists of two wells 50 and 60 feet deep.

Prior to completing the monitoring wells, it was felt an
additional well, 100 feet deep and adjacent to nest 1,
was needed to further define the vertical extent of the
contaminant plume. Because the hollow-stem auger rig
was no longer available, arrangements were made to
use a cable-tool rig. The well was constructed over a
period of 2 days, which was somewhat slower than any
of the other methods previously used (but typical of
cable-tool speeds). A 6-inch casing was driven several
feet, a bit was used to break up the materials inside the
casing, and then the materials were removed from the
casing with a dart-valve bailer. This procedure was
repeated until 100 feet was reached; then the well
casing and screen were screwed together and lowered
down the hole. The 6-inch casing was then pulled back,
which allowed the hole to collapse about the well, which
was constructed of stainless steel exactly like the other
nine monitoring wells. All drilling equipment and well
construction materials were steam-cleaned prior to
use.

Appraisal of the sampling results of the monitoring wells
and the domestic wells in the area produced the pictorial
representations shown in Figures 1-14 and 1-15. Figure
1-14 conceptually illustrates a downgradient cross
section of the TCE plume in the vicinity of monitoring
nests 2,3, and 4. Figure 1-15 shows the likely extent of
the VOC contaminant plume. This map includes a
limited amount of data from privately owned monitoring
wells located on industrial property just upgradient of
monitoring nests 2 and 4. The dashed lines indicate the
probable extent of the contaminant plume based on the
dimensions of the plume where it passes beneath the
developed area along the Rock River.

This monitoring situation clearly indicates the role
different drilling and construction techniques can play in
a ground-water sampling strategy. In each instance,
much consideration was given to the effect the methods
used for construction and sampling would have on the
resultant chemical data. Where quantitative results for
a fairly "quick" preliminary investigation were not
necessary and driving sandpoints was too difficult, air-
percussion rotary methods were deemed acceptable.
For the placement of the permanent monitoring wells,
wells that may become crucial for contaminant source
identification and possibly for litigation, the hollow-stem
auger was the technique of choice. Finally, when the
hollow-stem auger was not available, a cable-tool rig
was chosen. Since only one hole was to be drilled, the
relative slowness of the method became less important.
Also, the depth of completion (100 feet) in the cavey
sand and gravel made the cable-tool preferable over
the hollow-stem. In addition, each method chosen was
capable of maintaining an open hole without the use of
drilling mud, which could have affected the results of the
chemical analyses of the ground water.
JNest4
Land Surface
A
Nest 2
i
Top of Saturated Zone (Water Table)
\\
\ 500-1000 Hg/L ) \\
100-250M9/L
<100 H9/L
\
Figure 1-14. Cross Section A-A' through Monitoring nests 2,3, and 4, Looking in the Direction of
Ground-Water Flow (from Wehrmann, 1984)
18

-------
                       OLOE FARM/TRESEMER SUBDIVISION
                           21 DOMESTIC WELLS
                           SAMPLED 10/3-4/83
                                10 A
Figure 1-15. General Area of Known TCE Contamination (from Wehrmann, 1984)
Summary

Critical considerations for the design of ground-water
quality monitoring networks include alternatives for wel
design and drilling techniques. With a knowledge of thq
principal chemical constituents of interest and the local
hydrogeology,  and  appreciation of subsurfac
geochemistry, appropriate materials for well design an
drillingtechniques can be selected. Wheneverpossible
physical disturbance and the amount of foreign material^
introduced into the subsurface should be minimized.

The choices of drilling methods and well construction
materials are very important  in every type of ground-
water  monitoring  program.  Details of network
construction can introduce significant bias into monitoring
data, which frequently  may be corrected only by
repeating the process  of well  siting, installation,
completion, and development. This can be quite costly
in time, effort, money, and loss of information. Undue
expense  is avoidable if planning decisions are made
cautiously with an eye to the future.

The expanding scientific literature on effective ground-
water  monitoring techniques  should be  read  and
evaluated on a continuing basis. This information will
help supplementguidelines, such as this.forapplications
to specific monitoring efforts.

-------
References

Aller, and  others, 1989, Handbook  of Suggested
Practices forthe design and installation of ground-water
monitoring wells:  National Water Well Association,
Dublin, OH EPA 600/4-89/034.

Barcelona,  M.J., J.P. Gibb, J.A. Helfrich, and E.E.
Garske, 1985, Practical guideforground-water sampling:
Illinois State Water Survey,  U.S.  Environmental
Protection  Agency,  Robert S. Kerr  Environmental
Research  Laboratory, Ada,  OK, and  Environmental
Monitoring and Support Laboratory, Las Vegas, NV.

Barcelona, M.J., 1984, TOC determinations in ground
water: Ground Water, v.22, no.1, pp.18-24.

Barcelona,  M.J., J.A. Helfrich, E.E. Garske, and J.P.
Gibb, 1984, A  laboratory evaluation of ground water
sampling  mechanisms:  Ground Water  Monitoring
Review, v.4, no. 2, pp. 32-41.

Barcelona,  J.J., J.P. Gibb, and R.A. Miller, 1984. A
guide to the selection of materials for monitoring well
construction and ground-water sampling. Illinois State
Water Survey Contract Report 327. Illinois State Water
Survey, Champaign, IL.

Berg, R.C., J.P. Kempton, and A.N.  Stecyk. 1981,
Geologyforplanning in Boone and Winnebago Counties,
Illinois: Illinois State Geological Survey Circular 531,
Illinois State Geological Survey, Urbana,

Brown, K.W., J. Green, and J.C. Thomas, 1983, The
influence of selected organic liquids on the permeability
of clay liners:  Proc. of the Ninth Annual Research
Symposium: Land Disposal, Incineration, and Treatment
of Hazardous Wastes. U.S. Environmental Protection
Agency SHWRD/EPCS, May 2-4, 1983, Ft. Mitchell,
KY.

Burkland, P.W., and E. Raber, 1983, Method to avoid
ground-watermixingbetweentwoaquifersduringdrilling
and  well  completion procedures: Ground  Water
Monitoring Review, v. 3, no. 4, pp. 48-55.

Campbell,  M.D. and J.H. Lehr, 1973, Water well
technology: McGraw-Hill Book Company, New York,
NY.

Fenn, D., E. Cocozza, J. Isbister, 0. Braids, B. Yare, and
P. Roux, 1977, Procedures manual for ground water
monitoring at solid waste disposal facilities (SW-611).
U.S. Environmental Protection Agency, Cincinnati, OH.

Gibb, J.P.,  R.M.  Schuller,  and  R.A. Griffin, 1981,
Procedures for the collection of representative water
quality data  from monitoring wells:  Cooperative
Groundwater Report. Illinois State Water and Geological
Surveys, Champaign, IL.

Gillham, R.W., M.J.L Robin,  J.F. Barker, and J.A.
Cherry, 1983, Groundwater monitoring and sample
bias: American Petroleum Institute Publication 4367,
Environmental Affairs Department.

Illinois State Water Survey and Illinois State Geological
Survey, 1984, Proceedings of the 1984 ISWS/ISGS
Groundwater  Monitoring Workshop. February 27-28,
Champaign, IL.

Illinois State Water Survey and Illinois State Geological
Survey, 1982, Proceedings of the 1982 ISWS/ISGS
Groundwater  Monitoring Workshop. Illinois Section of
American Water Works Association. February 22-23,
1982, Champaign, IL.

Johnson, T.L., 1983,  A comparison of well nests  vs.
single-well completions: Ground Water Monitoring
Review, v. 3, no. 1, pp. 76-78.

Johnson, E.E., Inc., 1972,  Ground water and wells.
Johnson Division, Universal Oil Products Co., St. Paul,
MN.

Keith, S.J., LG. Wilson, H.R. Fitch, and D.M. Esposito,
1982, Sources of spatial-temporal variability in Ground-
Water Quality Data and Methods of Control: Case
Study of the  Cortaro  Monitoring  Program, Arizona:
Proc. of the Second National Symposium on Aquifer
Restoration and Ground Water Monitoring. National
Water Well Association, May 26-28, Columbus, OH.

Lewis, R.W.,  1982, Custom designing of monitoring
wellsforspecific pollutants and hydrogeologicconditions:
Proc. of the Second National Symposium on Aquifer
Restoration and Ground Water Monitoring. National
Water Well Association, May 26-28, Columbus, OH.

Luhdorff, E.E., Jr., and J.C. Scalmanini, 1982, Selection
of drilling method, well design and sampling equipment
for wells for monitoring organic contamination: Proc. of
the Second National Symposium on Aquifer Restoration
and  Ground Water Monitoring, National Water Well
Association, May 26-28, Columbus, OH.

Mackay, D.M., P.V. Roberts, and J.A. Cherry, 1985,
Transport  of  organic contaminants in  groundwater:
Environmental Science & Technology, v. 19, no. 5,  pp.
384-392.

Miller, G.D., 1982, Uptake and release of lead, chromium
and trace level volatile organics exposed to synthetic
well casings: Proc. of Second National Symposium on
                                               20

-------
Aquifer  Restoration  and Ground Water Monitoring.
National Water Well Association, May 26-28, Columbus,
OH.

Minning, R.C.,  1982, Monitoring well design  aid
installation: Proc. of Second National Symposium jn
Aquifer  Restoration  and Ground Water Monitoring.
National Water Well Association, May 26-28, Columbijis,
OH.

Mobile  Drilling  Company,  1972,  Soil sampling
equipment-accessories: Catalog 650. Mobile Drillipg
Company, Indianapolis, IN.

Morrison, R.D. and P.E. Brewer, 1981, Air-lift sampldrs
for zone of saturation monitoring:  Ground Wat,er
Monitoring Review, v. 1, no. 1, pp. 52-55.

Naymik, T.G. and M.E. Sievers, 1983, Groundwater
tracer experiment (II)  at Sand Ridge State Forest,
Illinois: Illinois State WaterSurvey Contract Report 33J4,
Illinois State Water Survey, Champaign, IL.

Naymik, T.G. and JJ. Barcelona, 1981, Characterization
of a  contaminant plume in groundwater, Meredosi|a,
Illinois: Ground Water, v. 16, no. 3, pp. 149-157.

O'Hearn, M., 1982,  Groundwater monitoring at the
Havana Power Station's ash  disposal  ponds  arid
treatment lagoon: Confidential Contract Report, Illincjis
State Water Survey, Champaign, IL.

Perry,  C.A. and R.J. Hart, 1985,  Installation of
observation wells on hazardous waste sites in Kansas
using a hollow-stem auger: Ground Monitoring Review,
v. 5,  no. 4, pp. 70-73.

Pettyjohn, W.A. and A.W. Hounslow, 1982, Organic
compounds and ground-waterpollution: Proc. of Second
National Symposium on Aquifer Restoration and Ground
Water Monitoring. National Water Well Associatiop,
May  26-28, Columbus, OH.

Pettyjohn, W.A.,  W.J. Dunlap,  R. Cosby,  and J.
Keeley,  1981, Sampling  ground water  for organic
contaminants: Ground Water v. 19, no. 2, pp.  180-189.
Pfannkuch, H.O., 1981, Problems of monitoring network
design to detect unanticipated contamination: Proc. of
First  National  Ground  Water Quality Monitoring
Symposium and Exposition,  National Water Wejll
Association, May 29-30, Columbus, OH.

Pickens, J.F., J.A. Cherry, R.M. Coupland, G.E. Grisak,
W.F. Merritt, and B.A. Risto, 1981, A multi-level device
for ground-water sampling: Ground Water Monitoring
Review, v. 1, no. 1, pp. 48-51.

Rinaldo-Lee, M  . B ., 1983, Small- vs. large-diameter
monitoring wells: Ground Water Monitoring Review, v.
3, no. 1, pp. 72-75.

Scalf, M.R., J.F. McNabb, W.J. Dunlap, R.L. Cosby,
and  J.  Fryberger,  1981, Manual of  ground-water
sampling procedures:  NWWA/EPA Series,  National
Water Well Association, Worthington, OH. EPA-600/2-
81-160.

Schalla, R.,and R.W. Landick, 1985, Anew valved and
air-vented surge plungerfor developing small-diameter
monitor wells: Proc. of Third National Symposium and
Exposition on Ground-Water Instrumentation, National
Water Well Association, October 2-4, San Diego, CA.

Sosebee, J.B., Jr. and others 1982. Contamination of
groundwater samples with PVC adhesives and PVC
primer from monitor wells: Environmental Science and
Engineering, Inc., Gainesville, FL.

Torstensson,  B.A.,  1984, A  new system  for ground
water monitoring: Ground Water Monitoring Review, v.
4, no. 4, pp. 131-138.

Voytek, J.E., Jr., 1983, Considerations in the design
and  installation  of monitoring  wells:  Ground  Water
Monitoring Review, v. 3, no. 1, pp. 70-71.

Walker, W.H., 1974, Tube  wells, open  wells, and
optimum ground-water resource development: Ground
Water, v. 12, no. 1, pp. 10-15.

Wehrmann, H.A., 1984, An investigation of a volatile
organic chemical plume in northern Winnebago County,
Illinois: Illinois State WaterSurvey Contract Report 346,
Illinois State Water Survey, Champaign,

Wehrmann, H.A., 1983, Monitoring well  design and
construction: Ground Water Age, v. 17, no. 8, pp. 35-38.

Wehrmann, H.A., 1983, Potential nitrate contamination
of groundwater in the Roscoe Area, Winnebago County,
Illinois: Illinois State WaterSurvey Contract Report 325,
Illinois State Water Survey, Champaign, IL.

Wehrmann, H.A., 1982, Groundwater monitoring for fly
ash leachate, Baldwin Power Station, Illinois Power
Company. Confidential Contract Report, Illinois State
Water Survey, Champaign, IL.
                                              21

-------
                                            Chapter 2
                                 GROUND-WATER SAMPLING
introduction

Background
Ground-water sampling  is conducted to provide
information  on the condition of subsurface water
resources. Whether the goal of the monitoring effort is
detection or assessment of contamination, the
information gathered during sampling efforts must be of
known quality and be well documented. The most
efficientway to accomplish these goals is by developing
a sampling protocol, which is tailored to the information
needs of the program and the hydrogeology of the site
or region under investigation. This  sampling protocol
incorporates detailed descriptions of sampling
procedures and other techniques that, of themselves,
are not sufficient to document data quality or reliability.
Sampling protocols are central parts of  networks  or
investigatory strategies.

The need for reliable ground-water sampling procedures
has  been  recognized for years by a variety  of
professional,  regulatory, public, and private  groups.
The technical basis for the use of selected sampling
procedures for environmental  chemistry studies has
been developed for surface-water applications overthe
last four decades.  Ground-water quality monitoring
programs, however, have unique needs and goals that
are fundamentally different from previous  investigative
activities. The reliable detection and assessment  of
subsurface contamination require minimal disturbance
of  geochemical and hydrogeologic conditions during
sampling.

At this time, proven well construction, sampling, and
analytical protocols for ground-water sampling have
been developed for many of  the  more problematic
chemical constituents of interest. However, the
acceptance of these procedures and protocols must
await more careful documentation and firm regulatory
guidelines for monitoring program execution. The time
and  expense of characterizing actual subsurface
conditions place severe restraints on the methods that
can  be employed.  Since the technical basis for
documented, reliable drilling, sample collection, and
handling procedures  is in the  early stages  of
development, conscientious efforts to document method
performance under real conditions should be a part of
any ground-water investigation (Barcelona and others,
1985; Scalf and others, 1981).

Information Sources
Much of the literature on routine ground-water monitoring
methodology has been published in the last 10 years.
The bulk of this work has emphasized ambient resource
or contaminant resource  monitoring (detection and
assessment),  rather than case  preparation  or
enforcement efforts. General references that are useful
to the design and execution of sampling efforts are the
U.S.  Geological Survey (1977), Wood (1976), the U.S.
Environmental  Protection Agency (Brass and others,
1977; Dunlap and others, 1977; Fenn and others, 1977;
Sisk, 1981) and others (National Council of the Paper
Industry, 1982;Tinlin, 1976). In large part, these works
treat sampling  in  the context of overall monitoring
programs, providing descriptions of available sampling
mechanisms, sample collection techniques, and sample
handling  procedures. The impact of  specific
methodologies on the usefulness or reliability of the
resulting data has received little discussion (Gibb and
others, 1981).

High-quality chemical data collection is essential in
ground-water monitoring  programs. The  technical
difficulties involved in "representative" samplings have
been recognized only recently (Gibb and others, 1981;
Grisak and  others, 1978). The long-term collection of
high-quality ground-water chemistry  data  is more
involved than merely selecting a sampling mechanism
and agreeing on sample handling procedures. Efforts to
detect and assess contamination can be unrewarding
without accurate (i.e., unbiased) and  precise  (i.e.,
comparable and complete) concentration  data  on
                                               22

-------
ground-waterchemical constituents. Also, the expense
of data  collection and management argue tor
documentation of data quality.

Gillham and others (1983)  published a  very useul
reference on the principal sources of bias and imprecision
in ground-water monitoring results. Their treatment is
extensive and stresses the minimization of random
error, which can enter into well construction, sample
collection, and sample handling operations. They furth'er
stress the importance of collecting precise data ovjer
time to maximize the effectiveness of trend analysis,
particularly for regulatory purposes. Accuracy also is
very important, since the ultimate reliability of statistical
comparisons  of results from different wells (e.g.,
upgradient versus downgradient samples) may depend
on differences  between mean values for selected
constituents from relatively small replicate sample sets.
Therefore,  systematic error must be controlled  by
selecting proven methods for establishing  sampling
points and sample collection to ensure known levels pf
accuracy.

The Subsurface Environment
The subsurface environment may be categorized broad y
into two zones, the unsaturated or vadose zone and the
saturated zone. The use of the term "vadose" is morje
accurate because isolated saturated areas may exist in
the unsaturated  zone  above the  water table  of
unconfined aquifers.

Investigators  have discovered  recently  that th
subsurface is neither devoid of oxygen (Winograd an!
Robertson, 1982) nor sterile  (Wilson and  McNabl
1983; Wilson and others, 1983).  These facts  m
significantly influence the mobility and persistence
chemical species, as well as the transformations of th
original  components  of  contaminant mixture
(Schwarzenbach and others,  1985)  that have bee
released to the subsurface.

The subsurface environment also is quite different f ror i
surface water systems  in that vertical gradients in
pressure and dissolved gas content have been observed
within the usual depth ranges of monitoring interest
(i.e., 1 to 150 m [3 to 500 ft]). In some cases, these
gradients can be linked to well-defined hydrologic ojr
geochemical processes. However, reports of apparently
anomalous geochemical processes have increased in
recent years,  particularly at contaminated sites
(Barcelona and Garske, 1983;Heatonand Vogel, 1981
Schwarzenbach and others, 1985; Winograd and
Robertson, 1982; Wood and Petraitis, 1984).

The subsurface environment is not as readily accessible i
as surface water systems, and some disturbance is
necessary to collect samples of earth  materials or
ground water. Therefore, "representative" (i.e., artifact
or error free) sampling is really a function of the degree
of detail needed to characterize subsurface hydrologic
and geochemical conditions and the care taken to
minimize disturbance of these conditions in the process
(Claasen,  1982).  Each well or boring represents  a
potential conduit for short-circuited contaminant
migration or ground-water flow,  which  must be
considered a potential liability to investigative activities.

The subsurface environment is dynamic over extended
time frames and the processes of recharge and ground-
water  flow are  very  important  to a thorough
understanding of the system. Detailed descriptions of
contaminant distribution, transport, and transformation
necessarily rely on the understanding of basic flow and
fluid transport processes. Short-term investigations may
only provide a snapshot of contaminant  levels or
distributions. Since water-quality monitoring data are
normally collected on discrete dates, it is very important
that reliable collection methods are used to assure high
data quality over the course of the investigation. The
reliability of the  methods  should  be  investigated
thoroughly during the preliminary phase of monitoring
network implementation.

Although the scope  of this discussion is on  sampling
ground water for  chemical  analysis, the same data
quality requirements apply to water-level measurements
and to hydraulic conductivity testing.  These hydrologic
determinations form the basis for interpreting chemical
constituent data and  may well limit the validity of fluid or
solute transport model applications. Hydrologic
measurements must be included in the development of
the quality assurance/quality control (QA/QC) program
for ground-water quality monitoring networks.

The Sampling Problem  and Parameter Selection
Cost-effective water-quality sampling is difficult in
ground-water systems because provenfield procedures
have not been extensively documented. Regulations
that call for  "representative  sampling" alone are not
sufficient to  ensure  high-quality data collection. The
most appropriate monitoring  and sampling procedures
for a ground-water quality network will depend on the
specific purpose of the program. Resource evaluation,
contaminant detection, remedial action assessments,
and litigation studies are purposes for which effective
networks can be designed once the information needs
have been identified. Due to the time, personnel needs,
and cost of most water-quality monitoring programs, the
optimal network design should be phased so as to make
the most of the available  information as it is collected.
This approach allows for the gradual refinement of
program goals as the network is implemented.
                                                23

-------
Two fundamental considerations are common to most
ground-waterqualitymonitoringprograms: establishing
individual sampling points (i.e., in space and time) and
determining the elements of the water sampling protocol
that will be sufficient to meet the information needs of
the overall program. The placement and number of
sampling points can be phased to gradually increase
the scale of the monitoring program. Similarly, the
chemical constituents of initial interest should provide
background ground-water quality data from which a list
of likely contaminants may be prepared as the program
progresses. Table 2-1 shows candidate chemical and
hydrologic parameters for both detective and
assessment monitoring activities. Special care should
be taken to account for possible subsurface
transformation of the principal pollutant species. Ground-
water transport of contaminants can produce chemical
distributions that vary substantially overtime and space.
In particular, transformation of organic compounds can
change substantially the identity of the original
contaminant mixture (Mackay and others, 1985;
Schwarzenbach and others, 1985).

Detective Monitoring
Chemical Parameters*
pH. 2", TOC, TOX, Alkalinity, TDS, Eh, CI", NO,; SO/, PO,',
SK3,, Na: K*, Ca", Mg", NH,', Fe, Mn
Hydrologic Parameters
Water Level, Hydraulic Conductivity
Assessment Monitoring
Chemical Parameters*
pH, Q". TOC, TOX, Alkalinity. TDS, Eh, CI', NO,: SO.: P0,s,
SiO,, B, Na', K", Ca", Mg", NH/. Fe, Mn. Zn. Cd, Cu, Pb, Cr.
NS. Ag, Hg, As, Sb, Se, Be
Hydrologic Parameters
Water Level, Hydraulic Conductivity
•Q ' = specific conductance, a measure of the charged species in
solution.
In this respect, monitoring in the vadose zone is attractive
because it should provide an element of "early" detection
capability. The methodologies available for this type of
monitoring have been under development for some
time. There are distinct limitations, however, to many of
the available monitoring devices (Everett and McMillion,
1985; Everett and others, 1982; Wilson, 1981; Wilson,
1983), and it is frequently difficult to relate observed
vadose zone concentrations quantitatively to actual
contaminant distributions in ground water (Everett and
others, 1984; Lindau and Spalding, 1984). Soil gas
sampling techniques and underground storage tank
monitors have been commercially developed that can
be extremely useful for source scouting. Given the
complexity of vadose zone monitoring procedures and
the need for additional investigation (Bobbins and
Gemmell, 1985), implementing these techniques in
routine ground-water monitoring networks may be
difficult.

This chapter addresses water-quality sampling in the
saturated zone, reflecting the advanced state of
monitoringtechnology appropriate forthis compartment
of the subsurface. There are a numberof useful reference
materials forthe development of effective ground-water
sampling protocols, which include information on the
types of drilling methods, well construction materials,
sampling mechanisms, and sample handling methods
currently available (Barcelona and others, 1985;
Barcelona and others, 1983; Gillham and others, 1983;
Scalf and others, 1981; Todd and others, 1976). To
collect sensitive, high-quality contaminant concentration
data, investigators must identify the type and magnitude
of errors that may arise in ground-water sampling.
Figure 2-1 presents a generalized diagram of the steps
involved in sampling and the principal sources of error.
Table 2-1. Suggested Measurements for Ground-
Water Monitoring Programs

Contaminant detection is generally the most important
aspect of a water-quality program, and must be assured
in network design. False negative contaminant readings
due to the loss of chemical constituents orthe introduction
of interfering substances that mask the presence of the
contaminants in water samples can be very serious.
Such errors may delay needed remedial action and
expose either the public or the environment to an
unreasonably high risk. False positive observations of
contaminants may call for costly remedial actions or
more intensive study, which are not warranted by the
actual situation. Thus, reliable sample collection and
data interpretation procedures are central to an optimized
network design.
Step
In-Situ Condition
Establishing a Sampling Point
Field Measurements

t
Sample Collection

Sample Delivery/Transfer
Field Blanks, Standards

Field Determinations

Preservation / Storage

Transportation
Sources of Error
Improper well construction/
placement; inappropriate
materials selection
Instrument malfunction;
operator error
Sampling mechanism bias;
operator error
Sampling mechanism bias;
sample exposure, degassing,
oxygeriation; field conditions
Operator error; matrix
interferences
Instrument malfunction;
operator error; tield conditions
Matrix interferences; handling/.
labeling errors
Delay; sample loss
Figure 2-1. Steps and Sources of Error in Ground-
Water Sampling
24