EPA/600/2-91/022
June 1991
STATE-Or-THE-ART FIELD HYDRAULIC
CONDUCTrVTTY TESTING OP COMPACTED SOILS
by
Joseph 0. Sai and Dav.d C. Anderson
K. W. Brown & Associates, Inc.
College Station, Texas 77840
Subcontractor to
PEI Associates, Inc.
Cincinnati, Ohio 45246
EPA Contract No. 6B-03-3413
Work Assignment No. C-23
PN 3741-23
Project Officer
Walter E. Grube, Jr.
Waste Minimization, Destruction and Disposal Research Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45263
-------
DISCLAIMER
The infcrmation in this document has been funded by che U.S.
Environmental Protection Agency under Contract No. 68-03-3413, to P£I
Associates, Inc. It has been subjected to the Agency's peer review and
administrative review, and it has been approved for publication as an EPA
docu-T.er.t . Mention of trade names or cornmercial produces does not
constitute ar. endorsement or recormendat ion for use.
ii
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial
products and practices frequently carry with them the increased
generation of materials that, if improperly dealt with, can threaten both
public health and the environment. The U.S. Environmental Protection
Agency is charged by Congress with protecting the Nation's land, air, and
water resources. Under a mandate of national environmental lawa, the
agency strives to formulate and implement actions leading to a compatible
balance between human activities and the ability of natural systems to
support and nurture life. These laws direct the EPA to perform research
to define our environmental problems, measure the impacts, and search for
solutions.
The Risk Reduction Engineering Laboratory is responsible for
planning, implementing, and managing research, development, and
demonstration programs to provide an authoritative, defensible
engineering basis in support of the policies, programs, and regulations
of the EPA with respect to drinking water, wastewater, pesticides, toxic
substances, solid and hazardous wastes, and Superfund-related activities.
This publication is one of the products of that research and provides a
vital communication link between the researcher and the user community.
This report documents the available technical information and
current state-of-the-art of field hydraulic conductivity test methods fcr
compacted soils. This information will be useful to those involved in
policy development, research, design, construction, and monitoring of
soils compacted for service as waste management system liners or
infiltration barriers in cover systems.
E. Timothy Oppelt
Di rector
Risk Reduction Engineering Laboratory
iii
-------
ABSTRACT
The Congressionally mandated performance standard for soil liners of
hazardous waste management facilities is a hydraulic conductivity of
1 x 10 m/a or less. In response to this statutory requirement, the
U.S. Environmental Protection Agency has issued guidance requiring that
facilities demonstrate this hydraulic conductivity in field tests.
Hydraulic conductivity test methods currently used on soil liners
were evaluated for their ability to meet the minimum requirements for
field teats; i.e., that the test be capable of measuring hydraulic
conductivities at least as low as 1 x 10 ^ m/s and that the values
obtained be representative of the overall soil liner. Of the few methods
capable of meeting the minimum requirements, even fewer are both
practical to use and rarely give false low values. Based on the
advantages of the methods evaluated, the best and most practical
currently available technologies for evaluating hydraulic conductivity
are large single-ring inf ilt rometers ar.d sealed double-ring
ir.f iltrometers. If correction factors are needed to bring the values
_c
obtained with single-ring devices belcw 1 x 1C .t./s, confirmatory tests
should be conducted with sealed double-ring infiltrometers.
The size of infiltrometers usee
2
2 -t. . Also, at least three separate
test fill to allow characterization
soil lir.er.
or. soil liners should be at least
tests should be conducted or. each
of the spatial variability in the
A long-term study is needed to allow a comparative evaluation of
candidate hydraulic conductivity testing devices. A large collection
lysimeter should be incorporated into the study to give the true overall
hydraulic conductivity value with which other values should be compared.
This report was submitted in fulfillment of contract number 68-03-
3413 by K. W. Brown & Associates, Inc., subcontractor to PEI Associates,
Inc., under the sponsorship of the U.S. Environmental Protection Agency.
This report covers the period from February 12, 1988 to May 14, 1989.
iv
-------
CONTENTS
Disclaimer ii
Forewctd iii
Abstract iv
Figures v i
Tables vi i i
1 . Introduction i
2 . Conclusions 3
3. Reconnendations 6
4. Field Hydraulic Conductivity Methods 8
Air-entry perr.eameters 0
Bcrehcle methods 14
Gueiph pemeameter 14
Boutwell nethod 21
rcrcjs probes 25
Surface infiltration devices 31
ASTM double-ring inf i It roxeter 31
.Modified double-ring ir.fiitrcmeters 34
Sox mfiltrometer 39
S i.-.gle-r ir.g inf i It roxeter 45
Sealed double-ring infiltromecer 5C
Collect ior. lysimeters 55
Methods under current study 62
Velocity permeameter 62
Porous plate infiltroxeters 66
Soil cores evaluated in the laboratory 72
References 8 0
V
-------
FIGURES
Nur.be r Page
1 Diagram of Che equipment for the air-entry perrr.earr.eter
technique 10
2 Schematic of the Guelph perraeaneter 15
3 C versus H/a relationship for a range of soil textures ... 17
4 Schematic diagram of the Boutwell borehole perr.eameter ... 22
5 SAT permeability system 27
6 Field hydraulic conductivity of a soil line; obtained jsing
a BA? pe rrr.eamete r 30
7 AS7M couble-ring ir.f-It rometer 32
3 Modified double-ring ir.filtrorr.eter with a level recorder . . 35
9 ASTM double-ring ir.f lit rorr.eter with tensiometer 37
10 Infiltration rate plotted as a function of time, using data
from the inner ring of a double-ring infilttometer .... 40
11 Box inflitrometer with detail of enbeddea base support ... 41
12 Hydraulic conductivity of the sanc-bentcnite liner rreasured
using four box infiltrometers 44
13 Open single-ring infiltrometer 46
14 Shape factor for single-ring test 48
15 Schematic of a sealed double-ring ir.f lit rometer 51
16 Field hydraulic conductivity values measured with an
ur.derdram and four sealed double-ring in f i it rcmeters . . . 54
17 SDRI data from a site in Wichita, Kansas 56
16 A test fill equipped with an underdrain record section ... 5S
vi
-------
FIGURES (continued)
N'urr.be r Page
19 Schematic of a velocity permea.T.eter €3
20 Apparent hydraulic conductivity versus time as measured with
the velocity penr.eameter 65
21 Schematic of an automated tension inf iltro.-neter 68
22 Schematic of Guelph infiltrometer 69
23 Schematic of a fixed-wall permeameter 74
24 Flexible-wall permeameter "7 5
vii
-------
TABLES
Number Page
1 Hydraulic Conductivity of Compacted Clay Soils at Two Heap
Leach Pads in California 13
2 Field Hydraulic Conductivity Data Obtained by Using
Air-Entry Permeameter and Laboratory Permeability Test
With Shelby Tube Samples 13
3 Comparison of Three Methods of Determining Saturated
Hydraulic Conductivity--Guelph Ferrneameter, Air-Entry
Ferneameter, and Soil Core 2 0
4 Comparison of Saturated Hydraulic Conductivity Data From
Guelph, 3ore'r.cle, and Air-Entry Permea.T.eters 20
5 Field ar.d Laboratory Measurements or Saturated Hydraulic
Conductivity in a California Clay 29
6 Hydraulic Conductivity of Two Prototype Clay Liners Using
a Single-Ring and a Double-Ring Infiltrometer 49
' Range of Hydraulic Conductivity Values for Varying Leachate
Heads at Twc Clay-Lmed Landfills in Wisconsin 61
b Comparison of Field and Laboratory Hydraulic Conductivity
Data 7 6
viii
-------
SECTION 1
INTRODUCTION
In the Hazardous and Solid Waste Amendments (HSWA) of 1984, Che
Congress of the United Scares mandated that where compacced soil liners are
used, their hydraulic conductivity shall be 1 x 10 ^ m/s or less. In
response to this statutory performance standard, U.S. Er.vi rcnir,enta 1
Protection Agency (EPA) issued guidance requiring all proposed soil liners
ir. hazardous waste management facilities to demonstrate this hydraulic
conductivity in field tests. The intent of this requirement was to obtain
accurate and realistic evaluations of how the compacted soil lir.er would
perfcrx. under field conditions. Numerous studies have demonstrated chat
values obtained ir. laboratory hydraulic conduce ivicy cescs are not reliable
indicators of the performance of soil liners under field conditions. No
single study has beer, conducted, however, to document the range of field
hydraulic conductivity test methods that are capable of adequately
evaluating the field performance of a compacted soil liner.
Field hydraulic conductivity tests can danr.age soil lir.ers in several
ways. Such darr.age can occur in the form, of holes drilled or trenches cut
ir. tr.e liner to facilitate the installation of testing equipment. Daxage
ray alsc occur because field hydraulic conductivity tests can take as long
as several weeks to ccnplete, during which time both climatic events and
weatr.erir.g processes can substantially damage a soil lir.er (Rehage et al.
19 j 6!
Intensive rainfall, cry periods, and freezes are examples of climatic
events tr.at can daxage a soil lir.er. Weathering processes that can affect
exposes soil liners include shrinking anc swelling caused by moisture
fluctuations, evaporative drying, or freeze-thsw cycles. Fcr avoidance cf
damage cue to pro.onced exposure, a soil liner shculd oe covered with
protective layers immediately after its construction. This need for
lxxediace covering cf a soil liner makes it difficult to concuct fielo
hydraulic conductivity tests. Consequently, the EPA (1335) recommended
that field hydraulic conductivity tests be conducted or. a test fill.
A test fill is a test section of a soil liner constructed with the
soil and equipment and by the same procedures used in the construction of
an actual soil lir.er. Its primary function is to serve as a facility where
testing can be conducted to verify that the specified hydraulic
conductivity values can actually be achieved with a full-scale liner. For
the data generated from the test fill to be useful, the quality assurance
testing on the test fill must be well documented, ar.d construction of the
1
-------
test fill must be the same as that used for the full-scale facility (U.S.
EPA 1SB6). In addition, both the teat fill and the hydraulic conductivity
tests used on the test fill should provide a sound basis for meeting the
following objectives:
_ g
1) Accurately measuring hydraulic conductivities of 1 x 1C m/s ar.d
13we:.
2) Obtaining hydraulic conductivity values that accurately represent
the properties of the full-scale soil liner.
It is especially important that the hydraulic conductivity values
obtained be representative of the full-scale soil liner. Rcgcwski et al.
{19£4) defined the representative elementary volume (REV) of a soil liner
as the smallest volume above which the variance no longer decreases
significantly. The R£V will be different for every liner ar.d will be
highly dependent on the natural variability of the scil material ar.d the
level of quality assurance exercised during construction of the soil liner.
At least three field hydraulic conductivity tests will be required to
define the REV for a given soil liner.
This report documents the available technical inf cr.r.at ion on field
hydraulic conductivity test methods for soil liners. The methods discussed
are currently used and readily a va liable for cete rmir.ing the hydraulic
conductivity cf soils compacted in the field.
Section 2 summarizes the findings anc specifies which field hydraulic
conductivity methods meet the r.mirr.urr. requirements for evaluating soil
liners. Section 3 presents general recommendstions regarding areas in
w.-.ich additional study is needed.
Section 4 discusses all the applicable field hydraulic conductivity
rr.et.nods ar.c the parameters used to characterize the selected methods for
measuring hydraulic conductivity.
-------
SECTION 2
CONCLUSIONS
Various methods are currently being used to evaluate the field
hydraulic conductivity of soil lir.ers. Or.ly a tew of these rr.ethods,
however, car. reliably meet the following requirements for evaluating soil
lir.ers :
1) A method should be capable of accurately measuring hydraulic
conductivities of 1 x 10~^ m/s and lower.
2) The values obtained should be representative of che overall
hydraulic conductivity cf a soil liner.
Of the few methods capable cf meeting the preceding minimum requirements,
ever, fewer are relatively simple to use, rarely give fa.se low values, ana
provide definitive results in a reasonable time frame. Key findings cf
edcr. memoi evaluated are summarized ir. the following paragraphs.
Air-entry permeameters can cive rapid fie-d measurements cf
vertical hydraulic conductivity. They involve the introduction of water
ir.to a sealed test rinc and tr.e measurement of chances ir. volume of water
in a graduated supply reservoir. After the wetting front advances into the
soil, the water supply is cut off and the resulting maxin-m suction is
measured. The saturated hydraulic conductivity is calculated frcm a
combination of the depth to the wetting front and the maximum suction
values. This method can measure hydraulic conductivity values cf 1 x 10 3
m/s and lower, but it may not test a large enough area to give values
representative cf the overall soil lir.er.
Borehole methods such as the Guelph ar.d Boutwell permeameters are in-
hcle methods for measuring steady-state infiltration into unsaturated soil.
The Guelph permeameter uses a constant head ar.d can account for the
pressure, gravity, and soil capillarity components of flow. The Boutwell
permeameter car. determine both vertical and horizontal hydraulic
conductivity of a soil liner. Both methods can measure hydraulic
conductivities as low as 1 x 10 ® m/s. Neither method, r.owever, tests a
large enough area of a soil liner to give reliable hydraulic conductivity
values that are representative of the overall soil liner.
Porous probe methods such as the EAT permeameter typically use a cor.e-
sr.aped porous probe to provide a hydraulic connection with the soil pores.
A pressure transducer is used to measure gas pressure in a test container
3
-------
that is connected to the porous probe. When water flows out of the
container and into the soil pores, the pressure change in the container is
recorded and used to calculate hydraulic conductivity. This method has
been usee to measure hydraulic conductivities less than 1 x 10~^ m/s on
scil liners. The method does not, however, test an area of the soil liner
large enough to give reliable hydraulic conductivity values representative
of the overall scil liner.
The ASTK double-ring infiltrometers use an outer ring to ensure the
inner ring treasures only the vertical hydraulic conductivity of a soil.
This method has not been shown to be capable of either measuring hydraulic
conductivity values of less than 1 x 10 ® m/s or obtaining values
representative of an overall soil liner.
The standard ASTM double-ring equipment has been modified to permit
the measurement of hydraulic conductivities as low as 1 x 10 * m/s. The
size of the inner ring, however, is too small to give reliable hydraulic
conductivity values that are representative of an overall soil liner.
Box ir.f i It rometers use ar. array of sealed boxes that can measure
hydraulic conductivities as lew as 5 x 10 1 m/s. The aggregate area
evaluated by an array of four inf iltrometers (I . 4 4 sr?) appeared to be
sufficient to obtain representative values for hydraulic conductivity on a
scil liner that had received a high level of quality assurance per unit
area. The array of inf iltrometers also was able to define the spatial
variability of the soil liner. The primary disadvantages of this method
are that it is difficult to use and that skilled personnel would definitely
be required for installation, monitoring, caie cc.lection, and data
analysis .
Sir.gle-ring inf iltrometers measure the infile rat ion of water ponded
inside a la roe-ciameter ring Dressed into £ soil liner. These
-S
inf i it rort.ete r s car. measure hydraulic conductivities less than 1 x 10 ,t./s.
If the rinc covers an area of at least 2 n', it should be capable of
obtaining values representative of the overall soil liner. Care should be
taker, not rely on correction factors to reduce the hydraulic
conductivity values below 1 x 10 3 n/s.
Sealed double-ring infiltrometers can use a relatively large inner
ring (2.3 rr.fc) to measure infiltration. So configured, this method has
received substantial field testing and has a demonstrated capability both
tc measure hydraulic conductivity values below 1 x 10 m/s and to obtain
values representative of the overall soil liner. The method requires more
installation time than many methods, but it nas the advantages of both few
ambiguities in the experimental method and few possibilities for yielding
false low values. Care should be taken not to scour the soil liner surface
when water is placed in the sealed inner rir.c, ar.d temperature fluctuations
should be kept to a minimum.
Collection ly si meters collect ar. o measure the volume of liquid
percolating through a defined area of a soil liner. These systems are
capable of measuring hydraulic conductivities of less than 1 x 1C~9 m/s and
-------
of obtaining values representative of an overall soil liner. These tests
can, however, be time-consuming, and they may require skilled personnel.
They are relatively expensive to install, and great care should be taken
not to damage the flexible membrane that is used to collect the
ur.cerdrainage. Ultimately, this method has the potential for yielding the
most accurate values for hydraulic conductivity of a soil liner. In
practice, however, the response time of a collection lysimeter limits its
utility for short-term tests.
Velocity permeair.ete rs use a snail computer to deternine in situ
hydraulic conductivity from a falling-head infiltrometer. The method has
yet to be tested on compacted soil liners, and further tests are needed to
determine if it is capable of measuring hydraulic conductivities of less
than 1 x 10 m/s. As currently designed, the method evaluates too small
an area to give hydraulic conductivity values representative of an overall
soi1 liner.
Porcus plate inf iltrorr.eters use water infiltration under tension to
estimate hydraulic conductivity. In currently available porous plate
infi1trcraeters, the area evaluated is too small to obtain hydraulic
conduct ivit ies representative of an overall soil liner. No data are yet
available or. the use of this method on soils with hydraulic conductivities
_ g
less than 1 x 1C m/s. Research with this method is ongoing, and it
appears to have promise for distinguishing between overall field hydraulic
conductivity and the component of flow occurring through nacropores.
Cores can be collected frorr. a soil lir.er and used in a variety of
laboratory hydraulic conductivity tests. These tests are usually capable
of measuring hydraulic conductivity values less than 1 x 1C 3 m/s.
Laboratory samples, however, are typically tcc small to yield hydraulic
conductivity values representative of the overall soil liner.
5
-------
SECTION 3
RECOMMENDATIONS
To maximize the probability of obtaining a hydraulic conductivity
value that is representative of the overall soil liner, a test method
should have the following qualities;
1) The ability to measure accurately hydraulic conductivities of less
char. 1 x 1G ^ m / s .
2: Minimal requirements for skilled personnel during installation,
operatic-, data acquisition, data reduction, and interpretation of
results.
3) Few a:-\higuit ies in the experimental rtethod and few possibilities
fcr yielding false low values.
4; Inexpensive enough to allow at least three replicate tests tc
determine spatial variability of field hydraulic conductivity.
5; Sufficient area of coverage for each replicate test to ensure that
a statistically sound average number of micropores per unit area
is covered.
6) Sufficiently short time required to conduct each test to ensure
that it is practical for use.
Not sll cf these qualities are quantifiable, given the current state
of knowledge on Hydraulic conductivity test methods and soil liners. It is
possible, however, tc pinpoint where additional study is r.eedec and tc
suggest the best currently available methods.
For a better definition of the test methods that would .Tiaximize the
probability cf obtaining representative hydraulic conductivity values fcr
soil liners, a long-term soil-liner study incorporating both a large
collection -ysi.neter !to give true overall nydraulic conductivity* and an
array of candidate hydraulic conductivity testing devices is recommended.
The results could be used tc define the reliability of testing devices and
to characterize their sensitivity to spatial variability within the soil
liner. Ideally, several sized versions cf each test method should be used
in each arrangement tc cbtair. a better understanding cf t.ne minimum size
requirement for the test devices.
6
-------
Based on the advantages and disadvantages of aii the methods reviewed
in this document, the best and r.ost practical currently available
technologies for evaluating field hydraulic conductivity are as follows:
2
1) Single-ring iniilt rometers covering an area greater than 2 m .
2) Sealed double-ring infiltrometer with an inner ring covering an
area greater than 2 m^ .
Until conclusive studies are completed, it is re comraended that
correction factors (such as those commonly used with single-ring devices)
not be relied upon to reduce hydraulic conductivity values below 1 x 10
rr./s. Correction factors for evaporative losses also should be avoided. If
the uncorrected hydraulic conductivity values are higher than the maximum
allowable value (1 x 1C ' m/s), confirmatory tests should be conducted with
sealed double-ring infiltrometers.
Spatial variability of the soil liner must be characterized to ensure
the hydraulic conductivity values obtained are representative. At least
three field hydra ulie conductivity tests should be conducted, and the
values obtained from all three should be I x 10 m/s or less. Averaging
cf values should not be allowed as a .-neans of bringing values below
-------
SECTION 4
FIELD HYDRAULIC CONDUCTIVITY METHODS
This section presents the field methods for measuring soil hydraulic
conductivity and evaluates the applicability of the methods for determining
hydraulic conductivity of compacted soil liners. The metr.ori3 discussed are
those currently used to determine field infiltration, hydraulic
conduct ivity, or underdrsinage rates for soils with low hydraulic
conductivity values.
A wide variety of field hydraulic conductivity methods have been
published in the scientific and engineering literature. Many of these
methods are not adequate for evaluating soil liners because they car.
neither measure very low hydraulic conductivities r.or evaluate a large
er.cuch area to give values representative of the overall liner. Other
methods are adequate to obtain representative values on low-hvdraulic-
ccr.duct ivity soils, but they are complex and time-consuming, or they
require considerable effort on the part of skilled equipment operators.
Seme ether methods appear to give accurate and representative values
without the need for extensive involvement of skilled operators. The
purpose cf this section is to provide sufficient detail on each field
hydraulic conductivity method to determine which methods car. provide a
reliable evaluation of soil liner performance.
£sch subsection or. a field nydraulic conductivity method begins with a
brief introduction and a discussion of the underlying principles and test
conditions. Details are then providec or. the specific test methodology,
including a list cf all required apparatus, a summary of procedures used in
the ret hoc, calculations used to determine hydraulic conductivity, and
applicable quality assurance measures. Example data that may be generated
by the method are then presented. Finally, the subsection includes
comments or. the acva.ntages, disadvantages, ana general applicability of tne
method for testing the hydraulic conductivity of soil liners.
AIR-ENTRY FERMEAMSTER3
Air-entry permeameteis are used for rapid field measurements of
vertical hydraulic conductivity in initially unsaturated soil (Bouwer
1973! . This method uses Darcy's fundamental law for water flow through
soil to determine soil hydraulic conductivity above a water table.
3
-------
Principles
This method involves introducing water into a sealed te3t ring and
measuring the infiltration rate from changes in volume of water in a
graduated supply reservoir over time. A sufficient period of time is
allowed to let the wetting front advance approximately 1 tc 15 cir. before
the water supply is turned off. The resulting maximum suction pressure
that develops within the apparatus is measured to give the air-entry
pressure .
The depth of the wettir.g front is measured by visual examination
(Bouwer 13 "7 3) or ten sio meters (Topp an d Bir.r.s 1976) . Hydraulic
conductivity calculated by this method is for the wetting front, and Bouwer
(1966) stated that the saturated hydraulic conductivity My be estimated as
double that measured for the wetting front fcr most soils. For a heavy
clay soil, however, the estimated saturated hydraulic conductivity should
be estimated as four times that of the wetting front (Eouwer 1S66) .
Method
Apparatus--
Ficure 1 is a diagram of an air-entry permeameter. The following is a
list of the equipment needed to set up an air-entry permea.-neter:
1: A cylinder 20 to 30 cm in diameter and mere thar. 10 c.-. long.
2) A disk to dissipate energy and the puddling effect of water
applicat ion.
3: A reservoir (graduated cylinder) fcr water application.
4! A vacuum gauge fcr measuring suction pressure.
5: Tcp plate assembly.
£) Ar. air escape valve.
Prccedure--
7he following is a summary of procedures to follow in the use of the
air-er.try perrr.earieter :
1) Crive the cylinder about 10 cm into the soil with minimum
distu rbance.
2) Place a layer of sand inside the cylinder.
3; Flace the disk on the sand.
4; Secure top plate assembly and fill the reservoir wich water.
5; Keep the vacuum gauge valve closed and the air escape valve open
to allow air to escape.
9
-------
RESERVOIR
VACUUM GAGE
ADJUSTABLE CLAMP
WATER SUPPLY VALVE
DISK
AIR ESCAPE VALVE
TOP PLATE ASSEMBLY
GASKET
SAND
WETTING FRONT
FIGURE I DIAGRAM OF THE EQUIPMENT FOR THE AIR-ENTRY
PERMEAMETER TECHNIQUE (BOUWER 1978).
10
-------
6;
Open the water supply valve.
7) Close air escape valve when ail air has been driver, cut of the
system.
6) Measure infiltration rate by recording change in volume of water
in the reservoir over time.
5; Close rhe water-supply valve to halt the rr.cverr.ent of the wetting
f ront .
10', Open the vacuum gauge valve to measure the pressure inside the
cylinder. Convert the reading into meters ana insert into
Equation 1.
11; Record the minimum pressure and immediately remove the equipment
to determine the depth of the wetting front.
12) Determine the depth cf wetting by the tensiometer method or by tr.e
digging and visual examination method.
La.cu.ations--
Equotior. 1 is usee to calculate the nydraulic conductivity a: tne
front:
wf x «
K = ( 1 !
Ax (H ¦
-------
number of deterninationa of composite hydraulic conductivity values than
Che 11 estimated in their work.
The accuracy cf measuring the depth of the wetting front (W^) can have
a significant effect or. the accuracy of the calculated hydraulic
conductivity. Recent studies showed that several compacted .materials with
hydraulic conductivities cf 1 x 10"^ m/s and lower had effective porosities
of 10 to 20 percent cf the total porosities (Horton et al. 1 988). If the
depth co the wetting front is measured in a snail area, such as with
tensiometers, then the value obtained may not accurately reflect the
overall average depth to the wetting front. Of the two methods for
determining Wf, visual examination of a large cross-section of soil, as
recommended by Eouwer (1978), would appear to be the more accurate means.
Example Data
Knig.it ar.d Haiie (196<: used air-entry permeameters to estimate the
field ar.d laboratory hydraulic conductivity or. a scil lir.er in r.ortr.err.
Saskatchewan, Canada. The tests were performed as pare of quality control
or. a compacted mixture of natural till and ber.tor.ite. Of the 133 tests
carried cut in the field, 35 gave hydraulic conductiviey values that were
— Q
nig.ner than the design objective of 1 x 10 m/s. The authors attributed
the nigh values to leaks around the perimeter of the test ring.
Halle and Brcwr. (1988) used a 0. 5-m-diameter air-entry permeameter
-------
TABLE 1. HYDRAULIC CONDUCTIVITY OF COMPACTED CLAY SOILS AT TWO HEAP LEACH
PADS IN CALIFORNIA3
Hvdraul
ic Conductivity,
n/9
Site
Method Test
Fill 1
Test
Fill 2
Test Fill 3
Jamestown
In situ AEP
Mine
(10 tests/fill) 1. 4
x 10"9
6.2
X
10~9
1.0 x 10"9
SDRI 1.2
x 1C~9
3 . 7
X
10~9
1.2 x 13"S
Lab AEP
(10 tests/fill) 2.6
x lo"S
1 . 9
X
10~9
6.1 x :o"-
Full-scale liner (13 ha)
In situ AEP (13 tests)
2 . 6
X
10"9
Lafc AEP (15 tests)
2 . 5
X
10"9
Ca rsor.
In situ AEP (9 tests)
1 .0
X
10"10
Hill
SCSI
9.0
X
10"10
S Source:
.Haile ar.d Ercwn 1388
TABLE 2.
FIELD HYDRA'JLIC CONDUCT iv::
:Y DATA
03TAINED BY US:
:ng air-snt
PER.yEAV.ETER AND LABORATORY
PERMEAB
I LIT Y
TEST WITH
SHELBY T'J
Type
jf t e s
Number
Hydraulic Conductivity. m/s
.ner
of tests
Range
Average
-1
£ y. 10"11 to 6 X
I0"10
-1 ~
3 x 10 i"
2
45
1 y. 10 11 to 1 x
10-5
3 x 10"10
3
2 C
IxlO^toGx
o
1
o
3 x 10~10
r. ir-er.try
perseameter
_cCC
retcry
£
2
X
10 "
:o
tc 4
X
y ....J
o
1
o
pe
i T
r T.e siTv-E :e:
ixec-wail)
-------
BOREHOLE METHODS
Several borehole methods have been used to measure the hydraulic
conductivity cf soils. Bcrehcle permeametera are essentially in-hole
constant-head or fallir.g-head permeameters. The methods involve measuring
the steady-state infiltration rate of water into unsaturated soil frem a
cylindrical borehole. Two borehole methods that have been used to test the
hydraulic conductivity of soil liners are Guelph (Reynolds ar.d Elrick 1 986)
and Boutwell permeameters (Boutwell and Derric 1 98 6) . The Guelph
permeameter uses a Mariotte siphon to maintain a constant head of water
within the bcrehcle. The Boutwell permeameter, however, involves an
evaluation cf the flew into a cased borehole ana an evaluation of
subsequent flow after the addition of an uncased extension to the hole.
Details of these two methods are discussed in the following subsections.
Guelsh Perrr.eareter
Frir.ciples--
The Guelph permeameter uses a Marictte siphen over a 0.02- to 0.05-m
diameter well to measure the steady-state infiltra.xcri race of water into
unsaturated soil. The in-hcle Mariotte bottle device is usea tc establish
and maintain a constant head of C.C5 to C.35 m within the borehole and to
measure the corresponding flew rate. Two or rr.ore data pairs of the
constant need and flow rate a re recorded for each hole and suostituted into
equations to obtain field-saturated hydraulic conductivity, sorptivity, ar.d
the relationship oetween the hydraulic conductivity and the pressure head.
Method--
Aor.aratu.". --A schematic cf the Guelph permeameter is shown in Figure 2.
The main cor-ipcr.ent: s are as follows:
1) Air-inlet tuse.
2) Reservoir cap.
3) Calibrated inner reservoir tube.
Calibrated outer reservoir tube.
5) Cuter reservoir cutlet pert.
6) Permeameter tip with rubber stopper on the end.
7) Sliding air-tight seals.
8; Tripod assembly.
?rccecure--The procedure used is as follows:
1) Using a soil auger, excavate a borehcle of radius "a" to the
desired depth in the soil to be tested.
14
-------
CALIBRATED NNER
RESERVOIR TUBE-
AIR-INLET Tuee
RELEASE VALVE
SLIDING AIR-TIGHT SEALS
LIQUID SURFACE IN RESERVOIR
PRESSURE TRANSDUCER (OPTIONAL)
CALI0RATEO OUTER
RESERVOIR TUBE
THREADED COUPLING
OUTLET TUBE
TRIPOD ASSEMBLY
STEADY LIQUID
LEVEL IN WELL
OUTLET PORT
— PERMEAMETER TIP
RUBBER STOPPER
(drawing not to scale)
FIGURE 2. SCHEMATIC OF THE GUEL PH PERMEAMETER
(REYNOLDS AND ELRICK 1986).
15
-------
2) Assemble the perinea meter .
3) Fill inner and cuter reservoir tubes with liquid.
<1) Place permeameter in borehole.
5) Start permeameter by raising the air-inlet tube ouc. of the cutlet
pert .
6) Set the desired head of water (H) by adjusting the height of the
air-inlet tube.
7) Monitor the rate of fall of the liquid surface in the outer
reservoir tube (R) until a steady rate is attained.
3) Calculate the steady-state volumetric flow rate (Q) by multiplying
the rate of fall of the liquid surface (R) by the annular cross-
sectior.al area between the air-inlet tube and the reservoir tube.
9) Calculate K/a ar.d use the value to select the proportionality
constant (C) from Figure 3.
10) Calculate the field saturated hydraulic conductivity by using
either cr.e simultaneous equation approach oi the least-squates
ac:reach.
The simultaneous equation approach (Richards analysis; is used to
calculate the hydraulic conductivity by solving Equation 2. This involves
repeating steps 7 through 10 at least once because the simultaneous
equations require at least twe values of K and Q. The least-squares
approach uses two or more H-.evels ar.c the corresponding Q values (Reynolds
and Elric.< 1 53 5a) to solve Equation 3.
Ca leu lat ic.-.s - -Revr.olcs ar.c Eirick (1965o) presented an equation to
describe the steady-state discharge from a cylindrical well into
unsaturates sell. The equation incorporates the parameters for pressure,
gravity, and scil capillarity.
2ffSfs * CrfKfs * 2rH.^ = C Q (2)
r t t
PRESSURE GRAVITY CAPILLARITY
where H = steady depth of liquid in borehole, n
K^s = field saturated hydraulic conductivity, n/s
C = dirr.er.sior.less proportionality constant dependent on H/a
a = borehole radius, n
2
a = matric flux potential, m /s
n 3
Q = steady-state volumetric flow rate of liquid, m /s
Reynolds ar.d Elrick (1586) and Baurngartner et al. (1 537) provided a
simultaneous equation approach for solving Equation 2. Tne approach
involves measuring two or more steady volumetric flow rates (eg., 0- anc
16
-------
3.0
2.0
O
X
1
1
l
i
0
H/g
FIGURE 3. C VERSUS H/a RELATIONSHIP FOR A RANGE OF
SOIL TEXTURES (REYNOLDS AND ELRICK !985b).
17
-------
Q2)) and their corresponding steady depth of liquid in the borehole (e.g.,
and Hj!. After H1# Qj and Q2 have been substituted into Equation 2,
is eliminated from the two equations formed. The resulting equation is
then simplified to the following form:
Kfs " G2 °2 " G1 Q1 (3)
where
H2C1
tt(2K1H2(H2 - Hx) + a2(H1C2 -HjC^]
H1C2
^ ir [ 2H jH2 {H2 ~ Hx ) r a2(H;C2 -HjC^]
The appropriate C values for H. and H2 are C, and C2, respectively.
The value cf the diner, sion less proportionality :or, stant (C) is
obtained frcr. Figure 3, which gives C curves for three classes of soils.
Tr.ese curves apply to soils at field capacity or aner. Kcst loams and
structured clay soils are represented by the Guelph loan soil. Tr.e metric
flux potential is defined by Gardner (1956) as:
:
l-i
where K, , = hydraulic conductivity-pressure head relationship, r./s
. " = initial (assumed uniform) pressure head in tne soil, m
= matric flux potential, m /s
Sir.pler ar.d less labor-intensive spproxirr.at icr.s of hydraulic
conduct ivicy can be made by using the analysis of Laplace with only cr.e ri-
le vei determination. The following is tne form of the Laplace equation
used to determine hydraulic conductivity (Kf.) :
C Q
K*<, = ^ n l5)
(2 7T H + C r a )
wnere H - steady depth of liquid in borehole, rr.
K - - field saturated hydraulic conductivity, rr. / s
C = dimensionless proportionality constant dependent on H/=
a ¦= borehole radius, m
0 " steady-state volumetric flow rate of liquid, rr?/s
Quality Assurance--The Guelph method gives "point" measurements.
Several readings are needed tc ensure that the data obtained frcm a given
location accurately represent the hydraulic conductivity of that area.
18
-------
A heterogeneous porous medium can give erroneous hydraulic conduc-
tivity values. The presence of a large macropore will give a negative
hydraulic conductivity value when tests are done at two constant water
levels in the borehole (H-levels) and the resulting data are used in the
simultaneous equation approach to calculate the hydraulic conductivity.
Smearing of the soil on the surfaces of the borehole will give low
values of K-s that are not representative of the overall soil. Reynolds
and Elricic ( 1986) found a wire brush was useful for removing the smeared
soil.
Example Data--
Bradshaw ( 1 9 96) used a Guelph permeameter to measure hydraulic
conductivity in a silty clay soil at an initial water content of 3?
percent. Fifty-eight measurements of flow rates in 20 test holes indicated
the geometric near, of the field-saturated hydraulic conductivity to be
7.3 x 10 ^ .t./s with a multiplier/divider factor of 1.65 for the standard
deviation (usir.g Laplace analysis) . Fifty percent cf these test holes
provided usable data for the simultaneous equation analysis (Baumga rtr.e r et
al. 1987). The geometric mean for the simultaneous equation analysis was
2.2 x 10"* r./'s, witn a standard deviation factor of 2.1.
Table 3 presents a comparison cf the Guelph permeameter method (using
Laplace analysis) with the air-entry permeameter method and a soil core
method cn a range of soil types (Reynolds and Elrick 1965b! . The data
indicated hydraulic conductivity determined by the air-entry and Guelpn
metr.ods were one order of magnitude higher than the data from the soil core
method for fine sand to clay soils.
Stephens et al. (153S) presented data that compared the Guelph perme-
amet&r with experimental borehole poncir.g data and air-entry permeameter
data. Comparison cf these data {Table 4) indicated a general agreement cf
the hydraulic conductivity values determined by these methods. The compar-
ison was m.ade on soils with much higher hydraulic conductivities than clay.
Further studies are needed to determine the extent to whicn values will
agree cn lew-hydraulic-conductivity materials such as compacted 3cii
liners .
C c r.T.e r. t s - -
Guelph pentameters have not been widely used to determine hydraulic
conductivity on compacted soil liners. Studies conducted on low hydraulic
conductivity soils indicated, however, that the instrument is capable of
reasurina values as low as 2 x 10 9 m/s. For provision of reliable data at
- 9
the regulatory limit of 1 x 1C m/s, a method shculd be acie to provide
credible data below this limit.
Another shortcoming of the Guelph permeameter is that -t only evalu-
ates a small area of soil ! 3 x 10 ^ to 2 x 10 smet . For tr:_s reason, the
method is unlikely to give values that are representative zi the overall,
field hydraulic conductivity. There are currently r.c acta available to
assess the number of tests that would be .necessary to obtain representative
va1ues .
19
-------
TABLE 3. COMPARISON OF THREE METHODS OF DETERMINING SATURATED HYDRAULIC
CONDUCTIVITY--GUELPH PERMEAMETER, AIR-ENTRY PERMEAMETER, AND SOIL
CORE3
Saturated hydraulic
conduct ivity.
Soil Type Method*3 10 ^ m/s
Fox loamy AEP 33."0
sand GP 51.6C
SC 20.£0
Vine land AEP 4.0 6
vf s loa.T. G? 1. S 9
SC 0 . 74
Woilwich AEP 2.'Z
silty loam GF 1.E 6
SC 0.6 2
Trafalgar AEP C.2 C2
clay GP L'.1C6
SC C.02 5
Source: Reynolds and Elrick ISfiSfc
AEP = Air-entry perxea.iete:
GF = Guelph perrr.ea.T.eter
SC = Soil core
COMPARISON OF SATURATED HYDRAULIC CGNC'JCT IVITY CAT A FROM G'JELFH.,
BCREHCLE , AND AIR-ENTRY P E £ AM E T E ?, S 5
Method Saturated hydraulic ccr.ducc i vie v, rri/s
Guelph
Borehole
Air-entry
Source: Stephens ez al. 1933
. - X
7.7 x 10
2.3 x 10"8
3.5 x 10-3
1.1 >: 1 0 "3
- - >: 1C~9
tc 1 . A
-10
-------
Auger equipment tends to smear the walls of a borehole excavated in
moist-to-wet clay soils. Smearing partially seals the borehole surfaces,
which results in flow rates that are too low to be representative of the
soil. Thus, removal of a smear-layer is critical to obtaining quality data
with the Guelph permeameter method. Reynolds and Elrick (1966) found the
efficiency of smear-layer removal declined with increasing wetness of clay
soils. Soil liners in waste-management facilities are typically compacted
at moisture contents that would cause smearing.
Determination of hydraulic conductivity for compacted soils by the
Guelph permeameter method is difficult because the instrument is not suffi-
ciently sensitive to measure the small volumes of water involved. for
_ 5
example, hydraulic conductivity near the regulatory maximum of I x 10
it./s, measured changes in water volume would be approximately 0.2 cm^/n at
the H values typically used with the device. In addition, at sucn low flow-
rates, the volume of water evaporated from the instrument could be a major
source of error.
Boct wel1 Method
Princicles--
The Ecutwell method is a two-stage borehole hydraulic conductivity
test capable of determining both vertical and horizontal nydraulic
conductivity cf a soil. In stage one, hydraulic conductivity is determined
by casing a borehole and measuring flow through the bottom cf the borehole
(K,;. In stage two, the length of the borehole is extended but the casing
is not. This allows flow through botn the bottom and the uncased sidewalls
cf the borehole (Kj) to be measured. These two hydraulic conductivity
values can then be used to calculate botn horizontal and vertical hydraulic
conductivity, as shewn in calculations presented later in this subsection.
Vg» rice —
f-.cca ratus--~icure 4 is a schematic of the Boutweii ocrer.ole method.
The following apparatuses are needed for this method:
1) An auger or otr.er device suitable for excavating a oorenole at
least seven times deeper than the diameter of the hole.
2) Borehole casing.
3) Graduated standpipe.
Procedures — Procedures for the Boutweii method of determining
hydraulic conductivity are summarized here.
1) Auger to a depth five times the diameter of the borehole.
2) Place a casing inside the borehole and seal the space between the
casing and the borehole with grout.
3) fill the casing and standpipe with water and cover the open er.d of
the standpipe to prevent evaporation.
21
-------
STAGE I STAGE H
no
rvj
H
.sz.
f—d
I ¦<>" I
D
CLAY LINER
- CASING
GROUT
>5D
>5D
_2_
I
D
CLAY LINER
> 5D
> 5D
FIGURE4. SCHEMATIC DIAGRAM OF THE BOUTWELL BOREHOLE PERMEAMETER
(BOUTWELL B DERRIC 1986).
-------
4! Measure a aeries of flow rates and calculate (Equation 6); plot
the values cf K. as a function of time until steady conditions are
reached.
5) Remove the top cf the permeameter ar.d deepen the hole with an
auger cr a suitable thin-walled sampling tube; the ratio of L/D in
the uncased zone should be between 1 and 1.5.
6) Reassemble the perrr.eameter ana perform the falling-head test
again.
7) Measure a series of flow rates and calculate K2 (Equation 7); plot
the values of 5<2 as a function of time until steady conditions are
reached.
5) Using arbitrary values cf n, where rr. is defined as (K^/Kv)^'^,
calculate corresponding values of K2/K^ (Equation 8;,- plot the m.
ar.c X2/X. values.
9) Deterr.ir.e actual Kj/K^ (Equations 6 and "I .
1C) -sir.g the Kj/K. value determined in Step 5 above, find the
corresponding value of rr. f rem the graph cf K 2 / K - versus m
generated in Step 8.
Determine K,_ (Equation 9).
12) Determine Ky (Equation 1C).
C£lculsticns--Eouaticns for Stage I determination are as follows
; r. i e 1 1 & b 7 : :
d^
K: = lr. (H, / H 2) (6)
11 D (t2 " -:)
whers ci = diameter cf stanapipe, n
t-= initial tine, s
t2 = final tine, a
Hn= initial water level in standpipe, m
final water level in standoipe, m
L • r
D = diameter of borehole, m
K1= hydraulic conductivity value for Stage I test, m/s
Equations for Stage II K2 determination are as follows (Daniel 1987):
K2 - A B lr. (Hx/H2) (7)
d2 In[L/D + (1 + (L/D)^)0'5]
8d :l/d; (t2 - t.)
23
-------
and
B - 1/[1 - 0.5G2e _1•57 (L/D)]
where t, = initial time, 3
t 2 = final time, s
H. = initial water level in star.apipe, m
Hj = final water level ir. standpipe, m
D - cianeter of borehole, n
K. - hydraulic conductivity value for Stage I teat, m/s
K2 " hydraulic conductivity value for Stage II test, m/s
1 *» length cf extended hcle, m
Equations for determining -^/K^ ana m are as follows (Daniel 1 987):
in [L/D -r (1 r (L/D) 2 ) C ' 5]
K2/K. = n — (8)
In irr.L/D - (It (nL/D) 2 ) u ' 2 •
where 1 ¦= length cf extended hole, n
D = cianeter cf hole, n
K^ and *2 are given in Equations 1 and 8
r. = constant
Equations for determining horizontal and vertical hydraulic
induc-ivizy ;K^> and (KvJ are as follows:
X„ - mX. (9)
X - X. /r. ; 10 )
V 1
where Kj_ = horizontal hydraulic conductivity, ra/s
K " vertica 1 hydraulic cor.cuc:ivity, m/s
n « constant derived from Ecuation 9
Quality Assursnce--A homogeneous scil is assumed to be permeated in
Stages I and II of the Boucwell method. Tests performed in heterogeneous
scil rr.ay give erroneous hydraulic conductivity values. A Boutwell perme-
arr.eter test should, therefore, be performed on sections of the liner that
are far removed from the boundaries. Tests conducted close to the eages of
the liner can be affected by the boundary conditions and may, therefore,
produce values that are r.ot representative of the overall lir.er.
It is also necessary to assure saturated soil conditions around the
vicinity of the borehcie. This saturated condition may increase smearing
of the soil surface in the borehole, especially during the Stage Ii test.
Maxir.ur. effort should be made to minimize smearing of the soil surface
during the Stage II tests.
Example Data--
Boutweil and Donald (1592) used a Boutwell permeameter to determine
the field hydraulic conductivity of a natural clay and compared the values
24
-------
with those determined in the laboratory on Shelby tube samples in a
ccnatant-head perneameter. The average laboratory hydraulic conductivity
value was 7.2 x 10 ^ m/s. Values from the field tests were found to be
six tines higher than the laboratory values. The authors explained that
the difference was due to the jointed structure of this clay soil.
Boutwell and Derrick (1986) determined the field hydraulic conduc-
tivity cf soils at two sanitary landfills. The horizontal conductivities
of the stiff clay (CHJ at the two sites were 1 x 10 and ^3 x 10 ^
whereas the respective vertical conductivities were 6 * 13 and 6 x II *
rr./ s .
Csmer.ts--
Bcutwell permeameters can measure hydraulic conductivities of i * ;i '
r./s and less. The volume of soil tested in a single borehole, however, is
relatively small. Because of the small size of the test area, soil
macrcpores and other flaws ir. soil liner const rac: ior. may be missed when
this r.ethod is used. As a result, hydraulic conductivity determined by
this method may not be representative cf the actual field value.
Multiple tests could be conducted to evaluate a larger aggregate area.
For several reasons, however, running many small tests may not achieve the
desired test cojective cf obtaining a representative measure of the field
hycraulic conductivity. For example, even if rr.ar.y tests were conducted,
the snail s c a - e of each individual test would greatly increase the
probability tnat throuch-going nacropores (Anderscr. et al. 19 S 5 j would be
truncated. Such truncation of through-going macrcpores would significantly
reduce the hydraulic conductivity from the actual field value. Other
prcclerr.s include the large amount of potential smearing per unit volume of
soil sr.c the difficulty that may be encountered in distinguishing between
defective test results ana results reflecting the value of a small area of
the soil liner with a ncntruncatec macropore.
The Soucwell method is a relatively fast means of determining soil
hydraulic conductivity. It is also inexpensive, si::.ple, ar.c convenient to
use; however, interpretation cf results is open tc seme question. The
effects cf incomplete and variable saturation are unknown. The short test
periods cc net allow entrapped air to dissolve, and tne :r.et.'iod dees r.ot
account for the effects cf soil suction.
POROUS PROBES
These permeameters typically consist of a cone-shaped porous probe
that is either pushed or driven into the soil :Dar.iel 1987) . These probes
can be used in a falling-head or a const ant-nead mode to determine
hydraulic conductivity. One such commercially available porous prcte that
has been used to measure the hydraulic conductivity cf soil liners is
called the SAT permeameter. This device uses a falling-head method.
25
-------
Principles
The BAT perx.eameter uses a falling-head method to determine hydraulic
conductivity according to Darcy's law. A pressure transducer is used to
measure the gas pressure in an evacuated or partially filled and
pressurized test container. The container is temporarily connected to a
porous proce that is installed in the soil. When fluid flows out of the
container into the soil pores, a pressure change occurs and is recorded.
The rate of pressure change is a direct function of the hydraulic
conductivity of soil.
Methods
Apparatus-
Figure 5 depicts the major features of the 3AT pe rmeamete r. These
never features are as follows:
1) A porous probe to provide a hydraulic connection with soil pcres.
2) A flexible septum that seals the upper end cf the porcus probe.
3) Test adapter.
4) Hypcderrr.ic needle.
51 Extension pipe.
:cecures--
Tr.e following is a listing of the procedu res required in the use oi a
' per.-r.ear.eter ;
1/ Auger a hole 0.03 nr. (1 inch) in diaretfer tc the desired depth,
2: Push the porous prcbe under static lead into the augured hole.
3) Connect test adapters to the septum. (The hypodermic needle will
pierce the septum. as soon as this connection is made; the liquid
in the container will then flew to the porous proDe.J
<; Record the initial pressure from the pressure transducer and the
initial volume cf water in the test container.
5) Record the average pressure between times t, and t2-
6) Measure ana record the net pressure applied during tne test.
7; Lift adapter off the tip to stop flow and to er.c test. The test
container will instantly reseal tc maintain the pressure in tr.e
test container.
26
-------
EXTENSION P|P£
PRESSURE TRANSDUCER
TEST AOAPTER
TEST CONTAINER
SEPTUM
NEEDLE
SEPTUM
POROUS - PROSE
FIGURE 5. BAT PERMEABILITY SYSTEM
(PETSONK 1985).
27
-------
Calculaticr.a--
Fetacnk (1965) discussed some of the calculations useci to determine
hydraulic conductivity with the BAT permeameter. The equations moat often
used with this method are as follows:
Po Vo 1 A Pt
k = t:
F «Pt) P3 A t
where K = hydraulic conductivity, n/s
PQ = initial gaa pressure ir. test container, m of water
Vc = initial volume of gas in test container, m
=* average measured pressure between times t-1 ar.d t, m of
water
F -j - r.et pressure applied in test container, m of water
AF,./£t ¦ rate of pressure change, m of water/'s
A" = cross-sectional area of flew, m2
L » horizontal component of hydraulic conductivity gradient
? - flow f3ctor, m
2 * y
.n [ (y /d) - (1 -
-------
Average hydraulic conductivity values from field tests on a corr.pacted
soil liner in a teat fill were presented by Fetsonk et al. (1967) The
arithmetic mean conductivity values were B.O x 10 ^ (: 7.5! x 10 ^ m/s
for BAT and 6.5 x IC-1'" m/s for SDR! (sealed double-ring infiltrcmeter) .
Figure 6 shows a plot of the hydraulic conductivity and pressure readings
obtained with the BAT permearr.eter.
TABLE 5. FIELD AND LABORATORY MEASUREMENTS OF SATURATED HYDRAULIC
CONDUCTIVITY IK A CALIFORNIA CLAY
Depth, m BAT perxeameter, m/s Triaxial permeameter, in/s
o.ie 3.7 x io 4.5 x ic";;
0.45 4.4 X 1Q~10 6.3 X 10"11
° Source: Petsonk 1385
Ksier. and Yamamoto (1987 ) determined field and laboratory hydraulic
conductivity cn a test fill liner 36 m x 24 m x C.76 m. The data indicated
the SDS.I ana BAT permearr.eter results were as mucn as two orders of magni-
tude higher than the laboratory determined values. laboratory :ests ^ere
performed or. relatively undisturbed samples collected m Shelby tubes and
zested in a triaxial perxearece: . Hydraulic conductivity values deter-
mined by ;r.e 3AT perneaneter ranged from 6.2 x 10 to IS * 10 .t./
wnereas the SDRI indicated the hydraulic conductivity was at 2.7 x :C 5
rr / s . The la be ra to rv-det e rnsi ned hydraulic conductivities ranaec from
-11 -ri
1 x 10 *' to 2 x 10 rr./s at a confining stress of 172 k?a. Whereas tne
triaxial permeameter yielded consistently lew values, the BAT permeameter
yielded values ranging from, lower than the triaxial pernteaneter to nicher
than the SDRI.
Because the volume of soil tested by the BAT perneaneter is small, the
method is unlikely to yield values that are consistently representative of
the overall hydraulic conductivity of the soil being tested. Tne BAT
permeaneter can measure hydraulic conductivities of i x 10 ' m/s cr less;
however, it primarily measures horizontal hydraulic conductivity. This
nax.es interpretation of results for a soil liner at a RCRA-regulated
facility difficult.
Tr.is device provides a simple and inexpensive method of measuring
field hydraulic conductivity that requires Minimum time and experience.
Because it is a closed system;, errors due tc evaporation losses are
eliminated.
29
-------
5.60
0
(/)
UJ
a:
a.
4.00 -
3.20
2.40
O =PRESSURE
A 'PERMEABILITY
160 200 240 280
ELAPSED TIME, MINUTES
400
FIGURE 6. FIELD HYDRAULIC CONDUCTIVITY OF A SOIL LINER OBTAINED USING A BAT
PERMEAMETER (PETSONK et al. 1987).
-------
SURFACE INFILTRATION DEVICES
Included in this section are all field hydraulic conductivity methods
that measure the rate water enters a soil liner. These methods include
single- and double-ring infiltrometers of various designs. A method for
converting from infiltration rate to hydraulic conductivity is given in the
section on the modified double-ring infiltrometer. With all surface
infiltration nethods, care should be taken to protect the soil surface from
des iccat ion.
ASTM Double-Ring Infiltro.xeter
In ASTM Method D 3385, double-ring cylinders are used to determine the
rate of infiltration of water into soils. For more than a decade, this
method has been used widely for the evaluation of both percolation rate in
septic fields and infiltration rate in soils to be irrigated. Two open
cylinders, one inside the other, are driven into the soil and partially
filled with water. A constant water level is maintained on the soil by
continuously adding water to the soil in the rings. The volume of water
added in a given period is used to aetermine the rate of infiltration.
Prmciple--
The reasoning behind the double-ring method was to let the outer
annular space between the two rings serves as a buffer area in which the
lateral flow would be diminished. This would allow tne measurement of only
vertical infiltration from the inner ring. The test can be used at tne
soil surface or at a giver, depth in pits, as long as the test surface is
above the water table.
y.et'-.od--
This method is documented in the 1S6S Annua- Boo* at ASTM Standards,
Section 4, Volume 4.06, and tne reader is referred to ASTM Standard No. 3
3355-66 for details of the procedures and calculations. (The ASTM standards
are updated annually, and the most recent edition should be obtained.)
r.pca ratu3--Fioure 7 is an illust ration of an ASTM double-ring
ir.fi_tror.eter. The following list summarizes the equipment needed fo: the
ASTM method to determine the rate of infiltration of water into a soil:
Two cylinders 30 and 60 cm in diameter and 50 cm in height,
constructed of 1/8-inch steel beveled at the bottom.
2) One 200-liter (55-gallon) barrel for water supply to outer ring.
3; One lCCC-ml graduated Mariotte tube for measurement of the water
flow to the inner ring during the test.
<: Wooden block ana sledge hammer.
5; Aluminum disxs 1.3 cm thick with diameters slightly larger thar.
the infiltrometer rings.
31
-------
TO WATER SUPPLY
SHUTOFF CLAMP
POINT GAGE
MARIOTTE TUBE
HOOK GAGE
WATER LEVEL
VALVE
WATER
LEVELS
WATER
SUPPORT CLAMP
STEEL ROD
SOIL
infiltration ring
THREADED HOSE CONNECTOR
NOTE : Outer ring has b«en eliminated for simplification of the tllustroiior*.
FIGURE 7. ASTM DOUBLE-RING INFILTROMETER
(MODIFIED FROM ASTM 1988).
32
-------
6) Hock gauge, steel tape, or rule for use in controlling ar.d
measuring depth of water in the infiltrometer ring.
Procedure--Details of the procedure are in the 1 988 Annual Book of
ASTM Standards, Section 4, Volume 4.C0 (Standard No. D 3385).
Calculationa--Calc illations involve converting the volume of liquid
used during each time interval into an incremental infiltration rate for
both the ir.r.er and outer ring by use of Equations 13 and 14:
Inner Ring:
*IR VIR// !AIR x (13)
where "~R = infiltration rate of soil ir. inner ring, crn/h
V„_ » volume of liquid used during tirr.e interval to maintain
constant head in ir.r.er ring, cm^
Air = inner ring soil surface area, crr^
r.e interval, h
L'.:et :
ICR V0R''(A0R X
= Infiltration rate of soil in outer ring, c:n/n
V„_, •= Volume cf liquid used curing tirr.e lnctrva. ;o maintain
constant heac ir. outer ring, cm
o
Ano = Outer ring soil surface area, cr'
t = time interval, h
Quality Assurance--Drivinq in the rings with a 5ledgehanner can
fracture a test soil that is dry or stiff and lead to false high values.
In addition, the test rings should be protected frorri the environment,
inir.als, ar.d vandalism. Intercepted rainfall or high evaporative losses
may also lead to false values.
Example Data--
Bagcni ; 19B6) presented data for comparison of laboratory
properties of blended soil in a landfill ir. Wisconsin,. Average
anc double-ring hydraulic conductivities cf S.O x !C""V and 2.0
were reported.
Ccrr.mer.ts--
This equipment should not be used to measure infiltration rates on
compacted soil liners. According to ASTK (1938!, this method may be botr;
difficult to use ar.d unreliable in soiii with a hydraulic conductivity less
— 6 *
than about I x 10 n/s. Also, less of water due to evaporation may be
higher than the quantity of water permeating the soil.
The surface area of the inner ring of the ASTX double-ring
infiltrometer is larger than most laboratory per.mea.meters, out smaller than
several types of currently available field infiltrometers. No definitive
and field
laboratory
x 10"5 rr./s
33
-------
data base exists that conclusively indicates the minimum infiltrometer size
necessary for consistently yielding infiltration rates equivalent to that
of the overall soil liner. A substantial body of inforir.ation suggests that
the larger mf ilt rometers (>2 ir) do give a reliable indication of field
performance (Elsbury et al. 1566) . Consequently, equipment such as the
ASTM double-ring infilt rorr.eter, which evaluates relatively small areas,
should not be used tc test materials, such as soil liners, that may have
widely spaced macropores. Although these wisely spaced nacropores may
represent only a snail fraction cf the total porosity cf the soil lir.ei,
they also may serve as the conduits for most of the flow through the liner.
Rings used in this method can be driver, into the soil with a
sledgehammer, a heavy bulldozer blade, or other equipment. Caution should
be exercised when driving the rings into especially stiff or dry soil
because the soil may fracture, which can result m artificially high
hydraulic conductivities.
As with many field infi1trometers, hydraulic conductivity cannot be
directly determined with this device unless ail boundary conditions are
known (e.g., the hydraulic gradient and the extent of lateral flow! .
Modified ^ourile-P.ino Infiltrcmeters
Med i f i ca t i on s have beer, made to
ir.f iltrometer to make it mere sensitive
rates. T.nese modifications include the
sensitive water level measuring devices.
Principles —
The principles behind this metned are similar tc these presented
earlier fcr the ASTM double-ring inf ilt rcir.eters. The primary difference is
t.-.at t.-.e outer ring has been enlarged for further nir.iini2ar.icn of the
pctentiai fcr a lateral flow ccmpcnent in the infiltration rate measured
*itn the inner ring. Other differences from the ASTM couole-ri.ng metr.cd
are designed tc increase the ability of the equipment to measure
infiltration rates less than 1 x 10 13 m/'s.
Met nod--
The method is similar to that for the ASTM couble-ring infiltrometer
except for its use of a larger outer ring, sensitive water level measuring
devices, and the installation of ter.sioneters in the ring to measure soil
moisture tension at various soil depths. The sensitivity cf the system may
be increased further by preventing evaporative loss of water (i.e.,
covering the infiltrometers with suitable material that acts as a ir.oisturs
barrier;.
Acpa ratus--Figure 8 shows a modified couble-ring
uses larger rings, float valves to maintain a constant
rings, and level recorders. The main components of
follows:
3
the basic ASTM doubie-ring
for measuring low infiltration
use of a large outer ring and
inf i 11 roir.eter that
water level in the
the system are as
-------
RECORDER
PULSE GENERATOR
OUTER RING RESERVOIR
NNER RING RESERVOIR
FLOAT VALVES
FIGURE 8. MODIFIED DOUBLE-RING 1NFILTROMETER WITH A LEVEL RECORDER
(BROWN 1985).
-------
1) Two cylinders 30 and *7 6 cm in diameter and 30 cm high, constructed
cf 1/6-inch steel.
2) Supply tube to inner ring made of 15-cir.-diaxeter steel tubing.
3; A SS-gallon barrel to supply water to the outer rir.c
4; Adjustable float valves.
5) Level recorder.
6! Wooden block, sledgehammer, and carpenters level.
Figure 3 shows ar. ASTf. double-ring infiltroxeter setup moaified to
include trie use cf tensicmeters . Procedures fcr this device are similar to
those given ir. Section 4, except that a constant water level is no:
required. Changes in water level in the rings are penod.cally measured
wit:, a hook ca.ce.
Procedure--The following is a listing of the procedures used in this
xcdifiea double-rir.g ir.f i it r ometer :
1) Drive steel rings (inner and cuterl into soil as in the ASTM
ccusie-ring infi1troxeter xethod.
2) Ccr.n=:: float va 1 ve s to the ir.nex and cuter ring: adjust the level
of tr.e floats tc maintain a water level of IC tc 15 c.-i ir. tr.e
rings.
3) "Jse 6 pail to fill both rings with liquid to the same desired
depth, Taking sure tc- prevent scouring cf the sc.il surface by
using a splash ajard.
¦5) Connect _r.r.ec and outer rings tc their respective water supply
t UOg; .
5, Cover the rings with a sheet cf plastic or suitable xatenal to
prevent evaporation frorr. the liquic surface.
ci Cover t.-.e cper. end of the supply tube to prevent evaporation of
water f rex the tube.
7) Periodically record the liquid level in the rings with a level
recorder mounted on top of the supply tube.
Calculations—The voluxe of water used during each rr.easurea time
interval is converted into ar. incremental infiltration rate by the
fcllcwing equations:
I =
{d2 x h) / : T)
(15)
-------
INNER RING
TENSIOMETERS
OUTER RING
GROUT
FIGURE 9. ASTM DOUBLE-RING INFILTROMETER
WITH TENSIOMETER.
-------
where I = infiltration rate, m/s
d = diameter of liquid supply tube, m
D = diarr.eter of ring, m
h = liquid level drop in the supply tube, m
T = elapsed time between readings, s
The incremental infiltration rates are then plotted to show
infiltration rate change with time. This curve will show an exponential
decline in infiltration rate with tine. If the test is run to the point
where the soil is at or near saturation (i.e., steady-state), the curve
will be practically flat, which indicates a constant infiltration rate with
tine. If the depth to the wetting front is known, the hydraulic
conductivity can be calculated from the infiltration rate as follows:
K = I/H (16)
where K » hydraulic conductivity, m/s
H = hydraulic gradient = (h - L!/L
h « depth of water ir. the inner ring, m
L = depth to wet t ir.g front, m
I = infiltration rate, m/s
Equation, 17 is usee to calculate tr.e infiltration rate cf the soil
.. wr.er. : er.siox.eters are used with tr.e double-ring inf i It rometer, as shown in
Figure 9.
: = *[!+ (D,/'L#) - (L'/L,)] (17)
k = hydraulic conductivity, n/s
3^ = depth of liquid above soil Surface, m
= depth of wetting front, m
'J = soil moist ure tension at the wett ir.g f ror.t, m
Ter.sio.Tiecer readings are recorded during the hook gauge measurement
period and used to determine the depth of the wetting front in the rings.
The test is analyzed by plotting infiltration rate with ti.T.e. The wetting
front is determined when the tens icr.ete r at that depth registers zero soil
moisture ter.sio-..
Qua 1 i t v A; sura r.ce--2ua 1 i t v assurance considerations for this method
are similar to these giver. in the section on ASTM double-ring
ir.f ilt rometers. Care should De taker, to calibrate ail level recording
-devices and to install tensiorr.eters such that leakage along the bored hole
does .not occur.
Example Data--
Rogowski et al. (1934) used a modified double-ring infiltrometer
(barrel infiltration ring) to determine the hydraulic conductivity of a
coroacted soil liner. The cuter ring was 0.6 :r. in diameter and the inner
36
-------
ring was 0.3 m. Figure 10 shews the inner ring infiltration rate as a
function of time. The steady-state infiltration rate was approximately
1 x 10 ® m./'s .
A modified double-ring infiltrometer similar to that shown in Figure
8 was used tc determine the infiltration rate of two clay soils at sites in
Austin, Texas (Brown 1985) . The saturated infiltration rates of the two
- A — 7
clay soils were 6.7 x 10 and 1.3 x 10 m/s.
Corrr.ents--
With modifications similar to those shown in Figure 8, the ASTK
double-ring ir.f ilt rometer has beer, used to measure infiltration rates cf
_ * P
9.7 (z 8.1) x 10 n/s (Rogowski et al. 1985). The ability cf the test tc
ootain representative values is questionable, nowever, because of the
relatively small size of the the inner ring (0.3 m ir. diameter) .
Maercpores that can affect the infiltration rate by orders of magnitude
could have a wider average spacing than the diameter of the inner ring.
Bex I n f i It rcm.eter
This method involves the use of a sealed box and a relatively complex
star.dpipe arrangement to measure infiltration into soil in a 0.36-m^ area.
It has been used to measure hydraulic conduct ivit ies as low as £ x I0~iu
,T.S .
?rir.ciples--
Scx ir.fiitrc.Tieters work on the same principles as the ASTM coubie-ring
ir.filtrcr.eter (ASTM D-3385) . This method, however, differs from the ASTM
C-32S5 method by eliminating evaporative losses throucr. tr.e use of a sealed
inner rir.c {Andersen et al. 1934) . A float valve and a thermal expansion
reservoir assembly are used tc maintain a constant head on the soil in the
G. i-r. v C . 6-rr. domed stainless steel box. Diurnal changes occur ir. the
v: iurre of » = ;er ir. the ir.f i It romet e r as the result cf amoier. t air
temperature manges. These water velum,e changes are acccirr.cdated in tne
tnermal expansion reservoir placed arounc the float valve chamber.
Evaporative losses are eliminated by completely sealing the refill
reservoir i.star.dpipe) except for a small vent at the top for maintaining
atr.oscr.er ic pressure in the reservoir. Changes in water ievel in the
refill reservoir are used to est imate the infiltration rate cf the soil
under tr.e sealed domed box.
M«2 c r, o c - -
Acoarat-s--Figure 11 is an illustration of an installed box
ir.f ilt rometer . Its main features are as follows:
1) Stainless steel domed box.
2; Refill reservoir.
3) Thermal expansion reservoir and float valve assembly.
35
-------
>-a._
U_
15 20 23 30 35 40 45 50 55 60 65 70 75 80
0
5
TIME, DAYS
FIGURE 10. INFILTRATION RATE PLOTTED AS A FUNCTION OF
TIME, USING DATA FROM THE INNER RING OF A
DOUBLE-RING INFILTROMETER (ROGOWSKI et ai.
1984).
40
-------
RESERVOIR CAP
REFILL RESFRVOlR
THERMAL EXPANSION RESERVOIR
AND FLOAT VALVE ASSEMBLY
OOMEO LIQUID CHAMBER
infiltrometer
CONCRETE COLLAR
SOIL LINER
WATER SEALANT CEMENT
c
ANCHOR CEMENT
BASE SUPPORT
V
ANCHOR CEMENT
FIGURE II BOX INFILTROMETER WITH DETAIL OF EMBEDDED BASE SUPPORT
(ANDERSON ef al. 1984).
-------
?rocedure--The following is a listing of the procedures used with an
installed box infiltrometer:
1) Dig the cuter ring trench 11.5 m diameter).
2) Dig a trench on a level 0.6-m by 0.6-m surface of the soil liner
inside the outer ring trench.
3! Fill the inner ring trench with anchor cement to approximately
three-fcurths of the trer.ch depth and immediately lower the dorr.ed
steel box into the trench.
4) Fill the cuter ring trench with anchor cenent tc about three-
fourtr.s cf the trench depth and immediately lower the cuter ring
into the trench.
5) Allow the cement to cure for 45 minutes.
6) Fill the steel box with liquid to a depth of SDout 15 cm ana check
for leaks between the box and the soil.
7) Fill the outer ring until tr.e coned steel box is covered with
liquid; allow the box to rill witn liquid.
3; Attach the float valve and thermal expansion reservoir.
9; Attach refill reservoir.
11) Fill the refill reservoir with liquid and attach the cap.
11) Periodically measure the level cf liquid ir. the refill reservoir
and the temperature of the water ir. tne outer rir.c.
Calculations — The following form cf Carey's equation is _;ed tc
calculate the hydraulic conductivity of the soil liner:
x = v/{ath; : 13)
wr.ere = volume of liquid that flowed from the refill reservoir, p."
A « crcss-secticr.a- area over wr.ich inriltrstior. oecurree, m'
T = time in which liquid volume was measured, s
H = hydraulic gradient » (h -r L)/L
h = depth of water in steel domed box, m
L = soil liner thickness, m
Equation 19 is used to determine the liquia volume change cue tc
thermal expansion of liquid and steel:
dV = Vq d? <3W - 3S) (19;
where dV = change in liquid volume due to temperature change, jr."
V = volume of liquid at measured temperature, m^
42
-------
dT - change ir. temperature, C
Bw " coefficient of volume expansion of liquid, l/cC
3S ¦ coefficient of volume expansion of steel, 1/°C
Quality Assurance—The quality of tne data from this test depends on
obtaining all data cn concurrent tests at the same time during the day tc
offset the fluctuation in water level measurements due to differential
thermal expansion of water and steel. The temperature cf the liquid should
be recorded and used to adjust the measured volume of liquid to that at a
specified temperature (Equation 19) . Data from this test should be plotted
to determine when the infiltration rate of the soil has reached a steady
state.
A trench and anchor cement should be used to install the box into the
soil liner to control leakage of liquid a round the bottom of the
ir.filtrometer sidewslls. The inflltrometer aiso must be insulated from
heat tc minimize temperature changes in water ar.c the steel material used
to construct the box.
Example Data--
Ar.de rs or. et al. (I960 used box mf i It romete rs tc evaluate hydraulic
conductivity cf a 0.12-it-t.nick scil liner. The liner was constructed of a
mixture of 30 percent coarse sand ar.d 20 percent ber.tonite. The mixture
was compacted tc greater tr.an 100 percent maximum density as determined by
ASTM Method 2-633. Four box infiltrometers were used simultanecusly or.
the soil liner to evaluate the variability of steady-state nydraulic
conductivity values. The overall near, steady-state value for the
— If"
in.filtrometers was 4.2 1.2) x 10 m/s.
A small error is introduced into the me6surenier.es made with this
device when a temperature change occurs. For eacr. degree centigrade change
in temperature, a change of approximately 0.04 x will occur in the depth of
water initially in the infiItrometer.
Daily fluctuations occurred in the hydraulic conductivity values, but
the trend ir. the fluctuations was the same for ail four inf iltrometers
(Figure 12; . This appears to indicate that the cause of aaily variation
was not measurement errors, but rather actual fluctuations ir. the
infiltration rate with time.
Comments--
Ecx infiltrcmeters are capable of measuring r.ydrau 1 _c conductivity
values substantially lower than 1 x 10 m/s. It is unlikely that a single
bcx ir.f iltrcmeter would cover sufficient area (0.36 r") to determine the
overall field hydraulic conductivity. In the study cy .-.r.sersor. et al.
(1984), however, the box inf i it rometer was used ir. arrangements cf four to
allow definition of both the overall field hydraulic conductivity and the
spatial variability of the liner. The number cf infiltrometers used (four)
appeared to be sufficient to define spatial variability. Ir. addition, ir.
the single study conducted, the aggregate area evaluated by the
ir.filtrometers (1.44 m ) appeared to be sufficient to obtain representative
hydraulic conductivity values. It should be noted that the test site was
42
-------
-o #2
-o #3
©-
-a n 4
6
>
'>
i-
u
o
(J
o
D
<
C
O
>
I i 10
¦ "9
9H
e-
7-
6-
5J
0---0 MEAN
i 10
—i—
156
-1—
376
336
396
416
<36
456
TIME, HOURS
FIGURE 12. HYDRAULIC CONDUCTIVITY OF THE SAND-BENT0N1TE
LINER MEASURED USING FOUR BOX INFILTROMETERS
(ANDERSON et ol- 1984).
44
-------
2
snail (8 m ) and that it required a larger effort pet unit area tc produce
a uniform soil liner than is typical for full-scale liners. Consequently,
although an aggregate area of 1.44 was sufficient to characterize the
spatial variability of the test site, it probably would be necessary tc
test a larger area under conditions more typical of full-scale liners.
The box ir. f i 11 r cms t e r is relatively difficult to use and would
definitely require skilled personnel for installation, monitoring, data
collection, and data analysis. Temperature fluctuations can greatly
influence the apparent infiltration with this method. Consequently, data
should be collected at the same time each day, and the values obtained
should be adjusted tc compensate for thermal expansion or contraction or
the water and test equipment.
Sir.cle-Rino Infiltrcmeter
A single-ring infiltrometer consists of a metal cylinder 20 to 60 cm
in diameter, which is pressed or driven into the soil. Infiltration is
measured by ponding water inside the cylinder ana by meas-rmc tne rate
that the free surface falls, or by measuring the rate at which water must
be added tc maintain a constant depth m the cylinder (Say and Daniel
1535a) . A network of tensiorr.eters may be used to determine the depth of
the wetting front, which is subsequently used to calculate the hydraulic
conductivity of the soil.
r nr.ciples--
The sinale-rir.g method is similar to the double-ring method except for
the elimination of the outer ring. When the wetting front exceeds the
depth of the curied cylinder, water flows from the ring vertically and
laterally. The lateral flow may cause tne measured infiltration rate to be
higher than expected, and a correction factor (Day 1984) can be used to
account for the effect of lateral flows on the deterrr.ir.ed saturated
hydraulic conductivity values.
Xethcc--
Apparatus--A schematic diagram of a single-ring inf lit rc.meter is snc.vn
in Figure 13. Essential features of the in f i i t: ome t e r ir.cl-de the
following:
1) Metal cylinders 2C to 60 cm in diameter.
2) A hook gauge or suitable method of maintaining constant water
level in the cylinder.
3) Te.nsiometers.
4; Plastic sheeting tc cover the ring.
Procedure--The following is a listing of tne procedures usea witn
single-ring infiitrometers:
1) Carefully install tne single-ring cylinder into the soil liner
-------
OUTER RING
GROUT
FIGURE 13. OPEN SINGLE-RING INFlLTROMETER.
-------
2) Install hook gauge or Mar:o::e tube (or suitable method for
maintaining constant water level) ,
3) Install tensiometers at different depth3 to monitor depth of the
wetting front. (Typical depths are 0.15, 0.30, and C.46 m.)
4) Ada water to the ring, making sure not to scour the soil, and
cover the inner ring with plastic sheeting.
5) Recsrc tirr.e, volume of water used to maintain the constant head,
anc tens iorr.eter reading.
Calculatlon--The constant-head equation (Equation 20) or the falling-
head equation (Equation 21) can oe used to calculate the saturated
hydraulic conductivity of the soil.
K = Q W (H A t)
L ln(H2/H1)
(20)
(21;
where K = hydraulic conductivity, m/s
L - depth to wetting front, rr.
H = head less through liner, rr.
2
crcss-sectional area of soil in the ring, rr.
•s_apsed time setween readings, s
volume of water that infiltrated into the soil, mJ
initial. depth of water m cylinder, rr.
f water in cylinder after tixe interval t, i
Cualir -• Assurance — The value of the hydraulic conductivity obtained
frc.-i the sir.gle-ring infiltrorr.ec.er when Equations 20 or 21 are used is
multiplier by a shape factor (F) , which varies from 0 to I (Day 19 9 4) tc
account for lateral spreading (Figure 14). The correction factor is based
cr. the geometry cf the soil lir.er, the infiltrometer, and the degree of
ar.isotropy of the scil. Hydraulic conductivity versus time relationship
= ."ic_ic also be plotted tc show a steady- state saturates hydraulic
conduct ivit v :.i5 beer, attained.
Exarr.p.e Tito-
Stewart and Nolan (1987) reported hydraulic conductivity data or. six
scil lir.ers as determined by 0.56sincle-ri.no infilt rorr.eters. The hy-
v Q w O
draulic conductivities varied from 2.3 x 1C to 3.0 y. 10 rr./s . Labora-
tory hydraulic cor.ductivity values from fixed-wall permeameters that used
Shelby tube samples taken from the soil inside the infiltrometers after
completion of infiitrorr.eter tests ranged from 1.3 x 1C "i(^ to 5.5 k 10-1^
rr./ s .
Bagchi (1986) reported data cn the permeability cf blended soil used
tc construct a landfill soil liner. These data indicated field hydraulic
conductivities determined with a 0.15-m diameter single-ring pe rir earr.ete r
47
-------
0.8
0.6
o.e
0.6
0.4
0.8
0.6
0.4
0.2
D= DIAMETER OF RING
T = EMBEDMENT DEPTH
H= LINER THICKNESS
FIGURE 14. SHAPE FACTOR (F) FOR SINGLE-RING TEST (DAY 1984).
48
-------
and a double-ring permeamete r
diameter outer ring) to be 4
r- V,
( 0 .1 5-m-diameter inner
x 1Q-7 and 6 x 10 7 m./s,
7
ring and 0.61-m-
respect ively,
compared with a laboratory permeability of 3 x 1
rn/s .
Triad Engineering Consultants, Inc. (1986) used a 55-gollcn open drum
(0.6 -t. in diameter) 33 a single-ring infiitroneter tc determine field Hy-
draulic conductivity of a compacted clay iir.er ir. West Virginia. The over:
all hydraulic conductivity of the Iir.er was determined cq be
n/5 * .... o n .. , rt~lC
a r.d
5.6 x 10 10
2.8 x 10~10,
Average data frcrn the three rings used are 8.0 x 10
6.0 x^lO n/s, compared with the average laboratory value cf
2 . C
10
m/ s .
Day and Dar.iei (1985b) determined the field hydraulic conductivity of
prototype clay liners by using single-ring and double-ring
ir.r lit rometers
Results are surrjr.a rized in Table 6.
The s ingle- and
dcubie-r
ing data were within one order cf
rnagn i t ude .
TABLE 6.
HYDRAULIC CONDUCTIVITY OF TWO
PROTOTYPE CL
LINERS USING
SINGLE-RING ANO A DOU3LZ-
KING INF I LTP.CMETER
Type of Diameter cf
•4y
dra
u 1 i c
£ c i ( rr.
ring ring, rn
conductivity, ri./s
i"* *i ^ •.'
Single C . 5 £
1 . 2
X
10 I
Single 0.56
1 5
X
10" '
Single 1.22
1.2
X
lO"8
Single 1.22
i. . J
X
10~8
Double 0.30 a r.d 0.51
i.e
X
lO"3
c ; a y 2
Single 0.56
1 . 0
X
10"?
£-r:gie 0.5 6
3. 5
X
10" ''
Single 0.56
3 . 4
X
I0_t
S ir.gie 1.22
1. 6
X
10-6
Dcucle C.3C and 0.51
2 . 5
X
IC'6
Colder Associates (1964} used two 3. 1-m-diar.eter infiltration rings to
determine the in situ hydraulic conductivity of a compacted fill ir.
Georcia . hydraulic conductivity values for both rinos ranced from
_ 1 _ a .is
t.3 x 10 ~ " tc 1.5 x 1C n/s, with an average value of ".6 x 10 "" m/s .
Ccr.r.ent s--
L-rge single-rir.g mf ilt ro.T.eters can measure hydraulic conduct iviry
values as lew as 1 x 10 it./ s. Zf the ring is cf sufficient diar.ete t, the
nycraulic conductivity values obtained should be representative of the
overall soil liner. Until there is a definitive data base documenting tne
minimum rinc diameter needed to obtain these representative values, a total
area of at least 2 tr/ within the rinc is recommended.
49
-------
Errors can be introduced into the hydraulic confine: ivity values as a
result of lateral flow and evaporative losses. There is no danger of
falsely concluding that a soil lir.er meets a specific performance standard,
however, as long as the uncorrected hydraulic conductivity values are less
than I * 10 ' m/s. If it is necessary to use a correction factor to reduce
hydraulic conductivity below i x 10~® n/s, additional tests should be
conducted either with a double-ring i.nfiltroxeter or by using measures to
eli.T.inate evaporative losses.
Single-ring infiltrcmeters have the advantages of be.r.g simple and
inexpensive. Care should be taken, however, to make s.re that thermal
expansion of the water dees not yield false low values. Other possible
sources of false low values are the incerception of rainfai 1 by the open
ring or incorrect calculation of the hydraulic gradient. These sources
of error can be avoided by U3ing a sealed ring and a free drainage layer
beneath the known thickness of a soil liner, respectively.
Sealed Double-Sing Inf i It rom.et er
This method involves -he use of a sealed ring or box tnat Measures
infiltration i r. to a relatively large area. The sealed double-ring
infilt rorr.eter iSDRI) has a sealed inner ring that permits measurement cf
low infiltration rates and rr.ir.ir.izes procle.T.s with temperature fluctuations
and evaporation (Elsbury et al. 1333). It is relatively easy to operate,
but as with all infiltration ring devices, it r.ust be carefully installed
to assure tr.at no leakage occurs around the edges.
The sealed double-ring inf i It ronr.et er consists of =r. inner rinc and an
outer ring. The fiberglass inner rir.g is 1.5 rr^ ir dia.tieter and has a
sloped top that extends only 12 c.t. above the lir.er surface . The outer
ring, which is r.ace cf 3 ."-rr.-ior.g aluminum panels, is usee to pane water
around the inner ring to insure vertical percolation in the inner rir.c.
Flo*' is measured during an infiltration test by using a f low-rneasurener.t
bag attached to the ir.filt rorr.eter through Tvgor. tubing. Any water flow cut
of the infiltrcneter into the ground is replaced by water frcn the oag.
Flow measurement is initiated by filling the bag with a known weight of
water, connecting the bag to the inflltrorr.eter, ar.d periodically retrieving
and reweighir.g the bag.
rri.iciplas--
Ar. SZRl jses the double-ring infiltration principle to measure
hydraulic conductivity cf saturated soil near the ground surface. This
consists of ponding water on the ground surface and .-r.easuring trie amojr.t of
water that enters the soil ir. a given tirr.e interval. Tne double-ring
arrangement is designed to assure that the flow fror, the inner ring is
vertical. If the hydraulic gradient is known, the infiltration rate car. De
used tc determine hydraulic conductivity with Equation 23.
Method--
Ascaratus — A schematic of ar. SDRI is shewr. m Figure 15. The
following are the essential features:
-------
OUTER Ring wall
MEASUREMENT BAG ASSEMBLY ON CONCRETE BLOCK
SEALED INNER RING
v VALVE
BENTONlTE GROUT
FIGURE 15. SCHEMATIC OF A SEALED DOUBLE-RING INFlLTROMETER
(MODIFIED FROM ELSBURY el ol. 1988).
-------
i;
A sealed fiberglass inner ring.
2) An open metallic cuter ring.
3) Flow-.T.ea3ure.T.er.t, bag and ball valve a::ac,h.ner.:s.
4) Tygcn tubing.
5) Tensioxeters for monitoring the wetting frcr.t.
Procedure--The following is a listing of the procedures used with
sealed dcuble-ring infiltrcmeters:
1) Position the inner and outer rings on the soil l.ner ana mar* the
soil surface around the edges of the ring.
2) Re.T.cve the rings anc excavate trenches alone tr.e mariced -mes oy
one of the following xetheds :
a) Sxcavace the inner ring trenches !5 cm wide y. 15 cm deep; with
a masonry harcr.er or chain saw (care should be taken not tc
excavate within the area under which the inner ring will be
placed;.
bj Excavate the outer ring trenches (11 cm wide x < 1 on deep)
with a ditch-witch.
2> Fill the trenches with a suitable grout (eg, ientor.ite) and
embed both rings into the trenches.
i; Fill the inner ring with water tc a depth cf 3 err, ar.d test rc-r
leaks around the edges.
5) Install three ter.sioneters at depths ci J. 15, C ,
each ir. the area between t.ne rings to monitor the pre
wetting front.
c> Slowly fill the cuter ring with water in a ::jr.r.e: that not
scour the soil ana rvuady tr.e water.
?> Purge all trapped air ir. the inner ring cy tapping on the
infiltrcrr.eter with a bloc* of wood.
8) Till the measurement bag with water (squeeze to remove all air
bubbles before closing the attached ball valve).
5) Dry and weigh the bag before connecting it to the teed tube
leaaing to the inner ring.
10) Irx-.ediately open the valve and record beginning tirre for the
measurement period.
52
-------
11) Periodically close the valve, remove 'he bag, record the tirr.e, and
reweigh the bag after it has been dried.
Calcuiations--The infiltration rate (I) and the hydraulic conductivity
are calculated by Equations 22 ar.d 23:
I - Q/(A t!
(22)
where I = infiltration rate, m/s
Q = volu.T.e o£ flow, m
2
A = cross-secticnal area of flow, rr.
t = ti.T.e interval curing which flow Q occurred, 3
The calculated infiltration value (1) is then jseJ tc determine
hydraulic conductivity (K) :
K = Z / H
(23)
K = hydraulic conductivity, m/s
H = hydraulic gradient = (h + D!/D
n = depth cf water or. soil, .n
3 = depth of wett.r.g front, t.
I = infiltration rate, rvs
Oucii'-V Assurance — As with a _
i.T-cor t ar.t for the sell surface to
installs tier, cf ihe ir.filt rometer
Three tensior.ete rs should be instal
the inner ring as possible (C. 15 to
surface mf _ _t ra t _q:: :-e:;.odi, it is
be protected f roir. sesicc&tior. durir.c
ana prior tc filling «.;r. water,
led at each soil deptr: and as close to
0.20 mi .
Scn.e forir. of insulation, such as 1-inch-thick styrofoar. sheeting, is
needed tc rrir.irr.ize temperature changes during the test . Also, the
r.easurer.er.t bags should be attached and removed at approximately tne sarre
ter.peratures so that any error due to temperature variations would be srr.ail
ccr.oared with the water flux ir.to the soil.
An SD."vI is based or. maintaining vertical seepage oen;;:!". tr.e inner
ring and having material beneath the liner cr.at is free-eraining. Steady-
state fl;w is maintained ar.a tne temperature cifferer.ce netu'ser. water in
the inner ar.c cuter ring is minimized.
l. y. c _ 6 Data"-
Elsbury et ai. (1566) used a ccmfcir.at ion of ar. urderdrain, soil cores,
2
and SDR I s that covered ar. area of 2.1 m to determine hydraulic
conductivity of a prototype soil liner in Houston, Texas (Figure 15). The
underdram showed the SDRI values tc be close tc cr.at cf tr.e overall liner.
Field hydraulic conductivity of the liner, as measured oy the 5CRIS, ranged
from. 3.; x 10""7 to 2.6 x 10-D m/s( whereas the soil ccres_ taken froir: the
soil liner yielded values ranging from 1 x 10 ~~ tc 4 x 10 ~" i::/s
Data from SDRIs installed in South Carolina indicated that the
- 9
hydraulic conductivity of test-fill son liners ra.naed from 5 x 10 to
53
-------
10"
10'6
10'1
SDRI NO. I
RECORD
SECTION
SDRI NO. 3
SDRI NO. 2
SDRi NO. 4
10
20
30
TIME, DAYS
FIGURE 16. FIELD HYDRAULIC CONDUCTIVITY VALUES MEASURED
WITH AN UNDERDRAW AND FOUR SEALED DOUBLE-
RING INFILTROMETERS (ELSBURY et al. I988).
54
-------
- 9
7.5 x 10 rr./s (Anderson and Sai 1988 ). An SDRI usee on a test-fill soil
— Q
in New York found the hydraulic conductivity to be 1.1 x 10 ' m/s (Anderson
and Sai 1988) .
Daniel and Trautwein (1936) repotted data from a site north of Niagara
Tails, New York. Hydraulic conductivity (K) of two soil covers were
- a _ - n
reported to be 6 x 10 and 8 x 1C *" m/s.
Hsien and Yamamoto (1987) performed field and laboratory hydraulic
conductivity tests on a 36 m x 2<5 m x 0.76 m c-ay liner in a test fill.
The SDRI data indicated that the average hydraulic conductivity was
2.7 x 10 m/s, whereas a BAT permeameter gave a value of 3 x 10 n/s.
These field results averaged one order of magnitude higher than the
laboratory data.
Woodward-Clyde Consultants 11967) provided EPA with results of field
infiltration tests using S2RIs on scil liners (Figure 17). The data
indicated a final infiltration rate of 2.0 x 10 m/s at the end of 192
days .
Comments--
An S3RI is capable of measuring hyaraulic concuctivieies of less than
1 x 10"9 rr/s. The 5DRI also covers a relatively large area of soil (the
inner ring covers an area of 2.3 rrL} . Tests conducted with SDRIs appear to
confirm that the equipment can yield values representative of the overall
soil lir.er (Els bury et a 1. 1 S S B j .
In rest cases, compacted soil liners are unsaturated anc the
infiltration rate is not in a steaay state. A plot of infiltration rate
versus time is needed to determine when infiltrat ion thtuug.n me soil ;-.u;
reaor.eu a steady state. Tests must be run for several weeks to achieve
this desirea steady-stste condition.
Although S3RI tests require greater installation time than many
methods, few ambiguities exist in this experimental method. Periodic
recording of tine anc the .measurement bag weignt are the only tasks
performed after the equipment is installed. The SDRI appears to yield
high-quality data and presents few possibilities for yielding false lew
values.
collection lysimeters
Collection lysimeters are p.aced beneatn the compacted soil lir.er to
collect liquid percolating through the lir.er (Elsbury et ai. iyst; . A
perforated pipe is cfter. installed at the low er.c of tr.e lysimeter to
collect and move any accumulated liquid to a solid pipe and subsequently cc
an access point outside the boundary of t:ic compactec soil liner, where the
liquid can be extracted and analyzed. Lysimeters have been used to monitor
the cuantity and quality of the leachate from landfills m Wisconsin and
Canada.
C. C
-------
I
A « TlMt
00232 2 m
1itO*
101 «? i?0 i?0 "JS H4 )5? i«0 16A irr. ifM i*j
7?
0
0
OAYS
FIGURE 17. SONi DATA PROM A SITE IN WICHITA, KANSAS (WOODWARD-CLYDE CONSULTANTS 1907).
-------
Principles
Lysimeters operate on the pcincipie of collecting and .T.easuring tne
volume of liquid percolating through a cefmed area of an overlying soil
lir.er. This method often involves installing a lysirneter around a porous
drain field that is positioned directly under a compacted soil liner. The
device functions by intercepting downwa rd-mo vi r.c water and diverting it
inco a collection system. Hydraulic conductivity is calculated by using
Darcy's law and tne measured flow (seepage) rate. Lysimeters should be
installed at locations well above the water table to avoid the possibility
of a rise in the water table affecting liquid flow into the lysirneter.
details of collection lysimeters have been ciscussec by fCmet and
lir.dorff ( 1 93 3), Kmet et al. ( 1 96 5:, Elsbury e; al. (1988), Reades (1966),
Gordon et al. (1 908) , ana Bacopoulos (1986) . A summary of this information
is presented ir. the following subsections.
A pea r a t us - -
: iv.:e 15 is a schematic of a collection lysirneter. Essential
ieaturc-s of the system are as follows:
1: The collection lysirneter (a wide variety cf materials have been
used to construct lysimeters) . Tr.e durability cf the material
usee should be a primary selection corisicerstion because the
lysirneter cannot be replaced once it is installed. The iysir.eter
should be constructed with a r.or.react ive material (such as a
geor.ccbrar.ej that is less permeable than the overlying ccmpacteo
soil liner. The selected material should ce chemically compatible
with the underlying soil and the expected liquid tc be collected.
2; Clean and highly permeable granular material (less than 51 passing
a 2CC-mesn sieve) fcr oacxfill into the lysirneter. (Material
should be adequately sized to prevent puncture cf the membrane
used ir. tr.e lysim.eter . )
3; Transfer pipe(s) with substantial perforated sections inside the
lys.me t e r.
As shown in Figure 18, the collection lysirneter has a general
underdram and an underdrain record section. Data on underdrainage were
collected from the record section only. This test-fill design allowed a
buffer zone cf underdrainage around the zone used tc calculate hydraulic
conductivity. The outer general underdrain assures that only vertical
underdrainage and no boundary artifacts wculd affect the hydraulic
conductivity values measured m the ur.aerdrair. record section.
? rocedure--
Ccl lection lysimeters are installed either w-thir. or below l h =
ccmcactec soil lir.er. Lysimeters installed within tr.e soil liner give
information about flow rates and leachate migration more quic.
-------
LtNCR BOUND ART
7
UNOCAORAIN
RtCO«0 SCCTion
VENT
unoerdrain
RECORD
section -
GENERAL
U*OERC>RA!N
outfalls
unocpdrain
R£CO«D
SECTION
OUTPALL
GENERAL
UNDERDRA1N
PONO WALL
UN^EPORaiN
RECC^O
S£CTi0.s
OUTF6LL
i-Fi - Thick
SOIL LINER
QRaiNAGE
geotextile placed
BETWEEN UNDERORAtN
AND LINER —
rono wall
1
UNO€RORiiN
RECORD
SECTION
VENT —7
y
RECOBO SECTION
general UNOERORA1N
7
Pl»STiC Sheeting
AND GEOTEXTILE
PlICEO BELOW
GENERAL UNOERORtif.'
scale in fee t
figure 18 A TEST FILL EQUIPPED WITH AN UNDERDRAIN
RECORD SECTION (MODIFIED FROM ELSBURY e1 ol.
I988).
58
-------
installed below the soii liner. The following is a listing of the
procedures f;r collection lysimeters:
1) Install a flexible membrane lir.er (at least a 40-mm-thick sheet)
of appropriate size at a predetermined depth in or below the
lir.er; Km.et (1983) recommenced a ranee of lysirr.eter sizes for use
in compacted clay soils.
2; Install transfer pipes (perforated ar.d solid: to ouuet location.
3) Construct raar.hole and sampling riser: connect a transfer pipe to
the sampling riser and the riser to the rr.annoie.
4) fill ly si meter with sand ar.d wet to field capacity befcre
compacting soil liner or. top to tne specifies thickness.
5; Periodically collect anc measure accumulated liquid. (Kit:. Icw-
hysraulic-cor.ductivity soil liners it may oe years befcre liquid
accumulates in the lysimeter.)
6; Calculate nydraulic conductivity.
aquatic", is usea to ca.cj.ate
:scted soil lir.er above the collecti;
tne hydraulic
n lysimeter.
Q/H.A
(21)
;ne re
.yarau_ic conductivitv, m/s
•clume cf liquid collected par unit time,
iydraulic gradient » (h - L) /h
: rcss-sect lor. area of lysimeter, rr.^
.iquid head cr. soil liner, m
:tr.cacted soil lir.er thickness,
Quality .-.ssurar.ee--
ks wit.-, every other method discussed in this cccument, the accuracy of
determining t.-.e quantity of liquid collected by trie ciiieoticr lysiTeter
cieterr.ir.es tr.e quality of the data produces. The greater tne quant-ty cf
liquid rcllectes, tr.e lower will be the error involves in .T.easurir.c the
volume of liquid that percolated through tne soil liner. Tne lysimeter
used should oe large enough to insure tnat the system intercepts a
representative number cf macropores.
Collection lysimeters uses ir. compacted soil ..ne: tests require
substantial time for steady flow conditions tc d^velc_ :wcc/.^ cr mor.tr.s) .
Tests should be run and data plotted tc show that steady-state conditions
nave ceer. achieved.
59
-------
Example Data
2
Bacopoulos {1986~ monitored flow rates from three 225-m lysi-tieters
within and below a compacted clay liner to determir. e the hydraulic
conductivity of the compacted clay liner at the Keele Valley Landfill site
in Toronto, Canada. Flow rates from the lysimeters, which were installed
within trie liner in 1984, indicated that the hydraulic conductivity at the
end of Jane 1 965 was less than 1 x 10~lvl m/s. No flow was ODserved m the
lysimeters installed below the clay lir.er at the end of the measurement
period.
Reades (1966) presented detailed monitoring data from the Keele Valley
Landfill. Laboratory hydraulic conductivity data were compared with data
from the 225-m^-area lysimeter. The hydraulic conductivity data frorr. the
lys:reter_ installed within the clay lir.er were found to range fioin
S.O x 10 "" to 9.8 x 10 lm rr,/s, with a mean cf 6.6 >: 10 m/s. Tr.e
laboratory hydraulic conductivities of samples from the site djnr.c _ c r.e
three const ruction seasons had ceor.etri c means ranamo from x 12 ** to
_ i i ' ¦
£ . 2 x i: -1 m/s.
Elsoury st al. (1966i presented data on a collection lysi.xeter iron a
site m Houston, Texas, ar.d compared t.ne data ::o:r. tcu: sealec double-ring
1.-.f ilcror.ecers with laboratory data. The nydreui; c cui.uuct _ vi t y of tr.e
ccmoacted soil liner determined witr. tr.e lysimeter wis 1.6 x 1C ® x./s,
-7
whereas tne SC?.I snowed the conductivity to range :r;r. j . 5 x IC to
2. i x IC c r./s (Figure 15).
Gorier, et al. (1983) presented .:. fcr::.c'. .on on ccl.ccu:i. iysi.T.eters a:
three active, fu.i-scale, clay-lir.ei lar.af.l_s ir. w.sccr.sm. Five to seven
years of data or. the quantity cf liquid collettec f:o~ these lysimeters
were used to estimate field hydraulic conductivity by using Daroy's law and
tr.e rr.eas_red or estimated leachate heac. "able presents the data from
two cf tr.e landfills. These data indicate that tr.e hydraulic conductivit v
- 9
of the compacted soil liner at the two sites was lower than 1 x 10 m/s.
The average laboratory hydraulic conductivity values tcr compacted
clay samples from Portage County. Landfill ranged frcr. 1 x 10 to 3 x iC
" rvs, whereas the values for Saux County Landfii.. ra.ngeo from S x 10
to A x IC *v rr./s.
Con-ant s
Collection lvsimeters car. measure hydraulic conductivities of less
_Q
than 2 x 10 n/s. If tr.e lysimeters are _arge enough, they are also
capable of yielding hydraulic conductivity values that are representative
cf the overall soil liner. The collection lysimeters used by Eisbury et
al. (1583) were approximately 19.4 m , whicr. proved tc be large enough to
yield representative values. The main drawback of this .method is tr.at it
can take months to obtain steady-state r.ycrau.ic conductivity va.ues.
L'r.lixe ir.fi It rcneters, which give ir._tial.y nign value; tr.at cecrease tc a
low steady-state value, collection lys ur.ete r values typically oegin veiy
low and gradually build up tc a higher sceauy-state vaiue. Consequently,
-------
the teaming period may well interfere with const ruct ior. of a facility, and
the adequacy of a particular soil liner design nay not be known until the
end cf a long test period.
TABLE 7. RANGE OF HYDRAULIC CONDUCTIVITY VALUES FOR VARYING LEACHATE HEADS
AT TWO CLAY-LINED LANDFILLS IN WISCONSIN
Leachate
Hydrauiic
Locat ion
head, m
Gradient
conductivity, m/s
Fortage County
0.03
1 . 0
9.1 y. 1 U" U
Landfill
0.15
1 .1
a.3 x 10" 1 1
0 . 30
1 . 2
7.6 x 10 ~ 11
0.61
1 . 4
6.5 x 10"11
0.91
1 . 6
5.7 x 10
1.22
1.8
- -1 l
5.1 x i u
1 . 52
2 . 0
4.6 x 10-11
3 . C 5
3 . C
3.0 x 10";;
6 . 10
5 . 0
: 8 x ic";;
9.15
7 _ q
1.3 y. ic";"-
Sauk County
C . 0 3
I . G
n . 0 V. 1 C * "
.an::i.;
C .15
1 . 1
3.6 x 1C
. 30
1 . ^
* i ¦ n~ *v
:. 6 i
X . 4
2.o x i 0 *v
0 . 91
l . 6
- - v .0-i0
I . 22
1 . b
2.2 :o";C
1 . 52
2 . 0
2.0 x : o ~1 ^
3.05
3 . C
i.i x ;c"4v
6 . 10
£ . 0
-.s x io"^;
9.15
•
_ . U. V
Tests comparing collection lysimeters and large i.-.f il: rc."-e:ers have
shcwr. tr.at the values obtained by each re: hoc a:e relatively close
Consequently, if a collection Ivsimeter is cesirec, concurrent testing cf
the soil liner should be conducted with large in f i it i c-.t.s ts: s . This
procedure could save rr.or.tr,s of waiting for results definitive enough to
begir. con st ruct ior. of a facility.
Construction of a collection lysimeter is tixe-ccnsuming and requires
skilled personnel. Special care must oe exercised to ensure tna: tr.e
flexiole membrane underlying the collection field is not can,aged curing
construction. Because aucn damage could result ir. fi-se- 1 ju r.yar;.-iic
conductivity values, concurrent testing of tne soil liner with, large
ir.f i It rometers is recommended. Ultimately, a large collect ion iysnt.eter
has the potential for yielding the most accurate values for quantification
of the volume and rate of liquid that moves through a compacted soil liner.
In practice, however, the response tirr.e cf a lysimeter lin.its its utility
for short-term tests.
6 1
-------
METHODS UNDER CURRENT STUDY
Velocity permeameters and porous plate infiltrcmeters are both under
current study for their applicability to measuring the hydraulic
conductivity cf compacted soils. Sufficient data are not currently
available to determine the adequacy of either method ir. tnis application;
however, both methods show promise and are briefly discussed here.
Velocity Permeameter
A velocity permeameter is used to measure the in situ hydraulic
conductivity cf a soil. This method is based or. monitoring the time it
cakes a water column to fall through a distance, A n. The rate of fall of
tr.e water column depends on the rate of water er.trv into a soil core
enclosed within the coring tube. A Hewlett Packard 41CX calculator
equipped with a timing module is used to determine the race o: fall cf the
water column. The rate is then jsed as input to a program in the
calculator tr.at computes tr.e hydraulic conductivity.
Principles--
A velocity permeameter uses a small computer to ueterm.ne in situ
r.ydra_lic conductivity from falling-head infiltration data. The method
involves tr.e use of a calculator equipped with a timing module to enter
data cn change of water level in a water column witn time. These data are
_sed ir. a computer program to convert the inf ormaticr. ir.ee hydraulic
¦cor.d-.ct ivity values. This method involves detern.inat icr« of a series cf
hydra-lie conductivity values on the same in situ soil core and plotting
- r. s s v z. 1 j s s w i ^ r. z. i rr. e
Apparatus--Ficure IS is a schematic ciagra:" cf tr.e velocity
permeameter. The main components of tr.is permeamete:" ra as follows:
Soil coring device.
2: Head tube with scale.
3; Outer tube with driver and coupler.
<; Opening for filling with water.
5; Hewlett Packard 41CX calculator equipped wi.cn a timing module and
an expanded memory module.
6) Printer.
Procedure—The following is a listing of tr.e procedures for installing and
using a velocity permeameter:
i i Select test a rea.
62
-------
-HEAO TUBE SCALE
U
¦hole for filling water
¦OUTER TU9E
-DRIVER
coupler
SOIL CORER
a) FRONT VIEW
4
HEAD TUBE
»h
¦ENO Or HEAD TUBE
SCALE
WWW' Is '*¦
W0- —r
w
¦SOIL CORING
TUBE
^ SOIL
WETTING
FRONT
(b) OPERATIONAL schematic
FIGURE 19. SCHEMATIC OF A VELOCITY PERMEAMETER
(MODIFIED FROM KANWAR el al. 1987 ).
63
-------
2)
Position coring device (corer) on soil and drive it into the soil
with the driver.
3) Attach outer and head tubes to corer by using the attached coupler
(Figure 19).
4) Fill the tubes with water through the tube opening.
5) Reccrd the time and water level in head tube.
5) Reccrd and store the tine it takes the water -eve. in the head
tube cc fall through a fixed increment ct distance !dh) measured
cff the head tube scale.
) Calculate the rate of change of flow with respect to head
; cv/dh).
3) Ceterrr.me the hydraulic conductivity of the soil in corer by using
the computer program and the Hewlett Packard (H?!
-------
<» 2 i 'O"5 ,
0
T 1 ; 1 1 : 1 1 ; 1 1
10 20 30 40 50 60
TIME DURING SINGLE FIELD TEST
TIME, min.
FIGURE 20. APPARENT HYDRAULIC CONDUCTIVITY VERSUS
TIME AS MEASURED WITH THE VELOCITY
PERMEAMETER (MODIFIED FROM MERVA 1987).
65
-------
Quality Assurance—The method depends on accurate determination of the
velocity of the fall cf water in the head tube, and this depends greatly on
the person recording the value and on the ratio cf the soil cross-sectional
area to the area ir. the head tube. The larger the soil core and the
smaller the diameter of the head tube, the more accurate the values
collected will be.
Example Data--
Rogers et al. (1937) used the velocity perme ameter tc determine
vertical and horizontal hydraulic conductivities cf Commerce silt loam soil
at a site in Baton Rouge, Louisiana. The range and mean values determined
in this study were compared' with data from 36 auger-hole tests from the
same area. The auger-hole tests, indicated the soil had a mean hydraulic
conductivity value of 5.1 x 10 D m/s at an average hole depth of 56 cm.
Hydraulic conductivity values measured with the velocity permeameter ranged
from 5.6 x 10 to 2.0 x 10 ^ m/s.
Co.T.T.er.ts--
Veiccity perir.eair.ee ers measure tr.e hydraulic conductivity of srr.all
areas, and the values obtained may not be representative of me cvera..
soil iir.er. This metnee has r.ot been tested or. compacted scil liners, ar.c
further tests are needed to determine whether it is capable of measuring
hydraulic conductivities of 1 x 12 ' m/s ana less. Difficulties have beer;
reported in sealing the edges cf the coring device under high liquid heads.
Also, the data acquisition and reduction technique involved ir. this method
.T«y den.and the t ime cf a skilled operator.
Porous Plate :r.filtrcmeters
?crc_s plate ir.f lit rometers use water infiltration under tension tc
estimate r.ydraulic conductivity. A porous plate is placed on tne soil
surface to supply water tc the scil, and the flow rate from a supply
reservoir is usea to calculate the infiltration cf water ir.tc tr.e soil.
Variations cf this infiltrcmeter nave been used ey several researchers
•i.e., Dirksen (1975), Clothier and white C13 61), Kocre et al. (192 6;,
AnV.er.y et al. ;1S33) , and Eaumcartner et al. (1987)]. The following
suosectior.s describe cnly two cf these designs because the basic principles
and methods are similar in ail the variations. The automated tension
ir. f i It i oneter ( An. ken. y et al. 196S) and the Guelph inf i 11 ronster
iBsuxgartr.er et al. 1 957 ) are the designs addressee.
Principles--
The automated tension infiltrometer (ATI; automatically measures tne
r.eight of water in a Kariotte column for use in calculating the volune of
water infiltrating into the soil. Water flows from the Mariotte coiurnr. tc
a porous plate placed on the soil surface, which distributes the .-ater to
the scil profile. Changes in water level in the Mariotte column are
measured automatically by using the difference in tension detween two
pressure transducers, one at the top and one at the bottom of the y.ariotte
column. This method requires two tests to solve a system of simultaneous
equations needed to calculate tr.e hycrauiic conductivity of tne soil. Each
test is run with a different size porous plate.
6 6
-------
The Guelph infiltrometer (GI) is a modified Guelph permeameter (GP)
with ar. attached disk, that allows it to be used for measurement3 at the
surface with little disturbance of the soil surface. The analysis is
similar to that used for the GP except that the GX uses a zero depth oi
water in the borehole (H = 0). In addition, the GX is set up on the soil
surface rather thar. in a borehole as with the GP. The field saturated
hydraulic conductivity is obtained by using twc or more measurements of the
wetted surface of the disk attachment.
Met hod--
The ATI involves determination of two series of data for infiltration
rate versus time. The test is run with a particular size porous plate
until a steady-state condition is achieved. This procedure is repeated
with a different size plate until a steady state is attained. The GI
r-ethod is similar to trie G? method except in those areas discussed ir. these
sufcsectiens.
Acc a rat u s - icure 21 is a schematic diagram of the ATI. The main
features are listed he re:
1; Soil cor. tact base (porous plate/ faceplate) 0.12 and 0 15 m
diameter with a hele per lC-n.-n^ area.
2; Better?. £r.a top transducers.
Air entry ports with va.ves .
6; Marictte column.
t: Data logger .
9: t0C-rr.esn nylon filter.
The schematic of the GI (Figure 22) snows the design of trie disk
attach-ent and a second Mariotte attachment (cartridge). Essential
features are listed here:
1! Nylon mesh disk.
2: Support screen.
3; Air tip seating washer.
< ; Existing main body of Guelph perrreameter .
-------
SEPTUM
TOP TRANSDUCER
to data logger
VALVES
WATER LEVEL
AIR ENTRY
PORTS
MARIOTTE COLUMN
WATER LEVEL
BUBBLING TUBE
BUBBLE TOWER
to data logger
BOTTOM TRANSDUCER
SOIL CONTACT BASE
FIGURE 21. SCHEMATIC OF AN AUTOMATED TENSION
INFILTROMETER (ANKENY el al. 1988).
68
-------
^CARTROGE (second Moriotte)
TO APPLY A NEGATIVE HEAD
FROM I TO 15 CM OR MORE
1 1
ATTACH TO THE AIR TUBE
> EXISTING main booy
OF GUELPH PERMEAMETER
AiR TIP SEATING
WASHER
TAPPERED CAVITY
SUPPORT SCREEN
NYLON MESH DISK
FIGURE 22. SCHEMATIC OF GUELPH iNFILTROMETER (BAUMGARTNER
et cl. 1907).
69
-------
5) Cartridge Cseccnd Mariotte supply tube).
Procedure—The following is a listing of the procedures used for
installation and operation of the ATI:
1) Assemble the inf ilt rometer with a known size porous plate.
2) Record Mariotte column diameter.
3) Place a 400-mesh nylon filter (air-entry value of about 250 mm of
water) or. the soil.
4) Place assembled infiltrometer on the filter (align contact base
with filter) .
5) Till Mariotte column with water.
6) Cper. ore cf the valves on the air-entry ports tc allow water to
flow tc the pcrous plate at a predetermined tension (ports can be
set to tensions of 0.02 to 0.5 rr. by operation of tr.e appropriate
valve).
7t Program, data logger to record paired readings cf top and bottom
transducers at regular intervals.
6) Repeat Steps 1 through ~> with a different size porous piste.
Procedures for the C-I are as follows:
ii Measure the radius cf the disMs; .
2) Assemble the infiltrsmeter.
3; Till the reservoir anc cartridge with water.
4) Place inf iltroneter on soil surface. (A thin layer cf sand is
required on the surface tc ensure good contact between the porous
plate ar.d the field soil.)
5) Start ir.f iitro.T.eter by raising the air-inlet tube out of the
cut let port.
6) Monitor the rate of fall of water level in the cartridge/reservoir
until a steady rate is attained.
*7) Record the radius of the disk (wetted suuface of disk attachment)
3) Change the disk attachment and repeat Steps 2 through 7.
S) Calculate the field saturated hydraulic conductivity by the
simultaneous equation approach by using Equation 31 or the least
squares analysis.
70
-------
Calculat ions--Eauat ion 27 is used with the ATI to calculate the
saturated hydraulic conductivity cf the soil:
Ksat = !Ilri " I2t2,/lri " r21 (27)
where ^sat = saCuratec:' hydraulic conductivity, m/s
r, and r0 are radii of the infiitrometer soil contact base, :n
J. Z
I. and 12 are the steady infiltration rates using soil contact
bases r^ and ^ respectively, n/s
Wooding (1963) presented calculations for use with the GI . For a
surface disk with zero poncing, he showed that the steady-state three-
dimer.s ior.a 1 intake rate is given by:
Q " r2Kfs - 4 ro(n (25)
where 0 = steady-state water flow rate, m"/s
r = radius cf the wetted surface of disk, r:.
K- « field saturated hydraulic cor.ductivicv, rr./s
-5 2 "
o_ = matnc flux potential, ;» / s
for the two-ring method shewn ir. the procedures, K-e is obtained by
use of the simultaneous equation approach:
Kfs = -M-2Q2 " Ml°l !29!
wr.ere >!- = r2 ¦' s'ir 2 • r2 " L'-'
"2 = rl/(rlr2fr2 " rI'; (1/^;
r, = radius of wetted surface of Dis.< 1, :r
r~ = radius cf wetted surface of Disk 2,
^ -
Cn = steady-state flew rate using Disk 1, rr."/'s
Ci = steady-state flow rate using Disk 2, :r.J/'s
Quality Assurance--A primary problem with tr.e per-us plate method is
obtaining sufficient sensitivity in tne mst rur.ents :: measure low
hydraulic conductivity rates. Larger-diameter porous plates and small-
diameter reservoir tubes may be used to improve the ser.s.t i vity of cnese
instruments.
Example Data--
N"o data are availaole cn the use of an AT I w*t:« compacted clay soi.s.
Data provides Dy Ankeny et al. ;193S) for a coarse sane. snowed that, the
method can be used to measure hydraulic conductivities ci 1 x 10 ~ to
- h 7
£ x 1C m/s The studies were done on a 0.C1- to 0 . C2-m~ area of soil.
.Research is continuing on the use of tne tension ir.f i It romet c- r to determine
field hydraulic conductivity cf clay liners over a l^rcer area.
No field data are yet available on the use of cue u;
liners . A series cf tests on compacted soil liners '..ere
spring cf 1955 at the L'niversity of Gueipr..
i compacted soil
olannec for the
71
-------
C OT-T.e ncs —
The pcrocs plate methods which essentially evaluate unsaturatec
hydraulic conductivity, are currently being adapted for use on saturated
soils. No data are available on soils with hydraulic conductivities of 1 x
- 9 - *
10 n/s or less. Only small areas of scil have been evaluated with this
method; therefore, results may net be representative cf the overall soil
lir.er. Further testing of this method cn compacted clay soils is planned
by U.S., Canadian, and Australian researchers. Investigation of the effect
of bigger disk diameters on the determination is also planned. The method
has promise for distinguishing between the cverall field hydraulic
conductivity and the component of flow resulting from aacropores. At the
time c£ issuance of this report, however, research nad not progressed far
enough to suggest the er.cir.eerir.c corr-muriity' s j:e of tnis methed on soil
1 ir.ers .
SOIL CORES EVALUATED IN THE LABORATORY
A variety of methods are available for obtaining the hydraulic
conductivity cf scil cores in the laboratory. Several of these methods
nave ceen discussed by U.S. EPA (196 < : an a Go.d.T.an et al. (1 93 6; .
Laboratory m.ethods are not believed to provide a reliable indicator of
field performance of clay liners (U.S. EPA 1985;. Because widespread use
has been race of laboratory hydraulic conductivity tests as indicators of
field perforn-.ar.ce, some cf these methods are c:_efly discussed nere.
Principles
A ccmpactec soil liner car: ce eval_=cea cy collecting relat-vely
_ndist-roed cores for use in hydraulic conduct.vity tests in a laboratory
setting. Soil lir.er samples car. cither ic u^rved cut of tne soil cr
letted by pushing a thin-walled tuce in ti soil. The samples
i r. ir.c: - t, 1 n — wa - -L0Ci cubts Ccf, cicr.cr c.1? tssicc wn i ie in c.ri£ z uoc
.fixed-wall permeame t e r i or removed from the tuce, trimmed, and piacec ir. a
flexicle-wall permeameter.
The following discussion covers fixec-va11 and f1 exib 1 e-wa11
perrear.eters and tne constant-head retr.cz usee to calculate hydraulic
conductivity. The fallir. g-nead net nod for calculating hydraul ic
conduct ivity is not discussed in this document because it has limited
utility witn tests on low-hydraulic-conductivity compacted soils. In
general, high pressures are applied in tests or. compacted materials to
speed testing time. The pressures usee are typically equivalent to a
height of water of several meters. Consequently, sr.y liquid level arcp m
the water supply burette used in a falling-read method wouid be negligible
co-pared to the total applied pressure.
nCpaTcIUS—
"ixed-Wall Perneameter--The fol
fixed-wall permeameter (Figure 23!:
of the parts of a
-------
1) Double-zing base plate (optional) .
2) Gaskets.
3) Tcp plate with pressure input port.
4) Ir.ner ring leachate outlet.
5) Outer ring leachate outlet.
Flexible-Wall PerrTieameter--The following is a list of the parts of the
flexible-wall perir.eametet (Figure 2 4) ;
I! Steel pedestal under a porous plate.
2; Soil sample encased in a flexible memDrane.
3! Steel cap ever a pcrcus plate.
4) Liquid feed lines.
5) Liquid supply reservoir.
?rccedures--
Fix.e 3-Wa 11 Perxea^.ete r--The following is a listing of the procedures
required in the use of a fixed-wall perxearr.eter:
II Press stainless steel cylinder into soil liner.
2) Retrieve cylinder with soil; cap both ends of cylinder for
transportation to laooratory.
Rer.ove er.ti cap fro.T. totto.t, of cylinder.
4) Gently scrape away soil to level the surface of the sample tc
approximately 1.3 err. frorr. the bottCPi of the cylinder.
;) ?lsce geotextils faoric into the double - r ir.c insert
-------
< PRESSURE INPUT
r -r¦
¦1
1 . L>
N
¦- / /v
¦///>>
/
\
(INNER) LEACHATE OUTLET
(OUTER) LEACHATE OUTLETS
GASKET
LIQUID
STAINLESS STEEL
CYLINDER
¦INNER RING
SOIL SAMPLE
¦GASKET
DOUBLE-RING
BASE PLATE
FIGURE 23. SCHEMATIC OF A FIXED-WALL PERMEAMETER
(MODIFIED FROM DANIEL et ol 1985.)
74
-------
LIQUID SUPPLY
RESERVOIR
&
J
LIQUID FEED LINES
STEEL CAP
SOIL SAMPLE
ENCLOSED IN A
FLEXIBLE MEMBRANE
STEEL PEDESTAL
FIGURE 24. FLEXIBLE-WALL PERMEAMETER.
75
-------
10) Tighten down the top plate with hex nuts.
11) Apply liquid to the top of the soil.
12) Fit pressure source to assembled permeameter and place liquid
collection flasks on the inner 2nd outer leachate cutlets.
13) Apply low pressure to the liquid and record time and pressure.
14) Periodically record the time, volume of liquid in the flasks, and
the pressure on the liquid.
15) Pass liquid through the soil until steady-state flows have been
attained in both the inner ar.d outer compartments.
Flexible-Wa11 Permeameter--Flexible-wall hydraulic conductivity tests
are performed in tnaxial perjr.earr.eters . The following procedures apply to
tr.e perr.ear.eter shewn in Figure 2 4:
1) Extrude the soil sample from a Shelby tube.
Trim sample to des.red size.
3) Place porous plate or. tr.e steel pedestal.
1 Place trimmed sample on top of the porous plate.
5: Place a pcrous plate cr. tcp of the soil and put the steel cap or:
tcp cf the plate.
c) Place a flexible membrane over the steel cap, soil sample. and
steel pedestal.
") Secure flexible membrane to the steel pedestal ana steel cap with
suitable rubber bands.
5) Place prepared soil sample en base plate.
ji Ccr.r.ec" pressure lir.es and reservoirs.
1C) Apply appropriate confining pressure ar.d back pressure to saturate
scil fully.
11 i Apply liquid under required pressure t: the top cr the soil to
start test.
Calculaticns--
H.ydraulic conductivity tests dene in tr.e loocrscory an compacted scii
samples are typically conducted by the const one-head method. Calculations
used in these determinations are provided .".ere . The ccn3tc::t-:iead method
involves applying water to the soil with a constant pressure and measuring
the quantity of water that is discharged ir. a giver, penoa or time. Tr.e
76
-------
head less across the scil thus remains constant throughout the test, and a
forrr. of Carey's Equation (Equation 30) is used to calculate the hydraulic
conductivity of the soil.
k - Q/(A i t) (30!
where k = hydraulic conductivity, m/s
Q = quantity of flow ir. a giver, time, m"*
u
hydraulic gradient » K/L
hydraulic head difference,
L = deDth cf soil, m
2
A = area of flow, rr,
t = time for flow measurement, s
The two main techniques used to maintain constant head are 1) the use
of a large reservoir of water so that the volume of water lest from the
reservoir during the test will cause a negligible change in reservoir water
level, and 2) the use of a Kariotte bottle arrangement.
Quality A3surar.ce--
Laboratory samples are usually too small to include a representative
distribution cf the macrcfeatures present ir. a soil liner. Because scil
imperfections may ae widely spaced, it may not be practical to collect
samples ir. the field that are large enough to assure tnat the laboratory
test yields representative nydraulic conductivity values.
Example Data
A prototype compacted soil liner was constructec, and detern:ir.acior.s
cf hydraulic conductivity were made by laooratcry ana field methods.
JElsbury et al. 1968). Laboratory hydraulic conductivity values obtained
or. Snelfcy tube samples it. flexible-wall permeameters ranged from 1 x iC~^
to 1.2 x 10 r./s . Hydraulic conductivity values obtained with samples
compacted in the laboratory in fixec-wall permeameters rar.ced from 2 x 10
"" to 1 x IS ** m/s. Field hydraulic conductivity values obtamec with a
sealed double-ring infiltrcmeter ranged from 2.6 x 10 c to 2.0 x 10"' m.'s.
Nordquist et al. (1986) measured hydraulic conductivity cf Snelby ruse
samples to iter. from, three test-fill liners in Oklahoma and Utan. The average
hvdraulio conductivity value obtained from these samples in fixed-wall
-1 0
perrriearr.eters was 3 x 10 * m/s, whereas the average field hydraulic
conductivity value obtained with the air-entry permeameter was 3 x 10 ^
m/s. These data showed that the laboratory and fieid data were within one
order of magnitude (Table 2) .
Day and Daniel (1965) presented summary ar.d comparison data on labor-
atory and field hydraulic conductivity of a compacted clay liner (Table 8).
Hydraulic conductivity values determined in the laboratory were two to
three crders of magnitude lower than the field-determined values.
77
-------
TABLE S. COMPARISON OF FIELD AND LABORATORY HYDRAULIC CONDUCTIVITY DATA.
Type of hydraulic conductivity Average hydraulic conductivity, m/s
test Clay 1 Clay 2
- 0 - 9
Pond test an entire liner S x 10 4 x 10
Laboratory test (all types)
on laboratory compacted samples 1 x 10 ^ 2 x 10"*"
Laboratory tests (flexible-wall)
or. hand-carved samples 1 x 10 3 x lC~~i
Laboratory tests (flexible-wall)
on Shelby tube samples 1 x IG""1^ 3 x 10_1*
Laboratory tests on hand-carved samples
tri.T.T.ed into consolidation ring and
tested at an effective stress of 96 kPa 1 x 10 ' < x 10 lv"
- 8 — u
Ring infiltration test in field 5x10 3x10
s
laboratory hydraulic conductivity tests can measure very low hydraulic
conductivity values; however, using these values to represent field
conditions presents several problems. Lacoratory samples usee to determine
r.ydruji.c conductivities are typically too small to have a representative
distribution of the macr©pores present in the field. Consequently, t.'.ese
tests tend to give values that are one to three orders of magnitude lower
than the actual field values. Application of confining pressure or. the
sarple during flexible-wall tests also may seal cr reduce tne size of any
-acropcres present in the soil sample. The absence and/or reduction of the
rraeropores will result in a lower hydraulic conductivity than that of the
actual liner.
Side-wall leakage can be significant in laboratory tests, and every
effort should be made to minimize this effect in laboratory tests. Oo.ole-
rir.g cr flexible-wall permeameters (Figures 23 and 2-5) can be effective at
detecting or minimizing, respectively, side-wall leakage in laooratory
tests. Double-ring permeameters give indications of the relative magr.-tuae
of the side-wall leakage (i.e., no side-wall leakage is occurring i: the
flow rate per unit area in both the inner ring and outer ring are the
same). Test results should be considered invalid if tne flow rates per
ur.it area differ significantly (more than 50%) . Flexible-wall permeameters
recruire confining pressure to minimize side-wall flow. This confining
pressure can close macropores that may be present ir. the soil sample, and
thereby reduce the hydraulic conductivity below tne actual values that
would occur under field conditions.
-------
The termination point of a test is critical to t he usefulness cf the
test data for comparison with field data. Saturated hydraulic conductivity
tests should be run until steady-state values have been attained. A graph
of hydraulic conductivity with time is needed to determine if steady-state
hydraulic conductivity values have been obtained.
-------
REFERENCES
Aidabach, A. S., and C. E. Beer. 1971. Field Measurement of Hydraulic
Conductivity Above a Water Table With Air-Entry Per.T,eameter
Transactions of the ASAE. pp. 29-31.
Anderson, D. C., anci J. 0. Sai. 1 988. Unpublished Data From Private
Clients. K. w. Brown S Associates, Inc.
Anderson, d. C. , J. 0. Sai, ar.d A. Gill. 1905. Perr-eabilicy of
Cor.;an:r.a:ed Soil to Nutrient Solution. Prepared for the U.S.
Er.virorr.er.tal Protection Agency under Contract 68-03-ic j 6 .
Anderson, D. C., J. C. Sai, and A. Gill. 15 3 4 Surface Impoundment Soil
Liners: Permeability and Morphology of a Soil Liner Permeated cy
Acid, and Field Permeability Testing for Soil Liners. Prepared for
the 'J.S. Environmental Protection Agency under Contract 56-03-2943.
Ar.ser.y, M. D., ?. C. Xasper, and R. Hortc-r:. 1966 . Submitted to J. Soil
Science Sue. of AneriCs . (I:- Press) .
America.'. Society of .estmg and Materials (A;-X) . 190 6. Annual Boox. of
ASTM Standards . Section 4, Vol. 4 . C- 9 .
5aoopc_los, A. 1586. Devclcpnsr.t ar.d Operation of the Keelc- Valley
Landfill Site. Maple, Ontario, Canidi. Presented at Ninth Annual
Madisor. Waste Conference, September 9-1C, 1 586. Dept. of Engineering
Professional Development, University of Wisconsin, .Madison.
Bagchi, A. 1S6G. Landfill Geost ructure Construction Using Blended Soil.
In: Proceedings of International Symposium on Environmental Geotech-
nolcgy. vol. 1. Hsai-Yar.g Fang (Ed. ) . Envo Publishing Co. Inc
3aun.gcitr.er, N. , D. E. Elriok., and K. £. Bracshaw. 1967. In-Situ
Kyara_lic Conductivity Measurement of Slowly Pori.eable Materials 'Jsmg
Modified Guelph Permeameter and the Guelpn Infiitroreter. In:
Proceedings of the First National Outdoor Action Conference on Aquifer
Restoration, Ground Water Monitoring and Geophysical Methods. May 16-
21, 19 8 7.
Bcutwei1, G. and R. K. Derric. 1966. Sanitary Landfills in the
Saturated Zone. Paper presented at Waste Tech '66. National Solid
Waste y.anagerent Association, Chicago, Illinois.
ec
-------
Bcutwell, G. P., and V. R. Donald. 1982. Compacted Clay Liners for
Industrial Waste Disposal. Presented at cne American Society of Civil
Engineers National Meeting in Las Vegas, Nevada, April 26, 1582.
Bouwer, H. 1366. Rapid Field Measurement of Air-Entry Value and Hydraulic
Conductivity of Soils as Significant Parameters in Flow System
Analysis. Water Resources Research, 2:723-732
Bouwer, H. 1978. Groundwater Hydrology. McGraw-Hill, Toronto, Ontario.
4 80 pp.
Bradshaw, K. 1. 1986. An Assessment of the Standard Percolation Test.
Master's Thesis, University of Waterloo, Ontario.
Srown, K. K. 1965. Kiiiianscr. Cree* Wastewater Treatment Plant Irrigation
Field Infiltration Studies. Prepared for a Private Client by X. W.
Browr. i Associates, Inc., College Station, Texas.
Clothier, E. E., anc I. White. 1931. Measurement of Sorptivity and Soil
Water Diffusivity in the Field. Soil Sci. Soc. Am. J., 45:241-24 5.
Cc.-.iei, D, E. 1 3 37. .Hydraulic Conductivity Tests for Clay Liners. In:
Proceedings of Ninth Annual Synpos iu:ti or. Geotechr.ical and Geohydro-
logical Aspects of Waste Management, Feb. 1-6, 1967, Fort Collins,
Colorado.
Dsr.iei, D. £., and S. J. Tra-Jtweir.. 1906. Field Permeability Test for
Earthen lir.er . In: Proceedings, Ir.-Situ 1986, ASCE Specialty
Conference on Use of In-Situ Tests in Geotechnical Engineering,
Virginia Polytechnic Institute, ar.a State University, Blac*sburg,
Virsinia, June 22-25.
Dar.isl, Z. E. , D. C. Anderson ar.c S. S. Boyr.tor.. 1S85. Fixed-Wail versus
Flexiole= 11 Permeameters. H vrir 5 u 11 c 3arrlers :r. Soil a nd Roc <. ASTK
S7P 874, Ar-.ericar. Society for Testing and Materials . pp 107 - 12 6.
Day, S. R. 1594. A field Permeability Test for Compacted Clay Lir.ers.
Yaster's Thesis, University of Texas, Austin, TX.
-ay, S. R., and ~. E. Daniel. 1SS5A. Field Permeability Test for Clay
Lir.ers Hydraulic Sa r r ie rs in Soil and Rock. ASTM STF 374, American
Society fcr Testing and Materials, pp 276 - 238.
Day, S. R. , ar.c D. E. Daniel. 15653. Hydraulic Conductivity of Two
Prototype C_ay Liners. Journal of Geotecnnica1 Engineering, ASCE,
III I?) : 3 5 7 - 9 7 C.
Dirkser., C. 1975. Determination of Scil Water Diffusivity by Sorptivity
Measurements. Soil Sci. Soc. Am. Prcc., 33:22-27.
-------
Elsbury, 3. R., G. A. Sraders, D. C. Andersen, J. A. Rehage, J. 0. Sai, and
D. E. Daniel. 1988. Field Laboratory Testing of a Compacted Soil
Lir.er. Prepared for the Office of Research and Development, Hazardous
Waste Engineering Research Laboratory, U.S. Environmental Protection
Agency, PB-89-125942.
Gardner, W. R. 19S6. Some Steady-State Solution of the Unsaturated
Moisture Flow Equations Kith Applications to Evaporation From a Water
Table. Soil Science, 85 { 4) : 228-232 .
Golder Associates. 1986. In-Situ Permeability Test Procedures, Texas
Scclogist Facility. (Final Report). Submitted to U.S. Ecology.
Louisville, Kentucky.
Golder Associates. 1984. Compacted Fill In-Situ Permeability Test Report.
Submitted to Chenical Waste Management Inc., Marietta, Georgia.
Goldr.ar., L. J., L. I. Greenfield, A. S. Canie, G. L. Kingsbury, C. x.
Ncrtheir., and R. S. Truesdale. 1 303. Design, Construction, and
Evaluation of Clay Liners for Waste Management Facilities.
EFA/530-SW-8S . 'J . S . Environments- Protection Agency.
Gordon, M. E., P. X. Huebner, and T. Miazca. 1 S99 . Hyoraulic Conduct.viey
cf Tnree Landfill Clay Liners. ASAE (In Press) .
L e , . P .
Liners
r-. - A C A f-i «¦ (
B. S . 3rown. 19 8 6. Design ar.c Construction of I
3ld Tailings Facilities and Heap Leach Pads. Paps
the 2nd International Conference on C-cld Mining,
ritish Columbia, November 7-9, 1383.
Hcrtor., , M. L. Tho.-psor. and J. F. McBrice. 1 988. Determination of
Effective Pore sity cf Soil Materials. PS — S3 — 242 391 /AS.
¦Isien, . W . , a no ar.ar.cto .
'Waste Management Unit Clay L
Aspects cf Waste Management.
?-ilishers Inc., Chelsea, Ki.
198". Permeability Tests for Hazardous
L.ners . Geotecr.nica. and Gecr.yorolcgica—
D. J. A. van Zyi et al. (Ecs.)• Lewis
:r.igar..
;., H. A. Rizvi, J
Ir.: Proceedings
:e, December 1987,
. Ahmed, R. .-.orton, Cr
;f the Third Internal
Chic State University
. ana S. Z . Ka rley.
e:.al V.'or/.sncp cn Lar
, Co.u.''Joui , Chic.
Kir.et, ?. 13 5 3. Design cf Ccilection Basins (Collection Lys.meters i .
State cf Wisconsin Interoffice Correspondence/Me.",orandum. 47 pp.
Kr.et, P., ar.c D. E. Li.ndorff. 1983. Use of Collection Lysirc-ters in
Monitoring Sanitary Landfill Perforr.ar.ce. Presented at tne National
Water Well Association Conference on tr.e Characterization and
Monitoring cf Vadcse (Unsaturates 2or.e, Las vecas, Nevada, Deceinber
c - - C , I 9 S 3 .
82
-------
Kmet, P., G. Mitchell, and M. Gordon. 1 966. Leacr.ate Collection System
Design and Perforrr.ance--Wiscor.sir-'s Experience. Presented at ASTSWMO
National Solid Waste Forum or. Integrated Municipal Waste Management,
July 17-20, 1S88, Lake Buer.a Vista, Florida.
Kr.ignt, R. 9., and J. P. Haile. 196 4. Construction of the Key Lake
Tailings facility. In-. Proceedings of the International Conference
on Case Histories in Geotechr.ical Engineering, May 6-11, 1S84 . Vol.
McKeague, J. A., C. Wang, and G. C. Topp. 1982. Estimating Saturated
Hydraulic Conductivity From Scil Morphology. Soil Sci. Soc . Am. J,
<5 6:1239-1244.
Merva, G. E. 1SS7. The Velocity Per.Tieameter Technique for Rapid
Determination cf Hydraulic Conductivity In Situ. Presented at the
Workshop on land Drainage, December 1S87, Ohio State University,
C01 unX"us, Chic .
ar
D., G. J. Surch, and P. J. Walitrink,
hydraulic Conductivity of Forest Scil:
I 7 o - 8 8 1 .
1 9 S 6 . Preferential F.ow
£o-l Sci. Soc. A:r.. J.,
N-uruquis;, J. £. , G. S. Mason, Jr., and G. L. Olson. 1966. Clay Liner
Hydraulic Ccncuotivity. Comparison cf Field and Laboratory Tests.
In: Proceedings of Specialty Conference, l&ei/SE D i v / ASCE/'Cmcinriat i ,
Chic, Jjiy 3-10, 1966. pp. 401-406.
: n <,
•s, 0 r.:
A BAT Method fcr In-Situ Measurement ci Hydraulic
:, i w e c e r.
saturate;
5 4 o o .
Master's Thesj
srversity or
. 15? 5. A Device fcr In Situ Measurement cf Hydraulic
Ivity. Presented at the AGL Syrr.pos lum or. Advances in Hydraulic
and Tracer Methods. San Fra.nciso, CA. December 10, 1985.
:r.v, A X. , w. Romanows k i, ana V. Richards. 198"?. Field Perrr.eabi I
Testing fcr a RCRA Landfill Final Cover. Presented at Hazardous
Materials Conference. Santa Clara, CA. April 22, 159"
Peaces, I. 1 93 6. Laboratory and Field ?e rmeabi 111 v Tests or. Clay Liners.
Waste Tech. In: Proceedings of National Solid Waste Management ASSH
ar.; Waste Age Magazine. Chicago, IL. Oct. 20-22, 1566. 7 p.
Rer.age, J. A., R. L. Shiver, D. C. Anderson, W. . Crawley, M. L'nger a no C.
C:r
-------
Reynolds, W. D., and D. E. Elrick. 1985a. In-Situ Measurement of Field-
Saturated Hydraulic Conductivity, Sorptivity, and the a -Parameter
Using the Guelph Permeameter. Soil Science, 11C (4! : 292 -302 .
Reynolds, W. D.( and D. E. Elrick. 1965b. Measurement of Field-Saturated
Hydraulic Conductivity, Sorptivity, and the Conductivity Relationship
Using the Guelph Permeameter. In: Proceecings of the National Water
Well Association Conference on Characterization and Monitoring of the
Vadose ("Jr,saturated) Zone, Denver, CO, November, 1 985.
Reynclcs, W. D., and 0. E. Elrick. 1986. A Mechod for Simultaneous Ir-
Situ Measurement in the Vadose Zone of riela Saturated Hydraulic
Conductivity Sorptivity, and the Conductivity-rressure Head
Relationship. Ground Water Monitoring Review, £: 84-35.
Rogers, J. S., J. L. Fouss, ar.d C. E. Carter. 1 937. Field Experience With
a Velocity Head Perneaneter. (Ir. Press)
Rogc-wski, A
Doc n.ey
VUIaU
Jazooy,
Jr.,
Richie, W. M. Yazujian, ana P. J.
1964. Relationship of Laboratory s:.J fie.d Deternmed
: Conductivity in Compacted Clay Liner. Progress Report Phase
¦)W ¦ 7 c - " r " "i - n 1
S It * 4. „ W V V V * V .
Stephens, D. 5., M. Ur.ruh, J. ravler.a, R. 3. Know It on Jr., E . Kittson, and
w. Cox. 1556. Vadose Zone Characterize tier, of Lcu-Perr.eabi 1 ity
Sediments Using Fiela Permeameters . Ground Wat ex Mor.i to i _:.c Review.
Stewart, U. and T, W. N'olan. 1 967. Inflic ration Testing for hydraulic
Conductivity of Soil Liners. GTUODJ, i C (2 ) : 41 - 5 C .
loop, G. C., and M. S. 3inr.s. 1 976. Field Measurement of Hydraulic
Conductivity Witr. a Modified Air-Er.try Perneameter. Cm. J. Soil Sci.
i. '.96 4. A New System for G;
:orir.c Review, 4(4; : 131-136 .
Triad Engineering Consultants, Inc. 1966. Field Perr.eaoility Tests for
Compacted Clay Liner Panic Pond Upgrade. Union Carbide Corporation,
Sistersville, W. Virginia.
U.S. Environmental Protection Agency. 1986. Construction Quality
Assurance for Hazardous Waste Land Disposal Facilities. Tecnnical
Guidance Document. EPA/530-SW-66-031.
U.S. Environmental Protection Agency. 1965. Minimum Technology Guidar.ce
or. Double-Liner Systems for Landfills and Surface Impoundment s - -
Design, Construction and Operation. EPA/530-SW-65-014.
-------
U.S. Envi ror.rr.er.ca 1 Protection Agency. 1984. Sox- Properties,
Classification and Hydraulic Conductivity Testing. Technical Resource
Cocuner.t, EPA/53Q-SW-84-925 . 167 pages.
Weeding, R. h, 1968. Steady Infiltration From Shallow Circular Pond,
water Resources Res., 4:1259-1273.
Woodward-Clyde Consultants. 1987. Response to U.S. EPA Coramencs of
February 17, 1987, Technical Memorandum, Evaluation of Results of
Field and Laboratory Permeability Testing of Clay Liner and Corr.pleticn
Report No. 5-30, Construction of Test fill for Disposal Ceil No. 2.
-------
TECHNICAL REPORT DATA
(Please ret J Instruction! on ihe mtrtt btfote conptctt'
1. REPORT NO. 3.
EPA/600/2-91/022
3 i
4 title and subtitle
State-of-the-Art Field Hydraulic Conductivity Testing
o* CoTpacted Sci Is
5 REPORT OATE
June 1991
8. PERFORMING OROANIJaTIO* CODE
7 AUThORIS)
J. 0. Sai anc D. C. Anderson
B.PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORGANIZATION name and address
K. 'A. Brown ard Associates, Inc.
College Statior, TX 77840
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO
68-C3-3413, WA 0-23
12 SPONSORING AGENCY NAME AND ADDRESS
Risk Reduction Engineering Laboratory—Cincinnati, OH
3*'ice of Research and Development
U.S. Env'rornenta1 Protection Agency
C-ncirnati , OH <15268
13. TYPE OF REPORT AND PERIOD COVERED
Pro.iect Report
14 SPONSORING AGENCY COOE
EPA/600/14
15 SUPPLEMENTARY NOTES
Project Officer: Walter £. Grube, Jr. FTS: 684-7798 COMM: 513/569-7798
16. ABSTRACT
Th's report documents the available technical information on field hydraulic
conductivity test xethods for soil liners. The methods discussed are currently used
and reacily avei'able for determining the hydraulic conductivity of soils compacted in
the field.
Hydraulic conductivity test methods currently used on soil liners were evaluated
for their ability to neet the r-inimum requirements for field tests; i.e., thatthe
test be capable of measuring hydraulic conductivities at least as low as 1 x 10" m/s
ard that the values obtained be representative of the overall soil liner. Of the few
netr.ods capable of meeting the rinimum requirements, even fewer are both practical to
use and rare'y give false low values. Based on the advantages of the methods evaluated,
t.ne best and most practical currently available technologies for evaluating hydraulic
corduCtivity a^e large single-ring infi1trometers and sealed double-ring infi1trometers.
17. KEY WORDS AND DOCUMENT ANALYSIS
1. DESCRIPTORS
b. IDE NT IF IE RS/DPEN ENOEO TERMS
c. COSATl Field/Group
hydraulic conductivity, infiltration
field, test methods,
soil liners
IB. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS iThuRtpon)
UNCLASSIFIED
21. NO. OF PAGES
Q?
20 SECURITY CLASS (Thti paftj
UNCLASSIFIED
22 PRICE
eP* ftm 1220-1 (R»». 4-77) p«tviou» IBlTION IIOIIOLCTC ^
------- |